LEGIONELLA 2009
October 13-17, 2009 Institut Pasteur, Paris, France
SCIENTIFIC COMMITTEE
Paola BORELLA Elizabeth HARTLAND University of Modena and Reggio Emilia, University of Melbourne, Melbourne, Australia Modena, Italy Carol JOSEPH Carmen BUCHRIESER Health Protection Agency, London, United Institut Pasteur, Paris, France Kingdom
Nicholas P. CIANCIOTTO Jaana KUSNETSOV Northwestern University Medical School, National Public Health Institute, Kuopio, Finland Chicago, USA Christian LÜCK Didier CHE National Reference Center for Legionella, Institut de Veille Sanitaire, Saint-Maurice, Dresden, Germany France Carmen PELAZ Paul H. EDELSTEIN Instituto de Salud Carlos III, Madrid, Spain University of Pennsylvania Medical Center, Philadelphia, USA Rod RATCLIFF Institute of Medical & Veterinary Science, Jérôme ETIENNE Adelaide, Australia National Reference Center for Legionella, Lyon, France Michele S. SWANSON University of Michigan Medical Center, Ann Barry FIELDS Arbor, USA Center of Disease Control (CDC), Atlanta, USA Keizo YAMAGUCHI Jörg HACKER Toho University, Tokyo, Japan Robert Koch Institut, Berlin, Germany
ORGANIZING COMMITTEE
Carmen BUCHRIESER Institut Pasteur, Paris, France Jérôme ETIENNE National Reference Center for Legionella, Lyon, France Didier CHE Institut de Veille Sanitaire, Saint-Maurice, France
ENTIFIC
ACKNOWLEDGEMENTS
We are grateful to the following organisations for their generous financial support Nous remercions chaleureusement nos partenaires financiers
INSTITUT PASTEUR
EUROPEAN MOLECULAR BIOLOGY ORGANIZATION (EMBO)
FEDERATION OF EUROPEAN MICROBIOLOGICAL SOCIETIES (FEMS)
NETWORK OF EXCELLENCE “EUROPATHOGENOMICS” (NOE EPG) OF THE EUROPEAN COMMISSION
REGION ILE DE FRANCE
INSTITUT NATIONAL DE VEILLE SANITAIRE
ACKNOWLEDGEMENTS
We are grateful to our sponsors Nous remercions chaleureusement nos sponsors
SPONSORS
SPONSORS
OTHER PARTNERS
TABLE OF CONTENTS
Theme of the conference...... I
Scientific program...... II
Oral communications...... 1
Opening session...... 2 Session 1: Epidemiology and Clinical aspects: a) disease, diagnosis and treatment ...... 6 Session 2: Clinical aspects: b) detection and subtyping ...... 12 Session 3: Host - Microbe interactions: a) secretion systems and their substrates...... 18 Session 4: Host - Microbe interactions: b) metabolism and virulence ...... 24 Session 5: Immunology and host susceptibility...... 28 Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilm...... 33 Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence ...... 39 Session 8: Legionella prevention and control ...... 44 Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals ...... 48 Session 10: Epidemiology: b) outbreaks and population genetics and taxonom...... 54 Session 11: Genomics and Comparative genomics ...... 60 Session 12: Functional genomics ...... 65 Session 13: Gene regulation in the environment and the host ...... 70 Session 14: Cell biology - protozoa – macrophages...... 74
Posters ...... 78
Session 1: Epidemiology and Clinical aspects: a) disease, diagnosis and treatment ...... 79 Session 2: Clinical aspects: b) detection and subtyping ...... 107 Session 3: Host - Microbe interactions: a) secretion systems and their substrates...... 126 Session 4: Host - Microbe interactions: b) metabolism and virulence ...... 141 Session 5: Immunology and host susceptibility...... 158 Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilm...... 165 Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence 179 Session 8: Legionella prevention and control ...... 211 Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals ...... 267 Session 10: Epidemiology: b) outbreaks and population genetics and taxonom...... 277 Session 11: Genomics and Comparative genomics ...... 289 Session 12: Functional genomics ...... 297 Session 13: Gene regulation in the environment and the host ...... 299
Authors index...... 306
Detailed summary...... 315
Company profiles of the sponsors ...... 337
THEME OF THE CONFERENCE
The International Conference on Legionella is a premier international meeting held every four years to discuss new findings on Legionella pneumophila, other Legionella, and the disease caused by these organisms and also to identify knowledge gaps that exist in controlling Legionella infections of humans. These four-day meetings have been held in various locations in Europe and the USA. The 6th International Conference, Legionella 2005 in Chicago, was attended by many international leaders in Legionella research and attracted 400 attendees to participate in nearly 40 invited lectures, 200 poster presentations, three panel discussions, and numerous opportunities that fostered scientific collaboration.
The 7th International Conference in this series, Legionella 2009, will address a wide range of current research and trends related to Legionella. Sessions will be dedicated to epidemiology and clinical aspects; pathogenesis and immunology; genetics and genomics; ecology and evolution; physiology, regulation, and biochemistry. The Conference will be of interest to researchers; those with clinical, public health and epidemiological interests; regulatory agency representatives; representatives from industry with interests in Legionella and Legionnaires’ disease. Participation by students and scientists in the early stages of their careers is highly encouraged. Legionella 2009 will provide abundant opportunities for networking and exchange of information between senior scientists and professionals, new investigators and students.
La conférence internationale sur Legionella (LEGIONELLA 2009) est une rencontre internationale qui se tient tous les quatre ans pour discuter des nouvelles découvertes sur Legionella pneumophila et autres Légionelles, de la maladie causée par ces organismes ainsi que pour identifier des problèmes existant dans le contrôle des infections humaines à Legionella. Ces rencontres de quatre jours ont été organisées dans différentes villes européennes et aux Etats-unis. La 6ème conférence internationale, Legionella 2005 à Chicago, a vu la participation de nombreux leaders internationaux de la recherche sur Legionella et a attiré 400 personnes qui ont participé aux 40 séminaires invités, 200 présentations de posters, trois discussions générales, profitant également des nombreuses opportunités pour initier ou approfondir des collaborations scientifiques. La 7ème Conférence Internationale dans cette série, Legionella 2009, adressera de nombreuses questions de la recherche courante et de tendances concernant Legionella. Les thèmes abordés porteront sur l’épidémiologie et l es aspects cliniques, la pathogénicité et l’immunologie, la génétique et la génomique, l’écologie et l’évolution, la physiologie, la régulation et la biochimie.
La conférence sera d’un grand intérêt pour les chercheurs, qu’ils soient intéressés par des questions cliniques, de santé publique ou par l’épidémiologie mais aussi pour les représentants des agences de contrôle, de gestion du risque et de surveillance ; les représentants de l’industrie intéressés par Legionella et la maladie de Légionnaire. La participation des étudiants et des chercheurs en début de carrière est fortement encouragée. Legionella 2009 fournira beaucoup d’opportunités pour la création de réseaux de travail, l’échange d’informations entre chercheurs seniors, professionnels, jeunes chercheurs et étudiants.
I
SCIENTIFIC PROGRAM
Tuesday, October 13th, 2009
10:00 am-4:00 pm Ewgli meeting for Ewgli collaborators only, related to business aspects of the EWGLINET network and the transfer of the scheme to ECDC
2:00 pm-5:30 pm Registration
5:30 pm-6:00 pm Welcome addresses Tony Pugsley, Senior Executive-Vice President for Scientific Affaire, Institut Pasteur, Paris, France Carmen Buchrieser, Institut Pasteur, Paris, France
6:00 pm Opening Session
Chair: Nicholas P. Cianciotto, Northwestern University Medical School, Chicago, USA Barry Fields, Center of Disease Control, Atlanta, USA Environmental approaches to the prevention of legionellosis Carol Joseph, Health Protection Agency, London, UK Legionnaires' disease in Europe 1995 - 2008: Trends and challenges Jörg Hacker, Robert Koch Institut, Berlin, Germany Legionella. A model organism for Genomics and Cellular Microbiology
7:30 pm Welcome Reception at Institut Pasteur
Wednesday, October 14th, 2009
8:30 am-10:15 am Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment
Chair: Didier Che, Institut de Veille Sanitaire, Saint-Maurice, France 8:30am-9:00 am Miguel Sabria, University of Barcelona, Spain Clinical aspects of Legionnaires´Disease 9:00 am- 9:30 am Paul H Edelstein, University of Pennsylvania, Philadelphia, USA Antimicrobial therapy for Legionnaires' disease, still in need of improvement 9:30 am-9:45 am Maria Luisa Pedro-Botet, Spain Legionella pneumonia (LP) in immunosuppressed (IS) patients 9:45 am-10:00 am Frances Graham, New Zealand Changing Epidemiologic Trends of Legionellosis in New Zealand, 1979 - 2008 10:00 am-10:15 am Lauri Hicks, USA Travel-Associated Legionnaires' Disease Surveillance in the United States, 25-28 10:15 am-10:45 am Coffee break
II 10:45 am-12:30 am Session 2: Clinical aspects b) detection and subtyping
Chair: Jaana Kusnetsov, National Public Health Institute, Kuopio, Finland
10:45 am-11:15 am Christian Lück, University Dresden, Germany Microbiological diagnostic methods for the detection of Legionella infections 11:15 am-11:45 am Timothy G. Harrison, Health Protection Agency, London, UK Legionella pneumophila DNA sequence-based typing: progress, pitfalls and perspectives. 11:45 am-12:00 am Patrick Yang, Sweden Evaluation of a fully integrated point-of-testing system for the detection of viable Legionella pneumophila 12:00 am-12:15 am Valeria Gaia, Switzerland MALDI-TOF MS based protein mass fingerprinting: a rapid and reliable method for the identification of Legionella spp. 12:15 am-12:30 am Jonas Lundström, Sweden An improved and rapid sequence-based method for typing of Legionella pneumophila. 12:30 am-2:00 pm Lunch at Institut Pasteur 2:00 pm-3:30 pm Poster session 3:30 pm-5:15 pm Session 3: Host - Microbe interactions a) secretion systems and their substrates
Chair: Craig Roy, Yale University, USA
3:30 pm-3:55 pm Nicholas Cianciotto, North Western University, Chicago, USA Type II protein secretion, siderophores, and pigment and what they mean for Legionella ecology and pathogenesis 3:55 pm-4:20 pm Joseph Vogel, Washington University, St Louis, USA The Legionella Dot/Icm type IVB secretion system 4:20 pm-4:45 pm Antje Flieger, Robert Koch Institut, Berlin, Germany Legionella pneumophila phospholipases and their mode of secretion 4:45 pm-5:00 pm Li Xu Luo, USA Inhibition of Host Vacuolar H+-ATPase Activity by a Legionella pneumophila effector 5:00 pm-5:15 pm Mariella Lomma, France Legionella pneumophila infections involve bacterial F-box proteins 5:15 pm-5:45 pm Coffee break 5:45 pm-7:00 pm Session 4: Host - Microbe interactions b) metabolism and virulence
Chair: Philippe Glaser, Institut Pasteur
5:45 pm-6:15 pm Paul Hoffman, University of Virginia, Charlottesville, USA Host Microbe Interactions: Metabolism and Virulence 6:15 pm-6:45 pm Michele Swanson, Michigan University, USA Legionella pneumophila virulence as a response to starvation 6:45 pm-7:00 pm Yury Belyi, Germany Cytotoxic glucosyltransferases of Legionella pneumophila 7:00 pm-8:30 pm “Wine and cheese” party
III Thursday, October 15th, 2009
8:45 am-10:00 am Session 5: Immunology and host susceptibility
Chair: Ralph Isberg, Tufts University, Boston, USA
8:45 am-9:15 am Tom Hawn, University Washington, Seattle, USA Genetic variation & the pulmonary innate immune response to Legionella 9:15 am-9:30 am Catarina Nogueira, USA Dendritic cells rapidly undergo apoptosis to limit intracellular replication of Legionella pneumophila 9:30 am-9:45 am Tatiana Silveira, Brazil Pore formation is an Nlrc4/Ipaf inflammasome-dependent host cell response against virulent Legionella species that express flagellin 9:45 am-10:00 am Juliane Lippmann, Germany Type I IFNs mediate innate intracellular defense to Legionella pneumophila 10:00 am-10:30 am Coffee break 10:30 am-12:30 am Session 6: Microbe - Environment interactions a) cell biology, protozoa, biofilms
Chair: Agnes Ullmann, Institut Pasteur
10:30 am-11:00 am Michael Steinert, University Braunschweig, Germany Proteomic and functional characterization of outer membrane vesicles and Legionella-containing phagosomes 11:00 am-11:30 am Hubert Hilbi, University of Zürich, Zürich, Switzerland Signals produced and exploited by Legionella 11:30 am-12:00 am Howard A. Shuman, Columbia University, New York, USA Host determinants of Legionella infection 12:00 am-12:15 am Sophie Pécastaings, France A new method of producing Legionella pneumophila biofilms in a minimal medium 12:15 am-12:30 am Rafael A. Garduno, Canada Mature infectious forms (MIFs) and MIF-pellets as factors in the transmission of Legionnaires' disease 12:30 am-2:00 pm Lunch at Institut Pasteur 1:30 pm-3:00 pm Poster session 3:00 pm-4:30 pm Session 7: Microbe - Environment interactions b) detection in natural and artificial reservoirs - survival, persistence
Chair: Yann Héchard, Université de Poitiers, France
3:00 pm-3:30 pm Paola Borella, Modena University, Italy Risk factors for Legionella contamination and persistence in potential sources of infection 3:30 pm-4:00 pm Shin-ichi Yoshida, Kyushu University, Fukuoka, Japan Legionella physiology in the environment 4:00 pm-4:15 pm Rinske Valster, Netherlands Detection of hosts for Legionella pneumophila in freshwater by enrichment of free-living protozoa in a biofilm-batch system 4:15 pm-4:30 pm Sharon G. Berk, USA Long-term survival of novel amoeba-associated bacteria compared with Legionella pneumophila under conditions of hydration and desiccation 4:30 pm-4:50 pm Coffee break
IV 4:50 pm-6:05 pm Session 8: Legionella prevention and control
Chair: Maria Luisa Ricci, Istituto Superiore di Sanita, Rome, Italy
4:50 pm-5:20 pm Katherine Ricketts, Health Protection Agency, London, UK Travel associated Legionnaires' disease in Europe 2008: reporting and response 5:20 pm-5:50 pm Janet Stout, Special pathogens Unit, Pittsburg, USA Legionella and the Conundrum of Low Risk Buildings: A Very Modest Proposal 5:50 pm-6:05 pm John V. Lee, UK An international trial of quantitative PCR for monitoring Legionella in artificial water systems 7:30 pm Gala dinner
Friday, October 16th, 2009
8:30 am-10:15 am Session 9: Epidemiology a) nosocomial infections and Legionella control in hospitals Chair: Paul H. Edelstein, University of Pennsylvania Medical Center, Philadelphia, USA 8:30 am-9:00 am Loreen Herwaldt, University of Iowa Hospital, USA Decontamination of Hospital Water Supplies: A Review of the Literature and the University of Iowa's Experience 9:00 am-9:30 am Jérôme Etienne, University Lyon, France Nosocomial infections, Legionella strain characterization and host-related risk factors 9:30 am-9:45 am Mireia Coscollá, Spain Multiple Legionella strains infection detected by sequence analysis directly from respiratory samples 9:45 am-10:00 am Claressa Lucas, USA The impact of monochloramine introduction on Legionella colonization in a hospital potable water system. 10:00 am-10:15 am Susanne Surman-Lee, UK Lessons learnt from investigating hospital-acquired legionellosis and the remedial actions in England. 10:15 am-10:45 am Coffee break 10:45 am-12:30 pm Session 10: Epidemiology b) outbreaks and population genetics and taxonomy
Chair: Carmen Pelaz, Instituto de Salud Carlos III, Madrid, Spain
10:45 am-11:15 am Antoni Plasencia, Public Health Catalonia, Spain Managing public health crises: from legionellosis to pandemic influenz 11:15 am-11:45 am Rod Ratcliff, Infectious disease laboratories, Adelaide, Australia Ecological diversity within Legionella 11:45 am-12:00 am Sebastian Crespi, Spain Legionellosis Prevention Guidelines: Better, but Not Well
12:00 am-12:15 am Igor S. Tartakovskiy, Russia From large community outbreak in Verhnaya Pyshma to effective prevention of legionellosis in Russia. 12:15 am-12:30 am Joana Costa, Portugal Molecular evolution of dotA gene in Legionella pneumophila: contribution of natural environmental isolates.
V 12:30 am-2:00 pm Lunch at Institut Pasteur 2:00 pm-3:30 pm Poster Session
3:30 pm-4:45 pm Session 11: Genomics and Comparative genomics
Chair: Howard Shuman, Columbia University, New York, USA
3:30 pm-4:00 pm Carmen Buchrieser, Institut Pasteur, Paris, France Legionella pneumophila and Legionella longbeachae pathogenesis: new insights gained form comparative and functional genomics 4:00 pm-4:15 pm Brian Shelton, USA Polymorphic loci in Legionella pneumophila serogroup 1 4:15 pm-4:30 pm Olga Shevchuk, Germany Proteomic analysis of a Legionella-containing phagosome in Dictyostelium 4:30 pm-4:45 pm Ivan Moszer, France Genome re-annotation and web resources for Legionella pneumophila species 4:45 pm-5:15 pm Coffee break
5:15 pm-6:45 pm Session 12: Functional genomics
Chair: Michele Swanson, University of Michigan Medical Center, Ann Arbor, USA
5:15 pm-5:45 pm Elisabeth Hartland, University Melbourne, Australia Hidden protein treasures revealed in the Legionella genome 5:45 pm-6:00 pm Frank Schuren, Netherlands Epidemiological genome analysis of a large Dutch Legionella pneumophila strain collection identifies five markers highly correlated with pathogenic strains 6:00 pm-6:15 pm Tamara O'Connor, USA Solving functional redundancy amongst effectors of the bacterial pathogen, Legionella pneumophila 6:15 pm-6:30 pm Barbara Weissenmayer, Irland Analysis of the transcriptome of Legionella pneumophila under infection conditions using RNA.seq
VI Saturday, October 17th, 2009
9:00 am-10:15 am Session 13: Gene regulation in the environment and the host
Chair: François Vandenesch, National Reference Center for Legionella and INSERM U851, Lyon, France 9:00 am-9:30 am Klaus Heuner, Robert Koch Institut, Berlin, Germany The Flagellar Regulon of Legionella pneumophila and the Expression of Virulence Traits 9:30 am-9:45 am Tobias Sahr, France Two small ncRNAs jointly govern virulence and transmission in Legionella pneumophila 9:45 am-10:00 am Xavier Charpentier, USA Parasexual behavior in response to genotoxic stress in Legionella pneumophila 10:00 am-10:15 am Alcora prize 10:15 am-10:45 am Coffee break
10:45 am-12:30 am Session 14: Cell biology - protozoa - macrophages
Chair: Elizabeth Hartland, University of Melbourne, Melbourne, Australia
10:45 am-11:15 am Craig Roy, Yale University, USA Establishment of a vacuole that supports Legionella pneumophila replication 11:15 am-11:45 am Youssef Abu Kwaik, University of Louisville, Buffalo, USA Exploitation of conserved eukaryotic pathways by L. pneumophila 11:45 am-12:15 am Ralph Isberg, Tufts University, Boston, USA Doubling up as a way to combat resilience Closing remarks 12:15 am-12:30 pm
VII
ORAL COMMUNICATIONS
- 1 -
OPENING SESSION
- 2 - Opening Session Tu. 6:00 pm
Environmental approaches to the prevention of legionellosis
Barry Fields
Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, GA 30333, United States of America [email protected]
Legionellosis represents a considerable challenge for public health. The disease is completely preventable if persons are not exposed to environments which harbor Legionella. This is a difficult objective to achieve because legionellae are so prevalent in freshwater environments. There are some new and promising prevention measures focused on the environments legionellae inhabit such as the complex biofilm communities in which the bacteria reside. Emergent New techniques such as metagenomics and terminal-restriction fragment length polymorphism (T-RFLP) allow extensive characterization of microbial communities that surpasses traditional methods. We examined biofilm samples from environments with and without L. pneumophila using T-RFLP and DNA sequencing. This procedure identified both eubacteria and protists based upon 16S and 18S ribosomal DNA sequence. Identification of microbial communities capable of supporting Legionella will allow the development of strategies to prevent colonization by these bacteria. In addition to better characterization of microbial communities supporting Legionella there is a need to better understand the differences in pathogenecity of the many species and strains of legionellae in these environments. Enhanced molecular characterization of legionellae by sequenced-based typing coupled with established techniques like monoclonal antibody testing will allow us to recognize the more virulent strains and track their distribution. The resulting data suggests that certain strains require more robust prevention measures than others. This may result in the development of screening techniques to allow more focused approaches for certain virulent subpopulations of these bacteria. PCR is one such screening technique for rapid detection of legionellae nucleic acids. However, use of PCR on environmental samples remains controversial. We compared the culture and PCR detection of legionellae in a building water system during a disinfection study. These data reveal both the shortfalls and potential advantages of these procedures. In addition to improved molecular techniques, advances in engineering design and implementation are providing improved physical barriers to prevent disease. Recent data on the use of state-of-the-art drift eliminators on cooling towers show that these devices can greatly reduce the number of aerosolized bacteria. Continued investigation of new approaches for controlling legionellae in the environment should decrease the disease burden of these pathogens.
- 3 - Opening Session Tu. 6:30 pm
Legionnaires’ disease in Europe 1995 - 2008: Trends and Challenges
Carol Joseph
Health Protection Agency Centre for Infections, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
In 1986 the European Working Group for Legionella Infections (EWGLI) was formed to collaborate in the development of scientific knowledge on the epidemiology and microbiology of Legionnaires’ disease. In 1987 EWGLI members established the European Surveillance Scheme for Travel Associated Legionnaires’ Disease (EWGLINET). The scheme has been managed by the co-ordinating centre in London since 1993 and has taken on a major role in developing control and prevention strategies in Europe, supported the growth and development of national and international surveillance and initiated research projects to improve diagnostic capabilities for clinical and environmental investigations. It has also played a major role in developing European management policy through the introduction of European guidelines and the establishment of national legionella policies in EU and non-EU countries. All of these activities underpin and sustain the large body of knowledge and expertise now present in Europe on the control and management of legionella infections. Countries that participate in EWGLINET also produce an annual dataset of all their cases (non-travel and travel) for analysis of epidemiological and microbiological trends within and between European countries. In 1995, 24 countries submitted an annual dataset which amounted to a European total of 1255 cases and an overall European rate of 3.7 cases per million population. For the last seven years over 30 EU countries have reported an annual dataset and around 6000 cases now comprise the dataset in Europe each year, giving a rate of 10-11 cases per million population. A comprehensive review of these data show that ascertainment of cases has increased markedly in some countries, while in others it has not changed at all in the last 14 years. There are wide disparities in legionella incidence rates which range from <1 to >20 cases per million population per country and almost certainly reflect national differences in the amount of clinical, microbiological and legislative resource applied to this disease. There are also marked differences between countries in age standardised rates, cases by category of exposure and the number of outbreaks reported each year. This important European dataset has enabled the effectiveness of national surveillance programmes to be studied and differences and similarities in the epidemiological and microbiological patterns of infection across Europe to be analysed. Many challenges still exist: some countries need to enhance their surveillance activities, others to improve their control and prevention strategies and nearly all to reduce the overall level of under diagnosis of Legionnaires’ disease still apparent today. This keynote address will highlight important aspects of Europe’s national and international epidemiological data from 1995-2008 and the way in which EWGLI’s common values and collective goals have contributed to improvements in the detection, prevention and management of legionella infections. In 2010 responsibility for these surveillance activities will transfer to the European Centre for Disease Prevention and Control (ECDC) in Stockholm, Sweden. The ongoing commitment of EWGLI members to demonstrative effective public health action should ensure the network’s ability to continue to provide added value in Europe.
- 4 - Opening Session Tu. 7:00 pm
Legionella: A Model Organism for Genomics and Cellular Microbiology
Jorg¨ Hackera, Klaus Heunera, Michael Steinertb and Carmen Buchrieserc aRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; bInstitut fur¨ Mikrobiologie, Technische Universitat¨ Braunschweig, Spielmannstr.7, 38106 Braunschweig, Germany; cInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France [email protected]
Legionnaires’ disease is a notifiable disease worldwide. In 2008, in Germany, 522 cases of Legionnaires’ disease with a fatality rate of 7.9% were reported to the Robert Koch Institute (RKI). It is supposed that in the future the cases of Legionnaires’ disease will increase because of climate change and because of the growing group of elderly within the human population. Legionnaires’ disease is, however, not only a serious threat in the public health system, Legionella is also considered as a model organism to study the genome architecture of pathogens as well as the mechanisms of host pathogen interactions. Considering the genomes of Legionella pneumophila strains, it is important to mention that horizontal gene transfer is an essential mechanism for the evolution of these genomes, as it has been shown that the genomes of Legionella carry several genomic and pathogenicity islands. We were able to show that these genetic elements have the capacity of gene transfer and that genomic islands encode factors which contribute to pathogenesis and thus to our understanding of the cellular microbiology of this pathogen. L. pneumophila , however, is also used as a model organism to analyze the molecular mechanism of intracellular replication as well as of metabolism of a pathogen. Furthermore, it became evident recently that specific transcriptome patterns may monitor the replicative phase of Legionella as well as the transmissive (virulent) phase. A specific virulence factor of Legionella, the MIP protein (macrophage infectivity potentiator) plays a crucial role in the interaction of Legionella with different host cells, e.g. Dictyostelium discoideum, Acanthamoeba castellanii or human macrophages. MIP seems to trigger proteolytic processes and is, therefore, a major factor influencing the interaction of bacteria with their respective tissue layers. Taken together, L. pneumophila represents an interesting microbe to study the processes of bacterial genome evolution and host-parasite interaction as well.
- 5 -
SESSION 1 EPIDEMIOLOGY AND CLINICAL ASPECTS A) DISEASE, DIAGNOSIS AND TREATMENT
- 6 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment We. 8:30 am
Clinical aspects of Legionnaires’Disease
Miquel Sabria`
Hospital Germans Trias i Pujol, carretera de canyet, 8916 Badalona, Spain [email protected]
Legionellosis refers to the two clinical syndromes caused by bacteria of the genus Legionella. Pontiac fever is an acute, febrile, self-limited illness that has been serologically linked to Legionella species, whereas Legionnaires’ disease (LD)is the designation for pneumonia caused by these species. The virulence-associated subtype MAb2 cause the majority of cases of Legionnaire’s disease. Legionella-like amebal pathogens are microorganism infecting freshmater amebas than are in all probability members of the genus Legionella based on genomic studies. Serologic studies has implicated this LLAM in occasional cases of community acquired pneumonia (CAP). The most common risk factors for Legionnaires’ disease are cigarette smoking, chronic lung disease, advanced age, and immunosuppression. However, some patients did not have these classic risk factors. In hospitals, patients receiving corticosteroids and transplant recipients have the highest risk. Hospital-acquired cases are now being recognized among neonates and childs with immunosuppression or underlying pulmonary disease. Patients with swallowing disorders may contract legionnaires’ disease when contaminated water is used for drinking, mouth toilette, and nasogatric tube cleaning or enteral feeding (especially in hospitals or in long term care facilities). The symptoms and signs of Legionnaires disease may range from a mild cough and a slight fever to stupor with widespread pulmonary infiltrates and multisystem failure. Currently, extrapulmonary manifestations are rarely observed. In community-acquired LD, differences can be observed according to the sporadic or outbreak presentation. Moreover there are some clinical differences between nosocomial and community cases of legionella pneumonia. Clinical manifestations are more severe and lead to a higher mortality in nosocomial cases; to the contrary extrapulmonary manifestation are more common in community cases. Clinical data may differ in especial population as immunosupressed, HIV and old patients. Although the clinical manifestations often considered classic for LD may suggest the diagnosis, prospective comparative studies have shown that clinical manifestations are generally nonspecific and that Legionnaires’ disease is not readily distinguishable from pneumonia of other etiologies. Clinical manifestations occurring significantly more often in Legionnaires’ disease include diarrhea, confusion, and a temperature of >39◦C. Hyponatremia, and elevated values in liver function tests and CK are also common. Mortality rates for Legionnaires’ disease vary, depending on the patient’s underlying disease and its severity, the patient’s immune status, the severity of pneumonia, and the timing of administration of appropriate antimicrobial therapy. Mortality rates are highest among immunosuppressed patients who do not receive appropriate antimicrobial therapy early in the course of illness. With appropriate and timely antibiotic treatment, mortality from community-acquired Legionnaires’ disease among immunocompetent patients ranges from 0 to 11%.
- 7 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment We. 9:00 am
Antimicrobial therapy for Legionnaires’ disease, still in need of improvement
Paul H. Edelstein
Hospital of the University of Pennsylvania, Dpt of Pathology and Lab. Medicine, 3400 Spruce St., Philadelphia, PA19104, United States of America [email protected]
Antimicrobial chemotherapy for Legionnaires’ disease (LD) requires the use of antimicrobial agents that are active against intracellular Legionella spp., in the proper intracellular compartment. Cell infection studies, as well as animal model treatment studies, show that fluoroquinolones, macrolides, azalides, ketolides, tetracyclines and some other drug classes all have antibacterial activity, whereas β-lactams and aminoglycosides have poor to no activity. Of these agents, the fluoroquinolones and azithromycin have the most potent activity, with the ability to both kill bacteria and limit lung damage. All of the other agents effective in model systems possess mainly inhibitory activity without limitation of lung damage. Thus there are three categories of antimicrobial activity: very active drugs with either cidal or slowly reversible inhibitory activity, solely inhibitory drugs with short or no duration post-antibiotic effect and inactive drugs. Rigorous comparative clinical studies of patient outcome when treated with these different drug classes do not exist, because of the difficulty of conducting randomized trials for a rare disease, the wide range of severity of pneumonia and of underlying illnesses, and varying definitions of treatment effectiveness. Non-randomized studies and case reports clearly show that the inactive drug class is clinically ineffective, and that when used early in mild disease the other two drug classes are effective with no significant differences in survival or most other parameters. For more severely ill patients it is likely that treatment with levofloxacin results in a superior outcome to that obtained with erythromycin or clarithromycin treatment, as judged by hospital length of stay, complications and febrile days, but not survival. The results of the these observational studies show that randomized trials of LD treatment are feasible if parameters other than survival are measured. Whether the clinical outcome of azithromycin treatment is similar to that of the fluoroquinolones for severe disease is unknown. The weight of evidence is that combination therapy with rifampicin has little to no benefit and may be harmful; whether very short duration combination therapy with this drug is beneficial when only less active drugs are available is unclear. Over the last 20 years the overall fatality rate of LD has declined remarkably, from around 30% to about 12%, and with prompt active therapy of non-immunosuppressed patients the treated fatality rates are now around 1 to 5%. The decrease in fatality rates is due in part to prompt therapy and to policies of administration of LD-active drugs to patients with community-acquired pneumonia. However fatality rates remain high, around 30%, in some patients, especially in those who present with respiratory failure. In addition, long term morbidity remains a problem in some patients with severe disease. Challenges for the future are to construct multicenter randomized clinical trials, to determine if azithromycin has benefit over levofloxacin therapy, to develop educational programs to promote early treatment of LD and other forms of pneumonia, to study methods to reduce mortality and lung damage in patients with respiratory failure, and to determine the true frequency of post-pneumonia morbidity.
- 8 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment We. 9:30 am
Legionella pneumonia (LP) in immunosuppressed (IS) patients.
Maria Luisa Pedro-Botet, Raquel Nunez,˜ Nieves Sopena, Lourdes Mateu, Irma Casas, Neus Robert, Marian Garc´ıa- Nunez,˜ Celestino Rey-Joly and Miquel Sabria`
Hospital Germans Trias i Pujol, carretera de canyet, 8916 Badalona, Spain [email protected]
Background: Immunosuppression is considered a bad prognostic factor of LP. Nevertheless, a shorter delay in the initiation of appropriate treatment, the use of more effective drugs and probably a lower rate of hospital-acquisition may have improved the prognosis of this infection. The aim of this study was to describe the clinical manifestations of IS patients with LP and compare the outcome with non-immunosuppressed (NIS) patients. Method: We performed an observational prospective study with review of the LP database registered from 1983 to May 2009. Diagnosis was made by urinary antigen test (UAT), respiratory specimen culture, serology or sputum direct fluorescent antibody (DFA). Immunosuppression was defined as cancer, HIV infection and/or immunosuppressive treatment. Results: 157 IS and 302 NIS patients were included. Regarding IS patients, 99 had positive UAT, 30 seroconverted, 32 Legionella culture, 7 positive DFA and 3 positive necropsy. 80.9% were male with a mean age of 59.74±16.6 yrs. 55.1% had cancer (solid 41.6%, hematolologic 16.7%), 60% were on immunosuppressive therapy (corticoids 51.4%, chemotherapy 27.6%, other drugs 10.6%) and 12.7% were HIV-infected. 44.2% were Fine IV or V. Hospital and healthcare-acquisition were significantly more frequent in IS patients (p<0.001). IS patients were appropriately treated more promptly (p=.02) but no differences were observed in the use of macrolides/quinolones. Clinical complications and respiratory failure were significantly more frequent (p=.04 both) and septic shock and radiological extension tended to be more frequent in IS patients. Mortality was significantly higher (24.8 vs. 6%, P=.000) in IS patients. Conclusions: At present the morbimortality of IS patients with LP remains significantly higher than in NIS patients. Prompt diagnosis and initiation of adequate treatment are warranted. UAT must be done in all patients with pneumonia.
- 9 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment We. 9:45 am
Changing Epidemiologic Trends of Legionellosis in New Zealand, 1979 - 2008
Frances Grahama, Simon Kinghamb, Paul Whitec and David Harted aUniversity of Canterbury/Ministry of Health, P O Box 5013, 6011 Wellington, New Zealand; bDepartment of Geography, University of Canterbury, Private Bag 4800, 8041 Christchurch, New Zealand; cMinistry of Agriculture and Forestry, P O Box 2526, 6011 Wellington, New Zealand; dInstitute of Environmental Science & Research Ltd, Kenepuru Drive, 5022 Porirua, New Zealand frances [email protected]
Background Legionellosis, an infection caused by the bacterium Legionella, is a common cause of community-acquired pneumonia worldwide. In New Zealand the diagnostic testing and reporting of legionellosis has been conducted through a centralised laboratory since 1979 (the disease became notifiable in 1980). This represents an important repository of population- based data on the epidemiology of legionellosis which to date has never been comprehensively documented. The aim of this paper is to carry out a spatio-temporal analysis of Legionella in New Zealand which will add to our understanding of the epidemiologic trends of legionellosis. To achieve this we review legionellosis in New Zealand over the past three decades, focusing on changing incidence rates and occurrence of different species over this time-period. We also examine whether demographic characteristics such as ethnicity affects incidence. Methods We analysed cases using disease notification and laboratory surveillance data, 1979 to 2008. Poisson regression models were used incorporating demographic covariates. To account for the possibility of some cases being identified using different available testing methodologies, logistic regression was used to evaluate temporal changes. Statistical analyses were performed using SPSS (version 15) and thematic maps were created using ArcGIS version 9.3. Results Since 1979 when the first case of legionellosis was diagnosed in New Zealand, there have been 2695 laboratory-reported cases (annual incidence rate 2.6 cases per 100,000 population) as opposed to 1382 cases (annual incidence rate 1.5 cases per 100,000 population) which have been notified since 1980. Incidence was highest amongst European population and showed large geographic variations between district health boards responsible for the provision of health and disability services in twenty-one regions in New Zealand. The incidence of legionellosis in New Zealand remains high compared to other temperate developed countries with cases occurring both sporadically and in outbreaks. We found that L. pneumophila serogroup 1 was not the predominant isolate from clinical specimens and environmental isolates compared to other developed countries. Risk factors such as travel were not significant (6.6%) while environmental sources of infection accounted for 65% of cases. Conclusion Despite New Zealand showing some similarities with the legionellosis epidemiology of other temperate countries such as an increasing incidence, results derived predominantly from serological investigations suggest that the distribution of the Legionella species may not necessarily follow patterns observed in other parts of the world where the predominant species responsible for illness is L. pneumophila serogroup 1. This suggests that there is a distinctive legionellosis epidemiology for New Zealand.
- 10 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment We. 10:00 am
Travel-Associated Legionnaires’ Disease Surveillance in the United States, 2005-2008
Nicole Alexander and Lauri Hicks
Centers for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop C-23, Atlanta, GA 30333, United States of America [email protected]
Background: Timely reporting of travel-associated Legionnaires’ disease (TALD) cases with complete travel information can allow for early identification and prevention of additional cases. In 2005, the Centers for Disease Control and Prevention (CDC) initiated enhanced surveillance for TALD. CDC requests that state and local public health departments report TALD cases within 7 days of notification and include case travel dates and accommodation details. For each TALD case, the national legionellosis surveillance database is queried by a surveillance epidemiologist to determine if there are additional cases reporting overnight travel to the reported accommodation. Notifications describing the case are sent to the appropriate public health or industry representatives. For each cluster notification, CDC recommends the public health authority check for additional cases, notify the accommodation, and conduct an environmental assessment of the facility. We conducted a study to evaluate enhanced TALD surveillance efforts and to describe the epidemiology of TALD cases in the United States. Methods: We analyzed U.S. surveillance data for Legionnaires’ disease (LD) cases with symptom onset from 2005 to 2008. An LD case was defined as a U.S. resident fulfilling both clinical and laboratory diagnostic criteria (pneumonia or respiratory syndrome and a positive Legionella urinary antigen, culture or seroconversion). A TALD case was defined as LD in a patient with overnight travel in the two weeks before symptom onset. A travel-associated cluster was defined as two or more cases traveling to the same accommodation ≤ one year apart. Results: Of the 3550 LD cases reported in the time period, 904 (25%) were travel-associated; 144 in 2005, 222 in 2006, 271 in 2007 and 267 in 2008, resulting in an 85% increase in TALD reporting from 2005 to 2008. The majority of these cases were confirmed (868, 96%), had a Legionella urinary antigen laboratory diagnosis (844, 94%) and reported travel within the U.S. (719, 80%). The average case age was 60.3 years (range 1-97); 570 (69%) of 823 cases were male and 698 (94%) of 740 cases survived. Cases were residents of 44 states and reported travel to 49 states, 2 U.S. territories and 47 countries. Four hundred ninety- eight (55%) of 904 TALD cases reported at least one overnight stay at a commercial accommodation and 56 (6%) reported travel aboard a cruise ship. Four hundred twenty-two case notifications were sent to domestic and international public health partners or cruise ship management. Forty travel-associated clusters, in 15 states, 4 countries and on 7 cruise ships, were reported directly to CDC (n=21) or detected through surveillance (n=19); 2 in 2005, 9 in 2006, 16 in 2007 and 13 in 2008. Conclusions: Enhanced TALD surveillance efforts have contributed to several improvements, including an increase in TALD reporting and cluster detection, an overall increase in reports made to the national supplemental legionellosis reporting system and better communication among public health partners.
- 11 -
SESSION 2 CLINICAL ASPECTS B) DETECTION AND SUBTYPING
- 12 - Session 2: Clinical aspects a) detection and subtyping We. 10:45 am
Microbiological diagnostic methods for the detection of Legionella infections
Christian Luck¨
Institute of Medical Microbiology and Hygiene, German Reference Laboraotry for Legionella, Fiedlerstrasse 42, D-01307 Dresden, Germany [email protected]
Legionella species are responsible for 2 to 5% of nosocomial or community acquired pneumonia occurring both sporadically and in outbreaks. The majority of infections are caused by strains belonging to Legionella pneumophila serogroup 1. The diagnostic methods currently available, such as culture, serology, direct fluorescent antibody testing and urinary antigen detection, still constitute the basic diagnostic repertoire. However, none of the diagnostic tests presently available offers the desired quality with respect to sensitivity and specificity. Therefore, the standard performance is to use several diagnostic tests in parallel. The culture method which is 100% specific has a moderate sensitivity ranging from 30 to 80%. Despite this it should be obligatory especially when hospitalized patients with underlying diseases are investigated. Isolated L. pneumophila strains are typed serologically. For the identification of non-pneumophila species sequencing the mip gene is the method of choice. L. pneumophila usually grows within 3 to 4 days. Species other than L. pneumophila may grow at a slower rate. A positive culture is warranted as the prerequisite for epidemiological investigations. Urinary antigen detection is a valuable tool when L. pneumophila serogroup 1 is the causative agent. This is the case in the majority of community acquired cases. The sensitivities of various enzyme immunoassays varied. In general, sensitivity ranged from 90 to 95% for travel-associated infection, 75 to 85% for community-acquired infection and less than 50% for nosocomially acquired infection. Currently, several immunochromatographic assays are available to detect infections caused by L. pneumophila serogroup 1. Several studies showed possibly false positive results with some tests. Thus, caution is required when using these assays. For the surveillance of nosocomial legionellosis direct fluorescence antigen detection using a commercially available specific monoclonal antibody may be of greater value since this test detects all serogroups of L. pneumophila. Major disadvantages are reduced sensitivity, the time spent on the performance and the experience needed. Recently, significant advances have been made in the development of diagnostic tests which are based on the detection of nucleic acids. Real time PCR assays that detect all Legionella species are of great benefit in elucidating the role of non-pneumophila species as causative agents. The sensitivity in respiratory samples is considered high. Variable results were found when urine or acute phase serum samples are used for DNA amplification. The detection of antibodies in patient’s sera is still a valuable laboratory test in epidemiological investigations but is of little use in the acute phase of illness.
- 13 - Session 2: Clinical aspects a) detection and subtyping We. 11:15 am
Legionella pneumophila DNA sequence-based typing: progress, pitfalls and perspectives.
Timothy Harrison
Health Protection Agency Centre for Infections, Respiratory and Systemic Infection Laboratory, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
Since 2003 members of the European Working Group on Legionella infections (EWGLI) have been working to develop a standardised method for typing Legionella pneumophila utilising DNA sequence data. Initially a sequence-based typing (SBT) scheme, based on the comparison of DNA sequences of parts of six genes (flaA, pilE, asd, mip, mompS and proA), was adopted and preliminary work presented at the 6th International Conference in Chicago in 2005 demonstrated the great potential of this method. In the last four years the major developments have been the inclusion of a seventh gene, neuA, into the scheme, the formal designation of sequence types (STs) based on the seven gene allelic profile, improvements to the web-enabled database, development of real-time mapping to aid interpretation of the increasingly large dataset and, most significantly, the widespread utilization of the method by the workers across the globe. A key aim of the project was to ensure that data obtained in one laboratory could be reliably compared with data obtained elsewhere. This has been ensured by the rigorous standardisation of the method, use of an online sequence-quality tool, scrutiny of putative new alleles by the curators, and the provision of an external quality assurance programme. The SBT method was primarily developed as a typing tool and it is now widely used for this purpose in both national and international investigations. Recently published data have confirmed its utility for the investigation of community outbreaks in Canada, Russia, USA and the UK, for the investigation of nosocomial cases and for the investigation of travel-associated legionellosis. Furthermore the availability of large quantities of high quality, and directly comparable, data is for the first time allowing us to build up a global picture of strain distribution and significance. The submission of all SBT data to the EWGLI database has been actively encouraged and as of the 22nd July 2009, data for 3200 isolates had been submitted by colleagues in 19 countries comprising isolates from 36 countries. Of these, 3005 have a full seven-allele profiles resulting in the designation of 700 distinct ST. Among the 2683 ’unrelated’ isolates ST1 is the most frequent (12.7%) being recovered from 18 countries. Other common STs include ST47, ST23, ST42, ST62 and ST37: > 50% of isolates are accounted for by just 14 STs. Although many ST are represented in the database by only a few isolates, these were often obtained from geographically remote locations suggesting that many strains will subsequently prove to be common across the world (e.g. the three isolates of ST278 were isolated in Japan, USA and UK). Comparison of data for the 1869 (62%) clinical isolates and 1136 environmental isolates confirm and extend recently published data. There is a marked mismatch between the distributions of STs among clinical and environmental isolates. ST1 accounts for 21.4% of environmental isolates but only 7.4% of clinical isolates. Conversely ST47 and ST23 account for 13.6% and 10.7% of clinical isolates but 0.4% and 1.5%, respectively, of isolates recovered from the environment.
- 14 - Session 2: Clinical aspects a) detection and subtyping We. 11:45 am
Evaluation of a fully integrated point-of-testing system for the detection of viable Legionella pneumophila
Patrick Yanga, Barry Fieldsb, Tatiana Travisb, Randy Johnsonc, Richard Leec, Edgar Kamantiguec, Sara Fanc, Scott Boyetted, Rong Xue, Jing Chene, Weiqing Xue, Weimin Xiaoe, Kechao Yange and Jie Lie aCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, AK 30333, United States of America; bCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, GA 30333, United States of America; c2Gen-Probe Incorporated, 10210 Genetic Center Dr, San Diego, AK 92121, United States of America; d4GE Water and Process Technologies, 4636 Somerton Rd, Feastervl Trvs, AK 19053, United States of America; e4GE Water and Process Technologies, 3W002, 1800 Cailun Road, Zhangjiang Hightech Park Pudong, 201203 Shanghai, China [email protected]
Legionella pneumophila (LP) can adapt to stressful environmental conditions, such as high temperature or chemical treatment, by entering into a viable but non-cultivable state (VBNC). Although PCR recognizes VBNC cells, the lack of differentiation between DNA from live and dead LP organisms leads to false positive results in the absence of viable LP organisms. In this study, a new point-of-testing platform that runs a complete sample-to-result Real Time Transcription- Mediated Amplification (TMA) assay was evaluated to assess its ability to detect viable LP organisms. When challenged with purified LP RNA and DNA, the TMA assay system preferentially amplifies RNA (present only in viable cells) over DNA. LP harvested from BYE broth at post-exponential phase was subjected to heat (70◦C for 20 minutes) or bleach treatment (3 ppm hypochlorite NaClO for 12 hours). The TMA assay’s LP detection rate reduced by 4- or 5-fold post heat or bleach treatment, respectively, compared to the PCR assay, which exhibited no reduction in the LP detection rate. The study was expanded to include environmental samples (swabs and bulk water, n=111). Samples were collected pre- treatment (n=55) and post-treatment (n=56) in a hospital known to be colonized with legionellae and were evaluated with the TMA assay and culture. Compared to culture, the sensitivity and specificity of the TMA assays were 100% (29/29) and 82.9% (68/82), respectively. Among the LP culture-negative samples (n=82), the TMA assay detected LP RNA in 14 samples. The viability of LP bacteria was confirmed by amoebae co-culture in 3 samples. In summary, the TMA assay run on the point-of-use platform displayed high sensitivity and provided a more accurate estimate of viable LP organisms.
- 15 - Session 2: Clinical aspects a) detection and subtyping We. 12:00 am
MALDI-TOF MS based protein mass fingerprinting: a rapid and reliable method for the identification of Legionella spp.
Valeria Gaia and Mauro Tonolla
Istituto cantonale di microbiologia, via Mirasole 22A, 6500 Bellinzona, Switzerland [email protected]
MALDI-TOF MS (matrix assisted laser desorption ionization - time of flight mass spectrometry) is becoming a popular tool for the identification and classification of bacterial and fungal species since commercial database applications are available for diagnostic and research purposes. The identification of Legionella spp. was formerly based on serological tests and is now mainly performed by sequencing of the mip gene (macrophage infectivity potentiator). The aim of this study was to use intact cell MALDI-TOF MS (ICMS) for the identification of Legionella spp. For this purpose, SuperSpectraTM were created by choosing specific biomarkers shared by different strains of the same Legionella species isolated from different environments. A panel of 506 Legionella strains comprising 36 reference strains and 470 other isolates representing 38 different species were analyzed by ICMS in combination with the SARAMIS (spectral archive and microbial identification system) application from Anagnostec. Choosing appropriate sets of biomarkers it was possible to create specific SuperSpectraTM that recognize the more frequently isolated Legionella species. These SuperSpectraTM were tested for their ability to identify Legionella strains isolated from water samples, cooling towers, potting soils and patients’ specimen collected by the National Reference Center and previously identified by standard methods (mip-sequencing). With these SuperSpectraTM it is possible to correctly identify strains belonging to L. pneumophila, L. bozemanii, L. anisa, L. londiniensis, L. taurinensis, L. micdadei, L. sainthelensi and L. longbeachae. The identification of Legionella spp. by MALDI-TOF is very quick and easy and also has the advantages of being time- and cost-saving as compared to sequence-based identification. This method allows screening and identification of a large number of colonies within a few minutes.
- 16 - Session 2: Clinical aspects a) detection and subtyping We. 12:15 am
An improved and rapid sequence-based method for typing of Legionella pneumophila.
Johan Lindha, Jonas Lundstrom¨ b and Gorel¨ Allestamc aSwedish Institute for Infectious Disease Control, Sweden (SMI), Dep Parasitology, Mycology, Water and Environmental Microbiology (PMV, SE-171 82 Solna, Sweden; bSwedish Institute for Infectious Disease Control (SMI), Dep Parasitology, Mycology, Water and Environmental Microbiology (PMV, SE-171 82 Solna, Sweden; cSwedish Institute for Infectious Disease Control (SMI), Dep Parasitology, Mycology, Water and Environmental Microbiology (PMV), SE-171 82 Solna, Sweden [email protected]
In July 2009 approximately 700 different types of Legionella pneumophila could be found at the European Working Group for Legionella Infections (EWGLI) homepage (http://www.hpa-bioinformatics.org.uk/legionella/legionella sbt/php/sbt homepage.php). The homepage works as a valuable tool for epidemiological typing of clinical and environmental isolates. It is important that the protocol of the method used is safe, quick and affordable. According to Sequence-Based Typing (SBT) protocol for Legionella pneumophila, Version 4.1, genomic DNA is extracted and amplified using different primer sets, each specific for a specific gene (flaA, pilE, asd, mip, mompS, proA, neuA). After purification, the amplicons are sequenced directly with forward and reverse primers and sequences compared to known sequences in the EWGLI database. Although the protocol is easy to follow and give satisfactory results, there is today with technical development possibilities for improvements. We have modified the above method and thus been able to shorten the time of procedure from 1-2 workingdays to 2h and made it safer and cheaper. By the use of M13-flanked specific primers, Ampli Taq gold fast enzyme and BigDyeTerminator v1.1 we have developed an initial PCR which is ready within 25 minutes and sequencing reactions within 30 minutes. Purification steps are performed by dilutions and ethanol precipitation. Since the new sequencing primers are M13 reverse and forward sequences the same stock solutions can be used in all sequencing reactions. Hence, less chan-ce of contamination and longer sequences will be obtained. Initially, we got 100% correlation with the NEW RAPID SBT-method of strain 32-36 in the last Legionella SBT Quality Assessment. Furthermore, a similar approach has been used in order to identify Legionella species with DNA-sequencing of the mip gene. Data and comparisons between the two different SBT-protocols will be discussed.
- 17 -
SESSION 3 HOST - MICROBE INTERACTIONS A) SECRETION SYSTEMS AND THEIR SUBSTRATES
- 18 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates We. 3:30 pm
Type II protein secretion, siderophores, and pigment and what they mean for Legionella ecology and pathogenesis
Nicholas Cianciotto
Northwestern University, 320 E Superior St, Searle 6-541, Chicago, IL 60611, United States of America [email protected]
The secretion of both protein and non-protein substrates is critical to the ecology and pathogenesis of Legionella pneumophila (Lp). Type II protein secretion (T2S) is a multi-step process. Proteins destined for secretion first translocate across the inner membrane via the Sec or Tat pathway. On delivery into the periplasm, unfolded proteins then assume their tertiary conformation. Finally, the proteins are translocated across the outer membrane by a multi-protein complex that is specifically dedicated to T2S. In effect, a ”pseudopilus”, using energy generated at the inner membrane, acts like a piston to push the proteins through an outer membrane pore (secretin). More than 25 proteins substrates are secreted via Lp T2S, including a wide variety of degradative enzymes, proteins that have striking similarity to eukaryotic proteins, and proteins that are predicted to encode novel activities. Based upon mutant analysis, T2S is critical for Lp extracellular survival in water at low-temperatures, surface translocation (sliding motility), intracellular growth within various amoebae, intracellular infection of macrophages and lung epithelial cells, and infection of the mammalian lung. Among the key effectors are a protease and ribonuclease that promote infection of amoeba hosts and a chitinase that appears to facilitate persistence in the lung. Lp also secretes a non-protein, high-affinity, ferric iron-chelating siderophore known as legiobactin. Purified legiobactin stimulates the growth of Lp under low-iron conditions and displays the characteristics of a carboxylate siderophore. Legiobactin mutants are defective for growth in the lungs of infected mice, indicating that siderophore secretion is important for Lp pathogenesis. Finally, Lp secretes a pyomelanin pigment that, among other things, catalyzes ferric iron reduction and thereby may facilitate the import of ferrous iron, serving a function that is complementary to legiobactin-mediated ferric iron acquisition. Interestingly, both the T2S mutants and legiobactin mutants show a much greater growth defect during in vivo lung infection than during in vitro infection of macrophages or epithelia, providing evidence that bacterial phenotypes in addition to intracellular growth are critical for pathogenesis and that secreted factors are in part mediating those processes.
- 19 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates We. 3:55 pm
The Legionella Dot/Icm type IVB secretion system.
Joseph Vogel
Washington University School of Medicine, 660 S. Euclid Avenue, Campus Box 8230, St. Louis, MO 63110, United States of America [email protected]
The bacterial pathogen Legionella pneumophila is able to survive and replicate within phagocytic cells by inhibiting the normal interaction of the bacteria-containing vacuole with the endocytic pathway, thereby preventing phagosome- lysosome fusion. Instead, L. pneumophila creates a novel compartment, the ”replicative phagosome”, where it is able to multiply to high numbers prior to lysis of the host cell. Similar to many intracellular pathogens, L. pneumophila utilizes several specialized secretion systems as part of its virulence strategy. These include a type II secretion system (T2SS) and several type IV secretion systems (T4SSs), but not a type III or a type VI secretion system. The T2SS exports a variety of enzymes necessary for optimal growth in bacteriological media and within host cells. Most L. pneumophila strains also encode several type IV secretion systems, including a type IVB Dot/Icm secretion system that functions to mediate intracellular survival and replication of this bacterium. Inactivation of the Dot/Icm T4SS results in mistargeting of the Legionella-containing vacuole (LCV) into the endocytic pathway of the host cell, thus thwarting intracellular growth of this pathogen. The Dot/Icm system is used to secrete a large number of substrates, including toxins, into the host cell. Export of Dot/Icm substrates typically requires the presence of a c-terminal signal sequence, although at least one substrate contains a second, internal signal-sequence. In addition, some substrates require the chaperone-like molecules IcmS/IcmW, which have been classified as type IV adaptor proteins, for their proper export. Due to the large number of T4SS substrates exported by Legionella, inactivation of individual substrates often confers no apparent intracellular growth defect presumably due to functional redundancy. The Dot/Icm secretion apparatus is made up of at least twenty-six proteins, including several cytoplasmic proteins and a large number of inner and outer membrane proteins. The complex is localized specifically to the bacterial poles where it exports substrates in a spatial-temporal manner. The Dot/Icm secretion machine is comprised of two major subcomplexes: the ”core subcomplex” and the ”T4CP subcomplex”. The core subcomplex spans the bacterial inner and outer membranes and consists of the lipoproteins DotC and DotD, the outer membrane protein DotH/IcmK, and the inner membrane proteins DotF/IcmG and DotG/IcmE. The T4CP subcomplex consists of the inner membrane proteins DotL/IcmM, DotM/IcmP, and DotN/IcmJ. DotL, the putative type IV coupling protein (T4CP), has been proposed to play a key role as the major inner membrane receptor for Dot/Icm substrates. Interestingly we also detected an interaction between DotL and the type IV adaptor proteins IcmS and IcmW via an IcmS/IcmW-binding domain in DotL. We are currently exploring the molecular role of DotL, the type IV adaptors and polar localization of the T4SS in substrate export by the Legionella Dot/Icm secretion system.
- 20 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates We. 4:20 pm
Legionella pneumophila phospholipases and their mode of secretion
Antje Flieger
Robert Koch Institut, Burgstrasse 37, 38855 Wernigerode, Germany fl[email protected]
Legionella pneumophila expresses secreted and cell-associated phospholipase A (PLA) and lysophospholipase A (LPLA) activities belonging to at least three enzyme families. The first family, the GDSL hydrolases, consists of three enzymes; PlaA, PlaC, and PlaD; with LPLA and PLA activities displaying the amino acid signature motif ’GDSL’. PlaA represents the major secreted LPLA and PlaC in addition to secreted LPLA/PLA activities shows cholesterol acyltransferase activity. Both enzymes possess a signal peptide and are exported via the Lsp type II secretion pathway. Whether PlaD is exported out of the bacterial cell has not completely been clarified. The second group, designated PlaB-like enzymes, contains the cell-associated and very potent PLA/LPLA, PlaB, which is the first characterized member of a novel class of bacterial phospholipases. Since L. pneumophila plaB knock out mutants show reduced haemolytic capacity, we conclude that the protein is acting from the bacterial cell surface; however its mode of export still remains elusive. The third group, the patatin-like proteins (PLP), comprises eleven members (VipD, PatB, VpdA, PatD, PatE, VpdC, VpdB, PatH to PatK). VipD shows LPLA and PLA activities and interferes with vesicular trafficking when expressed in yeast and therefore is possibly involved in the intracellular infection process. At least three L. pneumophila PLP (VipD, VpdA and VpdB) are injected into the host cell via the type IVB secretion system Dot/Icm. Others like PatD are rather acting from the inside of the bacterium to support bacterial establishment by contributing to usage of intracellular lipid storage compounds. In summary, L. pneumophila possesses at least 15 different phospholipase enzymes which show overlapping enzymatic activities and act both inside and outside of the bacterium. In case of the latter, Legionella uses a variety of export modes for proper placement of the effectors.
- 21 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates We. 4:45 pm
Inhibition of Host Vacuolar H+-ATPase Activity by a Legionella pneumophila effector
Li Xua, Xihui Shena, Andrew Bryanb, Simran Bangac, Michele Swansonb and Zhao-Qing Luoa aDept. of Biological Sciences, Purdue University, 915 West State Street, West Lafayette, IN 47907, United States of America; bUniversity of Michigan Medical School, Department of Microbiology & Immunology, Ann Arbor, MI 48109, United States of America; cDept. of Biological Sciences, Purdue University, 915 West State Street, West Lafayette, AK 47907, United States of America [email protected]
One hallmark of lysosome is its low luminal pH that is important for the activities of many hydrolyzing enzymes responsible for efficient digestion of phagocytosized contents. To survive and replicate in phagocytes, successful intracellular pathogens have evolved various mechanisms to circumvent the challenge posed by lysosomal killing. These include the ability to thrive in an acidic environment, active escape to the cytosol or strategies for maintaining a neutral pH in their phagosomes. One salient feature associated with infection of the intracellular bacterial pathogen Legionella pneumophila is the maintenance of a neutral pH of the Legionella containing vacuoles (LCVs) that supports its intracellular growth, and nonpathogenic mutants often are concomitantly trafficked to an acidic compartment. In eukaryotic cells, organelle acidification is mediated by the vacuolar H+-ATPase that translocates proton into target compartments in a process energized by ATP hydrolysis. The recent discovery of the association of v-ATPase with LCVs points to the necessity for active modulation of v-ATPase activity by the bacterium. By screening L. pneumophila proteins that cause a yeast phenotype similar to its v-ATPase mutants, we have identified a substrate of the L. pneumophila Dot/Icm type IV secretion system that specifically inhibits the activity of the proton transporter. This protein, termed SidK inhibits v-ATPase-mediated ATP hydrolysis by directly interacting with its VatA subunit, which is responsible for hydrolyzing ATP. We also found that expression of sidK is highly induced right after stationary bacteria were diluted into fresh medium, suggesting that SidK plays a more important role in the early phase of infection. Our results reveal a mechanism used by an intravacuolar pathogen to control the luminal pH without active removal of v-ATPase from its vacuolar membranes.
- 22 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates We. 5:00 pm
Legionella pneumophila infections involve bacterial F-box proteins
Mariella Lommaa, Delphine Dervins-Ravaulta, Hayley Newtonb, Fiona Samsomb, Matthieu Julesa, Tobias Sahra, Elizabeth Hartlandb and Carmen Buchriesera aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bUniversity of Melbourne, Department of Microbiology and Immunology, 3010 Victoria, Australia [email protected]
The human pathogen Legionella pneumophila encodes three proteins (Lpp2082, Lpp2486, Lpp0233) containing each an F-box domain and additional protein-protein interaction domains, reminiscent of eukaryotic F-box proteins known to constitute the SCF-type E3 ubiquitin ligase. Ubiquitination is a fundamental cellular process exploited by pathogens to promote their survival and replication in host cells. Here we show that the three F-box proteins of L. pneumophila are Dot/Icm secreted effectors and we provide evidence for their implication in the ubiquitination of proteins associated with the Legionella- containing vacuole. Single, double and triple mutants deficient in the different F-box protein encoding genes are impaired in the efficiency of infection in a wide range of in vitro like Acanthamoeba castellanii, THP-1 human monocytes derived macrophages and lung epithelial A549 cells, as well as in vivo infection like lung infection of A/J mice. Furthermore, a Yeast Two Hybrid screen identified two possible targets for one of the F-box proteins (Lpp2082). Further investigations to validate these targets in in vivo assays, such as pull-down and co-immunoprecipation, are under way to determine at which stage of the infection they may act and whether they are degraded by the proteasome following ubiquitination.
- 23 -
SESSION 4 HOST - MICROBE INTERACTIONS B) METABOLISM AND VIRULENCE
- 24 - Session 4: Host - Microbe interactions: b) metabolism and virulence We. 5:45 pm
The Legionella pneumophila Developmental Cycle: Role in Survival and Pathogenesis
Paul S. Hoffman
UNIVERSITY OF VIRGINIA HEALTH SYSTEMS, Division of Infectious Diseases and International Health, 409 Lane Road, Charlottesville, VA22901, United States of America [email protected]
In natural environments, Legionella pneumophila is essentially an obligate intracellular parasite, which differentiates late in infection into resilient, hyper-infectious cyst-like forms that can persist in biofilms in a semi-dormant state for years. In contrast, in laboratory culture, differentiation aborts in stationary phase, although suspension of SP bacteria in water enables further differentiation to a resilient intermediate survival form. In ciliates, vegetative L. pneumophila escape digestion and rapidly differentiate into cysts that are excreted within fecal pellets. Finally in the nematode C. elegans, L. pneumophila both replicates and differentiates into cysts within the gut and does not invade tissue. All of these examples indicate that L. pneumophila is extremely successful at persisting in natural environments from which they can be transmitted to humans. Dimorphic lifestyles are common to many obligate intracellular pathogens, where division of tasks (replication and virulence) is relegated to distinct morphological forms and include various fail safe mechanisms to maintain fidelity. Here we explore the two forms of L. pneumophila and examine the role of nutrition as a barrier to extracellular growth and inappropriate germination of cysts. If indeed dormant cysts are the most infectious form of L. pneumophila, they must be preloaded with effectors and the Dot/Icm Type IV delivery system must be activated upon contact with a suitable host cell. Moreover, our studies suggest that the major outer membrane protein complex, through inter-chain disulfide bonding, might contribute to the structure of the developing cyst and to proper function of the Dot/Icm secretion system.
- 25 - Session 4: Host - Microbe interactions: b) metabolism and virulence We. 6:15 pm
Legionella pneumophila virulence as a response to starvation
Michele Swanson
University of Michigan Medical School, Department of Microbiology & Immunology, Ann Arbor, MI 48109, United States of America [email protected]
Legionella pneumophila is commonly found in fresh water supplies, both natural and man-made. When ingested by free-living amoebae or protozoa, the virulent bacteria avoid digestion and grow inside the host cells. After a period of replication, motile bacteria escape from the host cell, and the cycle is repeated. When humans inhale water that contains L. pneumophila, the bacteria can also survive and replicate within lung macrophages. Using genetic, molecular, and microscopic approaches, our laboratory has learned that in macrophage cultures L. pneumophila alternates between two distinct cell types. A resilient, infectious, motile form is equipped for transmission from one host cell to another, whereas a sensitive, non-motile cell type grows in host cell vacuoles. Furthermore, the L. pneumophila life cycle is controlled by metabolic cues. When nutrients are plentiful within the host cell, transmissive bacteria switch to the replicative form and grow to large numbers. Conversely, when amino acids and fatty acids become scarce in the host cell, L. pneumophila stops replicating and expresses its panel of virulence traits to promote transmission. Components of the machinery that equips L. pneumophila to alter its life style in response to metabolic cues will be discussed, including the PhtC and PhtD nucleoside transporters, the RelA and SpoT stringent response enzymes, and the DksA transcriptional regulatory protein.
- 26 - Session 4: Host - Microbe interactions: b) metabolism and virulence We. 6:45 pm
Cytotoxic glucosyltransferases of Legionella pneumophila.
Yury Belyia, Michael Stahlb and Klaus Aktoriesb aGamaleya Research Institute, Ulitsa Gamalei 18, 123098 Moscow, Russian Federation; bAlbert-Ludwigs University, Albertstrasse 25, 79104 Freiburg, Germany y [email protected]
Legionella pneumophila produces three proteins - Lgt1, Lgt2 and Lgt3, which modify serine-53 in eukaryotic elongation factor eEF1A by a retaining glucosylation reaction. Lgt1 and Lgt3 are intrinsic to all tested L. pneumophila strains, while Lgt2 can be found only in a portion of isolates. As shown by western blot, autoradiography and quantitative PCR after reverse transcription Lgt1 is produced during stationary phase of Legionella growth in liquid medium and late during bacterial infection of Acanthamoeba castellanii. In contrast to these data, Lgt3 is elaborated in pre-logarithmic phase of cultivation in broth and early in bacterium-amoeba interaction. These results demonstrate differential regulation of lgt expression in Legionella and suggest specific roles of each enzyme in bacterial virulence. Synthetic decapeptide 50-GKGSFKYAWV-59 of eEF1A is efficiently modified by Lgt1. Glucosylation rate of this peptide is higher than that of full length eEF1A, suggesting that a specific conformation of the elongation factor is the preferred substrate of the enzyme. Gene bank search on the basis of the decapeptide for similar eukaryotic protein sequences retrieved Hsp70 subfamily B suppressor 1 (Hbs1) as another target for glucosylation by Lgt1. Indeed, recombinant Hbs1 and the corresponding recombinant fragment (303-GKASFAYAWV-312) were glucosylated by Lgt1. Interestingly, full length Hbs1 was modified more efficiently than full length eEF1A. Enzymatic activity of the glucosyltransferases leads to inhibition of protein synthesis assayed by in vitro transcription/translation and in vivo methionine incorporation and outcomes in death of target eukaryotic cell. It remains to be studied whether Lgt-catalyzed glucosylation of eEF1A and Hbs1, which both are important components of translational machinery of a cell, not only results in additive effects on inhibition of protein synthesis but is also important for proliferation of Legionella in host cells.
- 27 -
SESSION 5 IMMUNOLOGY AND HOST SUSCEPTIBILITY
- 28 - Session 5: Immunology and host susceptibility Th. 8:45 am
Genetic variation and the pulmonary innate immune response to Legionella
Tom Hawn
University of Washington School of Medicine, Box 356523, BB1219, Seattle, WA 98195-6, United States of America [email protected]
The innate immune system enables the host to differentiate self from invading microbes, discriminate among pathogens, and initiate a cascade of inflammation that influences formation of the acquired immune response as well as host survival. Toll-like receptors (TLRs), which are critical mediators of the immune response to pathogens, contain polymorphisms that regulate cellular function and are associated with susceptibility to human infection. Our research seeks to understand the role of these polymorphisms in the pulmonary innate immune response to Legionella pneumophila (Lp). We have identified polymorphisms in the TLR pathway that are associated with regulation of the cellular immune response as well as clinical susceptibility to Lp. We recently identified a common variant in Toll-like Receptor 1 (TLR1), T1805G, which regulates the immune response to lipopeptides, molecules present in the cell wall of many bacteria including Legionella. PBMCs from TLR1-deficient individuals (TLR1 1805GG) had impaired IL-6 responses to stimulation with Lp suggesting that TLR1 mediates a significant proportion of the initial inflammatory response to Legionella. However, in a case-control genetic study with 101 patients with Legionnaire’s disease (LD) and 498 controls, we found no association of TLR1- deficiency with clinical outcome. In contrast to TLR1, we recently identified 3 polymorphisms in RP105 (CD180), which are strongly associated with protection from LD. Although RP105 regulates TLR2 (as a heterodimer with TLR1 or 6) and TLR4-mediated signaling, its role in the immune response to Lp is not currently known. These are the first associations of RP105 polymorphisms with any infection. We are currently investigating the functional significance of these polymorphisms and their role in Lp pathogenesis. With the identification of individuals with variation of different components of the human innate immune response, we are examining the mechanism of TLR regulation of immunity in order to elucidate how humans successfully control Legionella infection.
- 29 - Session 5: Immunology and host susceptibility Th. 9:15 am
Dendritic cells rapidly undergo apoptosis to limit intracellular replication of Legionella pneumophila
Catarina Nogueiraa, Tullia Lindstenb, Amanda Jamiesonc, Christopher Casea, Sunny Shina, Craig Thompsonb and Craig Royd aYale University School of Medicine, Section of Microbial Pathogenesis BCMM room345, 295 Congress Av., New Haven, CT 6536, United States of America; bUniversity of Pennsylvania, Department of Medicine and Pathology and Laboratory Medicine, Abramson Family Cancer Research Institute, Philadelphia, PA 19104, United States of America; cUniversity of Vienna, Campus Vienna Biocenter 4 , Max F. Perutz Laboratories, Dr. Bohr Gasse 9, A-1030 Vienna, Austria; dYale University, 295 Congress Ave, BCMM, rm 347, New Haven, CT 06536-0812, United States of America [email protected]
Dendritic cells (DCs) are specialized phagocytes that play a crucial role in pathogen detection and induction of subsequent innate and adaptive immune responses to eliminate pathogens from the infected host. There are several examples where DCs have been shown to be more efficient at restricting the intracellular replication of pathogens compared to macrophages, a property that could prevent DCs from enhancing pathogen dissemination. To understand how immune responses are generated against intracellular pathogens that can infect DCs, we investigated the mechanisms by which mouse DCs are able to restrict replication of Legionella pneumophila. We show that both DCs and macrophages have the ability to interfere with L. pneumophila replication through a cell death pathway mediated by caspase-1 and Naip5. L. pneumophila that avoided Naip-5 dependent responses, however, showed robust replication in macrophages but remained unable to replicate in DCs. Apoptotic cell death mediated by caspase-3 was found to occur much earlier in DCs following infection by L. pneumophila compared to macrophages infected similarly. The activation of this pathway is type IV secretion system (TFSS)-dependent, since bacteria with nonfunctional TFSS are unable to induce this rapid cell death pathway. Eliminating the pro-apoptotic proteins Bax and Bak or overproducing the anti-apoptotic protein Bcl-2 were both found to restore L. pneumophila replication in DCs. Thus, DCs have an intrinsic cell autonomous mechanism that rapidly activates apoptosis to limit pathogen replication.
- 30 - Session 5: Immunology and host susceptibility Th. 9:30 am
Pore formation is an Nlrc4/Ipaf inflammasome-dependent host cell response against virulent Legionella species that express flagellin
Tatiana Silveira and Dario S. Zamboni
Department of Cell Biology and Microbial Pathogenesis. University of Sao˜ Paulo, Medical School Ribeirao˜ Preto, FMRP/USP, Bandeirantes Avenue, 3900, 14049-900 Ribeirao Preto, Brazil [email protected]
Legionella pneumophila inhabit a wide variety of naturally occurring and man-made aquatic systems and causes a severe form of pneumonia called Legionnaires’ disease in humans. Essential for L. pneumophila pathogenesis is its capacity to replicate within alveolar macrophages. Therefore, the bacterium utilizes a type IVB secretion system called Dot/Icm to modulate phagosome biogenesis and to create an intracellular niche that supports bacterial multiplication. We have previously demonstrated that the Nod-like receptors (NLRs) Naip5 and Ipaf trigger caspase-1 activation in response to bacterial products delivered in the host cytoplasm. Previous studies have demonstrated that L. pneumophila triggers pore formation in macrophages by mechanisms dependent on the Dot/Icm secretion system. We found that regardless of Dot/Icm activity, pore formation does not occur in macrophages deficient in caspase-1 and Nlrc4/Ipaf. Flagellin, which is a known agonist for the Nlrc4 inflammasome, was required for pore formation as flaA mutant bacteria failed to induce cell permeabilization. By using 11 different Legionella species, we found robust pore formation in response to L. micdadei, L. bozemanii, L. gratiana, L. jordanis and L. rubrilucens, and this trait correlated with flagellin expression by these species. Our results suggest that pore formation is a highly coordinated host cell response dependent on host Nlrc4 and caspase-1 and on bacterial flagellin and type IV secretion systems.
- 31 - Session 5: Immunology and host susceptibility Th. 9:45 am
Type I IFNs mediate innate intracellular defense to Legionella pneumophila
Juliane Lippmanna, Vincent Van Laaka, Julia Eitela, Sunny Shinb, Gregory A Taylorc, Norbert Suttorpa, Craig Royd and Bastian Opitza aCharite´ University Medicine Berlin, Dept. of Infectious Diseases and Pulmonary Medicine, Augustenburger Platz 1, 13353 Berlin, Germany; bYale University School of Medicine, Section of Microbial Pathogenesis BCMM room345, 295 Congress Av., New Haven, CT 6536, United States of America; cDepartments of Medicine, Molecular Genetics and Microbiology, and Immunology, Duke University, and GRECC, VA Medical Center, 508 Fulton Street, Durham, GA NC 27705, United States of America; dYale University, 295 Congress Ave, BCMM, rm 347, New Haven, CT 06536-0812, United States of America [email protected]
Host innate immune responses to Legionella pneumophila depend on (I) the Toll-like receptors, (II), the NAIP5/IPAF inflammasome and (III) IFNγ-mediated antibacterial mechanisms. In our study we investigate a further innate defense mechanism potentially directed against Legionella infection, which is governed by type I IFNs (IFNα/β). Our results showed that both wildtype (wt) and the flagellin-deficient ∆flaA L. pneumophila induced IFNβ in murine bone marrow macrophages (BMMs) dependent on their T4SS. While in C57BL/6 BMMs growth of wt Legionella was rapidly terminated, BMMs lacking IFNAR allowed substantial bacterial replication. We also observed increased bacterial replication of Legionella ∆flaA in IFNAR-/- compared to C57BL/6 BMMs, while growth was restricted in C57BL/6 BMMs upon treatment with IFNβ in a dose-dependent manner. Endogenous IFNα/β as well as exogenous IFNβ negatively affected bacterial numbers in replication vacuoles, but appeared not to influence ER recruitment to the vacuole or its fusion with lysosomes. Our results suggest involvement of immunity-related GTPases, but do not argue for a major role of autophagy, caspase-1-dependent cell death, reactive oxygen or nitrogen species, indolamine 2,3-dioxygenase or transferrin receptor-mediated iron-supply in IFNα/β-mediated defense. Taken together, IFNα/β-mediated mechanisms might contribute to the host innate immune response against L. pneumophila infection.
- 32 -
SESSION 6 MICROBE - ENVIRONMENT INTERACTIONS A) CELL BIOLOGY, PROTOZOA, BIOFILMS
- 33 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms Th. 10:30 am
Proteomic and functional characterization of outer membrane vesicles and Legionella-containing phagosomes
Michael Steinerta, Olga Shevchuka, Frank Galkaa, Albert Haasb, Klaus Heunerc, Sun Nyunt Waid and Su- sanne Engelmanne aInstitut fur¨ Mikrobiologie, Technische Universitat¨ Braunschweig, Spielmannstr.7, 38106 Braunschweig, Germany; bInstitut fur¨ Zellbiologie, Universitat¨ Bonn, Ulrich-Haberland-Str. 61a, D-53121 Bonn, Germany; cRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; dDepartment of Molecular Biology, Umea˚ University, 6K och 6L, Sjukhusomradet,˚ SE-901 87 Umea,˚ Sweden; eInstitut fur¨ Mikrobiologie, Ernst-Moritz-Arndt- Universitat, F.-L.-Jahn-Str. 15, D-17487 Greifswald, Germany [email protected]
Intracellular growth of L. pneumophila within phagocytic cells depends on a complex pathogen-host cross-talk. We have used a proteomic approach to analyze specific aspects of both sides of the Legionella-host interaction. On the pathogen side we elaborated comprehensive proteome reference maps for outer membrane vesicles (OMVs). These spherical bilayered structures are produced by L. pneumophila during extra- and intracellular growth. OMVs manipulate phagocytic cells in many ways. However, despite diverse enzyme activities which may contribute to tissue destruction during human infection, isolated OMVs do not kill host cells but promote protozoa growth. On the host cell side we performed a proteome analysis of Legionella-containing phagosomes isolated from the social amoeba Dictyostelium discoideum. Cell fractionation, two-dimensional gel electrophoresis and MALDI-TOF MS combined with genomic data identified 157 phagosome host proteins. In addition to proteins with an evident role in phagosome maturation, we identified proteins for which a function remains to be elucidated. A comparative proteomic approach of isolated phagosomes containing different Legionella strains revealed virulence- and time-specific changes. A reference flow chart suggests so far unrecognized consequences of infection for host cell physiology and actin degradation on phagosomes, and a putative cysteine proteinase inhibitor interference with lysosomal enzyme sorting and activation processes.
- 34 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms Th. 11:00 am
Signals produced and exploited by Legionella
Hubert Hilbi
Institute of Zoology, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland [email protected]
Legionella pneumophila is an amoebae-resistant opportunistic pathogen that employs a biphasic life cycle to replicate in host cells and spread to new niches. In contrast to many other bacteria, L. pneumophila apparently does not regulate gene expression by producing diffusible ”autoinducer” signaling molecules such as acylated homoserine lactones. We recently identified the lqs (Legionella quorum sensing) gene cluster, which encodes the autoinducer synthase LqsA, the putative cognate sensor kinase LqsS and the response regulator LqsR. LqsA was found to be a pyridoxal-5’-phosphate- dependent enzyme that produces a diffusible α-hydroxyketone signaling molecule (3-hydroxy-pentadecan-4-one) termed LAI-1 (Legionella autoinducer-1). Functional studies and transcriptome analysis revealed that lqsA, lqsS and lqsR regulate phagocyte interactions, bacterial sedimentation and a genomic ”fitness” island. L. pneumophila employs a conserved mechanism to replicate in amoebae and macrophages within a unique membrane- bound compartment, the ”Legionella-containing vacuole” (LCV). Formation of LCVs is a complex and robust process involving the Icm/Dot type IV secretion system and more than 100 translocated ”effector” proteins that are thought to subvert host cell signaling and vesicle trafficking pathways. We identified and characterized Icm/Dot substrates, which bind specific host cell phosphoinositide (PI) lipids, and thus exploit these trafficking signals to anchor to the LCV membrane. PI-binding effector proteins interfere with host cell vesicle trafficking and (i) promote the interaction with organelles, (ii) catalyze guanine nucleotide exchange of small GTPases, or (iii) bind to PI-metabolizing enzymes, such as the PI 5-phosphatase OCRL1 (Oculocerebrorenal syndrome of Lowe). OCRL1 is implicated in retrograde endosome to Golgi trafficking and decorates LCVs in amoebae and macrophages. LCVs avoid the fusion with lysosomes, intercept the early secretory pathway and interact with the endoplasmic reticulum (ER). We purified intact LCVs from Dictyostelium discoideum amoeba by immuno-magnetic separation. To this end, we used a primary antibody against an Icm/Dot substrate exclusively localizing to LCVs and a secondary antibody coupled to magnetic beads. The proteome of purified LCVs was determined by liquid chromatography followed by tandem mass spectrometry and revealed more than 560 host proteins. These include a number of small GTPases implicated in endosomal or secretory vesicle trafficking. Thus, LCVs seem to communicate not only with the early secretory pathway (ER to cis-Golgi), but also with the late secretory pathway (trans-Golgi to endosomes), as well as with early and late endosomal trafficking pathways.
- 35 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms Th. 11:30 am
Host determinants of Legionella infection
Howard Shuman
Columbia University Medical Center, Department of Microbiology & Immunology, 701 West 168th Street, New York, NY 10032, United States of America [email protected]
Legionella sp. are ubiquitous in aquatic environments and are considered an ”accidental” pathogen of people. We used two approaches to find out about the host factors that determine the outcome of a Legionella infection. In the case of amoebae, we examined environmental samples for amoebae that could graze on wild-type Legionella pneumophila. Amoebae representing a wide variety of morphotypes and phylotypes could be easily recovered that grazed on L. pneumophila. This suggests that not all amoebae are equally susceptible to lethal infection by L. pneumophila. These differences may be due to variation in the abilities of L. pneumophila effector proteins to target host functions. In a completely independent approach we screened libraries of small molecules with known targets for their ability to inhibit Icm/Dot effector translocation to macrophages. We found that a broad array of small molecule inhibitors could completely block effector translocation. In some cases, these compounds could protect macrophages from lethal infection by L. pneumophila by targeting a host function required for infection. For example, active host cell receptor tyrosine phosphate phosphatases are required for L. pneumophila infection of macrophages.
- 36 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms Th. 12:00 am
A new method of producing Legionella pneumophila biofilms in a minimal medium
Sophie Pecastaings´ a, Mathieu Berge´a, Karine Dubourgb and Christine Roquesa aFaculte´ de Pharmacie - Laboratoire de Microbiologie Industrielle, 35 ch des Maraˆıchers, 31062 Toulouse, France; bInstitut du Thermalisme - Universite´ Victor Segalen´ Bordeaux 2, 8 rue Ste Ursule, 40100 Dax, France [email protected]
Little is currently known about L. pneumophila surface colonization and biofilm formation. Some authors argue that L. pneumophila is not able to produce biofilms without another prokaryote or eukaryote species, while some studies show that L. pneumophila biofilms can be produced using a rich medium but depend on planktonic growth. In order to better understand L. pneumophila’s capacity to produce a biofilm under minimal conditions, we optimized a method using a low nutrient concentration medium. At the same time, biofilm structure was investigated using confocal laser scanning microscopy. Three environmental isolates of L. pneumophila were inoculated in 24-well microplates containing the tested growth media. To eliminate planktonic bacteria, the media were frequently renewed. Adhered population of L. pneumophila was monitored by culture counts or by quantitative PCR, in order to establish biofilm formation kinetics. Confocal laser scanning microscopy (CLSM) was used to determine the 3-D structure of biofilms. Among 9 growth media tested, a medium consisting of a mix of mineral salts and a low concentration carbon source was selected. This medium did not support planktonic growth. However, after 7 days of incubation, biofilms were constituted of 5.75±0.12 log CFU/cm2. The adhered population remained stable up to 3 weeks after initial inoculation, whichever strain was tested. Quantification was similar by culture counts or q-PCR. Subsequent analysis showed that initial bacterial concentration (102 or 106 CFU/well) had no effect on the final adhered biomass. In situ CSLM observations of adhered L. pneumophila stained with Syto9 revealed a typical biofilm structure, comprising cell clusters ranging up to 300µm. Staining the carbohydrates with a fluorescent lectin indicated that the L. pneumophila biofilm matrix had a specific makeup: clusters or sheets of carbohydrates were scattered between cells instead of constituting a slime in which cells were embedded. The model described in this study is an innovative protocol that allows the formation of mono-species L. pneumophila biofilms grown in minimal conditions, differing from biofilms formed in a rich medium. This method would be of interest in determining the effectiveness of disinfectants used in combating water system biofilms.
- 37 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms Th. 12:15 am
Mature infectious forms (MIFs) and MIF-pellets as factors in the transmission of Legionnaires’ disease
Rafael A. Gardunoa, Celia Limaa, Gary Faulknera, Sharon G. Berkb, Michele Merchatc and David S. Allana aDalhousie University, Department of Microbiology and Immunology, Tupper Bldg. 7th floor, 5850 College Street, NS B3H-1X5 Halifax, Canada; bTennessee Technological University, Center for the Management, Utilization and Protection of Water Resources, 1020 Stadium Drive, P.O. Box 5033, Cookeville, TN TN 38505, United States of America; cClimespace, 185 rue de BERCY, 75012 Paris, France [email protected]
Legionnaires’ is an environmental disease, always transmitted from the environment to humans, with no human-to-human transmission. Previously we suggested that the interaction with the ciliate Tetrahymana predisposes L. pneumophila to better survive in the environment and infect humans. Here we report that differentiation of L. pneumophila into mature infectious forms (or MIFs) is a requirement for the survival of this pathogen in protozoa, but not humans, and that the protozoa-mediated packaging of MIFs into pellets or vesicles is a factor that would only occur in freshwater that provides fitness advantages to L. pneumophila. Methods: Specimens for electron microscopy were fixed in glutaraldehyde- osmium, and embedded in epoxy resin. Specimens for immuno-gold labeling were fixed in formaldehyde and embedded in the hydrophilic resin LR White. Immuno-fluorescence was performed with an antibody specific for the major outer membrane protein (MOMP) of L. pneumophila. Infectivity was evaluated by plaque assay, and survival against various environmental stresses by bacterial viable counts. Results: Differentiation mutants of L. pneumophila with defects in rpoS and letA were readily digested by Tetrahymena tropicalis and had growth defects in Acanthamoeba castellani, but grew well in HeLa cells and macrophages indicating the sharp contrast of outcomes between protozoa and mammalian hosts. Pellets containing live MIFs, produced by T. tropicalis, were infectious to macrophages and A. castellani, and protected the contained MIFs from desiccation, UV light, and biocides commonly used in cooling towers. Finally, pellets or vesicles specifically labeled with the MOMP-specific antibody were found in potable water systems, suggesting that packaging of MIFs into complex infectious particles does happen in water systems. Conclusions: The interaction with protozoa apparently forces L. pneumophila to differentiate into MIFs (regarded as the transmissive forms of L. pneumophila), and may result in the formation of complex infectious particles (pellets from ciliates or amoeba, and vesicles from amoeba). The selective pressure to differentiate into MIFs is not present in the human host, and packaging of MIFs into complex infectious particles would not occur in humans. Therefore, these events constitute important factors to consider in the transmission of Legionnaires’ disease and in explaining the lack of human-to-human transmission.
- 38 -
SESSION 7 MICROBE - ENVIRONMENT INTERACTIONS B) DETECTION IN NATURAL AND ARTIFICIAL RESERVOIRS - SURVIVAL, PERSISTENCE
- 39 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence Th. 3:00 pm Risk factors for Legionella contamination and persistence in potential sources of infection
Paola Borella
Department of Public Health Sciences University of Modena and Reggio Emilia, Via Campi 287, 41100 Modena, Italy [email protected]
Bacteria of the genus Legionella normally inhabit fresh water or wet soils, but the major reservoirs for human infection are the artificial aquatic environments, particularly hot water systems and cooling towers. The peculiar abilities to survive as intracellular parasite of free-living protozoa and to be associated with biofilms are responsible for the frequent contamination of man-made water systems as well as difficulties in eradicating legionellae from contaminated sites, and the scarce efficacy of biocides. Among risk factors contributing to explain Legionella spp. colonisation, those related to buildings, water distribution systems, water composition and germ characteristics have been studied. From a large survey on public and private structures (hospitals, hotels, homes) conducted in Italy, aimed to search, isolate and subtype Legionella strains from hot water, the main risk factors for contamination were structural such as the building age (hotels) and the centralisation of water distribution systems (homes), but differences according to Legionella species and serogroups were also observed. Concerning the new water distribution systems, plumbing materials, constructive techniques and dimensions may have a role in colonisation, but we also remark that water left inside plant between system testing and facility occupation may be associated with initial contamination. After colonization is established, case onset can depend on the contamination characteristics such as to have L. pneumophila serogroup 1, germ levels >10,000 CFU liter−1 and/or high percentage of positive points. Water temperature is essential to control contamination, but other physical-chemical parameters such as pH, hardness, conductivity, copper and residual chlorine have been studied for their capability in inhibiting/favouring Legionella colonisation and persistence. Genetically defined strains of potentially virulent L. pneumophila may become resistant to treatments favouring the long-term persistence of the germ. In addition, recent reports suggest that weather conditions, particularly humidity, and local water supplying can contribute to Legionella infection risk through direct contamination of water sources or introduction of micronutrients and/or commensal microorganisms. Climate may also have a direct influence on airborne dissemination of contaminated aerosols from cooling towers, whereas inadequate disinfectant procedures and/or insufficient maintenance interventions contribute to cooling towers colonisation. Competition with other aquatic microorganisms, also in relation to the production of bacteriocin-like substances, has not been sufficiently studied, although their knowledge might lead to more effective measures for controlling Legionella spp. persistence in the environment.
- 40 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence Th. 3:30 pm Legionella physiology in the environment
Shin-Ichi Yoshida
Department of Bacteriology, Faculty of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, 812-8582 Fukuoka City, Japan [email protected]
Legionellae are usually cultured at 35-37 C in artificial BCYEα and BYE media. The media are amino acid-rich and Fe3+-rich, but do not contain glucose and sodium salt. However, legionellae mainly proliferate in protozoa and live in biofilms in fresh water and soil where temperature is lower than 30 C, nutrition is poor, Fe3+ concentration is low, and sometimes it is salty. Legionellae have tactics to overcome such unfavorable conditions. Here, we present our new findings on the physiology of legionellae in the environment. 1. Tactics of growing at low temperatures a) L. dumoffii TEX-KL has a 66 kbp plasmid which enable the organism to proliferate at low temperatures. The responsible gene on the plasmid is one of tra region gene homologous to Agrobacterium tumefaciens TraA gene and was named TraA(Ti). The gene codes nickase and helicase which may accelerate the chromosome duplication at temperatures below 30 C. b) L. pneumophila Philadelphia-1 icmN gene product is not a structure protein of Icm/Dot type IV secretion apparatus but is thought to be an inner membrane protein. icmN mutants are reported to be defective in growing inside amoeba. We confirmed that icmN is important for L. pneumophila to grow in amoeba at low temperatures. 2. Biofilm formation Among legionellae, only L. pneumophila could form biofilm when cultured in BYE medium. The architecture of the biofilm varied according to the temperature at which the biofilm was formed. At 37 and 42 C, filamentous cells formed the biofilm. Whereas at 25 C, short rods formed the biofilm. 3. Isolates from salty hot spring L. pneumophila was isolated from hotsprings containing NaCl. The isolates could grow on the BCYEα containing more than 100mM NaCl. It is known that salt-tolerant legionellae are non-pathogenic but the isolates were observed to proliferate rapidly in J774 cells. 4. Glucose utilization L. pneumophila utilizes glucose via the Entner-Doudoroff pathway. Mutants of the genes coding the enzymes of this pathway could proliferate in vitro but not in amoeba. Glucose uptake was found to be necessary for intracellular growth in amoeba.
- 41 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence Th. 4:00 pm Detection of hosts for Legionella pneumophila in freshwater by enrichment of free- living protozoa in a biofilm-batch system
Rinske Valstera, Bastiaan Eggingb, Bart Wullingsa and Dick Van Der Kooija aKWR, Watercycle Research Institute, Groningenhaven 7, 3430 BB Nieuwegein, Netherlands; bLaboratory of Microbiology, Wageningen University, Dreijenplein 10, 6703 HB Wageningen, Netherlands [email protected]
Legionella pneumophila is widespread in natural freshwater environments and in man-made water systems, e.g., warm water installations and cooling towers. In vitro studies have shown that 17 species of free-living protozoa can serve as host for L. pneumophila. However, it cannot be excluded that other protozoa can act as host as well. Therefore, the objective of this study is to apply a method to amplify and subsequently identify protozoa that serve as host for L. pneumophila. Water samples (600 ml) collected from engineered fresh water systems (n = 21), were inoculated with L. pneumophila at concentrations between 4.2 × 103 and 5.0 × 104 mip copies liter−1. Cylinders of polyethylene (ø 20 mm, S 20 cm2) were added at S/V-ratio of 0.2 cm−1 to promote a biofilm for growth of L. pneumophila during incubation at 37◦C for 20 days. In 90% of the samples, growth of inoculated L. pneumophila was observed, when the host Hartmannella vermiformis was also added. In about 60% of the samples, a two- to three-log increase of the L. pneumophila concentration was observed with q-PCR in the water phase and in the biofilm. In 56% of these biofilm-batch tests, growth of indigenous H. vermiformis was detected with q-PCR, but growth of Acanthamoeba spp. was not observed. T-RFLP analyses of 18S rRNA gene fragments revealed that the eukaryotic diversity decreased during the incubation period. These observations suggest that certain host protozoa such as the frequently observed H. vermiformis, were able to compete effectively with other eukaryotes. In tests with cooling water and treated sewage 18S rRNA gene sequences related to Echinamoeba thermarum and Diphylleia rotans dominated in the biofilms with growth of L. pneumophila. These Amoebozoa have not been identified as host for L. pneumophila and further studies using FISH analysis are in preparation to demonstrate that these organisms can act as host. Results of the test show also which protozoa did not serve as host. For example, in the test with water from a warm water installation dominated sequences related to Sphaeroeca volvox, but no growth of L. pneumophila was observed.
- 42 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence Th. 4:15 pm Long-term survival of novel amoeba-associated bacteria compared with Legionella pneumophila under conditions of hydration and desiccation
Sharon G. Berka, Mary Faroneb, Wesley Willeforda, James Whitea and Michelle Lowea aTennessee Technological University, Center for the Management, Utilization and Protection of Water Resources, 1020 Stadium Drive, P.O. Box 5033, Cookeville, TN TN 38505, United States of America; bMiddle Tennessee State University, Biology Department, 1301 Main Street, Murfreesboro, AK 37132, United States of America [email protected]
Introduction. Previously we reported finding novel amoeba-associated bacteria from cooling towers and natural habitats. The majority of isolates were related to Legionella pneumophila, but were not culturable outside the amoeba host. The bacteria reproduced in amoebae and completely destroyed all amoeba cells in a co-culture. We now report results of studies regarding the survival of several of these bacterial isolates (and a few recent isolates) compared with the survival of L. pneumophila. Methods. To study the long-term survival of the novel bacteria and their subsequent ability to infect amoebae, we co-cultured the bacteria with Acanthamoeba polyphaga until the amoebae lysed, and then sampled the lysate periodically to determine how long the bacteria remained infectious to amoebae. To determine whether the novel bacteria survive desiccation, 50 µl samples of the lysates were placed onto empty sterile petri dishes and allowed to dry in a desiccation chamber at 5-9% relative humidity. Periodically over several months, dried lysate samples were rehydrated with sterile distilled water; and fresh washed amoebae were added to the rehydrated bacteria to test for infections. Results. Of 11 novel isolates tested, only 1 had a survival pattern similar to that of L. pneumophila, which remained infectious to amoebae in a hydrated state for 43 days, but did not survive 1 day in the desiccated state. Eight of the 11 isolates survived in the aqueous state longer than L. pneumophila (in the range of 50 to 130 days), and 2 isolates are still under investigation, but have survived up to 26 days to date. All novel isolates (except the one noted above with L. pneumophila) survived much longer than did L. pneumophila in the desiccated state. Most were still infectious in the range of 38 to 78 days, and 2 isolates were still infectious after 15 months. Conclusion. From previous research, we found that novel amoeba-associated bacteria are found more frequently in environmental samples than is L. pneumophila, and they do not grow on conventional media. The current finding that most of these novel isolates survive environmental conditions far longer than L. pneumophila may help to explain their prevalence compared with that of L. pneumophila. Some of the novel isolates have been shown to be capable of infecting human macrophage cells. Research is needed to determine the significance of such prevalent and hardy amoeba pathogens with respect to human infections.
- 43 -
SESSION 8 LEGIONELLA PREVENTION AND CONTROL
- 44 - Session 8: Legionella prevention and control Th. 4:50 pm
Travel associated Legionnaires’ disease in Europe: Reporting and response
Kate Ricketts, Rekha Yadav and Carol Joseph
Health Protection Agency Centre for Infections, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
EWGLINET collects data on cases of travel-associated Legionnaires’ disease from 35 European countries. The co- ordinating centre in London maintains a database of accommodation sites associated with these cases to enable the detection of clusters; the European guidelines lay out standards for the investigation of such. 8177 cases have been reported to EWGLINET since the scheme began in 1987. The number of cases submitted annually has RISEN over time as the number of countries taking part has increased and as national surveillance schemes have strengthened. This paper presents data for 2008, the most recent year of results, and places it in the context of the scheme’s historical data. Cases have risen from an average of 140 per year between 1987 and 1999 to an average of 684 per year between 2000 and 2007 (with a peak of 946 cases in 2007). In 2008, 866 cases were reported by 21 EWGLINET countries and 2 non-EWGLINET countries (America and Australia). The cases visited a total of 63 countries; Italy was associated with the highest number of cases (n=182). The proportion of cases with a travel history to countries outside Europe has shown a notable increase in recent years. This is to be expected as travel becomes more accessible, but is increasingly leading to clusters of cases being identified in countries where EWGLINET has no control over the investigation and the implementation of control measures. Since the introduction of the European guidelines in July 2002, EWGLINET has detected 688 clusters. 108 of these were detected in 2008, of which 35.2% would not have been identified without the scheme’s international database. The guidelines require that, when these clusters are located in European countries, they must each be investigated. In 2008 EWGLINET published the names of twelve accommodation sites on its website for failure to adhere to the agreed standards. The case fatality rate (CFR) observed in the EWGLINET dataset has decreased over time. Whilst the CFR was 6.2% between 1987 and 2007, it fell to 4.9% in 2008. This may be a result of reporting bias as cases are notified to EWGLINET in a more timely manner without the final information on patient outcomes, and may also reflect better diagnosis and reporting of less seriously ill cases across Europe. Through its national collaborators, EWGLINET fulfils an important public health role, ensuring the rapid detection and investigation of travel-associated clusters and the prevention of further cases in travellers. The EWGLINET investigation guidelines mandate that all such clusters in Europe are investigated to internationally accepted standards, and help to ensure the safety of travellers within Europe. With the planned transfer of the EWGLINET scheme to ECDC in 2010, new challenges are being presented. Ageing populations, warmer climates and improved diagnostics will tend towards an increase in the number of cases reported to the scheme and, as people’s travel patterns become more global, a truly international approach to tackling travel-associated Legionnaires’ disease will be required.
- 45 - Session 8: Legionella prevention and control Th. 5:20 pm
Legionella and the Conundrum of Low Risk Buildings: a Very Modest Proposal
Janet E. Stouta and Victor L. Yub aUNIVERSITY OF PITTSBURGH, Dept of Civil and Envir. Engineering, Special Pathogens Laboratory, 1401 Forbes Ave. Suite 209, Pittsburgh, PA PA15219, United States of America; bUNIVERSITY OF PITTSBURGH, Special Pathogens Laboratory, Pittsburgh, PA PA15219, United States of America [email protected]
Healthcare-acquired Legionnaires’ disease (LD) occurs with greater frequency in hospitals heavily colonized with Legionella pneumophila; the criteria of >30% of outlets positive with serogroup 1 has been surprisingly robust in predicting occurrence of disease. Disinfection of the drinking water systems have terminated outbreaks. Legionnaires’ disease is now diagnosed with greater frequency worldwide. As a result, more building water systems are being implicated as possible sources, although a definitive causal link usually cannot be made. These buildings include apartment buildings, commercial office buildings and hotels that house residents that are mostly immunocompetent, i.e. ”low risk” buildings. Unfortunately, even for low risk buildings, unvalidated and arbitrary concentration-based targets (colony-forming units) have been used to assess risk. We have found > 30% positive outlets in buildings associated with community-acquired Legionnaires’ disease. It is unclear, however, whether the parameters used to assess the risk of LD in high risk healthcare facilities can be extrapolated to low risk buildings. Expensive disinfection systems are being placed in many more buildings than is likely necessary from a public health perspective. For example, should a building colonized with L. anisa be disinfected? Moreover, sustained maintenance may be impractical resulting in disinfection systems becoming ineffective over time. A pragmatic restrained approach should be taken until sufficient objective data demonstrates that specific parameters are predictive of increased risk in these low risk buildings. Our modest proposal is to prospectively perform annual Legionella culturing of low risk buildings in which a case of LD has occurred. One time disinfection can be applied, followed by annual environmental culturing for Legionella as well as a prospective clinical survey to document subsequent cases. In this way, a database can be constructed and building parameters and incidence of LD can be prospectively accumulated. An international cooperative study would yield results that would provide the first step to using evidence-based medicine to assess risk for contracting LD in low risk buildings.
- 46 - Session 8: Legionella prevention and control Th. 5:50 pm
An international trial of quantitative PCR for monitoring Legionella in artificial water systems
John V. Leea, Martin Exnerb, Valeria Gaiac, Phillipe Hartemannd, Christian Luck¨ e,Beatrice´ Pangonf, Maria Luisa Riccig, Miquel Sabria`h and Susanne Surman-Leei aWater & Environmental Microbiology Reference Unit, Health Protection Agency, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; bInstitute for Hygiene and Public Health, University of Bonn, Sigmund Freud Str. 25, D-53105 Bonn, Germany; cIstituto cantonale di microbiologia, via Mirasole 22A, 6500 Bellinzona, Switzerland; dDepartement´ Environnement et Sante´ Publique, Universite´ CHU de Nancy, 9, Avenue de la Foretˆ de Haye, 54505 Vandoeuvre-les-Nancy, France; eInstitute of Medical Microbiology and Hygiene, German Reference Laboraotry for Legionella, Fiedlerstrasse 42, D-01307 Dresden, Germany; fService de Microbiologie, Centre Hospitalier de Versailles, 78157 Le Chesnay CEDEX, France; gDepartment of Infectious, Parasitic and Immune-mediated Diseases - Istituto Superiore di Sanita` -, Viale Regina Elena, 299, 161 Rome, Italy; hHospital Germans Trias i Pujol, carretera de canyet, 8916 Badalona, Spain; iFood Water and Environmental Microbiology Network - London Laboratory, Health Protection Agency, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
Quantitative PCR (qPCR) is increasingly used to monitor for Legionella in water systems. Current national guidelines and legislation are based on detection by culture. However there is no consistent correlation between qPCR and culture. Consequently qPCR results cannot be used as a direct substitute for culture making interpretation against current guidelines problematic. There is a need to develop guidelines that include the use of qPCR and are based on extensive field experience. We report a multi-centre international trial using a commercial qPCR method complying with AFNOR XP T90-471. The objective of this study was to define the action thresholds of (qPCR). through monitoring the dynamics of qPCR and culture results in different types of water systems. The study evaluated: 1) action thresholds based on qPCR compared to action thresholds based on culture and 2) action thresholds based on an increase in the qPCR signal. 8 laboratories participated from 6 countries. The methods were adjusted to ensure comparability of their respective detection limits. A set of quality control samples was analysed by each laboratory and all the results were satisfactory (within 0.5 log deviation). Each laboratory analysed >100 samples collected while monitoring at ≥5 systems weekly over 2 - 3 months. >800 samples were analysed (>200 from cooling systems, >600 from hot and cold water systems). Comparing results to presence or absence by culture, for cooling towers PCR for Legionella species and L. pneumophila gave a positive predictive value (ppv) of 28% and 48% respectively and negative predictive value ( npv ) of 100% and 98%. For domestic water the ppv was 35 and 41% and npv was 90 and 95% respectively. This confirms the potential value of qPCR for screening. It was possible to establish qPCR thresholds based on existing culture alert and action levels that overall suggested similar actions to the culture results. Trend analysis for results from specific systems was also undertaken. This study confirms that it is possible to derive algorithms for monitoring systems by qPCR and establish a set of guidelines that will inform regulators and public health practitioners on the practical use and value of qPCR for monitoring of legionellae
- 47 -
SESSION 9 EPIDEMIOLOGY: A) NOSOCOMIAL INFECTIONS AND LEGIONELLA CONTROL IN HOSPITALS
- 48 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals Fr. 8:30 am
Decontamination of Hospital Water Supplies: A Review of the Literature and the University of Iowa’s Experience
Loreen Herwaldt
University of Iowa College of Medicine, 200 Hawkins Drive, Iowa City, Iowa, 52242-1081, United States of America [email protected]
Several different approaches have been used to control growth of Legionella spp. in healthcare facilities’ plumbing systems, including: superheating the water and flushing the plumbing system, hyperchlorinating the water, adding supplemental chlorine, adding chlorine dioxide, adding copper-silver ions, or exposing the water to ozone, and exposing the water to ultraviolet light. All of these methods have been successful in some situations and have failed in others. Most studies assessing the efficacy of these disinfection methods have been quasi experimental (e.g., comparing rates of water contamination or nosocomial infections caused by Legionella spp. in a pre-intervention period with those during a post-intervention studies. Most studies have not had control groups and many have been done in hospitals with outbreaks of infections caused by Legionella spp. This presentation will review the literature on disinfection of hospital plumbing systems and will describe the University of Iowa’s experience with suppressing growth of Legionella spp. in a complex plumbing system over approximately 30 years.
- 49 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals Fr. 9:00 am
Nosocomial infections, Legionella strain characterization and host-related risk factors
Christophe Ginevra, Sophie Jarraud and Jer´ omeˆ Etienne
Centre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France [email protected]
Nosocomial cases or hospital outbreaks of Legionnaires disease have been linked to contamination of hospital water supplies in numerous reports. In France, prevention measures in hospitals have been generalized and the overall rate of nosocomial cases dropped from over 20 to 7% in 2008. In comparison to community-acquired cases, case-fatality, mean age, presence of cancer or blood diseases are higher for nosocomial cases. A virulence-associated epitope (recognized by the MAb 3/1 -Dresden Panel) was searched in a European collection of Legionella strains and the proportion of MAb 3/1-negative serogroup 1 isolates was significantly higher in nosocomial cases compared with community- (P=0.0055) or travel-associated (P<0.0005) cases. In water systems, MAb 3/1-negative strains are also predominant which means that the distribution of serotypes seen in nosocomial infections merely reflects the colonization of the water distribution system. MAb 3/1 positive strains harbor the lag-1 gene, whereas MAb 3/1 negative strains either lost lag-1 or contain mutated lag-1 gene. Molecular typing of 202 nosocomial L. pneumophila serogroup 1 strains isolated in France between 1998 and 2007 (no case being related) identified 63 isolates belonging to the Paris clone (31%). In France from 1998 through 2007, 80 Paris clone isolates were detected in patients with community-acquired Legionnaires disease, representing 8.2% of culture-proven cases. The crude mortality rate was significantly higher for patients infected with the Paris clone (38%). Female sex, advanced age, steroids or other immunosuppressive therapies, and a history of cancer or hematologic malignancy were more frequent among patients infected by the Paris clone. The particular tropism of the Paris clone for immunocompromised patients could explain the high prevalence of hospital-acquired Legionnaires disease caused by this strain. The Paris clone grouped ST1 or ST1-related isolates with a specific PFGE pattern and different Mab subgroups: Philadelphia (MAb 3/1-positive), or France/Allentown (MAb 3/1-positive) or Olda (MAb 3/1-negative). ST1 isolates have been detected all over the world. During the same time period, only 3 culture-proven cases with the Lorraine clone were identified. The Lorraine clone is a significant cause of Legionnaires disease in France, causing around 10% of cases of culture-confirmed disease, most of which are sporadic community cases. Smoking is more frequent among patients infected by the Lorraine clone than among patients infected by sporadic strains. Isolates of the Lorraine clone are rarely isolated from the environment. In contrast to the Paris clone, the isolates of the Lorraine clone have more homogeneous characteristics: they are of ST47, MAb France/Allentown (MAb 3/1-positive) with a specific PFGE pattern. ST47 isolates have been detected in several European countries (England, Wales, the Netherlands, Greece, Spain, etc.). In France, the most prevalent ST in culture confirmed cases was ST23. In 2008, isolates of ST23 were cultured in 40 of the 197 culture proven cases (20.3%). None of them was nosocomial cases.
- 50 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals Fr. 9:30 am
Multiple Legionella strains infection detected by sequence analysis directly from respiratory samples
Mireia Coscolla´ and Fernando Gonzalez-Candelas´
Universitat de Valencia, Pol´ıgono La Coma S/N, 46980 Paterna, Spain [email protected]
Mixed bacteria infections are common in respiratory tract infections, with different bacterial species isolated in cultures. But mixed infections with different strains of the same species have been studied only in long-term or chronic infections, such as those produced by Mycobacterium tuberculosis or Helicobacter pylori in the digestive tract. Legionellosis is an acute infection and outbreaks are studied by typing pure Legionella cultures from infected patients, assuming that only one strain is causing the disease. Here, we present evidence for different L. pneumophila strains with different sequence profiles infecting patients of a Legionellosis outbreak. The data were obtained by sequencing around 30 PCR products of several loci from respiratory samples of two outbreak patients. We recovered EWGLI profiles with six loci (fliC, pilE, asd, mip, mompS, proA) from respiratory samples of 15 patients of an outbreak produced in the late summer of 2008 in Alzira and Carcaixent (Valencia, Spain). We detected two main profiles, one (2, 10, 14, 10, 21, 4) present in 7 samples and the other (21, 14, 28, 15, 15, 29) in 3 samples. Three additional profiles were present in one and two samples and they showed mixed patterns with alleles of the two more common profiles. Sequences of these mixed profiles showed polymorphisms in several positions, which could result from the simultaneous presence of different Legionella variants in the corresponding sample. These results indicated that these mixed profiles could be chimeric, resulting from amplifying different strains in different loci. In order to test this possibility, we cloned and sequenced PCR-amplification products from all loci for three patients with a mixed profile. After obtaining around 30 sequences for each locus of the three patients, we detected several variants in two of the three patients analyzed. Sequences obtained from one of the patients resulted in several alleles for fliC, pilE, mompS and mip, while the more conserved loci (proA and asd) presented only one variant. Sequences for other patient resulted in several alleles only in fliC and mompS (but pilE was not analyzed in this patient). The third patient presented only one variant each locus. In summary, two of the three patients analyzed showed evidences of at least two Legionella variants during the acute infection. These results indicate that probably more than one strain was infecting some patients. This could be due to co-infection from the same environmental source or, alternatively, to independent infections in a very short period. Although our data cannot discriminate between these alternatives, these results clearly suggest that Legionella infection patterns could be more complex than previously assumed.
- 51 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals Fr. 9:45 am
The impact of monochloramine introduction on Legionella colonization in a hospital potable water system.
Lauri Hicksa, Tatiana Travisb, Linden Witherellc, Nicole Alexandera, Thomas Taylorb, Dawn Terashitad, Matthew Moorea, Claressa Lucasb and Barry Fieldsb aCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop C-23, Atlanta, GA 30333, United States of America; bCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, GA 30333, United States of America; cEnvironmental Health Services, Inc., 777 South Prospect St., Burlington, VT 5401, United States of America; dCounty of Los Angeles Public Health, 313 N. Figueroa St., Los Angeles, CA 90012, United States of America [email protected]
Background Healthcare-associated Legionnaires’ disease (LD) is often attributed to Legionella colonization of a facility’s potable water system. Previous studies have shown that healthcare facilities supplied with municipal water containing monochloramine, as compared to free chlorine, have a lower risk of healthcare-associated LD. Monochloramine is formed when ammonia and chlorine are mixed in water. It is more stable and has better biofilm penetration than chlorine. Two prospective studies evaluated Legionella colonization in large buildings before and after the conversion of the municipal water supply to monochloramine disinfection and both showed marked decreases in colonization post-conversion. We conducted a prospective environmental study to evaluate the impact of on-site injection of monochloramine into a hospital potable water system. Methods The study site was a large, acute care hospital supplied by municipal water containing chlorine. The hospital’s potable water was colonized with Legionellaand remediation had not been attempted. At each sampling period, a 1-liter water sample and a biofilm swab were collected from the facility’s heat exchanger and each of 27 sink faucets or showers, randomly selected within each floor and wing, and cultured for Legionella. Pre-chloramination samples were collected over a four month period immediately before introduction of monochloramine into the distribution system. Beginning on December 16, 2008, chlorine and ammonia were injected into the hot potable water supply to form monochloramine, and the first post-introduction sampling was completed December 22, 2008. Three additional sampling periods over 3 months were completed. Legionella and amoebae colonization rates were compared before and after monochloramine introduction using Fisher’s exact test. Monochloramine and total chlorine levels were measured at one proximal and one distal site for each sampling period. Results Monochloramine and free chlorine levels were undetectable in the pre-monochloramine period and had a mean detection level of 1.39 mg/L and .56 mg/L, respectively, after introduction. Legionella colonization ranged from 20 to 22 (71%-79%) of 28 sites positive in the four pre-monochloramine periods. Colonization fell significantly to 4 (14%) of 28 sites positive in the first post-monochloramine period (p< 0.001) and continued to decline for each subsequent sampling period. Legionella colonization was undetectable in the final sampling period. Isolates were predominantly L. pneumophila serogroup 1, monoclonal antibody subtype 1,6. There was no change in amoebae colonization rates. Conclusion Addition of monochloramine into a hospital potable water system rapidly reduced Legionella colonization and ultimately resulted in undetectable levels of Legionella. On-site monochloramine introduction may be a promising option for lowering risk of healthcare-associated LD. Further studies should be conducted in other hospital and large building settings to assess the feasibility, broader applicability, and long term sustainability of managing a monochloramine injection system.
- 52 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals Fr. 10:00 am
Lessons learnt from investigating hospital-acquired legionellosis and the remedial actions in England.
Susanne Surman-Leea, John V. Leeb, Sandra Laib, Falguni Naikc, Carol Josephc and Timothy Harrisond aFood Water and Environmental Microbiology Network - London Laboratory, Health Protection Agency, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; bWater & Environmental Microbiology Reference Unit, Health Protection Agency, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; cHealth Protection Agency Centre for Infections, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; dHealth Protection Agency Centre for Infections, Respiratory and Systemic Infection Laboratory, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
Nosocomial legionellosis is a relatively rare occurrence in England and Wales with 77 cases reported between 2000-2008. Because the death rate for hospital-acquired cases is high (35%) compared to community-acquired cases (12%) extensive investigations are always carried out when nosocomial cases are suspected. Since 2000, nine hospitals have had clusters of Legionnaires’ disease associated with them with one hospital responsible for two clusters and one further sporadic case. In 8/9 of the hospitals with clusters, the cause was due to contamination of the distributed water systems. Experience has shown that there is a the need for a multidisciplinary approach to the investigations to verify the hospital as the source and to ensure that remedial actions are effective. It is essential that: •The team includes microbiologists who understand Legionella ecology. They need to have sufficient engineering experience to identify potential sources and faults within the systems design, operation and management and should not limit themselves to investigating previously accepted sources; •Common problems seen were: deadlegs and blind ends in the plumbing; lack of temperature control of both the hot and cold distributed water; unbalanced water flow in systems; control valves inappropriately adjusted; lack of insulation; inappropriate materials of construction; unsuitable quick-fix repairs; filling the system with water prematurely; delays between commissioning and occupation of buildings etc. Investigations revealed potential problems and sources which had not previously been recognised including: •global warming producing increases in the temperature of cold mains supply water; •previously unrecognised potential sources which had high levels of Legionella one of which was a steam cleaner used in the hospital deep clean initiative to clean hospital wards and corridors; •inexperience and poor competence of water treatment providers and risk assessors: for example in failing to identify appropriate or sufficient sampling points and high risk areas and areas housing high risk patients; •nebulisers cleaned by rinsing in contaminated tap water •laboratory competence; ie:- ◦poor commercial laboratory performance resulted in routine monitoring failing to identify Legionella colonisation that resulted in a false sense of security ◦a clear understanding of the factors affecting the validity of the environmental culture results including the potential interference from background flora. ◦Selecting multiple isolates for confirmation and SBT typing is important to verify the source as the outbreak strain may not always be the predominant Legionella •problems with inappropriate disinfectants or inadequate disinfectant concentrations, In the UK chlorine dioxide is frequently used as a first line remedial measure but drinking water regulations limit the maximum level at 0.5 mg/L which we find is inadequate for treatment of colonised systems. Effective treatment may take years at elevated levels (2-3 mg/L) and require long term intensive monitoring for Legionella Although in all cases the systems responsible for cases were not designed, managed and maintained according to the UK Approved Code of Practice and Guidance, in the hosptials to be discussed there were also additional factors which increased the risk to patients.
- 53 -
SESSION 10 EPIDEMIOLOGY: B) OUTBREAKS AND POPULATION GENETICS AND TAXONOMY
- 54 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy Fr. 10:45 am
Managing public health crises: from Legionellosis to pandemic influenza
Antoni Plasencia
Health Department,. Catalonia, Spain., Roc Boronat 81-95, 8005 Barcelona, Spain [email protected]
Situations regarding the management of collective risks for population’s health require a systematic approach. This implies a differentiation between the concepts of alert, emergency, alarm and crisis, according to the response provided, either at the technical and scientific level, or at the societal level. Besides taking into account risk perception individual and community determinants, different mediating factors can contribute to prevent the occurrence of a public health crisis. Such factors, including leadership, investigation, participation, intervention, communication and evaluation, are at the core of the public health response and need to be operationally formulated. The cumulative experience in the management of Legionellosis risks and outbreaks in different contexts provides useful clues for the systematic management of situations potentially conducive to public health crises. Some of these lessons are also applicable to the responses in the management of the current A(H1N1) influenza pandemia, and should be taken into account to foster efforts towards improving crises preparedness and to promote a reasonable level of confidence in the public.
- 55 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy Fr. 11:15 am
Ecological diversity within Legionella
Rod Ratcliff
Institute of Medical and Veterinary Science, Infectious Diseases Laboratories, Frome Road - PO Box 14, Rundle Mall, 5000 Adelaide, Australia [email protected]
The species and strain diversity within Legionella reflect the evolutionary forces to which they have been subjected over time as they adapted to the various available environmental reservoirs. What are now recognised as species are the end result of these forces with each strain clade adapted to and occupying a distinct ecological niche. Random mutation causes the slow accumulation of genetic diversity. Natural selection results genetic refinement as the dominant strain purges the ecological niche of less fit siblings. As a consequence each strain cluster with a different ecological role corresponds to a discrete DNA sequence cluster for every gene in the genome (ref). However, it can be unclear which level of diversity within the sequence cluster corresponds to ecologically distinct populations. Current species delineation has been largely based on phenotypic distinctions overlayed with relatively crude DNA relatedness
- 56 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy Fr. 11:45 am
Legionellosis Prevention Guidelines: Better, but Not Well
Sebastian Crespi
Policlinica Miramar, Camino La Vileta 30, 7011 Palma de Mallorca, Spain [email protected]
BACKGROUND In recent years, there has been a spate of laws, regulations and guidelines aimed at preventing legionnaire’s disease. Although these regulations are generally based on the same broad prevention principles, significant differences are known to exist between them when put into practice. Further, the effectiveness of these prevention guides is not well established. Identifying the most contentious aspects to the different regulations and analysing their underlying available scientific evidence would be helpful so as to offer improvements. METHODS We have analysed 7 legionellosis prevention guidelines and regulations (European, Spanish, British, French, Dutch, Australian and American guidelines) in order to identify the significant differences between them. Alongside this, we have performed a PubMed database search for articles published during the period 1990-2008 on primary and secondary prevention for legionellosis in different settings (hospitals and other healthcare settings have been specifically excluded). Cost-benefit analysis and adverse-effects studies in relation to prevention of legionnaire’s disease were also included in the search. RESULTS We have identified significant differences between the analysed prevention guides across five areas: 1) Health-based Legionella targets; 2) Action levels following Legionella monitoring; 3) Recommended disinfection procedures; 4) Legionella testing in domestic water systems; 5) Risk Assessment procedures. From the more than 800 original or review articles screened on primary and secondary prevention, we have not found either controlled epidemiological studies or cohort or case-control observational studies. Nearly all published studies on prevention refer to control measures, secondary prevention and individual case studies. There are no cost-benefit studies or quantified studies on the possible adverse effects of the proposed preventative interventions. CONCLUSIONS AND DISCUSSION The current guidelines offer a framework for action and have improved previous situations but there is a need of significant repair. The lack of controlled or observational studies on the effectiveness of the proposed guidelines is the Achilles’ heel in current prevention systems. These studies are required in order to allow the main recommended prevention measures to be supported or refuted. Cost-benefit analysis are also needed. New prevention guidelines should incorporate rigorous assessment of the level of scientific evidence behind the main recommendations.
- 57 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy Fr. 12:00 am
From large community outbreak in Verhnaya Pyshma to effective prevention of legionellosis in Russia.
Igor S. Tartakovskiya, Yulia Deminab, Olga L. Voroninaa, Yulia S. Alyapkinac and Tatiana I. Karpovaa aN.F. Gamaleya Institute for Epidemiology and Microbiology RAMS, Gamaleya str., 18, 123098 Moscow, Russian Federation; bFederal service of surveillance in the sphere of the protection of consumers’ rights and human welfare, Vadkovski per. 18, 127994 Moscow, Russian Federation; cJSC ’Syntol’, Timirayzevskaya st., 42, 127550 Moscow, Russian Federation [email protected]
The first cases of legionellosis were reported in the former USSR 29 years ago.However only few cases of legionellosis notify by year in the modern Russia.In summer 2007 a large community pneumonia outbreak occured in Verhnaya Pyshma,Sverdlovsk region of Russia.More then one hundred people were hospitalized and 5 patients died.In 74 cases diagnosis of Legionnaires’disease was confirmed by EWGLI and WHO standards of laboratory diagnosis of legionellosis.Clinical symptoms were typical for Legionnaires’disease and most part of patients had underlying diseases.The main hypothesis of the epidemic was linked to massive contamination by legionellae of city hot water supply during summer prophylactic works.During some weeks considerable volumes of water were stalled in 2 km long pipe and the tempertature within system reached 30oC.Afterwards water supply was restored and masses of contaminated water under higgh pressure were delivered to inhabitants.Time points of the start of the epidemic (considering incubation period) were in good correlation with the date of water supply interruption.More then 70 Legionella pneumophila and Legionella spp. strains were isolated from water samples of different origin in Verhnaya Pyshma after cleaning and desinfection of water supply.Molecular typing by SBT protocol showed difference between clinical isolate of L.pneumophila serogroup 1 and other environmental isolates.New cases of Legionnaires’disease in Verhnaya Pyshma have not reported last 2 years.Two national guidelines and two national recommendations fro prevention and control of Legionnaires’disease introduced by Federal service of surveillance in the sphere of the protection of consumers’rights and human welfare in 2007-2009.These documents included different aspects of monitoring Legionella in potential dangerous water systems by bacreiological and real-time PCR methods,standardization of diagnostic tool and harmonization of epidemiological surveillance of Legionnaires’disease in Russia accordance EWGLI and WHO guidelines.
- 58 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy Fr. 12:15 am
Molecular evolution of dotA gene in Legionella pneumophila: contribution of natural environmental isolates.
Joana Costaa, Igor Tiagoa, Milton Da Costab and Antonio´ Ver´ıssimoa aCentro de Neurocienciasˆ e Biologia Experimental, Dep. Zoologia, Universidade de Coimbra, 3004-517 Coimbra, Portugal; bDep. Bioqu´ımica, Universidade de Coimbra, 3001-401 Coimbra, Portugal [email protected]
Given the role of DotA protein in establishing successful infections and the diversity of host cells interacting with Legionella pneumophila in nature, it is possible that this gene product is a target for adaptive evolution. Allelic forms of dotA may reflect fitness variations of L. pneumophila strains in distinct environmental niches. Despite the ubiquitous character of L. pneumophila in fresh water natural environments, most population studies are focused on isolates from man-made environments, namely air conditioning-systems, potable water distribution systems, and plumbing fixtures and on clinical-related strains. Previous studies suggest the existence of intraspecific recombination of dotA among L. pneumophila isolates from man-made environments and clinical-related strains. We investigated the contribution of L. pneumophila isolates from natural environments, namely, streams, rivers, lakes, ground waters, to the molecular evolution of this crucial virulence-related gene. The population genetic structure of L. pneumophila was inferred from the partial sequences of rpoB and dotA of 304 worldwide strains, including isolates from natural and man-made environments and clinical-related strains. The isolates were clustered in discrete groups obtained from the rpoB partial sequence analysis, matching the three different L. pneumophila subspecies. While five major clusters were inferred from the partial dotA sequence analysis. The topology of the two trees was not congruent since most strains had different relationships with each other and with L. pneumophila reference and type strains, depending on the considered gene. Incongruence between the phylogenetic trees may result from dotA gene recombination and/or positive selection. Moreover, the vast majority of the isolates from natural environments were clustered in a discrete group including few clinical-related strains. The Ka/Ks ratios demonstrated that this group, contrarily to all others, has been under strong diversifying selection with respect to amino acid change. The importance of recombination has become increasingly clear in recent years, both as a fundamental process in strain diversification and as a mechanism by which strains acquire virulence factors. The large number of strains, their diverse origin, together with the inclusion of reference strains of L. pneumophila allowed us to expand our view of the population genetic structure of this species. Strains for entire dotA gene sequencing were selected within the previously defined clusters. The alignment of all DotA amino acid sequences allowed the identification of several isoforms. The degree of amino acid diversity was different within each of the DotA clusters, being the DotA from the group comprising the vast majority of the isolates from natural environments the only one under strong diversifying selection. Amino acid variations were not randomly distributed, and the estimates of both positive and purifying selection at each amino acid site showed that the periplasmatic regions PP2 and PP4 were highly variable. These DotA isoforms may reflect the dispersal of advantageous alleles that may attain high frequencies among genetically related individuals by means of sporadic selection events. Furthermore, the existence of these isoforms strongly suggests differences in the persistence of certain strains in a given environment and may be associated with different levels of virulence verified for some strains.
- 59 -
SESSION 11 GENOMICS AND COMPARATIVE GENOMICS
- 60 - Session 11: Genomics and Comparative genomics Fr. 3:30 pm
Legionella pneumophila and Legionella longbeachae pathogenesis: new insights gained form comparative and functional genomics
Christel Cazaleta, Laura Gomez-Valeroa, Mariella Lommaa, Christophe Rusnioka, Nora Zidaneb, Christiane Bouchierb, Sophie Jarraudc,Jer´ omeˆ Etiennec and Carmen Buchriesera aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bPlate-forme Genomique,´ Pasteur Genopole´ °R Ile de France, 25-28, Rue du Dr. Roux,, 75724 Paris, France; cCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France [email protected]
Although best known for its ability to cause severe pneumonia in people whose immune defences are weakened, Legionella pneumophila and Legionella longbeachae are two species of a large genus of bacteria that are ubiquitous in nature. Despite the abundance of dozens of species of Legionella in aquatic reservoirs, 90% of human disease is caused by L. pneumophila. An exception is Australia and New Zealand where L. pneumophila accounts for 45.7% and Legionella longbeachae for 30.4% of legionellosis cases. Interestingly, the two Legionella species live in different environmental niches. L. pneumophila is mainly found in natural and artificial water circuits while L. longbeachae is mainly found in soil. Adaptation to the host environment and exploitation of host cell functions are critical to the success of intracellular pathogens. Thus, in order to better understand the genetic basis allowing these two pathogens to multiply intracellularly and to cause disease, we have undertaken a comparative and functional genomics approach. The analysis of the complete genome sequences of L. pneumophila had revealed a variety of unexpected features unique to Legionella, such as the extended array of eukaryotic-like proteins, promising candidates for involvement in host pathogen interactions. Many of these proteins, such as tetratrico peptide repeat, ankyrin, F-Box, and coiled-coil domain proteins or serine-threonine protein kinases, apyrases and a sphingosine-1-phosphate lyase, are predicted or have been shown to modulate host cell functions to the pathogen’s advantage. Comparison of L. pneumophila with the complete genome sequence of L. longbeachae strain NSW150 revealed features specific to this pathogen. L. longbeachae encodes additional secretion systems and seems to interact differently with the host, as many of the eukaryotic domain proteins identified in L. pneumophila as well as many substrates of the Dot/Icm type IV secretion system are different between these two species. Furthermore, L. longbeachae does not encode flagella, explaining the differences in mouse susceptibility, does not seem to have a biphasic life cycle like L.pneumophila, as revealed from transcriptome analysis and encodes a capsule. The comparative genome analysis thus revealed species specific differences some of which may be related to their different environmental niches.
- 61 - Session 11: Genomics and Comparative genomics Fr. 4:00 pm
Polymorphic loci in Legionella pneumophila serogroup 1
Brian Sheltona, Paul Foglea, Barry Fieldsb, Natalia Kozakb, Laurel Carterc, Amgad Salehc and John Lesliec aPathCon Laboratories, 270 Scientific Drive, Suite 3, Norcross, Georgia, AK 30092, United States of America; bCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, GA 30333, United States of America; cKansas State University, Department of Plant Pathology, 4024 Throckmorton Plant Sciences Center, Manhattan, Kansas, AK 66503-3114, United States of America [email protected]
The genomic core of a large number of Legionella pneumophila serogroup 1 isolates was screened for conserved genes common to all isolates. Identified conserved genes were assessed to determine the level of sequence variability within them. Seven loci common to all serogroup I strains tested had a high level of polymorphism with most variations occurring in the third base position, suggesting that these regions encode functional proteins. PCR primers were developed that were specific to the conserved regions in each of these seven genes. DNA from five ST1 and five ST36 L. pneumophila isolates recovered from diverse geographic regions was extracted, amplified by PCR and the resulting gene fragments sequenced. All seven fragments from the five ST1 isolates were identical to their homologues in the sequenced L. pneumophila strain Paris, while all seven fragments from the five ST36 isolates were identical to their counterparts in the sequenced strain Philadelphia-1. Any of the seven gene fragments could differentiate the four fully sequenced L. pneumophila serogroup 1 genomes currently in GenBank, i.e., Corby, Lens, Paris, and Philadelphia-1. These gene fragments can be used to provide rapid preliminary information regarding genetic variability in clinical and environmental Legionella populations.
- 62 - Session 11: Genomics and Comparative genomics Fr. 4:15 pm
Proteomic analysis of a Legionella-containing phagosome in Dictyostelium
Olga Shevchuka, Susanne Engelmannb, Michael Heckerb, Albert Haasc, Klaus Heunerd and Michael Steinerta aInstitut fur¨ Mikrobiologie, Technische Universitat¨ Braunschweig, Spielmannstr.7, 38106 Braunschweig, Germany; bInstitut fur¨ Mikrobiologie, Ernst-Moritz-Arndt-Universitat, F.-L.-Jahn-Str. 15, D-17487 Greifswald, Germany; cInstitut fur¨ Zellbiologie, Universitat¨ Bonn, Ulrich-Haberland-Str. 61a, D-53121 Bonn, Germany; dRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany [email protected]
Legionella pneumophila, the agent of Legionnaires’ disease, replicates intracellularly within specialized phagosomes of human macrophages and amoebae. In this study we have developed a protocol for the isolation of Legionella- containing phagosomes from D. discoideum. Cell fractionation, two-dimensional gel electrophoresis and MALDI-TOF MS combined with genomic data identified 157 phagosome host proteins. Phagosomal proteins were analysed for physical and functional interaction using the STRING database. Network analyses of phagosomal proteins allowed the identification of protein clusters relevant for molecular functions associated with pathogenic phagosomes. Comparative proteomics of phagosomes containing highly virulent L. pneumophila Corby versus less virulent L. hackeliae revealed distinctive protein expression patterns, e.g., an abundance of RhoGDI and FttB in L. hackeliae degrading phagosomes versus little RhoGDI and FttB in L. pneumophila Corby replicative phagosomes. We present a kinetic dissection of phagosome maturation including the complex alterations of the phagosome protein composition. A reference flow chart suggests thus far unrecognized consequences of infection for host cell physiology, actin degradation on phagosomes and a putative cysteine proteinase inhibitor interference with lysosomal enzyme sorting and activation processes.
- 63 - Session 11: Genomics and Comparative genomics Fr. 4:30 pm
Genome re-annotation and web resources for Legionella pneumophila species
Laura Gomez-Valeroa, Pierre Lechatb, Zoe Rouyc, Claudine Mediguec, Carmen Buchriesera and Ivan Moszerb aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bInstitut Pasteur, Integration´ et Analyse Genomiques,´ 28 rue du Docteur Roux, 75724 Paris Cedex 15, France; cCEA /DSV /FAR /IG/ Genoscope Laboratoire de Genomique Comparative, 2 rue Gaston Cremieux, 91057 Evry Cedex, France [email protected]
Since the publication of the first complete genomes of Legionella pneumophila in 2004, their sequences and related annotations have been available through the LegioList web site. This genome browser was derived from a generic data structure and user interface, originally designed for the genomes of Escherichia coli (Colibri) and Bacillus subtilis (SubtiList) more than a decade ago. Thus, a comprehensive update of this database proposing enriched capabilities for queries, navigations and comparisons through a new web resource was desirable. Furthermore, given the high number of new genomes published and the functional work undertaken on many of the L. pneumophila proteins since publication of the sequences, a re-annotation of the genomes was undertaken to have a reliable reference. Re-annotation was performed using the annotation platform MaGe (http://www.genoscope.cns.fr/agc/mage), considering L. pneumophila Paris as the reference genome. Physical locations of coding sequences (CDS) underwent a systematic reassessment, using specialized procedures for bacterial gene identification (e.g. coding potential, synteny information). In particular, about a hundred new small genes were identified, a number of start codons was revised, and several pseudogenes were annotated. Gene products were surveyed using standard approaches (BLAST similarity, protein motifs, but also literature searches), and gene names were revised accordingly, putting emphasis on a systematic and consistent nomenclature. The new annotation information was integrated in an updated version of LegioList. This genome database is now embedded in a multi-genome framework, GenoList (http://genolist.pasteur.fr/GenoList), which holds genome information of more than 700 prokaryotic organisms imported from the Genome Reviews repository (EBI), and manually curated genome data, such as those described here (four L. pneumophila genomes - Paris, Lens, Philadelphia and Corby - accessible at http://genolist.pasteur.fr/GenoList/Legionella). Furthermore, using this multi-genome server, users can store their favorite lists of organisms, also other than Legionella, and access them in any working session. Together with browsing and query capabilities similar to those of the previous version of LegioList, GenoList offers additional functionalities and enables comparative genome analysis. For example, it is possible to identify the genes that are specific for one group of bacteria (e.g. Legionella) but that are absent in a selection of other bacteria, export these genes as a tab-separated list, get their protein sequences, and run a multiple alignment on a subset of these sequences.
- 64 -
SESSION 12 FUNCTIONAL GENOMICS
- 65 - Session 12: Functional genomics Fr. 5:15 pm
Hidden protein treasures revealed in the Legionella genome
Elizabeth Hartland
University of Melbourne, Department of Microbiology and Immunology, 3010 Victoria, Australia [email protected]
As part of its pathogenesis, Legionella invades human alveolar macrophages and epithelial cells where it replicates in a vacuole that evades phagolysosome fusion. The Legionella containing vacuole (LCV) not only avoids interaction with the endocytic pathway but recruits vesicles that traffic in the secretory pathway. The Legionella genome encodes a number of proteins with similarity to eukaryotic proteins that are predicted to interfere with host cell processes. Two of these unusual proteins belong to the CD39/NTPDase1 family of nucleotidases which are found almost exclusively in eukaryotic organisms. In mammals, NTPDases are expressed on the extracellular face of the plasma membrane or the lumenal face of intracellular organelles where they control levels of nucleoside tri and di-phosphates. The extracellular enzymes regulate physiological processes such as thrombosis and inflammation through hydrolysis of extracellular ATP and ADP. The Legionella NTPDase, Lpg1905, plays a role in intracellular replication and hydrolyses ATP/ADP as well as GTP/GDP. To understand the importance of substrate specificity to Lpg1905 function, we determined the crystal structure of Lpg1905 in complex with nucleotide analogues. Based on the structure, we introduced amino acid changes to alter the substrate specificity of the enzyme and found that ATP/GTPase activity appeared to be critically important for replication in macrophages. Although the precise function of Lpg1905 in virulence is not yet fully understood, our work to date has revealed that the protein is a near perfect structural and functional mimic of mammalian NTPDases that may assist Legionella virulence by influencing the mammalian host inflammatory response and/or intracellular organelle integrity.
- 66 - Session 12: Functional genomics Fr. 5:45 pm
Epidemiological genome analysis of a large Dutch Legionella pneumophila strain collection identifies five markers highly correlated with pathogenic strains
Ed Yzermana, Jeroen Den Boera, Martien Caspersb, Arpit Almalc, Bill Worzelc, Walter Van Der Meerd, Roy Montijnb and Frank Schurenb aRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands; bTNO Microbial Genomics, P.O. Box 360, 3700 AJ Zeist, Netherlands; cGenetics Squared Inc, 401 W. Morgan Road, Ann Arbor, AK MI 48108, United States of America; dVitens, Snekertrekweg 61, 8912 AA Leeuwarden, Netherlands [email protected]
Discrimination between pathogenic and non-pathogenic strains within a bacterial species is currently underexplored. Genomic analyses have clearly shown the enormous variability in genome composition between different strains of a bacterial species. In this study we have used Legionella pneumophila, the causative agent of Legionnaire’s disease, to search for genomic markers related to pathogenicity. During a large surveillance study in The Netherlands well- characterized patient-derived strains and environmental strains were collected. We have used a mixed-genome microarray to perform comparative-genome analysis of 257 strains from this collection. Out of 3360 markers 480 DNA markers showing clear variation in presence between individual strains were selected for further analysis. Unsupervised statistical analysis of these markers showed the enormous genomic variation within the species but did not show any correlation with a pathogenic phenotype. We therefore used supervised statistical analysis to identify discriminating markers. Genetic programming was used both to identify predictive markers and to define their interrelationships. A model consisting of five markers was developed that correctly predicted 100% of the pathogenic strains and 62% of the environmental strains. This result was confirmed with a reserved test set not used in creating the model, with a correct prediction of 100% of the pathogenic strains and 69% of the environmental strains. This novel approach for identifying predictive markers for L. pneumophila pathogenicity can be applied to all bacterial species, allowing for better discrimination between strains well equipped to cause human disease and relatively harmless strains.
- 67 - Session 12: Functional genomics Fr. 6:00 pm
Solving functional redundancy amongst effectors of the bacterial pathogen, Legionella pneumophila
Tamara O’Connora, Dana Boydb and Ralph Isberga aHoward Hughes Medical Institute, Tufts University School of Medicine, 150 Harrison Ave, Boston, AK 2111, United States of America; bHarvard School of Medicine, 200 Longwood Ave., Boston, AK 2115, United States of America [email protected]
Legionella infect multiple hosts ranging from amoebae to macrophages. This process is dependent on a type IV secretion system that translocates bacterial proteins, or effectors, to the host cell cytoplasm where they manipulate a variety of host cellular processes. Despite the identification of numerous effectors, their host cell targets and mechanisms of action remain largely unknown because mutants lacking their encoding genes do not exhibit any discernable phenotype. The ability of the bacterium to grow despite the loss of effector function has been attributed to functional redundancy. Thus, defining functionally redundant gene sets is a major obstacle in elucidating the role of these proteins and ultimately, understanding the interplay between the host and bacterium during infection. To address this, we used a microarray-based technique called transposon site hybridization (TraSH) to assess the importance of Legionella genes for intracellular growth in the presence or absence of host factors that mediate the formation of a Legionella replication vacuole. To do this, a library of Legionella mutants was used to infect untreated Drosophila cells or cells individually depleted of vesicle trafficking components using RNA interference. In doing so, we have i) identified host conditions under which Legionella mutants deleted of a single effector-encoding gene exhibit a growth defect; ii) assigned effectors to functional pathways based on the growth of their corresponding mutants under each host condition tested and; iii) used these classifications to identify combinations of genes that, when deleted in pairs, result in growth defects in untreated Drosophila cells. Further characterization of these mutants in natural hosts of Legionella, including amoebae and primary macrophages, demonstrate that functionally redundant relationships between effectors are host cell type specific. This work has established a directed approach for defining functionally redundant proteins employed by a bacterial pathogen and conditions under which the role of these proteins can be further characterized based on phenotypic analysis.
- 68 - Session 12: Functional genomics Fr. 6:15 pm
Analysis of the transcriptome of Legionella pneumophila under infection conditions using RNA.seq
Barbara Weissenmayer, James Prendergast, Amanda Lohan and Brendan Loftus
UCD Conway Institute for Biomolecular and Biomedical Research, Belfield, 4 Dublin, Ireland [email protected]
An important component in the evolution of virulence in Legionella species has been its association with free-living amoebae such as Acanthamoeba castellanii in the environment, where it is thought to have co-evolved with this important natural host. We have investigated the infection of Acanthamoeba castellanii with Legionella pneumophila Philadelphia 1 using next generation sequencing in order to provide a thorough and quantitative assessment of the Legionella transcriptional response. We find that approximately 90% of the genome is expressed under 37C infection conditions. Significant expression differences in genes coding for metabolic and signalling pathways are observed underlying the switch from a replicative to a transmissive phase. We confirm and expand previous findings from a microarray based study. Transcripts expressed during the replicative stage were mainly involved in energy production and conversion, translation and biogenesis, and amino acid transport. During the transmissive phase a strong up-regulation of genes important for cell motility and signal transduction was observed. Interestingly, at late infection time points enzymes necessary for the transposition of mobile genetic elements are also up-regulated. An advantage of the use of RNA.seq is the ability to not only monitor changes at the protein coding level but also characterize extensive expression difference within non-coding regions of the genome. We use this data to map operon structures, putative riboswitches and a number of different regulatory RNAs associated with infection in a eukaryote host. We identify two regions of approximately 47Kb and 16kb respectively sufficiently different from the published genome sequence to represent evidence for a possible genetic transfer event. Both regions appear to contain Legionella virulence determinants, in one case elements of the Dot/Icm secretion system, and the other an operon coding genes for effectors required for efficient recruitment of Endoplasmic Reticulum proteins.
- 69 -
SESSION 13 GENE REGULATION
IN THE ENVIRONMENT AND THE HOST
- 70 - Session 13: Gene regulation in the environment and the host Sa. 9:00 am
The Flagellar Regulon of Legionella pneumophila and the Expression of Virulence Traits
Christiane Albert-Weissenbergera,b, Tino Schulzc, Tobias Sahra, Carmen Buchriesera and Klaus Heunerc aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bInstitut fur¨ Molekulare Infektionsbiologie, Rontgenring¨ 11, 97070 Wurzburg,,¨ Germany; cRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany [email protected]
Background: Bacterial flagella are highly complex molecular machines. They are surface organelles that are made up of over 40 different protein components that mediate bacterial motility. To ensure maximal efficiency and accuracy during flagellar biogenesis, bacteria use hierarchical regulatory networks to control the ordered expression of the individual components of the flagellar organelle. Expression of the flagellum is known to be genetically linked to the expression of virulence traits and is modulated by different environmental factors. In order to obtain a better understanding of the flagellar regulon of L. pneumophila, we further evaluated the role of the regulatory proteins FleQ, FleR, RpoN and FliA. Methods: In silico analysis of the four L. pneumophila genomes sequenced revealed 45 genes organized in 10 genomic regions were predicted to participate in the flagellar regulon. The mutants generated (fleQ, rpoN, fleR, fliA of L. pneumophila Paris) were analysed by phenotypic investigations, transcriptome analysis, Western blot analysis and electron microscopy.Results: In the FleQ, FleR, and RpoN mutant strains the expression of the flagellin is repressed and the mutants are non-flagellated. Together with RpoN, FleQ enhances transcription of about 22 out of 45 flagellar genes, which code for the basal body, hook, and regulatory proteins. Unexpectedly, FleQ but not RpoN enhances the expression of the fliA gene. FliA activates the flagellar class IV genes. It is also obvious, that there are some differences between the strain L.pneumophila Corby and Paris. Further details of the flagellar regulon and of regulators involved in virulence gene expression will be discussed.Conclusions: Our model for the transcriptional regulation of flagellar genes has now been modified. The class II genes are controlled by FleQ and RpoN, whereas expression of the flagellar class III gene fliA is influenced by a FleQ-dependent but RpoN-independent manner. However, FleR did not induce the transcription of class III genes. FliA induces expression of the flagellar class IV genes leading to the complete biosynthesis and assembly of a functional flagellum. Our results indicate that FleQ of L. pneumophila regulates gene expression in an RpoN-dependent as well as RpoN-independent manner.
- 71 - Session 13: Gene regulation in the environment and the host Sa. 9:30 am
Two small ncRNAs jointly govern virulence and transmission in Legionella pneumophila
Tobias Sahra, Holger Bruggemann¨ a, Matthieu Julesa, Christel Cazaleta, Mariella Lommaa, Christiane Albert- Weissenbergera,b and Carmen Buchriesera aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bInstitut fur¨ Molekulare Infektionsbiologie, Rontgenring¨ 11, 97070 Wurzburg,,¨ Germany [email protected]
Legionella, the single genus of the family Legionellaceae, are gram-negative, rod-shaped bacteria, naturally found in freshwater habitats and moist soils. As facultative pathogens, they can persist in the environment in biofilms, but grow and replicate intracellular in its natural hosts, mainly free-living protozoa like A. castellani. As adaptation to the alternation between intra- and extracellular environments, L. pneumophila developed two discrete physiological stages: a replicative form, non-virulent, but proficient in avoiding the host cells defense mechanisms and a transmissive form that is motile, resistant to environmental stress and highly infectious. To control the transition and hence expression of a variety of virulence-linked traits, Legionella is using mainly the two-component system LetA/LetS and the global repressor protein CsrA. Trying to determine, how both regulators act coordinately to govern switching between these two physiological states, we constructed letA and letS mutants and investigated the molecular mechanisms underlying this differentiation process. While replication was independent of LetA/LetS, the phenotypic and the transcriptional pattern of the mutants bring evidence that the two component system is indispensable to convert Legionella into a fully virulent and motile form. Furthermore, we show for the first time that L. pneumophila employs small ncRNAs, RsmY and RsmZ, that link the LetA and CsrA regulatory network. Single mutants have no (rsmY) or less pronounced (rsmZ) impact on virulence, whereas the rsmYZ double mutant shows a drastic defect in intracellular growth and transmission to a new host. Electrophoretic mobility shift assays demonstrate that LetA interacts with a specific DNA sequence upstream of the ncRNAs region and that both small ncRNAs bind CsrA in vitro. Thus, we conclude that LetA - and in lesser amount also RpoS - activates transcription of RsmY and RsmZ, which then sequester CsrA to abolishing its post-transcriptional repressive activity. However, the RsmYZ-CsrA-pathway seems to have no influence on the flagella synthesis. We suggest that rather, LetA independent of ncRNA regulate motility in L. pneumophila by influencing LetE, and probably cyclic-di GMP levels.
- 72 - Session 13: Gene regulation in the environment and the host Sa. 9:45 am
Parasexual behavior in response to genotoxic stress in Legionella pneumophila
Xavier Charpentiera and Howard Shumanb aDepartment of Microbiology, Columbia University Medical Center, 701 West 168th Street, New York, AK NY 10032, United States of America; bColumbia University Medical Center, Department of Microbiology & Immunology, 701 West 168th Street, New York, NY 10032, United States of America [email protected]
Chromosomal DNA is continuously exposed to damage, repair and replication. Many bacteria respond to DNA damage by activating a regulatory network leading to induction of a DNA repair system. This phenomenon, called the SOS response, also triggers a transient mutagenic state that may be important for adaptive evolution. The SOS response has been found in most gram-positive and gram-negative bacteria. However, genetic and phenotypic evidence suggest that Legionella pneumophila lacks a prototypic SOS response. When exposed to UV radiation L. pneumophila responds by a limited increase in mutagenesis. However, UV radiation strongly induces competence, a genetically programmed physiological state that confers the ability to take up DNA from the environment. Competence in L. pneumophila is also strongly stimulated by other genotoxic stresses including antibiotics of the fluoroquinolone family, used for the treatment of L. pneumophila infections. Competence and subsequent bacterial transformation increase genetic diversity and purge detrimental mutations, thus serving a function similar to meiotic sex in higher organisms. By responding to genotoxic stress with limited mutagenesis and the induction of competence, L. pneumophila favors genetic diversity over genetic variability. This parasexual strategy may have enabled L. pneumophila to acquire foreign genes and contributed to emergence of this human pathogen.
- 73 -
SESSION 14 CELL BIOLOGY PROTOZOA / MACROPHAGES
- 74 - Session 14: Cell biology - protozoa - macrophages Sa. 10:45 am
Establishment of a vacuole that supports Legionella pneumophila replication.
Craig Roy
Yale University, 295 Congress Ave, BCMM, rm 347, New Haven, CT 06536-0812, United States of America [email protected]
The vacuole containing Legionella pneumophila undergoes a series of maturation events. Initially derived from the plasma membrane, this vacuole interacts with host vesicles and organelles through a series of reactions that are facilitated by the Dot/Icm secretion system. Host and bacterial proteins that are involved in these maturation processes will be discussed. Special emphasis will be placed on effectors proteins translocated into host cells by the Dot/Icm system. Host substrates engaged by these effectors will be presented as will the function of those host proteins. Spatial and temporal control mechanisms used by effectors to localize to vacuoles will also be discussed.
- 75 - Session 14: Cell biology - protozoa - macrophages Sa. 11:15 am
Exploitation of conserved eukaryotic pathways by L. pneumophila
Chris Pricea, Souhaila Al-Khodora, Tasneem Al-Quadana, Marina Santicb, Fabien Habyarimanaa, Awdhesh Kaliab and Yousef Abu Kwaika aUniversity of Louisville, 319 Abraham Flexner Way Rm 410, Louisville, KY 40202, United States of America; bUniversity of Rijeka, Brace Branchetta 20, 51000 Rijeka, Croatia [email protected]
Molecular Mimicry by an F-box effector of Legionella pneumophila hijacks a conserved polyubiquitination machinery within macrophages and protozoa. The ability of Legionella pneumophila to proliferate within various protozoa in the aquatic environment and within macrophages indicates a remarkable evolution and microbial exploitation of evolutionarily conserved eukaryotic processes. Ankyrin B ( AnkB) of L. pneumophila is a non-canonical F-box-containing protein and is the only known Dot/Icm-translocated effector of L. pneumophila essential for intra-vacuolar proliferation within both macrophages and protozoan hosts. We show that the F-box domain of AnkB and its two L9P10 conserved residues are essential for intracellular bacterial proliferation and for rapid acquisition of polyubiquitinated proteins by the Legionella -containing vacuole (LCV) within macrophages, Dictyostelium discoideum, and Acanthamoeba. Interestingly, translocation of AnkB and recruitment of polyubiquitinated proteins in macrophages and Acanthamoeba is rapidly triggered by attached extracellular bacteria within 5 min of bacterial attachment. Ectopically expressed AnkB within mammalian cells does not associate with endosomal, lysosomal, ER, or golgi compartments but is localized to the periphery of the cell. Interestingly, ectopically expressed AnkB in mammalian cells recruits polyubiquitinated proteins to the periphery of the cell where AnkB is localized, and restores intracellular growth to the ankB mutant. While ectopically expressed AnkB−9L10P/AA variant is localized to the periphery of the cell, it does to recruit polyubiquitinated proteins and fails to trans-rescue the ankB mutant for its intracellular growth defect. Direct in vitro interaction of AnkB with the host SKP1 component of the ubiquitin ligase complex is demonstrated. Importantly RNAi-mediated silencing of expression of SKP1 renders the cells non-permissive for intracellular proliferation of L. pneumophila. The role of AnkB in exploitation of the polyubiquitination machinery is essential for intrapulmonary bacterial proliferation in the mice model of Legionnaires’ disease. Therefore, AnkB exhibits a novel molecular and functional mimicry of eukaryotic F-box proteins to exploit an evolutionarily conserved polyubiquitination machinery for intracellular proliferation within evolutionarily distant hosts and to manifest disease in mammals.
- 76 - Session 14: Cell biology - protozoa - macrophages Sa. 11:45 am
Doubling up as a way to combat resilience
Ralph Isberga, Tamara O’Connora, Yewande Adepojub and Matthew Heidtmana aHoward Hughes Medical Institute, Tufts University School of Medicine, 150 Harrison Ave, Boston, AK 2111, United States of America; bTufts University School of Medicine, 150 Harrison Ave, Boston, AK 2111, United States of America [email protected]
Over 190 substrates of the Legionella pneumophila type IV Dot/Icm system have been identified. Deletion of genes for individual substrates usually have little effect on the efficiency of intracellular growth in either macrophages or amoeabae. In fact, a strain that is competent for intracellular growth has now been constructed that is deleted for over 60 of these substrates. Lack of phenotypes has hampered the investigation of the role of these proteins in intracellular growth. The absence of phenotype has been explained by potential functional redundancy among the substrates. The accepted solution to functional redundancy is to construct strains that lack multiple genes and demonstrate phenotypes that require the simultaneous knockout of these genes. Choosing combinations of genes among 190 or more candidates is extremely difficult. To overcome this problem, we have developed a strategy to generate phenotypes through combining siRNA analysis of host cells with insertion mutagenesis. This has allowed the identification, by cluster analysis, of L. pneumophila proteins that appear to act in the same biochemical pathway, and to construct strains that are defective for intracellular growth due to the simultaneous loss of translocated substrates that act on parallel biochemical pathways.
- 77 -
POSTERS
- 78 -
SESSION 1 EPIDEMIOLOGY AND CLINICAL ASPECTS A) DISEASE, DIAGNOSIS AND TREATMENT
- 79 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P1
Mutational paths towards increased fluoroquinolone resistance in Legionella pneumophila
Iyad Almahmoud, Elisabeth Kay, Dominique Schneider and Max Maurin
LAPM, CNRS UMR5163, Universite´ Joseph Fourier, Institut Jean Roget, 38042 Grenoble, France [email protected]
Fluoroquinolone resistance has been poorly studied in Legionella pneumophila, an intracellular pathogen responsible for legionellosis. Our goal was to further characterize molecular mechanisms involved in fluoroquinolone resistance in this species. Eight independent lineages were founded from a common fluoroquinolone-susceptible L. pneumophila ancestor and propagated by serial passages in moxifloxacin-containing culture medium. We identified the substituted mutations that affected the DNA topoisomerase II-encoding genes, determined the order of substitution of the mutations leading to the stepwise MIC increases of moxifloxacin over evolutionary time and demonstrated their direct involvement in the resistance process. Adaptation occurred through parallel stepwise increases in the moxifloxacin MICs up to 512- fold the MIC for the parental strain. Mutations affected the topoisomerase II-encoding genes gyrA, parC and gyrB, reflecting a high degree of genetic parallelism across the independent lineages. During evolution, the T83I change in GyrA occurred first, followed by G78D or S80R in ParC and D87N in GyrA, or S464Y or D426N in GyrB. By constructing isogenic strains, we showed that the progressive increase in resistance was linked to a precise order of mutation substitution, but also to the coexistence of several subpopulations of bacteria bearing different mutations. In conclusion, specific mutational trajectories were identified, strongly suggesting that intermolecular epistatic interactions between DNA topoisomerases underlie the mechanism of fluoroquinolone resistance in L. pneumophila. Our results suggest that L. pneumophila has strong potential to become resistant to fluoroquinolone compounds and warrant further investigation of resistance in clinical and environmental strains of this pathogen.
- 80 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P2
Evaluation of a fluorescence-based method for the rapid detection of Legionella in environmental water samples
Cristina Gruasa, Isidro Alvarez´ b, Carlos Laraa, Cristina Garc´ıaa, Demetris Savvac and M. Victoria Arrugaa aUniversity of Zaragoza, Miguel Servet, 177, 50013 Zaragoza, Spain; bCobrial Laboratory, Plaza Extremadura, 4 Bis, 22004 Huesca, Spain; cUniversity of Reading, Whiteknights, P.O. Box 221, RG6 6AS Reading, United Kingdom [email protected]
The method considered as the ”gold standard” for the detection and characterisation of Legionella suffers from a number of drawbacks such as the long time necessary for the cultivation of the organism, the fact that there are cells that may be infectious but can not be cultured and the fact that it may be difficult to isolate the organism from contaminated samples. A number of alternative methods have been proposed to overcome these problems including the use of methods based on the polymerase chain reaction and the use of probes. This study compared the traditional culture method as used in public health laboratories with a method based on the use fluorescent probes available and commercially as a kit (ScanVIT-Legionellao).ˆ The results demonstrate the usefulness of the ScanVIT- Legionellaoˆ method as a rapid diagnostic tool despite the fact that any Legionella detected is not isolated for further characterisation.
- 81 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P3
Legionella infection - a survey in the Greater Copenhagen area from 2000-2007
Nicolai D. Jørgensen, Jette M. Bangsborg and Ida Gjørup
Copenhagen University Hospital Herlev, Herlev Ringvej 75, 2730 Herlev, Denmark [email protected]
During 2000-2007, 123 patients, from whom clinical information was available, were diagnosed with Legionella pneumonia (LP) in three major hospitals in the Greater Copenhagen area. Diagnostic criteria were a positive PCR in a respiratory sample (111 patients (90 %), of which 48 cases (43 %) were culture-positive), and/or a positive urinary antigen test (LUT) (49 patients (40 %)), and symptoms compatible with pneumonia. Male : female ratio was 1.67, mean age 64.7 years (range 25-88), and the overall 30-day mortality 21 % . Thirty-seven cases (31 %) were verified or possible nosocomial cases, 19 (16 %) were travel-associated, and 64 (53 %) cases were community-acquired. Sixty-seven patients (55 %) required ventilatory and supportive treatment in the Intensive Care Unit. Sixty-three (54 %) patients received combination therapy with a macrolide plus a fluorquinolone. Duration of fever and hospitalization, and outcome in relation to treatment modalities awaits further analysis. Preliminary data, however, suggests that although general recommendations for the treatment of LP were in effect, frequent changes in antibiotic therapy for the individual patient occurred, hampering comparisons of the efficacy of recommended regimens. In conclusion, LP remains a relatively rare, but serious disease in our area, with a high frequency of nosocomial cases until 2006. Epidemiological data concerning age, risk factors etc. is concordant with results from other geographical areas. However, non-L. pneumophila serogroup 1 infections constitute more than 50 % of the culture-verified cases, in contrast to other studies.
- 82 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P4
Atypical Legionella maceachernii Infection in a Severly Immunocompromised Patient.
Vladimir Drasara and Tomas Szotkowskib aNational Legionella Reference Laboratory, Public Health Institute, Masarykovo namesti 16, 682 01 Vyskov, Czech Republic; bDepartment of Hemato-Oncology, University Hospital, I.P.Pavlova 6, 775 20 Olomouc, Czech Republic [email protected]
Background. L.maceachernii is a species that is rarely encountered in both environmental and clinical samples.Since 1985, only 5(4 fatal)cases have been reported outside Europe.All were individuals with severe pneumonia and underlying immune defects. We present the first culture proven case of atypical infection in Europe. Introduction. A 29- year-old male patient was admitted to hospital with symptoms of progressive dyspnoea ans wheezing cough in March 2007.Two years ago, he underwent allogenic hematopoietic stem cell transplantation due to polycythemia vera. Long- term immunosuppresive therapy necessary to treat extensive chronic graft-versus-host disease(GvHD) led to frequent and severe infectious complications. Results. Computed tomography revealed bilateral pulmonary infiltrates suspected to be of fungal origin. Bronchoalveolar lavage(BAL) was performed and voriconazole administered.Routine examination of the BAL fluid yielded Aspergillus spp.To our surprise, selective culture for Legionella showed several colonies of L. maceachernii after 7 days.The finding was later serologically confirmed by a four fold rise in specific L.maceachernii titres.Levofloxacin was therefore added to the therapy for safety’sake(10 days after the BAL was performed).A month later, pulmonary infiltrates persisted without significant improvement. Severe sepsis(caused by Staphylococcus aureus and polyresistant Pseudomonas aeruginosa ) developed subsequently, treated with meropenem and vancomycin.One month later,CT showed improvement of lung infiltrates.Their aetiology remained unclear. An immediate environmental investigation was carried out in 3 locations where the patient had lived or stayed, including the hospital.L. maceachernii was not found in any of the potable water systems tested. Conclusion. L.maceachernii causes severe pneumonia. Despite severe immunosuppresion, the patient did not develop typical symptoms of legionella pneumonia infection as reported in the literature. The role of L. maceachernii detected with Aspergillus spp. in the pulmonary infiltrates remains unexplained. L.maceachernii is a rare species in the Czech environment. Over the last 20 years, it has been isolated only twice.From hot water of a big hotel and from thermal water used for medicinal purposes.
- 83 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P5
Early ”point-of-care” diagnosis of Legionella infection by detection of IgM with Lateral flow test
Pernille Elverdala, Ian C. Skovstedb, Nina Garrettc, Iain Hargravec and Søren A. Ulduma aStatens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark; bStatens Serum Institut, SSI Diagnostica, Herredsvejen 2, 3400 Hillerød, Denmark; cBBInternational, Golden Gate, Ty Glas Avenue, Llanishen, CF14 5DX Cardiff, United Kingdom [email protected]
The aim with this study was to evaluate a lateral flow prototype for detection of IgM antibodies to Legionella pneumophila (Lpn) serogroup (sg) 1 and 3. Materials and Methods: Sera from patient with Legionnaires’ disease (LD): 124 sera from 82 patients with Lpn sg 1 infection and 18 sera from 10 patients with Lpn sg 3 infection. Sera from patients without LD: 88 sera, which have been tested positive for antibodies to other bacteria than Lpn, and 51 sera from blood donors. All sera were tested in a lateral flow prototype for detection of IgM antibodies against Lpn sg1 and Lpn sg 3. For comparison all sera were also tested by an in-house immunofluorescence antibody test (IFAT) and an in-house Lpn sg 1 and 3 IgM ELISA. Results and conclusion: The results from the 124 sera from the LD patients were divided into four time intervals according to days after onset of symptoms (1-7, 8-14, 15-21 and 22 days or more). Under these conditions it was found that the lateral flow test had the highest sensitivity of the three assays in all four time intervals for both the sg 1 sera (30; 65; 91 and 75%), and the sg 3 sera (83; 80; 100 and 100%). The three assays were also tested for false positive results. It was found that the lateral flow prototype had an overall false positive rate of 14% for the sg1 test and 8% for the sg 3 test. The bacterium causing the false positive background was Campylobacter jejuni/coli. Among the blood donors only 5-7% had a weak reaction in the test. From this study we can conclude that it is possible to achieve a high sero-positive percent early in the course of the disease by specific detection of IgM antibodies by lateral flow as 39 % of the LD patients were positive within seven days after onset of symptoms. The lateral flow prototype should be optimised by blocking some of the false positive results. This lateral-flow assay could emerge as the future ”point-of-care” Lpn serological test Additional data will be presented
- 84 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P6
In-house ELISA for detection of IgM and IgG to Legionella pneumophila serogroup 1, 3 and 6 as a routine test
Pernille Elverdal, Charlotte S. Jørgensen and Søren A. Uldum
Statens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark [email protected]
The aim with this study was to evaluate an in-house enzyme linked immunosorbent assay (ELISA), as a routine test for detection of IgM and IgG antibodies to Legionella pneumophila (Lpn) serogroup (sg) 1, 3 and 6. Methods: 4520 sera from 3824 patients have been tested in the in-house ELISA over a period of 10 months. Ninety-five of these sera were from 57 patients with confirmed Legionnaires’ disease (LD). 295 sera from blood donors were also tested to estimate the false positive ratio. All sera were tested in an in-house ELISA with lipopolysaccharid (LPS) extracts from Lpn sg 1, 3 or 6 as antigens and with conjugates to human IgG and IgM. For IgM testing all sera were treated with goat anti-human IgG (Biodesign International L15406G) to absorb IgG. Results and conclusion: The results from the 4520 sera showed that 6% of the 3824 patients were positive for Lpn antibodies. Approximately 4% of the 3824 patients were positive for IgM Lpn antibodies and 3% were positive for IgG Lpn antibodies. Approximately 1% of the blood donor sera were positive for IgM antibodies, whereas 2% were positive for IgG antibodies. The results from the 57 confirmed LD patients showed that 74% of the patients were positive for IgM (67%) or IgG (46%) antibodies to Lpn. The samples from the 57 confirmed LD patients were divided into four time intervals, according to days after onset of symptoms (1-7, 8-14, 15-21 or more than 21 days). Samples from each patient can only occur one time in each interval and not all patients had samples in each interval. It was found that the in-house ELISA had a high sensitivity in the last two time intervals of 82% or more, but even within the first week of onset of symptoms 19% of 16 LD patients were positive in the ELISA test, all were positive for IgM antibodies only. In the second week after onset of symptoms 54% of 24 LD patients were positive for antibodies to Lpn, 50% of these were IgM positive and 21% were IgG positive. After three weeks 82% were IgM positive and 53% were IgG positive (17 samples). Within the fourth week the IgM positive starts to decrease (79%) and the IgG positive starts to increase (63%, 24 samples). If the IgM and IgG positive samples are combined we get a sensitivity of 92% after four weeks or more. From this study we can conclude that it is possible to achieve a high sero-positive percent early in the course of the disease by detection of IgM and IgG antibodies separately in an in-house ELISA. IgM antibodies can be detected early, but we cannot exclude the IgG test, because this study has shown that some patients only develop IgG antibodies. Additional data will be presented
- 85 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P7
Travel associated legionnaires disease: Clinical features of 17 cases
Haluk Erdogana, Askin Erdoganb, Huseyin Lakadamyalic, Aynur Yılmazd and Hande Arslane aBaskent University Medicine Faculty, Department of Infectious Disease and Clinical Microbiology, saray mh, kızlarpınarı cd,no:1, 7400 Alanya, Turkey; bBaskent University Medicine Faculty, Department of Gastroenterology, saray mh, kızlarpınarı cd, no:1, 7400 Alanya, Turkey; cBaskent University Medicine Faculty, Department of Pulmonary Disease,, saray mh, kızlarpınarı cd,no:1, 7400 Alanya, Turkey; dBaskent University Medicine Faculty, Department of Neurology, saray mh, kızlarpınarı cd,no:1, 7400 Alanya, Turkey; eBaskent University Medicine Faculty, Department of Infectious Diseases and Clinical Microbiology, Fevzi C¸akmak Caddesi, 10. Sok. No:45 Bahc¸elievler, 6490 Ankara, Turkey [email protected]
Legionnaires’ disease is a systemic infectious disease that primarily involves the lungs and has a mortality rate of 5% to 30. In this report, which to our knowledge is the first series reported from Turkey, we present 17 cases of travel-associated legionnaires’ disease. All patients studied (mean age, 61.1 ± 9.5 years; range, 39-77 years) were tourists. Diabetes mellitus was found in 7 of those patients and 6 patients smoked cigarettes. Twelve patients had gastrointestinal symptoms such as diarrhea, loose stools, abdominal pain, nausea, vomiting. A change in mental status or headache were found in 10 patients. Relative bradycardia was found in 9 patient. Hepatic dysfunction, hyponatremia, hypophosphatemia and creatine phosphokinase elevation were the most common laboratory findings. Legionnaires’ disease was severe in 11 patients which required intensive care unit. The main diagnostic method was Legionella urinary antigen test. Diagnose was also confirmed by culture in 3 patients. Appropriate antibiotic therapy was initiated in all patients on admission day. Four of the 17 patients died. In conclusion, clinicians should remain vigilant about the diagnosis of LD in patients with severe community-acquired pneumonia, especially in the presence of extrapulmonary involvement, risk factors for LD, and history of recent travel. As our cases, mortality is still high in sporadic cases despite early appropriate treatment.
- 86 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P8
Legionnaire’s disease presenting with travellers diarrhea and resulting acute respiratory distress syndrome: A case report
Haluk Erdogana, Askin Erdoganb and Hande Arslanc aBaskent University Medicine Faculty, Department of Infectious Disease and Clinical Microbiology, saray mh, kızlarpınarı cd,no:1, 7400 Alanya, Turkey; bBaskent University Medicine Faculty, Department of Gastroenterology, saray mh, kızlarpınarı cd, no:1, 7400 Alanya, Turkey; cBaskent University Medicine Faculty, Department of Infectious Diseases and Clinical Microbiology, Fevzi C¸akmak Caddesi, 10. Sok. No:45 Bahc¸elievler, 6490 Ankara, Turkey [email protected]
Introduction: Acute respiratory distress syndrome (ARDS) is characterized by the acute onset of bilateral infitrates on chest radiography and the development of breathlessness within hours to days of an inciting event. The most common risk factor of ARDS is severe sepsis and the prognosis is extremely poor. Here, we report a case of ARDS associated with legionnaires disease. Case report: A 54-year-old male tourist with diabetes mellitus presented at our institution with diarrhea and high fever. He had watery diarrhea with 10 times in previous day. At the time of admision, the patient had a temperature of 39.5◦C, a blood pressure of 140/70 mm Hg, a heart rate of 102 beats/min, a respiratory rate of 24 breaths/min. His oxygen saturation was 93% while breathing room air. Chest examination revealed ralles and the chest x-ray showed right-lower zone infiltrates. Legionella urinary antigen was positive on admission day, so the patient was diagnosed as having legionnaires disease. Treatment with levofloxacin (1000 mg/day) was started in the first 4 hours on admission after taking suputum cultures. The patient had progressive dyspnea and cyanosis was developed. Analysis of arterial blood gas revealed a pH of 7.29, a CO2 pressure of 30 mm Hg, an O2 pressure of 37 mm Hg, and an oxygen saturation of 62% with 5 L/min nasal oxygen support. Thirty-two hours after admission, the patient was admitted to the medical intensive care unit for ventilator support. Roentgenogram of the chest worsened and showed bilateral infiltration, indicating ARDS. Rifampicin (600 mg/day) was added levofloxacine. During the course of several days, the patient’s clinical status continued to deteriorate. He was hypotensive and required dopamine infusions for blood pressure support. Five days later, vancomycin (2 gr/day), cefepearazone-sulbactame (6 gr/day), chlaritromycine (1000 mg/day) and ciprofloxacine (1200 mg/day) were substituted for the initial antibiotics. Subsequent days clinical status improved and the body temperature was below 37.5◦C on the eleventh day. Legionella pneumophila serogroup 1 isolated on sputum culture. This strain was susceptible to levofloxacine (<0.003 mg/dL). On the patient’s request, he was flown back to his country by air ambulance. One year later, the patient again came back as a turist and fully recovered from the illness. Conclusions: ARDS may complicate the course of legionnaires disease. Early appropiate therapy can be life saving.
- 87 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P9
Legionella Source Identification in the Netherlands: 2007-2008
Sjoerd Eusera, Petra Brandsemab, Helma Ruijsb, Ed Yzermana, Jacob Bruina, Sacha Bleekera and Jeroen Den Boera aRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands; bNational Institute for Public Health and the Environment, Antonie van Leeuwenhoeklaan 9, 3721 MA Biilthoven, Netherlands [email protected]
Background: The Legionella Source Identification Unit was installed in 2002 as part of a national outbreak detection programme in the Netherlands. The aim of the project is to improve source identification and thereby preventing outbreaks of LD by swift elimination of the source. Objective: Describing the sampling results of potential sources of LD by the Legionella Source Identification Unit in 2007 en 2008 in the Netherlands. Method: Treating physicians reported incident cases of LD to the Municipal Health Services (MHSs), which subsequently identified potential sources of LD by interviewing the patient (or his/her relatives). Patients who stayed abroad for five or more days of their 2-10 day incubation period were (arbitrarily) not considered to have a Dutch source of infection and were excluded from the sampling procedure. The Legionella Source Identification Unit sampled the potential sources of LD when at least one of four criteria was fulfilled: (1) the source was linked to another LD patient in the past ; (2) the LD patient was part of a geographical cluster (3 or more patient within an radius of 1 km) ; (3) the LD patient resided in a hospital or nursery home in the incubation time ; (4) Legionella spp. was isolated from respiratory secretions or long tissue of the LD patient. The collected samples were concentrated by filtration and resuspention and cultured on BCYE agar. Legionella isolates were serotyped and Amplified Fragment Length Polymorphism (AFLP)- and Sequence Based Typing (SBT)-techniques were used to genotype the environmental strains. These genotyping results were then compared with those of the patient isolate, when available. Results: Legionella spp. was identified in the environmental samples of 73 (32%) of the 229 potential sources of LD that were investigated. Legionella pneumophila was found in 38 (17%) of the sources, Legionella non-pneumophila was found in 35 (15%) of the sources. In six patients, the Legionella pneumophila strains isolated from respiratory secretions or long tissue were genetically indistinguishable from the strains that were identified in the environmental samples. Conclusion: Almost one third of all sampled potential sources in 2007 and 2008 were positive for Legionella spp. However, in the majority of Legionnaires’ disease patients, no source could be identified. This indicates that increased efforts on source identification for Legionnaires’ disease are needed.
- 88 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P10
Saunas and Legionnaires’ disease in the Netherlands
Sjoerd Eusera, Wim Van Der Hoekb, Ed Yzermana, Jacob Bruina, Sacha Bleekera and Jeroen Den Boera aRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands; bNational Institute for Public Health and the Environment, Antonie van Leeuwenhoeklaan 9, 3721 MA Biilthoven, Netherlands [email protected]
Background: Sauna bathing is considered safe and relaxing and has been suggested to inhabit health benefits for visitors with cardiovascular, dermatological or infectious diseases. However, visiting saunas could also comprise certain health risks, for example with respect to Legionnaires’ disease (LD). The Legionella Source Identification Unit, as part of a national outbreak detection programme in the Netherlands investigated several saunas that were identified as potential sources of LD. Objective: Describing the sampling results of the Legionella Source Identification Unit with respect to saunas as potential sources of LD in the Netherlands in the period 2002-2008. Methods: Treating physicians reported incident cases of LD to the Municipal Health Services, which subsequently identified potential sources of LD by interviewing the patient (or his/her relatives). Patients who stayed abroad for five or more days of their 2-10 day incubation period were (arbitrarily) not considered to have a Dutch source of infection and were excluded from the sampling procedure. The Legionella Source Identification Unit sampled the water from potential sources of LD. Collected samples were cultured and serotyped to identify the presence of Legionella isolates. Amplified Fragment Length Polymorphism (AFLP)- and Sequence Based Typing (SBT)-techniques were used to genotype the environmental strains. These genotyping results were then compared with those of the patient isolate(s), when available. Results: The Legionella Source Identification sampled 15 saunas that were identified as potential sources of infection for a total of 27 LD patients. Of these saunas, 12 (80%) were positive for Legionella spp.: 9 (60%) for Legionella pneumophila ; 3 (20%) for Legionella non-pneumophila. In two saunas, the Legionella pneumophila strains that were retrieved from the collected samples were genotypically indistinguishable from the patient isolates. Conclusion: Saunas are not merely the health promoting facility they are often held for, but are also potential sources for Legionnaires’ disease. Increased attention of health services for this questionable role of saunas could increase the number of identified sources of Legionnaires’ disease, as many sources remain unknown at the moment.
- 89 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P11
Variations in Legionella seroprevalence across Australia - a national serology study
Jessica Gorman, Paul Van Buynder, Chantal Ferguson, Angus Cook, Thomas Riley and Philip Weinstein
School of Population Health, University of Western Australia, 35 Stirling Highway, Crawley, Western Australia, 6009 Perth, Australia [email protected]
Background In contrast to the worldwide incidence of Legionnaires’ disease, where Legionella pneumophila causes approximately 90% of cases, Legionella longbeachae infections in Australia occur as frequently as L. pneumophila infections. However, there is not a uniform pattern of notified disease across the country. The more westerly states of Australia, Western Australia and South Australia consistently report greater rates of legionellosis from L. longbeachae infection, whereas states on the Eastern side of Australia (Victoria, New South Wales and Queensland) have a predominance of cases attributed to L. pneumophila. To assess the community prevalence of Legionella in Australia and compare this to notification rates, a national serosurvey was conducted. Methods A multistate, community-based seroprevalence study was conducted on 1004 blood donor volunteers. Laboratories examined for evidence of exposure of Legionella spp. Data were analysed to determine any differences in laboratory results between states and compared to notification rates. Results In 2006 the notification rate of legionellosis in Australia was 1.4 per 100,000 population, a total of 349 cases nationwide. The state of Victoria had 69 cases with a rate of 0.8 per 100,000 population. Of these cases in Victoria, 13 were attributed to L. longbeachae infection and 51 to L. pneumophila infection, showing a predominance of L. pneumophila infection in this State. The serosurvey results in Victoria show that out of 201 participants, only 5 individuals had evidence of exposure to L. pneumophila while 33 participants showed evidence of exposure to L. longbeachae. Conclusion The serosurvey results differ substantially from the pattern that would have been anticipated by the State notification rates. In Victoria, due to the legionellosis notification rates, it would be expected that seroprevalence of L. pnuemophila be significantly higher than the seroprevalence of L. longbeachae, however this hypothesis was not supported. The discrepancy between the seroprevalence and notification rates warrants further investigation. One suggested hypothesis would be that testing patterns and laboratory procedures for clinical disease differs between states, leading to a difference in notifications from the different serotypes.
- 90 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P12
Epidemiological Surveillance of Legionnaires’ Disease in Portugal, 2004-2008
Teresa Fernandesa, Carlos Gomesa, Judite Catarinoa, Cristina Furtadob, Maria-Jesus Chasqueirac, Marta Nascimentod, Joao˜ Felicianoa, Jose´ Giriaa and Teresa Marquesc aDirecc¸ao-Geral˜ da Saude,´ Alameda D. Afonso Henriques, 45, 1049-005 Lisbon, Portugal; bInstituto Nacional de Saude´ Dr. Ricardo Jorge, IP, Avenida Padre Cruz, 1649-016 Lisbon, Portugal; cFaculdade de Cienciasˆ Medicas,´ Microbiology Department, Chronic Diseases Research Centre-CEDOC, Campo Martires´ da Patria,´ 130, 1169-056 Lisboa, Portugal; dHospital de Santa Cruz, Av. Prof. Reinaldo dos Santos, 2790-134 Carnaxide, Portugal [email protected]
Objectives: Legionnaires’ Disease (LD) was included in the Portuguese Mandatory Notification Communicable Diseases Surveillance System (DDO) in December 1998. In April 2004, the Legionaries’ Disease Integrated Epidemiological Surveillance Program (Program) was created to improve the diagnosis, notification and epidemiological investigation of cases, promoting environmental studies and molecular characterization of Legionella strains isolated. The results of the first five years of the Program are presented. Methodology: To analyse the long-term trends (2000-2008) and performance of the Program, it was compared with the DDO and the hospital discharges (GDH) databases. The Program database (2004-2008) was used for epidemiological analysis and assessment of laboratory diagnosis methods use. The EWGLI case definition was used. The calculation of reporting rates were based on the middle of the year population estimates. Typing of clinical isolates (2004-2008) was performed by the two methods recommended by EWGLI: Monoclonal antibodies typing (MAbs) and Sequence Base Typing (SBT), using seven alleles. Results: In 2004-2008 a total of 400 cases were reported within the Program, compared with 338 DDO reports. This gives a crude annual reporting rate of 0.76/100,000 inhabitants. The GDH numbers were continuously higher than DDO and Program reports. During that period, six clusters were detected. The case-fatality ratio was 7%, 79% of cases were male and the most affected age groups were the 40-59 and ≥80 years. In cumulative terms, there were high numbers from May to December, with a peak in September-October. The use of urinary antigen detection diagnosis increased (from 44% in 2004 to > 80% since 2006) and the diagnosis by culture decreased (11% to 6%). Characterisation of 22 clinical isolates resulted in 5 sequence types (ST) only described in Portugal(total of 13 isolates), according with the EWGLI SBT database. Conclusions: From the comparison of LD data from hospital discharges (GDH) with data from clinical reports, we conclude that the reporting rate is increasing but still underestimating the real incidence. The Program showed clear added value. The increase in the number of cases detected is probably related with the increase in the diagnosis by urinary antigen detection. As in many countries, the culture method is rarely used, which compromises the microbiological characterisation of the disease in Portugal and the epidemiological links with the environmental investigations. These conclusions highlight the need to promote the diagnosis, mainly by culture, and notification of cases.
- 91 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P13
Predominance of L. pneumophila serogroup 1 ST-23 in Grenoble, France.
Elahe Hajmirbabaa, Marilyne Rummelharda, Brieuc Gestina, Jacques Croizea and Max Maurinb aCHU de Grenoble, Universite´ Joseph Fourier, BP 217, 38043 Grenoble, France; bLAPM, CNRS UMR5163, Universite´ Joseph Fourier, Institut Jean Roget, 38042 Grenoble, France [email protected]
Legionellosis is a prevalent disease in France, especially in Rhone Alps region. About 30-40 severe cases are hospitalized annually in Grenoble University Hospital (GUH), including about 25% in intensive care units. The true incidence of the disease is probably underestimated since many patients present with less severe clinical manifestations and remain undiagnosed. We studied 29 strains of L. pneumophila serogroup 1 isolated from 29 patients living in Grenoble area and hospitalized in GUH between 2001 and 2009 for community-acquired pneumonia. Strains were studied using the MLST procedure recommended by EWGLI, which is based on amplification and sequencing of a portion of seven genes (asd, pilE, neuA, momps, proA, flaA, and mip). The following MLST sequence types were found: ST-9 (1 strain), ST-20 (1 strain), ST-23 (14 strains), ST-48 (1 strain), ST-62 (3 strains), ST-75 (1 strain), ST-146 (3 strains), ST-299 (1 strain), and 4 strains with previously unrecognized ST. Thus, ST-23 represented 48.3% of the human strains isolated in GUH. We could not correlate infection with a ST-23 strain and residence of patients, year of hospitalization or severity of disease. We suspect L. pneumophila sg1 ST-23 to be a predominant species in water adduction systems in Grenoble. Alternatively, patients may have been infected from a common source that remains to be defined. We will test by MLST L. pneumophila sg1 strains isolated in water adduction systems or in environment during the same period to verify these hypotheses. Our observation may also suggest that ST-23 is a well-adapted environmental strain and/or a particularly virulent strain in humans.
- 92 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P14
Occurrence of Legionella non-pneumophila species in Denmark.
Christina Wiid Olsen, Mette Kjærgaard, Sanne Søgaard Nielsen and Søren A. Uldum
Statens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark [email protected]
From 1993 to 2008 Legionella non-pneumophila species were isolated from respiratory samples obtained from 27 cases of Legionnaires’ disease (LD); all cases were Danish citizens. All isolates have been identified to the species level by mip sequencing (RM Ratcliff et al. 1998, J. Clin. Microbiol. 36:1560-67). The distribution of the species were; L. micdadeii (n=12), L. bozemanii (n=9), L. longbeachae (n=4), L. anisa (n=1), L. dumofii (n=1). Almost all of the cases were primarily diagnosed by a positive PCR for Legionella non-pneumophila. In the period from 1997-2008, 32 isolates of Legionella non-pneumophila were obtained from water samples from 31 different sites in Denmark. The environmental isolates were also mip sequenced for species identification, and showed a rather different distribution; L. anisa (n=13) was the most present species in the environmental samples followed by L. micdadeii (n=6) and L. londiniensis (n=4). From 1993 to 2008 L. pneumophila were isolated from 583 cases of LD in Denmark. This means that app. 5% of all culture verified cases of LD were caused by a Legionella non-pneumophila species. Infections with Legionella non- pneumophila are rare but the number is probably underestimated, as the infections are difficult to diagnose. In Denmark several laboratories routinely use PCR that can detect ”all” Legionella species, from approximately 40% of the non- pneumophila PCR positive samples, it was possible to culture the bacteria. On this basis it can be estimated that Legionella non-pneumophila species accounts for 5 - 10 % of all LD cases in Denmark. Even though the infections are rare, we think that it is recommendable to use PCR which can detect Legionella non-pneumophila for the routine diagnosis of Legionnaires’disease.
- 93 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P15
An evaluation of the sensitivity of The Xpect Legionella test - an immunochromatic test for detection of Legionella pneumophila antigen in human urine samples.
Christina Wiid Olsen and Søren A. Uldum
Statens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark [email protected]
Early diagnosis of Legionnaires’ disease (LD) can be vital for the patients. Several methods are available for the diagnosis of LD, but urinary antigen tests are the most widely used, and have shown to be both specific and to give an early diagnosis. However, all available urinary antigen tests have the general disadvantage that they only have a satisfactory sensitivity for Legionella pneumophila serogroup (sg) 1. In this respect it is interesting that Oxoid recently has introduced the Xpect Legionella test for detection of L. pneumophila sg 1 and sg 6. We have evaluated the sensitivity of the Xpect kit with a collection of 69 urine samples from 67 patients with verified LD. A) 46 samples, selected from culture verified LD cases: L. pneumophila non-sg 1 (n=16), sg 1 subtype non-Pontiac (n=18) and sg 1 subtype Pontiac (n=13). B) 22 urine samples, randomly selected from LD cases, confirmed by urinary antigen test (n=17), PCR (n=4) or by a significant rise in antibodies to L. pneumophila (n=1). For a preliminary evaluation of the specificity five urine samples positive in pneumococcal urinary antigen test and five urine samples from patients with a positive Mycoplasma pneumonia respiratory PCR were examined, all 10 samples were negative. All the urine samples were simultaneously tested in the Binax Legionella urinary antigen EIA. The overall sensitivity for the 46 samples from group A was; 19.6% for the Oxoid Xpect kit and 69.6% for the Binax EIA kit. The sensitivity for the urine samples from patients with a non-sg 1 infection (n=16) was 0% and 43.8% respectively. For the sg1 samples (n=31) 29.0% and 80.7% respectively, (Pontiac (n=13) 61.5% and 84.6%, non-Pontiac (n=18) 7.1% and 77.8%). Among the non-sg 1 cases, eight cases (eight patients) of sg 6 infection was included, none of these were positive by the Xpect kit (three were positive by Binax EIA). For group B the Xpect kit showed a sensitivity of 45.5% and the Binax kit of 95.5%. In conclusion the Xpect kit lacks sensitivity for L. pneumophila non-sg 1 and some subgroups of sg 1, and it seems not to be sensitive for sg 6 at all. We cannot recommend this new kit for the diagnosis of LD.
- 94 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P16
Legionella urinary antigen test as a routine diagnostic assay, one year experiences from Denmark
Christina Wiid Olsen and Søren A. Uldum
Statens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark [email protected]
During 2008 we (SSI, Copenhagen) examined 1601 urine samples in our routine laboratory for Legionella pneumophila urinary antigen; 1140 samples by the Binax EIA kit and 461 samples by the Biotest EIA. Out of those, 93 samples were positive, of which 15 became negative after heat treatment. The remaining 78 positive samples were from 62 patients, and of these 11 with borderline positive urine samples by the Binax EIA were excluded as no other tests were performed and none of them were considered as LD cases (not pneumonia and not notified). Consequently we had 51 patients with positive urine samples, considered as verified LD case. In 2008, 61 cases of LD, from which a urine sample was obtained within a month from onset of symptoms, were diagnosed at SSI. Out of the 61 LD cases, 51 (se above) had at least one positive urine sample in the antigen tests (83.6% of all). Thirty of the cases were positive by Legionella PCR (respiratory samples). Of the 30 PCR positive patients, 21 also had a positive urinary antigen test (9 negative), which corresponds to a sensitivity for urinary antigen detection of 70%. A urine sample from one patient was negative in the antigen test, but the patient had a significant increase in antibodies to L. pneumophila. A total of 29 patient were positive by the urinary antigen tests, but had no samples investigated (n=22), or were negative, by other test. At local Danish hospitals, 37 cases of LD, from which a urine samples was collected, were recorded. Of the 37 cases, 23 were diagnosed by urinary antigen tests (Binax NOW or Biotest EIA), (62.2%). Of the 37 cases, 25 were tested positive by PCR, of which 11 also had a positive urine sample. This corresponds to a sensitivity for urinary antigen detection of 44%. Overall, with PCR and serology (one case) as references, the sensitivity for diagnosing LD by detection of Legionella urinary antigen in Denmark in 2008, was only 58.9%. The discrepancies seen between SSI and local laboratories can be due to the use of different kits. The manufacturers do not recommend heat treatment of samples to remove unspecific binding, but in our laboratory this is a routine procedure. The 15 samples which became negative after heat treatment, corresponds to a false positive rate of 1.32%. Even worse were the 11 patients with borderline ”false” positive results by the Binax EIA. To avoid false positive results we use a higher cut-off value in our routine setting than the cut-off value given by the manufacturer (3xnegative control).
- 95 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P17
Cross-reactions in ELISA IgM tests for Legionella pneumophila and Bordetella pertussis.
Katarzyna Pancer, Wlodzimierz Gut and Hanna Stypulkowska-Misiurewicz
National Institute of Public Health - NIH, Chocimska 24, 00-791 Warsaw, Poland [email protected]
Serological examinations were still the main method used for identification of infections due to Legionella pneumophila in Poland. In 2005-2006 the majority of reported cases of legionellosis were detected among children. Moreover almost 70% of received serum samples were single sera. In this situation the main problem of diagnosis of legionellosis was an appropriate interpretation of results of serological examinations. The aim of this study was to determine cross-reactivity between two ELISA IgM tests: one for L.pneumophila and second for Bordetella pertussis. In both tests LPS was the main antigen. Selected 15 serum samples (with previously determined level of IgM antibodies to L.pneumophila) were absorbed with antigens: - L.pneumophila sg 1 (ATCC 33152) - L.pneumophila sg 12 (ATCC 43290) - B.pertussis (clinical isolate) - L.pneumophila sg 1 + B.pertussis The ELISA IgM for L.pneumophila and ELISA IgM test for B.pertussis were done using non-absorbed and absorbed sera. The absorbed with L.pneumophila sg 1 sera were also examined using ELISA IgM test for M.pneumoniae. The index reduction of reactivity was determined (NAbs/Abs). The value of IgM index for sera absorbed with L.pneumophila sg 1 strains and tested by ELISA IgM L.pneumophila assay ranged 1,5 to 61,0; tested by ELISA IgM B.pertussis assay- from 1,13 to 2,66. For sera absorbed with B.pertussis antigen index IgM was from 3,7 to 23,8 (tested by ELISA IgM B.perussis) and from 1,51 to 3,0 (in ELISA IgM test for L.pneumophila). The IgM index determined in sera absorbed with both antigens were 3,36-62,6 (in ELISA IgM test for L.pneumophila) and 13,0-47,7 (in ELISA IgM test for B.pertussis). On the base of obtained results we were able to confirm infections due to L.pneumophila in 4 children. Moreover co-infection due to L.pneumophila and M.pneumopniae were found in 2 another children. The positive result of IgM antibodies to B.pertussis was found in 7 sera, among them in 3 sera the high reduction of IgM antibodies to L.pneumophila was also observed. The problem of interpretation of B.pertussis IgM ELISA tests is closely related to cross-reactivity with antigen L.pneumophila and unclear status of vaccination against B.pertussis in examined children. Statistical analysis indicated possibility of misidentification legionellosis as a B.pertussis infection because of the cross-reactivity between ELISA IgM L.pneumophila sg1 and ELISA IgM B.pertussis (CL95%). It might indicate the necessity of simultaneous examination of both infections in children.
- 96 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P18
Trends observed in Legionnaires’disease (LD) in a hospital of Calatonia (Spain) (1983- 2008)
Irma Casas, Raquel Nunez,˜ Maria Luisa Pedro-Botet, Lourdes Mateu, Nieves Sopena, Neus Robert, Fatima El Kanouni, Celestino Rey-Joly and Miquel Sabria`
Hospital Germans Trias i Pujol, carretera de canyet, 8916 Badalona, Spain [email protected]
Background: Active surveillance for legionellosis was initiated in 1983 to control hospital-acquired Legionellosis in a 650- bed hospital in Badalona (Catalonia). Earlier diagnosis and changes in treatment of LD may have allowed the recognition of this infection in patients with different demographic characteristics and risk factors and may have influenced clinical appearance and outcome. The aim of this study was to determine trends in demographic and clinical data, individual risk factors, diagnostic methods and outcome over time of a series of community and hospital-acquired LD collected from 1983 to 2008. Methods: Trends in case number, mean age, rate of males, smokers an alcoholics, individual risk factors, underlying diseases, extrarespiratory symptoms, diagnostic techniques, complications and outcome were calculated by multivariate lineal regression analysis per year. Results: We obtained data from 469 cases from 1983 to 2008. The community cases rose from 22,3% (period 1983-1988) to 91,3% (peri od 2004-2008). The rate of smokers, alcoholics, diagnosed by culture and by serology significantly decreased and the rate of diagnosed by urinary antigen tests significantly increased over time. Conclusions: Since 1983, the incidence of community-acquired LD has remarkably increased. There have been some differences in demographic characteristics, risk factors and diagnosis methods over time.
- 97 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P19
Has the Legionella urinary antigen test changed Legionella pneumonia?
Lourdes Mateu, Maria Luisa Pedro-Botet, Raquel Nunez,˜ Nieves Sopena, Fatima El Kanouni, Irma Casas, Marian Garc´ıa- Nunez,˜ Celestino Rey-Joly and Miquel Sabria`
Hospital Germans Trias i Pujol, carretera de canyet, 8916 Badalona, Spain [email protected]
Background: At present the Legionella urinary antigen test (LUA) is the hallmark of the diagnosis of Legionnaires’ disease and may have influenced the demographic and the clinical data of this disease. The aim of this study was to compare demographic data, risk factors, clinical characteristics and outcome of patients with community-acquired pneumonia (CAP) by Legionella pneumophila serogroup 1 diagnosed with LUA with those diagnosed with other tests. Methods: Patients were selected from a database of Legionella pneumonia prospectively collected from 1983 to 2006 in a 630-bed university hospital. We divided the patients into two groups: the LUA group (LG) including patients diagnosed with LUA and Non LUA group (NLG) including patients diagnosed with tests other than LUA. Results: We studied 248 patients, 247 (76.5%) diagnosed with LUA and 58 (23.5%) with other tests. The mean age was significantly higher in LG (p 0.006). Smoking habit and alcohol intake was more common in NLG than in LG. There were no differences between the groups concerning underlying diseases but LG had cancer (p 0.01) and were on immunosuppressive treatment (p 0.027) more frequently than NLG. According to clinical data, headache was more common in LG (p.006) and no differences in FINE score and other clinical data were observed. In relation to outcome, respiratory failure was more common in LG than in LNG (p 0.007). 13 patients died (5.5%), all being in the LG, although the differences between the groups were not significant (p 0.1). Conclusions: Age and immunosuppression of patients with Legionella CAP are increasing over time. The availability of LUA has not significantly changed the clinical appearance of this infectious disease but has notoriously increased the number of cases diagnosed.
- 98 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P20
Recurrent Infection Due to Legionella pneumophila serogroup 1 in an immunocom- promised patient
Carmen Pelaza, Manuel Garc´ıa-Cenozb, Beatriz Baladronc, Aurelio Barricarteb, Jesus´ Castillab and Fatima´ Irisarreb aLegionella Reference Laboratory; Centro Nacional Microbiolog´ıa; Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain; bInstituto de Salud Publica´ de Navarra, c) Leyre, 15, Pamplona, 31003 Navarra, Spain; cCentro Nacional de Microbiolog´ıa. Instituto de Salud Carlos III, Crta. Pozuelo Majadahonda, KM 1, 28220 Majadahonda, Spain [email protected]
We describe a case of recurrent Legionnaires’disesase in an immunocompromised patient who was admitted in hospital for surgical intervention of Glioblastoma multiforme. On November 2008, a case of nosocomial legionnaires’disease was notified to the Public Health Service of Navarra, Spain. The case was a woman, aged 69 years, with date of hospital admission on November, 3th. On November 13th she was diagnosed of Legionella pneumonia with positive urine antigen and culture. Patient strain was identified as L. pneumophila SG1, Pontiac, AFLP CNM 039b, PFGE-SfiI CNM 010 and SBT 62. L. pneumophila SG 1 was recovered from several places from the hospital, but neither of them matched with the patient strain. However, the patient strain was recovered from the hospital (a shower room) one year before. On February 2009, the patient was diagnosed of community pneumonia, as no hospital admissions had been reported in the previous 10 days of the onset of symptoms. The patient had remained at home since the last hospital discharge, except for a visit to Radiotherapy Unit 16 days before the onset. Urine antigen and culture were positive for Legionella again. The second culture of this patient was indistinguishable from the first one. Water samples were recovered from the patient home, but all of them were negative for Legionella. Although a re-infection could be possible with the same Legionella strain from some community environmental source, these data could also suggest that the second infection was a reactivation of a persistent focus of infection that was not apparent during three months.
- 99 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P21
Wet cooling systems as a source of sporadic Legionnaires’ disease: a geographical analysis of data for England and Wales, 1996 to 2006
Kate Rickettsa, Carol Josepha and Paul Wilkinsonb aHealth Protection Agency Centre for Infections, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; bLondon School of Hygiene and Tropical Medicine, Keppel Street, WC1E 7HT London, United Kingdom [email protected]
This paper reports on the findings of a case-control study which aimed to quantify the relationship between the occurrence of sporadic cases of Legionnaires’ disease and proximity to wet cooling systems (WCSs). The dataset for analysis comprised 1163 sporadic, community-acquired cases of Legionnaires’ disease in England and Wales, with onset between 1996 and 2006, and 11630 postcode controls randomly sampled in proportion to population size and matched on region, age group and sex. The relationship between risk of Legionnaires’ disease and distance from a WCS was analysed by conditional logistic regression. Cases and controls had a mean age of 56.3 years; 79.3% were male. Cases lived appreciably closer to WCSs than their controls (mean distance of cases = 2.11 km, controls = 2.58 km; mean difference 0.47 km (95% CI: 0.28, 0.65)). The un-adjusted odds ratio for disease within 1 km of a WCS compared with over 6 km (a distance taken to reflect background rates of Legionnaires’ disease) was 1.94 (95% CI: 1.48, 2.56), and was 1.43 (95% CI: 1.12, 1.83) when adjusted for socio-economic deprivation and population density. Population attributable risk calculations based on odds ratios by distance band suggest that residential proximity to a WCS represents the source of infection for over a third of sporadic community-acquired cases in England and Wales. A substantial proportion of sporadic, community-acquired cases of Legionnaires’ disease may be linked to cooling towers, evaporative condensers and other WCSs, an observation that has important implications for health protection policies.
- 100 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P22
Factors Associated with in-Hospital Mortality of Legionnaires’ Disease (LD): A Prospective Multicenter Study of 595 Patients (pts) in France.
Christian Chidiaca, Silene Pires-Cronenbergerb, Pierre Weinbreckc, Didier Ched, Christine Campesed, Sophie Jarraude, Philippe Vanhemsb and Group Studyf aService de Maladies Infectieuses et Tropicales, Hopitalˆ de la Croix Rousse, 103 Grande Rue de la Croix Rousse, 69004 Lyon, France; bService d’Hygiene` Epidemiologie´ et Prevention,´ Hopitalˆ Edouard Herriot, 5 Place d’Arsonval, 69003 Lyon, France; cCHU Dupuytren, 2, av Martin Luther-King, F87042 Limoges, France; dInstitut de Veille Sanitaire, 12 rue du Val d’Osne, 94415 Saint Maurice, France; eCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; fUCBL1 et HCL, maladies Infectieuses, F69317 Lyon, France [email protected]
Background: Risk factors for acquisition of LD have largely been described, but the factors associated with survival of pts with LD are not often reported in large cohorts. The aim of this study was to identify the determinants associated with in-hospital mortality in pts with LD. Methods: French hospitalized pts with LD based on abnormal chest XR and positive Legionella pneumophila serogroup 1 urinary antigen test were prospectively included (April 2006-June 2007). In-hospital death was considered for survival analysis. Kaplan Meier method and Cox regression model were used to identify the independent variables associated with mortality according to the hazard ratio (HR) with its 95% CI. Results: From April 2006 to june 2007, 595 pts with confirmed LD were included (mean age 61.2 ± 16, M/F sex ratio: 2.7, 9). 74.6% had at least one identified risk factor: smoking (52.6%), diabetes mellitus (15.8%), malignancy (8.4%). Main clinical symptoms at onset were: fever >38.5◦C (86.2%), respiratory (74.4%), digestive (30.9%), and neurological signs (39.0). 56.7% of pts had a Pneumonia Severity Index class of IV-V at admission. 29% of pts required ICU. The median time between onset and hospitalization and onset appropriate antimicrobial therapy (macrolide or quinolone) were 4 days (d) and 4d respectively. Median length of stay was 10d. Death occurred in 58 pts (9.7%). The probability of survival at 10d, 20d, 30d, and 60d was 93%, 87%, 78%, and 65% respectively. The variables independently associated with mortality were: female gender (HR 1.8, 95% CI 1.0-3.0), age by 10 years increase (HR 1.3 95% CI 1.0-1.5), and ICU admission (HR 4.2, 95% CI 2.3-7.8). The delays between onset and hospitalization, onset and diagnosis, and onset and appropriate antibiotherapy were not significantly associated with survival. Conclusion: These results contribute to improve our knowledge on presentation and prognosis of LD in hospitalized pts, and stress the need for a reduction of delays in terms of diagnosis and treatment, particularly for pts with the most severe prognosis.
- 101 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P23
Legionellosis in Poland 1997-2009. The Epidemiology diffent from EU Countries
Hanna Stypulkowska-Misiurewicz and Katarzyna Pancer
National Institute of Public Health - NIH, Chocimska 24, 00-791 Warsaw, Poland [email protected]
Legionella infections are registered more often in developed countries than in less developped ones as in Poland The aim of the study was to analysed the developpment of legionella infections suveillance system in Poland to explain the origin of differences in epidemiology of registered cases between Poland and some EU contries Material and methods: laboratory and epidemiological data colected in NIPH-NIH since 1998 Results:The first impact for work on subject was ceated in 1997 by official note from WHO that EWGLI registered swedish patient with legionellosis infected in polish spa treatment centre. The epidemiological investigation with examination of water system were demanded. Finantial grant from our gouvernement and contacts with some EWGLI members facilitated to develope the pactical laboratory methods for examination of patients and water systems for legionella infections. Obligatory registration of legionellosis was introduced in 2002 and legal regulation for obligatory control of water in hospitals enter to power in 2007. In 1997- 2008 EWGLI registered the 30 tourists from EU countries as probably infected in Poland but only 8 cases of polish citizens infected abroad. Such difference is not only due to diagnostic facilities but probably due to travelling population: from Poland younger and more healthy population is travelling abroad so pneumonia symptoms are not severe.Survey of children cases of pulmonary diseases has shown high level of IgM in many of them indicating probability of Legionella infection. Conclusion: Uunexpensive, reliable, tests to diagnose are needed for not severe cases of legionellosis as some are misdiagnosed as imported influenza
- 102 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P24
Report of two cases each with two separate episodes of Legionnaires’ disease
Søren A. Ulduma, Jette M. Bangsborgb and Lars Lemmingc aStatens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark; bCopenhagen University Hospital Herlev, Herlev Ringvej 75, 2730 Herlev, Denmark; cAarhus University Hospital Skejby, Brendstrupgardsvej˚ 100, 8200 Arhus˚ N, Denmark [email protected]
We report two patients that contracted Legionnaires’ disease (LD) twice. In each case, the time span between the two episodes was approximately three years. Both patients had a favourable outcome. Case 1: A 76 year old man with Morbus Waldenstrom¨ contracted LD during a travel to Greece in October 2005. The diagnosis was confirmed by PCR and subsequent isolation of Legionella pneumophila serogroup 1 subgroup Benidorm by culture. Two urine samples were also positive by Legionella urinary antigen test (Biotest). No blood samples were collected. Three years later (September 2008) the patient was hospitalised again with fever and pneumonia. A lower respiratory tract (LRT) specimen was positive for L. pneumophila by PCR, a urine sample was negative by Legionella urinary antigen test (Binax), but an IgM seroconversion (negative to positive) to L. pneumophila serogroup 1 was recorded (in-house ELISA) in two blood samples collected four and 36 days after onset, respectively. The PCR positive sample was unfortunately negative by culture, but the serological results confirmed a current infection. Case 2: A 66 year old male with chronic lymphocytic leukaemia was hospitalised with high fever and pneumonia in June 2006. LD was diagnosed by PCR and subsequent isolation of L. pneumophila serogroup 1 subgroup Bellingham. A urine sample was negative by the Biotest urinary antigen test, but testing the same urinary sample with the Binax kit revealed a positive result. No blood samples were collected. One year later (August 2007) a urine sample and a LRT sample was investigated for Legionella. The LRT sample was negative by PCR and the urine sample was borderline positive by the Binax kit. The patient was again hospitalised with pneumonia in May 2009. LD was diagnosed by PCR and subsequent isolation of L. pneumophila serogroup 1 subgroup Bellingham. A urine sample collected four weeks after onset was negative by the Binax kit but a blood sample from the same day showed a high level of IgM antibodies to L. pneumophila sg 1 and sg 6. Conclusion: These recent cases show that LD can occur more than once in a lifetime in patients with immune dysfunction, even in the absence of specific immunosuppressive therapy. The risk of recurrent infection highlights the need for preventive measures also outside hospitals, i.e. in private homes and in the community, for this group of patients.
- 103 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P25
Isolation of Legionella bozemanii from a lymph node in a healthy young woman
Søren A. Ulduma, Naser Dadkhahb and Jens Jørgen Christensena aStatens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark; bKøge Hospital, Department of ENT, Lykkebækvej 1, 4600 Køge, Denmark [email protected]
We report the detection and isolation of Legionella bozemanii from a biopsy from a lymph node in an otherwise healthy 18 year old woman. The patient was seen in an ear-nose-throat outpatient clinic and she presented with a swollen lymph node at the neck. She was otherwise healthy without actual or recent respiratory symptoms. A biopsy of the lymph node was drawn and investigated by partial 16 S rRNA gene sequencing. The sequence was blasted against known sequences at the NCBI gene bank and the best taxon match was Legionella bozemanii. Sample material was thereafter run in a Legionella qPCR (TaqMan based PCR detecting all Legionella species by primers to the 5 S gene and L. pneumophila by primers to the mip gene) and spread on a BCYE ager plate, in an attempt to isolate the bacteria. The PCR was positive for Legionella species but negative for L. pneumophila. The culture was negative; a new biopsy was requested for culture. The sample was drawn approximately 40 days after the first sample was collected. At that time the lymph node was shrunken but still swollen. The patient had not received antibiotic treatment in the intervening period. PCR was still positive for Legionella spp non-pneumophila. Legionella grew on the BCYE plate and the isolate was identified as Legionella bozemanii by mip sequencing (RM Ratcliff et al. 1998, J. Clin. Microbiol. 36:1560-67). After confirmation of the persistent infection, azithromycin treatment was initiated. A month later, although the node was still present the patient was well-being. To our knowledge, this is the first report of isolation of Legionella spp. from a lymph node from a patient without a preceding respiratory infection.
- 104 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P26
A case of recurrent meningitis associated with positive Legionella urinary antigen test and seroconversion to Legionella pneumophila
Søren A. Ulduma, Hanne Vebert Olesenb, Mikala Wangc and Pernille Elverdala aStatens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark; bAarhus University Hospital Skejby, Pediatric Department, Brendstrupgardsvej˚ 100, 8200 Arhus˚ N, Denmark; cAarhus University Hospital Skejby, Department of Clinical Microbiology, Brendstrupgardsvej˚ 100, 8200 Arhus˚ N, Denmark [email protected]
We report a case of recurrent purulent meningitis in a fourteen year old girl with a cochlear implant. There were no respiratory symptoms during any of the incidents. During the second episode of meningitis a urine sample as well as a serum sample was positive for Legionella pneumophila (antigen and antibodies respectively). No other obvious agents (neither by culture, PCR, antigen tests or serology) were found in the initial two episodes, which were three months apart. A third incident occurred three months later, and Streptococcus (S.) pneumoniae serotype 15 was cultured from cerebrospinal fluid. During the second episode (six days after onset) a urine sample was investigated for S. pneumoniae antigen and found negative. By coincidence the sample was also tested with the Binax Legionella urinary antigen EIA kit, and came out positive (OD = 0.898, with a cut off value of OD = 0.099). The result was confirmed by testing the sample after heat treatment (boiling for 5 min). Subsequently a new urine sample (three days later) and two blood samples for Legionella antibody test were obtained. The urine sample was negative, but the serum samples were highly positive for IgM antibodies to L. pneumophila (serogroup 3 and 6) in an in-house ELISA. Two serum samples collected during the first episode of meningitis (three months and two and a half months earlier), were then tested for antibodies to L. pneumophila. A very pronounced rise in the specific IgM antibody level was observed between the first and second sample (negative to highly positive), especially for L. pneumophila serogroup 3. The high level persisted in two blood samples collected during the third incident. No IgG antibodies were detected in any of the six samples. We consider these results indicative of a L. pneumophila involvement in the first two episodes of meningitis; however Legionella was probably not a factor in the third. We speculate that the cochlear implant could play a role as basis/surface for growth of Legionella.
- 105 - Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P27
Post-antibiotic effect of various antibiotics on Legionella pneumophila strains isolated from environmental samples
Seher Birteksoz¨ a and Zuhal Zeybekb aIstanbul University, Istanbul University, Faculty of Pharmacy, Department of Pharmaceutical Microbiology, 34134 Istanbul, Turkey; bIstanbul University, Faculty of Science, Department of Biology, Section of Fundamental and Industrial Microbiology, 34134 Istanbul, Turkey [email protected]
The aim of our study is to examine the PAE of azithromycin, clarithromycin, ciprofloxacin and levofloxacin against L. pneumophila strains isolated from several water systems of different buildings in Istanbul. Azithromycin, clarithromycin, ciprofloxacin and levofloxacin MICs were determined by microbroth dilution technique as described by CLSI. PAEs were determined by a standard viable method where bacteria in the logarithmic phase of growth were exposed for 1 hour to antibiotics and antibiotics were removed by centrifugation and repeated washing. The PAE was defined as PAE = T-C, where T is the time (in hours) required for the count in the test culture to increase 1 log 10 above the count observed immediately after centrifugation and C is the corresponding time for the controls. Experiments were performed in duplicate.The MICs were determined in a range of 0,0156-0,0312 µg/ml for azithromycin, 0,0078-0,0156 µg/ml for clarithromycin, 0,0156 µg/ml for ciprofloxacin and 0,0078-0,0156 µg/ml for levofloxacin. The PAEs of L. pneumophila strains exposed for 1h to 1xMIC and 4xMIC of azithromycin, ciprofloxacin, clarithromycin and levofloxacin ranged from 0,70 to 2,75 h and from 2,95 to 4,5 h; from 1,45 to 5,75 h and from 1,7 to 7,20h; from 1,85 to 3,55 h and from 2,95 to 4,8 h; from 1,25 to 2,65 and from 2.0 to 4.75h, respectively. All of the antibiotics have PAEs on L. pneumophila strains. When the concentration of antibiotics were increased, the duration of PAE was prolonged. The duration of PAE was prolonged related to the increasing concentrations. The findings of this study may have important information for the optimal timing of the doses during therapy with these antibiotics.
- 106 -
SESSION 2 CLINICAL ASPECTS B) DETECTION AND SUBTYPING
- 107 - Session 2: Clinical aspects a) detection and subtyping P28
Evaluation of a PCR assay to detect Legionella pneumophila in respiratory samples
Beatriz Baladrona, Virginia Gilb, Consuelo Elolab and Carmen Pelazb aCentro Nacional de Microbiolog´ıa. Instituto de Salud Carlos III, Crta. Pozuelo Majadahonda, KM 1, 28220 Majadahonda, Spain; bLegionella Reference Laboratory; Centro Nacional Microbiolog´ıa; Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain [email protected]
Introduction: Culture is the gold standard for detection of Legionella in clinical samples, but it has a low sensitivity and requires several days to obtain results. PCR assays have been developed with different target (mip gene, 16s rRNA) providing results in a short period of time. The aim of this study was to test a conventional PCR assay using the mip gene to detect Legionella in clinical samples. The results were compared with two multiplex PCR assays previously published (1, 2), used as a conventional monoplex PCR. Material and method: 72 clinical samples were analysed, including 20 sputum and 38 bronchroalveolar lavage (BAL) samples among others. Culture was performed with all samples. Urinary antigen results were reported only in 17 of the 72 clinical samples. Primers were designed using mip gene as the target with a conventional PCR. PCR conditions of the multiplex PCR (1, 2) were adapted to a conventional monoplex PCR. Sensitivity of the method was determined with ten-fold dilution series of the reference L.pneumophila SG1 and SG6 strains. Reference strains of L.pneumophila serogroups 1 to 14, L.bozemanii, L.anisa, L.jordanis, L.micdadei and L.longbeachae SG1 and other species different from L.pneumophila were used to test the specificity. Samples showing discrepant results between the three PCR assays, culture and urinary antigen results were examined by sequencing (3). Results and discussion: Eight samples were culture positive and 18 were PCR positive (17 of them with urinary antigen positive). Discrepant results were: 1) Two samples PCR positive with urinary antigen no reported, these positive results were confirmed by sequencing. 2) A false negative PCR result, with a positive urinary antigen and culture negative, could be due to the low amount of bacteria present in the sample. The sensitivity of the culture was of 47%, being 14.28% for the sputum. PCR sensitivity was lower than expected (1.2x102 ufc/reaction). PCR specificity was 98%, no cross- reactivity was found and it increased to 100% by sequencing the samples that showed discordant positive results. The results obtained with this PCR were compared to those of the multiplex PCR assays, the results were similar to one of the PCR (1) and better than the other (2). Conclusions: The PCR method evaluated increased the sensibility in detection of Legionella compared to culture. However sensitivity was low compared to other PCR methods. PCR could be a very useful tool as a screening method previous to culture. References:(1) Welti, M, et al., (2002), Diagnostic Microbiol and Infect Disease, 45: 85. (2) McDonoughet, E, et al., (2005), Mol Cell Probes, 19: 314. (3) Ratcliff, RM, et al., (1998), J Clin Microbiol, 36:1560.
- 108 - Session 2: Clinical aspects a) detection and subtyping P29
Occurrence of Legionella pneumophila sequence types circulating in clinical and environmental areas of Northern Italy and comparison with other regions of the world
Annalisa Bianchi, Marina Tesauro, Michela Consonni and Maria Gabriella Galli
Dipartimento di Sanita` Pubblica, Microbiologia e Virologia. Universita` degli Studi di Milano, via Pascal 36, 20133 Milano, Italy [email protected]
We compared the sequence types (ST) designed by sequence-based typing (SBT) of 4 clinical and 12 environmental strains of Legionella pneumophila serogroup 1 (LP1), isolated from hospital facilities for the mentally disabled and an associated hospital. The strains were selected after a retrospective surveillance of 565 clinical records (2002-09) and investigations of water circuits. It was possible to correlate two clinical strains with the corresponding environment, which were collected from showers that had exposed the patients (ST685: 2,10,18,10,1,1,9 and ST16: 2,10,18,10,2,1,9) and two clinical strains present in the same structure (ST1: 1,4,3,1,1,1,1). The other environmental strains were isolated from water showers in the departments where there were confirmed or suspected clinical cases. All the strains (seven) from the first structure had ST188 (2,6,17,6,13,3,11), two from the second structure had ST34 (3,13,1,25,14,9,6) and the last from the third structure, which was colonized in a single point from LP1, gave an ST694 (6,1,22,30,6,10,11). From this analysis, it appears that the distribution of environmental strains is homogeneous, with only one ST for structure, unlike in the hospital that was colonized by at least three strains simultaneously, with different STs. The results were compared with the EWGLI international database: the four STs (1, 16, 34 and 188) were already known in literature. Their presence was confirmed among clinical and nosocomial cases, especially for ST1, the most frequent and distributed worldwide. Two STs were new to the database and therefore never isolated and sequenced in Italy or any country: ST685 was isolated both from a patient and from the water, ST694, which was found exclusively in the third control structure (no cases of legionellosis and low number of pneumonia), was unknown in the literature and we could only speculate on its possible minor virulence and/or distribution. The application of the SBT in epidemiological surveys is important to establish the source of infection, if it were possible to analyze the clinical strains. The large-scale implementation of SBT and international comparisons may be useful to gain the genotypic knowledge of circulating environmental strains and eventually determine their possible virulence, verifying their presence in the clinical setting.
- 109 - Session 2: Clinical aspects a) detection and subtyping P30
Legionella longbeachae Sg 1 infections linked to Potting Compost
Diane Lindsay, Alistair Brown and Giles Edwards
Scottish Haemophilus Legionella Meningococcus Pneumococcus Reference Laboratory, Stobhhill Hospital, 133 Balornock Road, G21 3UW Glasgow, United Kingdom [email protected]
While investigating a human case of Legionella longbeachae Sg 1 infection associated with a UK-produced brand of potting compost, four different species of Legionella were isolated from the compost using conventional and immuno- magnetic separation techniques. The identity of the strains was verified by polyclonal serology and mip speciation. The isolates identified from the compost were L. anisa, L. londinensis, L. sainthelensis and L. longbeachae Sg 1 (A) and B). The patient and compost L. longbeachae Sg 1 isolates were genotyped using Amplified Fragment Length Polymorphism (AFLP). One isolate (A) from the compost and the patient isolate were found to be indistinguishable but different from isolate (B). Over a year later, a further culture positive human case of L. longbeachae Sg 1 was identified. The AFLP type of the patient isolate was different from the first isolate (A) case but identical to isolate (B). In this second case, compost samples were also tested and L.longbeachae Sg 1 was isolated. The AFLP type was identical in both the patient and compost isolates. The patient and compost from the second case was indistinguishable from isolate (B) in the first case compost. There was geographical and product association with both cases. In the second case, the concentration of L. longbeachae Sg 1 in the compost was calculated and found to contain 4000 cfu/g. In total, nine L. longbeachae Sg 1 strains including reference, patient and environmental were typed by AFLP and two distinct genotypes or AFLP types were found; both differed from that of our reference strain. Sequence based typing (SBT) was attempted on the strains but only 2 house keeping genes could be amplified and those were considered novel sequences (EWGLI SBT database). This study highlights the diversity of Legionella species found in compost and further confirms the associated risk factors. In addition with the strong link to two human cases, the inclusion of hazard warnings on potting compost should be included considering the profile of many gardeners
- 110 - Session 2: Clinical aspects a) detection and subtyping P31
Evaluation of an oligochromatographic test for Legionella pneumophila detection in respiratory samples
Diane Lindsaya, Alistair Browna, Jose A Carrillob and I Rodr´ıguez-De La Rosab aScottish Haemophilus Legionella Meningococcus Pneumococcus Reference Laboratory, Stobhhill Hospital, 133 Balornock Road, G21 3UW Glasgow, United Kingdom; bVircell SL, Plaza Dom´ınguez Ortiz 1, Pol. Ind. Dos de Octubre, 18320 Santa Fe (Granada), Spain [email protected]
OBJECTIVES To evaluate a new rapid commercial assay (Speed-oligo°R LPN, Vircell) for the detection of Legionella pneumophila DNA in respiratory samples based on a PCR and rapid probe hybridization test device. METHODS A 134 base pair fragment of the Legionella pneumophila mip gene is amplified and the PCR product detected by means of a probe hybridization dipstick included in the kit. The amplified products flow into the strip for a double hybridization with colloidal gold conjugates bearing complementary probes, for the specific amplicon and the internal control, and with specific probes immobilized in the membrane (test line and PCR amplification control line). Final reading of the results is accomplished visually after 5 minutes incubation. To check specificity, DNA from 11 commonly seen respiratory bacteria and viruses including Adenovirus, Influenza B, Parainfluenza 1, 2 and 3, Respiratory syncytial virus and Chlamydophila psittaci, Haemophilus influenzae, Streptococcus pneumoniae and Mycoplasma pneumoniae were tested. Legionella pneumophila serogroup (Sg) 1 Philadelphia, Knoxville, OLDA, Bellingham, Camperdown and Benidorm, Legionella pneumophila Sg 2-14, L. anisa, L. micdadei, L. bozemanii Sg 1 and 2, L. feeleii Sg 1 and 2, L. hackeliae Sg 1 and 2, L. longbeachae Sg 1 and 2, L. cherrii, L. gormanii. L. oakridgensis, L. jordanis, L. erythra Sg 1 and L. dumoffii were also tested for specificity of the kit. Sixty-seven patient respiratory samples previously tested in an ”in house” 16s RNA PCR assay for all Legionella species were analysed by Speed-oligo°R LPN. RESULTS There was no cross reactivity observed with the 11 common respiratory viruses and bacteria tested. All the Legionella pneumophila serogroups gave a positive signal with Speed-oligo°R kit; however faint test lines were observed with L. bozemanii Sg 2, L. feelii Sg 1 and 2, L. longbeachae Sg 2 and L. oakridgensis. The test gave a sensitivity score of 100% in the 67 respiratory samples analysed. 19 samples were positive by the ”in house” PCR assay and the Speed-oligo°R LPN (11 L. pneumophila Sg 1, 2 L. pneumophila Sg 3, 1 L. pneumophila Sg 5, 1 L. anisa, 1 L. longbeachae Sg 1 and 3 unknown Legionella species). The 19 samples were from patients that were identified as either presumptive or definitive cases as they were also positive in one or more other laboratory tests (urinary antigen, IFA serology, in-house PCR and/or culture). The specificity of the test was 89.5% because one of the positive samples was considered an L. anisa by IFA serology and other as L. longbeachae Sg 1 by IFA serology and culture. However cross reactions are possible in respiratory samples if we consider the cross reactivity with the tested strains. The remaining samples were negative with all the assayed techniques. Two of those negative samples gave invalid results because the amplification control line did not appear, probably due to the presence of some inhibitors in the samples. CONCLUSION Speed-oligo°R LPN is a fast and sensitive assay for L. pneumophila detection in clinical samples. It showed high correlation when results were compared with standard methods.
- 111 - Session 2: Clinical aspects a) detection and subtyping P32
Molecular typing of Legionella Pneumophila strains isolated in Italy from 1987 to 2008: preliminary results
Stefano Fontana, Maria Luisa Ricci, Desiree´ Mineo, Federica Pinci and Maria Scaturro
Department of Infectious, Parasitic and Immune-mediated Diseases - Istituto Superiore di Sanita` -, Viale Regina Elena, 299, 161 Rome, Italy [email protected]
Legionella pneumophila (Lp) is a facultative intracellular pathogen, responsible for a severe form of atypical pneumonia known as Legionnaires’ Disease. Lp is widespread in natural and artificial aquatic environments, and occasionally infects humans through inhalation or aspiration of contaminated aerosols. In the last ten years the number of the Italian notified cases increased about tenfold. To improve the microbiological and epidemiological surveillance of legionellosis, in 2006 the Italian Center for Disease Control (CCM) has funded a project to characterize Lp at molecular level, in order to identify the most frequent genomic types circulating in the national territory and their relative distribution in the different regions. 198 clinical isolates, collected from 1987 to 2008 in various Italian regions were analysed in this study. For each strain, monoclonal antibodies (MAb) typing was performed by Indirect Immunofluorescence assay. The Amplified Fragment Length Polymorphism (AFLP) method was used to determine the genomic pattern according to the EWGLI (European Working Group for Legionella Infection) standard protocol. The various genomic profiles were subjected to UPGMA analysis, to assess their genomic relationships. About 90% of Lp isolates belonged to the serogroup 1 (sg 1), and 10% was represented by serogroups 3, 4, 5, 6, 7, 8, 9, 14. MAb typing of the Lp sg 1 isolates demonstrated a predominance of the Philadelphia subgroup (47.5%), followed by Knoxville (21.8%), Benidorm (12.8%), Olda (8.9%), Allentown (6.7%), Oxford (1.7%) and Bellingham (0.6%). The AFLP analysis distinguished 31 different genomic profiles among the 179 sg 1 isolates (90%). Two profiles, 10c and 5a, were prevalent (45%). The profile 10c was significantly more represented in Latium, as compared to the other regions. As expected no correlation was evidenced between the phenotypes and the genomic profiles. These preliminary results highlight the molecular distribution of Lp strains in Italy showing the prevalence of 10c and 5a genomic profiles. Sequence Based Typing (SBT) method will be also performed for all the isolated Italian strains in order to compare the obtained allelic profiles, with those of the EWGLI database. Moreover, a possible relationship between AFLP and SBT typing will be evaluated.
- 112 - Session 2: Clinical aspects a) detection and subtyping P33
Detection of Legionella DNA by Real-Time PCR in Respiratory Samples.
Laura Franzin, Daniela Cabodi and Nicoletta Bonfrate
Amedeo di Savoia Hospital, Corso Svizzera 164, 10149 TORINO, Italy [email protected]
Introduction: Culture is still the gold standard for the diagnosis of Legionella infection and is the most specific diagnostic procedure. However several days are required to obtain a positive result. DNA techniques are promising for rapid detection. In this study a real-time PCR was used for the direct detection of Legionella from clinical samples in comparison to culture. Methods: Culture and real-time PCR was performed on 132 respiratory samples (33 sputum, 52 bronchoalveolar lavage fluid, 42 bronchial aspirate, 5 lung tissue) from 130 patients with pneumonia. Legionella infection (LD) was confirmed in 57 patients by at least one of these positive tests: urine antigen detection (ELISA), serology (fourfold rise in antibody titers by indirect immunofluorescence and/or microagglutination test) and culture. Aliquots of the untreated, heat-treated and acid-washed samples were plated on BCYE, BMPA and MWY. The plates were incubated at 37◦C for 15 days. Real-time PCR for L.pneumophila (Lp) and Legionella spp (L.spp) was performed with iCycler after DNA extraction. Results: Culture was positive from 39/59 (66.1%) respiratory samples of 57 confirmed cases of LD. Lp serogroup 1 was isolated from 36 samples of 34 patients, Lp serogroup 3 from one and L.bozemanii from two patients. Real-time PCR was positive in a total of 56/132 (42.4%) samples. 55/59 (93.2%) samples from LD patients were positive by real-time PCR, while 72/73 (98.6%) samples from patients without LD were negative. Discrepancy was observed in four cases (7%) of patients with LD and only in one case (1.4%) in patients without LD. Five patients with proven LD were positive by real-time PCR for L.spp and negative for Lp, two of these was culture positive for L.bozemanii. All respiratory samples positive for culture were real-time PCR positive, except one where few colonies of Lp serogroup 3 were isolated. Conclusion: LD laboratory diagnosis is better achieved by combined urine antigen detection, serology and culture. Results of this study showed good concordance between the criterion based on traditional methods and real-time PCR. This sensitive technique is useful for early detection of Legionella in patients with suspected LD who produce sputum, but it should be critically evaluated.
- 113 - Session 2: Clinical aspects a) detection and subtyping P34
Use of automated online tools in External Quality Assurance programs for sequence- based typing uncovers both strengths and weaknesses in technique and proficiency
Norman Frya, Massimo Mentastia, Anthony Underwoodb, Garan Jonesa and Timothy Harrisona aHealth Protection Agency Centre for Infections, Respiratory and Systemic Infection Laboratory, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; bHealth Protection Agency Centre for Infections, Statistics, Modelling and Bioinformatics Department, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
The seven locus European Working Group for Legionella Infections (EWGLI) DNA Sequence-Based Typing (SBT) scheme for Legionella pneumophila, is now well-established internationally (see http://www.hpa-bioinformatics.org.uk/ legionella/legionella sbt/php/sbt homepage.php). Since the description of the original method in 2003, we have sought to assess the ability of national and regional reference laboratories to correctly type coded strains of L. pneumophila isolates by introducing a proficiency program using standard protocols. This program has also allowed rigorous evaluation of modifications to the methodology. To date a total of five distributions have been sent with a total of 31 centres in 22 countries participating (from Europe, Canada, Japan, Singapore and the USA) in the latest round. Each panel, containing coded L. pneumophila isolates, including at least one epidemiologically related pair, was distributed to participating centres with detailed instructions. Participants were requested to submit data online using the Sequence Quality Tool (SQT). This tool allows users to upload trace files in standard file formats, automates base- calling using phred and phrap software, contig assembly, trimming, and matching against a reference library. A quality report is automatically generated giving the allelic profile and Sequence Type (ST). Completed results from the most recent distribution were received from 31/32 centres of which 24 (77%) reported the correct allelic profiles/STs for the entire panel (5/5); four centres achieved 4/5 and the remaining two centres 0/5. Deficiencies identified in those laboratories not scoring 100% included; failure to use latest protocols, poor storage of reagents, and lack of experience of sequencing. The development of a novel web-based tool (the SQT) has facilitated the automated assembly and quality assessment of sequence chromatograms and since its development has allowed user-independent scoring in this program. A major strength of DNA sequence-based typing schemes is the portability of the data. However, these programs reveal two main shortcomings: inexperience and poor laboratory practice. Use of properly trained staff and continuous monitoring of laboratory performance is vital to facilitate improvements, highlight technical problems and identify training needs in all aspects of the procedure to ensure high confidence in laboratories responsible for reporting such data.
- 114 - Session 2: Clinical aspects a) detection and subtyping P35
Implication of sequence - based typing method for epidemiological investigation of Legionella pneumophila infections
Darja Kesea, Rok Kogoja and Boris Kopilovicb aInstitute of Microbiology and Immunology, Medical Faculty, University of Ljubljana, Zaloska 4, 1000 Ljubljana, Slovenia; bInstitute of Public Health Koper, Vojkovo nabrezje 4/a, 6000 Koper, Slovenia [email protected]
The aim of the study was to perform amplification and sequencing of Legionella pneumophila DNA directly from clinical specimens and from potable water. Genomic DNA was extracted from lung autopsy specimens obtained of two patients, and from bronchoalveolar lavage fluid specimen of the third patient. Furthermore, DNA was also prepared from L. pneumophila isolated from the potable water of the building where patients were stationed and from an epidemiologically unrelated strain from our collection, serving as a positive control. Isolation of DNA was performed using the MagNA Pure Compact automated nucleic acid isolation system (Roche Diagnostics, Germany). A real time PCR assay using Legionella species r-gene Primers/Probe mix (Argene Biosoft, France) was prepared and found to be positive for Legionella spp. for all tested samples. Subsequently, genomic DNA was amplified using the primers targeting seven Legionella genes comprising, flaA, pilE, asd, mip, mompS, proA and neuA described by the EWGLI (1,2). The PCR program was modified in the annealing step: 45 cycles were ran with the annealing temperature set between 50◦C and 62◦C depending on the primer pair used. Obtained amplicons were sequenced in both directions using the BigDye Terminator v3.1 Cycle Sequencing Kit in the ABI Prism 310 Genetic Analyzer (Applied Biosystems, USA). The sequences of the seven genes obtained by direct amplification were analyzed and allelic profiles of L. pneumophila DNA were determined. The SBT profiles of L. pneumophila for all clinical samples were equal, namely 1, 4, 3, 1, 1, 1, 1 and they were congruent with the SBT profile of L. pneumophila isolated from the water. L. pneumophila sg.1 serving as a positive control had different allelic type (8, 10, 1, 15, 18, 1, 6). Sequence-based typing proved to be a valuable, high resolution tool for epidemiological investigations of L. pneumophila infections and in the assessment of the source of infection. Furthermore, it could be uses in the cases where no Legionella isolate from the patient specimen is available. References: 1.Gaia V, et al., (2005), J Clin Microbiol; 43, 2047; 2.Ratzow S, et al., (2007), J Clin Microbiol; 45, 1965
- 115 - Session 2: Clinical aspects a) detection and subtyping P36
Community acquired pneumonia due to Legionella wadsworthii .
Marylin LecsO-Bornet¨ a, Florence Brossiera, Isabelle Leratb, Jean Gabarrec, Capucine Morelot-Panzinib and Vincent Jarliera aGroupe Hospitalier Pitie-Salp´ etriˆ ere,` Laboratoire de Bacteriologie-HYgi´ ene,` 75013 PARIS, France; bGroupe Hospitalier Pitie-Salp´ etriˆ ere,` Service de Pneumologie et Reanimation,´ 75013 PARIS, France; cGroupe Hospitalier Pitie-´ Salpetriˆ ere,` Service d’Hematologie´ Clinique, 75013 PARIS, France [email protected]
We report a case of community acquired pneumonia due to Legionella wadsworthii in an immunocompromised host. It is, at our knowledge, the second report of a case due to this non-pneumophila Legionella. The patient, an 88-year-old man, treated for pre-lymphocyte leukemia, was admitted for pneumonia and transferred in intensive care unit. The Gram stain performed on bronchoalveolar lavage (BAL) showed 10 to 50 intra-macrophagic Gram-negative bacilli per microscopic field. Culture on BCYE agar provided, 2 days later, numerous colonies of Legionella wadsworthii. After a treatment by spiramycin plus levofloxacin for 7 days then spiramycin for 14 days, the patient recovered and was discharged.
- 116 - Session 2: Clinical aspects a) detection and subtyping P37
DNA microarray-based genotyping of Legionella pneumophila serogroup 1
Christian Luck¨ a, Stefan Moneckea, Klaus Heunerb, Peter Slickersc and Ralf Ehrichtc aInstitute of Medical Microbiology and Hygiene, German Reference Laboraotry for Legionella, Fiedlerstrasse 42, D- 01307 Dresden, Germany; bRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; cCLONDIAG GmbH, Lobstedter¨ Straße 103-105, 7749 Jena, Germany [email protected]
Legionella pneumophila is commonly detected in environmental specimens and can cause pneumonia (Legionnaires’ disease) if transmitted to susceptible persons. Due to the ubiquitous prevalence of legionellae in water supply systems, strains isolated from patients and environmental sources have to be compared by molecular typing methods. For this purpose, a diagnostic microarray was developed and used to characterise a collection of L. pneumophila serogroup 1 isolates. These included strains from the European Working Group on Legionella Infections strains collection as well as strains from the German Reference Laboratory. Genes used as targets were selected from previous studies (Thurmer¨ et al. J Med Microbiol 2009:58, 588-595, Glockner¨ et al. Int J Med Microbiol 298:411-428). They include genes associated with lipopolysaccharide synthesis and variable genetic elements. For comparison, the seven gene sequence- based typing (SBT) scheme of the European Working Group on Legionella infection (Gaia et al. J Clin Microbiol 2005: 43: 2047-2052; Ratzow et al. J Clin Microbiol 2007: 45: 1965-1968) was applied. The identification of strains was highly reproducible, discriminating and epidemiologically concordant. The DNA array facilitated rapid and reliable detection of strain-specific determinants within 24 hours. Under conditions used in a routine laboratory the array can be used for rapid and high-throughput typing of L. pneumophila serogroup 1.
- 117 - Session 2: Clinical aspects a) detection and subtyping P38
Utility of a Legionella pneumophila real-time PCR assessed using respiratory and serum samples from proven cases of Legionnaires’ disease.
Massimo Mentastia, Norman Frya, Draga Tchipevab, Baharak Afshara, Chantal Palepou-Foxleya, Patrick Kimmittb and Timothy Harrisona aHealth Protection Agency Centre for Infections, Respiratory and Systemic Infection Laboratory, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; bUniversity of Westminster, Department of Biomedical Sciences, 115 New Cavendish Street, W1W 6UW London, United Kingdom [email protected]
PCR for Legionella pneumophila has been described in many studies; however its clinical utility (sensitivity, specificity, reliability, added-value) remains largely unevaluated. We sought to determine some of these parameters using respiratory and serum samples obtained from patients with proven Legionnaires’ disease (LD). Since January 2007 respiratory specimens and sera were requested from all patients from whom a positive L. pneumophila urinary antigen (UAg) test result was obtained in our laboratory. These specimens, together with a number of historic samples, were examined on the LightCycler (Roche) using a previously established PCR specific for the L. pneumophila macrophage infectivity potentiator gene. Samples were extracted using the MagNA Pure Compact (Roche) and tested using a hybridisation probe assay which includes an internal process control. Nested-direct sequence-based typing (NDSBT) was attempted on some samples. Respiratory specimens were available from 100 patients with proven LD: 93 by detection of L. pneumophila UAg, two by direct immunofluorescence and five outbreak-associated. L. pneumophila was subsequently grown from 43 samples all of which were PCR positive: 31/57 culture-negative samples were also PCR positive (74% positive overall). Although the crossing point values for culture-negative, PCR positive samples were frequently high (i.e. low DNA content) they often yielded complete, or partial, DNA-sequence data when examined using NDSBT. A serum sample was available from 114 patients: 23 UAg-positive and culture-positive, one culture-positive and 90 UAg- positive. Of these 35 (31%) were PCR-positive. A respiratory sample was available for 35 of the 114 patients of which 24 were culture-positive: 16 (67%) sera from culture-positive patients were PCR-positive compared to only 1 (9%) serum from the 11 culture negative patients (p=0.005). When a respiratory sample can be obtained, real-time PCR offers a rapid approach to the diagnosis of LD and allows epidemiological typing data to be obtained in a timely manner. The clinical sensitivity is significantly higher than that of culture, but is likely to be lower than UAg detection for L. pneumophila serogroup 1. While sera are sometimes PCR- positive the sensitivity only approaches that of respiratory samples in culture-proven patients where antigen/DNA loading is likely to be highest.
- 118 - Session 2: Clinical aspects a) detection and subtyping P39
Detection of Legionella pneumophila using real-time PCR in the presence of other Legionella species
Anne Nor, Bente Wallervand Ofte and Ole-Jan Iversen
SINTEF Techology and Society, Lab.Centre 3rd fl east, E. Skjalgssonsgt 1, N-7465 Trondheim, Norway [email protected]
Background: This study is based upon a published real-time PCR assay (1) where primers used are common to various Legionella spp while the probes are specific to L. pneumophila. Problems arise when samples have multiple Legionella spp. that may mask L. pneumophila during melting curve analysis. The aim of the study was to vary the experimental conditions to reduce the risk of false negative PCR results of L. pneumophila. Methods: Water samples were analyzed by three different real-time PCR assays. The PCR was monitored on a LightCycler 2.0 device (Roche Diagnostics, Germany), starting with an initial denaturation step for 10 min at 95◦C, and proceeding with 50 cycles of amplification followed by a melting curve analysis with a heating rate of 0.1◦C /s. The three assays are described in Table 1. TABLE 1. Description of real-time PCR assays used in this study: Assay Amplification Melting curve analysis Reference 1 95 ◦C 5 sec 60 ◦C 15 sec 72 ◦C 15 sec 62 ◦C - 85 ◦C This study 2 95 ◦C 5 sec 58 ◦C 15 sec 72 ◦C 15 sec 40 ◦C - 85 ◦C (1) 3 95 ◦C 5 sec 58 ◦C 15 sec 72 ◦C 15 sec 40 ◦C - 85 ◦C (1) 62 ◦C - 85 ◦C(a) This study a Second melting curve analysis executed immediately after the previous Results: The 200 water samples analyzed gave the following results: Samples positive for assay 1: 67 Samples positive for assay 2: 64 Samples positive for assay 3: 69 Forty-six samples were positive for all three assays, 19 were positive for assay 1, 18 were positive for assay 2 and 3, two were positive for assay 1 and 3, and three were positive for assay 3. Conclusion: Based on these experiments, the published real-time PCR assay (assay 2) can be improved with respect to detection of L. pneumophila in the presence of other Legionalla spp by increasing both the annealing temp (assay 1) and the initial temp of melting curve analysis (assay 3). 1) Stølhaug, A., Bergh, K. (2006) Appl. Environ. Microbiol 72, 6394-6398.
- 119 - Session 2: Clinical aspects a) detection and subtyping P40
Evaluation of a new immunochromatographic rapid test for the identification of legionella from cultures.
Teresa Pardoa, Dori Fraileb, Almudena Rojasa, Luis Gonzalez´ b, Jose´ M. Delgadoa, Antonio Fuertesb and Jose´ Rojasa aVircell SL, Plaza Dom´ınguez Ortiz 1, Pol. Ind. Dos de Octubre, 18320 Santa Fe (Granada), Spain; bSGS Tecnos SA. Laboratorio de Microbiolog´ıa de Medio Ambiente y Prevencion.,´ Traspaderne 29, Edificio Barajas 1, 28042 Madrid, Spain [email protected]
Legionella is an important cause of both community-acquired and nosocomial pneumonia. Legionella pneumophila serogroup 1 is involved in most legionellosis reported cases, although other Legionella species are often found in samples from water analysis: tanks, air conditioning towers, heated swimming-pools. Water analysis for Legionella is mandatory in many countries and involves culture in microbiological media. Identification of colonies can be performed with specific immunoreagents. A new immunochromatographic test, VIRapid LEGIONELLA CULTURE, designed for the identification of Legionella in samples from bacterial cultures has been evaluated. The test includes two strips: one for Legionella genus (L) identification and a second one with two test lines, one for L. pneumophila species (Lpn) and one for L. pneumophila serogroup (sg) 1 (Lpn1). A combination of monoclonal antibodies, targeting specific constituents of the bacteria, adsorbed on the colloidal gold and the membrane, are used in the test. Three groups of samples have been analyzed to evaluate the test performance. I) Bacterial strains from reference collections (NCTC, DSMZ, ATCC): 14 L. pneumophila sg 1 strains; L. pneumophila sg 2 to 14; 15 non-pneumophila legionellae (L. longbeachae sg 1 and 3, L. bozemanii sg 1 and 2, L. dumoffi, L. micdadei, L. gormanii, L. oakridgensis, L. taurinensis, L. sainthelensi, L. rubrilucens, L. cincinnatiensis, L. quinlivanii, L. spiritensis and L. anisa.); and 13 non-legionella (NOL) bacteria. II) 141 isolated colonies from 64 water samples received at SGS Tecnos (Spain) for legionella testing; identification was carried out with a commercial latex slide test (Oxoid, UK). III) A group of clinical isolates comprising 7 Pseudomonas sp and 1 Acinetobacter sp specimens from Jaen´ Hospital (Spain). In group I, all the strains were correctly classified by VIRapid. In group II, 17 Lpn1, 13 Lpn plus 5 L colonies were identified, with 100% agreement with the reference technique for L and Lpn, and 94% for Lpn1. All the clinical isolates were negative in VIRapid. All together, the following results were obtained: 29/31 Lpn1, 28+29/26+31 Lpn, 20+57/20+57 L and 0/21+106 NOL, resulting in 98% sensitivity and 100% specificity. In conclusion, VIRapid LEGIONELLA CULTURE is a reliable, easy to handle and quick test to identify legionella from cultures.
- 120 - Session 2: Clinical aspects a) detection and subtyping P41
Novel automated MLVA typing assay for a rapid and high-throughput Legionella pneumophila genotyping tool
Daniel Sobrala, Christine Pourcelb, Fabienne Loisy-Hamona, Gilles Vergnaudb, Anne Gerard´ c and Pierre Le Cannc aCEERAM (Centre Europeen´ d’Expertise et de Recherche sur les Agents Microbiens), 1, allee´ de la Filee,´ 44244 La Chapelle sur Erdre, France; bDepartment of Genetics and Microbiology , University of Paris XI, CNRS, UMR8621, 91405 Orsay, France; cLERES (Laboratoire d’Etude et de Recherche en Environnement et Sante),´ EHESP Avenue du Professeur Leon-Bernard,´ 35043 Rennes, France [email protected]
A fast, easy-to-use and reliable genotyping method that allows real-time epidemiological surveillance would be helpful to monitor the source and spread of Legionella pneumophila. A previous Multiple Loci VNTR Analysis (MLVA) assay (Pourcel et al., 2007) demonstrated that MLVA is a valuable approach to perform epidemiological investigation of outbreaks caused by L. pneumophila. In that present work, we change, adapt and extend the existing MLVA scheme to provide high capacity analysis in an easy-to-perform way. We describe a new optimized automated multiplex capillary-based assay that ensures a higher level of discrimination and reduced time and cost requirements. Excellent epidemiological performance criteria such as discriminatory power, typability or reproducibility were observed with a set of 79 well-characterized unrelated clinical isolates provided by the reference EWGLI collection. Thus, our MLVA typing system appears to be a high-throughput screening method that combines convenience and performance criteria. It was then used for a long-term epidemiological monitoring of L. pneumophila population in Rennes, France. In less than two days, we type clinical isolates obtained from 5 patients, who contract legionellosis during the epidemic cases of 2000 and 2006 in Rennes, and 210 environmental isolates collected between 2000 and 2009 in Rennes. Twenty-three different genotypes were identified among the 215 L. pneumophila isolates. The majority cluster is almost exclusively composed of environmental isolates extracted from hot water supply systems and was not responsible for the known Rennes epidemic cases. Clinical isolates belonged to the ’Philadelphia’ group. We confrmed the outbreak source as a contaminated cooling tower from a mall and revealed an unsuspected one in an agro-industrial factory cooling tower. This study confirms previous work showing that clinically significant strains are rarely found in managed water systems and further illustrates that the prior knowledge of L. pneumophila diversity is a major tool to better control legionellosis. This work demonstrates that our MLVA procedure is perfectly suited to investigate the origin of legionellosis cases. We hope that these results will permit this technique to be used during future studies to systematically type any new isolate and include the data on internet accessible databases such as http://mlva.u-psud.fr.
- 121 - Session 2: Clinical aspects a) detection and subtyping P42
Sensitivity and specificity of the detection of Legionella pneumophila using PCR assays targeting three different genes
Nadezhda Temezhnikovaa and Alexandr Taranb aPrichernomorskaya Antiplague Station, 90 Kunikova Street, Krasnodar region, 353919 Novorossiysk, Russian Federation; bFederal State Institution Public Health ’Stavropol Research Antiplague Institute, 13-15 Sovetskaya Street, Stavropolskiy region, 353900 Stavropol, Russian Federation tnd [email protected]
Background Bacteriological methods, although relatively slow, labour-and time-consuming, remain a gold standard in laboratory diagnosis of legionellosis and Legionella pneumophila detection in specimens from industrial conditioning and water supply systems. Development and implementation of molecular assays will potentially allow for a significant increase in sensitivity and reduction in turnaround times for the detection of Legionella pneumophila in clinical and environmental specimens. In the current study we evaluated the performance of three molecular assays (Legionella REF, Russia; Amplisens°R Legionella pneumophila-FL, Russia; GENePak°R DNA PCR test, Russia) based on PCR detection of specific targets in ftsZ, mip i lpn genes respectively. Materials and methods Evaluation was conducted using bacterial suspensions prepared using L. pneumophila Philadelphia I strain (ATCC 33152), water specimens collected from systems of drinking water supply, conditioning and heating (n=51), and sputum specimens collected from patients with pneumonia of unknown etiology (n=10). All specimens were tested twice using three different molecular assays. All clinical specimens, as well as specimens obtained from the water supply and conditioning systems, were tested using standard bacteriological methods as well. Crude DNA was extracted using DNA-sorb kits (Central Epidemiology Research Institute, Moscow, Russian Federation). Results Both GENePak°R DNA PCR test and Amplisens°R Legionella pneumophila-FL molecular tests demonstrated higher (100 CUF/ml) sensitivity when testing the microbial suspension compared to the Legionella REF assay (1000 CUF/ml). Results of PCR test of all clinical specimens, as well as specimens collected from the systems of heating, water supply, and conditioning, were negative. No Legionella pneumophila were isolated from the clinical and environmental specimens using bacteriological methods. We concluded that all three molecular assays are 100% specific for the detection of Legionella, while GENePak°R DNA PCR test and Amplisens°R Legionella pneumophila-FL are more sensitive (sensitivity threshold 100 CUF/ml).
- 122 - Session 2: Clinical aspects a) detection and subtyping P43
Genomic markers of Legionella pneumophila Paris strains
Mike Vergnesa, Christophe Ginevrab, Max Maurina, Philippe Normandc, Jean Thiouloused,Jer´ omeˆ Etienneb, Sophie Jarraudb, Elisabeth Kaya and Dominique Schneidera aLAPM, CNRS UMR5163, Universite´ Joseph Fourier, Institut Jean Roget, 38042 Grenoble, France; bCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; cEcologie Microbienne UMR5557 CNRS, Universite Lyon1, 69622 Villeurbanne, France; dBiometrie´ et Biologie Evolutive CNRS UMR 5558, Universite Lyon1, 69622 Villeurbanne, France [email protected]
The causative agent of legionellosis, Legionella pneumophila, colonizes all water networks providing the entry gate for human infections. To offer efficient treatments of both contaminated patients and polluted environmental installations, fast and accurate identification of the bacterial source of the infection is necessary. Two methodologies currently prevail for Legionella: 1/ phenotypic methods are based upon biochemical or immunological traits, the most frequently used method being serotyping. 2/ Two molecular methods are based on the diversity of nucleic acids. The first one is Pulsed Field Gel Electrophoresis (PFGE) and the second is Sequence Based Typing (SBT). These molecular methods, used in clinical bacteriology, have distinguished four groups within L. pneumophila serogroup 1, which is the major infectious Legionella group: L. pneumophila Philadelphia, Paris, Lorraine and Lens isolates. This work aimed at characterizing new genomic markers with two main features: 1/ a significantly higher discrimination power within L. pneumophila serogroup 1 strain Paris, and 2/ a higher identification power of the source strain during an infection episode. From the available genome sequences of L. pneumophila, we identified six different types of Insertion Sequence elements (IS), all potentially functional. We used them as genomic markers in hybridization experiments with genomic DNA from 93 isolates of L. pneumophila serogroup 1 strain Paris, including both clinical and environmental isolates, which have been responsible for different epidemic episodes in France between 1995 and 2007. Three major results were obtained: 1/ one of the six IS types, ISL3, revealed a high discrimination power, with 63 % of the tested isolates revealing a unique Restriction Fragment Length Polymorphism (RFLP) profile. 2/ In most cases, each clinical isolate revealed an ISL3 RFLP profile that was recovered in at least one environmental isolate from the same geographical location, thereby suggesting the discovery of the source of infection. 3/ A single clone, showing a unique RFLP profile, was repeatedly recovered during infectious episodes occurring over several years in specific geographical locations, probably suggesting re-emergence of the same clone and/or inefficient decontamination of the corresponding environments.
- 123 - Session 2: Clinical aspects a) detection and subtyping P44
Characterization of Legionella pneumophila isolates in Norway by sequence typing and monoclonal sero- and subgroups.
Elisabeth Wedege, Karin Bolstad, Elisabeth Fritzsønn and Dominique A. Caugant
Norwegian Institute of Public Health, P.O.Box 4404 Nydalen, NO-0403 Oslo, Norway [email protected]
Legionellosis has been a notifiable disease in Norway since 1980. Two large outbreaks occurred in 2001 and 2005 with 28 and 103 cases, respectively. The source of the first outbreak was a cooling tower, whereas the latter outbreak was caused by an industrial air scrubber. In between these two outbreaks about 25 cases were reported yearly, corresponding to an incidence of 0.6 per 100 000. About half of the cases were infected abroad. At our reference laboratory we have analysed 43 patient and 93 environmental L. pneumophila isolates (n=136), sampled between 2001 and 2009. Monoclonal sero- and subgroups were determined by dot-blotting of whole-cell suspensions with the Dresden panel of monoclonal antibodies (Mabs). Sequence types (ST) was analysed according to Ratzow et al. (2007). Most isolates (n = 106; 77.9%) belonged to serogroup 1, and 86 (81.1%) expressed the virulence-associated epitope for Mab 3/1. Among the serogroup 1 isolates, Benidorm with 51 (48.1%) isolates and France/Allentown with 21 (19.8%) isolates were the most frequent subgroups, followed by OLDA (10.3%), Bellingham (8.5%), Knoxville (7.5%), Philadelphia (4.7%), and Heysham (0.9%). The isolates were assigned to 28 STs, of which 15 STs were represented among the 43 patient isolates. Thirty-six (90%) serogroup 1 patient isolates expressed the Mab 3/1 epitope, and the majority belonged to subgroups Benidorm and France/Allentown. Patients in the 2001 outbreak were infected by a subgroup France/Allentown strain of ST40, whereas those in the 2005 outbreak strain were infected by subgroup Benidorm of ST15. Environmental samples taken in relation with the two outbreaks showed the corresponding characteristics. Patients in other parts of the country usually fell ill with other serogroup 1 or non-serogroup 1 strains. In 2006, a subgroup 4 Portland strain dominated in samples taken from biodegradation ponds close to the air scrubber that caused the 2005 outbreak, but this strain was not isolated from patients. This strain did not carry the neuA gene, nor did two serogroup 3 and 10 environmental isolates, as well as a subgroup Philadelphia patient isolate. Thus, these isolates could not be assigned an ST number.
- 124 - Session 2: Clinical aspects a) detection and subtyping P45
Three sporadic cases of L. longbeachae Legionnaires Disease in Austria, 2008 and 2009
Daniela Schmida, Alexander Indraa, Marion Blaschitza, Barbara Robla, Petra Hasenbergera, Ingrid Gruner¨ a, Wolfgang Prammerb, Gerhard Leitnerc and Gunther¨ Wewalkaa aAustrian Agency for Health and Food Safety, Wahringer¨ Straße 25a, A - 1096 Vienna, Austria; bKlinikum Wels- Grieskirchen, Grießkirchner Straße 42, A - 4600 Wels, Austria; cLandeskrankenhaus Leoben, Vordernberger Straße 42, A - 8700 Leoben, Austria [email protected]
In 1980, L. longbeachae was first identified in two Californian patients with pneumonia. Since then, most cases of infection with L. longbeachae have been identified in Australia. Infections with L. longbeachae are associated with contact to potting soil and occur predominantly in immunocompromised patients. In Europe, Legionella longbeachae is rarely associated with human disease or isolated from the environment. According to the European Working Group for Legionella Infection (EWGLI) database, there have been 22 cases of infection with L. longbeachae reported between 1995 and 2007. We report on three sporadic cases of community-acquired pneumonia caused by Legionella longbeachae in Austria. Methods of establishing the diagnosis included tests for urinary Legionella antigen, serum antibody detection, analyses of respiratory secretions by light-cycler PCR, culture, and sequence determination of the 16S rRNA gene of the culture isolates confirmed by sequence-based classification using the mip gene database. Source identification included analysis of potting soil samples from case’s garden and relevant water system samples. The environmental samples were analysed by light-cycler PCR and culture. Three patients with community-acquired pneumonia from three of the nine Austrian provinces were hospitalized in July 2008 (n=1) and June 2009 (n=2). Tests for urinary Legionella antigen were negative in all three cases. Following detection of L. non-pneumophila DNA in bronchoalveolar fluid by PCR, L. non-pneumophila was isolated from bronchoalveolar fluid and identified as L. longbeachae by sequencing of the 16S rRNA gene and the mip gene. The median age of the three male cases was 67 years (range 61-70). All three cases had an immunocompromised comorbidity and one case died 9 days after disease onset. All three cases reported exposure to potting soil prior to disease onset. Potting soil samples obtained from commercially available potting soils used for gardening by two of the cases were available for analyses. Although culture did not grow Legionella, Legionella longbeachae DNA was detected by PCR. All analysed water samples were negative for Legionella. In 2008, a total of 7 cases of L. longbeachae infection were reported in Europe. The increasing number of cases and detection of L. longbeachae in potting soil indicate that potting soil may gain increasing importance as alternative reservoir for Legionella spp. in Europe.
- 125 -
SESSION 3 HOST - MICROBE INTERACTIONS A) SECRETION SYSTEMS AND THEIR SUBSTRATES
- 126 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P46
The mechanism of action of VipA, a Legionella pneumophila actin-binding effector
Irina Francoa, Nadim Shohdya and Howard Shumanb aDepartment of Microbiology, Columbia University Medical Center, 701 West 168th Street, New York, NY NY 10032, United States of America; bColumbia University Medical Center, Department of Microbiology & Immunology, 701 West 168th Street, New York, NY 10032, United States of America [email protected]
Legionella pneumophila, the causative agent of Legionnaires’ disease, invades and replicates inside phagocytic cells. For a successful infection, Legionella has to avoid degradation by the host cell and establish its own replicative niche. In fact, immediately after uptake the bacteria prevent fusion with the lysosome and recruit ER- and Golgi-derived material. This results in the formation of a specialized Legionella-containing vacuole (LCV) and is accomplished by modification of host cell pathways by the action of Legionella effector proteins, which are translocated into the host cell via the Icm/Dot type IVB secretion system. The Legionella effector collection contains to date more than 150 proteins. Most of these are not required for intracellular replication, which makes it challenging to identify them and define their precise functions. One successful method for their identification relied on ectopic expression in yeast and screening for defects in Vacuolar Protein Sorting (Vps- phenotype). Among the Icm/Dot substrates affecting vacuolar trafficking is the VipA protein (VPS inhibitor protein A). In this work we identified actin as an interacting partner for VipA in pull-down assays. Actin polymerization assays showed that purified, recombinant VipA protein interferes in the growth of F-actin filaments, exerting effects on both actin nucleation and elongation. A VipA mutant, VipA-1 no longer causes a defect in vacuolar trafficking and exhibits a decreased affinity for actin, indicating that actin binding is related to the normal function of VipA. Localization of VipA in eukaryotic cells was assayed by ectopic expression of VipA-GFP in yeast and mammalian cells. VipA-GFP localizes to distinct puncta in both cell types. In yeast the puncta are associated with the same type of pre-vacuolar compartment seen in mutants defective for multivesicular body formation and partially co-localizes with actin patches. In mammalian cells, VipA-GFP associates with short cortical actin filaments that remain after treatment with Cytochalasin D. In contrast, the VipA-1-GFP mutant protein exhibits homogeneous cytosolic localization in both yeast and mammalian cells. These results indicate that the ability of VipA to bind actin is related both to its specific subcellular location as well as its role in modulating organelle trafficking.
- 127 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P47
Molecular characterization of the Dot/Icm-translocated AnkH and AnkJ eukaryotic- like effectors of Legionella pneumophila
Fabien Habyarimanaa, Souhaila Al Khodora, Chris Pricea, Marina Santicb and Yousef Abu Kwaika aUniversity of Louisville, 319 Abraham Flexner Way Rm 410, Louisville, KY 40202, United States of America; bUniversity of Rijeka, Brace Branchetta 20, 51000 Rijeka, Croatia [email protected]
The eukaryotic-like AnkH and AnkJ proteins of Legionella pneumophila are required for intracellular proliferation. In this report, we show that expression of ankH and ankJ is temporally triggered by a network of interacting regulatory cascades. Intracellular bacteria translocate AnkH and AnkJ and 5 other Ank proteins into the macrophage cytosol via the Dot/Icm type IV secretion system. The IcmSW chaperones are essential for translocation of AnkJ but not AnkH. The ankH and ankJ mutants are severely defective in intrapulmonary proliferation in mice. The 10 C-terminal residues and the ANK domains of AnkH and AnkJ are required for translocation. Expression of AnkH and AnkJ fusions within HEK293 cells show a punctuate distribution in the cytosol but no association with endocytic vesicles, the Golgi apparatus or the ER. Interestingly, the defect in intracellular proliferation of the ankH or ankJ mutants is rescued in HEK293 cells expressing the respective effector. We conclude that AnkH and AnkJ are effectors translocated by the Dot/Icm system by distinct mechanisms and modulate distinct cytosolic processes in the host cell. This is the first demonstration of a trans-complementation of an effector mutant of L. pneumophila through expression of the respective effector in the host cell.
- 128 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P48
Influence of regulatory elements of Legionella pneumophila on expression of lipopolysaccharides in liquid cultures
Johanna Kuhn,¨ Cacilia¨ Muller,¨ Esteban Fernandez Morera and Jurgen H. Helbig
Institute of Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany [email protected]
The intracellular pathogen Legionella pneumophila has a very complex life style. After uptake by natural host cells or monocytes followed by the ”pregnant pause” inside the phagosomes the transmissive, infective phenotype switches to the replicative, non-infective type. After several multiplication steps bacteria are motile again, lyse the host cells and are able to start the next infection cycle. Responsible genomic elements were described almost in detail. The aim of our study was to investigate, if deletion mutants of regulatory elements differ in their LPS expression compared to the wild type. For this we have tested strain Lp02 and its csrA-, flaA-, fliA-, letA-, milX-, and rpoS-mutant. Strains were cultivated in liquid media until exponential or post-exponential growth phase representing the replicative, non-infective phase and the transmissible, infectious phase inside host cells, respectively. Cell-wall bound and shed LPS were tested by ELISA for quantification of six different LPS-epitopes. The expression of the serogroup 1-specific epitope recognized by MAb 8/5 as well as the MAb 88/3-epitope are not influenced in any mutant, neither as cell-wall bound nor as shed LPS. Phase variability is also not valid. It is to assume that these structures are essential for the bacteria. LPS carrying the epitopes of MAbs 3/1, 8/4, 12/2 and 88/1 were phase variably expressed. With exception of MAb 3/1-positive LPS all other components were not shed, which demonstrates their failed influence on modulation of the early host cell phagosomes. Nevertheless, MAb 8/4-epitope is over-expressed in the post-exponential growth phase. The ”virulence-associated” MAb 3/1-epitope is phase variably expressed and shed. Compared to the wild type, the concentration of this LPS is up to fivefold higher in the released LPS fractions. Taken together, LPS expression is influenced by known regulatory genes and its favourites the Legionella-LPS as a co-factor of virulence development.
- 129 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P49
Comparative analysis of the role of protein-kinases in Legionella pneumophilavirulence
Eva Herveta, Xavier Charpentierb,Melanie´ Leveta, Anne Vianneya and Patricia Doubleta aUMR 5240 Microbiology, Adaptation and Pathogeny CNRS-Lyon1-INSA-BayerCropscience, BAT A. Lwoff, 10 rue Dubois. Campus de la Doua, 69622 Villeurbanne, France; bDepartment of Microbiology, Columbia University Medical Center, 701 West 168th Street, New York, AK NY 10032, United States of America [email protected]
Legionella pathogenic strains are emerging from the environment after intracellular multiplication in amoeba. Its ability to exploit the basic cellular mechanisms of numerous protozoa also enables Legionella to infect human macrophages and pneumocytes of alveolar lungs. The complex regulatory circuits involved in the control of host functions and the intracellular life of L. pneumophila are still poorly understood. We focused on a particular hypothesis according to which the infectious cycle of L.pneumophila would be controlled by protein phosphorylation. Analysis of the genomes of L. pneumophila suggested the occurrence of 5 genes encoding eukaryotic- like protein-kinases. First, we have characterized the biochemical activities of these enzymes in order to establish if these protein-kinases are functional in terms of phosphorylation. Each enzyme has been overproduced and purified. Its ability to autophosphorylate and phosphorylate protein substrates has been assessed by performing in vitro phosphorylation assays. The specificity of protein-kinase has been determined by analysing the phosphoaminoacids, and catalytic mechanism of the phosphorylation reaction has been established by site-directed mutagenesis experiments. Then, we have investigated the role of each protein-kinase in the virulence of Legionella. L. pneumophila mutants deleted of one protein-kinase encoding gene have been constructed by insertional mutagenesis. Virulence of the mutant strains towards amoeba and macrophages has been compared to the parental strain. One mutant strain was impaired in virulence, more precisely in the early steps of the infectious cycle. This result demonstrates, for the first time, that at least one protein-kinase plays a key role in Legionella virulence. Finally, in order to establish if these protein-kinases could control host cell functions, we have determined whether these enzymes are translocated into the host cell cytoplasm during infection. By using a real-time translocation assay, we have established that four among the five kinases were indeed effectors of the type IV secretion system Dot/Icm. Our final aim is at characterizing at the molecular level the precise role of protein-kinases in L. pneumophila virulence. Host cell proteins that are phosphorylated and/or interact with Legionella protein-kinases during infection will be identified. This will provide data about the signal transduction pathways controlled by Legionella, which allow bacteria to effectively infect eukaryotic cells.
- 130 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P50
Host Prenylation of Legionella CaaX Box Proteins
Stanimir Ivanova and Craig Royb aYale University, 295 Congress Ave, BCMM, rm 347, New Haven, AK 06536-0812, United States of America; bYale University, 295 Congress Ave, BCMM, rm 347, New Haven, CT 06536-0812, United States of America [email protected]
Subcellular localization is an important component of protein function. Localization of proteins to membrane-bound organelles is often facilitated by lipidation. In eukaryotes, one form of protein lipidation involves the covalent modification of a cysteine residue located in a CaaX motif in the protein by farnesyltransferase (FT) or geranylgeranyltransferase (GGT) enzymes. To determine whether Legionella pneumophila effectors might subvert host lipid transferases to assist in their subcellular localization, in silico analysis of the proteome was conducted and ten polypeptides were identified containing a CaaX motif predicted to be recognized by host lipid transferases. Immunofluorescence analysis of ectopically produced Legionella CaaX Box Proteins (CBPs) revealed localization at the plasma membrane and at endomembrane compartments. For many of the CBPs, substitution of serine for the cysteine residue in the CaaX box motif interfered with membrane localization. Inhibitors of host lipid transferases also redistributed the Legionella CBPs to the cytosol. Additionally, inhibition of host prenyl transferases during the early stages of Legionella infection perturbed vacuole remodeling and resulted in an increase in the percentage of Legionella residing in LAMP1-positive vacuoles. These data indicate that several Legionella CBPs are prenylated by host FT and GGT enzymes after translocation of into the host cytosol, and that host prenylation facilitates localization of these effectors to membranes. Because perturbation of prenyltransferase activity interfered with trafficking of vacuoles containing Legionella, we hypothesize that lipidation plays an important role in regulating the activities of the CBPs spatially and temporally during infection.
- 131 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P51
The role of the secreted effector protein SidJ in virulence of Legionella pneumophila
Kwang Cheol Jeong and Joseph Vogel
Washington University School of Medicine, 660 S. Euclid Avenue, Campus Box 8230, St. Louis, MO 63110, United States of America [email protected]
L. pneumophila replicates inside alveolar macrophages and causes an acute, potentially fatal pneumonia called Legionnaires’ disease. Growth of L. pneumophila inside phagocytic cells is completely dependent on the Dot/Icm type IVB secretion system (T4SS), which alters the endocytic pathway of host cells. Recently a large number of Dot/Icm substrates have been identified, including the SidE family of proteins (SdeA, SdeB, and SdeC). We previously reported that members of the SidE family are secreted into host cells, where they localize to the cytoplasmic face of theLegionella containing vacuole (LCV), and are important for the virulence ofL. pneumophila inside amoebae. We have also demonstrated by an immunofluoresence assay that the SidE proteins are present on the LCV early during infection but the signal disappeared from the LCV at later points of the infection. In this study, we examined the interaction between the SidE family proteins and SidJ, another T4SS substrate that is encoded within the sdeCBA locus. We have now obtained data supporting a model where SidJ modulates the function of SidE proteins within host cells, thereby allowingL. pneumophila to survive and replicate within host cells. Our model is based on four lines of evidence. First, the SidE proteins and SidJ are each equally required for optimal intracellular growth ofL. pneumophila. Deletion of all four sidE genes or sidJ alone resulted in partial attenuation of growth of L. pneumophila in the protozoan host Acanthamoeba castellanii . Second, over-expression of SdeA in a L. pneumophila strain lacking sidJ completely inhibited the intracellular growth of L. pneumophila. Third, SdeA is toxic when expressed in yeast, but this toxicity can be suppressed by co- expression of SidJ. Finally, we have observed that disappearance of the SidE proteins on the LCV at later points of infection requires SidJ. Taken together, these data suggest that the SidE proteins function as toxins and SidJ down modulates SidE function, thereby preventing lethal intoxication of host cells during intracellular growth ofL. pneumophila.
- 132 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P52
The Trb/Tra conjugation/type IVA secretion system of Legionella pneumophila Corby
Monika Lautner, Christine Ott and Klaus Heuner
Robert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany [email protected]
Background: Exchange of genetic information by horizontal gene transfer is an important mechanism for the evolution of bacterial genomes. Primarily the conjugation machinery represents a basic mode for the transfer of DNA between bacterial cells or between bacterial and eukaryotic cells. Recently we identified and characterized a new trb/tra conjugation type IVA secretion system in the genome of Legionella pneumophila Corby. Two similar versions of this conjugation system are localized on two seperate genomic islands (Trb-1 and Trb-2) integrated within the tRNAPro gene (lpc2778) and the tmRNA gene (lpc0164), respectively. Both islands exhibit an oriT region and encode all essential trb/tra genes require for conjugation. We also could represent that Trb-1 as well as Trb-2 can exist in an integrated chromosomal or an episomal circular form. Methods: Generation of mutant strains by reciprocal recombination after natural transformation of the construct. Conjugal transfer was performed on BCYE agar plates for 24 h at 30 ◦C. Southern blot hybridization and PCR was used to analyse the distribution of several trb/tra genes among various Legionella strains. Results: By deleting the site-specific integrase (int-1, lpc2818) of Trb-1, we could demonstrate that the excision and forming of an episomal circle is int-1 dependent. The whole genomic island can be transferred to other Legionella strains and integrated site-specifically into the genome of the transconjugants. The int-1 mutant was also tested to determine their abilities to transfer Trb-1 by conjugation to the same recipient. First results indicate that conjugation frequencies were lower, compared to the parental strain. Additionally we analysed the distribution of the trb/tra gene cluster within the genus Legionella by Southern hybridization. On the basis of these results we could conclude that similar or identical trb/tra regions also present in various L. pneumophila and non-pneumophila strains. Conclusions: Here we described a new type IVA secretion system of L. pneumophila Corby encoding a conjugation system, which mediates the transfer of chromosomal DNA regions in Legionella. The trb/tra system may also be widespread within the genus Legionella. Furthermore we showed that the site-specific integrase, present on the genomic island is necessary for the mechanism of excision and integration. Functional analysis of the mobile genomic island Trb-1 is already in progress.
- 133 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P53
The Role of Type II Secretion during Lung Infection by Legionella pneumophila
Kessler McCoy-Simandle and Nicholas Cianciotto
Northwestern University, 320 E Superior St, Searle 6-541, Chicago, IL 60611, United States of America [email protected]
The type II protein secretion (T2S) system of Legionella pneumophila is required for optimal intracellular infection of macrophages as well as growth in the lungs of infected A/J mice. To determine if T2S also promotes infection of lung epithelial cells, we tested multiple, independently-derived T2S mutants of L. pneumophila strain 130b for their relative ability to infect three different alveolar epithelial cell lines. All of the mutants, but not complemented derivatives, were impaired ca. 10-fold for growth within a human type II epithelial line (A549), a human type I epithelial line (WI-26 VA4), and a murine alveolar epithelial line (TC-1). To determine if T2S also influences infection of activated macrophages, we tested ability of the T2S mutants for their ability to infect and survive in gamma interferon activated macrophages. The mutants behaved similarly to wild type, exhibiting an interferon dose-dependent inhibition of growth. To further investigate the role of T2S in infection, ongoing experiments are assessing the inflammatory responses that occur in the murine lung following infection with a T2S mutant vs. the wild-type strain.
- 134 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P54
Legionella pneumophila PmiA contributes to infectivity to macrophages and protozoa by stabilizing DotA, a component of the Icm/Dot typeIV secretion apparatus.
Masaki Miyake, Yasunori Iwasaki, Tsuyoshi Hayashi, Hitomi Koike and Yasuyuki Imai
University of Shizuoka School of Pharmaceutical Sciences, 52-1 Yada, Suruga-ku, 422-8526 Shizuoka, Japan [email protected]
Legionella pneumophila is an opportunistic pathogen which replicates within protozoan hosts in natural environment. The ability of L. pneumophila to cause pneumonia in human is dependent on intracellular replication within alveolar macrophages. The type IV secretion apparatus Icm/Dot translocates multiple effectors into host cells and is essential for the ability of L. pneumophila to evade endocytic fusion, to remodel the phagosomes by the endoplasmic reticulum, and to replicate intracellularly. We previously identified pmiA gene which is associated with intracellular replication within host cells. The encoded protein PmiA was predicted as a transmembrane protein by secondary structural analysis. We confirmed that PmiA is a membrane protein by subcellular fractionation of a L. pneumophila strain expressing M45 epitope-tagged PmiA (GB112/pM45-PmiA) and Western blotting analysis using antibodies against the M45 tag. We examined whether PmiA is associated with L. pneumophila activities dependent on the Icm/Dot system, because PmiA has commom features to the Icm/Dot system as to intracellular growth factors of L. pneumophila which locate in the bacterial membrane. As a result, a mutation of pmiA gene decreased all of Icm/Dot-dependent activities which we examined, such as evasion of bacterial phagosomes from lysosomes, pore-formation to mammalian cells, translocation of effectors to host cells, apoptosis induction of host cells, conjugative transfer of RSF1010 plasmid, and extracellular secretion of DotA which is a component of the Icm/Dot system. It is suggested that PmiA is involved in the Icm/Dot functions. With regard to the decreased DotA secretion in the pmiA mutant, Western blotting of whole bacterial lysates using anti-DotA polyclonal antibodies was performed to examine the production of DotA. It was observed that the DotA band was fragmented in the pmiA mutant. The complementation of the pmiA gene prevented the fragmentation of DotA. There was no difference of the mRNA expression level of DotA between the pmiA mutant and the parent strain on RT-PCR. Furthermore, by the pull-down assay using the whole bacterial lysate of GB112/pM45-PmiA by means of antibodies against DotA or M45 tag, it was shown that PmiA interacts with DotA. We conclude that PmiA stabilizes DotA by their interaction to maintain the regular function of the Icm/Dot system.
- 135 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P55
Characterization of the type IVB secretion system coupling protein IcmO/DotL of Legionella pneumophila
Catherine A. Mueller and Howard Shuman
Columbia University Medical Center, Department of Microbiology & Immunology, 701 West 168th Street, New York, NY 10032, United States of America [email protected]
The Icm/Dot type IVB secretion system of Legionella pneumophila is essential for intracellular growth of the bacteria in all known hosts. It translocates a relatively large number of diverse effector proteins (> 150) into both protist and host cells. These translocated effectors mediate the intracellular events that are required for survival and replication in host cells. In the absence of a functional Icm/Dot type IVB secretion system, effectors are not translocated and the bacteria are unable to to avoid intracellular degradation. In contrast to the type IVA secretion system, for which parts of the structure have been solved, the interactions between the different components and the structure of the type IVB secretion system remain mostly unclear. One component of the Type IVB secretion system, IcmO/DotL, is homologous to the known type IV coupling proteins TrbC (R64/IncI) and VirD4/TraG (Type IVA Vir/IncP). Type IV coupling proteins (T4CP) are thought to act as intracellular receptors targeting secreted substrates to the secretion machinery and possibly preparing them for secretion. Interactions between the IcmO/DotL coupling protein and components of the type IVB secretion machinery and effectors were studied using a bacterial two-hybrid approach. As previously reported (Vincent et al. 2006), IcmO/DotL was found to interact with IcmJ/DotN. In addition, purification of IcmO/DotL in complex with its interaction partners will help characterize the role of the coupling protein in substrate secretion.
- 136 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P56
The Legionella pneumophila type II secretion system secretes an endoglucanase, a eukaryotic-like protease, and a novel effector that is required for virulence.
Meghan Pearce and Nicholas Cianciotto
Northwestern University, 320 E Superior St, Searle 6-541, Chicago, IL 60611, United States of America [email protected]
The Legionella pneumophila (Lp) type II secretion pathway (Lsp) secretes at least twenty-five effectors and is required for full virulence. We now report three newly identified Lsp effectors, designated as CelA, LegP and Lpg0264. CelA proved to be the first example of an endoglucanase produced by Lp, although it was not required for infection of protozoa (Hartmannella vermiformis, Acanthamoeba castellanii) or mammalian hosts (U937 macrophage-like cells, A/J mice). The second effector, LegP, shared similarity with eukaryotic zinc-metalloproteases although it was most similar to prokaryotic hypothetical proteins. We have confirmed, using anti-LegP antibodies, that the secretion of LegP by legionellae growing in extracellular broth is dependent upon type II (but not type IV) secretion, whereas others have demonstrated intracellular translocation of LegP to be type IV dependent. Together, these data are the first to suggest that an effector may use two different secretion systems to exit the bacteria depending on its environment. LegP was not required for infection of amoebae (H.vermiformis, A.castellanii) or mammalian cell lines (U937 cells and WI-26 VA3 and A549 epithelial cells). In an in vivo competition assay, legP mutants competed well with wild-type Lp for growth, suggesting that LegP is not required for lung infection. The last effector, Lpg0264, was a novel protein. Multiple independently derived lpg0264 mutants were significantly impaired for infection of protozoa (H. vermiformis, A. castellanii), and mammalian cell lines (U937, A549, WI-26 VA3). In competition-based mouse infection, Lp lacking Lpg0264 exhibited an 80-fold decrease in growth compared to wild type. This is the largest in vivo defect of any type II effector mutant tested to date and the first case of a type II-dependent effector mutant being impaired in both environmental and mammalian hosts. These data suggest a large role for Lpg0264 in Lp ecology and virulence.
- 137 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P57
Sliding motility by Legionella pneumophila
Catherine Stewart and Nicholas Cianciotto
Northwestern University, 320 E Superior St, Searle 6-541, Chicago, IL 60611, United States of America [email protected]
Legionella pneumophila exhibits sliding motility when it is grown on a buffered charcoal yeast extract (BCYE) containing 0.5 to 1.0% agar. After 7 to 22 days of incubation, spreading legionellae appear in an amorphous, lobed pattern that is most manifest at 25-30◦C. All nine L. pneumophila strains examined displayed the phenotype. Surface translocation was also exhibited by some, but not all, other Legionella species examined. L. pneumophila mutants that were lacking flagella and/or type IV pili behaved as the wild type did when plated on low-percentage agar, indicating that the surface translocation is not swarming or twitching motility. A translucent film was visible atop the BCYE agar, advancing ahead of the spreading legionellae. Based on its abilities to disperse water droplets and to promote the spreading of heterologous bacteria, the film appeared to manipulate surface tension and, as such, acted like a surfactant. L. pneumophila type II secretion mutants, but not their complemented derivatives, were defective for both surface translocation and film production. In contrast, mutants defective for type IV secretion exhibited normal surface translocation. Thus, L. pneumophila exhibits a form of surface translocation that is most akin to ”sliding motility” and uniquely dependent upon type II secretion. A screen of mini-Tn10-mutagenized L. pneumophila was utilized to identify genes involved in sliding motility and/or surfactant production/secretion. From the 1536 mutants examined, 5 were found to be defective for both sliding motility and surfactant production. One mutant contained an insertion in the gene encoding the outer membrane protein TolC. A second had its insertion in a gene encoding a putative acetyltransferase, and a third in a gene encoding a potential acyl-CoA synthetase. The remaining two had insertions in genes encoding hypothetical proteins. By testing these mutants, we have determined that sliding motility and surfactant production are not required for intracellular growth of L. pneumophila in Acanthamoeba castellanii or U937 macrophage-like cells.
- 138 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P58
Characterization of the IcmS/IcmW-DotL interaction in type IVB substrate secretion
Molly Sutherlanda, Carr Vincenta, Victor Tsenga, T. Linh Nguyena and Joseph Vogelb aWashington University in St. Louis, 660 S. Euclid Avenue, Campus Box 8230, Saint Louis, MO 63110, United States of America; bWashington University School of Medicine, 660 S. Euclid Avenue, Campus Box 8230, St. Louis, MO 63110, United States of America [email protected]
The Legionella pneumophila type IVB secretion system is encoded by twenty-six dot/icm genes. This secretion system is used by the bacterium to inject a large number of substrates into the host cell cytoplasm and is required for the bacterium’s ability to survive and replicate within host cells. In recent years there has been significant progress on the identification and function of Legionella type IV secretion effectors, but the molecular mechanism of substrate secretion is still not well understood. Therefore, we have focused on determining how substrates are secreted, specifically the interaction between the type IVB secretion apparatus and substrates. The type IVB secretion system contains two subcomplexes: the core subcomplex (DotC, DotD, DotF, DotG, and DotH) and the type IV coupling protein subcomplex (DotL, DotM, DotN, IcmS, and IcmW). DotL, the type IV coupling protein (T4CP), is a putative ATPase and has been proposed to act as the major membrane receptor for Dot/Icm substrates. Substrates are targeted for export via a C-terminal signal sequence and can be divided into two main classes based on their requirement for the type IVB adaptor proteins, IcmS/IcmW. Several IcmS/IcmW-dependent substrates have been shown to bind IcmS/IcmW via internal regions of the proteins distinct from their C-terminal signal sequences. Intriguingly, DotL was also shown to bind to IcmS/IcmW, although the importance of this observation was unknown. To ascertain DotL’s role in substrate secretion, we performed a large-scale mutagenesis of dotL resulting in the isolation of forty-four dotL mutants, each containing a single amino acid change. Within a subset of dotL mutants isolated to be defective for growth inside host cells, one mutant was of particular interest because it was unable to interact with IcmS/IcmW. This mutant suggests that the binding of IcmS/IcmW to DotL is important for export of substrates into host cells. To further elucidate the significance of the DotL-IcmS/IcmW interaction, we are currently mapping the IcmS/IcmW- binding domain of DotL and are exploring its role substrate export by the Legionella type IVB secretion system.
- 139 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates P59
Cytochromes c3 and c4 and TolC promote the production and secretion of the Legionella pneumophila siderophore
Emily Yip and Nicholas Cianciotto
Northwestern University, 320 E Superior St, Searle 6-541, Chicago, IL 60611, United States of America [email protected]
Iron plays an essential role in intracellular infection and virulence of Legionella pneumophila, however, the molecular mechanism of Legionella iron acquisition is relatively unknown. Previously, a transposon mutagenesis study revealed a locus of cytochrome c maturation (ccm) genes that promote L. pneumophila growth under iron-limited extracellular conditions and during intracellular growth in protozoa and macrophages. To assess the role of ccm in L. pneumophila siderophore (legiobactin) expression, supernatants of representative ccm mutants (ccmB, ccmC, ccmF) were tested for CAS activity and the ability to restore growth of a ferrous-iron uptake mutant (feoB) in deferrated media. The ccm mutants exhibited a decreased in CAS activity from 30% to 70% compared to that of wild type. In addition, their supernatants failed to restore growth of the feoB mutant. In gram-negative bacteria, ccm genes encode the transport of heme across the inner membrane and the subsequent maturation of apocytochrome c into cytochrome c. To examine the role of c-type cytochromes in siderophore expression by L. pneumophila, mutations were constructed in the genes encoding cytochrome c1, c3, c4 and c5. Although the c3 and c4 mutants grew normally on iron-limiting media and inside protozoa, they had a decrease in CAS activity. The c1and c5 mutants exhibited an opposite phenotype in which they produced a wild- type level of CAS activity but were defective for intracellular replication. These data indicate that Ccm and two of the cytochrome c molecules are necessary for optimal expression of siderophore by L. pneumophila. Despite intensive studies on siderophore uptake, little is known about how siderophores are secreted across the gram-negative outer membrane. Recent work demonstrated that the outer membrane protein TolC is involved in export of enterobactin by Escherichia coli. Interestingly, multiple independently derived tolC mutants of L. pneumophila produced supernatants with reduced siderophore activity and an inability to restore growth to the feoB mutant. These data indicate that TolC is also involved in the secretion of legiobactin, suggesting that this outer membrane protein constitutes a general siderophore secretion pathway in Gram-negative bacteria.
- 140 -
SESSION 4 HOST - MICROBE INTERACTIONS B) METABOLISM AND VIRULENCE
- 141 - Session 4: Host - Microbe interactions: b) metabolism and virulence P60
Legionella pneumophila requires polyamines for intracellular growth - Potential role of the L. pneumophila chaperonin in modulating polyamine levels in host cells.
Gheyath Nasrallaha, Angela Riverollb, Audrey Chongc, Lois E. Murraya, Rafael A. Gardunod and David S. Alland aDalhousie University, Department of Microbiology and Immunology, Tupper Bldg. 7th floor, 5850 College Street, AB B3H-1X5 Halifax, Canada; bNovartis Animal Health Canada Inc., Victoria site, Aquarium, R&D, 797 Victoria Road, Route 116, PE C0A-2G0 Victoria, Canada; cNIAID, National Institutes of Health - Rocky Mountain Laboratories, 903 South 4th Street, Laboratory of Intracellular Parasites, Hamilton, 59840, United States of America; dDalhousie University, Department of Microbiology and Immunology, Tupper Bldg. 7th floor, 5850 College Street, NS B3H-1X5 Halifax, Canada [email protected]
A portion of the cellular pool of the Legionella pneumophila chaperonin, HtpB, is found on the bacterial cell surface where it can mediate invasion of HeLa cells. Internalized L. pneumophila continues to produce HtpB and releases it into the Legionella-containing vacuole, suggesting that it may play post-invasion roles. To determine whether HtpB reaches the host cell cytoplasm we infected CHO cells with L. pneumophila expressing translocation-reporter genes comprised of the catalytic domain of Bordetella’s adenylate cyclase fused at the N- or C-terminus of HtpB. These fusions induced increases in host cellular cAMP levels, suggesting that HtpB is translocated into the cytoplasm of Legionella-infected cells. To identify potential intracellular functions and eukaryotic cytoplasmic partners, HtpB was ectopically expressed in the genetically tractable eukaryote Saccharomyces cerevisiae. HtpB induced pseudohyphal growth in yeast by activating a signaling cascade controlled by Ras2. A yeast-two-hybrid screen showed that HtpB specifically interacted with S- adenosyl methionine decarboxylase (SAMDC), an essential enzyme required for polyamine biosynthesis and encoded by SPE2. We propose a hypothetical working model in which HtpB stabilizes SAMDC, causing an increase in polyamine levels, which in turn triggers the signaling cascade leading to pseudohyphal growth. Due to the implications of this model for the potential functions of the translocated/cytoplasmic HtpB, we are experimentally testing it. A plasmid harboring SPE2 was transformed into yeast, and the transformants showed pseudohyphal growth, enabling us to conclude that HtpB might induce pseudohyphal growth by altering the biosynthesis of polyamines via SAMDC, as predicted by the model. Two pharmacological inhibitors of SAMDC significantly reduced L. pneumophila intracellular multiplication in L929 cells and macrophages, suggesting that polyamines are important in the biology of L. pneumophila. However, it remains to confirm that host cells treated with the inhibitors have reduced polyamine levels. These results collectively suggest the importance of polyamines in Legionella biology, and implicate HtpB as a factor used by L. pneumophila to modulate the levels of intracellular polyamines in host cells.
- 142 - Session 4: Host - Microbe interactions: b) metabolism and virulence P61
Characterization of L. pneumophila PlaB, a member of a novel phospholipase A family and its potential role in virulence
Jennifer Bendera, Kerstin Rydzewskib, Markus Broichc, Eva Schunderb, Klaus Heunerb and Antje Fliegera aRobert Koch Institut, Burgstrasse 37, 38855 Wernigerode, Germany; bRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; cRobert Koch Institut, Nordufer 20, 13353 Berlin, Germany [email protected]
Legionella pneumophila possesses several phospholipases capable for host-cell manipulation and lung damage. Recently, we discovered the major cell-associated hemolytic phospholipase A PlaB sharing no homology to described phospholipases and being dispensable for intracellular replication in vitro. Nevertheless, we could show that PlaB is the major lipolytic activity in L. pneumophila cell infections and that PlaB utilizes a typical catalytic triad of Ser-Asp- His for effective hydrolysis of phospholipid substrates. Crucial residues were found to be located within the N-terminal half of the protein and amino acids embedding these active sites were unique for PlaB and homologs. Thus, PlaB is the first described member of a new class of lipolytic enzymes. We further displayed that catalytic activity towards phosphatidylcholine, but not phosphatidylglycerol is directly linked to hemolytic potential of PlaB. While the function of the prolonged PlaB C-terminus remains to be elucidated, it is essential for lipolysis as removal of fifteen amino acids already abolishes enzyme activity. As PlaB plays an important role as a virulence factors in a guinea pig infection model, we further analyzed its potential role in Legionella pathogenicity and its distribution among clinical isolates and non- pneumophila species. All tested clinical isolates showed comparable, while out of the non-pneumophila species, only L. gormanii and L. spiritensis possessed similar lipolytic activities as L. pneumophila and comprised plaB-like genes. Interestingly, phosphatidylcholine-specific phospholipase A activity and hemolytic potential was more pronounced in L. pneumophila. Therefore, hydrolysis of the eukaryotic membrane constituent phosphatidylcholine and generation of potential lipid second messengers triggered by PlaB could be an important virulence tool for Legionella pathogenicity.
- 143 - Session 4: Host - Microbe interactions: b) metabolism and virulence P62
Legionella pneumophila outer membrane protein LbtU promotes both siderophore utilization and pathogenesis
Christa Chatfield and Nicholas Cianciotto
Northwestern University, 320 E Superior St, Searle 6-541, Chicago, IL 60611, United States of America c-chatfi[email protected]
Legionella pneumophila (Lpn) growth is dependent on the ability of the bacterium to acquire iron, and we recently demonstrated that siderophore (legiobactin) is required for optimal Lpn growth in the lungs of experimentally-infected mice. The synthesis of legiobactin by Lpn serogroup-1 strain 130b requires the lbtA gene, whereas the export of legiobactin appears to require the adjacent lbtB. In an attempt to identify other factors involved in legiobactin production or uptake, we investigated the function of the ORF (tentatively named lbtU) that is located immediately upstream of lbtAB and whose promoter region, like that of lbtAB, contains a putative Fur-binding site. The lbtU gene is conserved in terms of both sequence and location in the genomes of all four sequenced strains of Lpn. LbtU is predicted to be an outer membrane protein with beta-barrel structure and at least 16 transmembrane domains. Inactivation of lbtU resulted in a 1000-fold loss in the ability of Lpn to grow on iron-limited, solid media. lbtU mutants were also unable to utilize legiobactin-containing supernatants to grow on low-iron media. These data suggest that LbtU is required for legiobactin uptake. Since Lpn lacks TonB, the energy-transducing molecule that typically conjoins with outer membrane receptors to mediate siderophore uptake by gram-negatives, LbtU may represent or be part of a novel mechanism for siderophore and iron acquisition. Multiple, independently-isolated LbtU mutants were impaired at least 100-fold for growth in amoebae and macrophages. In the A/J mouse model of Legionnaires disease, lbtU mutants were severely impaired for survival and replication in the lung, since the CFU recovered were approximately 100 fold reduced from that of the wild-type parent. In contrast to the infection phenotypes of lbtU mutants, legiobactin- deficient lbtA mutant strains were not impaired for in vitro intracellular growth and were only 5-10 fold defective in vivo. These data indicate that LbtU performs a function(s) in addition to that of mediating legiobactin uptake. In sum, the outer membrane protein LbtU performs a vital role for both Lpn iron acquisition and pathogenesis.
- 144 - Session 4: Host - Microbe interactions: b) metabolism and virulence P63
LHBP1, a novel immunogenic gylcosaminoglycan binding protein of Legionella pneumophila, is produced during legionellosis and contributes to alveolar epithelial cells adhesion.
Carla Duncana, Akriti Prasharb, Patrick Tanga, Mauricio Terebiznikb, Donald E Lowa and Cyril Guyarda aOntario Agency for Health Protection and Promotion, 81 A resources road, ON M9P 3T1 Toronto, Canada; bUniveristy of Toronto, 1265 military trail, ON M1C 1AC Toronto, Canada [email protected]
Adhesion of Legionella pneumophila serogroup 1 (Lp1) to host cell surface and extracellular matrix (ECM) is an essential step in the successful colonization of its human host. Human lung epithelial cells represent more than 95% of the surface area of the alveoli and constitute a niche for the replication and dissemination of Lp1 in the lungs. Adherence of Lp1 to human lung epithelial cells is inhibited by heparin when this molecule is used as a competitor of host cell sulfated glycosaminoglycans (GAGs). In the mouse model, co-instillation of heparin during intra-tracheal Lp1 challenge has a protective effect on the alveolar-capillary barrier and prevents bacterial dissemination. Based upon these findings, our central hypothesis is that interactions (adherence/signaling/internalization) of Lp1 with human lung epithelial cells is mediated by one or more GAG-binding adhesins. Using heparin as a GAGs prototype, Legionella Heparin Binding Protein 1 (LHBP1) was identified by affinity chromatography followed by MALDI-TOF analysis. In silico analysis suggested that LHBP1 is a recently acquired collagen-like protein containing a putative signal sequence and multiple tandem repeats. PCR and sequencing analysis of lhbp1 with 21 clinical isolates of L. pneumophila serogroup 1 show this is a polymorphic gene that is conserved among Lp1 circulating strains. Southern Blot analysis using lhbp1 probe detected homologous open reading frames in 7 different clinical serogroups while none of the non-pneumophila environmental isolates tested carried this gene. Protease accessibility experiments using increasing concentration of proteinase K demonstrated that LHBP1 is a surface exposed protein of Lp1. Immunoblot analysis showed that 75% of patient sera from pulmonary legionellosis react with a recombinant LHBP1 (rLHBP1). Affinity chromatography and ELISA assays were used to demonstrate that rLHBP1 retains LHBP1 heparin binding properties. Deletion of collagen like tandem repeats from LHBP1 greatly decreased immunogenicity and partially decreased binding to heparin. In addition to heparin sulphate, rLHBP1 binds to sulphated sugars such as chondroitin 6 sulfate, dextran sulphate and mannose suggesting that LHBP1 is a Legionella ECM binding proteins. In ELISA assays, an in-frame deletion ∆lhbp1 mutant showed a significant reduction in binding to biotinylated heparin compared to wild type. Adhesion assays with human lung epithelial cell line NCI-H292 show that the ∆lhbp1 mutant is impaired in binding to NCI-H292 cells when using 0.1% Triton X-100 to lyse epithelial cells. We have shown that LHBP1 is a surface exposed and immunogenic collagen-like protein, produced in vitro and during human infection. We have demonstrated that LHBP1 is a GAG binding surface protein encoded by a polymorphic gene unique to L. pneumophila species. Variable multiple tandem repeats of LHBP1 are involved in its heparin binding properties and immunogenicity. An lhbp1 isogenic mutant is impaired in binding to heparin and human lung epithelial cells. Altogether our data suggest that LHBP1 is a new GAG binding adhesin involved in the pathogenesis of Lp1.
- 145 - Session 4: Host - Microbe interactions: b) metabolism and virulence P64
The bdhA-patD Operon as a virulence determinant, revealed by a novel large-scale aproach for virulence-attenuated Legionella pneumophila mutants
Philipp Aurassa, Birgit Plessb, Kerstin Rydzewskic, Gudrun Hollandb, Norbert Bannertb and Antje Fliegera aRobert Koch Institut, Burgstrasse 37, 38855 Wernigerode, Germany; bRobert Koch Institut, Nordufer 20, 13353 Berlin, Germany; cRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany fl[email protected]
In our search for novel Legionella pneumophila virulence factors, we developed an agar plate assay, designated the scatter screen, which allows screening for mutants deficient in infecting Acanthamoeba castellanii amoebae. Likewise, an L. pneumophila clone bank consisting of 23,000 transposon mutants was investigated here, and 19 different established Legionella virulence genes, for example, dot/icm genes, were identified. Importantly, 70 novel virulence-associated genes were found. One of those is L. pneumophila bdhA (lpg2316), coding for a protein with homology to established 3- hydroxybutyrate dehydrogenases involved in poly-3- hydroxybutyrate metabolism, such as Sinorhizobium meliloti BdhA. Here we show that L. pneumophila BdhA indeed possesses in vitro 3-hydroxybutyrate dehydrogenase activity. Our study further revealed that bdhA is cotranscribed with patD, encoding a patatin-like protein of L. pneumophila showing phospholipase A and lysophospholipase A activities. In addition to reduced lipolytic activities and increased poly-3- hydroxybutyrate levels, the L. pneumophila bdhA-patD mutant inefficiently replicated in amoebae and U937 macrophages. Our data show that the operon is involved in poly-3-hydroxybutyrate utilization and phospholipolysis and reveal the bdhA- patD operon as a novel virulence determinant of L. pneumophila. In summary, the screen for amoeba-sensitive Legionella clones efficiently isolated mutants that do not grow in amoebae and, in the case of the bdhA-patD mutant, also human cells.
- 146 - Session 4: Host - Microbe interactions: b) metabolism and virulence P65
Investigations into the metabolism of Legionella pneumophila
Vroni Herrmanna, Eva Eylertb, Wolfgang Eisenreichb, Monika Lautnera and Klaus Heunera aRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; bTU Munich, Department of Chemistry, Lichtenbergstr. 4, D-85747 Garching, Germany [email protected]
BACKGROUND: Many virulence factors of L. pneumophila have been reported, but less is known about nutrition of the bacteria, especially inside host cells. When nutrients become limiting, a regulatory casdade triggers the differentiation from the replicative form, with high metabolic activity, to the transmissive form. L. pneumophila uses amino acids as primary energy and carbon sources, and although glucose is assmilated, it is thought not to be important for bacterial growth. The aim of this study is to gain insight into the metabolic role of glucose in L. pneumophila, in vitro and in host cells. METHODS: We used 13C-isotopologue profiling to study the carbon metabolism of L. pneumophila. The bacteria were 13 13 supplied with [U- C6]glucose and [1,2- C2]glucose during in vitro growth, respectively. The metabolic products were analyzed via magnetic resonance (NMR) spectroscopy and mass spectrometry (MS). Similar experiments were performed with the natural host organism Acanthamoeba castellanii. To determine the expression of genes for key enzymes of glucose metabolism, Reverse Transcription-PCR analysis was done in different growth phases of L. pneumophila. In addition, operon structures of these genes were defined also by RT-PCR. RESULTS: We found high 13C-incorporation rates for the amino acids alanine, aspartate, glutamate, glycine, and proline as well as for 3-hydroxybutyrate in L.pneumophila. Genes involved in glucose catabolism were mainly transcribed during the exponential growth phase and were organized in several cistrons. A. castellanii synthesised a number of amino acids from glucose, as well as glycerol and fatty acids. CONCLUSION: Here, we demonstrate that glucose is used for de novo synthesis of several amino acids in L. 13 pneumophila. The tricarboxylic acid cycle is complete and active. Experiments with [1,2- C2]glucose indicate that the pentose phosphate pathway is important for glucose catabolism. Moreover, we present an experimental background for in vivo-studies of the intracellular metabolism inside A. castellanii host cells.
- 147 - Session 4: Host - Microbe interactions: b) metabolism and virulence P66
Genetic analysis and function of redundant ompS genes in Legionella pneumophila
Paul S. Hoffmana, Serhiy Vitkob, Poornima Gourabathinib and Fanny Ewannb aUniversity of Virginia Dept. Infectious Diseases, 409 Lane Rd Rm 2143, Charlottesville, 22908, United States of America; bUniversity of Virginia, MR4 Bldg Room 2146, 409 Lane Road, Charlottesville, AK 22908, United States of America [email protected]
Background: The major outer membrane protein oligomer (OmpS) of the Philadelphia strain of Legionella pneumophila(Lpp) is composed of three subunits cross-linked by inter-chain disulfide bonds; with one subunit covalently bound to peptidoglycan. Genomic analysis reveals two additional ompS homologues (ompS2 and ompS3) located in an operon separate from the essential ompS1 gene. Other sequenced strains contain up to 5 ompS homologues which includes gene duplications. The three Lpp OmpS proteins are highly conserved in the N-terminal region which includes 2 conserved cysteines and divergent in the C-terminal region, where only OmpS1 contains an additional 2 cysteines. This study was initiated to asses the contribution of each ompS gene to the composition, disulfide cross-linking arrangements and function of the OmpS oligomer. Methods: qRT PCR and GFP reporter fusions were used to monitor expression of the ompS genes. Allelic exchange mutagenesis was used to create ompS2-3 operon deletion mutants in Lpp strain Lp02 and in AA100. Mutant strains were analyzed for infectivity in amoeba and cell line models and oligomer composition was analyzed by SDS-PAGE and alkylation of cysteine residues with methoxy-polyethylene glycol-maleimide (Mal-PEG) and immunoblot analysis. Results: Analysis of gene expression (qRT PCR/GFP reporter) showed that all three ompS genes were expressed equally across growth cycle in vitro, suggesting the OmpS oligomer is a hetero-trimer potentially composed of three different subunits. The Lp02∆ompS2-3 mutant displayed a slow growth phenotype; whereas, the LpAA100∆ompS2-3 mutant was essentially wild type for growth. The ompS1 mutation could only be obtained in LpAA100. All ompS mutants were avirulent for amoeba, macrophages and cell lines. SDS-PAGE analysis of outer membrane material revealed no differences in OmpS protein composition between ompS mutants and WT strains. Mal-PEG labeling confirmed the monomeric subunit composition of the OmpS oligomer in each of the mutants based on change in monomer mass due to alkylation of cysteines. Conclusions: Our findings indicate that the Lpp ompS genes are equally expressed and contribute a subunit to the oligomer. In contrast to ompS1 (Lpp only), the ompS2-3 operon is nonessential for growth. OmpS synthesis and oligomeric assembly appear to be tightly coordinated with cell division. While the porins of other bacteria consist of three subunits produced from a single gene, the OmpS oligomer of L. pneumophila is unique in its heteromeric composition. Based on the profound effect that ompS mutations have on infectivity, we suggest that the OmpS oligomer might provide a novel function, perhaps necessary for assembly or function of the type IV Dot/Icm secretion system.
- 148 - Session 4: Host - Microbe interactions: b) metabolism and virulence P67
Legionella pneumophila flagellum in the fateful journey from amoeba to macrophages.
Juliana I. Hori, Giuliano F. Morgantetti, Catarina V. Horta, Marcelo De S. F. Pereira, Liliana M. Massis and Dario S. Zamboni
Department of Cell Biology and Microbial Pathogenesis. University of Sao˜ Paulo, Medical School Ribeirao˜ Preto, FMRP/USP, Bandeirantes Avenue, 3900, 14049-900 Ribeirao Preto, Brazil [email protected]
Legionella pneumophila, the etiologic agent of Legionnaires’ disease, has evolved sophisticated mechanisms to subvert the host response allowing its replication and survival in amoebae. It has been speculated that the L. pneumophila interaction with aquatic protozoa generated a pool of virulence traits during the evolution process that enabled Legionella to infect human macrophages. The bacteria possess a monopolar flagellum, which is important to motility in the aquatic habitats. However, the flagellin (FlaA), the major subunit of flagellum, is detected by pattern recognition receptors present in innate immune cells. It was demonstrated that Legionella is recognized by mammalian macrophages through Nod-like receptor proteins (NLRs) such as Naip5 and Nlrc4 (Ipaf), which trigger caspase-1 activation with consequent restriction of bacterial multiplication. Moreover, NLRs activation requires the presence of a competent type IV secretion system and flagellin. Here, we evaluated the natural history of the L. pneumophila flagellum during infection of amoeba, macrophages and mice lungs. We found that flaA mutants replicate in macrophages from C57BL/6 mice, while wild type and fliI mutant, which is non-motile, had a growth restriction. Experiments performed in vivo demonstrated that flagellin, but not motility was also required for Legionella restriction in a murine model of Legionaire’s disease. In addition, experiments performed with Nlrc4−/− mice demonstrated the profound requirement of Nlrc4 for flagellin-dependent restriction of L. pneumophila infection in vitro or in vivo. To further evaluate the requirement of Legionella flagellin, we performed experiments in axenic media and in amoeba. We found that the absence of flagellin did not interfere with bacterial growth neither in axenic media nor amoeba. However, in a competition assay in amoeba, wild type bacteria multiplied better than flaA and fliI mutants. Together, these results indicate that flagellin is required for effective amoeba colonization whereas the innate immune system has developed intracellular pattern recognition receptors that recognize flagellin and trigger host resistance.
- 149 - Session 4: Host - Microbe interactions: b) metabolism and virulence P68
The essential, membrane-bound oxidoreductase Com1 is required for growth and infectivity of Legionella pneumophila.
Max Jameson-Lee and Paul S. Hoffman
University of Virginia Dept. Infectious Diseases, 409 Lane Rd Rm 2143, Charlottesville, 22908, United States of America [email protected]
Background: Legionella pneumophila displays a developmental cycle in which vegetative replicating forms differentiate late in infection into resilient infectious cysts. Comparative proteomic analysis identified a 27 kDa protein which was enriched in cysts that is an important virulence factor in related intracellular parasites where it is known as Coxiella outer membrane protein 1 or Com1. In Legionella, subunits of the major outer membrane oligomer OmpS are cross- linked by inter-chain disulfide bonds that may provide structural integrity to the developing cyst forms. Com1 contains a thioredoxin fold similar to the archetypal disulfide bond oxidase DsbA, which might catalyze disulfide bond formation. Here we investigate the role of Com1 in catalyzing the assembly of disulfide bonds associated with periplasmic and outer membrane proteins. Methods: Targeted mutations were made to assess the role of the oxidoreductase system in L. pneumophila growth and infectivity of amoeba. Com1 mutants designed to trap interacting proteins were expressed in L. pneumophila and purified under conditions which preserve covalent disulfide bridges. Com1 was assessed for oxidoreductase activity and antibody was produced and used for cellular localization studies. Results: Com1 appears to be essential - resistance cassettes could be engineered downstream of com1, but not within the coding region. In contrast, a knockout mutation constructed in dsbA displayed no phenotype and was essentially wild type in amoeba, macrophage and cell line infection models. Cell fractionation and western blot analysis localized Com1 to the outer membrane and periplasmic fractions and biotinylation experiments on whole cells suggested that Com1 may span the outer membrane. In order to trap interacting proteins, a C112S mutation in the CxxC site and a second cis- proline P151T mutation were expressed from an inducible plasmid system in Lp02 strains. Interacting proteins were visualized under non-reducing conditions by SDS-PAGE and western blotting: these bands disappeared upon treatment with reducing agents. Conclusions: Com1 is an essential oxidoreductase found in the outer membrane. Many membrane-spanning and surface- expressed structural proteins in L. pneumophila cysts contain disulfide bonds which ostensibly require oxidoreductases to properly fold and assemble them. We suggest that L. pneumophila utilizes the essential, membrane-bound oxidoreductase Com1 to promote disulfide bond cross-linkage in the assembly of outer membrane proteins and other potential virulence determinants.
- 150 - Session 4: Host - Microbe interactions: b) metabolism and virulence P69
Metabolic reconstruction of the human pathogen Legionella pneumophila
Matthieu Jules, Christophe Rusniok, Sandra Duperrier, Holger Bruggemann¨ and Carmen Buchrieser
Institut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France [email protected]
L. pneumophila is a pathogenic gram-negative bacterium that naturally replicates within aquatic protozoa like Acanthamoeba castellanii. Five years ago, the genome sequences of three different isolates became available: the strains Paris, Lens [1] and Philadelphia [2]. Genomes greatly improved our knowledge of the host-pathogen interaction. It now allows more formal approaches dedicated to the understanding of the pathogen metabolism in terms of evolution. In contrast to its in vivo fast growing features (e.g. Tg∼1h), L. pneumophila is fastidious in vitro (Tg>2h). In order to decipher its in vivo needs, we performed a genome-scale metabolic reconstruction, from genome sequence annotation, but also from transcriptomic [3,4], biochemical and physiological data [5]. We built a MetaCyc based database for L. pneumophila, namely LppCyc, accounting for more than one fourth of the protein encoding genes. This database was first used to carry out a metabolic comparative analysis between intracellular, extracellular and intravacuolar pathogens. Interestingly, it revealed that L. pneumophila is ”metabolically” related to phytopathogens [6]. Using elementary flux modes (EFM) analyses, we explored the metabolic capabilities of this bacterium through an in silico model containing >200 enymatic and transport reactions. General network properties, such as robustness, were analysed and in silico deletion studies predicted essential genes in L. pneumophila. Similarly the model predicted only 7 essential amino acids for L. pneumophila indicating that L-arginine was not. This was confirmed experimentally [6]. The probable lack of the transaldolase reaction also showed up as a major, singular feature. As for Salmonella enterica, transcriptomic data [3,4] showed an overexpression of the Entner-Doudorrof pathway encoding genes during the exponential growth phase of infection. The model can neither explain that by a lack of NADPH nor a strict requirement of gluconeogenic fluxes for precursor biosynthesis but provided some putative explanations. [1] Cazalet, C. et al. (2004). Nat.Genet. 36, 1165-1173. [2] Chien, M. et al. (2004). Science 305, 1966-1968. [3] Bruggemann, H. et al. (2006). Cell Microbiol. 8, 1228-1240. [4] Jules, M. and C. Buchrieser (2007). FEBS Letters 581, 2829-2838. [5] Sahr, T. et al. (2009). Mol Microbiol. 72, 741-762. [6] Jules, M. et al. (2009). In preparation.
- 151 - Session 4: Host - Microbe interactions: b) metabolism and virulence P70
Virulence properties of Legionella pneumophila GDSL lipolytic enzymes
Christina Langa, Elena Rastewb, Sangeeta Banerjib, Sina Bartfeldc, Thomas F. Meyerc and Antje Fliegera aRobert Koch Institut, Burgstrasse 37, 38855 Wernigerode, Germany; bRobert Koch Institut, Nordufer 20, 13353 Berlin, Germany; cMax Planck Institute for Infection Biology, Department of Molecular Biology, Chariteplatz´ 1, 10117 Berlin, Germany [email protected]
Legionella pneumophila expresses a multitude of lipolytic enzymes falling into three groups that may be involved in pathogenesis. One of these groups, the GDSL hydrolases, comprises enzymes of prokaryotic and eukaryotic origin with phospholipase, acyltransferase, and hemolytic activity. Enzymatic activity depends on a conserved nucleophilic serine embedded into the GDSL motif as well as aspartate and histidin together building up the catalytic triad. The Legionella pneumophila genome codes for three GDSL-hydrolase genes: plaA, plaC and plaD. The three enzymes show lysophospholipase A (LPLA) and phospholipase A activity with PlaA being the major secreted LPLA. PlaC additionally displays acyltransferase activity. This activity is post-transcriptionally regulated by ProA, a secreted zinc metalloprotease of Legionella pneumophila. The sequences of PlaA and PlaC harbour N-terminal signal peptides for Lsp type II-dependent protein secretion, whereas the secretion mode of PlaD is still unclear. Since phospholipases are important virulence factors that have been shown to promote bacterial survival, spread and host cell modification or damage, we here aimed to investigate the contribution of GDSL enzymes to Legionella pneumophila virulence. Our studies of amoeba and macrophage infection showed that single and double gdsl-knock outs do not affect intracellular infection of Legionella pneumophila, pointing to overlapping functions of the three enzymes. Interestingly, the interleukin 8-release by lung epithelial cells infected with gdsl double mutants (plaA−/plaC− and plaA−/plaD−) and a plaA single mutant was reduced up to 60% in comparison to infection with the Legionella pneumophila wild type. Secretion of IL8 can be induced via the NfκB-signaling pathway. Accordingly, we found less p65 translocation from cytoplasm to nucleus with the gdsl mutants compared to the wild type. These results indicate that GDSL-enzymes interfere with signalling cascades during infection and contribute to the pathogen-directed manipulation of the host cell.
- 152 - Session 4: Host - Microbe interactions: b) metabolism and virulence P71
Expression of Legionella pneumophila efflux pump genes during different physiolog- ical states
Mourad Ferhat, Daniele` Atlan, Jean-Claude Lazzaroni and Christophe Gilbert
UMR 5240 Microbiology, Adaptation and Pathogeny CNRS-Lyon1-INSA-BayerCropscience, BAT A. Lwoff, 10 rue Dubois. Campus de la Doua, 69622 Villeurbanne, France [email protected]
In the environment, Legionella pneumophila is found in water and more precisely in amoeba that contributes to its replication and spread. Contamination of humans occurs after inhalation of contaminated aerosols and the bacteria is able to grow within macrophages. The intracellular life cycle include the transition from a replicative to a transmissive form characterized by an increased resistance to antibiotics. This increased resistance suggests a potential link between the efflux of toxic compounds by the bacteria and virulence. Indeed, one universal mechanism underlying drug resistance to various toxic compounds, namely multi-drug resistance (MDR), is the expression of efflux pumps that drive drugs outside the target cell. Five families of efflux pumps have been described on the basis of the inner membrane protein structure: the major facilitator (MF) superfamily, the ATP-binding cassette (ABC) family, the resistance-nodulation-division (RND) family, the small multi-drug resistance (SMR) family and the multi-drug and toxic compound extrusion (MATE) family Our previous work reported that TolC acts as a major efflux pump component in L. pneumophila Lens. We also assessed for the virulence of the tolC mutant in two eukaryotic hosts : the mutant presented severe intracellular growth defects and attenuated cytotoxicity in both Acanthamoeba castellanii (protozoa) and U937 cells (human monocyte cell line). The purpose of this study is to evaluate the expression levels of many potential efflux pumps genes in L. pneumophila Lens during growth in various conditions: liquid broth (exponential and stationary phase) and inside amoeba (replicative and transmissive phase). From bioinformatics analysis of Legionella whole genome, 35 genes were chosen for their potential involvement in efflux mechanisms. The expression of these target genes was followed by RT qPCR. Our results showed significant different expression levels of few target genes in relation to growing conditions. All the results will allow us to focus future work on particular efflux mechanisms in relation to their possible role in virulence.
- 153 - Session 4: Host - Microbe interactions: b) metabolism and virulence P72
Interaction between COPI and the Legionella pneumophila protein LpnE.
Hayley Newton, Ralf Schuelein, Margareta Aili, Joan Sloan and Elizabeth Hartland
University of Melbourne, Department of Microbiology and Immunology, 3010 Victoria, Australia [email protected]
Previously we have demonstrated that the SLR containing protein LpnE is an important virulence factor of Legionella pneumophila . LpnE is involved in the initial establishment of the unique L. pneumophila replicative vacuole. Reporter assays demonstrate that LpnE is not translocated into the host cell via the Dot/Icm secretion system however it is secreted by L. pneumophila and is prevalent in outer membrane vesicles that can manipulate host cell trafficking. A yeast two- hybrid screen identified εCOP, a component of the coat protein complex COPI, as a putative host binding partner of LpnE. This direct interaction has been confirmed by recombinant protein overlays and the ability of GST-εCOP to bind native LpnE from L. pneumophila lysate. The binding of LpnE to εCOP is dependent on the N-terminus of LpnE with LpnE52- 375 unable to interact with εCOP. Furthermore LpnE-GFP, but not GFP-LpnE, co-localizes with COPI components when overexpressed in eukaryotic cell lines. The first 123 amino acids of LpnE are sufficient for this localization. Co- localization of LpnE-GFP and COPI is disrupted by both the ARF inhibitor Brefeldin A and the COPI specific inhibitor 1,3-Cyclohexanebis(methylamine). Infection of LdlF cells, a derivative of CHO-K1 cells with a temperature sensitive εCOP mutation, demonstrates that efficient infection is dependent on εCOP although this does not appear to be linked to the ability of the L. pneumophila containing vacuole to avoid lysosomal fusion.
- 154 - Session 4: Host - Microbe interactions: b) metabolism and virulence P73
Structural investigation into the Legionella CD39 NTPDase Lpg1905
Patrice Riedmaiera, Fiona M Sansoma, Travis Beddoeb and Elizabeth Hartlanda aUniversity of Melbourne, Department of Microbiology and Immunology, 3010 Victoria, Australia; bMonash University, ARC Centre for Excellence in Structural and Functional Microbial Genomics, 3800 Clayton, Australia [email protected]
Legionella pneumophila is an opportunistic pathogen that establishes a replicative vacuole within alveolar macrophages upon inhalation of contaminated aerosols, resulting in the onset of severe atypical pneumonia. Lpg1905, a CD39/NTPDase-1 homologue, belongs to the NTPDase family of enzymes. These enzymes hydrolyse terminal phosphoanhydride bonds of extracellular nucleoside di- and tri-phosphates, such as ATP and ADP. Lpg1905 is secreted by L. pneumophila and is required for intracellular replication in macrophages and amoebae and essential for virulence in a mouse lung infection model. This mutagenesis study investigated the importance of several residues essential for catalysis and intracellular replication of L. pneumophila in macrophages. Residues were chosen based on their proximity to the substrate after analysis of the crystal structure of Lpg1905 in complex with an ATP analogue. Non-conservative mutations of Arginine 56 abolished enzymatic function of Lpg1905, whilst a conservative mutation to Lysine (R56K) reduced ATPase activity by ∼60% and converts Lpg1905 to an ATPase with complete abolishment of ADPase activity. This residue is implicated in binding the substrate during hydrolysis via interaction with the γ- and β-phosphates of ATP and ADP during hydrolysis. Asparagine 302 does not directly interact with the substrate and both non-conservative and conservative mutations had no effect on function. The non-conservative mutation of Tyrosine 346 to Alanine (Y346A) reduced ATPase activity by ∼80% compared to wild type. This residue is potentially involved in positioning the substrate for hydrolysis via π-π stacking of the aromatic rings between the substrate adenine and the Tyrosine residue. Analysis of the effect of R56K and Y346A mutation in Lpg1905 was investigated in THP-1 macrophages. The Lpg1905 Y346A mutant was unable to survive as well as wild type Lpg1905 and Lpg1905 R56K mutant in THP-1 macrophages, suggesting survival and intracellular replication of L. pneumophila is dependent on a minimal threshold of ATPase activity rather than ADPase activity.
- 155 - Session 4: Host - Microbe interactions: b) metabolism and virulence P74
Characterization of the major phospholipase A/ lysophospholipase A activity of Legionella pneumophila
Eva Schundera, Patrick Adamb, Futoshi Higac, Katharina Remerd, Udo Lorenze, Michael Steinertf, Jennifer Benderg, Antje Fliegerg and Klaus Heunera aRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; bInstitut fur¨ Pathologie, Universitatsklinikum¨ Tubingen,¨ Liebermeisterstraße 8, 72076 Tubingen,¨ Germany; cDepartment of Internal Medicine, 207 Uehara, Nishihara town, 903-0215 Okinawa, Japan; dInstitute for Molecular Infection Biology, Roentgenring 11, 97070 Wuerzburg, Germany; eZentrum Operative Medizin, Universitat¨ Wurzburg,¨ Oberdurrbacher¨ Str. 6, 97080 Wurzburg,¨ Germany; fInstitut fur¨ Mikrobiologie, Technische Universitat¨ Braunschweig, Spielmannstr.7, 38106 Braunschweig, Germany; gRobert Koch Institut, Burgstrasse 37, 38855 Wernigerode, Germany [email protected]
BACKGROUND: The etiopathology of guinea pigs infected with L. pneumophila is comparable to the clinical and histological features of human Legionnaires‘ disease. The bacteria reach the lung, replicate in alveolar macrophages, spread and cause pneumonia. Phospholipases are well known virulence factors. They hydrolyse cell membranes, modulate the host immune response and are able to destroy lung surfactant. The PlaB protein is a recently identified cell-associated phospholipase A/ lysophospholipase A of Legionella pneumophila. METHODS: We characterized the PlaB protein via human blood hemolysis assays, phospholipase A activity assays, Reverse Transcription PCR, RACE transcription start determination and sucrosegradient ultracentrifugation. To investigate the role of PlaB in vivo, the guinea pig infection model was used. RESULTS: In vitro, we could detect a growth phase-dependent expression and enzymatic activity of PlaB. Active PlaB-Protein was located on the surface exposed side of the outer membrane. In guinea pigs, plaB- deficient bacteria were impaired for replication and dissemination. CONCLUSION: The experiments revealed that PlaB is the major-cell associated phospholipase A/ lysophospholipase A of L. pneumophila with an additional hemolytic activity which plays a role in guinea pig infection. The mechanism of action and the transport to the outer membrane needs further investigation.
- 156 - Session 4: Host - Microbe interactions: b) metabolism and virulence P75
Increased expression of the plasminogen activator Lpa of Legionella pneumophila in low-magnesium medium
Leen Vranckx, Veerle Saels and Jozef Anne´
Rega Institute for Medical Research, Minderbroedersstraat 10, 3000 Leuven, Belgium [email protected]
Outer membrane proteins (Omps) are temporally and spatially the first proteins to interact with the environment or with the host cell. Therefore Omps are considered critical determinants for virulence. In a search for Legionella pneumophila genes encoding putative virulence-related Omps, we identified a gene encoding the plasminogen activator Lpa. This protein belongs to the class of omptins, a family of surface proteases/adhesins that share high sequence identity and a conserved beta-barrel fold in the outer membrane. The omptins are multifunctional, and individual omptins exhibit different virulence-associated functions. We previously showed that Lpa activates the zymogen plasminogen into plasmin (Vranckx et al., 2007), a potent serine protease that can degrade fibrin and extracellular matrices and can activate collagenases. Lpa can also degrade alpha-2- antiplasmin, the main inhibitor of plasmin. In this way the formed plasmin is protected against the action of the inhibitor. These features could be important for the spread of Legionella across the lung tissue. Expression of Lpa could hardly be detected when wild-type Legionella was grown in standard rich BYE medium. Therefore we looked for conditions in which Lpa expression was increased. In Salmonella enterica the expression of many virulence-associated genes is controlled by the PhoP/PhoQ two-component system, a signal transduction system that responds to low environmental Mg2+ concentration. For example the activation of PgtE of Salmonella, an Lpa homologue, is Mg2+-controlled. Also in L. pneumophila Mg2+ limitation seems to be a regulatory factor: it has been shown that the resistance of L. pneumophila against cationic antimicrobial peptides is induced in low-magnesium medium (Robey et al., 2001). We analyzed the expression of Lpa in L. pneumophila grown in a chemically defined medium containing three different Mg2+ concentrations and quantified Lpa expression by Western blotting with Lpa-specific antibodies and the amount of fluorescence produced by the green fluorescent protein under the control of the Lpa promoter. Using these methods a sharp increase in the expression of Lpa with decreasing Mg2+ concentration was observed, even though at low Mg2+ levels the overall growth rate was hampered. These results demonstrate that Mg2+ is a regulating factor in the expression of Lpa. At current the relevance of the increased Lpa-expression in low-magnesium conditions is further investigated. Vranckx, L., De Buck, E., Anne,´ J. & Lammertyn, E. (2007). Legionella pneumophila exhibits plasminogen activator activity. Microbiology, 153: 3757-3765. Robey, M., O’Connell, W. & Cianciotto, N.P. (2001). Identification of Legionella pneumophila rcp, a pagP-like gene that confers resistance to cationic antimicrobial peptides and promotes intracellular infection. Infection and Immunity, 69: 4276-4286.
- 157 -
SESSION 5 IMMUNOLOGY AND HOST SUSCEPTIBILITY
- 158 - Session 5: Immunology and host susceptibility P76
How can an excutioner caspase control Legionella pneumophila infection
Anwari Akhtera, Thirumala Kannegantib, Mark Wewersc and Amal Amerd aThe Ohio State University, 460 W 12th avenue, BRT 1014, Columbus, OH 43210, United States of America; bSt Jude’s Children’s Hospital, 570 St Jude Place, suite E7004, Memphis, AK 38105, United States of America; cThe Ohio State University, 473 W 12th avenue, DHLRI room 201 Ohio, Columbus, AK 43210, United States of America; dThe Ohio State University, 460 W 12th avenue, BRT 1014, Columbus, AK 43210, United States of America [email protected]
Legionella pneumophila (L. pneumophila), the causative agent of a severe form of pneumonia called Legionnaires’ disease, replicates in human monocytes and macrophages. Particularly, caspase-1 activation is detected during L. pneumophila infection of restrictive wild-type murine macrophages while absent in permissive human cells. Therefore, most inbred mouse strains are restrictive to L. pneumophila infection except for those that do not activate caspase-1 in response to L. pneumophila such as the A/J, Nlrc4-/-, Ipaf-/-, and caspase-1-/- derived macrophages. During apoptosis, initiator caspases cleave and activate the executioner caspases-3 and 7, which in turn are responsible for the majority of proteolytic events that ultimately result in the demise of the cell. We found that the activation of caspase-7 can occur independently of the typical apoptosis pathway. In murine macrophages caspase-7 was activated by caspase- 1 in response to L. pneumophila infection and bacterial flagellin and required a functional Naip5. Remarkably, the inflammatory cytokines IL-1B and IL-18, also activated downstream of caspase-1, did not play a major role in restriction of L. pneumophila infection. Moreover, mice lacking caspase-7 and their macrophages allowed substantial L. pneumophila replication. Permissiveness of caspase-7-/- macrophages to the intracellular pathogen was due to defective delivery of the organism to the lysosome and to delayed pyropototic cell death during early stages of infection. Interestingly, caspase- 7 activation was not detected in human monocytes when infected with L. pneumophila. The activation of caspase-7 through the caspase-1 inflammasome complex in response to infection by Legionella pneumophila links an executioner caspase into the inflammatory pathway. These results reveal a new mechanism for caspase-7 activation downstream of the caspase-1 inflammasome and present a novel biological role for caspase-7 in host defense against an intracellular bacterium. Moreover, these data offer a new animal model for the understanding of L. pneumophila pathogenesis during human infection.
- 159 - Session 5: Immunology and host susceptibility P77
The early immune response in Legionella longbeachae serogroup 1 mice infection
Ivana Gobina, Zlatko Trobonjacab, Darinka Vuckovica, Gabrijela Begica, Tiana Grubesica, Miljenko Dorica and Milorad Susac aDepartment of Microbiology, School of Medicine, University of Rijeka, Brace Branchetta 20, 51000 Rijeka, Croatia; bDepartment of Physiology and Immunology, School of Medicine, University of Rijeka, Brace Branchetta 20, 51000 Rijeka, Croatia; cDiagnostikzentrum Ulm, Einsteinstraße 59, 89077 Ulm, Germany [email protected]
Background: Legionella longbeachae serogroup 1 is an important human pathogen responsible for almost half of all cases of legionellosis in Australia. The primary transmission mode of this microorganism is inhalation of dust from contaminated compost or soil. It was shown recently that, in comparison to other Legionella spp., L. longbeachae has 5 an unusual high lethality potential with LD90 of only 10 cfu and causes, in contrast to classic Legionnaires’ disease, a unique focal bronchopneumonia, with a massive recruitment of inflammatory cells in the lungs. In the present study we characterised the cell infiltrate in the lungs, as well as the cytokine profiles in the sera and lungs of infected mice. Methods: C57Bl/6 mice were intratracheally inoculated with 105 cfu of L. longbeachae serogroup 1. For immunohistological procedures the lungs were fixed in formalin, dehydrated and embedded in paraffin. Sections were stained either with HE or labelled with different monoclonal antibodies (anti-CD11b, anti GR-1, anti-CD4, anti-CD8, anti-B220). The lungs were digested by collagenase/DNase and leukocytes were separated using a percoll gradient. The cells were phenotypied using monoclonal antibodies specific for the following leukocyte surface antigens: anti-CD11b, anti-GR- 1. Three colour immunofluorescence analysis was performed by BD FACS Caliburoˆ flow cytometer using CellQuestoˆ software. 72 hours after infection levels of IFN-gamma, TNF-alpha and IL-12 in the sera and lungs were determined by enzyme immunoassays. Results: Immunohistological staining showed an infiltration of CD11b+ and GR1+ cells in the peribronchial and perivascular areas of the lungs at an early phase of infection followed by infiltration of CD8+, CD4+ and B220+ cells. CD11b+ cells were predominant in the cell infiltrate and three cell subpopulations could be distinguished: CD11b+dim /GR-1− CD11b+bright /GR-1− and CD11b+bright/GR-1+ cells. The cytokine profiles in the lungs and sera of infected mice demonstrated a strong Th1 response, characterized by an increase of IFN-gamma, IL-12 and TNF-alfa. Conclusion: CD11b+/GR-1+ neutrophils are predominant cells in the early immune response to L. longbeachae infection accompanied by a strong Th1 cytokine profile.
- 160 - Session 5: Immunology and host susceptibility P78
Polyreactivity of a mAb against human erytrocyte glycophorin A with the major outer membrane protein from Legionella pneumophilia
Jan Kolberg, Ø Ihle, Arne Høiby and Audun Aase
Norwegian Institute of Public Health, P.O.Box 4404 Nydalen, NO-0403 Oslo, Norway [email protected]
Polyreactive antibodies can bind to a variety of structurally unrelated self and non-self antigens. We have previously described a mouse mAb 124,D-7 (IgM) obtained after in vitro immunization with haemoglobin-free ghosts of human erythrocyte as antigen (Kolberg & Blanchard, 1991). The antibody agglutinated directly all human red cells irrespective of blood type, except the rare erythrocytes which lack the major sialoglycopprotein, glycophorin A. Inhibition of agglutination with purified glycophorin A and peptides suggested that the epitope was located within residues 35-40. Recent use of this antibody in assays with bacterial antigens revealed unexpected findings. The mAb showed strong reactions by ELISA and immunoblotting with some bacterial species including Legionella and moderate to weak reactions with other species. For L. pneumophilia, we detected by immunoblotting one main reacting band whereas for the other examined species there were several bands. Total proteins from L. pneumophilia were separated by isoelectric focusing followed by SDS-PAGE. The immunoreactive band was identified as the major outer surface protein by mass spectrometry (MALDI-MS) analysis. Flow cytometric analyses with live and heat-killed L. pneumophilia was performed to see whether the epitope for mAb 124,D-7 was exposed on the bacterial surface. The bacteria did not bind mAb 124,D-7. In contrast, the control mAb 3/1 (Dresden panel) directed against LPS showed high reactivities against both live and dead bacteria whereas Monofluo (BIO-RAD) recognizing the major outer membrane protein only bound to heat-killed bacteria. Sequencing of the genes encoding the heavy chain of the mAb revealed germ-like DNA sequences which are characteristic for polyreactive antibodies. These antibodies constitute a first line of defence against bacterial pathogens. It is therefore possible that Legionella infections in humans could cause the activation of B-cells producing polyreactive antibodies, and one might speculate whether such antibodies also could be autoreactive with erythrocyte sialoglycoproteins. Kolberg J & Blanchard D (1991). Immunol Lett 30: 87
- 161 - Session 5: Immunology and host susceptibility P79
Phase variable changes and intracellular delivery of LPS components of Legionella pneumophila serogroup 1 strain Corby and its MAb 3/1-negative mutant
Katja Reichardta and Jurgen H. Helbigb aTU Dresden, Institute Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany; bInstitute of Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany [email protected]
L. pneumophila is characterized by a high degree of its LPS diversity. Modulating effects of LPS components concerning bacterial transmission and host cells alterations are under consideration because strains of serogroup 1 carrying the so- called virulence-associated LPS epitope, recognized by the monoclonal antibody (MAb) 3/1 cause most of the community- acquired legionellosis. Using double immunofluorescent staining serogroup 1 strain Corby and its MAb 3/1-negative mutant TF 3/1 possessing a point mutation in the lag-1 gene were tested for their intracellular LPS pattern. Switching and shedding of LPS components was investigated in A. castellanii and U937 cells. MAb 3/1 recognizing long and short chain LPS components and a phase-variable short chain LPS component recognized by MAb 59/1 were chosen for strain Corby whereas LPS components of the TF 3/1 mutant strain were detected by MAb 8/4 and also by MAb 59/1. We evidenced that LPS components of the MAb 3/1-positive L. pneumophila strain Corby are expressed phase variably during the intracellular lifecycle. The expression of MAb 3/1-positive LPS components on the bacterial surface occur continuously during the intracellular lag-phase and egress phase whereas the replicative bacteria show a loss of these epitopes. MAb 3/1-positive LPS components were delivered coevally in the intracellular host cell environment to a different extent during these stages. In contrast, the expression of the MAb 59/1-positive LPS epitopes increases up to the release of intracellular replicated bacteria and is repressed by the egress of bacteria of the MAb 3/1-positive strain. Unlike to strain Corby the MAb 3/1-negative mutant TF 3/1 does not deliver LPS components, but it shows a continuous expression of LPS components recognized by MAb 59/1 during the whole intracellular lifecycle. Considering potential intracellular targets for modulating effects of Legionella-LPS, we fractionated infected host cells (A. castellanii, HeLa-cells) and tested the obtained cellular fractions (cytoplasm, membrane and membrane organelles, nucleus and cytoskeleton) by ELISA detecting bacterial LPS components of strain Corby. The obtained data revealed positive LPS signals in two fractions: cytoplasm and cytoskeleton. To differentiate bacterial bound and shed LPS components we used confocal laser scanning microscopy for detecting phallotoxin-labelled cytoskeleton in HeLa cells and intracellular LPS components labelled with LPS-specific MAbs for both strains. A colocalization of delivered LPS components with host cell cytoskeleton was not feasible but comparing infected and uninfected HeLa cells during replicative stages, clear structural differences of the cytoskeleton were visible for strain Corby, indicating a degradation of these host cell structures. Our findings substantiate that the LPS expression is regulated environmentally and therefore the shedding of LPS is one form of the equipment of L. pneumophila in order to modulate host cells.
- 162 - Session 5: Immunology and host susceptibility P80
Comparative immunological studies of sera from patients hospitalized during a large outbreak of Legionnaires’ Disease.
Elisabeth Wedegea, Karin Bolstada, Anita Kanestrømb, Eva H. Lindbergb and Øystein Simonsenb aNorwegian Institute of Public Health, P.O.Box 4404 Nydalen, NO-0403 Oslo, Norway; bØstfold Hospital Trust, Cicignongt. 19, NO-1606 Fredrikstad, Norway [email protected]
In 2005, a long-distance outbreak of Legionnaires’ Disease (LD) occurred from an industrial air scrubber in Norway (Nygard et al., 2008). From most patients admitted to the local hospital with pneumonia, sera were obtained at admittance and after 1 and 3 months. They were analysed in IFA (Meridian) and ELISA (Virion/Serion) with antigen pools of different L. pneumophila serogroups. From patients with confirmed LD and probable LD, the serum with the highest antibody level was also immunoblotted with whole-cells from the Benidorm serogroup 1 outbreak strain as antigen. Our aim was to compare the responses in ELISA and IFA and to study if the antibodies were directed to the serogroup-specific lipopolysaccharide (LPS) of the outbreak strain. Confirmed LD (n = 45) was defined as pneumonia with a positive urinary antigen or culture test. Twenty-six (58%) of these patients showed positive responses in both ELISA and IFA, 7 (15%) were positive in one and 12 (27%) negative in both tests. On immunoblots, all but one of the patients with both positive ELISA and IFA levels had IgG and/or IgM antibodies that reacted with LPS of the outbreak strain. Seven more patients with positive antibody levels in either ELISA or IFA or negative levels in both assays also showed an LPS reaction. Thus, a total of 32 patients (71%) with confirmed LD had LPS-specific antibodies. Probable LD, defined as pneumonia with positive ELISA and/or IFA antibody levels, was identified in 38 patients. Twenty-seven patients (71%) had positive responses in both assays, while 11 (29%) were positive in one of the assays. Among these, 29 (76%) patients, showed LPS-specific IgG and/or IgM binding on immunoblots. Significant correlation coefficients were obtained between antibody levels in ELISA and IFA for both patient groups. In conclusion, the results of the ELISA and IFA, based on polyvalent antigens, corresponded for the majority of the patient sera. A high proportion of the confirmed and probable LD patients had antibodies directed to the serogroup-specific LPS of the outbreak strain, indicating that they had been infected by this strain.
- 163 - Session 5: Immunology and host susceptibility P81
Cytomegalovirus latency does not influence clinical presentation and outcome of infection with Legionella pneumophila serogroup 1
Rob Rentenaara, Peter De Jagerb, Peter Schneebergerc, Bart De Witd and Peter Weverc aDept. of Medical Microbiology, Radboud University Nijmegen Medical Center, P.O. Box 9101, 6500 HB Nijmegen, Netherlands; bDept. of Intensive Care and Emergency Medicine, Jeroen Bosch Hospital, P.O. Box 90153, 5200 ME s- Hertogenbosch, Netherlands; cDept. of Medical Microbiology and Infection Control, Jeroen Bosch Hospital, P.O. Box 90153, 5200 ME s-Hertogenbosch, Netherlands; dLaboratory for Clinical Chemistry and Haematology, Jeroen Bosch Hospital, P.O. Box 90153, 5200 ME s-Hertogenbosch, Netherlands [email protected]
Herpesvirus latency in mice induces a prolonged state of macrophage activation that confers resistance to infection with the intracellular bacterial pathogens Listeria monocytogenes and Yersinia pestis. This suggest that herpesvirus latency provides immune benefits to the host (Barton et al. Nature 2007;447:326-9). To address this hypothesis in a human setting, we examined if cytomegalovirus (CMV) serostatus influenced clinical presentation and outcome of infection with the intracellular bacterium Legionella pneumophila serogroup 1. We retrospectively analyzed CMV serostatus in 28 patients with proven legionellosis (urinary antigen positive). CMV serostatus was related to C-reactive protein (CRP) level, white blood cell (WBC) count and CURB-65 score on admission and to intensive care unit (ICU) requirement and death during hospitalization. Anti-CMV IgG antibodies were detected in 15/28 (54%) patients. CRP level and WBC count on admission did not differ between CMV-seronegative and CMV-seropositive patients (CRP 367 ± 147 (mean ± SD) vs 353 ± 136 mg/L and WBC count 13,4 ± 4,8 vs 14,0 ± 5,4 x109/L). Likewise, CURB-65 scores on admission were identical in both groups (2 ± 1). In the CMV-seronegative and CMV-seropositive groups, 7/13 (54%) vs 6/15 (40%) patients were admitted to the ICU (P > 0.05, ns). In the CMV-seropositive group, 2/15 (13%) patients died of multiple organ failure while all CMV-seronegative patients survived. Thus, cytomegalovirus latency does not influence clinical presentation and outcome of infection with L. pneumophila serogroup 1. Immune benefits of herpesvirus latency might be difficult to demonstrate in humans.
- 164 -
SESSION 6 MICROBE - ENVIRONMENT INTERACTIONS A) CELL BIOLOGY, PROTOZOA, BIOFILMS
- 165 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P82
A missing link in Legionella host-parasite interactions
Ann Karen Brassingaa, Jason Kinchenb, Meghan Cuppc, Paul S. Hoffmand and Costi Sifrie aUniversity of Manitoba, Buller Bldg., Rm. 418, 45 Chancellor’s Circle, MB R3T2N2 Winnipeg, Canada; bUniversity of Virginia, MR4 Bldg., Rm. 4072, 409 Land Road, Charlottesville, VA 22908, United States of America; cUniversity of Virginia, MR4 Bldg., Rm. 2143, 409 Lane Road, Charlottesville, VA 22908, United States of America; dUniversity of Virginia Dept. Infectious Diseases, 409 Lane Rd Rm 2143, Charlottesville, 22908, United States of America; eUniversity of Virginia, MR-4 Bldg., Rm. 2138, 409 Land Road, Charlottesville, VA 22908, United States of America [email protected]
Caenorhabditis elegans nematodes are natural microbivores commonly isolated from garden compost and potting soil in which members of the Legionella genus are also natural inhabitants. Since cases of legionellosis have been epidemiologically linked to exposure to potting soil, we aimed to determine whether C. elegans could harbor and disseminate infectious forms of Legionella within a soil environment. In simulated soil assays, we show that Legionella colonizes the intestinal tract of nematodes. Moreover, legionellae replicate within the nematode intestinal tract, do not invade surrounding tissue, and are excreted as differentiated forms with ultrastructural features similar to cyst-like forms observed in host cell-lines. Susceptibility of C. elegans to Legionella is influenced by the evolutionary conserved host innate immune system governed by the p38 mitogen-activated protein kinase (MAPK) and DAF-2 insulin/insulin growth factor-1 receptor signaling pathways that, when elicited, activate transcription of genes encoding antimicrobial products. Intestinal colonization of Legionella pneumophila icm/dot mutant strains does not affect the survival rate of nematodes in comparison to wild-type strains. However, nematodes colonized with L. pneumophila dotA and icmT mutants have significantly higher levels of germline apoptosis than those colonized with wild-type Legionella. Furthermore, apoptosis- deficient nematodes are resistant to Legionella infection. Taken together, the results suggest that C. elegans germline apoptosis is influenced by the Dot/Icm system, perhaps through the production and secretion of anti-apoptotic effector molecule(s), as has been observed in human macrophages. In conclusion, we hypothesize that nematodes of the genus Caenorhabditis are natural hosts for Legionella and that mechanisms for interacting with host cell programmed cell death pathways that is a hallmark of Legionella pathogenesis may have evolved through metazoan-bacterial interactions.
- 166 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P83
In vitro analysis of interaction of Acanthamoeba castellanii with Legionella pneumophila of the filamentous morphological form
Sock Hoai Chan and Chun Chau Sze
School of Biological Sciences, Nanyang Technological University, 60, Nanyang Drive, 637551 Singapore, Singapore [email protected]
Legionella pneumophila in the environment thrives by parasitizing on its protozoan hosts, such as amoeba. In addition, it can be found in biofilms in association with other microbial species, and under specific conditions, is able to form biofilm on its own. It has been reported that when the ambient temperature was raised beyond 37◦C, the cell morphology of L. pneumophila in axenic biofilms elongated from rods to filaments. Much of our knowledge on the interaction of Legionella with its amoeba host has been derived from investigation of infection by rod-shaped Legionella cells; infection by the filamentous form is still largely uncharacterized and remains to be investigated. Our study is aimed at understanding the ecological relevance of L. pneumophila filamentation, in particular the interaction between the filaments and amoeba. We are interested in whether the filaments are as infective as rods, or how this filamentous form can impact the Legionella- amoeba interaction in a way which may provide it with an ecological advantage. Using in vitro infection assays, we compared the infection profiles of the amoeba Acanthamoeba castellanii by L. pneumophila Philadelphia-1 filaments with respect to its rod forms. We observed differential interaction dynamics between the two types of infection. The features of amoeba infection by the filamentous form and its possible implications in relation to the higher proliferative potential of the filamentous form will be discussed.
- 167 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P84
Modulation of host cell phosphoinositide metabolism by Legionella pneumophila
Sabrina Engelhardta, Roger Meierb, Curdin Ragaza and Hubert Hilbia aInstitute of Zoology, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland; bInstitute of Biochemistry, Schafmattstr. 18, CH-8093 Zurich, Switzerland [email protected]
Phosphoinositide (PI) lipids play a central role in signal transduction, cytoskeleton rearrangements and vesicle trafficking in eukaryotic cells. Accordingly, PIs represent a prominent target for Legionella pneumophila and other pathogenic bacteria to subvert host cell processes and to establish a replicative niche. We recently found that L. pneumophila exploits host PIs to anchor to the Legionella-containing vacuole (LCV) bacterial effector proteins that are translocated by the Icm/Dot type IV secretion system. These effectors include SidC and SidM, which specifically bind to phosphatidylinositol 4-phosphate (PtdIns4P) present on LCVs. L. pneumophila modulates the PI levels on LCVs in an Icm/Dot-dependent manner. While more than 80% of LCVs harboring wild-type L. pneumophila accumulate PtdIns4P, vacuoles harboring icm/dot mutant bacteria lack this PI. L. pneumophila might modulate PI levels on LCVs (i) indirectly by interfering with small GTPases and PI-metabolizing enzymes of the host cells, or (ii) directly by secreting bacterial PI phosphatases or PI kinases into the host cells. The Icm/Dot substrates RalF and SidM are guanine nucleotide exchange factors (GEFs) that recruit and activate the small GTPases Arf1 or Rab1, respectively. In turn, activated Arf1 and Rab1 GTPases recruit and activate PI-metabolizing enzymes, such as PI 4-kinase IIIβ (PI4KIIIβ), or the PI 5-phosphatase OCRL1, respectively. Depletion of Arf1 and PI4KIIIβ but not Rab1 by RNA interference reduced the levels of the PtdIns4P- binding effector SidC on LCVs, suggesting that Arf1 and PI4KIIIβ promote the accumulation of PtdIns4P on the LCV membrane. However, LCVs harboring an L. pneumophila mutant strain lacking the Arf1 GEF RalF accumulated the same amount of SidC as LCVs harboring wild-type L. pneumophila. In light of these results L. pneumophila likely modulates PI levels on LCVs directly. Currently, we explore the role of a putative PI phosphatase for LCV formation.
- 168 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P85
Proteome analysis of Legionella vacuoles isolated by immuno-magnetic separation
Ivo Finsela, Simon Urwylera, Curdin Ragaza, Alexander Schmidtb, Ruedi Aebersoldb and Hubert Hilbia aInstitute of Zoology, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland; bInstitute of Molecular Systems Biology, ETH Zurich, Wolfgang-Pauli-Str. 16, 8093 Zurich, Switzerland ivo.fi[email protected]
Intracellular growth of Legionella pneumophila within phagocytic cells depends on the formation of a unique ”Legionella- containing vacuole” (LCV). To form LCVs, the pathogen employs the Icm/Dot type IV secretion system (T4SS), which translocates more than 100 different ”effector” proteins into host cells. These effectors promote uptake of the bacteria, communication with vesicle trafficking pathways and formation of a replication-permissive LCV. However, the role of individual Icm/Dot substrates has been difficult to define, because LCV formation is a robust process, and the effectors seem to be functionally redundant. We recently developed a simple and rapid method, which allows the isolation of LCVs from the social amoeba Dictyostelium discoideum. The two-step purification scheme involves immuno-magnetic separation using an antibody against the Icm/Dot substrate SidC, which localizes exclusively to LCVs, followed by a subsequent gradient centrifugation step. The use of D. discoideum producing a fluorescent LCV marker and fluorescently labeled L. pneumophila allowed tracking the enrichment of LCVs by light microscopy. The proteome of purified LCVs was determined by reversed-phase micro-capillary liquid chromatography coupled to electrospray ionization tandem mass spectrometry (LC-MS/MS) and revealed 566 host proteins. LCV proteins include novel components of the endosomal pathway, as well as the early and late secretory trafficking pathways. We currently adopt the LCV isolation protocol to purify vacuoles harbouring L. pneumophila mutant strains lacking distinct effector proteins and determine the proteome of these LCVs. A comparison with LCVs harbouring wild-type L. pneumophila should provide insights into the function of individual effector proteins. Furthermore, using quantitative proteomics approaches, we will quantitatively determine the changes in host protein levels on LCVs over time.
- 169 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P86
Copper indirectly controls the proliferation of Legionella pneumophila in drinking water heterotrophic biofilms
Maria Salome´ Giao˜ and C. William Keevil
University of Southampton, School of Biological Sciences, Boldrewood Campus, SO16 7PX Southampton, United Kingdom [email protected]
Drinking water biofilms are known to provide a protective haven for several pathogens, including Legionella pneumophila. Conversely, several factors, such as temperature, nutrients, water chemistry and pipe material, can play an important role in the incorporation and survival of this pathogen when imbedded in heterotrophic biofilms. The aim of this study was to study the influence of copper surfaces on L. pneumophila in heterotrophic biofilms and in pure cultures. To study the incorporation of L. pneumophila into drinking water biofilms, a two stage chemostat was used. The first stage consisted in a seed vessel inoculated with microorganisms collected from tap water. The second stage consisted of three vessels in which coupons of different materials (PVC, Copper and PEX) where immersed to grow the heterotrophic biofilm. Upon immersion of coupons the second stage was also challenged with L. pneumophila NCTC 12821. Coupons were removed at several time points and sessile cells where scraped and quantified for cultivable L. pneumophila and L. pneumophila cells labelled with a specific peptide nucleic acid (PNA) probe targeting 16S rRNA. For the pure culture experiments, biofilms were formed in 6-well plates on copper and PVC surfaces in drinking water. At several time points the coupons were removed and the biofilm scraped and quantified as described above. In this case, live and dead L. pneumophila cells were also quantified using the BacLightTM viability kit. All experiments were performed at 30◦C. PNA quantification results showed that L. pneumophila can incorporate into drinking water heterotrophic biofilms, in the absence of amoebae, and persist for at least 32 days, although losing cultivability in <24 hours. This demonstrates a community composition effect on pathogen cultivability. Comparing the different materials it was observed that the numbers of PNA-labelled L. pneumophila were slightly higher on copper (5.81 x 105 cells cm−2) than on PVC or PEX surfaces (2.82 x 105 and 1.76 x 105 cm−2, respectively). To confirm whether PNA-labelled cells were viable or dead on copper L. pneumophila biofilms were formed in pure culture and it was observed that L. pneumophila retained cultivability for at least 32 days. Moreover the PNA-positive numbers (8.44 x 105 cells cm−2) were lower than the numbers of BacLight-viable cells (3.11 x 106 cells cm−2) but higher than the cultivable numbers (4.80 x 105 CFU cm−2), suggesting that the PNA probe does not label dead cells and that the L. pneumophila cells recovered from heterotrophic biofilms and detected by PNA were still alive. Our laboratory and others have previously demonstrated that copper can reduce the numbers of cultivable L. pneumophila in mixed species biofilms; however it is clear here that copper does not influence the survival of L. pneumophila in pure culture. This study provides insight into the indirect non-toxic effects of copper on L. pneumophila and indicates that the metal probably inhibits beneficial heterotophic species that are supporting the cultivability and growth of the pathogen in drinking water biofilms.
- 170 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P87
In vitro antagonistic activity among clinical and environmental strains of Legionella spp, strains showing inter and intraspecific differences
Rafael Gomez-Lus´ a, Carmen Pelazb, Carmen Rubioc, Raquel Becerrila, Francisco Javier Castilloc, Estrella Duran´ c, Concepcion´ Garc´ıac, Silvia Lobez´ a, Mercedes Ocac and Soledad Salvoc aDepartment of Microbiology . University of Zaragoza. IUCA., Domingo S/N, 50009 Zaragoza, Spain; bLegionella Reference Laboratory; Centro Nacional Microbiolog´ıa; Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain; cDepartment of Microbiology. Hospital Cl´ınico ”Lozano Blesa”., Juan Bosco 15, 50009 Zaragoza, Spain [email protected]
In previous studies we found that 228 clinical and environmental isolates of three different families ( Enterobacteriaceeae, Pseudomonadaceae and Vibrionaceae) have the ability to inhibit legionellae. Thus, we demonstrated that six Legionella species and eight serogoups of L. pneumophila were susceptible to a variety of bacteriocins or bacteriocin-like inhibitory substances (BLIS) (1).To our knowledge, Legionella spp have not been investigated as bacteriocin producer, neither the inhibitory effect of Legionella-BLIS against gram-negative and gram-positive bacteria (GNB and GPB) as well as against Legionella isolates. Twenty nine Legionella strains, including 18 species and 11 L. pneumophila isolates, were used. Legionella-BLIS were obtained as 8 mm BCYE agar disks as previously described (1). To detect the BLIS activity against Legionella, the diffusion assay method was used. BCYE-α agar plates were inoculated with each Legionella receptor strain and disks were placed with a forceps. Inhibitory activity was also analyzed against a set of GNB and GPB (reference and clinical strains) by using a similar technique with recommended media. All the Legionella species and serogroups tested were BLIS producers and the pattern of legionellae inhibition was distributed among the strains with an apparent ecological meaning. L. pneumoniae sg1 strains producing BLIS were active against L. bozemanii sg1 and sg2, L. dumoffii, L. feeleii sg2, L. gormanii, L. longbeachae, L. micdadei and L. oakridgensis. The relationship between production of -and sensitivity to- BLIS could be of significant interest for Legionella spp, because there are at least the following three patterns: Good producer and good susceptibility, such as L. pneumoniae sg1; Good producer and moderate susceptibility, such as L. bozemanii sg2; Good producer and bad susceptibility, such as L. jordanis. Conclusions: In the present study we found that 29 Legionella spp strains were BLIS producer. Differences in sensitivity of legionellae to BLIS produced by Legionella spp could support its use in the classification of legionellae. In addition, legionellae produce BLIS active againts GPB (Staphylococcus and Streptococcus), but ineffective against GNB (16 genera).
- 171 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P88
Investigating Legionella pneumophila physicochemical surface properties to explore the intimate interactions with host cells.
Florence Gosselina, Fabien Gaboriaudb, Julien Simoneta, Edith Angela, Gregory´ Franciusb, Christophe Ginevrac, Sophie Jarraudc and Laurence Mathieua aEcole Pratique des Hautes Etudes, LCPME, UMR CNRS 7564, Poleˆ de l’Eau, 15 avenue du Charmois, 54505 Vandoeuvre-les-Nancy,` France; bLaboratoire de Physique Chimie et Microbiologie pour l’Environment, UMR CNRS 7564, 405 rue de Vandoeuvre, 54602 Villers-les-Nancy,` France; cCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France fl[email protected]
Bacteria have complex parietal structures in direct contact with the medium which constitute a zone of exchanges with their local environments. Adhesion to host cells is a significant and crucial phase of the cell-cell interactions and governs the capacity of infection of bacteria. This adhesion mechanism is dependant on both bacteria parietal structures (LPS, pili, flagella,...) and their physicochemical properties (electrostatic interactions, hydrophobicity,...)(1). But, only few studies approached the initial interaction step between Legionella and host cells. And to our knowledge, none explored the physicochemical properties of Legionella surfaces to explain the adhesion phenomenon to host cells. In this pioneering study, our objectives were to investigate the physicochemical surface properties (charge and hydrophobicity) of different L. pneumophila sg1 (Lp1) strains and, to explore their influence on the infection kinetics of a host model, Acanthamoeba castellanii. Nine clinical Lp1 strains from Dresden panel (2) and the Lp1 Lens strain from both clinical and environmental origins were studied. Among them, six were MAb3/1 positive. Physicochemical properties of Lp1 strains were characterized at the micrometric scale using microelectrophoresis and at the nanometric scale with atomic force microscopy (AFM) (3). The first allows the measurements of electrophoretic mobility (electrostatic interactions) and the second quantifies the hydrophobicity and specific adhesion forces of surface biopolymers. Different ionic strengths (1 to 100 mM) at neutral pH of the medium were studied in order to determine the consequences on the physicochemical surface properties and infection capacity of the Lp1 strains. Our results showed that the LPS nature of the Lp1 strains contributed to their negative electrical charges and consequently to their electrophoretic mobilities. Indeed, the ionic strength-mobility patterns of MAb3/1 positive Lp1 were significantly lower (absolute value) than MAb3/1 negative Lp1. This could explain the strong potential of epidemicity of the MAb3/1 positive strains, as they presented lower charge distributed throughout the wall and/or the polymer fringe, leading to weak electrostatic repulsions. This hypothesis was reinforced by infection assays with A. castellanii which showed that the more the charge was low, the more the infection was important (quantitatively and in terms of kinetics). Complementary information, particularly the cell surface imaging (presence of flagella and pili) and the mapping of cell hydrophobic domains, were obtained with AFM done on the Lp1 strains with contrasted mobility patterns. These data could lay the molecular basis of the L. pneumophila adhesion mechanisms to host cells, which are missing today. Aknowledgment: The authors thank S. Jarraud and C. Ginevra from CNRL INSERM-U851 who provided all the Legionella strains. References: (1) Clements A. et al., (2008), PLoS One, 3(11), e3817. (2) Helbig J. et al., (2002), Eur. J. Clin. Microbiol. Dis., 21, 710. (3) Gaboriaud F. et al., (2006), Colloids and Surfaces B: Biointerfaces, 52, 108.
- 172 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P89
Large particles of Legionella pneumophila formed inside Acanthamoeba castellanii have a complex spherical LPS-architecture that does not depend on special serotypes
Clemens-Johannes Schumachera and Jurgen H. Helbigb aTU Dresden, Institute Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany; bInstitute of Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany [email protected]
Extensive outbreaks of community- acquired Legionnaires’ disease were dominantly caused by Legionella pneumophila serogroup 1 strains that express the ”virulence-associated” LPS-epitope recognized by the monoclonal antibody (MAb) 3/1 of the ”Dresden Panel”. The aim of our study was to compare the LPS-architecture of different serotypes regarding the so-called ”large particles” formed in natural host cells. Using confocal-laser-scanning and conventional microscopy we focussed primarily on differences between a MAb 3/1-positive strain and four MAb 3/1-negative strains belonging to different serogroups of L. pneumophila. For every strain we used two different monoclonal antibodies that recognise sub- and serogroup-specific LPS-epitopes. To elicit if the spherical LPS-architecture depends on bacteria of different growth- phases we checked the used antibodies for growth-phase-variability by quantitative ELISA. Growth phase dependence of LPS expression in liquid culture was found for several epitopes, but there was no correlation to ”large particle” formed in vivo. According to our aims we did not find any differences in the LPS-architecture of the large particles. All examined strains showed a complex spherical LPS-architecture in correlating morphological structured particles that appeared after intracellular growth in Acanthamoeba castellanii. Furthermore we observed large bacterial particles with this particular LPS-structure that appear inside of cyst-forming amoebae that have not been described before. Different LPS-types on bacterial surfaces seem to have various functions. Among the used strains there were no indices for differences in quantity or quality of the formed ”large particles”. Therefore, we suggest the ability to form clusters does not account for high infective potency of 3/1-positive strains. Maybe high hydrophobic properties of MAb 3/1 recognized epitopes enhance airborne transmission.
- 173 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P90
Influence of temperature and flow velocity on the proportion of Legionella pneumophila in natural biofilms
Cecile´ Maurice-Blanc, Sylvie Viboud and Dominique Fontvieille
UMR CARRTEL (Centre Alpin de Recherche sur les Reseaux´ Trophiques des Ecosystemes` Limniques), Domaine Universitaire, 73376 Le Bourget du Lac Cedex, France [email protected]
Two hot springs from the Aix-les-Bains catchment area are permanently colonized by Legionella pneumophila. While thermal baths and health care treatments of course no longer relate on these springs, they still feed the Tillet river hydrological network that itself is a tributary of Lake Bourget. The study aimed to figure out what was the fate of L. pneumophila in biofilms grown along the river from the point of hot springs connection down to the lake. We observed both, changes in the physiological status of attached bacteria (culturable vs viable but non culturable bacteria) and a decrease in their concentration downstream the point where hot springs waters join the river down to the lake. In a second part of the study, effects of temperature and flow velocity on the concentration of L. pneumophila in natural biofilms were experimented. The experiment intended to simulate the fate of L. pneumophila contained in biofilms fragments when they are moved from the hot sulfur water where they developed downstream to the conditions of the recipient river. For this experiment, glass slides were exposed to biofilm colonization during 15 days in a settling tank of the Aix les Bains National Thermes. The slides were then installed in biological reactors fed with river water where both, flow velocity and temperature were under control. Two flow velocity and three temperatures have been tested. High flow favored the culturable form of L. pneumophila at the expense of other culturable bacteria. This result was confirmed by direct analysis on the Aix- Les-Bains hydrological network, in which we observed higher concentrations in culturable L. pneumophila along the fastest part of the river. However there was no significant change in the proportion of all L. pneumophila by comparison to the total bacterial community (whatever their physiological status). As for the temperature of the recipient water, the experiment showed that the closest it was the thermal water, the higher was the concentration of culturable L. pneumophila. A decrease or no change in the concentration of other culturable bacteria occurred simultaneously. Our results then suggested that high flow of the river in summer conditions would increase the proportion of culturable L. pneumophila in biofilms, thus impairing self-purification processes.
- 174 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P91
Interrelationship between Legionella pneumophila and free-living amoeba at low temperature
Akira Ohnoa, Naoyuki Katob, Chikako Hakiia and Keizo Yamaguchia aDepartment of Microbiology and Infectious Disease, Faculty of Medicine, Toho University, 5-21-16 Omori-nishi, Ota-ku, 143-8540 Tokyo, Japan; bDepartment of Chemistry, Faculty of Medicine, Toho University, 5-21-16 Omori-nishi, Ota-ku, 143-8540 Tokyo, Japan [email protected]
Legionaires’ disease appeared with the spread of artificial warm water systems. It is widely thought that legionella acts as parasite for protozoa, mainly free-living amoeba in nature, and legionella switched host cell from protozoa to alveolar macrophage very similar to amoeba via aerosol from man-made warm water system. Therefore, pathogenicity of Legionella pneumophila against host cells has been studied at warm temperature, usually body temperature. However, legionella is living in natural water system that the water in the temperate regions is usually below 20 degrees C in temperature except for summer season. It is little understood how free-living amoeba such as acanthamoeba and haltomanella interacts with legionella at low temperature. We reported that an intracellular kinetics assay of Acanthamoeba castellanii using different strains and doses showed decreases of intracellular counts of L.pneumophila at temperatures below 20 degrees C with significant differences to 25 degrees C, further, over-expression of type 1 metacaspase, a stimulator of A.castellanii encystations, was associated with a reduced number of bacteria within amoeba at temperature below 20 degrees C (ohno A et al. Appl Environ Microbiol 2008,74:4585-4588) This study aimed to clarify interrelationship between free-living amoeba and L.pneumophila at low temperature, especially in connection with encystations.
- 175 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P92
Spatial Arrangement of Legionella Colonies in Lab-Scale Model Cooling Tower Systems
Michael Taylora, Kirstin Rossb and Richard Benthama aFlinders University, Department of Environmental Health, Environmental Health, Flinders University, GPO Box 2100, 5001 Adelaide, Australia; bDepartment of Applied Science, Batchelor Institute of Indigenous Tertiary Education, PO BOX 106 PARAP NT, 804 Darwin, Australia michael.taylor@flinders.edu.au
Legionella has been shown to multiply within biofilms in both anthropogenic and natural water sources, either in association with bacteria and algae or as an intracellular parasite of free living amoebae. To investigate these interactions visually, laboratory derived biofilms grown in simulated cooling tower systems were tagged using fluorescent in-situ hybridisation (FISH). The probes targeted Legionella, protozoa, and heterotrophic bacteria. Lipopolysaccharide was counterstained using DAPI. These tagged films were then imaged using confocal scanning laser microscopy (CSLM) and 3D representations of each sample were generated. Within these films, although the overall architecture of the film was not uniformly flat, the distribution of microorganisms throughout the thickness of the film was mostly homogenous. In contrast to this relatively even distribution, Legionella appeared as small, separate microcolonies, with distinct boundaries, suggesting that their formation is mediated by specific environmental conditions. In some instances Legionella were observed in close proximity to viable protozoans, in other cases microcolonies were not aggregated in such a manner. Statistical analysis using Pearson’s correlation coefficient of these images confirmed positive associations between Legionella and protozoa was a relatively common occurrence in the sampled biofilms. These data also suggest protozoa as a crucial host for Legionella replication in biofilms ’in situ’. This work represents the first time that Legionella has been fluorescently tagged within a complex, naturally generated multispecies biofilm.
- 176 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P93
Destructive enzyme activities in the secretome of Legionella pneumophila
Jana Tiefenaua, Frederike Fresea, Frank Galkaa, Sun Nyunt Waib, Susanne Engelmannc and Michael Steinerta aInstitut fur¨ Mikrobiologie, Technische Universitat¨ Braunschweig, Spielmannstr.7, 38106 Braunschweig, Germany; bDepartment of Molecular Biology, Umea˚ University, 6K och 6L, Sjukhusomradet,˚ SE-901 87 Umea,˚ Sweden; cInstitut fur¨ Mikrobiologie, Ernst-Moritz-Arndt-Universitat, F.-L.-Jahn-Str. 15, D-17487 Greifswald, Germany [email protected]
Secreted effector molecules are critical for the extracellular pathogenicity of Legionella pneumophila, which is characterized by considerable tissue destruction, including extracellular matrix degradation and focal septal disruption. Specific secretion machineries which are responsible for the subfraction of secreted proteins (soluble supernatant proteins, SSPs) and the production of bacterial outer membrane vesicles (OMVs) both contribute to the protein composition of the extracellular milieu of this lung pathogen. In the present study we performed a comprehensive proteome comparison of proteins secreted by different secretion systems (SSPs) and the OMV fraction of proteins of L. pneumophila. Protein identification and assignment analysis revealed a total of 181 supernatant proteins, 33 of which were specific to OMVs and 107 of which were specific to the remaining SSP. A functional classification showed that a large proportion of the identified OMV proteins are involved in the pathogenesis of Legionnaires’ disease. Additionally, the binding of immunofluorescently stained OMVs to alveolar epithelial cells, as visualized by confocal laser scanning microscopy, suggested that there is delivery of a large and complex group of proteins and lipids in the infected tissue in association with OMVs. The zymography and enzyme assays performed in our study revealed that the SSP and OMV fractions possess proteolytic and lipolytic enzyme activities which may contribute to the destruction of the alveolar lining during infection. The observed proteolytic effects could be due to several identified proteins as shown by extracellular matrix degradation assays. Hence, SSP and OMV may promote the dissemination of L. pneumophila within the lung tissue by degrading extracellular targets of the lung epithelium barrier.
- 177 - Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P94
Diversity and identity of free-living protozoa in unchlorinated drinking water
Rinske Valstera, Bart Wullingsa, Geo Bakkerb and Dick Van Der Kooija aKWR, Watercycle Research Institute, Groningenhaven 7, 3430 BB Nieuwegein, Netherlands; bVitens Water Technology, Oude Veerweg 1, 8019 BE Zwolle, Netherlands [email protected]
Proliferation of Legionella pneumophila in warm tap water installations implies the presence and growth of free- living protozoa, serving as its host. However, information about host protozoa, e.g,. Hartmannella vermiformis and Acanthamoeba spp., and other free-living protozoa in engineered water systems is scarce, which may be attributed to the limitations of microscopic and cultivation techniques for detection of protozoa. To overcome these difficulties, we applied cultivation-independent molecular techniques targeting parts of 18S rRNA genes for the detection and identification of free-living protozoa, which are predominating in two unchlorinated drinking water supplies. Samples (< 20◦C) were collected from treated water at the plant, distributed water and distribution system biofilms of groundwater water supply A with a low concentration of natural organic matter (NOM) in treated water (< 0.5 ppm of C) and groundwater supply B with a high concentration of NOM (7.9 ppm of C). The study revealed that H. vermiformis, detected with q-PCR, was present in both treated water types at concentrations between 1 and 30 cells liter−1. H. vermiformis was observed in one of the eight biofilms of supply B at 4.3 cells per 10 cm2 and in none of the biofilms of supply A. The high level of NOM in supply B corresponded with elevated levels of active biomass (ATP) and with elevated concentrations of H. vermiformis in the distributed water in the summer. Acanthamoeba spp. were detected in one treated water sample at 4 cells liter−1 and in two biofilms at about 1 cell per 10 cm2 of supply A. In treated water and biofilms from both supplies, highly diverse protozoan communities were observed, with sequences that clustered with Amoebozoa, Cercozoa, Choanozoa, Ciliophora, Euglenozoa, Myzozoa and Stramenopiles. None of the OTUs (18S rRNA gene sequences that shared ≥ 99% similarity) were observed in both supplies, demonstrating that unique protozoan communities were present. In supply A 54 OTUs were observed and in supply B 72 OTUs. However, the estimated values for protozoan richness did not differ significantly between the two supplies. These observations demonstrate that water supplies, which comply with all quality criteria, contain highly diverse protozoan communities.
- 178 -
SESSION 7 MICROBE - ENVIRONMENT INTERACTIONS B) DETECTION IN NATURAL AND ARTIFICIAL RESERVOIRS - SURVIVAL, PERSISTENCE
- 179 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P95 Bacterial Colonization and Incidence of Legionella in Cooling Towers of Southern General Company for Fertilizer Production in Basrah City
Hadeel Al-Hadithia and Jundi Mehdib aFaculty of Pharmacy / University of Jordan / Amman, Amman/ Jordan, 11942 Amman, Jordan; bPolytechnique, Baghdad / IRAQ, don’t know Baghdad, Iraq [email protected]
Abstract During the period between April-August 2002, Sixty six water samples were collected from Inlets, Insides and Outlets of cooling towers (22 samples each, once a week). Samples were screened for bacterial colonization with special focus on Legionella. Percentage occurrence of predominant bacteria present in water samples collected from Inlets included: Klebsiella (95.45%), Enterobacter (90.9%), Pseudomonas (86.36%), Escherichia coli (81.81%), and Legionella (72.72%). While those predominant in water Inside cooling towers were: Pseudomonas (100%), Staphylococci (81.81%), Legionella (81.81%) and Bacillus (72.72%). Incidence of Legionella recoverd from Insides and Outlets of cooling towers was similar . Four species of Legionella were identified in water samples collected from Inlets, Inside and Outlets of cooling towers, in the following consecutive frequencies L. pneumophila ( 72.72%, 81.81% and 81.81%) , L. jordans (18.18, 36.36 and 27.27%), L. dumffii (9.09, 22.72 and 22.72%) and L. oekidegenes (4.54, 13.63 and 18.18%). The increased percentage occurrence of Legionella in water collected from Inlets and Insides of cooling towers, as well as those of Pseudomonas and Staphylococci suggests that Legionella is also a hardy organism , being potentially survive as free organism despite water disinfection .
- 180 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P96 Immunomagnetic capture enhance Legionella pneumophila serogroup 1 detection and recovery from environmental samples.
Severine´ Allegra, Franc¸oise Berger, Philippe Berthelot, Florence Grattard, Bruno Pozzetto and Serge Riffard
Universite´ Jean Monnet, Faculte´ de Medecine,´ 15 rue Ambroise Pare,´ 42100 SAINT-ETIENNE, France [email protected]
Rapid and sensitive techniques are needed to detect and optimize the quantification of Legionella spp. in environmental samples. The main objective of this study was to develop an immunomagnetic separation (IMS) method to improve the recovery, by plate culture, of L. pneumophila sg1 (Lp1) from environmental water samples. IMS was performed with magnetic Dynabeadso` coated with a monoclonal antibody (Mab, Dresden panel) directed towards the lipopolysaccharide of Lp1. The test was first evaluated, in combination with culture, on a panel of 39 Lp1 strains isolated from different water systems and then on 54 environmental samples (4 biofilms, 10 cooling towers and 40 sanitary hot waters). The performance of IMS was evaluated by comparing the rate of bacteria recovered by combining IMS and culture to that obtained after culture according the norm NF T90-431. This IMS-culture assay was found able to detect 33 of the 39 independent strains of Lp1 (a recovery percentage > 50% was considered significant), a similar specificity pattern than when using a genus-specific anti-Legionella polyclonal antibody, but with a better sensitivity (101±4% vs 89±4%; P <0.05), especially when water samples were spiked with increased amounts of interferent flora. From the 54 environmental samples, 8 and 35 samples were tested negative and positive by both IMS-culture and standard culture. Four samples gave discrepant results and the last 7 were not interpretable by standard culture. From these 7 latter samples, 4 were found positive for Lp1 after IMS-culture and the last 3 became positive with this technique after an acid treatment. When the quantitative results of the two techniques were compared, the IMS-culture was found more sensitive than standard culture, with a significant difference (P <0.05), especially in samples displaying a high level of contamination with heterotrophic microflora. Our results demonstrate that the IMS-culture assay is adapted to the detection of Lp1 in highly contaminated environmental samples (biofilms or water from cooling towers). Works are in progress: (i) to broaden the spectrum of recognized Legionella strains by adding Mab(s) with complementary specificity and, (ii) to evaluate Legionella viability by combining IMS with ATP bioluminescence assay, qPCR and flow cytometry.
- 181 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P97 Use of DGGE to study and follow Legionella population in environmental waters
Sandrine Baylea, Benoit Roiga, Sophie Jarraudb,Jer´ omeˆ Etienneb and Laurent Garrellyc aEcole des Mines d’Ales,` 6 avenue de clavieres, 30319 Ales, France; bCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; cGL Biocontrol, Parc Scientifique et technique Georges Besse, 30035 Nˆımes, France [email protected]
3 GL-Biocontrol innovup ERIEE Nimes The tools of the molecular biology and the real-time PCR in particular, are used in numerous laboratories to detect and to quantify Legionella and Legionella pneumophila in environmental waters. In particular, surface waters, ground waters, and distribution waters are the purpose of surveys when they are used to feed cooling facilities and sanitary hot water systems. Upon request of French sanitary authorities and to give the measures more reliable, the AFNOR (French Agency of standardization) published in 2006 a norm establishing the requirements concerning the performances of Legionella PCR methods. ISO (international Standard Organisation) take up also the works. On environmental samples, genomes of L. species (other than L. pneumophila) are much more frequently detected than L. pneumophila using PCR method. Classic methods on culture media do not allow the culture of all the species of Legionella resulting from environmental waters. GVPC and BCYE are specific adapted media for pneumophila species. Therefore it is necessary to use molecular methods, to describe the different species present in samples. Among these, DGGE (Denaturing Gradient Gel Electrophoresis) is a nucleic acid based (DNA or RNA) technique which can be used to profile and identify dominant members of the microbial community based on their genetic fingerprint. The aims of this study was to conduct DGGE method on L. pneumophila and 18 Legionella species recognized as human pathogens, and to apply this method directly on environmental samples. We defined a specific DGGE reference profile for L. pneumophila and for 13 Legionella species studied. Three facilities (1 cooling tower and 2 hydrotherapic establishments) were investigated to study the ecology of Legionella in the water networks. Results showed that certain species found in the primary resource disappear in the system while others settle down and proliferate. This method, as a supplement to the quantification of Legionella spp. using real-time PCR, allows to refine the risk assessment and to define one or several species settled in a particular resource and to follow it during time..
- 182 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P98 A comparison of culture methods and fluorescent in situ hybridization (FISH) for detecting Legionella in commercial potting mixes
Harriet Whiley, Michael Taylor and Richard Bentham
Flinders University, Department of Environmental Health, Environmental Health, Flinders University, GPO Box 2100, 5001 Adelaide, Australia richard.bentham@flinders.edu.au
Legionella longbeachae has proven to be an organism of major clinical significance in reported cases of Legionellosis in Australia. The primary environmental reservoir and carrier of infectious materials in L. longbeachae infections has been shown to be commercial potting mixes. A route of transmission of the organisms to susceptible hosts has yet to be established. Historically, positive detection of Legionella from environmental samples has been obtained by culture methods incorporating acid treatment of the sample, followed by 5-7 day culture on selective media containing glycine, vancomycin, polymixin and cycloheximide (GVPC). This methodology has to be adapted according to the content of the potting mix and results are highly variable. This combined with the very high microbial load in potting mixes has resulted in a working limit of detection of >1000 organisms per gram. In the paper we present successful non-culture based methods for detection form potting mixes including real-time PCR and fluorescent in situ hybridization (FISH). These methods have been demonstrated to be more sensitive and generally more reproducible than culture techniques and to providing results within hours rather then days. They also have their own shortcomings, namely an increased sensitivity to environmental conditions which may result in the generation of false negatives/positives. Culture methods were compared to FISH and fluorescent microscopy for detection of Legionella within potting mixes. Results for culture were unsatisfactory with fungal growth and spore-forming bacteria frequently eclipsing the emergence of Legionella colonies. Positives culture results were obtained in 1/8 samples tested. FISH provided positives results in 8/8 samples tested. The potential for false positive/negative results has not been eliminated by this technique, but the rapid turn around and increased sensitivity of the technique shows promise as an alternative to an already unreliable culture method.
- 183 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P99 Occurrence of Legionella-like amoebal pathogens and other amoeba-associated microorganisms in cooling towers and a municipal water system
Mary Faronea, Sharon G. Berkb, Tu Vua, Jeremy Friesena, Christopher Garnera, Tim Jacobsa, Brian Choatea, Celia Gendrona, John Gundersonc and Anthony Faronea aMiddle Tennessee State University, Biology Department, 1301 Main Street, Murfreesboro, AK 37132, United States of America; bTennessee Technological University, Center for the Management, Utilization and Protection of Water Resources, 1020 Stadium Drive, P.O. Box 5033, Cookeville, TN TN 38505, United States of America; cTennessee Technological University, Biology Department, Pennebaker Hall Rm 207, Cookeville, AK TN 38505, United States of America [email protected]
Between 2007 and 2008, water samples from 85 municipal fire hydrants and 10 cooling towers in the middle Tennessee region of the United States were examined for the presence of amoebae infected with intracellular microorganisms. Infected amoebae were identified in 7 of the fire hydrants and 5 of the cooling towers. Six of these infections were successfully passed into laboratory strains of Acanthamoeba polyphaga. Only one of the isolates is culturable on BCYE agar and real-time PCR analysis of extracted DNA established that none of the isolates were Legionella pneumophila. Sequence analysis of 16S rDNA shows that two of the isolates are closely related to the Legionella lytica clade, while another isolate appears to be Legionella-like amoebal pathogen 14 (LLAP-14) with a fourth isolate closely related to LLAP-14 but nonculturable. A fifth isolate is closely related to Legionella drancourti but is also nonculturable. Previous studies by our group have resulted in the isolation of infected amoebae containing organisms similar to the L. lytica clade with greater frequency than those infected with L. pneumophila. We have also isolated novel intracellular bacteria with similarities to Coxiella burnetii. To begin to assess the distribution of these and another newly described organism, mimivirus, in constructed water sources, DNA was extracted from cooling tower and hydrant samples and amplified with primers specific for L. lytica, Coxiella-like amoebal pathogens , and mimivirus. Of the 10 cooling towers, 4 were positive for L. lytica DNA and one for DNA of the Coxiella-like organisms. Two of 9 fire hydrant samples were positive for the L. lytica clade and one sample was positive for mimivirus DNA as confirmed by nested PCR. This may represent the first detection of mimivirus in the United States.
- 184 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P100 Detection of Legionella spp. in potting soil from different sources
Jacob Bruina, Simona Casatib, Jeroen Den Boera, Valeria Gaiab and Ed Yzermana aRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands; bIstituto cantonale di microbiologia, via Mirasole 22A, 6500 Bellinzona, Switzerland [email protected]
Background: The presence of legionellae in potting soil is not routinely investigated. Also, the compost facilities where potting soil is produced are not routinely maintained for the control of legionellae. Potting soil contains substrates that can be contaminated with legionellae and therefore can cause Legionnaires’ disease (LD) by spreading aerosols while processed. The aim of this study was to investigate the presence of legionellae in potting soil produced in two different countries. Methods: In this study we investigated 127 samples of potting soil; 101 samples from The Netherlands and 26 from Switzerland. The samples from The Netherlands were collected at various compost facilities and at retailers of potting soil distributed over the country. The samples from Switzerland were collected from 8 green waste collection sites, including 3 compost facilities and 5 gardening waste collection sites. All samples were divided in equal parts and each sample was investigated in two laboratories, one in The Netherlands, the other in Switzerland. In both laboratories the investigation of legionellae was performed using their own routine laboratory method. Results: The combined results of both laboratories showed that 24 out of 127 samples (18,9%) were culture positive. Nine samples (out of 101: 8,9%) originated from The Netherlands and 15 samples (out of 26: 57,6%) originated from Switzerland. Overall, 52 different Legionella spp. were found. Nine samples contained only 1 strain of Legionella spp. and 15 samples 2-5 strains. Six strains of L. pneumophila SG 1 were cultured and 32 strains of L. pneumophila SG 2-15. Fourteen strains belonging to L. non-pneumophila were cultured, including the species considered potentially pathogenic for humans (L. bozemanii, L. micdadei and L. oakridgenis). The culture results from the laboratories differentiated mainly in a higher percentage of positive cultures for L. non-pneumophila in the Swiss laboratory. Conclusions: There is a big difference in the presence of legionellae between potting soil produced in Switzerland (57,6% culture positive) and potting soil produced in The Netherlands (8,9% culture positive). In Australia and Japan, working with potting soil is already recognized as a risk factor to contract LD. This study showed that also in Europe potting soil may be a potential source for acquiring LD.
- 185 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P101 Compost facilities as alternative reservoirs of Legionella spp.
Simona Casatia, Jacob Bruinb, Lisa Conzaa and Valeria Gaiaa aIstituto cantonale di microbiologia, via Mirasole 22A, 6500 Bellinzona, Switzerland; bRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands [email protected]
Data on the presence of Legionella outside the aquatic environment are scant. Alternative ecosystems that could act as a reservoir for Legionella spp. have to be investigated to identify unconventional contaminated substrates that can produce bioaerosols: potting soils are an example in Australia, in Switzerland and in the Netherlands. In the present study, we investigated 8 green waste collection sites including 3 compost facilities (private manufacturers) and 5 gardening waste collection centres (community centres) situated in the southern part of Switzerland (Canton Ticino). Of these, 3 were short-term centres (green waste kept on the site for a maximum of 3 months) and 2 long-term centres (with no structured management of the green waste). A total of thirty-one samples were collected between Mai and October 2008. Samples were taken in each centre from the fresh material, in the 3 compost facilities from different piles and in the 2 long-term gardening waste collection centres from the decomposing pile. In the interpretation of the results, we considered all samples collected in each centre as a single collection point. Legionella strains were isolated from 6 (75%) out of the 8 green waste collection points analysed. The 3 compost facilities and the 2 green waste collection centres (long-term storing) were positive. In contrast, Legionella seems to be absent (or not detectable by culture) in 2 out of the 3 green waste collection centres with a short-term storing system of the fresh green material. The compost, final product of the composting chain, was always positive in the 3 facilities. In contrast, the fresh green waste seems to be free of Legionella; only in one site we could isolate Legionellae. L. pneumophila sg 1-15, L. bozemanii, L. cincinnatiensis, L. jamestowniensis, L. micdadei and L. oakridgensis were isolated from 6 of the 8 sites. The Legionella contamination ranged from 103 to 108 CFU/g. Compost facilities seem to be an important reservoir for Legionella. Although the importance for human health of the presence of Legionella spp. in compost is unknown and the risk assessment has yet to be defined, bioaerosols produced from these elements should not be underestimated.
- 186 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P102 Chlorine regulation of virulence genes expression in enviromental Legionella pneumophila isolates
Beatrice Casini, Paola Valentini, Francesca Torracca, Angelo Baggiani and Gaetano Privitera
University of Pisa, Dept Experimental Pathology, M.B.I.E., via S. Zeno 35-39, 56123 Pisa, Italy [email protected]
The Legionella pneumophila type IV protein secretory systems are composed of membrane-associated and transmembrane proteins, which are involved in the delivery of virulence factors modulating host defenses. Lvh locus, encoding a type IVA secretion system, is essential for intracellular growth, but may also contribute to Legionella ability to multiply in plumbing system and persist in the presence of toxic biocides. Aim of this work was to investigate the effect of exposure to chlorine on virulence genes expression and the relationship with reduced susceptibility to chlorine in some strains isolated from the Azienda Ospedaliero-Universitaria Pisana water network. Representative Legionella pneumophila sg 1 strains, isolated from the hospital water system prior to (March 2002- April 2003) and following (May 2003-June 2009) the start of hyperchlorination treatment, were analysed for chlorine- susceptibility to 2.5 mg/L of chlorine in accordance with BS EN 1040:1997, using L. p. Philadelphia (ATCC 33152) as reference strain. To investigate if virulence genes expression is regulated by exposure to chlorine, the mRNA level of the type IVA secretion system (lvh) and other virulence genes, such as mip (macrophage infectivity potentiator) and the region rtxA (entry gene), was quantitatively determined using relative qRT-PCR analysis on cells exposed to chlorine and harvested at different time points, using untreated cells as internal control. L. p. Philadelphia was also treated and used in comparative analysis. Strains isolated after water hyperchlorination showed reduced chlorine-susceptibility, being still detectable after 120 minutes of exposure to chlorine, while pre-chlorination strains showed a 5 log reduction of the bacterial count after 60 minutes. The quantification of virulence genes expression in pre- and post-chlorination treated strains showed that, compared with L. p. Philadelphia, the expression of all virulence genes was markedly and more promptly up-regulated in post-chlorination isolates: the expression levels reached a peak after 30 minutes of chlorine treatment compared with 120 minutes needed for pre-chlorination strains. Moreover expression was about 32768 and 16 fold higher than those observed respectively in L.p. Philadelphia and in pre-chlorination isolates. For their survival, successful pathogens have developed different defence strategies to counter stress situations. Exposure to chlorine seems to up-regulate the virulence genes expression in some strains of Legionella pneumophila sg 1. The increased ability to survive on adverse environmental conditions through the expression of virulence genes that regulate the internalization in host-cell, may account for the environmental adaptation and persistence of Legionella spp. colonization of water systems.
- 187 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P103 Detection and quantification of viable Legionella cells from environmental water samples by combined use of ethidium monoazide and real-time PCR
Bin Changa, Toshitsugu Tagurib, Kanji Sugiyamac, Junko Amemura-Maekawaa, Fumiaki Kuraa and Haruo Watanabea aDepartment of Bacteriology I, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, 162-8640, Tokyo, Japan; bNagasaki Prefectural Institute for Environmental Research and Public Health, Ikeda, 2-1306-11, 856-0026, Oomura City, Nagasaki, Japan; cDepartment of Microbiology, Shizuoka Institute of Environment and Hygiene, 4-27-2, Kita-ando, Aoi-ku, 420-8637, Shizuoka City, Shizuoka, Japan [email protected]
Legionellae are ubiquitous inhabitants of biofilms in aquatic environments and moist soil, replicating as intracellular parasites of protozoa. The bacteria widely exist in man-made water systems and cause legionellosis in humans. Because it takes 4-7 days to isolate Legionella organisms from environmental and clinical samples, we have worked on development and improvement of rapid detection methods in order to identify sources and routes of infections caused by Legionella at an early stage. In this study, ethidium monoazide (EMA) treatment was used together with real-time PCR targeting 16S or 5S rRNA genes for specific detection of DNA from viable Legionella cells. EMA could covalently link to the genomic DNA and inhibit PCR amplification of dead Legionella cells. Combined use of EMA and real-time PCR targeting 16S rRNA gene could specially and rapidly determine the number of viable cells in environmental samples. However, detection sensitivity of real-time PCR was low (≥100 CFU/100 ml). In order to increase sensitivity, EMA treatment and real-time PCR targeting 5S rRNA gene (1 CFU/100 ml) was tried. As a result, the number of Legionella cells estimated by the combination of EMA treatment and real-time PCR targeting 5S rRNA gene was found to be larger than that determined by plating, suggesting that EMA treatment could not completely inhibit PCR amplification targeting 5S rRNA gene of DNA from dead cells. It may be attributed to the difference between the lengths of PCR fragments targeting 16S and 5S rRNA genes and/or between the effects of EMA binding to the DNAs of the two genes. In conclusion, combined use of EMA and real-time PCR targeting 16S rRNA gene is appropriate to specifically detect viable Legionella cells from environmental samples.
- 188 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P104 Free-living amoebae and Legionella in bioaerosols from composting facilities of southern Switzerland
Lisa Conza, Simona Casati and Valeria Gaia
Istituto cantonale di microbiologia, via Mirasole 22A, 6500 Bellinzona, Switzerland [email protected]
Legionella is present worldwide in natural freshwater environment and human warm water system. The bacterium was also isolated in soil, potting soil and compost. In Australia Legionella longbeachae is the infectious particle associated with gardening while in Ticino (southern part of Switzerland) Legionella pneumophila and L.bozemanii were the agent mainly found in compost and potting soil. Legionella multiplication in amoebae was demonstrated using co-cultural methods. To date, 13 species of amoebae, some of which human pathogens, were identified as reservoirs and vectors for Legionella. Amoebae are able to survive in critical environmental conditions by forming cysts and may thus protect their endosymbionts such Legionella spp. and have an impact on their infectious capacity and in their in vitro cultivability. The role of amoebae as a vector during the aerosolisation of contaminated substrates (for example water, soil, potting mixes, composting material) has not yet been investigated in detail. Only little research has been carried out on the simultaneous presence of amoebae and Legionella in bioaerosols. The aim of this study is to investigate whether or not the composts and the bioaerosols developed from pile fermentation contain viable Legionella and amoebae. Bioaerosols were collected by a Coriolis µ air sampler. The protozoa were cultured on non-nutritive agar plates covered with E.coli as a food source. Legionella spp. were cultivated on GVPC agar plates and by co-culture with axenic Acanthamoeba polyphaga. Seven green waste collection centers were sampled. The aerosols of five centres were positive for Legionella after co-culture and two were positive for amoebae. Amoebae and Legionella spp. were isolated from composts of six and five centres, respectively. Four of the positive compost facilities were followed for 4 months. Legionella and amoebae were found in all 4 composts analysed. L.pneumophila and L.bozemanii were also detected in bioaerosols hovering over the heap but not nearby. This study showed that viable amoebae and Legionella are both present in composts and bioaerosols. Further studies are needed to investigate the spread from compost to bioaerosol of Legionella and the dispersion of these bioaerosols.
- 189 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P105 Legionella Colonization of the Czech Environment and the Associated Health Risk : A Ten Year Study.
Vladimir Drasara, Radomir Polcarb and Norman Fryc aNational Legionella Reference Laboratory, Public Health Institute, Masarykovo namesti 16, 682 01 Vyskov, Czech Republic; bFactor.E, Masna 5, 602 00 Brno, Czech Republic; cHealth Protection Agency Centre for Infections, Respiratory and Systemic Infection Laboratory, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected]
Introduction. Following the first nosocomial outbreak of L.pneumophila sg.3 in Prague,1998, clinical and environmental investigations were initiated.The occurence of particular Legionella species, serogroups and subgroups were compared with the aim of finding new approaches to risk assessment. Methods. Representative sampling was performed according to accredited standard operating procedures and EU Guidelines,2003.Isolates were identified by a quantitative microagglutination test with in-house rabbit hyperimmune sera raised against 52 Legionella species and serogroups. Strains with unusual serological patterns were identified by mip sequencing and the clinical L.pneumophila sg.1 isolates typed by sequenced-based typing(SBT). Results. 71% of hotels(n=60),69% of spa hotels(n=48),89% of hospitals(n=60) and 77% of cooling towers(n=52) were colonized with Legionella. Except hotels, where sg.6 was the most frequent isolate, sg.1 prevailed in the remaining facilities.Followed by sg.6,sg.3 and sg.5/10.The subgroup 3/1+ occured at low frequency(8-13%). The saprophytic L.spiritensis and L.rubrilucens dominated among Legionella species. Waters from 7 spas did not contain legionella at the source, but they were found during storage and distribution.Thermal waters were the most colonized.As many as 4 distinct L.pneumophila and 7 other Legionella species were isolated from one network. From 1998-2008,52 clinical strains were isolated.L.pneumophila sg.1 predominated(48%). The subgroup 3/1+ accounted for 80% of these,followed by sg.3 and sg.6. Serogroups 2,4,5,8 also appeared. Five strains belonged to L.parisiensis-anisa-bozemanii group, L.micdadei and L.maceachernii.14 different SBT occurred among L.pneumophila sg.1 isolates.Two profiles were new, and ST 62 was the most common. Five travel associated cases were notified from hotels colonized with 3/1+.Four other 3/1+ infections came from thermal waters used for medicinal purposes.Only two L.pneumophila sg.6 were reported from a boatel.Nosocomial cases provided a broader spectrum of serogroups where sg.3 and sg.6 prevailed.Community acquired infections showed a similar picture. Conclusion. The majority of large buildings are colonized with legionella and it appears that the mere presence of the 3/1+ strain represents a risk.As opposed to travel associated cases, hospital patients are susceptible to a wider range of L.pneumophila serogroups,colonizing the pipework.Community acquired pneumonia cases comprised both 3/1+ and some other serogroups and species.These results are consistent with those from Denmark and some other EU countries. They support a proposal to include the SBT of L.pneumophila in risk assessment in addition to determining the legionella concentration which is mandated by legislation.
- 190 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P106 Rapid Identification of Legionella Species using Whole Cell MALDI-TOF Mass Spectrometry
Michal Drevineka and Vladimir Drasarb aNational Institute for NBC Protection, Kamenna 71, 26231 Milin, Czech Republic; bNational Legionella Reference Laboratory, Public Health Institute, Masarykovo namesti 16, 682 01 Vyskov, Czech Republic [email protected]
Legionella species can be divided into several subgroups, which differ in their aggressiveness to human organism. Current techniques for the identification of Legionella are mainly based on serological methods, but the monoclonal antibodies are restricted to subtypes of L. pneumophila sg.1 and polyclonal antibodies are not commercially available. Furthermore, the Legionella genus exhibits wide range of serological cross-reactions (e.g. L. pneumophila sg.5,8 with L. gormanii, dumoffii, worsleiensis), which make the discrimination and identification within the genus difficult. We present simple, rapid and robust method based on whole-cell MALDI-TOF MS fingerprinting for characterization and discrimination of species within Legionella genus using species-specific biomarkers or their combination. Mass spectral fingerprints obtained for more than 50 Legionella species showed characteristic profiles with tens to more than a hundred peaks. The signals have been categorized by the m/z ratio, relative intensity and frequency of the appearance within the mass spectral fingerprints of the strains. In several cases characteristic signals, belonging to only one species, were observed. Other species can be discriminated using a combination of the biomarkers, which was found to be unique for each species. BioTyper software was used for the automated construction of reference peak list and consequently a reference library based on main spectra was created. The library was used for identification of environmental isolates and the results compared to polyclonal antibody tests. In all cases the classification was possible down to the species level; in most cases even the identification of sub-species or strain was achieved. Furthermore, the method is able to identify serologically indistinguishable species, eg. L. parisiensis, L. bozemanii, L. anisa or L. spiritensis, L. taurinensis. This study demonstrates that MALDI-TOF MS is a powerful method for rapid identification within Legionella genus, which might be useful for tracking source of infection and for epidemiologic studies in general.
- 191 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P107 Usefulness of Real-Time PCR for Detection of Difficult Growing Legionella spp. from a Water Supply.
Laura Franzin, Daniela Cabodi, Nicoletta Bonfrate and Valerio Demarie
Amedeo di Savoia Hospital, Corso Svizzera 164, 10149 TORINO, Italy [email protected]
Introduction: Culture is the gold standard for Legionella detection from environmental samples. However real-time PCR is sensitive and rapid method. In this study we evaluated the usefulness of this method to detect Legionella from a hospital water supply contaminated by Amoeba. Methods: We examined 74 water samples, collected from a hospital water supply (cold water, hot water tank and hot tap water of patients room) during eight-year period. Legionella culture was performed from five liter samples concentrated by filtration. Real-time PCR was performed on one liter of 49 water samples, after filtration and DNA extraction. Quantitative detection of L. pneumophila and Legionella spp. was performed with iCycler and Bio-Rad reagents. Amoeba culture was performed from 100 ml of 67 water samples. Identification of Acanthamoeba spp. was performed by PCR using specific primers (mitochondrial 16SrRNA). Eterotrophic bacteria plate count was performed at 37◦C and 25◦C. Free chlorine and pH were also determined. Results: Culture was always negative, except in the last sampling, where very little and slow-growing colonies were found after subculture only in two samples (600 cfu/L). Colonies were identified as Legionella by specific 16SrRNA and mip primers. Real-time PCR was negative for L. pneumophila in all 49 samples examined, but positive for Legionella spp. in 45 (91.8%) samples. The detection limit of the method was 80 GU/L and the lower quantification limit was 480 GU/L. In 27 samples low level of Legionella spp GU/L was found. The values of 18 samples ranged from 5x102 to 3x105 GU/L. Amoeba was found by culture in 6 (8.9 %) samples. Identification of strains by PCR showed negative results for Acanthamoeba spp. Free chlorine was <0.1 mg/L. Conclusions: Culture is still considered the gold standard for Legionella detection, providing strains available for further research. In this study real-time PCR showed persistent positive results for years despite negative culture, suggesting the presence of non culturable or difficult growing strains of Legionella spp. Moreover the presence of Amoeba in the water supply could enhance Legionella growth and possibly predict a greater risk of infection.
- 192 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P108 Legionella and Amoeba from Water Samples
Laura Franzin, Daniela Cabodi, Nicoletta Bonfrate and Paola Fresi
Amedeo di Savoia Hospital, Corso Svizzera 164, 10149 TORINO, Italy [email protected]
Introduction: Legionella is ubiquitary in acquatic environment, where is living in symbiosis with Amoebae. The aim of the study was the detection of Legionella and Amoeba from different water supplies. Methods: 1513 water samples from 22 hospitals were examined: 1387 hot water (5 L) and 126 cold water (5 L). 44 cold water samples (1 L) were analysed from air conditioning systems. Samples were concentrated by membrane filtration and Legionella was quantitatively determined by culture and real-time PCR. Aliquots of direct, heat-treated and acid-treated concentrates were plated on BCYE, BMPA and MWY for culture. After incubation at 37◦C, suspected colonies were typed and quantified after 15 days. Quantitative detection of L.pneumophila and Legionella spp. was performed with iCycler and Bio-Rad reagents. For Amoeba culture, concentrate from 100 mL of water was plated on Non Nutrient Agar with Escherichia coli and incubated at 25◦C and 37◦C up to 7 days. Amoeba isolates were tested by PCR using Acanthamoeba spp specific primers. Results: Legionella was isolated from 789 (52.1%) water samples of 90% hospitals. L.pneumophila serogroups 1, 2, 3, 5, 6, 8, 15 and 2-15, Legionella spp, L.bozemanii, L.anisa and other fluorescent Legionella species were found. Culture of Amoeba was positive in 202 (13.4%) samples of 81.8% hospitals. Legionella and Amoeba were both present in 10.2% samples of water supply, while 3.2% were only positive for Amoeba. Amoeba occurrence was significantly higher in hospital water samples with Legionella count >103 cfu/L. Legionella was found in 6.8% and Amoeba in 50% of cold water samples from air conditioning systems. Isolation rate of Amoeba was higher when culture was incubated at 25◦C than 37◦C (68.8% vs 12.9%) and 14.8% of strains was identified as Acanthamoeba spp. Weak correlation between results of Legionella culture and PCR values was observed, with GU/L higher than cfu/L. Conclusions: Legionella isolation rate was higher from hospital water supply. Amoeba was found in higher percentage from air conditioning systems water, especially for culture performed at 25◦C. As Legionellae multiply intracellulary in Protozoa that contribute to the spread of the bacteria, Amoeba occurrence may be considered a factor promoting the increase of Legionella colonization in water systems.
- 193 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P109 Prevalence and degree of Legionella colonization in cooling towers and associated variables.
Marian Garc´ıa-Nunez˜ , Sonia Ragull, Maria Luisa Pedro-Botet, Lourdes Mateu and Miquel Sabria`
Hospital Germans Trias i Pujol, carretera de canyet, 8916 Badalona, Spain [email protected]
Cooling towers have been frequently implicated in community Legionella outbreaks. Since July 2003, Spanish regulation obligates test water from cooling tower systems for Legionella and perform a chlorine shock if colonization is above 1,000 CFU/L. This last measure cause important logistic and economic consequences troubles. Other guidelines increase the cut off to above 10,000 CFU/L Legionella counts, considering little concern levels between 1,000 and 10,000 CFU/L. Prevalence of Legionella colonization in cooling towers, degree of colonization, and its relationship with the outbreak are under known. The aim of this study was to know the prevalence and degree of Legionella colonization in cooling towers in Catalonia (Spain) and to found associated variables. Two hundred seventy-nine cooling towers were randomly screened for physic-chemical and microbiological determinations during a three year period (2003-2006). Physic-chemical determinations: Temperature: 17.9 ± 5.8 ◦C; conductivity: 2.27 ± 2.87 ms/cm; turbidity: 41.87 ± 249.5 N.F.U.; iron: 0.38 ± 0.8 ppm; total hardness 456.4 ± 475.2 ppm CaCO3; carbonate hardness: 127.0 ± 151.1 ppm CaCO3; calcium hardness: 266.7 ± 242.6 CaCO3; alkalinity (p) 56.9 ± 91.6 ppm CaCO3; alkalinity (m): 340.3 ± 328.9 ppm CaCO3 and Langelier Index (LI): 1.1 ± 1.7. According the standard ISO6222:99, the 43% of the samples had inocula >10,000 CFU/ml of total aerobic counts. However, the 70% of theses samples had negative results for Legionella. The total bacterial counts were correlated with the temperature and inversely correlated with alkalinity (p) and the LI. Legionella was detected in thirty-nine samples (13.8%) by the standard procedure ISO11731:98. The Legionella inocula ranged from 50 CFU/L to 2,000,000 CFU/L. According to the Spanish regulation of hazard categories: the 30% of the positive cooling towers were between 100-1,000 CFU/L (revise the system and take samples after 15 days), 25.6% of the positive cooling towers were between 1,000-10,000 CFU/L (corrective actions, disinfection and sampling after 15 days) and the 20% were above 10,000 CFU/L (stop the system and disinfection). The Legionella isolation was associated with higher total bacterial counts at 37◦C and 22◦C (p<0.001) (the 87.5 % of the positive Legionella samples had total bacterial counts >10,000 CFU/ml) and higher levels of carbonate hardness (p=0.001).
- 194 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P110 Evaluation of the usefulness of a new direct immunofluorescence assay (ScanVIT- Legionella TM) for monitoring hospital water systems contaminated with Legionella spp.
Monica Giacomuzzi, Savina Ditommaso, Marino Gentile and Carla Zotti
Universita` degli Studi di Torino, Via Santena 5 bis, 10126 Torino, Italy [email protected]
The most commonly used method for environmental surveillance of Legionella is the standard culture technique by which Legionella organisms can be isolated and their number in environmental samples estimated. A recent diagnostic test for environmental testing is the ScanVIT-Legionella TM method (Vermicon, Munich Germany) based on the detection of Legionella using gene probe technology which enables quantification as well as simultaneous detection of cultivable Legionella and L. pneumophila within three days. With this study we wanted to compare the efficiency of the ScanVIT- Legionella TM test versus a conventional culture method for the quantification of Legionella spp. in hospital water samples in daily hospital practice. Culture method: Legionella spp. were isolated by concentrating 1 liter of water using 0.22 µm polycarbonate filter membranes. ScanVIT-Legionella TM : Legionella spp. were isolated by concentrating 50 ml of water filtered through 0.45 µm nitrocellulose filter membranes. The detection of Legionella spp. takes place on a cultivated filter (marked gene probes enter the bacteria and bind to the matching signatures within the cells). The results are analyzed using a fluorescence microscope. All bacteria that light up green belong to the genus Legionella; all bacteria that light up both green and red belong to the species L. pneumophila. The tests were carried out on a total of 79 water samples from 10 hospitals in Piemonte. Legionella was isolated in 72% and in 67% of the samples by the ScanVIT-Legionella TM test and the culture method, respectively. ScanVIT test had a sensitivity of 90%; agreement between the two methods was 82%. Calculation of the coefficient of Cohen’s kappa showed good concordance between the two methods (K=0.64; P<0.001). 9 samples tested positive with the ScanVIT test but negative with culture: the reason for the difference could be due to the low number of Legionella in the samples (20-80 Legionella CFU/L), which is below the detection limit of the culture method. In the 48 samples that tested positive with both methods, the Legionella concentration detected by the culture method was consistently higher than that detected with the ScanVIT test. A statistically significant difference between the results obtained with the two test methods emerged at the Wilcoxon test (P<0.001). Therefore the test may be recommended for investigating the presence of Legionella (qualitative testing). Given the simplicity of colony identification by fluorescence, the ScanVIT test can be used in laboratories where staff are not experienced in identifying typical colonies of Legionella.
- 195 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P111 Research of Legionella contamination in dental units’ waters and aerosols samples
Ays¸ın C¸otuk, Duygu Goksay¨ , Nihal Dogru˘ oz¨ and Esra Ilhan Sungur
Istanbul University, Faculty of Science, Department of Biology, Vezneciler, 34134 Istanbul, Turkey [email protected]
Legionella species may be present in a variety of man made water systems, including cooling towers, spas, shower heads and dental units. Legionella species which can be spread via aerosols and result in two forms of disease: a pneumonic form called Legionnaires’ disease and a non-pneumonic form called Pontiac fever. In dentistry, contaminated water may be aerosolized and inhaled by the patients and dentists. Legionella can enter the DUs from drinking water mains and are able to proliferate. In USA, a single fatal case of Legionella dumoffii pneumonia has been reported in a dentist exposed to contaminated water from his practice. The aim of this study was to analyze the presence of Legionella species in outlet water and aerosols of high speed drills. Samples were collected from 20 private dental offices in the morning before patients arrived. Water samples were taken from the high-speed drills and these samples were studied using filtration method. Aerosol samples which come from high-speed drill were collected by active sampling using the portable microbiological air sampler in dental surgery room. All samples were inoculated onto BCYEA (with GVPN) and incubated at 37◦C for 14 days. Water temperatures, free chlorine and pH values were measured. Legionella sp. were not detected in any water and aerosol samples. Average tempareture of water was recorded as 23.2 ◦C. Nine out of 20 dental units were detected low levels of free clorine (0.3-3 ppm). The pH values of water varied between 6.71- 7.76. There might be two potential factors playing a role about presence/absence of Legionella in DUWLs. Firstly, the existence of microflora in the same aquatic environment during the isolation of Legionella is important. Legionella must have relationship with other bacteria and free-living amoebae in order to survive. High concentrations of Gram negative bacteria, which generally produce bacteriocines may inhibit or create stress condition on Legionella. Therefore, Legionella were unable to grow or to be identified by classical culture tests. The other factor which should be noted here, some of Legionella spp. are in viable but non culturable form, which is become difficult to detect using classical culture method.
- 196 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P112 A Legionella typing method for ecological studies: Infrequent Restriction Site Polymerase Chain Reaction (IRS PCR)
Delphine Jakubeka, Matthieu Le-Bruna, Gerard Leblonb, Michael Dubowb and Marie Bineta aEDF R&D, Departement´ LNHE, 6, quai Watier, 78400 Chatou, France; bDepartment of Genetics and Microbiology , University of Paris XI, CNRS, UMR8621, 91405 Orsay, France [email protected]
The family of Legionellae comprises more than 50 species some of which are pathogenic to humans. The specie Legionella pneumophila is responsible for about 90% of the diagnosed cases of Legionellosis. These bacteria, which have an aquatic habitat are found in natural waters (lakes, rivers, moist soil) but also in artificial waters (water heaters, pools, cooling towers). Although this bacterium has a public health interest, few studies on their ecology have been performed. It was previously shown that some populations of Legionella present a great diversity and an important spatio-temporal variation. The reference method used for the identification of Legionella are: growth on a specific culture media followed by agglutination tests to identify species, molecular typing by Pulsed Field Gel Electrophoresis (PFGE) to identify strains and sequencing by Sequence Based Typing (SBT) to study their phylogeny. These methods, although widely used in epidemiological investigations, are not adapted to an ecological study on a large scale as they are fastidious, long and time-consuming. Thus, a typing method, easy to implement and fast to execute is required. S. Riffard has developped the typing method named Infrequent Restriction Site Polymerase Chain Reaction (IRS PCR) to the analysis of Legionella genus. Then he conducted a comparative study between IRS PCR and PFGE over fifty strains and showed that this method has good reproducibility, discriminating power, and is fast and easy for Legionella typing. However, this study was conducted on too few samples to validate the method for ecological study. So we optimized and validated the IRS PCR technique on several reference strains and 500 environmental strains of Legionella. The discriminating power and the phylogenetic ability of IRS PCR were respectively compared to PFGE and SBT on about 100 environmental strains. As well, to guarantee the representativity of the ecological diversity, the minimum Legionella strains to type in each sample was determined on rarefaction curves. We showed that the IRS PCR is simple, fast, reproducible and discriminated at the strain level, so an appropriate method to the large scale ecological study of Legionella that we plan in 2010.
- 197 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P113 Occurrence of legionella and other pathogenic microbes in communal waste water treatment plants with low water temperature
Jaana Kusnetsova, Piia Airaksinena, Tarja Pitkanen¨ a, Ari Kauppinena, Asko Vepsal¨ ainen¨ a, Ilkka Miettinena and Eila Torvinenb aNational Institute for Health and Welfare, Water and Health Unit, P.O.Box 95, FI-70701 Kuopio, Finland; bUniversity of Kuopio, Department of Environmental Sciences, P.O.Box 1627, FI-70211 Kuopio, Finland jaana.kusnetsov@thl.fi
An abundant growth of legionella and mycobacteria is possible in industrial waste water treatment plants, especially in active sludge basins with aeration. A few of those water plants have also recently been associated with Legionnaires’ disease cases and a Pontiac fever outbreak in Finland. The earlier studies have focused on only the industrial sites with relative warm waste water. This time communal waste water plants treating much colder waste water were studied. A total of 19 aeration basins from 17 communal waste water treatment plants were studied for legionella, mycobacteria, E. coli, intestinal enterococci, campylobacteria, C. perfringens, noroviruses, adenoviruses, amoebae, flagellates and ciliates. From only one water treatment plant of the studied 17 (6%) viable legionella bacteria was possible to isolate (105 cfu/l of Legionella sp.; mean 103 cfu/l). When compared to earlier findings with industrial waste water treatment plants where up to 109 cfu/l of legionella have been isolated (57% of the plants, mean 108 cfu/l), the communal waste water plants seem to be much less contaminated with legionella. Instead, mycobacteria were isolated from 16 of the 17 plants (94%, up to 108 cfu/l, mean 107 cfu/l). When compared to the industrial plants with up to 109 cfu/l of mycobacteria (100%, mean 108 cfu/l), the communal plants seem to be almost as frequently contaminated with mycobacteria, but with lower concentrations. E. coli, intestinal enterococci and C. perfringens were found in each communal plant (100%), but campylobacteria only from 47%. Noroviruses existed in 71% and adenoviruses in 100% of the plants. Amoebae were seen in 94%, flagellates 88% and ciliates 53% of the plants. Thus, opposite to legionella, these other microbes were common in the waste water systems. Because of the small number of the legionella positive sites in this study, the exact correlations between legionella and other environmental factors was not possible to determine. The studied communal plants were treating rather cold waste water (17◦C), which may explain this weaker legionella growth in these systems compared to the industrial waste water systems (31◦C). The occurrence of legionella in different waste water systems has yet to be determined individually.
- 198 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P114 Prevalence of Legionella strains from cooling towers and laboratory diagnosed legionellosis cases in New Zealand
Robert Laua, Brian Caughleya, David Harteb and Rob Deaconc aCollege of Sciences, Massey University at Wellington, New Zealand, Wallace Street, 6021 Wellington, New Zealand; bInstitute of Environmental Science & Research Ltd, Kenepuru Drive, 5022 Porirua, New Zealand; cEnvironmental Laboratory Services Ltd, 85 Port Road, 5010 Lower Hutt, New Zealand [email protected]
The study compared the prevalence of Legionella strains isolated from cooling towers in New Zealand with those in laboratory diagnosed legionellosis cases over a twelve month period. In 2008, 696 cooling towers in New Zealand were tested for Legionella strains. A total of 3904 water samples from these cooling towers were tested. During the same period the Legionella species that caused 74 laboratory diagnosed legionellosis cases were compared. Of 80 Legionella isolates from cooling towers, 10 (12.5%) were Legionella pneumophila serogroup 1, 10 (12.5%) Legionella anisa and 9 (11.2%) Legionella pneumophila serogroup 8. Of the 74 human legionellosis cases, 21 (28.4%) involved Legionella pneumophila serogroup 1, 38 (51.4%) Legionella longbeachae species. Between 2004 and 2007, of 283 legionellosis cases, 98 (34.6%), 36 (12.7%), 15 (5.3%) involved Legionella pneumophila serogroup 1, Legionella longbeachae serogroup 1 and Legionella micdadei respectively. There were 17 deaths among legionellosis cases between 2004-2008. In 2008 of 74 legionellosis case,25 (33.8%)were Legionella pneumophila and 38 (51.4%) Legionella longbeachae cases, while for cooling towers, 41 (51.3%) were Legionella pneumophila serogroups and 1 (1.2%) Legionella longbeachae species. Legionella anisa was more frequent in cooling towers (12.5%) than in legionellosis cases (0.0%) and this may reflect the lower virulence of this species. The presence of Legionella pneumophila strains in cooling towers in New Zealand means that regular monthly monitoring for Legionella bacteria, well managed biocide treatment and properly maintained cooling towers are essential in preventing outbreaks of Legionnaires’ disease. References: Australian Standard/New Zealand Standard (2000). AS/NZS 3666 Part 3:2000 Air-handling and water systems of buildings - microbial control - performance based maintenance of cooling water systems. Greig J E, Carnie JA, Tallis GF, et al. (2004). An outbreak of Legionnaires’ disease at the Melbourne Aquarium, April 2000: investigation and case-control studies. Med J Aust 180; 566. Doleans A, Aurell H, Reyrolle M, et al. (2004). Clinical and environmental distribution of Legionella strains in France are different. J Clin Microbiol. 42; 458.
- 199 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P115 Impact of water quality on the transfer and the survival of aerosolised Legionella .
Matthieu Le-Bruna, Thi-Lan Hab, Sylvie Soreaua and Marie Bineta aEDF R&D, Departement´ LNHE, 6, quai Watier, 78400 Chatou, France; bCentre Scientifique et Technique du Batiment,ˆ 84 avenue Jean Jaures,` 77447 Marne-la-Vallee,´ France [email protected]
Cooling circuits select thermophilic microbial population. Among these populations, bacteria of the genus Legionella, of which certain species and in particular Legionella pneumophila is responsible for an atypical pneumopathy the legionnaires’ disease are detected. The contagion is made by inhalation of sprays of the bacterium. The risk assessment associated with this pneumopathy is incomplete due to some lack of knowledge. The characterization of the transfer water / air of the bacteria in condition establish a starting point in the evaluation of the transfer in the atmosphere of Legionella. In this study, the transfer water / air and the post aerosolization survival of the bacterium was studied by various metrological approaches according to the nature of the water. Two waters of industrial origin and a deionised water were estimated. The main obtained results are: i) the strong impact of the quality of water on the phase of transfer water / air of culturable Legionella, ii) post aerosolisation fate is modulated by the water quality.
- 200 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P116 Metrological evaluation of a system of collection of biological aerosols in a cooling tower of a french nuclear power plant.
Matthieu Le-Brun, Marie Binet and Celine´ Bouteleux
EDF R&D, Departement´ LNHE, 6, quai Watier, 78400 Chatou, France [email protected]
Cooling circuits select thermophilic microbial population. Among these populations, bacteria of the genus Legionella, of which certain species and in particular Legionella pneumophila is responsible for an atypical pneumopathie the legionnaires’ disease are detected. The contagion is made by inhalation of sprays of the bacterium. The risk assessment associated with this pneumopathy is incomplete due to some lack of knowledge. The characterization of the term of transfer water / air of the bacterias in cooling towers in real conditions establish a starting point in the evaluation of the transfer in the atmosphere of Legionella. The proposed study aimed at estimating the behavior of a collector of biological sprays in industrial conditions towards a certain number of criteria, namely the point of sampling, the duration of collection, the presence of a pre-concentrator by virtual impaction, the various metrological approaches of measure of the population of Legionella (culture, molecular biology, laser cytometry). The obtained results indicate: i) the possibility of defining a range of linearity between the volume of pumped air by the device and collected population (total mesophilic aerobic flora) ii) the revealing that in the conditions of functioning, the system of collection protects partly the population of the most vulnerable legionella (i; e. the cultivable fraction), iii) the approach by laser cytometry should allow for additionnal developments to estimate a term of global transfer of Legionella.
- 201 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P117 Distribution of Legionella species from the environmental source of public facilities in Korea
Hae Kyung Leea, Jeong Im Shima, Woo-Sik Kimb, Jae-Yon Yua and Yeon Ho Kanga aKorea Centers for Disease Control and Prevention, 194, Tongil-lo, Eunpyeong-gu, 122-701 Seoul, Republic of Korea; bChungCheongNam-Do Institute of Health & Environment, 441- Gayang 2-dong, Dong-gu, 300-801 Dajeon, Republic of Korea [email protected]
Background: Legionella species are ubiquitous in natural and airtifical water environments and associated with legionellosis. The survey of Legionella contamination in airtifical water environments of the public facilities such as buildings, hotels, public baths, and hospitals has been recognized as a significant factor for the prevention and control of legionellosis. The purpose of the study was investigating the distribution and diversity of Legionella species from the public facilities in Korea. Methods: A total of 564 Legionella were collected from public facilities in 5 metropolitan cities and 8 provinces from June to September in 2008. Direct fluorescent antibody assay was used to identify Legionella isolates and genetic diversity was done by sequence-based typing (SBT). Results: Of the 564 isolates, 482(85.5%) were L. pneumophila and 82 (14.5%) were other Legionella species. In L. pneumophila, serogroup 1 was 55% (265/482) and L. anisa among non-L. pneumophila species was 48.8% (40/82). L. pneumophila sg 1 was dominant in the facilities such as buildings, hotels, public baths, hospitals, and factories, while L. pneumophila sg 6 was dominant in springs. The second dominant strains, however, showed the difference depending on the facilities as L. anisa in the buildings (10.8%), L. pneumophila sg 5 in public baths (21.6%), L. pneumophila sg 6 in factories (12.0%), and L. pneumophila sg 7 in hospitals (13.0%). 109 L. pneumophila sg 1 isolates randomly selected were grouped into 34 SBT profiles. In 34 SBT profiles profile 3 (1,4,3,1,1,1), as the dominant, was accounted for 45% and distributed widely in all 13 areas. While profile 25 (7,12,17,3,35,11) was shown in 5 areas, profile R (3,4,1,1,14,9) as L. pneumophila sg 1 (Philadelphia-1, ATCC 33152) was only in one area. Conclusions: The distribution of Legionella species showed some difference among types of facilities or water samples. L. anisa was more dominant in large buildings than other facilities such as public baths and hot springs. SBT profiles of L. pneumophila sg 1 represented diversity in Korea.
- 202 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P118 Improvement of airborne Legionella survival in presence of organics and minerals in waters
Julien Simoneta, Thi-Lan Hab, Enric Robineb and Laurence Mathieua aEcole Pratique des Hautes Etudes, LCPME, UMR CNRS 7564, Poleˆ de l’Eau, 15 avenue du Charmois, 54505 Vandoeuvre-les-Nancy,` France; bCentre Scientifique et Technique du Batiment,ˆ 84 avenue Jean Jaures,` 77447 Marne- la-Vallee,´ France [email protected]
If the Legionella pneumophila occurrence in the hydrous environments is largely documented, very few works attempted to characterize Legionella in aerosols, its only transmission mode. To our knowledge, no study exists on the Legionella pneumophila behaviour in aerosols in relation with the water quality. Therefore, within the framework of a research project (1), it was proposed to characterize the survival of Legionella pneumophila in aerosols according to their physiological states in water. The originality of this study was the approach integrating both the source (water), its abiotic factors (organic and mineral matters), and the transmission vector (aerosols). The biological model was Legionella pneumophila Lens strain from a clinical origin (2). An aerosolization chamber was used to generate controlled and reproducible L. pneumophila aerosols (3). Collection of the airborne Legionella was performed using liquid impingement method (3). L. pneumophila were detected by methods providing different informations on their physiological states: cultivability, membrane permeability (PI staining), hybridized Legionella (FISH) and total bacteria number (DAPI staining). In order to study various Legionella physiological states, laboratory assays were carried out on four different water qualities (mineral or distilled waters with or without 2 mgTOC/L added) and after a period of starvation of 11 days at 37◦C. Controls were performed in parallel without starvation (T0). Contrary to controls, the survival of airborne L. pneumophila was modified in aerosols generated from the 11-days suspensions starved in the four waters tested. Indeed, the presence of minerals in water led to a better survival of the airborne L. pneumophila: < 1 % cultivable Legionella and in average 22% of Legionella with non-damage membranes were detected in distilled water compared to more than 50 % in waters with minerals. In the same way, the TOC supplementation of distilled water allowed an increase by a factor 5 of the cultivability of airborne Legionella pneumophila. No difference appeared for FISH method between the four water qualities (in average > 60 % for hybridized L. pneumophila). This research enabled to better determine the Legionella survival in relation with their ”history” in water, in the perspective of a more relevant management of this microbial risk. Acknowledgments: The results of this study were obtained within the scope of a research programm funded by the French National Research Agency and coordinated by Pr. Jacques Frere` from the LCME laboratory in Poitiers, France. The authors thank all the partners of this project for their discussion, and particularly Dr. S. Jarraud from CNRL who also provided the Legionella strain and Pr. J. Frere` who provided the organic matter. References (1) Research projet named “ Legioa´ eropatho´ ” supported by ANR (French National Research Agency). (2) Centre National de Ref´ erence´ des legionelles´ (F-Lyon) - INSERM U851 provided the Legionella strain. (3) Deloge-Abarkan M., Ha T.L., Robine E., Zmirou-Navier D. & Mathieu L., (2007), JEM, 9, 91.
- 203 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P119 Opportunity catch-study on Legionellosis surveillance. Support legionellosis surveillance for Microbiology. A collaborative effort
Ma Antonia Carratala´a, Juan Bayoa, Antonio Salazara,M◦ Dolores Oceteb, Concepcion´ Gimenob, Remedios Gunab, Lorena Jordan´ c, Ana Ralloc and Enrique Mirac aCentro de Salud Publica´ de Valencia. Conseller´ıa de Sanidad., San Vicente Martir 83, 46007 Valencia, Spain; bUnidad de Microbiolog´ıa. Consorcio Hospital Universitario de Valencia, Avenida Tres Cruces n◦ 2, 46007 Valencia, Spain; cAmbientalys Consultor´ıa y Analisis,´ Les Ones n◦ 6, Bajo Derecha, 46980 Paterna, Valencia, Spain [email protected]
Background: In Valencia metropolitan area 144 cases of Legionnaires’ disease (LD) were notified without identify the infectious source during 2004-2008. Being the mean incident of 3,1/ 100.000 habitants. The cases present a seasonal distribution with maximum incident peaks in February, May-June and September- October. The mean ages of the cases were 62 years and 14 deaths occurred. Of the 144 studied cases, 2 were assumed travel-associated, 116 are from community area, 7 were nosocomial and 5 from unknown origin . L. pneumophila serogrup 1 was isolated from all cases. Clinical origin of the cases was not known in 1, and in 3 Pontiac’s Fever was diagnosed, pneumonia was diagnosed for the rest. Aims: Study the domiciles as source of infection of LD when no other source is found. Methods: From March to June 2009, 8 cases were notified with no infection source associated. Duplicate water samples were taken from every case: 2 from the home of the case, one from the more used point and the other from the minor used point. Two more duplicate samples were taken as control, one in the nearest public establishment from case home and the other far away from the house . A total of 60 samples were taken. Isolation of Legionella bacteria were done parallel by culture and PCR Results: L. pneumophila was no detected in the 60 samples. Legionella spp. was detected in 3 samples (5 %), 2 of them (3.33 %) by culture and 1 (1.67 %) by PCR. From the 8 studied cases, Legionella spp. was detected in samples from 3 cases (37.5 %), in 1 case (12.5 %) from the domicile and the other 2 (25 %) from the control samples. 66 years was the mean age of the studied patients, not existing significant differences with the mean age of the patients declared during the period 2004-2008. Conclusions: 1)It does not seem that infection source of sporadic cases of LD is related to survival in patients domiciles water, there for, probably Legionella has others reservories that still we do not know. 2) Realize an opportunity catch-study on Legionella surveillance for 2 years for LD prevention.
- 204 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P120 ”Legionella spp on sea”: preliminary results of a surveillance program on the Italian Military Boats.
Maria Teresa Montagna, Christian Napoli, Roberta Iatta, Teresa Marsico and Salvatore Barbuti
University of Bari, Piazza G. Cesare 11, 70124 Bari, Italy [email protected]
Introduction. In 2006, the Italian Ministry of Health, in an official document concerning Legionella prevention measures, reported the opportunity of an environmental surveillance on cruise ships, even in absence of legionellosis cases. With reference to this problem, some Authors reported a wide Legionella contamination on tourist cruise ships. On the contrary, few data have been reported regarding military ships. In light of these data, the Apulian Regional Epidemiological Centre (OER) is carrying on a surveillance program on the military ships berthed at the Apulian ports of embarkation (POE). Methods. According to the study protocol, all the ships leaving for an at least 7 days trip or as long as to need the water supplying by removing salt from sea water are enrolled. Water will be sampled before leaving (T0) when the tanks are filled with water coming from the POE water system and at the end of the trip (T1) when the tanks are filled with desalinized water. Results. Till now, 25 water samples at T0, coming from two boats, have been tested: 9 (36%) resulted positive for Legionella pneumophila serogroup 13 (88.9%) and serogroup 7 (11.1%), showing a microbial count range of 200-11.500 ufc/L. Discussion. When comparing these preliminary results with data coming from scientific literature, the military boats show contamination level lower than the ones reported from tourist cruise ships. However, a great attention must be paid to the prevention protocols currently in use in order to better protect our soldiers’ health.
- 205 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P121 Dynamics of Legionellae in a Mediterranean coastal river subjected to different pollutions
Nathalie Parthuisota, Philippe Lebarona, Aurelie´ Touron-Bodilisb and Julia Baudart-Lenfanta aUPMC Univ Paris 06, UMR7621, Laboratoire d’Oceanographie´ Biologique de Banyuls, Observatoire Oceanologique,´ Avenue du Fontaule,´ 66650 Banyuls sur mer, France; bEDF R&D, Departement´ LNHE, 6, quai Watier, 78400 Chatou, France [email protected]
More than a quarter of a century after their discovery, Legionellae continue to pose public health problems and still arouse a great interest within the scientific community. At present, the behavior of this hydrotelluric bacterium in its natural environment is poorly understood. A better understanding of the events which lead to an epidemic could be achieved by determining the sources of Legionellae in the environment and by a better understanding of the interactions between Legionellae and their natural environment. The objective of this work was to better understand the complex ecology of Legionellae, in particular by determining the factors responsible for their occurrence and their persistence in rivers. The difficulties inherent in the detection of Legionellae by traditional culture techniques required the use of alternative methods adapted to the quantification of Legionella pneumophila in surface water and also to the analysis of the genetic diversity of Legionellae species. The innovative development of immunodetection of L. pneumophila cells associated with a viability marker combined with solid phase cytometry, allowed us to quantify potentially pathogenic viable cells in river water and sediment. The fingerprinting technique, CE-SSCP(1), made it possible to study the dynamics of the genetic structure of Legionellae species present in river during an annual cycle. The results of this study carried out on a Mediterranean coastal river, allowed us to confirm the ubiquitous nature of Legionellae, cells being present in significant and recurrent amount from upstream to downstream. The study of L. pneumophila dynamics revealed spatial variations of this species in the river studied, with a concentration gradient from upstream to downstream, and weak seasonal variations, with maxima in summer and minima in winter. The study of the genetic diversity of Legionellae species showed a very diverse genetic composition from upstream to downstream but without any gradient of species along the river. In addition, the impact of anthropogenic discharges, and mainly those of thermal establishments using warm water, was shown on L. pneumophila physiological activity and on the genetic structure of Legionellae populations. (1) CE-SSCP : Capillary Electrophoresis Single Strand Conformation Polymorphism
- 206 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P122 Recovery of Legionella from aerosol sampling
Anthony Pinona, Anne-Marie Jandosa, Nelsie Berthelotb, Alain Vidalb, Franck Laurentc, Sandrine Obertib and Michele` Vialettea aInstitut Pasteur de Lille, 1, rue du Professeur Calmette, 59019 Lille, France; bVeolia Environnement - Centre de recherche sur l’eau, Chemin de la Digue BP 76, 78603 Maisons-Laffitte, France; cCentre d’Analyses Environnementales, 1 Place de Turenne, 94417 Saint-Maurice, France [email protected]
Human contamination by Legionella occurs by inhalation of an aerosol generated from a colonized, generally artificial, hydrous environment. However, official recommendations for Legionella monitoring concern maximum allowable content in water, as there is no recognized standardized method to detect or count Legionella in the air. The aim of this study was to evaluate survival of Legionella in aerosols; air samplers were compared to select a system that permits recovery of Legionella from artificially contaminated aerosols produced in defined conditions. Aerosols were produced from contaminated water with a medical generator in a 75 litres aerosolisation pilot. Samples were collected from the pilot and analysed to evaluate Legionella recovery. In a first step, 3 commercial air samplers were compared: the Coriolis µ (Bertin Technologies, France) and the CIP10M (Arelco, France), which both impact bacteria in a liquid, and the gelose impactor Sampl’air (Laboratoires AES-Chemunex, France). Solid impaction was not found suitable for quantitative evaluation of Legionella recovery. Regarding liquid impaction, the Coriolis µ provided better recovery results and was thus retained for the second step. A complete experimental design was conducted to study the impact of 4 factors on Legionella recovery: strain (reference or epidemic L. pneumophila, L. anisa), concentration in aerosolised water, relative humidity in the pilot, and time between aerosolisation and sampling. Samples were analysed to determine numbers of culturable bacteria, viable bacteria (culturable and non-culturable), and total bacteria (live and dead cells). It was thus possible to estimate a ”physical” loss of bacteria during the aerosolisation / air sampling process, a loss of viability, and a loss of culturability of Legionella. About 10% of all aerosolised bacteria were collected by the sampler. Results from this study indicate that Legionella exhibit a moderate loss of viability when aerosolised and sampled from air, whereas a high loss of cultivability may be obtained, thus resulting in a significant part of viable bacteria not being detected by the culture method. Other techniques, especially targeting viable bacteria, should thus be preferred.
- 207 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P123 PCR-quantification of viable Legionella in water samples: development of a propidium monoazide pretreatment of bacteria directly applied on membrane filters
Sami Slimania, Christophe Ginevrab, Maelle Molmeretb, Eric Dusserreb,Celine´ Mazurec, Monique Reyrollea, Jer´ omeˆ Etienneb and Sophie Jarraudb aCentre National de ref´ erence´ des Legionella, Centre de Biologie Est, Groupe Hospitalier Est, 59 Boulevard Pinel, 69677 Bron, France; bCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; cBioRad, 3 avenue Raymond Poincare,´ 92430 Marnes la Coquette, France [email protected]
Background: Recently, Ethidium and Propidium MonoAzide (E/PMA) has been proposed as a way to reduce PCR signal originated from dead bacteria by selectively entering damaged cells and by blocking PCR amplification of bacterial DNA via photoactivation. Purpose: The objective of this study is to develop a PMA pretreatment of Legionella water samples, allowing the PCR- based quantification of solely viable Legionella, which could be applied directly on membrane filters. Methods: Water samples inoculated with 106 bacteria/mL of viable or heat-killed Legionella were used to define the conditions of PMA pretreatment. Different concentrations of spiked- viable Legionella as well as more complex water samples spiked with different concentrations of both viable and heat-killed Legionella with and without Pseudomonas aeruginosa (mimicking aqueous interfering flora) were also studied. All samples were then quantified using iQ-CheckTM Quanti L. pneumophila Kits (Bio-Rad). Results: Optimal PMA pretreatment of samples containing 106 bacteria/mL involved a serie of 2 incubations of 5 minutes each with 6 µM of PMA in the obscurity followed by 5 minutes of light exposition on ice. The inhibitory effect of PMA pretreatment on Legionella quantification was -0.23 log +/- 0.08 and -3.86 log +/- 0.29, on viable and heat- killed Legionella samples, respectively. Over-estimation of viable Legionella occurred at 108 bacteria/ml of heat-killed Legionella. Interfering flora didn’t have any effect in the study conditions. Conclusion: PMA pretreatment applied directly on membrane filters followed by real-time PCR allows rapid and reliable quantification of viable Legionella in laboratory-controlled conditions.
- 208 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P124 Correlating numerical tools and experimental measurements for airborne microorgan- isms concentrations : application to the spreading of Legionella pneumophila from cooling towers.
Eric Tarnauda, Thi-Lan Hab, Enric Robineb, Frederic Togneta, Cyrille Turmeauc, Yannick Morelc and Laurence Rouila aINERIS, Parc Alata BP 2, 60550 Verneuil en Halatte, France; bCentre Scientifique et Technique du Batiment,ˆ 84 avenue Jean Jaures,` 77447 Marne-la-Vallee,´ France; cCentre d’Etude du Bouchet (DGA), BP 3, 91710 Vert-le-Petit, France [email protected]
Keywords: bioaerosols, atmospheric dispersion modelling, air sampling, Legionella pneumophila, cooling tower. Legionella pneumophila is a causative agent of respiratory illness in humans. Most of community outbreaks of legionellosis have been linked with an airborne transmission of pathogen from cooling towers. During the outbreak in Pas de Calais, France, 2003, it was observed that L. pneumophila could be transported in air at least 6 km from the source (Nguyen et al., 2006). To support investigations during epidemics, dispersion models could be used to optimize air sampling strategy for detecting airborne L. pneumophila and evaluate how far contaminated aerosols would be spread from cooling towers. The objective of this research is to investigate the performance of available plume models for predicting concentrations of airborne microbes, by iterative model-to-experimental data comparisons. A dispersion field campagn of non-hazardous biological aerosols was performed at CSTB during August 25 to 29, 2008. Spores of Bacillus globigii were disseminated from the roof of a building on site. Air was sampled at 5 various locations from the source. The sampling sites were selected from preliminary numerical simulations with the ADMS model developed by CERC (Cambridge Environnemental Research Consultants - www.cerc.co.uk),a Gaussian dispersion model using current understanding of the structure of the atmospheric boundary layer. To simplify calculations, biological aerosols paths were estimated by assuming the bioaerosol as an inert stable particle during the transport. Atmospheric dispersions of aerosols were computed following local daily weather forecasts (temperature, humidity, wind speed and direction). Streamlined buildings surrounding the emission source were also included to evaluate the experiments location . Slit and six-stage Andersen samplers were used for collecting the airborne microorganisms directly onto TSA agar. Wetted-wall cyclones and SKC Biosamplers were also run for sampling air into distilled water. Liquid samples obtained that way were then plated on TSA agar. Typical colonies of B. globigii were enumerated after 16-20 hours at 37◦C. The results showed a correlation between predictions and in situ measurements. Spores of B. globigii were detected at all sampling locations selected following simulations with the ADMS. These first observations suggest the potential operational use of models in case of epidemics of legionellosis or all other threat linked to biological aerosols dispersion. Further works including numerical simulation with the 3D lagrangian dispersion model MSS (Micro Swift Spray) developed by Aria Technologies (www.aria.fr), will be carried out. This model account for real topography and buildings. Simulation results, in term spores numbers at sampling site will be compared to the sampling observations. This work was partially supported by AFSSET. Nguyen, T. M. N., Ilef, D., Jarraud, S., Rouil, L., Campese, C., Che, D., Haeghebaert, S., Ganiayre, F., Marcel, F., Etienne, J. and & Desenclos, J.-C. (2006). The Journal of Infectious Diseases, 193, 102-111.
- 209 - Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P125 Susceptibility of extra- and intracellular Legionellae to silver and a method to detect the concentration of microbicidal active silver in water
Catharina Unger and Paul Ch. Luck
TU Dresden, Institute Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany catharina [email protected]
The antimicrobial effect of silver was tested in different circulating water systems. In some of these field applications a reduction of the TVC and the Legionella-count was detected, but in some others not, although there was an adequacy concentration of silver detected by AAS (atom absorption spectrometry). To clarify these we compared the antimicrobial effect of silver against intracellular grown (in Acanthamoeba) and agar- cultivated Legionella pneumophila. In comparison with extracellular Legionella the intracellular grown Legionella were less susceptible against silver. It seems that the Legionella in the tested water systems are intracellular-grown in occurring amoeba. To get an idea how many of the detected silver is microbicidal active and not bound to organic material or in complexes (e.g. silver-thiosulfate or silver-rhodanide) a specific bio-assay that detects active silver was developed. Therefore different indicator-strains (Staphylococcus aureus, Pseudomonas aeruginosa and Legionella pneumophila) were tested with different silver-concentrations in Aqua dest. The reducing effect was allegorized graphically and the regressions were determined. Following the concentration of microbicidal active silver in a water sample can be calculated. Our results show, that there might be a discrepancy between the concentration of silver, determined with the AAS and the concentration of biocidal silver determined in the bio-assay. The established bio-assay is a suitable procedure to assign the concentration of silver in water systems considering the different aspects like complexation and fixation of silver ions to organic matrix.
- 210 -
SESSION 8 LEGIONELLA PREVENTION AND CONTROL
- 211 - Session 8: Legionella prevention and control P126
Quality and quantity detection of Legionella in environmental water samples
Yulia S. Alyapkinaa, Igor S. Tartakovskyb, Tatiana I. Karpovac, Oksana V. Sadretdinovac, Jakov Alexeeva and Irina V. Novokshonovad aJSC ’Syntol’, Timirayzevskaya st., 42, 127550 Moscow, Russian Federation; bGamaleya Institute for Epidemiology and Microbiology RAMS, Gamaleya str., 18, 123098 Moscow, Russian Federation; cN.F. Gamaleya Institute for Epidemiology and Microbiology RAMS, Gamaleya str., 18, 123098 Moscow, Russian Federation; dFederal service for surveillance for protection of consumers rights and human welfare of Russian Federation, Varshavskoe shosse, 19 a, 117105 Moscow, Russian Federation [email protected]
Regular quantitative monitoring of Legionella in the potential risk water environments and prevention of high concentration accumulation is real way to prevention of legionellosis. We developed a quantitative real-time PCR kit for simultaneous detection of Legionella pneumophila and Legionella spp. DNAs in environment samples. It is based on the determination of Legionella pneumophila mip and Legionella spp. 16SrRNA gene specific targets and internal positive control target that allows to except false-negative results. Besides we developed aquatic environment samples DNA extraction kit that allows to concentrate big volume samples and carry out of DNA extraction with maximum purity and yield. Elaborated real-time PCR kit and DNA extraction kit were employed for quantity detection of L.pneumophila and Legionella spp. during inspection of water samples from cooling towers , central air condition system and autonomous water supply system of industrial sets. More than 400 of water and biofilm samples obtained from different regions of Russia were analysed. L.pneumophila was found in 27% samples, Legionella spp. were found in 56% samples.Real- time PCR results coincided with cultural method results. Modification of kit for simultaneous monitoring of Legionella pneumophila , Legionella spp. and Pseudomonas aeruginosa in hospital water samples was developed. Both modification of kit were recommended by Federal service for surveillance for protection of consumers rights and human welfare of Russian Federation for rapid quantification of Legionella in water samples as alternative tool.
- 212 - Session 8: Legionella prevention and control P127
Rapid and accurate enumeration of active Legionella pneumophila in aquatic environments by microcolony method
Takashi Baba, Naoko Inoue, Nobuyasu Yamaguchi and Masao Nasu
Graduate School of Pharmaceutical Sciences, Osaka University, 1-6, Yamada-oka, Suita, 565-0871 Osaka, Japan [email protected]
Background: Rapid methods to enumerate active Legionella pneumophila in aquatic environments are essential in order to monitor water quality and to prevent of the outbreaks of Legionella infections. Conventional plate counting requires 7-10 days to obtain reliable results. PCR-based method is sensitive and widely used, while the method detects both dead and active L. pneumophila cells. In this study, microcolony technique , which detect bacterial microcolonies (10-50 µm) at early stage of their proliferation, was examined for the specific detection of active L. pneumophila. Methods: L. pneumophila serogroup 1 (JCM7571) and water samples were used for this study. Cooling tower, hot spring and bath water samples were collected at Osaka and Hyogo, Japan. Bacterial cells in each sample were trapped on filters and they were incubated on selective medium for Legionella spp. (WYOα) at 37◦C. For specific detection and enumeration of L. pneumophila, the microcolonies were stained with anti-L. pneumophila SG1-14 fluorescent antibody. Results: This method allowed rapid detection (within 48 hours) of active L. pneumophila in aquatic environments and the results were almost equal to those determined by the conventional culture methods. Microcolonies were formed in some samples where no colony-forming cells were detected, that is, certain population of L. pneumophila cells in aquatic environments can form microcolonies but not visible colonies. Conclusions: Microcolony method combined with fluorescent antibody staining allowed rapid and accurate enumeration of active L. pneumophila within 48 hours incubation. This method may contribute to reduce the risk of the outbreaks of Legionella infections. This study was in part supported by the JSPS Grant-in aid for Scientific Research (A) (20249007).
- 213 - Session 8: Legionella prevention and control P128
Prevalence of Legionella in different aquatic environments in Hungary
Zsofia´ Barnaa, Csaba Bognar´ b, Judit Krisztina Horvath´ b, Mihaly´ Kad´ ar´ a, Anita Szaxa and Marta´ Varghaa aNational Institute for Environmental Health, Gyali´ ut´ 2-6, H-1097 Budapest, Hungary; bNational Institute for Epidemiology, Gyali´ ut´ 2-6, H-1097 Budapest, Hungary barna.zsofi[email protected]
Two years after the implementation of the EGWLI Guidelines in Hungary the legislation of environmental Legionella monitoring is still incomplete. There is no regulation for cooling towers or hot water systems, only the monitoring of hot whirlpool spas is mandatory. To emphasize the need for a more extensive regulation, the Legionella prevalence in different man-made aquatic environments was investigated by standard culture methods in the 2006-2009. The Legionella isolates were typed using monovalent sera, rep-PCR and PFGE. The majority of the investigated water distribution systems (102 objects) was colonized by Legionella (66 %), 44 % of the samples were positive, 25 % had Legionella counts above 1000 CFU/L. Legionella titers up to 106 CFU/L were observed. In the cooling towers and spa pools (21 and 19 objects, respectively), the prevalence of Legionella was lower than expected (<15 %), but once colonized, concentrations of Legionella were high (up to 106 CFU/L). All isolates were identified as L. pneumophila. The serotypes 1, 6 and 3 were most frequently isolated (27,5 %, 17,7 % and 14,0 %). The exact serotype could not be determined for 21,3 % of the strains. The extent of Legionella prevalence in the investigated hotels and hospitals poses serious risk of infection. The rate of colonization was above 75 %; the counts frequently exceeded the 1000 CFU/L guide value (up to 4 x 105) However, the number of reported nosocomial or travel associated Legionella cases is low (< 1/ million). Mainly L. pneumophila 2-14 serotypes were isolated from these facilities, except for those associated with the legionellosis cases which were colonized by high counts of L. pneumophila 1. The incorrect water management practices may be responsible for the high prevalence of Legionella, the water temperature was between 20 and 55 ◦C for 75,2 % of the samples. The results support the necessity of an environmental Legionella monitoring scheme and regulation in Hungary. We aim at a national legislation that would prevent rather than follow the first major outbreak.
- 214 - Session 8: Legionella prevention and control P129
Development and certification of a standard reference material (SRM) for Legionella detection and quantification by quantitative PCR in environmental samples.
Maud Baumea, Laurent Garrellyb, Philippe Guarinic,Hel´ ene` Hostc,Hel´ ene` Frenkield,Sebastien´ Boutone, Pierre- Olivier Fraissef, Monique Reyrolleg and Sophie Jarraudh aCentre National de Ref´ erence´ des Legionella, Hospices Civils de Lyon, Groupement Hospitalier Est, 59 boulevard Pinel, 69677 Bron, France; bGL Biocontrol, Parc Scientifique et technique Georges Besse, 30035 Nˆımes, France; cA.G.L.A.E., 629 Av. de la Republique,´ 59000 Lille, France; dBioRad, 3 avenue Raymond Poincare,´ 92430 Marnes la Coquette, France; ePall-GeneSystems, 1, rue du Courtil - Centre CICEA, 35170 Bruz, France; fLNE, 1, rue Gaston Boissier, 75724 Paris, France; gCentre National de ref´ erence´ des Legionella, Centre de Biologie Est, Groupe Hospitalier Est, 59 Boulevard Pinel, 69677 Bron, France; hCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France [email protected]
Background The Legionella spp and Legionella pneumophila Real Time PCR techniques have been used for many years in France (French Standard XP T90-471) and also in Europe to quantify Legionella in waters. They are reliable methods, much faster than the classic culture method, that enable detection of contaminated waters, real time monitoring and assessment of Legionella burden in risky installations and rapid outbreak investigations. Yet, variability in testing has been observed in interlaboratory studies which can be explained by the lack of a common standard (data not published). The aim of this work was to set a primary measurement standard to reduce the quantification variability. Choice of the DNA In 2007 and 2008 studies have been conducted to evaluate a number of candidate Legionella DNA standards. The candidates needed to fit several criteria regarding DNA (genomic DNA from L. pneumophila ATCC 33152) and its stability, batch homogeneity, and commutability to assure it could be used by every laboratory. Three candidates were chosen that were submitted to an interlaboratory assay to decide between them. Certification of the chosen DNA In 2009 eleven French laboratories contributed to an interlaboratory assessment to determine precisely the quantity of DNA contained in the chosen standard. The measurements have been performed by limiting dilution assay (LDA), and a statistical analysis of the results then allowed assigning a reference value and a confidence interval to the SRM. Use of the SRM This standard will be used by the PCR kits manufacturers to connect their measurement standards and by the laboratories using PCR, as described in the French standard NF T90-471 (currently under revision). The interest here is in harmonization of the commercially available materials as well as providing a standard that can be used to validate in-house produced standards. In this way, the SRM would contribute to reducing variability in quantification from one laboratory to the next and should make comparison between results from different laboratories easier and more accurate.
- 215 - Session 8: Legionella prevention and control P130
Impact of oxidizing disinfectants against co-culture of Legionella pneumophila and Acanthamoebae
Mathieu Dupuya, Stephane´ Mazouaa, Florence Bernea, Nathalie Garrecb, Pascaline Herbelinc, Sandrine Obertid, Marie- Hel´ ene` Rodiera, Sylvie Soreauc, France Wallete, Yann Hecharda and Nelsie Berthelotd aUniversity of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France; bVeolia Environnement - Centre d’Analyses Environnementales, 1 Place Turenne, 94417 Saint Maurice cedex, France; cEDF R&D, Departement´ LNHE, 6, quai Watier, 78400 Chatou, France; dVeolia Environnement - Centre de recherche sur l’eau, Chemin de la Digue BP 76, 78603 Maisons-Laffitte, France; eEDF - Service des Etudes Medicales,´ 22-28 rue Joubert, 75009 Paris, France [email protected]
Free-living amoebae might be pathogenic by themselves and it is now clearly established that they are a reservoir for bacterial pathogens, such as Legionella pneumophila. Not only the amoebae could protect the intracellular bacteria but they also enhance their virulence. Therefore, it is important to study the efficiency of treatments commonly used in cooling towers against Legionella. In this aim, we compared the effect of various oxidizing treatments (Cl2, ClO2, NH2Cl) against: (i) free Legionella pneumophila, (ii) Acanthamoebae (trophozoite), (iii) Acanthamoebae (trophozoite) infected by L. pneumophila, (iv) Acanthamoebae (cyst) and (v) vacuole infected by L. pneumophila. All the samples were treated for 1h and the biocide concentration was followed to calculate the Ct99,9%. Results indicate that (i) all disinfectant were efficient in our conditions, (ii) the most efficient was ClO2, (iii) there was no difference of susceptibility between infected or non-infected trophozoite Acanthamoebae (iv) most treatments were efficient on L. pneumophila but some treatments were efficient at the same level towards extracellular or intracellular bacteria (i.e. NH2Cl) while others are less efficient on intracellular bacteria (i.e. Cl2 and ClO2), (v) the dose needed to achieve a reduction of 3 log was 10 times higher for cyst than for trophozoite. In conclusion, Acanthamoebae could be inactivated by oxidizing disinfectants if the right dose is used. Under classical doses used in cooling towers trophozoite Acanthamoebae and intra-cellular Legionella could be successfully treat. Cyst Acanthamoebae could be inactivated under stronger treatment conditions.
- 216 - Session 8: Legionella prevention and control P131
Legionella growth and thermal treatment in copper domestic hot water network
Fabienne David, Alain Vidal, Nelsie Berthelot, Armelle Freval and Sandrine Oberti
Veolia Environnement - Centre de recherche sur l’eau, Chemin de la Digue BP 76, 78603 Maisons-Laffitte, France [email protected]
In order to define the materials the most adapted in domestic hot water, it’s essential o to investigate their colonization abilities by microorganisms as Legionella and o to evaluate impact and efficiency of the disinfection method. Several materials (copper, galvanized steel, stainless steel...) were tested in hot water network pilot unit. Legionella was grown during recirculation of water at 40◦C with similar hydraulic condition of real network. The results demonstrated that copper and galvanized steel are a similar colonization of Legionella in water. In biofilm, copper temporarily limit the growth of Legionella. 71 weeks are necessary to observe Legionella in biofilm at the same level of galvanized steel. Different treatments are commonly used for the eradication of Legionella in hot water distribution networks (continuous chlorination,) but appear to be inadequate for eradicating the bacteria in the biofilm. Hot water flushing (70◦C - 30 min) seems to be efficient in biofilm for the majority material but this treatment is punctual. In steel pipe, when this treatment followed by water recirculation at 40◦C, the Legionella re-growth at the same level after 1 week in water and 4 weeks in biofilm was observed. This comportment is different for copper. At the same condition, an effective and long-lasting reduction of Legionella in the biofilm (3 log) was observed. In the water, Legionella was detected one week after the treatment but the concentration stay lower than the quantification level (< 250 UFC/L) during 24 weeks. In biofilm, Legionella start to reappear after 16 weeks but punctually. A trial in real copper network was carried out and validates the pilot result. An immediately reduction of Legionella in water was observed (3 Log) and was maintained during 10 months in spite of a low water temperature (37◦C). This thermal treatment didn’t often use because of technical constraint. Nevertheless, in copper pipe it allows contributing at the control of biofilm contamination. In conclusion, the material the most adapted to prevent Legionella contamination in sanitary hot water network seems to be copper. If no material can guarantee the total Legionella absence of water, copper could to limit the risks.
- 217 - Session 8: Legionella prevention and control P132
Correlation between early diagnosis, clinical expiry and respiratory insufficiency in patients with pneumonia caused by Legionella
Roberto Bettellia, Raffaele Russoa, Giuseppe Russoa, Vincenzo Castaldoa, Mario Maglioccaa, Marilena Stancoa, Giampaolo Buonopanea and Anna Maria Rossib aA.O. S.G. Moscati, Contrada Amoretta, 83100 Avellino, Italy; bAgenzia Regionale per la Protezione Ambientale della Campania, Via Scavate Case Rosse, 84100 84100, Italy [email protected]
Introduction: : in the Campania Region of Italy the following cases of legionellosis were notified: 4 cases in the years 2001/2003 - 5 cases in 2004 - 38 cases in 2005 - 31 cases in 2006 - 48 cases 2007 - 57 cases in 2008. Results: during the observation period considered , of a total of 131 cases diagnosed of pneumonia 16 cases were from legionella, 15 cases showed favorable outcome and one exitus. Of these, 14 cases were men and women; the median age is 57.3 years for men and 56 years for women. The onset of symptoms for 16 patients was distributed as follows : 16 patients had hyperpyrexia, 6 patients had cough with or without production of secretion, 4 patients had asthenia and diarrhea, 3 patients had dyspnea and articular pains, 2 patients had temporo-spatial disorientation and vomiting, 1 patient had cefalea . At entry time 5 patients had a SpO2> 90%, at discharge time 11 patients had a SpO2> 90%. The 9: in the Campania Region of Italy the following cases of legionellosis were notified: 4 cases in the years 2001/2003 - 5 cases in the year 2004 - 38 cases in the year 2005 31 cases in the year 2006 - 48 cases in the year 2007 - 57 cases in the year 2008. of patients were treated with OLT, their average age was 60.7 years, 3 patients were treated with VAM + IOT and one patient was treated with NIV (non-invasive ventilation) . At discharge 2 patients at home were needed to OLT and both patients were suffering from BPCO. The average length of hospital stay was 15.3 days, the 4 patients associated with forms of severe respiratory failure requiring IOT + VAM, or NIV, had an average duration of hospital stay of 42.5 days . The 12 patients with good respiratory compensation who have not made OLT or who have made O2 therapy alone in first / second day, had an average length of stay in hospital of 6.3 days. Of the 16 total cases diagnosed, 13 cases had the diagnosis on the first day of hospitalization. 196 enviromental samples were collected by CRL with 24% by >103 CFU/L Lp1 Conclusions: The patients whose clinical course was characterized by severe hypoxemic respiratory insufficiency or hypoxemia-hypercapnia, and for that reason they have been subjected to different methods of invasive mechanical ventilation (IOT + VAM) or not invasive (NIV), they were associated with a very long average stay and a complicated clinical course with a case of exitus. On the contrary, the patients with a good respiratory compensation have a clinical course and optimal lengths of stay in hospital substantially contained. The data show that the early diagnosis is associated with a lower incidence of severe respiratory insufficiency.and that the application of preventive measures in a hospital is of fundamental importance for the control of legionella infections. In fact, up to now, no cases of nosocomial legionellosis has registered.
- 218 - Session 8: Legionella prevention and control P133
Source or coincidence? Factors contributing to source identification of Legionella infections
Sacha Bleekera, Ed Yzermana, Sjoerd Eusera, Jacob Bruina, Nico Nagelkerkeb and Jeroen Den Boera aRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands; bUnited Arab Emirates University, Faculty of Medicine and Health Sciences, Department of Community Medicine, 17666 AL AIN, United Arab Emirates [email protected]
Background: Sources of infection for Legionnaires’ disease (LD) patients are hard to identify. However, outbreaks are usually preceded by seemingly unrelated sporadic cases. The question is whether the relation of a newly reported case to a registered historical potential source is based on coincidence, or indicates an outbreak in an early stage which urges for immediate action. Objective: The overall objective of the study is to sort out indicators for the identification of sources of LD. The ultimate aim is to develop and implement an evidence based guideline for Municipal Health Services (MHSs) that quantifies the probability for each potential source to be the source of infection with Legionella bacteria. For this purpose, (a) the locomotive behaviour of LD patients will be compared to that of matched controls; and (b) the probability of detected geographical clusters based on coincidence will be calculated. Methods: Patients. We used a matched case-control study design, for which the patients were extracted from an existing database that is the result of an ongoing national systematic data collection on LD infections in the Netherlands. This database includes characteristics of LD patients, addresses of potential sources visited by LD patients during their incubation period, and laboratory data (serotyping and genotyping) on Legionella isolates cultured from patient materials or water samples from potential sources. Patients were eligible for the current study if the following criteria were met: (1) at least 5 days from the 2-10 day incubation period were resided in the Netherlands and (2) during the incubation period a potential source was visited which was linked to another LD patient in the past (’source clustered’) or the LD patient lives within an radius of 1 km distance of an other LD patient with a maximum of difference in disease onset of 6 months (’geographically clustered’). Controls. For each eligible patient controls (preferably 5) were drawn from Municipal Population Registries. Cases and controls were matched on age, gender, zip code and season of disease onset of the cases. Data Collection. Data were collected by questionnaires (general characteristics, health, use of alcohol, tobacco and medication, locomotive behaviour during 14 days). Additionally, cases and controls from the geographically clustered set indicated the locations visited during 14 days on a tailor-made map containing their community in 2.5 square kilometer lattice. Data Analysis. The case control data will be analysed in a matched pair case control analysis. Demographic data from Statistics Netherlands Registers will be used to determine the probability of 2 unrelated patients to live in the same postal area code. Results: In total 544 subjects are enrolled: 148 cases (87 ’source clustered’ and 57 ’geographically clustered’) and 396 controls. At present, further results are not available yet, but will be presented at the Legionella 2009 conference.
- 219 - Session 8: Legionella prevention and control P134
An Environmental Investigations in a Response to the Notification of Travel- associated Legionnaires’ Disease from European Working Group for Legionella Infection Network and World Health Organization, Thailand, 2006-2007
Rome Buathong, Wanna Hanshoawarakul, Sopon Iamsirithaworn and Kumnuan Ungchusak
Bureau of Epidemiology, Depart ment of Disease Control, Ministry of Public Health, 11000 Nonthaburi, Thailand [email protected]
Background: During 2006-2007, the Bureau of Epidemiology (BOE) was notified four events of travel-associated Legionnaires’ Disease (LD). Nine patients were confirmed by urine antigen detection. The environmental assessments were conducted in four suspected hotels to determine source of infection. Methods: We interviewed hotel owners and concerned staff. The water and cooling systems were reviewed. Water’s temperatures and residual chlorine concentration were measured. All suspected sources of infections were collected for bacterial culture. Biological studies for Legionella pathogen were tested. Results: There were nine cases aged 24 - 68 years. All of them were confirmed L. pneumophila serogroup1 with 8 cases were European and other one was Australian. All four hotels were located on Andaman coast and provide services for 7-22 years. One hotel used a cooling tower, the other used split-type air conditioners. Hot water’s temperature was below optimum (50 oC) and residual chlorine of cold water was substandard found in bathroom of hotels. City tap water was supplemental by untreated water brought from natural reservoirs at the time of water supply shortage in the hotels. Stagnation of water in the pipeline during temporary closure for repair after tsunami range from 1- 6 months promoted overgrowth of Legionella spp. in the water system. The positive yields of Legionella spp in environmental samples were 34 % (38/155). Positive samples were water and swabs from bath pillar tap (40%:17/43), shower nozzles (29%:17/59), water storage tanks (20%3/15) and trays under air-conditioner (10%: 2/20). No specimens collected from cooling tower was positive (0/6) Among 42 Legionella spp. isolates were subsequently identified as L. pneumophila serogroup1 (52%), serogroup6 (10%) and L. bozemanii(17%). Conclusions: Legionella were identified more in water supply than cooling system. The stagnation of water in pipes and utensils would promote the growth of Legionella. Hyperchlorination and thoroughly cleaned of water system were conducted. Follow up bacterial culture were done to ensure the safety of the hotels. Hotels in tourist area must be motivated to participate in the voluntary Legionella control program of the Department of Health.
- 220 - Session 8: Legionella prevention and control P135
Progress in the surveillance and control of Legionella infection in France, 1998-2008
Christine Campesea, Catherine Mainea, Sophie Jarraudb and Didier Chea aInstitut de Veille Sanitaire, 12 rue du Val d’Osne, 94415 Saint Maurice, France; bCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France [email protected]
In France, the notification of Legionnaire’s disease (LD) is mandatory since 1987. Following a study showing an important under-reporting of the disease, the surveillance system was strengthened in 1997. In parallel, the urinary antigen detection test was introduced and a guideline for prevention and control of the disease was implemented. Thereafter, the sensitivity of the surveillance system increased (10% in 1995 to 33% in 1998). From 1998 to 2008, a total of 11 147 cases were reported. Between 1998 and 2005 the incidence of LD increased gradually, by an average of 20% per year, reaching 2.5/105 in 2005 and then slightly decreased (2.0/105 in 2008). The delay between dates of onset and notification has decreased over the period from 28 days in 1998 to less than 7 days in the recent years. During this 10-year period, the majority of cases was diagnosed by urinary antigen test, and an isolate was available for 18% of cases. The median age of cases was 60 years [range 0-103], the M/F sex ratio was 2.9 and the case fatality rate was 13%. An exposure during travel was suspected for 17% of cases. Nine percent were considered as potentially hospital-acquired infections, this percentage decreased over the 10-year period from 21% in 1998 to 7% in 2008. Over this period, 14 outbreaks occurred (outbreak: more than 10 cases) with 380 cases notified. The median number of cases was 22 [range 11-86] and more than 40 cases were reported in only, one outbreak (Lens, 2003-2004). A single outbreak was identified in 2007 and none in 2008. Cooling towers were the most probable source of infection for 13 outbreaks (confirmed source: 8 outbreaks; suspected source: 5 outbreaks). The surveillance and control systems for LD in France continue to evolve on a regular basis. In recent years, several regulations concerning prevention and control were implemented in hospital settings and home for elderly people. At the end of 2004, a new regulation about the cooling towers was introduced to better control the dispersion of Legionella from these sources. The 1997 guideline for prevention and control was updated in 2005. All these measures have contributed to strengthen the French surveillance system and to improve our ability to prevent, detect and control LD.
- 221 - Session 8: Legionella prevention and control P136
Legionella Concentrations In Hotel Hot Water Systems Not Associated With Cases Of Legionnaire’s Disease: a Power Law Distribution?
Sebastian Crespia and Juan Ferrerb aPoliclinica Miramar, Camino La Vileta 30, 7011 Palma de Mallorca, Spain; bAsociacion´ Empresarial Hostelera de Benidorm y Costa Blanca, Via Emilio Ortuno,˜ 5-1a, 3501 Benidorm, Spain [email protected]
BACKGROUND Maintaining high temperatures (50-60 degrees centigrade) in hot water systems is widely recommended to prevent legionellosis in hotels. In turn, Legionella analyses on hot water systems are used to assess the effectiveness of established preventive measures, but there are large differences between the different regulations regarding tolerance limits, varying between maximums of 100 FCU/L and 10000 FCU/L accordingly. Uncovering the distribution patterns for Legionella concentrations in hotel drinking water systems not associated with cases of legionnaire’s disease and using preventive programmes may help to better define tolerability limits and improve preventive policies. METHODS Retrospective analysis has been carried out of the results from 1,435 Legionella analyses obtained from two different laboratories over three consecutive years in hotel hot water samples from two Spanish tourist regions -Valencia and the Balearic Islands - not associated to legionellosis cases and where maintaining temperature at 50-60 degrees centigrade was obligatory. All samples included in the final analysis complied with the same inclusion criteria set in advance and have been processed in each case as per the ISO 11731 method. The results obtained at each laboratory were grouped in logarithm categories, in accordance with Legionella concentrations obtained, and histograms of the frequencies tabulated for each category were obtained. RESULTS The distribution pattern for Legionella concentrations in both analysed tourist regions was very similar. The frequencies for the category with results below 100 FCU/L in both cases were near 80%. The frequencies for the remaining categories followed a weakly decreasing slope as concentrations increased, the final distribution being similar to a power law (80-20 rule). In both analysed regions, over 12% of samples showed values above 1000 FCU/L and just over 4% of samples showed concentrations above 10000 FCU/L. CONCLUSIONS AND DISCUSSION The hotels not associated with legionellosis cases and with preventive systems based on maintaining high temperatures in hot water frequently showed concentrations above 1000 FCU/L. This finding brings up doubts as to the suitability of the tolerable ranges often recommended in prevention regulations. Further, the similarity of legionella population distribution in two different regions, and obtained independently by two laboratories, suggests that we may be facing a general pattern. If later research confirms that we are facing a ’power law distribution’ it would be complicated to establish objective ’normal’ ranges for hot water systems kept at a constant temperature of 50-60 degrees centigrade.
- 222 - Session 8: Legionella prevention and control P137
The use of power ultrasound as an effective pretreatment step during hypochlorite disinfection of Legionella pneumophila-contaminated drinking water systems
Priscilla Declerck, Louise Vanysacker and Frans Ollevier
Laboratoy of Aquatic Ecology and Evolutionary Biology, Zoological Institute, Katholieke Universiteit Leuven, Charles Deberiotstraat 32, 3000 Leuven, Belgium [email protected]
Microorganisms commonly occur in well organized communities called biofilms. Such communities provide favourable micro-environments to pathogens like Legionella pneumophila, known as the causative agent of Legionnaires’ disease. Legionnaires’ disease is generally considered a preventable illness, as efficient treatment and control of Legionella concentrations in the water concerned, will lead to a significant decrease in disease cases. The aim of this study was to evaluate the use of power ultrasound as a valuable pretreatment step in the hypochlorite disinfection process of L. pneumophila-contaminated drinking water systems. Experiments were performed using four replicate rotating biofilm reactors fed with filter-sterilized (0.2 µm) dechlorinated tap water. Each reactor consisted of two concentric cylinders. The inner cylinder (V=1 L) contained a rotor with twenty removable clear polyvinylchloride slides mounted in parallel to the reactor hydrodynamic conditions. By using a Hazen-Williams coefficient of 120, a water distribution pipe consisting of galvanized iron was simulated. At the start of each experiment (day 0), reactors were inoculated with the following non- Legionella consortium: Escherichia coli, Aeromonas hydrophila, Flavobacterium breve and Pseudomonas aeruginosa. L. pneumophila serogroup 1 was inoculated in each reactor at day 5. Ultrasound pretreatments were performed using a DG-100 probe Disintegrator at a fixed operating frequency of 36 kHz and a specific energy of 343 kJ L-1 for 0 (control), 0.5, 1 and 5 min. Following ultrasound pretreatment, sodium hypochlorite was added in a final concentration of 0.1, 1 or 5 ppm. Hereafter, biofilm samples were withdrawn from each reactor at selected time points of 1, 5, 10, 15 and 30 min. The effect of ultrasound and hypochlorite treatments on the viability of both non-Legionella and L. pneumophila was evaluated using viable plate counts. At the moment analyses are still ongoing. Results will be presented and discussed at the conference.
- 223 - Session 8: Legionella prevention and control P138 v-PCR : an innovative method for quantification of viable Legionella in water samples
Pilar Delgado-Viscogliosia, Matthieu Bernierb, Lydie Solignaca and Jean-Marie Delattrea aInstitut Pasteur de Lille, 1, rue du Professeur Calmette, 59019 Lille, France; bIPL sante´ environnement durables Midi- Pyren´ ees,´ Rue du cheneˆ vert, 31682 LABEGE, France [email protected]
The evaluation of the risk associated with Legionella has traditionally been performed using culture based-methods. However, culture requires long incubation times and results can be under-estimated. Quantitative PCR (qPCR) method can mitigate the main drawbacks of culture. Actually, qPCR results are obtained in several hours instead of days and VBNC Legionella can also be detected. However PCR fails to discriminate between live and dead bacteria due to the persistence of DNA in cells after death, potentially leading to false-positive results and overestimation of the infection risks. The aim of the study was to improve qPCR by developing an assay (v-PCR, for viability PCR) to monitor viable Legionella concentrations in water. v-PCR combines 1) a pre-treatment of samples with an intercalating agent followed by 2) conventional qPCR. Ethidium monoazide bromide (EMA) has been used. The ability of v-PCR to differentiate viable from nonviable L. pneumophila was confirmed with permeabilizing agents such as heat, toluene or isopropanol. v-PCR suppressed more than 99.9% of the L. pneumophila signal in non viable cultures and was able to discriminate viable cells in mixed samples. v-PCR operated in the same way using other Legionella spp. A total of 64 domestic hot-water samples were examined for the presence of L. pneumophila by v-PCR, conventional qPCR and culture. The recovery rate of the v-PCR was equal to or higher than the culture method and lower than or equal to the conventional qPCR. v-PCR was used to successfully monitor in vitro the disinfection efficacy of heating to 70◦C, glutaraldehyde and chlorine curative treatments. Field studies including routine monitoring of Legionella in a representative number of cooling towers showed weak differences between viable (v-PCR) and total (qPCR) counts, suggesting that most Legionella cells are alive cells in the studied facilities. The v-PCR method appears to be a promising and rapid technique for enumerating Legionella in water, that offers an added value (viability notion) to the Legionella monitoring by conventional qPCR. Moreover, v-PCR unlike culture method, is suitable for rapid response in emergency situations. v-PCR-treated samples should be more representative of the potentially pathogenic Legionella populations than those evaluated by qPCR
- 224 - Session 8: Legionella prevention and control P139
The Water Safety Plan Applied for the Control of Legionella Infections in an Italian Hospital
Oscar Di Marinoa, Aldo Maria Capraa, Lorenza Camponovoa, Stefano Meladab, Francesco Garusib and Gian- franco Garusib aAzienda Ospedaliera San Gerardo di Monza, Via Pergolesi, 33, 20052 Monza, Italy; bSanipur srl, Via S. Quasimodo, 25, 25020 Flero, Italy [email protected]
The San Gerardo University Hospital in Monza is an advance hospital structure, which includes also the Faculty of Medicine of the University Milano Bicocca and comprises different sub-structures located north of Milan in the new province of Monza. Since 2002 hospital San Gerardo possessed a Legionella Surveillance Program and it is provided with a continuous disinfection system for potable and sanitary hot water pipeworks. Dosed disinfectant is chlorine dioxide. Cooling tower water is also disinfected with bromine-based products. After the publication of WHO book ”Legionella and the prevention of Legionellosis”, the San Gerardo hospital, strongly wanted to keep track of the work carried out since then and to achieve a structured Legionella risk management program. A Water Safety Plan (WSP) has been accomplished, as suggested by the WHO, and it is now well-accepted and fully functional. Risk assessment has been carried out with the cooperation of experts and the monitoring has been realized by the adoption of a new bioluminescence technique (measure of ATP of active biomass by the luciferin reaction) which allows the real-time measure of the total active biomass (rapid quantification of the microbiological contamination). WSP has been put in practice thanks to the collaboration between the Health and Technical Dept.’s staff. Risk assessment showed several critical points among cold and hot water networks, cooling towers, medical devices. Applied control measures included temperature monitoring, residual disinfectant concentration measurements and real-time total active biomass determination. Real-time monitoring is essential to quickly take on the right preventive and corrective actions, this is why we chose bioluminescence for microbiological contamination monitoring. The effectiveness of Water Safety Plan is periodically verified by sampling the water networks, medical devices and the cooling tower for the quantification of the Legionella content. Analyses show that the WSP is correctly implemented and functional. ATP measurement assures a rapid intervention in case of contamination.
- 225 - Session 8: Legionella prevention and control P140
Antimicrobial Activities of Some Bacillus Strains Isolated from Tap Water against Legionella pneumophila Strains
Nihal Dogru˘ oz¨ a, Zuhal Zeybekb and Ays¸ın C¸otuka aIstanbul University, Faculty of Science, Department of Biology, Vezneciler, 34134 Istanbul, Turkey; bIstanbul University, Faculty of Science, Department of Biology, Section of Fundamental and Industrial Microbiology, 34134 Istanbul, Turkey [email protected]
Legionella pneumophila (especially L.pneumophila serogroup 1) has known as causing Legionnaires Disease. These bacteria interact with the other microorganisms such as Bacillus living in the same environment. It is known that there are several antimicrobial substances which are produced by Bacillus. The effect of antibacterial substances of Bacillus spp. on L.pneumophila growth is important to prevent Legionnaire’s disease. This study was aimed to determine capable of inhibiting Bacillus strains (B. pumilus, B. circulans, B. amyloliquefaciens, B. stearotermophilus, B. licheniformis, B. megaterium, B. subtilus, Brevibacillus brevis) against the growth of L. pneumophila. In this study a total of 27 Bacillus strains isolated from the water and identified by API CHB (Biomeriux). So, it has been investigated anti - legionella activity of Bacillus strain by agar well - diffusion method. 27(93 %) Bacillus strains inhibited at least three L. pneumophila strains. The inhibition zones of L. pneumophila strains were found between 5-60 mm. 2 ( 7% ) Bacillus strains did not inhibite all tested L. pneumophila strains. The results show that the inhibition effect of Bacillus species against L. pneumophila strains is strain-specific and some Bacillus strains could be used as biological control of L. pneumophila in water systems.
- 226 - Session 8: Legionella prevention and control P141
Effective environmental sampling strategies for monitoring Legionella spp. contami- nation in hot water systems.
Monica Giacomuzzi, Savina Ditommaso, Marino Gentile, Angela Ruggenini Moiraghi and Carla Zotti
Universita` degli Studi di Torino, Via Santena 5 bis, 10126 Torino, Italy [email protected]
Background: The prevention and control of legionellosis in hospital settings involves, among other measures, environmental sampling. The data yielded by sampling constitute an important means of risk assessment and provide a valid basis on which to plan remedial (cleansing and disinfection) and preventive (maintenance) intervention. This retrospective study had two objectives: 1) to evaluate the utility of biofilm sampling at distal sites 2) to pick out any correlation between the results obtained from samples of recirculation water and those yielded by the sampling of distal sites within the same system. Methods: samples of hot water and biofilm were collected from June 1999 to March 2008 from 41 hospitals in Italy’s Piemonte region: a total of 787 sampling runs were carried out, which comprised 6443 water samples and 4045 biofilm samples. Each sampling run involved taking twelve samples: a 1-L water sample from the boiler outlet, a 1-L water sample from the recirculation loop of the hot water plant, and ten 1-L hot-water samples and ten biofilm samples from distal sites. We analyzed: 1) results of the samples (water and biofilm) taken from the same site and 2) results of the water samples taken from the recirculation loop and water samples taken from the distal sites during the same sampling run. Results: 1)data from 3910 pairs of samples (water/biofilm) showed that in 81% of the pairs, the results were concordant and in only 2% of the pairs was Legionella isolated from the biofilm sample alone. 2)data from 299 sampling runs showed that in 79% of the runs the results obtained from samples of recirculation water were concordant with water samples obtained from distal sites. In 63 of the runs (21%), the results were discordant: specifically, in 57 sampling runs recirculation water proved negative when distal sites proved positive. However in these 57 discordant sampling runs, the percentage of distal sites that proved positive during each run exceeded 30% in only 4 occasions and only 3 samples had a bacterial load > 10,000 cfu/L. Conclusion: This study indicates that 1) water sampling alone (i.e. without biofilm) has a sensitivity of 92%: biofilm sampling yields positive results in 2% of outlets (taps and showers) in which Legionella is not found in water. Thus, water sampling alone could be a viable approach, which would simplify procedures and reduce the costs of laboratory analyses; 2) sampling water from the recirculation loop can provide a useful indication of contamination of the entire water system; this means that sampling at distal sites can be limited to particular situations, such as epidemiological investigations following reported cases or the evaluation of disinfection procedures.
- 227 - Session 8: Legionella prevention and control P142
Expression of recombinant warnericin RK in Escherichia coli
Nicolas Girardin, Jean-Marc Berjeaud and Yann Hechard
University of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France [email protected]
Warnericin RK is a 22 residue cationic peptide with strong anti-Legionella activity isolated from S. warneri RK. Due to its specific anti-Legionella activity, this molecule should attract practical interest for environmental or medical use. However researchers are likely to encounter difficulties in obtaining workable quantities of peptide and recombinant expression is one way for producing suitable amounts of warnericin RK. The 66 bp synthetic gene encoding warnericin RK was cloned into an Escherichia coli expression system (pQE30 and pQE70). Warnericin RK was expressed, in E. coli M15 strain, through three distinct constructs: as His-tagged warnericin RK on N-terminus or C-terminus and as an unmodified warnericin RK. Production of recombinant peptides was confirmed by anti-Legionella activity assays with the E. coli crude extract and by western blot. Subsequently, fusion peptides were purified by metal affinity chromatography while the native peptide was purified using reverse-phase HPLC. The recovered peptides were then characterized by ESI-mass spectrometry and evaluated for their anti-Legionella activity. Comparing the anti-Legionella activities of the His-tagged peptides to the native peptide, we found that the His-tag decreases the activity of warnericin RK on Legionella by two times for the C-terminus tagged peptide and by four times for the N- terminus tagged peptide. This is the first report of the heterologous expression of warnericin RK, an anti-Legionella peptide. This system would be used for production of warnericin RK for structural studies or production of warnericin RK variants for structure/function studies.
- 228 - Session 8: Legionella prevention and control P143
What strategy for detecting airborne Legionella pneumophila?
Thi-Lan Haa, Laurence Mathieub and Enric Robinea aCentre Scientifique et Technique du Batiment,ˆ 84 avenue Jean Jaures,` 77447 Marne-la-Vallee,´ France; bEcole Pratique des Hautes Etudes, LCPME, UMR CNRS 7564, Poleˆ de l’Eau, 15 avenue du Charmois, 54505 Vandoeuvre-les-Nancy,` France [email protected]
Many uncertainties concern yet the relationship between the presence of airborne Legionella pneumophila (Lp) and the human contamination risk. The difficulty is mainly due to the lack of well-characterized methods for assessing the concentration of microorganisms in air. The variability of results depends on the sampling technique used and on the analytical method. The aim of this study was to assess the relevance of methods currently available for sampling and detecting this airborne pathogen. The methodology is based on laboratory experiments, with a controlled production of aerosols of Lp in a vertical wind tunnel where the samplers are placed. Three commercially available biocollectors based upon impaction into a liquid, the CIP 10-M, the BioSampler and the Coriolis were evaluated for their physical sampling efficiency and the stress they cause on collected Lp. Different assays on aerosols samples were performed by culture method, immunofluorescence detection, in situ hybridization and real-time PCR techniques. Physical collection efficiency was calculated as the ratio of the total collected bacterial concentration to the airborne Lp concentration upstream from the sampler. Test bioaerosols were measured at about 0.8 µm and closed in size to planktonic Lp isolated from aquatic environments. The collection rates varied from 100% for the BioSampler, 4% for the Coriolis and less than 1% for the CIP 10-M. The bacterial injury determined from the total and culturable concentrations of Lp in the collection liquid revealed to be not dependent on the sampler used. It seems just to result from the impact of the microorganism onto the surface liquid during impingement. Whatever the microbial assay applied, Lp were detected in all aerosol samples, the concentration mainly depending on the collector and its sampling efficiency. When airborne concentration decreased, detection by immunofluorescence was more suitable than the others based on culture method or molecular techniques. The results contribute to improve the reliability of measurements of airborne microorganisms for the characterization of the human exposure to Legionella pneumophila. The choice of the sampling strategy might also be addressed as regards the environmental conditions in which measurements would be performed and the characteristics of biological aerosols dispersed from realistic sources (size, concentration). Experiments are in progress to document these aspects and to improve the knowledge on the emission of airborne Lp from cooling towers. Acknowledgments: This work has received a financial support from the Agence Franc¸aise de Securit´ e´ Sanitaire de l’Environnement et du Travail. The authors gratefully thank the Institut National de Recherche et Securit´ e´ for helpful discussion and Electricite´ de France, Bio-Rad and Bertin Technologies for their technical support.
- 229 - Session 8: Legionella prevention and control P144
Detergent-like activity of warnericin RK, a specific anti-Legionella peptide
Julien Verdona, Mirjam Falgeb, Helke Maierb, Heike Bruhnb, Christian Lacombea, Jean-Marc Berjeauda, Michael Steinertc, Cornelius Faberb, Roland Benzb and Yann Hecharda aUniversity of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France; bUniversity, Wuerzburg, 97074 Wuerzburg, Germany; cInstitut fur¨ Mikrobiologie, Technische Universitat¨ Braunschweig, Spielmannstr.7, 38106 Braunschweig, Germany [email protected]
Warnericin RK is the first antimicrobial peptide described to be active against Legionella pneumophila. Strikingly, this peptide displays a very narrow range of antimicrobial activity, almost limited to the Legionella genus, and a hemolytic activity. A similar activity has been reported for δ-lysin, which is a well-known staphylococcal hemolytic peptide. We thus hypothesized that Legionella could have a specific feature explaining its sensitivity. Our study was aimed to understand the mode of action of warnericin RK in order to explain the specific sensitivity of Legionella. We first reported that warnericin RK permeabilized artificial membranes in a voltage-independent manner and that the pores were rather cations selective. Warnericin RK also permeabilized Legionella cells. Osmotic protection experiments on erythrocytes showed that warnericin RK does not formed well-defined pores, suggesting a detergent-like mode of action, as previously shown with δ-lysin. Warnericin RK structure was defined by NMR. The peptide adopted an amphiphilic α-helical structure, consistent with the proposed mode of action. In addition, we showed that Legionella displayed a high sensitivity to detergents. We thus hypothesized that Legionella might have a specific membrane composition. A resistant variant, able to grow at a concentration 32 fold higher than MIC, was isolated after repeated transfers of L. pneumophila on culture medium containing increasing concentrations of warnericin RK. As we assumed that Legionella membrane composition could modulate the sensitivity, phospholipids were identified by direct electrospray inonization- mass spectroscopy and fatty acids were identified by GC-MS. The resistant variant showed a higher amount of branched- chain fatty acids than the wild type strain suggesting that they could be, at least partially, responsible for sensitivity. We conclude that the detergent-like mode of action of the peptide and the high sensitivity of Legionella to detergents could explain the specificity of warnericin RK towards these bacteria.
- 230 - Session 8: Legionella prevention and control P145
A novel system for the control of legionella in the water system of the bone marrow transplant unit at Glasgow Royal Infirmary - nine years experience.
John Hooda, Gordon Cheapeb, David Gormlieb, Alex Meada, Ian Powrieb, Gordon Willsa and Evonne Curranc aGlasgow Royal Infirmary, Department of Clinical Microbiology, G4 0SF Glasgow, United Kingdom; bGlasgow Royal Infirmary, Estates Department, G4 0SF Glasgow, United Kingdom; cHealth Protection Scotland, 1 Cadogan Square, Cadogan St, G2 7HF Glasgow, United Kingdom [email protected]
Background: A nine - bedded Bone Marrow Transplant Unit (BMTU) was commissioned at Glasgow Royal Infirmary in early 1999. Previously, in 1990, a patient in the old BMTU died of Legionella pneumophila serogroup 1(Lp1) infection, acquired there. No patient was allowed to shower in that Unit between 1990-99. When the new Unit was proposed the Staff demanded that the water system be safe enough to allow patients to shower, wash and bathe. The following system was designed and commissioned in 1999, and was operational until the Unit closed and transferred to a new Unit (Beatson Cancer Centre) in April 2008 . Water came directly from the main and passed through a 0.5 micron (absolute) Memcor o filter. It was then dosed with chlorine dioxide (ClO2) to 0.5ppm, chilled to 12 C and stored in a glass-lined tank. The cold water was then circulated around the ward where instantaneous water heaters [IWH] (Stiebel Eltron) were employed at each sink or shower. Each shower unit was programmed to flush automatically for 1 min each day. The Oncology Unit upstairs employed the same filtered water containing ClO2. Hot water, however, was provided by a unit calorifier (from the treated cold water). This was circulated at 60oC with return at 57oC. It was then mixed at each outlet and shower with the treated cold water by thermostatic mixing valves. All drinking water was sterilised bottled. Objective: to monitor the effectiveness of the legionella-control systems in both water supplies. Methods: continuous surveillance of all outlets in the BMTU and selected outlets in the Oncology Ward for legionella (5L samples by filtration until July 2007 and thereafter 1L samples by centrifugation). Results: No isolation of Lp from 1050 samples from outlets in the BMTU, but 11 isolations of L. anisa. No isolation of Lp from 763 samples from Oncology ward outlets, but 89 isolations of L. anisa. The isolation of L. anisa resulted in the outlet or shower being cleaned with 500ppm ClO2. The difference between the two systems is statistically significant with a Relative Risk Ratio of 11.13 and 95% confidence limits between 5.99 and 20.69. The p value is <0.01 using Fisher’s exact 2-sided test. There were no cases of legionellosis. Conclusions: we have demonstrated over a 9 year period that a circulating cold water system that is filtered to 0.5 micron, dosed with 0.5ppm o ClO2, chilled to 12 C and employs point of use IWHs is an extremely effective means of controlling Legionella spp. in a highly susceptible population in a hospital with a significant past history of hospital-acquired legionella infections. IWHs are more effective in controlling legionella than a conventional domestic supply i.e. mixing treated cold water with circulating hot water at 60oC.
- 231 - Session 8: Legionella prevention and control P146
16 months experience of legionella control/water quality in a new cancer centre (including a bone marrow transplant unit): a comparison of instantaneous water heaters versus a conventional domestic hot/cold water system.
John Hooda, Gordon Cheapeb, Mel Aitkenc, Alex Meada and Gordon Willsa aGlasgow Royal Infirmary, Department of Clinical Microbiology, G4 0SF Glasgow, United Kingdom; bGlasgow Royal Infirmary, Estates Department, G4 0SF Glasgow, United Kingdom; cBeatson West of Scotland Cancer Centre, Estates Department, Gartnavel General Hospital, G12 0YN Glasgow, United Kingdom [email protected]
The Beatson West of Scotland Cancer Centre opened in February 2008 serving a population of 2.6 million. It is the second largest cancer centre in the UK. It has 3 in-patient levels. Level 4 consists of a Bone Marrow Transplant Unit (BMTU) with 10 single protective isolation rooms. There are a further 18 protective isolation rooms for haemato-oncology patients. With experience gained from the system in the BMTU at Glasgow Royal Infirmary (GRI) since 1999 (see Abstract no 217) it was decided to commission a similar water system for Level 4 Beatson, that would offer robust legionella control for these highly susceptible patients. The Gartnavel General site in Glasgow has never been associated with nosocomial legionellosis , therefore a decision was taken not to include the addition of a biocide (chlorine dioxide) to this system initially but to do continuous microbiological surveillance. Level 4 Cold water from main through 0.5 micron Memcor filter then into fibreglass tank. Water then chilled to 7oC and continuously circulated. Each outlet has an instantaneous water heater (IWH) [Steibel Eltron, 18kW, 3 phase]. Showers purge automatically once daily. Levels 0 to 3 Cold as above into tanks. Then a conventional hot and cold domestic water system. Circulating hot water from 2 calorifiers (out at 70oC and return at 60oC). Hot water blended at outlets with cold employing TMVs. Methods Each shower or sink outlet in Level 4 checked monthly (in rotation) for presence of legionella and Total Viable Counts (TVCs) Levels 0 to 3 a selection of ward outlets checked each 2 weeks (in rotation) for TVCs and a selection of shower outlets checked for both TVCs and legionella. 1L sample for legionella by centrifugation. TVC from 1ml of water sample. Sampling from April 2008 to July 2009 Results Legionella Level 4: 594 1L samples NEGATIVE for all legionella Level 0 to 3: 473 1L samples NEGATIVE for all legionella TVC’s (Acceptable = <100@ 22oC and <10@ 37oC) Level 4: 1210 water samples, 1086 had normal TVCs with 124 abnormal TVCs Level 0 to 3: 1210 water samples, 706 had normal TVCs with 594 abnormal TVCs On the water quality results there is a statistically significant difference between the 2 systems. The Relative Risk Ratio is 4.06 (95% confidence limits between 3.39 and 4.864). The p value is <0.01 using Fisher’s exact 2-sided test. Looking at the same TVC’s but results at ten times the acceptable level (Level 4:1192/18 vs Level 0-3: 948/262) the Relative Risk Ratio rises to 14.55 (95% confidence limits between 9.1 and 23.3). The p value is <0.01. Conclusion Preliminary results show Level 4 has a robust system for the control of legionella - without a biocide. The conventional domestic water system in levels 0-3 is also effective in legionella control. However general water quality is significantly better in the system employing IWH. This is further evidence that systems employing IWH are highly effective for legionella control.
- 232 - Session 8: Legionella prevention and control P147
Monochloramine Efficacy on Biofilms in a Model Water Distribution Pipe Rig
Nazmiye Ozlem Sanli Yurudu, Esra Ilhan Sungur, Irfan Turetgen,¨ Nihal Dogru˘ oz,¨ Ays¸ın C¸otuk and Duygu Goksay¨
Istanbul University, Faculty of Science, Department of Biology, Vezneciler, 34134 Istanbul, Turkey [email protected]
Biofilms formed on pipe walls used for potable water distribution are known to cause to serious public health problems by protecting and/or supporting pathogenic microorganisms. Some potentially pathogenic bacteria such as Legionella pneumophila was frequently detected in biofilms. Biocides are used to get under control potentially pathogenic bacteria and keep biofauling at minimal levels in potable water systems. While monochloramine is considered more stable than chlorine, there is not any knowledge about the effect it has on water distribution pipe biofilms in Turkey. In this investigation, we tested monochloramine efficiency against developed biofilms from unamended distribution water in a laboratory polypropylene pipe rig designed to model a water distribution pipe systems for 8 months. Pipe segments were cut monthly and effect of 2 ppm monochloramine on the biofilm was analyzed in terms of aerobic heterotrophic bacteria and L. pneumophila after zero time, 1st and 3rd hours of exposure. The number of aerobic heterotrophic bacteria was assayed by colony formation on R2A agar. Also the number of total and respiring bacterial cells in the same samples were respectively determined by direct counting of 4’,6-diamidino-2-phenylindole (DAPI) and 5-cyano- 2,3-ditolyl tetrazolium chloride (CTC). Heterotrophic bacterial colony counts on agar media declined by minimum 0.38 log and maximum 2.64 log reduction after 3 hours of monochloramine treatment in August and September respectively. The numbers of respiring bacteria declined in response to monochloramine treatment by lower than 1 log reduction after 3 hours of monochloramine treatment and much less than did the number of culturable bacteria during the experiment. Respiring cell counts detected by CTC were higher than plate counts. Legionella pneumophila were not found in both treated and untreated biofilm samples during the 8 months-period.
- 233 - Session 8: Legionella prevention and control P148
Antibacterial and Hemolytic Activities of Amsonia orientalis (Syn. Rhazya orientalis) Decne. against Legionella pneumophila strains
Ayten Kimiran-Erdema,Gul¨ Cevahir-Oz¨ b, Elif Ozlem¨ Arslan-Aydogdu˘ a and Elif Yuzbas¸ıo¨ glu˘ b aIstanbul University, Faculty of Science, Department of Biology, Vezneciler, 34134 Istanbul, Turkey; bIstanbul University, Faculty of Science, Department of Biology, Section of Botany, 34134 Istanbul, Turkey [email protected]
Amsonia orientalis Decne. belonging to Apocynaceae is a medicinally important plant which has very restricted distribution only in Turkey and Greece on the world. Literature on A. orientalis, which is critically endangered in nature according to IUCN categories, is very limited in number. Therefore, the objectives of this research were to investigate antibacterial and hemolytic activities of extracts obtained from the plant against eleven different Legionella pneumophila strains isolated from water system in Istanbul. In the current study, A. orientalis that has been used as a biological material was identified by Dr. Osman Erol, Department of Botany. The samples were stored at the Herbarium, Faculty of Science, Istanbul University (Herbarium no: 40081 ISTF). The different parts of A. orientalis were extracted with deionise water, ethanol, methanol, aceton, chloroform, n-hexane. Antibacterial activity of these extracts against L. pneumophila strains were carried out according to the agar diffusion method. Hemolytic activity was determined using human red blood cells. The results obtained in the present study suggest that A. orientalis extracts had the ability to inhibit the growth of L. pneumophila and the antibacterial activity differs with the applied extraction method. The ethanol leaf extract of A. orientalis was active against the tested L. pneumophila strains (%82) when compared to acetone extract (%64). Hovewer, the hexane leaf extract did not show antibacterial activity against any of the Legionella strains tested. In the present study, it was determined that chloroform stem+leaf extract, the hexane leaf extract and deionise water stem extract did not show hemolytic activity. The ethanol leaf extract exhibited only antibacterial activity with low hemolytic activity As a result, because stem+leaf extract which is not hemolytic activity shows antibacterial activity on L. pneumophila strains, it can be said that this extract is suitable for treating diseases caused by these bacterial strains. It has been thought that these extracts that are both non-hemolytic and antibacterial activity will be purified and continued to work for further studies.
- 234 - Session 8: Legionella prevention and control P149
The effect of various biocides on survival of different Legionella pneumophila strains
Ayten Kimiran-Erdema, Nazmiye Ozlem Sanli Yurudua, Elif Ozlem¨ Arslan-Aydogdu˘ a and Zuhal Zeybekb aIstanbul University, Faculty of Science, Department of Biology, Vezneciler, 34134 Istanbul, Turkey; bIstanbul University, Faculty of Science, Department of Biology, Section of Fundamental and Industrial Microbiology, 34134 Istanbul, Turkey [email protected]
Cooling towers are known to be one of potential sources for harboring, amplifying and disseminating legionellae. In order to minimize potential health hazards associated with Legionella bacteria, the excessive growth of legionellae must be prevented in cooling towers. Thus, one of the key issues for controlling legionellae in water systems is the proper use of the most effective biocide. The aim of this study is to evaluate the minimal inhibitory concentrations (MIC) of various biocides against different Legionella pneumophila strains and determine the actual dosages of active ingredient and contact time required to control these bacteria. The following five products were studied against Legionella pneumophila strains: Gemacide K4 Kloran, didecyldimethyl ammonium chloride, benzalkonium chloride, 4-isothiazolin 3-one and the mixture of benzisothiazol and isothiazolin-ones. The Minimal Inhibitory Concentration (MIC) was determined by a standard macrodilution method. Biocide susceptibility was assessed based on the modified methods of Skaliy et al. (1980). According to the results, it was shown that all biocides showed different inhibitory activity against all the strains and the efficacy of biocides depend on dosages and the origin of the strains (standard or environmental). When the susceptibility of Legionella pneumophila strains was compared according to MIC values, it was found that the SG 1 strain was the most resistant strain and the susceptibility of standard and SG 2-14 strains was found to be similar against biocides tested. The data demonstrate that the quaternary ammonium compounds (QACs) were found effective at the lower dosages than the dosages recommended by the manufacturer. Benzisothiazol and isothiazolin-ones, 4-isothiazolin 3-one which are isothiazolin mixtures were less effective against Legionella than QACs. The recommended dosages by manufacturer may not be proper for each system and each target microorganism. The results showed that effective biocide applications can be achieved by pre-determining MIC and minimum contact time of different biocides to kill target bacteria. In order to avoid the development of resistant strains of microorganisms, biocides must be applied to systems in correct dosages and contact times; otherwise it would be detrimental rather than beneficial.
- 235 - Session 8: Legionella prevention and control P150
Legionnaires’ disease associated with a new residential area - risk factors and remedial actions
Louise Hjelmar Krøjgaarda, Karen Angelika Krogfelta, Hans-Jørgen Albrechtsenb and Søren A. Ulduma aStatens Serum Institut, Department of Bacteriology, Mycology and Parasitology, Artillerivej 5, 2300 Copenhagen S, Denmark; bTechnical University of Denmark, Department of Environmental Engineering, Miljoevej, bygning 115, 2800 Lyngby, Denmark [email protected]
At the end of 2008 / beginning of 2009, two men from the same newly built residential area were diagnosed with legionnaires’ disease, one of them died. The new residential area consists of 225 apartments (210 were inhabited) spread over 6 blocks. The deceased man (Case I) had lived in the apartment for 2 months and had complained about low hot- water temperature. The other man (Case II) only stayed in the residential area for a few days, in an apartment which had not been occupied before. The strain isolated from both patients was Legionella pneumophila serogroup (sg) 1 subgroup Philadelphia (Sequence Type (ST) 1). The same type along with L. pneumophila sg 3, was detected in high numbers in water samples from the apartment where the deceased man lived, especially in water samples from the shower hose, but also in samples from the hot-water circulating system for the whole residential area. Several risk factors were identified during this small outbreak 1) water stagnation in the pipes in the apartments for longer periods before the residents moved in (Case II) 2) stagnant water in the shower hoses (Case I and II) 3) low flow in the water circulation system 4) low temperature in the hot water circulation (46◦C, Case I) and 5) presence of a virulent L. pneumophila strain in the water. Stagnant water in pipes seems to be a problem in new residencies since the water in the pipes are left at ambient temperature until people moves in During this period, a biofilm with Legionella can build up. In order to prevent high levels of Legionella in water pipe systems in new buildings, standard procedures to clean the systems should be applied before people moves in. To overcome the Legionella contamination from the water system, the water temperature was raised for longer or shorter periods (12 h-48 h), flow was increased, and shower hoses and heads where disinfected with chlorine. In an empty apartment, the effect of five minutes’ flush with hot water was tested. Only heat treatment (48 hours, boiler adjusted to 70-80◦C, tap water 65◦C) was effective to reduce the presence of Legionella.
- 236 - Session 8: Legionella prevention and control P151
Occurrence and prevention of legionella in water systems of a ferry ship after two Legionnaires’ disease cases
Jaana Kusnetsova, Lotta Simolab, Silja Mentulab, Petri Ruutub, Piia Airaksinena and Nhu Nguyen Tran Minhb aNational Institute for Health and Welfare, Water and Health Unit, P.O.Box 95, FI-70701 Kuopio, Finland; bNational Institute for Health and Welfare, P.O.Box 30, FI-00271 Helsinki, Finland jaana.kusnetsov@thl.fi
In 2007 three clusters of Legionnaires’ disease were associated with ships in Europe. One of them linked two truck drivers to a ferry ship harbouring Helsinki, Finland. The drivers stayed in the same cabin on successive cruises in August. They developed severe pneumonia six and four days after the cruises, but later recovered. The cases had positive legionella urine antigen test, and Legionella pneumophila serogroup 1 was isolated from one of them. After the first sampling of ship water systems in October, shock-chlorinations and continuous chlorination of cold water system, improved chlorination of spa water and temperature increase of hot water were implemented. Cold water pipelines were found to be heavily contaminated (Legionella micdadei, range from 14000 to 970000 cfu/l, n=3). After several control sessions, the concentration decreased to 510 cfu/l or less in cold water pipelines (n=3). No legionella was isolated from cold water tanks (n=9). In hot water system, concentrations of legionella were lower but L. micdadei was isolated from showers (range from <5 to 10000 cfu/l, n=8), taps (from 680 to 1500 cfu/l, n=2) and hot water returning from circulation (1900 cfu/l). After water temperatures were increased by 10◦C (leaving water to 64◦) and 7◦C (returning water to 45◦C), counts in returning water decreased to 1500 cfu/l. Only one shower remained positive (1000 cfu/l, n=6), and the taps were free of legionella (n=2). The first samples of the spa pools A and B having shared water circulation contained legionella up to 70 000 cfu/l (Legionella sainthelensi, 0 mg/l Cl2). The last four samplings gave negative results as chlorination system was repaired (0,24-1,4 mg/l Cl2). The pool C which was reserved for truck drivers only could not be completely sampled as it had been out of service. Comparison between clinical and environmental strains could not be performed as no L. pneumophila sg 1 was isolated from the ship water systems. Epidemiologically, however, the ship water systems seemed as the most probable source of these infections. In water systems of ships more efficient prevention methods against legionella should be implemented, which should be guided by national and international regulations.
- 237 - Session 8: Legionella prevention and control P152
Design, Operation and Maintenance of Building Services to Minimize the Risks of Legionellosis: a new European Guidebook for Practitioners
Sergio La Muraa, Cesare Maria Joppolob and Luca Alberto Pitera`c aSiram Spa - Aicarr, Via Bisceglie 95, 20100 Milano, Italy; bPolitecnico Milano, Piazza L Da Vinci 32, 20100 Milano, Italy; cAicarr, via M Gioia 162, 20100 Milano, Italy [email protected]
Important and effective efforts have been made, and are on-going, in order to develop a comprehensive overview of the sources, ecology, laboratory identification of Legionella, and of medical practices for outbreak management of Legionellosis. Autoritative documents do exist providing guidance on assessment and management of risks associated with potentially hazardous environments, such as cooling towers, pools and spa baths; these documents are useful to all those concerned with Legionella and health, including predominantly environmental and public health officers, health- care workers, researchers and special interest groups. However the documents identifying the necessary measures to prevent, or adequately control, the risk of exposure to Legionella bacteria in each particular system and application, cover with a quite lesser effectiveness the engineering field. In this respect, some differences do exist in the various national and international documents and also there is a lack of comprehensive coverage of the different phases of the building services life-cycle (design, installation, commissioning, operation and maintenance, retrofit and modification, etc.). The paper assess the situation of existing engineering texts and presents the on-going project of a new Guide Book covering the practical measures to prevent the risks of Legionellosis associated with drinking, cold and hot water systems, cooling towers and evaporative cooling, air humidification and Air conditioning systems. The Guide Book will be reviewed and published by REHVA, the Federation of European Heating and Air-conditioning Association - that represents more than 110 000 building engineers from 30 European countries; the working Group has been headed by AICARR (the Italian Air Conditioning Heating Refrigeration Association ) and involved experts from some other European countries.
- 238 - Session 8: Legionella prevention and control P153
Lessons from an outbreak of Legionnaires’ disease on a cruise ship
Sandra Laia, John V. Leea, Lorraine Sadler-Reevesb, Timothy Harrisonc, Carol Josephd, Allan Johnsone,Gorel¨ Allestamf, Birgitta De Jongg, Gillian Ashfordh, James Sedgwickh and Mathibalasingham Chandrakumarh aWater & Environmental Microbiology Reference Unit, Health Protection Agency, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; bKent Environmental Microbiology Services, Health Protection Agency,, Kennington Road, Willesborough, Ashford, TN24 0LZ Kent, United Kingdom; cHealth Protection Agency Centre for Infections, Respiratory and Systemic Infection Laboratory, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; dHealth Protection Agency Centre for Infections, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; eFood Water and Environmental Microbiology Network - London Laboratory, Health Protection Agency, 61 Colindale Avenue, NW9 5EQ London, United Kingdom; fSwedish Institute for Infectious Disease Control (SMI), Dep Parasitology, Mycology, Water and Environmental Microbiology (PMV), SE-171 82 Solna, Sweden; gStockholm County Council, Dept. of Communicable Disease Control and Prevention, Box 17533, SE- 118 91 Stockholm, Sweden; hKent Health Protection Unit, Preston Hall, Aylesford, Maidstone, ME20 7NJ Kent, United Kingdom [email protected]
Nine passengers on a 17-night cruise around Scandinavia and Europe aboard the Black Watch cruise ship in July 2007 were confirmed to have Legionnaires’ disease. Five passengers were hospitalized in Sweden and four in UK. Of these, three were culture-confirmed cases caused by a strain of L. pneumophila serogroup 5 (Dallas-1E). The remaining six confirmed cases were identified by PCR or serology. Initial environmental samples for legionella testing were collected in Sweden from cabins, the spa pool, swim exercise pool, shower, pool heat exchanger and condensate from an air conditioner. Following preliminary results from Sweden, samples from fresh water tanks and the restaurant fountain were taken in Germany as the ship was on its return to England. A thorough survey of the ship and further extensive environmental sampling were undertaken on four occasions when the ship docked in Dover. Investigations focussed on the ship’s water systems including the water supply, water storage, water usage and recreational water facilities. Patient diagnostic tests in Sweden and England included, urinary antigen tests, PCR, culture for Legionella and serology. Epidemiological investigation through questionnaires was carried out by the Health Protection Agency in the UK to establish the risk factors amongst the confirmed and suspected cases. There was strong microbiological evidence of a link between the ship’s water system and the three culture confirmed cases. A variety of different strains of L. pneumophila were isolated from the ship’s water systems including strains matching those from the patients. An extensive survey and risk assessment of the ship’s water systems identified a number of risk factors. Factors key to the success of this investigation were: good communication between countries; the use of diagnostic methods other than just urinary antigen testing, in particular culture; extensive and timely environmental sampling and full co-operation between the outbreak management team and the cruise ship’s owners and staff. In order to prevent outbreaks of Legionnaires’ disease on cruise ships there is a need for regular risk assessments of the water systems.
- 239 - Session 8: Legionella prevention and control P154
Preventing Legionella growth and controlling water quality in public spa pools
Robert Laua, Brian Caughleya and Rob Deaconb aCollege of Sciences, Massey University at Wellington, New Zealand, Wallace Street, 6021 Wellington, New Zealand; bEnvironmental Laboratory Services Ltd, 85 Port Road, 5010 Lower Hutt, New Zealand [email protected]
PREVENTING Legionella growth and controlling water quality in public spa pools Robert Lau1, Brian Caughley1, Rob Deacon2 1Institute of Food, Nutrition & Human Health, College of Sciences, Massey University at Wellington, New Zealand 2Environmental Laboratory Services Ltd, Lower Hutt, New Zealand. Abstract The New Zealand Standard NZS 5826:2000 Pool Water Quality covers essential criteria in the management and maintenance of pools regarding pool water quality. Our study aims to determine whether the criteria are adequate for preventing the growth of Legionella bacteria, indicator microorganisms and fungi in public spa pools. Between March and September 2007, ten public spa pools in Wellington, New Zealand were each sampled on four different days. Alkalinity, calcium hardness, free available chlorine (FAC), combined available chlorine (CAC), total dissolved solids (TDS) and pH were studied. Microbiological tests for Legionella bacteria, fungi, indicator microorganisms Staphylococcus aureus, Pseudomonas aeruginosa, faecal coliforms and heterotrophic plate count (HPC) bacteria were performed. On the whole, the chemical results which were outside the acceptable range were not far outside except for one sample which had a CAC value of 7.3 and well above the upper limit of 1.5. Legionella pneumophila serogroup 8 was isolated from one spa pool. All samples passed the Standard faecal coliforms, Staphylococcus aureus, Pseudomonas aeruginosa tests. All samples passed the Standard HPC test except for three. Fungi were found in two spa pools. Our findings suggest that adherence to the Standard NZS 5826:2000 Pool Water Quality and its chemical and microbiological criteria can safeguard public health. Legionella bacteria that can cause Legionnaires’ disease can grow in spas if chemical levels are not adequately managed within the acceptable range. References Kura F, Amemura-Maekawa J, Yagita K et al (2006). Outbreak of Legionnaires’ disease on a cruise ship linked to spa-bath filter stones a contaminated with Legionella pneumophila serogroup. Epid Infect, 134, 385. Nakadate T, Yamauchi K, Inoue H (1999). An outbreak of Legionnaires’ disease associated with a Japanese spa. Nihon Kokyuki Gakkai Zasshi 37,601. Standards New Zealand (2000). New Zealand Standard NZS 5826:2000 Pool Water Quality. Wellington, New Zealand.
- 240 - Session 8: Legionella prevention and control P155
Legionella risk assessment and control on board train: effectiveness of environmen- tal surveillance and bacteria genotyping
Patrizia Laurentia, Romina Sezzatinia, Gianluigi Quarantaa, Stefania Bocciaa, Rosarita Amorea, Cinzia Turnaturia, Francesco Dalla Torrea, Federica Bonintia, Marta Marinoa, Elio Munafo`b, Pasquale Del Nordb, Mario Russob, Massimo Accorsib and Gualtiero Ricciardia aInstitute of Hygiene Catholic University of Sacred Hearth, Largo Francesco Vito, 1, 168 Rome, Italy; bOccupational Health Service of the Italian Railways, Via F. A. Pigafetta 3, 154 Rome, Italy [email protected]
The aim of this work is to show the results of an environmental surveillance of Legionella spp. on board train, in order to evaluate the effectiveness of water tank decontamination. Clonal relatedness among the isolates was evaluated by Pulsed-Field Gel Electrophoresis (PFGE). From September 2006 we examined plumbing and toilets water tanks on passengers trains. The first cycle of decontamination procedures -performed when Legionella concentrations showed values > 1000 CFU/l- was completed in September 2008 (pre-decontamination step). October 2008 was considered as the start of post decontamination period. Microbiological detection of Legionella spp. was carried out according to the Italian ”Guidelines for legionellosis prevention and control”. According to the criteria of Tenover et al., a PFGE patterns designation was assigned if the electrophoretic profile differed by more than 3 bands. Unpaired data t-test was used to evaluate differences in contamination rate before and after decontamination phase. Nine-hundred sixty-five water samples were analyzed: positivity for Legionella (>1000 CFU/l) was found in 18% samples in the pre-decontamination period, and 1,5% in the post-decontamination. Four hundred fifty five strains of L. pneumophila were bio-chemically typed: the 57,9% belong to serotype 1. Patterns of isolates in pre-decontamination period showed 12 different clones, with 32 of 60 genotyped strains (about 50%) of serotype 1 belonging to the same clone. Pattern of isolates in post- decontamination period showed 4 different clones, with 3 of them present in pre-decontamination period, and 1 of new appearance. Lastly, 17 of 22 genotyped strains (about 73%) of serotype 1, were the same clone than 32 strains of pre- decontamination period. Comparing post-decontamination results to pre-intervention ones, we obtained a statistically significant decrease of the total bacterial count (t= -2.7508; p = 0.0030), confirming the interventions’ effectiveness. This study shows how environmental monitoring is an ideal approach to check rail water supply contamination, and also a good procedure to assess decontamination practices’ effectiveness, as well as seasonal related bacterial count increase. Genotyping is useful for qualitative evaluation: the persistence of clones after decontamination and the appearance of new clones suggests environmental favourable conditions to control with particular attention.
- 241 - Session 8: Legionella prevention and control P156
A controlled evaluation of deadlegs in Legionella colonization in a model plumbing system
Yusen Lin and Hsiu-Yun Shih
National Kaohsiung Normal University, Rm615, Cosmos Bldg., NKNU,, 62 Shen-chong Rd. Yanchou, 824 Kaohsiung, Taiwan [email protected]
Removal of deadlegs is a frequently recommended yet controversial intervention for control of Legionella colonization in water distribution system. Deadlegs may provide a ”stagnant” aquatic environment for Legionella to proliferate and chemical disinfectants or heat are difficult to penetrate. However, even if small amount of Legionella exist in dead legs, the effect may be inconsequential since minimal exchange of water between ”dead legs” and distribution lines will occur by definition. Thus, the objective of this study is to determine the role of deadlegs in Legionella control using a model plumbing system. Four 3-meter long PVC pipes (dia. 1”) were connected to a model plumbing systems to create ”deadlegs” (Deadleg 1 to 4). Test solution containing Legionella at approx. initial concentration of 3,000,000 cfu/ml was added into the model system. After initial recirculation and all four deadlegs were positive for Legionella, 2 ppm of chlorine was added to the recirculation system control the Legionella growth. Our results showed that though very low number of Legionella was recovered from deadlegs (0.2 - 8 cfu/ml), no Legionella was recovered from the recirculation water; ie. Legionella growth was inhibited even low concentration of Legionella was present in the deadlegs. After 30 days, the chlorine was neutralized by sodium thiosulfate, Legionella was detected in recirculation water 2 weeks after dechlorination. Based on our results, Legionella can be controlled by effective disinfectants even Legionella is present in deadlegs. Effective disinfection of water system is the key to successful infection control. The efficacy of removal of dead legs remains unestablished and the procedures are costly, labor-intensive and tedious.
- 242 - Session 8: Legionella prevention and control P157
Accuracy of Legionella isolation by US Labs in the ELITE Program pilot study
Claressa Lucasa, Thomas Taylora, Nicole Alexanderb and Barry Fieldsa aCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, GA 30333, United States of America; bCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop C-23, Atlanta, GA 30333, United States of America [email protected]
In the United States of America (US) commercial and local public health labs that test for Legionella environmental contamination are not accredited by any regulatory body. Historically, it has been difficult to gauge the accuracy and/or precision of these labs in isolating, identifying, and enumerating Legionella bacteria. The Environmental Legionella Isolation Techniques Evaluation (ELITE) Program, a proficiency testing scheme for labs within the US, was created to address these concerns. A pilot study was performed to establish baseline expectations. The results (n=286) as of June 12, 2009 were reported by 12 commercial, 4 state, 2 hospital, 1 county, and 1 federal lab. Participants processed panels consisting of six lyophilized samples designed to reflect potable water and reported results via an online interface. The majority (93%) of positive Legionella samples (86%) and negative control samples (100%) were correctly characterized by participants. For samples containing Legionella (n = 190), participants underestimated the cfu/ml by a mean of 1.25 logs with a standard deviation of 0.8 logs, a standard error of 0.07 logs, and a range of 0-3 logs. Incubation time (p = 0.3), incubation temperature (p = 0.3), and CO2 concentration (p = 0.5) did not significantly affect the accuracy of reported concentrations (cfu/ml).
- 243 - Session 8: Legionella prevention and control P158
Identification and purification of anti-Legionella peptides produced by staphylococcal strains
Adrienne Marchand, Julien Verdon, Christian Lacombe and Jean-Marc Berjeaud
University of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France [email protected]
We recently isolated a bacterial strain, Staphylococcus warneri RK, which displays an anti-Legionella activity as well as a haemolytic activity (3). These activities are related to three peptides secreted in the culture supernatant, two delta-lysins and an original peptide, named Warnericin RK (6). Warnericin RK is, to our knowledge, the first anti-Legionella peptide described in the literature (3). The aim of this study was to screen various staphylococcal strains for their anti-Legionella and haemolytic activity, and then purify and characterize the peptides involved in this activity. Ten from the 18 staphylococcal strains, chosen to be representative of the different species known to produce haemolytic peptides, displayed anti-Legionella and haemolytic activities. Four of them (S. epidermidis 567, S. haemolyticus, S. lugdunensis 967 and S. aureus 2850 ) appeared as active as S. warneri RK. Strains from these species were known to produce peptides which were already described. All these 4 strains display the same antibacterial spectrum restricted to the genus Legionella, as S. warneri RK (6), except Staphylococcus aureus 2850, which also had an anti-Pseudomonas activity. Peptides responsible for the anti-Legionella activity were purified with a new method using a hydrophobic interaction chromatography and a reverse phase HPLC, then analyzed by electrospray mass spectrometry. Most of the peptides were identified according to their molecular mass and those of data banks. Their anti-Legionella and haemolytic activities were measured. These peptides were, for the most part, already described. S. epidermidis PSMs were described to contribute to the systemic manifestations of Gram-positive sepsis (5). S. haemolyticus was described to produce 3 peptides which described to be gonococcal growth inhibitors (2). The SLUSH peptides of S. lugdunensis (1) and delta-hemolysin of S. aureus (4) are known to cause haemolysis. However, this is the first demonstration of their anti-Legionella activity. This study led us to select one peptide, PSMα produced by S. epidermidis 567, which displays not only an anti-Legionella activity which is comparable to that of Warnericin RK, but also a haemolytic activity which is about ten times lower than that of Warnericin RK. We are currently studying the effects of structural changes of this peptide and the Warnericin RK on their activities, in order to design an ideal anti-Legionella peptide without haemolytic activity, which could be used in therapeutic for example. References. 1. Donvito, B., J. Etienne, L. Denoroy, T. Greenland, Y. Benito, and F. Vandenesch. 1997. Infect Immun 65:95-100. 2. Frenette, M., R. Beaudet, J. G. Bisaillon, M. Sylvestre, and V. Portelance. 1984. Infect Immun 46:340-5. 3. Hechard, Y., S. Ferraz, E. Bruneteau, M. Steinert, and J. M. Berjeaud. 2005. FEMS Microbiol Lett 252:19-23. 4. Kreger, A. S., K. S. Kim, F. Zaboretzky, and A. W. Bernheimer. 1971. Infect Immun 3:449-465. 5. Mehlin, C., C. M. Headley, and S. J. Klebanoff. 1999. J Exp Med 189:907-18. 6. Verdon, J., J. M. Berjeaud, C. Lacombe, and Y. Hechard. 2008. Peptides 29:978-84.
- 244 - Session 8: Legionella prevention and control P159
Control of Legionella Colonization and Effects on Biofilm in a Hospital Water System Treated With Monochloramine
Isabella Marchesia, Patrizia Messib, Immacolata Anacarsob, Stefano Cencettic, Patrizia Marchegianoc, Giusep- pina Frezzaa and Paola Borellaa aDepartment of Public Health Sciences University of Modena and Reggio Emilia, Via Campi 287, 41100 Modena, Italy; bDepartment of Biomedical Sciences University of Modena and Reggio Emilia, Via Campi 287, 41100 Modena, Italy; cTeaching Hospital Policlinico of Modena, Via del Pozzo 71, 41100 Modena, Italy [email protected]
The control of Legionella contamination in water plants of big buildings may be a difficult challenge. Among disinfecting procedures, use of monochloramine already experimented for municipal waters is a promising method, and we evaluated their efficacy on planktonic and sessile populations of L. pneumophila contaminating a hospital water plant. A specific device aimed to distribute continues doses of monochloramine was installed on the hot water line, in order to obtain a concentration of approximately 1-2 mg/L at distal sites. At regular intervals, water and biofilm samples were collected to evaluate Legionella and Pseudomonas spp. contamination, total microbial count and amoebae presence using standardised cultural procedures. Before any treatment, L. pneumophila serogroups 9, 6 and 1 were isolated from hot water even at levels higher than 10 4 CFU/L. Treatment with continuous doses of monochloramine induced a 50% reduction in the number of contaminated sites just after one week. After one month, we documented the absence of legionellae in hot water collected at distal points before and after 1-2 min of flushing. The results were confirmed up to four months of treatment with monochloramine. Similar trends were registered for total count, whereas Pseudomonas spp. were recovered in some points, especially in the pre-flushing samples. Within biofilm, Legionella reduction after one week ranged between 98 and 100% and the reduction was substantially confirmed with time. When biofilm formation was studied on pipes treated with monochloramine since the beginning, we discovered a strong effect of monochloramine in limiting both total count and Legionella presence in comparison with non-treated pipes. No modification in the number of free-living amoebae was observed during the entire observation period. The proposed procedure shows a good efficacy in the medium period to control the planktonic Legionella concentration in a contaminated water distribution system of a big hospital. Additionally, an interesting reduction of legionellae associated with biofilm was observed. Studies comparing monochloramine with chlorine dioxide are in course, taking into consideration also the differences in terms of efficacy, selection of virulent strains, and pipes corrosion phenomena.
- 245 - Session 8: Legionella prevention and control P160
Gold nanoparticle interactions and impact upon a common biofilm source: Legionella pneumophila
Amber Stojaka, Stephen Klainea and Tamara McNealyb aClemson University, Institute for Environmental Toxicology, 509 Westinghouse Road, Pendleton, 29670, United States of America; bClemson University, Dept of Biological Sciences, 132 Long Hall, Clemson, 29634, United States of America [email protected]
There exists widespread concern of pathogenic bacteria colonizing and establishing biofilms in water systems thereby providing a direct exposure route of disease to the public. Legionellaspecies are widely distributed in all waters including, but not limited to, drinking and cooling systems, plant misters and spa pools. These warm water environments facilitate the colonization and establishment of biofilms containing Legionella. Bulk metals such as copper and silver have been attempted for use in removal of such biofilms but they have been unsuccessful in the permanent removal of biofilms. However, the physical properties of metal nanoparticles are unlike their bulk elemental or molecular form and behave based on their size, shape and surface chemistries. The objective of this research was to investigate interactions between gold nanoparticles (AuNPs) and Legionella pneumophila biofilms. This research evaluates planktonic virulence and growth kinetics, as well as biofilm structure, density, and composition. Transmission electron microscopy (TEM) micrographs indicate that after biofilm exposure to AuNPs, the nanoparticles integrate into the exopolymeric matrix and the cell surface and are taken up into the cell. However, there was no significant difference in viability of planktonic cells after exposure to the AuNPs. Biofilms also showed an alteration of morphology, as seen in confocal microscopy analysis, after exposure to AuNPs on both glass and silicon substrates. In treated biofilms L. pneumophila formed long chain-like assemblages in contrast to a thick, structured, densely packed, biofilm as seen in controls. Control experiments using the same size but polystyrene instead of gold spheres, showed no difference in morphology from untreated biofilms. This research suggests that planktonic L. pneumophila viability is not affected by exposure to AuNPs. However, L. pneumophila biofilms are morphologically altered and potentially destabilized as they are reduced in size and structure. This effect, specifically from AuNPs, may provide the potential for a more aggressive biocidal administration protocol in water cooling and recirculating systems if destabilization of biofilms can result in the use lower biocide concentrations for treatment but greater efficacy.
- 246 - Session 8: Legionella prevention and control P161
Relevance of indicators for legionella monitoring in cooling systems
Michele Merchata, Corinne Peyrona and Gilles Chaperonb aClimespace, 185 rue de BERCY, 75012 Paris, France; bCAPSIS, 1 rue terre neuve, 91940 Les Ulis, France [email protected]
Various treatments are implemented to minimize the Legionella concentration in the water cooling systems. In order to control their effectiveness, Legionella analyses are performed. But, new, indicators are also used, with the same unique objective of ensuring the control of Legionella. But, an analysis of water is not an end in itself, it can take a decision. If decisions are wrong, the consequences can be very important in terms of health, economic, environmental and media. The aim of this study is to compare in a pilot plant and industrial facilities of cooling system various indicators including, Legionella culture and PCR, ATP, mesophilic aerobic bacteria (MAB), amoebae culture or biofilm measurements, according to different water treatment strategies. In order to compare the analysis results, 10 litters of water are weekly sampled, brought to the laboratory to be homogenized and distributed in different vials before specifics analysis. For ATP analysis, 50 ml of water were taken from five points defined on the circuit. Monitoring is conducted under different conditions: colonization by the biofilm (without treatment), disinfected by shocks, chemical cleaning. Results show that whatever the indicator, the result of the measurement depends on the representativeness of sampling, relevance of analysis but also of water quality and water treatment strategy. Then it’s impossible to compare methods of analysis with sample collected from different facilities and without any information on the treatment. The monitoring of MAB does not seem to be a useful indicator because it’s spatial and temporal variation within a circuit. ATP is a useful indicator of water quality if the technique used is reliable, but that does not in itself guarantee the absence of Legionella. PCR is not a useful indicator on all types of water, and a misinterpretation often leads to an overestimation of risk and corrective actions inappropriate (especially Legionella species). Then, the choice of indicators and their interpretation must be made according to an objective, the type of installation, quality of water and chemical treatment strategy. The drift of a parameter is not necessarily correlated with a Legionella risk and should not induce a curative disinfection.
- 247 - Session 8: Legionella prevention and control P162
Optimal maintenance for outdoor ornamental fountains to avoid potential dissemina- tion of Legionella aerosols
Rosa Minarro˜ Del Morala, Ana Rubio Garciab and Jose Luis Guijarro Rodriguezc aAndalusian Public Health System, Enmedio 9, 14004 Cordoba, Spain; bAndalusian Public Health System, Mateo Inurria 9, 14001 Cordoba, Spain; cIBIS - Biomedical Research Institute of Seville, Mateo Inurria 9, 14001 Cordoba, Spain [email protected]
Objective: To assess the risk for Legionella Pneumophila contamination related to 86 outdoor ornamental fountains with previous maintenance program in the City of Cordoba. Methods: Design: Transversal study carried out at the City of Cordoba´ (Spain) during 2008 including all ornamental fountains registered. Assessment: Application of the algorithm included in the Technical Guide published by the Spanish Ministry of Health, that considers the following risk factors: • Structural Factors: water source, materials, aerosol generation, location; • Maintenance Factors: biocides, contamination, hygienic and mechanical status, treatment systems; • Operational Factors: water and system temperatures, renewal and filtration. The algorithm leads to obtain a Global Weighted Index from the results of the assessment of the three factors. This Global Index provides a summary of the risks that can be useful to determine the need of additional corrective measures in a more efficient way taking into account the specific characteristics of the facilities. Results: From the total 86 fountains registered, 20 were initially labelled as ”without aerosol generation potential” and no further investigation was performed. The other 66 fountains were included for the full assessment of the risk factors algorithm. No risk was detected for 63 fountains according to the Technical Guide. Two of the fountains needed corrective measures and just one of the fountains was candidate to be closed or to face important operational changes. Conclusions: Most of the fountains have aerosol generation potential and as a consequence they could disseminate legionella. However, most of them do not exhibit any risk as potential infecting sources due to their structural, maintenance and operational characteristics. Operating rules in our Autonomous Community for these facilities establish disinfection with 1-3 ppm Free Residual Chlorine. It is difficult to get those levels due to the volatility and deleterious effect of this biocide. Our results support a new approach to reconsider the use of such high concentrations of Chlorine or, at least, to limit its use just for those fountains at real risks. Downsizing Chlorine uses is an environmental challenge to reinforce. The risk assessment performed in our study calls attention to the priority of surveillance, monitoring and control of these facilities instead of general high-level chlorine disinfection approach.
- 248 - Session 8: Legionella prevention and control P163
The hospital tap waters: a risk of Legionella contaminated aerosol inhalation
Monique Reyrollea, Caroline Landelleb,Celine´ Mazurec, Marie-Christine Nicolleb, Sophie Jarraudd,Jer´ omeˆ Etienned and Philippe Vanhemse aCentre National de ref´ erence´ des Legionella, Centre de Biologie Est, Groupe Hospitalier Est, 59 Boulevard Pinel, 69677 Bron, France; blaboratoire hygiene hopital edouard herriot, place d’arsonval, 69008 lyon, France; cBioRad, 3 avenue Raymond Poincare,´ 92430 Marnes la Coquette, France; dCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; eService d’Hygiene` Epidemiologie´ et Prevention,´ Hopitalˆ Edouard Herriot, 5 Place d’Arsonval, 69003 Lyon, France [email protected]
Background: Hospital-acquired legionellosis is recognized to be a severe pneumonia with a mortality rate of 20% for immunocompromised patients. The contaminated hot water showers are the main source of contamination but the contaminated aerosol from taps remains few documented. A case of fatal nosocomial Legionella pneumophila infection due to exposure to contaminated water from a washbasin in a haematology unit was recently reported (Brulet A, et al., 2008, Inf Hosp Epid, 29, 1091). Our objective is an analytical approach to improve exposure assessment to Legionella in aerosols generated by tap waters. The aim of this study is to evaluate the contamination of the aerosol generated by colonized washbasin water. Methods: A colonized hospital tap was investigated. The water was running 30 min and 2 impingement technologies of aerosol collection were used: Impinger (Arelco) and CIP10M (Bio-Rad) air sampler. Each aerosol was collected in a DNA free buffer (BioRad) compatible with culture. The air collections were respectively 300 L in 3 mL for CIP and 360 L in 20 mL for Impinger. The water samples (WS), samples from 2 CIP and 2 Impingers were tested in each run and repeated 10 times. 20 WS and 40 aerosol samples were tested for Legionella using 2 methods: culture NFT90-431 and PCR XPT 90-471 using BioRad method (Aquadien kit, iQ-Check Legionella, Chromo4). The ATP value (Aquatools), Pseudomonas (NF EN ISO 16266) and microbial colonization were also analysed for 3 runs. Results: The WS of the tap were positive for culture (103-4 CFU/L Lp 1 and L. anisa) and for PCR (Lp and Lspp 104 GU/L). No cultivable Legionella were reported for the aerosol samples. 18/40 aerosol results were positive for PCR Lspp: 3-57 GU/well and 5/40 aerosol results were >DL and - 249 - Session 8: Legionella prevention and control P164 Prevention of Legionella and Pseudomonas colonisation for hot and cold hospital water systems: a one year experience of Waterclean°R system Monique Reyrollea, Rodolphe Beronb, Jerome Droguetb, Cindy Schurc, Georges Duboisd, Sophie Jarraude and Jer´ omeˆ Etiennee aCentre National de ref´ erence´ des Legionella, Centre de Biologie Est, Groupe Hospitalier Est, 59 Boulevard Pinel, 69677 Bron, France; bDAT hospices civils de lyon, 49 rue villon, 69373 lyon, France; cgroupe carso, 321 avenue jean jaures, 69362 lyon, France; dCHS de la savoie, bassens, 73011 chambery, France; eCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France [email protected] Background: Hospital-acquired legionellosis continue to be a common public health problem. The hot water system of the hospital is the main source of contamination and many reports related the difficulty of eliminating Legionella from the systems. Our objective is a preventive approach for the risk management. A treatment with elevation of temperature of cold and hot waters is a preventive action to avoid the contamination with Legionella and Pseudomonas. The aim of this study is a one year evaluation of the use of the Waterclean°R system in a renovated building of a specialized hospital. Methods: The renovation of the building (2 levels, 25 beds) consisted of a new water production with Waterclean Systems (Ciat) and a partial pipes replacement. The ”CATREL” procedure was applied. It defines actions and technical and analytical follow-ups to prevent the bacterial colonisation of the new water networks. As examples, survey of temperature was report in real time at Ciat Company. The water samples (WS) consisted in: 9 sampling points monthly analyzed from March 2008 to April 2009. For Pseudomonas: 36 WS tested using normative method NF EN 12780. For Legionella: 108 WS tested using 2 methods: culture NFT90-431 and PCR XPT 90-471 with Gene Systems technology. Results: The incoming water supply for hot water production and the untreated cold water (13-15◦C) were colonised with 102-3 UFC/L of L anisa. The hot water network treated with a raising of temperature up to 65◦C was negative for Legionella culture and PCR Lp <830 UG/L (QL) for 18/24 samples. A test room with treated waters (hot and cold) showed Lp 5 and L anisa positive culture in 12/24 WS. The other 24 sampling points of treated water were culture negative. The PCR Lspp results were positive for 93% samples with some values up to 106 UG/L. The culture of Pseudomonas was negative in 31/32, only a mix water of a shower was positive (630 UFC/L). The use of a thermal chock (70◦C) in July 2008 and March 2009 with Waterclean system induced negative results (culture and PCR Lp) for a month. Conclusions: The preventive use of Waterclean system could lead to a safe water network. The partial renovation of pipes and the ”partial” application of the CATREL procedure could explain the colonisation of some sample points. The curative action of Waterclean was useful to make the showers and the taps in the patient’s room safe. For a new hospital building, the hot temperature treatment for all the water supplies seems to be the best option so that hot and cold water is controlled and safe. Thanks for technical and financial supports: R.Meskel Caleffi, P.Gros, P.Couturier Ciat. - 250 - Session 8: Legionella prevention and control P165 Legionella in Campania: report of the year 2008 Anna Maria Rossia, Trofimena Lucibelloa, Mariangela Paganoa, Francesca Di Leoa, Antonio Coppolab, Anto- nio Petrosinoa, Giacomo Dentea and Giuseppe D’Antonioa aAgenzia Regionale per la Protezione Ambientale della Campania, Via Scavate Case Rosse, 84100 84100, Italy; bAgenzia Regionale per la Protezione Ambientale della Campania, Via Scavate Case Rosse, 84100 Salerno, Italy [email protected] Background: Since 2001 Regional Agency Environmental Protection (ARPAC) held the control on the spread of Legionella spp in Campania, by means the Regional Reference Laboratory for Legionella (LRRL). Methods: LRRL performs environmental investigations either after notification of cases to identify the source of infections either for surveillance plans in some hospitals as recommended by Italian Guidelines (Years 2000-2005) for Preventions and Control of Legionnaire’s Disease (L.D.) - . Swabs, air and water samples were collected from several sites (taps, showers, thermal circuit, aerosol) Result: In 2008 LRRL has carried out environmental controls after 76 reported cases of L.D. 54 cases were diagnosed in Campania hospitals and 22 cases were travel-related about tourists who stayed in our Region (11 from Italy and 11 from abroad). We examined 108 sites: 18 health-care facilities, 33 receptive-structures (hotel, thermal farm, wellness centres, camping), 38 private houses, 19 work-places. . Other 64 sampling were performed concerning surveillance plans of 4 big Regional Hospitals. A total of 1577 samples were collected to evaluate Legionella contamination of the water systems. Our results showed that Legionella in Campania is very diffused. Hotels (36%) and hospitals (88%) seems to be the structure more contaminated. We found absolute prevalence of Legionella pneumophila (Lp), and mainly Lp1 and Lp6 were isolated. Conclusion: Since 2005 in Campania the clinical diagnosis of LD has largely increased by using Urine Antigen Testing. Although Legionella is now a well known problem, in our Region the prevention measures adopted are still inadequate to reduce the risk. It is necessary to standardize control procedures and to train hotel managers and medical staff about risk assessment and risk management. - 251 - Session 8: Legionella prevention and control P166 Evaluation of Legionella Prevention and Control Measures at high risk facilities in the Municipality of Cordoba´ - Spain, 2005-2008 Ana Rubio Garciaa, Rosa Minarro˜ Del Moralb and Jose Luis Guijarro Rodriguezc aAndalusian Public Health System, Mateo Inurria 9, 14001 Cordoba, Spain; bAndalusian Public Health System, Enmedio 9, 14004 Cordoba, Spain; cIBIS - Biomedical Research Institute of Seville, Mateo Inurria 9, 14001 Cordoba, Spain [email protected] Environmental protection is essential to the quality of life for present and future generations. The EU lines of priorities for reducing the adverse health effects associated with certain environmental factors, include ”To prevent the risks associated with the facilities at risk in the transmission of certain diseases such as Legionella”. OBJETIVE To evaluate the actions undertaken to prevent and control biocontaminant legionella pneumophila serogroup I, in cooling towers and evaporative condensers (TyCC) in the municipality of Cordoba METHODOLOGY Cross-sectional study at the municipality of Cordoba (South of Spain) considering the period from 01/01/2005 to 31/12/2008 RESULTS AND CONCLUSIONS The number of cooling towers and evaporative condensers increased during our study period from 158 to 219. Manteinance plans were designed for the 100% of these facilities acording to local rules in force. The number of evaluation visits undertaken were 170, 313, 238 and 245 in 2005, 2006, 2007 and 2008 respectively. The number of incident cases of legionella increased substantially in the period of our study raising from 3 to 4, 15 and 5 cases for a 300.000 inhabitants population. Although efforts have proved to be useful to unify criteria and to facilitate communication among stake holders (evaluators, facilities owners, institutions and maintenance companies) the ultimate goal of the health strategy, to reduce the incidence of the disease, has not be reached. TyCC are low energy consumption facilities that does not need chemical products as refrigerants and exhibit a great cooling power. Exhaustive evaluation visits, maybe under too strict criteria, have lead to the replacement of these facilities with other technologies less clean and efficient from an enviromental poit of view. There is a need to design different strategies to include controls for other facilities that, although not related to disease outbreaks, are associated to sporadic cases which is the type of cases that are detected in our municipality. - 252 - Session 8: Legionella prevention and control P167 The Legionella chip: practical experience with this novel analytical tool to analyze Legionella positive water samples Michel Ossendrijvera, Adrie Atsmab and Frank Schurena aTNO Microbial Genomics, P.O. Box 360, 3700 AJ Zeist, Netherlands; bVitens, Snekertrekweg 61, 8912 AA Leeuwarden, Netherlands [email protected] A novel diagnostic tool (The Legionella chip) for the culture independent analysis of water samples for the presence of Legionella bacteria has been developed. An analysis with the Legionella chip leads to the following results: - Presence of Legionella (total) - Presence of L. anisa - Presence of L. pneumophila - Presence of mixtures of L. pneumophila and L. anisa - Presence of mixtures of multiple strains of L. pneumophila - Discrimination between pathogenic and environmental strains of L. pneumophila The analysis is based on DNA which can be extracted either directly from water samples or from Legionella colonies. Results can be obtained within 4-6 hours. Results from analyses performed on a variety of samples including colonies, drinking water and process water samples will be shown. - 253 - Session 8: Legionella prevention and control P168 Recognizing mixtures of Legionella strains Adrie Atsmaa, Michel Ossendrijverb, Martien Caspersb and Frank Schurenb aVitens, Snekertrekweg 61, 8912 AA Leeuwarden, Netherlands; bTNO Microbial Genomics, P.O. Box 360, 3700 AJ Zeist, Netherlands [email protected] Discrimination between pure bacterial cultures and mixed bacterial cultures is an underexplored field in microbiology. The standard microbial approach is to incubate a sample on defined culture medium and look for bacterial growth. In case of growth on agar plates single colonies can be retrieved and used for further analysis. When starting with a pure culture there is no problem but when the starting material contains two or more different strains (or species) the situation is more complex. In case there is no different colony morphology chances are small that the mixture will be recognized. Furthermore there is a possibility that one of the strains does not grow at all or much slower thereby escaping attention as well. Especially with slow growing bacteria such as Legionella there is a reasonable chance that such a situation may occur. To overcome this problem we have developed a molecular approach to recognize the presence of multiple strains in a (water) sample. This approach is based on the presence of so-called incompatible markers being present in the same sample. Incompatible markers are DNA markers that are never present in the same strains. One example of such a marker pair are the DNA markers representative for either serogroup 1 or serogroups 2-14 of L. pneumophila. Identification of incompatible marker pairs based on genomic analyses will be presented. - 254 - Session 8: Legionella prevention and control P169 Application of ribotyping for the identification of the infection source for a legionellosis case Alberta Stenicoa, Margit Seebera, Paul Huberb, Armin Oberlechnerb, Gabriele Spitalera, Ludwig Moroderc and Patrizia Rossic aA.P.P.A. Bolzano, Biological Laboratory, Via Sottomonte, 2, 39055 Laives, Italy; bAzienda Sanitaria dell’Alto Adige - Servizio di Igiene e Sanita pubblica, Vicolo dei Frati 3, 39031 Brunico, Italy; cAzienda Sanitaria dell’Alto Adige - Laboratorio Aziendale di Microbiologia e Virologia, Via Amba Alagi, 5, 39100 Bolzano, Italy [email protected] In December 2008 legionellosis was diagnosed in a 42 year old male patient, smoker and professional lifeguard. The first diagnosis was made using the urine antigen test, later a Legionella pneumophila serogroup 1 (sg1) was isolated from the patient’s sputum. The day after the hospitalization the local Hygiene and Public Health Service intervened for epidemiologic ascertainment. As possible infection source the warm water supply of the patient’s working place, a public swimming pool build three years before, was soon made responsible, but also the patient’s lodging was not excluded, a room in a small inn, even if at first this appeared a less probable infection source. Priority was given to the patient’s working place considering the high number of people visiting the public building and the consequent risk to which the population was exposed. Samples were taken from several warm water supply sources in the swimming pool, followed by an early thermal disinfection even before analysis results were available. In the meantime warm water samples were taken also from the shower at the patient’s lodging. In all samples, analyzed according to the method ISO 11731:1998, Legionella pneumophila sg1 was isolated. In the water samples taken in the swimming pool facilities the concentration ranged from 550 to 13000 CFU/L, while in the two samples taken from the shower room of the patient’s lodging the concentration was of 3400 CFU/L and 1500 CFU/L. An automated ribotyping analysis was performed on the isolated strains of Legionella pneumophila sg1 using a RiboPrinter Microbial Characterization System instrument (Qualicon, USA) and applying the EcoRI restriction enzyme. Other than expected, the results pointed out the similarity of the bacterial strain isolated from the patient to the strain isolated from the room’s shower samples, while all strains isolated from the swimming pool samples were genetically different. These results lead to several structural interventions both at the inn and at the public swimming pool, even if the latter was not directly involved as infection source for the described case. - 255 - Session 8: Legionella prevention and control P170 A Risk Management Approach for Preventing or Controlling Health Risks from Legionella bacteria William Thomas and Roger Shrapnel ISHEM Consulting, 43 Temple Row, B2 5LS Birmingham, United Kingdom [email protected] Legionella bacteria are ubiquitous in the natural environment where they pose no significant risk to health. In man-made water systems health risks from Legionella bacteria can range from negligible to very high risk depending on various factors including those affecting seeding, growth, aerosol generation, virulency and exposure and susceptibility of the exposed population. Assessing risks for water systems is a pre-requisite to determining the control measures required to enable their safe operations and also to comply with national health and safety regulations to prevent or control exposure to a hazardous biological agent. An approach and methodology is described which has been used successfully to manage risks of exposure to Legionella bacteria from a wide range of water systems. This is based on applying a risk management process including identifying and assessing the risk, eliminating, prevent or controlling any residual risk, and monitoring and reviewing the performance of the risk mitigation measures. Risks are assessed by considering the risks of exposure of Legionella bacteria and the likelihood of those individuals contracting Legionellosis by considering the potential risks for each water system of: - bacteriological seeding and growth; - aerosol generation, release and exposure; and - bacterial virulency and host susceptibility. After determining the SYSTEM RISK and considering any existing preventative measures to mitigate risk the water system is assigned a RESIDUAL RISK rating. Systems are assigned a low residual risk if the system risk is inherently low due to system design and/or operating regime or, for systems where the system risk is medium to high risk, if the risk control measures and management arrangements are considered adequate to mitigate system risks. For systems with a medium to high residual risk rating additional preventative measures will be required to ensure their safe operation. The development and implementation of a system-specific risk control scheme needs careful consideration of what risk mitigation measures are required for safe operation of each water system. These risk control measures need to be closely monitored and reviewed so that their overall performance can be assessed on an ongoing basis. Various risk management strategies are discussed and examples given of risk mitigation schemes applied to a range of industrial water systems. - 256 - Session 8: Legionella prevention and control P171 The Role of Corporate Standards and Best Practice Guidance for Managing Health Risks from Legionella bacteria in Multinational Organisations William Thomas, Roger Shrapnel and Kerry Horton ISHEM Consulting, 43 Temple Row, B2 5LS Birmingham, United Kingdom [email protected] Legionnaires Disease is recognised as a global issue but standards and best practice can vary from country to country. This is particularly evident when comparing acceptable (safe) and unacceptable (unsafe) concentrations of Legionella bacteria in water systems. For example the UK, European (EWGLI) and North American guidance recommends action level thresholds of 1, 10 and 100 colony forming units per millilitre respectively. Many multinational organisations whilst mindful of national and local regulations prefer to operate to a Corporate Standard and Best Practice Guide to ensure uniformity and consistency of approach across their business operations. This can be beneficial in ensuring a minimum standard is achieved and demonstrating a duty of care to all employees, but also challenging in parts of the world where local service provision and facilities do not meet the usual standards. This paper provides a synopsis of current regulations, standards and best practice guidelines for preventing and controlling exposure to Legionella bacteria in several countries, including those in Africa, Americas, Asia, Australasia and Europe. It discusses corporate strategies and approaches to setting standards and best practice guidance, considers minimum performance criteria and metrics for measuring compliance and draws on the experience of the authors in implementing and monitoring of such programmes. - 257 - Session 8: Legionella prevention and control P172 Legionella risk monitoring in a hospital water network through control of bacterial proliferation using a new generation of Quantitative ATP-metry technology Marc Raymond and Veliana Todorova AQUA TOOLS, 36 rue de la Falaise, 78126 AULNAY SUR MAULDRE, France [email protected] Progressive decrease in water quality is a favorable factor for increase of Legionella concentration in water networks, linked often to biofilm or amoeba. For minimizing Legionella risk in hospitals, operators need tools allowing better reactivity to detect degradation of water quality, identify possible Legionella proliferation locations and implement immediate corrective actions without waiting for culture results. New generation of bioluminescence method allow precise quantification of viable biomass in water samples and biofilm. Quench-GoneTM Aqueous kit (LuminUltraTM (Canada), supplied in Europe by Aqua-tools was used during Legionellosis cases investigation in a Paris hospital. This test quantifies Intracellular Adenosine Triphosphate (ATP) molecule contained only in living, cultivable or non cultivable, microorganisms. Results are delivered in pg ATP/ml & equivalents microorganism within 5 min. The method is used is microbial risk assessment, where, after a biological mapping of the installation, critical points are identified and installation- turned treatments can be applied and verified immediately, without long delay for results. Two cases of Legionellosis occurred in one Paris hospital. For immediate determination of the critical contamination points in the hospital, the water sanitary network was mapped for proliferation of active biomass using Quench-GoneTM Aqueous test. Parallel real time PCR Legionella spp. measurements were performed for characterization of the bacterial population. The investigation confirmed presence of important biomass proliferation, associated with high Legionella species contamination, only at the terminal points of the room of question. No global disinfection of the water network was required. After disinfection of the terminal points, the efficiency of the treatment actions could be verified within minutes. Our investigation using this method show promising results for fast identification of critical points for Legionella proliferation in sanitary water networks. The kits can be used for biological risk assessment and monitoring corrective actions and efficiencies of treatment strategies against proliferation of Legionella. - 258 - Session 8: Legionella prevention and control P173 Evaluation of automated nucleic acids extraction system for Legionella pneumophila detection in water samples Michael Treillesa, Delphine Espereta, Nadege Daguiera and Laurent Thieryb aLDA 50, Route de Bayeux, 50008 SAINT LO, France, Metropolitan; bQIAGEN S.A., 3 Avenue du Canada, LP 809, 91974 COURTABOEUF CEDEX, France, Metropolitan [email protected] The increasing number of deaths due to Legionella disease has resulted in growing public interest in problems associated with intra-cellular bacteria. As these bacteria are present in water, they can be found in several artificial ecosystems, like private boilers, air conditioning units, nuclear plants etc... Real-time PCR assays for Legionella have shown the potential for rapid, sensitive and specific detection of this pathogen. In France, a normalization standard (AFNOR XP T90-471) regulates Legionella pneumophila detection by real-time PCR. The most critical part of real-time PCR analysis is reliable and efficient extraction and purification of bacterial DNA from water samples. The aim of this study was to evaluate an automated system using magnetic beads, the QIAGEN EZ1 system, and determine performance in terms of yield carrying out validation according to the XP T90-471 standard request. - 259 - Session 8: Legionella prevention and control P174 Development of a real-scale test rig for Legionella elimination in biofilms in hot water distribution systems: thermal and chemical treatment evaluation Maha Farhata, Marie-Cecile´ Trouilhe´a, Marina Moletta-Denata, Enric Robinea, Christophe Foretˆ b and Jacques Frere` c aCentre Scientifique et Technique du Batiment,ˆ 84 avenue Jean Jaures,` 77447 Marne-la-Vallee,´ France; bBKG Water Solutions, 43 route de Ruaudin, BKG France SAS, 72230 Arnage, France; cUniversity of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France [email protected] Legionellosis, an acute pneumonia caused by bacteria of the genus Legionella (mainly L. pneumophila), kills 10% of patients. In France, the disease is notifiable and, in 2008, it has resulted in the average of one death every three days (1243 cases including 119 deaths). The management of Legionella risk is therefore a health issue of prime importance. Legionella are mainly encountered in biofilm which is present at the surface of pipes in cooling towers and hot water distribution systems (HWDS). Various disinfection procedures are used to eliminate this bacteria but none of them can eradicate it totally. Moreover, no reproducible and comparative method has been developed to evaluate the intrinsic efficiency of treatments against Legionella in biofilm. For this purpose, a pilot-scale 1 was constructed (CSTB, France) in order to simulate a real HWDS. It consisted in two identical loops entirely made of stainless steel. One of the two loops was used as a control loop whereas the other one was submitted to the treatment under assessment. This pilot-scale was contaminated with Legionella by water coming from a local river and from a cooling tower. Biofilm contained about 1 000 000 bacteria/cm2 and 1 000 CFU/cm2 of Legionella spp. Different treatments were tested and for each of them the following analyses were performed on tested and control water and biofilm: - Conventional culture: to enumerate culturable Legionella and culturable heterotrophic bacteria (GVPC and R2A counts respectively); - Real time PCR: to quantify Legionella spp. and Legionella pneumophila, (PCR L. spp. and PCR L. pneumophila counts respectively); - Microscopy by epifluorescence: to quantify the total cell density (DAPI counts); - Finger print methods: to study population dynamics and the microbial consortium accompanying Legionella. Two heat shock treatment sessions (70◦C for 30 minutes) were successively implemented under similar conditions in the treated loop. The first test showed only an effect on GVPC and R2A counts. Legionella spp. was no more detectable 24 hours later and then reached the initial concentration 48 and 72 hours after the treatment in water and biofilm respectively. The second test showed an observable effect only on GVPC counts in water where Legionella spp. increased provisionally by 2 log units after 48 hours. Two chemical treatment sessions (H2O2 combined with peracetic acid) were applied in a similar way to the treated loop. The tests showed a provisionally effect on DAPI, R2A, PCR L. spp., PCR L. pneumophila, and GVPC counts. According to the first test, Legionella spp. was no more detectable 24 hours after the treatment in water and biofilm. Then, the first Legionella spp. reappeared after 72 and 168 hours in water and biofilm respectively. The second test showed the same results in biofilm while legionella was undetectable anymore after 24 hours in water. The relationship between Legionella and amoeba species will be study in future works in order to better understand these results. Finger print analyses by SSCP (Single Strand Conformation Polymorphism) are still underway. The main results will be presented at this conference. - 260 - Session 8: Legionella prevention and control P175 Comparison of the SANIPACKING°R Cooling Tower Fill Material against Standard Polypropylene Fill Material in Recirculating Model Water System Irfan Turetgen¨ a, Nazmiye Ozlem Sanli Yurudua and Imke Nordenb aIstanbul University, Faculty of Science, Department of Biology, Vezneciler, 34134 Istanbul, Turkey; bGEA 2H Water Technologies GmbH, Kalscheurener Str. 92, 50354 Hurth,¨ Germany [email protected] SANIPACKING°R fill material and standard polypropylene (PP) fill material was compared in terms of biofilm formation potential and anti-Legionella activity within 4-months period using laboratory-scale recirculating water system. The experimental study was performed using a 100-liter polypropylene lab-scale recirculating model system under constant hydraulic conditions, which correspond more with the situation in cooling tower installations. Recirculating water system was experimentally infected with Legionella pneumophila standard strain (ATCC 33152) suspension (1 ml inoculum of L. pneumophila 105 cell.ml−1) and operated continuously until all experiments had been completed. Rather low final inoculum was provided to the model system to mimic the natural entry of L. pneumophila from supply water. No chemicals (disinfectant, pH regulators or anti-scaling agents) were added to the system, to exclude their effects on microorganisms and biofilm formation. Results showed that L. pneumophila were rapidly multiplied in a short time during the study within the model system. Depending on the L. pneumophila culture results on BCYE agar, no L. pneumophila was isolated from the SANIPACKING°R surface within the four-month period, whereas intensive colonization (>106 CFU.cm−2) occurred on standard PP material surface beginning from the first month. Heterotrophic bacterial counts on surfaces showed that significantly low accumulation recorded on SANIPACKING°R surface in comparison to standard PP material. Total bacterial counts on surfaces determined by epifluorescence microscopy (using DAPI) and revealed that significantly low counts of microorganisms are colonized on SANIPACKING°R surface. It could be concluded that SANIPACKING°R surface has an effective anti-Legionella activity as a fill material and has been significantly less colonized by surface-associated biofilm bacteria. - 261 - Session 8: Legionella prevention and control P176 Membrane composition is involved in high sensitivity to warnericin RK peptide Julien Verdona,Jer´ omeˆ Labanowskia, Tobias Sahrb, Christian Lacombea, Thierry Ferreirac, Carmen Buchrieserb, Jean- Marc Berjeauda and Yann Hecharda aUniversity of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France; bInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; cFRE CNRS 3091 Physiologie moleculaire´ du Transport des Sucres, 40 avenue du recteur pineau, 86000 Poitiers, France [email protected] Warnericin RK is the first antimicrobial peptide described to be active against Legionella pneumophila, a pathogen bacterium responsible for severe pneumonia. Strikingly, this peptide displays a very narrow range of antimicrobial activity, almost limited to the Legionella genus, and a hemolytic activity. We showed that warnericin RK is an amphiphilic alpha- helical peptide, which possesses a detergent-like mode of action. As warnericin RK is a membrane active peptide, we thus hypothesized that a specific feature of Legionella’s membrane is the key to explain its specific sensitivity. In literature, Legionella is described to possess a very particular membrane as it contains high amount of branched-chain fatty acids (BCFA) and phosphatidylcholine, an eukaryotic membrane component. An adapted strain, able to grow at a concentration 32 fold higher than MIC (i.e. 3.12 µg/ml), was isolated after repeated transfers of L. pneumophila with increasing concentrations of warnericin RK. As we assumed that Legionella membrane composition can modulate the activity of warnericin RK, fatty acids were identified by GC-MS and phospholipids were identified by direct electrospray inonization-mass spectroscopy (ESI-MS) and thin layer chromatography (1D TLC). Fatty acids analysis revealed that both stationary phase cells and adapted variant possess higher amount of short chain fatty acids (SCFA). Furthermore, the percentage of BCFA was significantly higher in the adapted variant than in WT cells. Phospholipids composition confirmed that Legionella possess only PC, PE, PG and CL as major phospholipids. Moreover, analyses showed that both the WT and adapted strains have similar composition, suggesting that the adaptation was not due to a change in phospholipids composition. A transcriptomic analysis of the WT and adapted strains was conducted in order to find genetic clues about resistance to the peptide. Microarrays results showed that, in the adapted strain, several genes implicated in membrane homeostasis were repressed. Taken together, the results demonstrated that BCFA and SCFA were more abundant in the adapted strain compared to the WT strain. Thus, Legionella specific fatty acids profile seems to play a critical role in the sensitivity to warnericin RK. - 262 - Session 8: Legionella prevention and control P177 Distribution of Legionella pneumophila serogroups and monoclonal antibody subgroups in environmental isolates from Russia (2005-2009) Olga L. Voroninaa, Marina S. Kundaa, Vera V. Bitkinaa, Vladimir G. Lunina, Oksana V. Sadretdinovaa, Tatiana I. Karpovaa, Paul Ch. Luckb and Igor S. Tartakovskiya aN.F. Gamaleya Institute for Epidemiology and Microbiology RAMS, Gamaleya str., 18, 123098 Moscow, Russian Federation; bTU Dresden, Institute Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany [email protected] Detection of potential virulent Legionella pneumophila strains is one of the important task of the artificial water systems monitoring. We used intarnationally standardised panel of reagents - Dresden panel for L. pneumophila environmental strains phenotypic typing. We demonstrated that most of the environmental strains from Russia (2005-2009) belonged to serogroup 6. Next in amount is a group of strains with serogroup 1. Most common strain was mAb ”Oxford”, that agree with date of enviromental isolates from England and Wales (2000 - 2008). 5 strains were mAb ”Benidorm” and were potential virulent. Such strains were commen for cooling towers and autonomous water system with stagnate water. We observed also strains with mAb 2, 3, 5, 7, 8, 9,10, 11, 12, 14. All strains we divided into 18 phenotyps with index discrimination - 0,91. So we demonstrated high efficiency of Legionella pneumophila environmental strains diversity by Dresden mAb panel. - 263 - Session 8: Legionella prevention and control P178 The use of windscreen wiper fluid without added chemicals in cars and commercial vehicles: A newly identified risk factor for Legionnaires’ disease Anders Wallenstena, Isabel Olivera, Kate Rickettsb, George Kafatosb, Nick Phinb, James Stuarta and Carol Josephb aHealth Protection Agency South West, The Wheelhouse, Bonds Mill, GL10 3RF Stonehouse, United Kingdom; bHealth Protection Agency Centre for Infections, 61 Colindale Avenue, NW9 5EQ London, United Kingdom [email protected] A source of infection is rarely identified for sporadic cases of Legionnaires’ disease. Among sporadic cases in the UK, professional drivers are five times more commonly represented than expected. We therefore investigated possible risk exposures in relation to driving or spending time in a car or other vehicle. A case control study including all surviving sporadic cases in England and Wales between 12 July 2008 and 5 March 2009 was carried out. Cases were contacted by phone and controls were consecutively recruited by sequential digital dialling matched by geographic area, sex and age group. Those who consented were sent a questionnaire asking questions on driving habits, potential sources in vehicles and known risk factors. The results were analysed with multivariable analysis. 75 cases and 67 controls were included in the study. Among those spending time in a vehicle 15/50 cases did not add any chemical to their windscreen wiper fluid while only 1/58 controls did not do so. Preliminary results for this exposure using a multivariable logistic regression model that included significant known risk factors showed of an association with being a Legionnaires’ case OR 21.4, 95%CI (1.7-264.7), p 0.02. Final data will be presented at the conference. This study has possibly identified a biologically plausible new risk factor for infection with Legionnaires’ disease. The bacteria could grow in the stagnant water of the windscreen wiper fluid reservoir, which is irregularly aerosolised onto the windscreen, and may then enter the vehicle through the ventilation system. This is an important public health finding. A simple recommendation to add chemicals to the windscreen wiper fluid may mitigate transmission of Legionella bacteria to drivers and passengers. - 264 - Session 8: Legionella prevention and control P179 Rapid detection of Legionella pneumophila in aquatic environment by using a microfluidic device Nobuyasu Yamagauchi, Yuko Uebayashi, Masashi Torii and Masao Nasu Graduate School of Pharmaceutical Sciences, Osaka University, 1-6, Yamada-oka, Suita, 565-0871 Osaka, Japan [email protected] Background: Rapid and simple methods to enumerate Legionella pneumophila in aquatic environments are essential for monitoring water quality and prevention of the outbreaks of Legionella infections. Conventional plate counting requires more than one week to obtain reliable results. PCR-based method is sensitive and widely used while it requires extraction of DNA from samples. In this study, we applied our microfluidic device to directly enumerate L. pneumophila in aquatic environment. Methods: L. pneumophila serogroup 1 (JCM7571) and cooling tower water samples were used in this study. A microfluidic device (size: 48 mm × 23 mm) was designed and fabricated by polydimethylsiloxane and a thin glass sheet. L. pneumophila cells spiked in cooling tower water were stained with anti-L. pneumophila fluorescent antibody in microchannel of the device. A fluorescence microscope-based system was used for counting of bacterial cells. The number of L. pneumophila cells in the samples was determined by both this microfluidic system and conventional fluorescence microscopy for comparison. Results: The number of L. pneumophila cells determined by the microfluidic device was correlated with microscopic count. The lower detection limit of the system was 5 × 104 cells/ml and the number of L. pneumophila cells in cooling tower water could be determined after concentration. Differences in quantitation of L. pneumophila cells between the microfluidic system and microscopy were less than 30% on average. The time required for quantitation of these bacteria by microfluidic devices was 10 - 30 min per sample. Conclusion: The microfluidic system fabricated in this study allowed rapid counting of L. pneumophila cells. This method may contribute to reduce the risk of the outbreaks of Legionella infections. This study was in part supported by the JSPS Grant-in Aid for Scientific Research (A) (21256002). - 265 - Session 8: Legionella prevention and control P180 Quantification of Viable Legionella pneumophila Cells using Propidium Monoazide combined with real-time PCR M. Adela Yanez˜ , Lorena Mart´ınez, Elena Soria, Raquel Murtula´ and Vicente Catalan´ LABAQUA, S.A., Pol. Ind. Atalayas, c) Del Dracma, 16-18, 3114 Alicante, Spain [email protected] All the culture-based methods applied to the analysis of Legionella require long incubation times due to the slow growth rate of the bacterium, they do not permit the detection of viable but non-culturable bacteria (VBNC), what still represents a public health hazard, and it is difficult to isolate legionellae in samples containing high levels of other microbiota. Numerous PCR-based methods have been developed to detect and quantify Legionella as an alternative to culture isolation since they have higher specificity, sensitivity and rapidity. However, the main drawback of these PCR-based methods is the difficulty in differentiating between live and dead bacterial cells, due to the persistence of DNA in the environment after cells have lost viability. In this work, a rapid detection method for viable Legionella cells combining propidium monoazide (PMA) and qPCR has been developed. This compound specifically intercalates and cleaves the genomic DNA from dead cells that are subsequently unable of being amplified by PCR. The capacity of PMA-qPCR to differentiate between L. pneumophila viable and non viable cells was studied using inactivated cells by a heating treatment at 95◦ for 5 min. The combined PMA-qPCR has demonstrated a signal reduction of 4.4-log in samples containing dead cells. In samples containing different viable/dead ratios, the use of PMA-PCR permitted the quantification of viable cells. The combined assay has been tested in 40 water samples from cooling towers in parallel with culture isolation method. The obtained data suggest the usefulness of the method for the specific detection of culturable and viable Legionella cells. In conclusion, the use of PMA prior qPCR is a promising technique, easy to be performed and with a great impact on DNA-based diagnostics of L. pneumophila in environmental samples. - 266 - SESSION 9 EPIDEMIOLOGY: A) NOSOCOMIAL INFECTIONS AND LEGIONELLA CONTROL IN HOSPITALS - 267 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P181 Usefulness of flow cytometry to assess the effect of heat treatment on Legionella in water systems. Severine´ Allegra, Franc¸oise Berger, Philippe Berthelot, Florence Grattard, Bruno Pozzetto and Serge Riffard Universite´ Jean Monnet, Faculte´ de Medecine,´ 15 rue Ambroise Pare,´ 42100 SAINT-ETIENNE, France [email protected] Background: Legionella (Legionnaires’ disease agent) are widespread in natural or man-made aquatic habitats. Heat- shock (30 min at 70◦C, WHO recommendation) is one of the treatments used to control the multiplication of Legionella, particularly in hospitals plumbing systems. Despite the apparent negative results obtained by the reference cell culture protocol, Legionella persist and re-colonize the ecosystem sometimes within a few weeks or even less. Depending on the environmental conditions, bacteria exhibit different viable physiological states, correlated or not with virulence. A flow cytometric assay (FCA) has been set up, to monitor Legionella viability while visualizing viable but non culturable (VBNC) cells. Legionella membrane integrity is evaluated by a Syto 9-PI (propidium iodide) double staining and is compared to Legionella growth on conventional plate culture media. Our results have demonstrated that FCA is a valuable tool to visualize VBNC cells of Legionella. Materials and Methods: FCA was used in an attempt to define heat-shock resistance patterns. 14 Legionella strains, isolated from 14 sampling points of a French hospital water system, were submitted to a 30 min heat-shock at 70◦C. Three of them (L. pneumophila sg1 strains A, B and C), epidemiologically- unrelated (as demonstrated by AP-PCR) were isolated several times (8, 8 and 13 times, respectively), from 1994 to 2008. Results: For 7 of the 14 strains, the FCA still detected 10 to 25% of VBNC cells which were able: (i) to produce ATP and; (ii) to be resuscitated after culture on amoebae. Between these 14 strains, L. pneumophila sg1 A, B and C strains exhibited also three different flow cytometric profiles. Responses to the 30 min heat treatment, in terms of number of VBNC generated were 10±3% for A, 59±8% and 53±4% for B and C respectively. Re-colonization of the hospital water system was more efficient by the strains B and C. Conclusion: Flow cytometry is an appropriate tool to predict the effect of heat-shock on water systems contamination with Legionella. In order to prevent nosocomial risks associated with these bacteria through the anticipation of their colonization capabilities, analyses of L. pneumophila sg1 VBNC cells are in progress. - 268 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P182 Resistance of Acanthamoeba spp. cysts to disinfection treatments Celine Coulona, Anne Collignonb, Gerald McDonnellc and Vincent Thomasa aSTERIS SA R&D, 18 route du Panorama, 92260 Fontenay-aux-Roses, France; bUniversite´ de Paris-Sud XI - Faculte´ de Pharmacie - Departement´ de microbiologie, USC INRA, EA 40-43, 5 rue Jean-Baptiste Clement,´ 92296 Chatenay Malabry, France; cSTERIS Limited, STERIS House, Jays Close, Viables, RG22 4AX Basingstoke, United Kingdom celine [email protected] Context In addition to their intrinsic pathogenicity, amoebae are known to potentially harbour pathogenic bacteria that could present a threat to humans and particularly hospitalized patients. These bacteria potentially survive in amoebal cysts. It is important to evaluate the effectiveness of chemical disinfectants used for surface disinfection against amoebal cysts. Methods Nine different Acanthamoeba spp. strains, including Acanthamoeba polyphaga and Acanthamoeba castellanii ATCC and field isolates were axenically grown in PYG medium before encystment in Neff’s medium. Cysts were diluted 1/10 in each disinfectant tested (see below), and incubated for 10 min at appropriate temperature. After neutralization in DE buffer, cysts were then serial-diluted on Escherichia coli lawns on NNA agar. After 7 days incubation at 28◦C, NNA plates were observed for trophozoites growth, and log reductions were calculated. Results The most effective treatments were bleach 2.5%, ethanol 70% and a peracetic acid-based formulation (STERIS 20), with a minimum 4-log10 reduction for each of the 9 strains tested, except for one strain presenting limited resistance to ethanol. Ortho-phthalaldehyde (OPA) alone at 0.55%, peracetic acid alone at 0.2% and a hydrogen peroxide-based product (SporKlenz) presented good activity on all strains (4-log10 reduction) except for 1 of 9 strains with OPA (1-log10 reduction), and for 2 of 9 strains for PAA (0.2 and 0.9-log10 reduction) and SporKlenz (1.7 and 3.6-log10 reduction). Bleach 0.25%, a commercial OPA-based product and glutaraldehyde 2% presented a moderate to low activity. Products with the lowest activity were hydrogen peroxide alone at 7.5% and a commercial, 2% glutaraldehyde-based product. Conclusions This study demonstrates that disinfectant efficacy can vary widely against amoeba, including known high level disinfectants that may be taken for granted as being effective under clinical practice. Amoebal cysts can be highly resistant to disinfection. - 269 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P183 Resistance of Chlamydia-like to disinfection treatments and survival on surfaces Celine Coulona, Anne Collignonb, Gerald McDonnellc and Vincent Thomasa aSTERIS SA R&D, 18 route du Panorama, 92260 Fontenay-aux-Roses, France; bUniversite´ de Paris-Sud XI - Faculte´ de Pharmacie - Departement´ de microbiologie, USC INRA, EA 40-43, 5 rue Jean-Baptiste Clement,´ 92296 Chatenay Malabry, France; cSTERIS Limited, STERIS House, Jays Close, Viables, RG22 4AX Basingstoke, United Kingdom celine [email protected] Context: Chlamydia-like are obligate intra-cellular bacteria, growing in amoebae or cells culture, known since 1997. They are believed to be responsible for human and veterinary infections like respiratory infections or abortions. These bacteria are widely spread in the environment because of their association with amoebae and thus represent a threat for the populations. In this context, it is important to evaluate the susceptibility of these bacteria to disinfection treatments and to determine their survival on surface. Methods Five different Chlamydia-like species were tested: Parachlamydia acanthamoeba (2 strains), Waddlia chondrophila, Protochlamydia naegleriophila, Criblamydia sequanensis and Simkania negevensis. Chlamydia trachomatis DSM19102 was also included for comparison. For disinfection tests, bacteria were diluted 1/10 in each disinfectant to be tested (alcohool, aldehydes, oxydants), and incubated for 10 min at appropriate temperature. After neutralization in DE buffer, bacteria were serial-diluted on amoebae or cells culture plates. After 7 days incubation at appropriate temperature, plates were observed for bacteria growth, and log reductions were calculated. For survival tests, suspensions of bacteria with or without blood were serial diluted, deposited in 96 wells plates and dried under laminar flow hood. Plates were then stored at room temperature. After 0, 7, 14, 21, 28, 35 and 42 days, suspensions of susceptible amoebal or mammalian cells were added in wells and plates were incubated at appropriate temperature for one week to allow intracellular growth of bacteria. Bacterial titer was then calculated to determine the reduction in bacterial viability after each tested time. Results All the strains tested were susceptible to disinfection treatments tested, except one (Parachlamydia acanthamoeba strain Hall Coccus) presenting limited resistance to hydrogen peroxide and another (Criblamydia sequanensis) presenting limited resistance to an aldehyde-based commercial product. This study also demonstrated a variable resistance of tested species to temperature, some like Protochlamydia naegleriophila resisting exposure at 55◦C for 10 min while others like Chlamydia trachomatis were completely inactivated. Concerning survival tests, most strains survived for more than 15 days on surfaces, with better survival in the presence of blood. Conclusion From these results, most Chlamydia-like species seem to be easily killed when they are outside their host cells. However most of these bacteria grow into amoebae which are able to encyst in unfavourable conditions. As cysts could provide further protection to intracellular Chlamydia-like, we will now study possible survival of Chlamydia-like in various amoebal cysts and their resistance to disinfection. - 270 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P184 Colonization of Legionella species in hospital water system in Turkey Haluk Erdogana, Hale Turanb, Riza Hasimogluc, Ozlem Kurt Azapd and Hande Arsland aBaskent University Medicine Faculty, Department of Infectious Disease and Clinical Microbiology, saray mh, kızlarpınarı cd,no:1, 7400 Alanya, Turkey; bBaskent University Medicine Faculty, Department of Infectious Disease and Clinical Microbiology, Hoca Cihan Mh, Saray c, No:1, 42000 Konya, Turkey; cBaskent University Medicine Faculty, Department of Infectious Disease and Clinical Microbiology, 35510 Izmir, Turkey; dBaskent University Medicine Faculty, Department of Infectious Diseases and Clinical Microbiology, Fevzi C¸akmak Caddesi, 10. Sok. No:45 Bahc¸elievler, 6490 Ankara, Turkey [email protected] Introduction: The hospital water distribution systems is the reservior for hospital associated legionnaires diseases. There are few reports investigating colonization of Legionella spp in Turkish hospital water systems. The goal of this study was to evaluate colonization of Legionella species in hospital water distribution systems in Baskent University Hospital and Medical Centers in four diffirent regions of Turkey. Methods: Water and swab samples were obtained from 4 hospital water systems from September 2006 to January 2007. Water samples were collected in 100-mL sterile containers and were concentrated by membrane filters with a pore size of 0.45µm. Heat treatment was used to eliminate other microorganisms from the samples, which were then spread on buffered charcoal yeast extract (BCYE-α) agar plates and glycine, vancomycin, polymyxin, cycloheximide (GVPC) agar plates. Cysteine-dependent colonies were identified by latex agglutination. Results: One hundred and twenty five water and swab samples were analyzed. Legionella was isolated at least one of samples in all hospital water samples. Legionella pneumophila was detected in 34 (27.2%) of the samples. The most frequently isolated species were L. pneumophila serogroups 1 (58.8%) and 6 (35.3%). Conclusions: Awareness has increased against legionnaires disease. Hospital water systems have been decontaminated. Hospital associated legionnaires disease has not been detected during consequent years. - 271 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P185 Preliminary Data on Legionella detection in Water Distribution Systems in Cameroon Marguerite Ndayo Wouafo, Ariane Nzouankeu and Guy Ejenguele Centre Pasteur du Cameroun, Rue Henry Dunant, 1274 Yaounde,´ Cameroon [email protected] After declaration of the first case of Legionellosis in Cameroon in 2007, the Centre Pasteur of Cameroon implemented the detection method of Legionella. The introduction of this new method was scheduled in order to investigate Legionella spp. colonization of water distribution systems (WDS) of large buildings including hospital, hotels and Off Shore Exploitations Sites (OSES) in an attempt to identify risk factors for Legionella spp. associated with WDS. METHODS: Water systems of 6 hotels, 6 hospitals and 3 OSES were investigated for the presence of Legionella spp. according to the NT90-431protocol. RESULTS: A total of 130 samples were collected, 77 from hotels, 27 from hospital and 26 from OSES. Legionella spp. was isolated from 41 (31.54%) samples. Of the total 41 positive isolates, 21 (51%) were L. pneumophila serotype 1, 8 (20%) were L.pneumophila serotype 2.15 and 12 (29%) were L. sp. CONCLUSION: These results showed that WDS of hospitals, hotels and OSES can be heavily colonized by Legionella spp. and may present a risk of Legionellosis. According to these preliminaries results, we have just implemented a survey protocol with ACIP funding (RIIP). Our aim is : (1) To improve awareness and recognition of Legionellosis a possible community or hospital-acquired pneumonia; (2) Provide current information about Legionella infections to healthcare professionals; (3) Suggest a proactive approach for prevention of Legionellosis by routine environmental cultures and suggest approaches for disinfection of Legionella in water systems. - 272 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P186 Sequence-based typing of Legionella pneumophila sg 6 strains isolated from hospital hot water systems in Poland. Katarzyna Pancer National Institute of Public Health - NIH, Chocimska 24, 00-791 Warsaw, Poland [email protected] The majority of hospital-acquired pneumonia due to Legionella pneumophila are caused by strains belonging to serogroup 1. However, L.pneumophila sg 6 strains were also found as an etiological agent of nosocomial legionellosis. The significant higher frequency of Legionella pneumophila sg 6 strains in polish hospital water systems than in non-hospital HWS was observed in our previous studies. The other (than serotyping) widely used method to characterize Legionella pneumophila strains is the sequence-based typing (SBT), but the majority of examined strains belonged to sg 1. The aim of this studies was to determine the SBT profiles of L.pneumophila sg 6 strains isolated from hospitals in Poland. Bacteria L.pneumophila sg 6 were found in water samples collected from 6 hospitals. Determination of STs (using 7 genes) was done according to Gaia et al. and Ratzow et al. and profiles were checked in EWGLI-SBT database. L.pneumophila sg 6 belonging to ST 421 were found in 2 hospitals. In one of them all tested strains of L.pneumophila sg 6 belonged to this ST 421. In second one - bacteria L.pneumophila sg 6 varied - some of them belonged to ST 421 and other to new/non-reported profile. Totally, new/non-reported profiles were found in 4 hospitals, but the number of determined profiles differed in those hospitals. Another found in EWGLI-SBT database profile was ST 114 - those strains were found in one hospital. The Sequence-Based Typing based on 7 allele was sufficient for differentiation of L.pneumophila sg 6 strains, even if they were environmental isolates. It was shown the strong colonization of one hot water system with one strain of L.pneumophila sg 6 (ST 421), but also heavy differentiation among strains belonging to serogroup 6 in another systems. - 273 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P187 + In-vitro study of H2O2 - Ag activity against Legionella and evaluation of its efficacy in hospital waterborne infection control Maria Luisa Riccia, Stefano Fontanaa, Laura Acheneb, Maria Scaturroa, Federica Pincia, Desiree´ Mineoa, Mas- simo Ottavianib and Enrico Veschettib aDepartment of Infectious, Parasitic and Immune-mediated Diseases - Istituto Superiore di Sanita` -, Viale Regina Elena, 299, 161 Rome, Italy; bDepartment of Environment and Primary Prevention - Istituto Superiore di Sanita` -, Viale Regina Elena, 299, 161 Roma, Italy [email protected] In Italy nosocomial Legionella pneumophila pneumonia remains an important cause of infection representing 7.1% of notified cases compared to 7.5% of community acquired cases. Its fatality rate is currently 33.3% (2008 data). Hospital hot water systems have been shown to be the main Legionella reservoirs and for this reason have usually been subjected to several disinfection treatments to minimize the risk of infection. However, their action has been proved not to be always effective in spite of their demonstrated in-vitro efficacy. The aim of this study was to investigate both in-vitro and + on-site activity of H2O2 - Ag solutions in the disinfection of hot water distribution systems. Factors affecting its efficacy + were also evaluated. In-vitro efficacy of H2O2 - Ag solutions was determined according to CEN/TC216 No. 461 norm conveniently modified. Mathematical modelling of kill curve experiments was performed by examining the results obtained using four disinfectant concentrations at seven sampling times. Two hospital water systems mainly consisting of galvanised steel and polyethylene materials were treated in continuous for 3 and 4 years, respectively, with a commercial + solution of H2O2 - Ag 1:103 so that nominal concentration of H2O2 in hot water was 10 mg/L at the introduction point. The actual concentrations of the disinfectant components were determined with a flow injection apparatus installed on-site immediately after the disinfectant addition point and at the most remote sampling point. Hydrogen peroxide was detected using a colorimetric reaction after addition of vanadate ion at alkaline pH. Silver was determined in the laboratory by atomic emission spectrometry with inductively coupled plasma (ICP-OES). Water samples were collected at fixed points to monitor Legionella CFU/L. Among the concentrations tested in vitro the solution containing 15 mg/L of H2O2 and 10 µg/L of Ag+ (roughly corresponding to the commercial formulation used to disinfect the two hospital water systems) gave a reduction of 4-5 log and 9 log after 24 and 48-hour contact time, respectively. Overall, a good correlation between the variables tested was demonstrated. An equation to evaluate the reduction of microbial concentration was developed + assuming that the ratio between H2O2 and Ag concentrations remains constant and equal to 103. Residual concentrations + of H2O2 and Ag measured in the two hospital water distribution systems were in the range 3.2-12.8 mg/L and 0.8-6.6 µg/L, respectively. A significant variability of disinfectant residue was detected at every monitoring point during the time. These fluctuations agreed with the variability of microbiological results and were primarily caused by the complex flow regime in the distribution system as well as by the interaction of the disinfectant with biofilm and pipe materials according to other findings described in literature. - 274 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P188 Contamination of legionella spp. and amoeba of water distribution systems in 3 hospitals of Shanghai, China Lili Tao, Bijie Hu and Zhaoyan Zhao Zhongshan Hospital, Fudan University, Fenglin Road 180, 200032 Shanghai, China [email protected] Objective: To survey Legionella spp. and amoebae contamination in hospital water systems in 3 inpatient healthcare facilities in Shanghai, China. Method: Three general hospitals with transplant departments were surveyed. Water samples were collected from each ward from the tap outlets of the water distribution system. Legionella spp. and amoebae were cultured from the water samples. Results: A total of 85 samples were collected from cool water outlets, 6 were collected from transplant departments, 12 from intensive care units, and others were obtained from common wards. 62 were contaminated by various amoebae (72.9%). 33 were contaminated by legionella spp. (38.8%), including15 by L. pneumophila1 (45.5%), 8 by L. pneumophila 2-14 (24.2%), 9 by other legionella spp., and 1 contaminated by the combination of L. pneumophila 1 and other legionella spp. The contamination rate of one hospital was as high as 73% (19/26), whereas another hospital has a very low contamination rate (1/25). In one transplant unit, 5 of 5 outlets were contaminated by L. pneumophila 1, and 5 of 12 ICUs yielded legionella spp. In regression models, the high contamination rate of legionella was related with the age of buildings and the presence of amoebae in the water distribution systems. Conclusion: There is a high risk for acquiring legionellosis in some of hospitals in China, inhouse detections of legionella should be taken in these hositals. - 275 - Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P189 Prospective Study of Legionnaires’ Disease Mortality Factors in France: Description of 21 Nososomial Cases, (April 2006 - June 2007) Silene Pires-Cronenbergera, Christian Chidiacb, Christine Campesec, Didier Chec, Sophie Jarraudd,Jer´ omeˆ Etienned, Roland Poiriere and Philippe Vanhemsa aService d’Hygiene` Epidemiologie´ et Prevention,´ Hopitalˆ Edouard Herriot, 5 Place d’Arsonval, 69003 Lyon, France; bService de Maladies Infectieuses et Tropicales, Hopitalˆ de la Croix Rousse, 103 Grande Rue de la Croix Rousse, 69004 Lyon, France; cInstitut de Veille Sanitaire, 12 rue du Val d’Osne, 94415 Saint Maurice, France; dCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; eService de Maladies Respiratoires, Centre Hospitalier du Pays d’Aix, #NOM?, 13616 Aix-en-Provence, France [email protected] Background: Legionella species are important pathogens causing community-acquired and nosocomial pneumonias. Most cases of legionnaire’s disease (LD) are due to Legionella pneumophila serogroup 1 (Lp1). In France, among reported cases, less than 10% are nosocomial. Among 595 LD cases included in a multicentric prospective study of LD mortality factors (April 2006-June 2007) we identified 21 nosocomial cases in order to describe their characteristics. Methods: Cases included were all cases with a clinical and radiological diagnosis of pneumonia due to Lp1. A confirmed nosocomial case was defined as a patient hospitalized during the entire incubation period (10 days) or more. A probable nosocomial case was defined as a patient hospitalized between 2 to 10 days before onset of symptoms. Data collection included: demographic, clinical and biological characteristics, risk factors, severity score and outcome. Statistical comparisons between nosocomial cases (NC) and community-acquired cases (CA) were performed using Fisher’s exact test or Mann- Whitney U-test. The survival probability was calculated using the Kaplan-Meier method. Results: Among the 21 cases, 12 (57%) were confirmed, and 9 (43%) were probable nosocomial cases. The mean age was 76 years (55-103 yrs) in NC vs. 61 years (17-100 yrs) in CA (p<0.001). No difference between the two groups was observed concerning, sex ratio (male/female) (1.33 vs 2.83, p=0.128), the presence of at least one risk factor, respiratory, digestive, and neurological signs and intensive care unit stay. Mortality was significantly higher among the NC (48% vs. 8.5%, p<0.001). A significant difference between the two groups was also observed concerning, presence of cancer or blood disease (43% vs. 7%, p<0.001), placement in Fine score risk classes IV-V at admission (p=0.002) and in the probability of survival (p=0.026). The probability of survival at 2d, 10d, 20d, and 90d was 95.2%, 81.0%, 65.4%, 57.2% and 99.1%, 93.4%, 90.9%, and 65.3% in NC and CA respectively. Median length of stay was 21 and 9.5 days in NC and CA respectively. Conclusion: Nosocomial acquired Legionella pneumonia occurred in elderly patients. The fatality rate in NC was higher than CA of LD. This might be explained by a different pattern of underlying diseases in these patients compared to community cases. These results suggest the high importance of early diagnosis of legionellosis is especially important in nosocomial cases in order to reduce the mortality. - 276 - SESSION 10 EPIDEMIOLOGY: B) OUTBREAKS AND POPULATION GENETICS AND TAXONOMY - 277 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P190 Distribution of serogroups, sequence types, and monoclonal antibody subgroups among Legionella pneumophila isolates from patients in Japan Junko Amemura-Maekawaa, Fumiaki Kuraa, Jurgen H. Helbigb, Bin Changa, Noriko Kanekoc, Yuko Watanabed, Junko Isobee, Masafumi Nukinaf, Hiroshi Nakajimag, Kimiko Kawanoh, Yuki Tadai and Haruo Watanabea aDepartment of Bacteriology I, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, 162-8640, Tokyo, Japan; bInstitute of Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany; cYamagata Prefectural Institute of Public Health, 1-6-6, Tokamachi, 990-0031 Yamagata, Japan; dKanagawa Prefectural Institute of Public Health, Shimomachiya 1-3-1, 253-0087 Chigasaki, Japan; eToyama Institute of Health, 17-1, Nakataikoyama, 939- 0363 Imizu, Japan; fPublic Health Research Institute of Kobe City, 4-6, Minatojima Nakamachi, 650-0046 Kobe, Japan; gOkayama Prefectural Institute for Environmental Science and Public Health, 739-1, Uchio, 701-0298 Okayama, Japan; hMiyazaki Prefectural Institute for Public Health and Environment, 2-3-2, Gakuen Kibanadai Nishi, 889-2155 Miyazaki, Japan; iInfectious Disease Surveillance Center, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, 162-8640, Tokyo, Japan [email protected] We collected a total of 86 unrelated clinical isolates of Legionella pneumophila in Japan, which were isolated during 1980- 2008. The major serogroup of the isolates was serogroup 1 (80.2%), followed by serogroup 5, 3, and 2. Interestingly, the male-to-female ratio of the patients due to L. pneumophila serogroup 1 was significantly higher than those due to L. pneumophila other serogroups (12.4 vs 2.0). We characterized all isolates by sequence-based typing (SBT) using seven loci and serogroup 1 strains additionally by monoclonal antibody (MAb) typing. The most prevalent subgroup was Benidorm (34.9% of all strains). MAb 3/1, which recognizes the virulence-associated epitope, positive isolates were 64.0% as almost the same as that by the pan- European study. A total of 53 sequence types were found among 86 isolates. By the minimum spanning tree analysis, seven clonal complexes were observed. The major ST was ST1 (seven isolates), which has been one of the most prevalent ST from patients and environments of the world, although six of seven ST1 strains were isolated before 1994. The second major ST was ST306, to which six isolates belonged. Next, each five isolates belonged to ST120 and ST138. All isolates belonged to ST306 or ST138 except for one ST306 isolate were suspected or confirmed to be derived from bath water, suggesting that these STs-strains would preferentially harbor in bath habitat. On the other hand, the sources of all isolates belonged to ST1 and ST120 remain to be unclear in Japan. ST23 was found among four isolates. Two large public-bath-facilities-associated outbreaks with hundreds patients were due to ST23 isolates. Other nine kinds of STs included three or two isolates. Remaining 39 STs were each represented by a single isolate. Identical ST could be found among isolates belonging to different MAb types in identical serogroups and vice versa. ST1 isolates consisted of OLDA (n=6) and Oxford (n=1). ST23 consisted of two Allentown/France isolates, one Philadelphia isolate, and one Oxford isolate. Ultimately, by the combination of SBT and monoclonal antibody, 86 isolates were divided into 58 types (discrimination index, 0.984), confirming the usefulness of the combination in epidemiological study. - 278 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P191 Sequence types of Portuguese clinical isolates of Legionella pneumophila and their correlation with geographic distribution and infection origin Maria-Jesus Chasqueiraa,Lucia´ Rodriguesa, Marta Nascimentob and Teresa Marquesa aFaculdade de Cienciasˆ Medicas,´ Microbiology Department, Chronic Diseases Research Centre-CEDOC, Campo Martires´ da Patria,´ 130, 1169-056 Lisboa, Portugal; bHospital de Santa Cruz, Av. Prof. Reinaldo dos Santos, 2790- 134 Carnaxide, Portugal [email protected] At the 22nd Annual Meeting of the European Working Group for Legionella Infections (EWGLI), in 2007, an improved Sequence-Based Type (SBT) database was implemented. Until the end of June 2009, this database included 3156 number of entries (1887 from clinical strains, 1202 from environmental strains and 67 from unknown origin) sent by 35 countries. The aim of the present study is to analyse the Portuguese data, including the geographic profile distribution and the correlation between Sequence-Type (ST) and the infection origin. Using the SBT database tools, we also evaluated the distribution of the Portuguese ST isolates in the world. We studied the clinical isolates received in our laboratories since 1987, including those isolated during the four years of the Surveillance Scheme for the Legionnaires’ disease implemented in 2004, in Portugal. In total, 72 clinical isolates of Legionella pneumophila (L. pneumophila) were analysed (67 serogroup 1 - sg1, and five non-sg1). Despite the low number of isolates, our collection contains the majority of the clinical isolates collected in Portugal since 1987, so it is possible that this sampling is representative of the profiles circulating in Portugal. The phenotype was determined using the Dresden panel of monoclonal antibodies (MAbs). The genotype of each isolate was determined using the EWGLI standard SBT method, including seven genes. Using the Dresden panel of MAbs, the 67 L. pneumophila sg1 isolates were divided into five different subgroups. Applying SBT, all but three of the isolates included in this study were typable by the seven genes. The 72 isolates were discriminated into 23 ST, thirteen of them being new in the EWGLI-SBT database. In addition, six new allele numbers were assigned by the curators after submitting our data to the database. Combining the phenotypic (MAb subgroup) and ST data for these strains resulted in a total of 31 profiles. These significant profile diversity observed is in concordance with the reports of the other countries. The results showed that the most prevalent strain was ST100 MAb subgroup ”Allentown/France” (18/67). The ST 100 is unique to Portugal and all of the isolates (32/72) came from nosocomial infections of the same hospital, in the area of Lisbon, over a period of several years. The ST 99, also unique to Portugal, was identified in four strains isolated in community-acquired infections in three different regions of Portugal (north, centre and Algarve). It was the ST more widespread in our country. Analysing the data of the EWGLI SBT database, we verify that a significant portion (21/72) of our strains belong to STs that occur throughout the world. This rate was biased by the disproportionately large number of strains obtained from one hospital (32/72). We also observed that six of the 23 STs (ST1, 23, 42, 44, 94, P4) identified in Portugal are widely spread in the world, being present in different European and non-European countries. - 279 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P192 A new hotel opened and wellcome Legionella: what’s wrong? Haluk Erdogana and Hande Arslanb aBaskent University Medicine Faculty, Department of Infectious Disease and Clinical Microbiology, saray mh, kızlarpınarı cd,no:1, 7400 Alanya, Turkey; bBaskent University Medicine Faculty, Department of Infectious Diseases and Clinical Microbiology, Fevzi C¸akmak Caddesi, 10. Sok. No:45 Bahc¸elievler, 6490 Ankara, Turkey [email protected] A 600 bed capacity new hotel opened on 15 May 2009 and then a few days later an outbreak of legionnaires disease occured. Five definitive cases and 1 presumptive case of legionnaires disease were identified between 20 May and 26 May 2009. One of them died. The all patients were tourist. The diagnosis was made by Legionella urinary antigen test and only presumptive case had high Legionella antibody titer. Legionella pneumophila wasn’t isolated any patients. The hotel water came from groundwater sourches and the municipal dirinking water network wasn’t used. The environmental samples were tested for colonization by Legionella species and the hotel water system had a high concentration (more than 102 CFU/100 mL) and an extensive colonization (11 of 13 samples) of L. pneumophila serogroup 1. The health ministry was informed. Immediate over chlorination and superheating of water were performed. This outbreak showed that a new hotel water systems have a risk of colonization with Legionella. Therefore, a control program for Legionella must be performed before opening. - 280 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P193 First population-based molecular epidemiology analysis of non-serogroup 1 Legionella pneumophila clinical isolates. Naglaa Elgngihya, Christine Pourcelb, Patrick Tanga, Nathalie Tijeta, Carla Duncana, Jonathan Gubbaya, Donald E Lowa and Cyril Guyarda aOntario Agency for Health Protection and Promotion, 81 A resources road, ON M9P 3T1 Toronto, Canada; bDepartment of Genetics and Microbiology , University of Paris XI, CNRS, UMR8621, 91405 Orsay, France [email protected] In Ontario, 39% of culture-confirmed cases of Legionella pneumophila(Lp) infections are caused by non-serogroup 1 isolates. This high percentage of non-serogroup 1 infections is in contrast with current literature. This finding suggests that non-lp1 Lp of Ontario might be well adapted to environmental survival and/or to host infection and it prompted us to conduct the first large scale molecular epidemiology analysis of culture confirmed non-serogroup 1 Lp infections. Among the 9 serogroups identified during this study, serogroups 6 and 8 were the most prevalent. EWGLI sequence based typing (SBT) and multiple-locus variable-number of tandem-repeat (MLVA) schemes were evaluated in a population-based study conducted with unrelated clinical isolates collected between 1980 and 2009. Twenty percent of the non-serogroup 1 isolates could not be typed with the neuA allele of the SBT scheme whereas all isolates were typable with the MLVA method. Discriminatory indexes of these 2 molecular typing methods were comparable and both schemes revealed a high genetic polymorphism. Cluster analysis revealed a high concordance between standardized SBT and MLVA. Interestingly, a large phylogenetic lineage comprising 55 to 65 % of the isolates and including 2 predominant molecular types was observed. Molecular types from this lineage were isolated over a 20 years period of time which suggests a high genetic stability. Comparative eBurst analysis of SBT data using the EWGLI database showed that all strains of this lineage belong to one major clonal group. This lineage also includes reference strains: Lp6 ATCC 33215, Lp12 ATCC 43290, Lp1 Philadelphia ATCC 33152 and emerging Lp1 Toronto EULV0827. Interestingly, some genotyping clusters tended to reflect the level of urbanization where patients were diagnosed suggesting that specific clusters have adapted to their respective environments. - 281 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P194 An Outbreak of Pontiac Fever Due to Legionella longbeachae Serogroup 2 Found in Potting Mix in a Horticultural Nursery in New Zealand Geoffrey Crampa, David Harteb, Nicholas Douglasa, Frances Grahamc, Mona Schousboed and Kate Sykesa aDepartment of Medicine, Gisborne Hospital, Tairawhiti District Health, Private Bag 7001, 4040 Gisborne, New Zealand; bInstitute of Environmental Science & Research Ltd, Kenepuru Drive, 5022 Porirua, New Zealand; cUniversity of Canterbury/Ministry of Health, P O Box 5013, 6011 Wellington, New Zealand; dCanterbury Health Laboratories, Canterbury District Health Board, P O Box 151, 8140 Christchurch, New Zealand [email protected] Background Previous outbreaks of Pontiac fever (a non-pneumonic illness caused by Legionella species) have invariably been associated with water droplet spread of Legionella species. We describe an outbreak of Pontiac fever in January 2007 in nine horticultural workers in Gisborne, North Island, New Zealand caused by L. longbeachae serogroup 2 present in potting mix and confirmed at the time of the investigation by the isolation of the same organism from sputum or the presence of species-specific antibodies in serum. In New Zealand an outbreak is defined as two or more cases associated with a single site with dates of onset within six months of each other. To our knowledge, there have been no previous reports in the medical literature of an outbreak of Pontiac fever caused by exposure to L. longbeachae serogroup 2 in potting mix. Methods An indirect immunofluorescent antibody test (IFAT) was performed to detect serum antibodies to heat-killed whole cell antigens from Legionella species commonly found in New Zealand. DNA for the PCR test was isolated from sputum and the gene targets were the Legionella 16S rRNA gene or the Legionella macrophage infectivity potentiator (mip) gene. Potting mix and water samples were collected from the workplace for Legionella culture. Results Nine of the ten exposed nursery workers were defined as having Pontiac fever on clinical grounds; two of these were sputum culture positive and three were serologically positive for L. longbeachae serogroup 2; none developed pneumonic features. The potting mix used by the exposed workers contained L. longbeachae serogroup 2 that was genetically indistinguishable from that isolated from the clinical samples. Three of the patients were admitted to hospital due to the severity of their symptoms but recovered without significant complications. This is the first recorded outbreak of Pontiac fever due to L. longbeachae serogroup 2 present in aerosolized potting mix. Conclusion As a result of this outbreak, we advocate introducing an industry standard ensuring the use of masks when handling potting mix and the attachment of masks to potting mix bags when sold to the public. Given the inconsistent and sometimes misleading messages on potting mix bags suggesting that they are pathogen free, we also advocate mandatory labelling warning consumers using compost of the risk of Legionella infection. - 282 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P195 Two groups of Legionella anisa isolates of environmental origin in Japan Fumiaki Kuraa, Junko Amemura-Maekawaa, Bin Changa, Atsuko Suzuki-Hashimotob, Masayuki Ichinoseb, Takuro Endoa and Haruo Watanabea aDepartment of Bacteriology I, National Institute of Infectious Diseases, Toyama 1-23-1, Shinjuku-ku, 162-8640, Tokyo, Japan; bTokyo Health Service Association, Ichigaya Sadohara-cho 1-2, Shinjuku-ku, 162-8402 Tokyo, Japan [email protected] Purpose: Legionellosis is caused mainly by inhalation of aerosols from water harboring Legionella. In previous EWGLI meeting held in Stockholm, we characterized Legionella spp. and serogroups of the isolates from a lot of cooling tower water both around Tokyo and throughout the country and reported that the rate of L. anisa isolates followed that of Legeinella pneumophila serogroup 1 isolates. As cases of L. anisa infection, outbreaks of Pontiac fever due to the infectious reservoir, fountains, at a restaurant and a hotel have been reported in the United States, indicating a potential risk of the bacterial infection. The identification of L. anisa has been performed, up to now, by whole-genome-DNA-DNA hybridization kit in Japan. In this study, we evaluated a convenient kit for identification of L. anisa (newly supplied latex agglutination test) and found that there were two groups of L. anisa, which were distinct serologically and in sequence types. Materials and methods: Forty-three isolates from independent facilities (40 isolates were derived from cooling tower water; two from spout water in bathhouses; one from bathtub water) and the type strain ATCC 35292 were included in this study. All isolates revealed white blue autofluorescence under long-wave ultraviolet irradiation and were identified as L. anisa by DNA-DNA hybridization kit (DDH-Legionella, Kyokuto Pharmaceuticals). We further characterized these strains using the latex agglutination kit specific for L. anisa (SLIDEX Legionella, bioMerieux) and sequencing of 16S rRNA gene (about 500 bp of the 5’-region). Results: By sequencing of 16S rRNA gene, L. anisa isolates were divided into 2 types, the type strain type (27.9%) and the heterologous type within three sets of 16S rRNA gene, of which one or two have one base deletion (72.1%). The latex agglutination test was positive in only 67% isolates, which was unexpectedly low. Most of the agglutination positve strains were assigned to the heterologous type. Discussion: All L. anisa isolates could be confirmed by the sequencing of 16S rRNA gene. In contrast, false negative results were obtained in the latex agglutination test, suggesting the test is not enough for identification of L. anisa in Japan. - 283 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P196 Development of a UK Legionnaires disease outbreak toolkit David Lemon, Ian Hall and Steve Leach Health Protection Agency, Proton Down, SP4 0JG Salisbury, United Kingdom [email protected] The overall objective of this project is to investigate and develop technical approaches based on a combination of geographical information systems, atmospheric dispersion and geospatial statistical modelling that will enable the HPA to identify the source of Legionella outbreaks more swiftly, efficiently and precisely, based on the home & work locations and travel histories of cases. Despite advances in understanding and increasing regulation, Legionella pneumophila infections continue to pose a threat to human health. There are a number of particular and specific public health challenges when trying to identify and control outbreaks, particularly those associated with ”geographically disseminated” sources linked to cooling towers and evaporative condenser units. A key aspect of the project is to be able to identify the source of the Legionella-contaminated aerosol that is causing the infections as quickly as possible, in order for remedial action to be taken with respect to the source so as to prevent further cases, and having found a potential source assess whether this remedial action is indeed impacting on the outbreak. Here we show analytical results that better characterise the incubation period distribution for Legionella based on specific outbreaks involving identified cases and the application of statistical inference models that are then able to characterise the period over which Legionella pneumophila has been released from a source. We also show some simple mapping and modelling approaches that are currently being developed to better geographically locate the potential sources of aerosol releases. Whilst outbreak investigations are well-guided by the extensive expertise of specialists in the UK the addition of a combination of computer-based visualisation and analytical approaches being proposed in this project would streamline existing procedures during outbreak investigations, help speed up the process of narrowing the geographic field for the location of putative implicated sources, provide more robust corroborative evidence regarding implicated sources at the end of the outbreak and in some instances decrease the time and resource required for the field epidemiological studies/investigations and consequent sample analysis. - 284 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P197 Description of clusters of Legionnaires’ disease associated with Spanish re-offender accommodation sites in the period 1987 to 2009 Rosa Canoa, Francisco Nogaredaa, Beatriz Baladronb, Carmen Martina and Carmen Pelazc aCentro Nacional de Epidemiologia. Instituto de Salud Carlos III, Monforte de Lemos 5, 28029 Madrid, Spain; bCentro Nacional de Microbiolog´ıa. Instituto de Salud Carlos III, Crta. Pozuelo Majadahonda, KM 1, 28220 Majadahonda, Spain; cLegionella Reference Laboratory; Centro Nacional Microbiolog´ıa; Instituto de Salud Carlos III, Majadahonda, 28220 Madrid, Spain [email protected] Introduction: The water supply systems of tourist accommodation sites (Hotels, spas, campsites, etc,) can be colonized by Legionella if control measures to prevent its multiplication are not put in place. Tourism is economically important in Spain. Every year Spain receives 50 million tourists who stay in over 9,000 establishments. The object of this study is to describe cases of legionellosis occurred in tourist re-offender accommodation sites in Spain. Methods:Legionellosis is a mandatorily reported disease in Spain. Cases associated with travel to Spain have been used, both cases of Spanish and foreign tourists reported to the national Centre of Epidemiology from 1989 to 2009. EU case definition and the European Guidelines were used in the reporting, handling and control of these cases. National Legionella Reference Laboratory receives isolates to be identified or typed voluntarily. Results:1,616 cases notified associated to travel in Spain caused 1,855 visits (some patients stayed at more than one accommodation site) and stayed at 1,170 sites. 49% of the cases (799/1,616) were sporadic and stayed at 900 sites. 51% (817/1.616) were cluster associated cases and stayed at 270 sites. These accommodation sites associated with more than one case were classified according to repeat visits in three categories. 89 out of 270 sites (33%) were re-offenders (the third category which implies higher risk) with repeat cases down the years despite taking control measures. 491 cases were involved in those sites. At the 79% (70/89) of the re-offending sites the repeat cases occurred with more than 2 years time interval (range 3 years in 36 sites to 10 years in 1 site. Legionella isolates were recovered from 66 out of 89 (74%) re-offender accommodation sites, and more than once in 40.5% of them (36/89) over the years. Persistence of Legionella was demonstrated in 26/89 (29%) re- offender accommodation sites, being most of them L. pneumophila SG 1. In 18 sites clinical and environmental isolates were matched, almost all of them Pontiac MAb subtype. Conclusions:The study showed that water supply systems were colonized with Legionella during the study period in re-offender accommodation sites. To identify the failures in actions carried out by the environmental inspectors at the sites associated with clusters is crucial in order to prevent new cases occurred despite the implementation of control measures. - 285 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P198 Characterization of a potentially novel Legionella species isolated from human clinical material Aleisha Reimer and Kathryn Bernard Public Health Agency of Canada, 1015 Arlington Street, MB R3E 3R2 Winnipeg, Canada aleisha [email protected] The genus Legionella currently consists of 52 species, roughly half of which have been implicated in human illness. Most clinical isolates of Legionella-like organisms referred to the Canadian National Microbiology Laboratory are easily identifiable to species level using a polyphasic approach; however, Legionella-like isolate 93L054 remained difficult to identify after extensive testing. Strain 93L054 was recovered from clinical material from a 63 year old male with respiratory illness in Montreal Canada and forwarded to our centre for identification. Original testing, including cellular fatty acid analysis, direct fluorescent antibody (DFA) assays, serum agglutination tests, and traditional biochemical tests, suggested this isolate was L. bozemanae-like; however, strain 93L054 did not autofluoresce and was only weakly reactive against L. bozemanae DFA reagents. Complete sequencing of the 16SrRNA gene in the current era suggests that the nearest relative is L. israelensis, rather than L. bozemanae, with 97.6% and 93.3% 16S rDNA sequence homology, respectively. Sequencing of secondary gene targets mip and rnpB revealed that strain 93L054 shares 81.7% and 86.8% homology with the type strain of L. israelensis, respectively, and is less than 87% homologous to all Legionella species tested. Strain 93L054 also differs from the type strain of L. israelensis in the production of oxidase, β-lactamase, and gelatinase. Genetic and phenotypic differences between 93L054 and other Legionella species, including L. israelensis and L. bozemanae, suggest that strain 93L054 is distinct from all known species and likely represents a novel species within the genus Legionella. - 286 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P199 The state of cooling towers Legionellae population in Verkhnyaya Pyshma after outbreak 2007. Olga L. Voronina, Marina S. Kunda, Vera V. Bitkina, Tatiana I. Karpova, Igor S. Tartakovskiy and Alexandr L. Gintsburg N.F. Gamaleya Institute for Epidemiology and Microbiology RAMS, Gamaleya str., 18, 123098 Moscow, Russian Federation [email protected] Verkhnyaya Pyshma Legionnaires’ disease ourbreak 2007 was the startig point of genomic population of Legionella investigation in Russia. Most important objects of monitoring in 2007-2009 years were cooling towers of industrial plant in Verkhnyaya Pyshma as most probable enviromental sourse of infection. In the observation period 5 series of water and biofilm samples were collected in different seasons. We isolated 75 Legionallae strains and characterized them, using sequence-based typing protocol EWGLI versions 3.0 and 4.0. 5 strains were Legionella spp.: 2 - L. dumoffii, 2 -L. gratiana and 1 - L. londeniensis. 70 strains were L. pneumophilla. They were divided into 23 ST with index discrimination - 0,919. Most common were ST 324, 7, 514 and 519, accounting for 54% of all. ST 324 and 519 predominated among water strains, and ST 324, 514, 7 - among biofilm strains. On the base of allelic profiles we made phylogenetic tree with help of programme PHYLIP version 3.68. There were 4 big clasters of ST demonstrated. All 7 targets of genomic analysis were various and had 5 - 8 alleles. The momp and pro fragments were most multiform. So the study showed the big diversity of enviromental Legionellae isolates from cooling towers. The seasonal changes in ST diversity were noted. - 287 - Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P200 Genetic relationships analyzed by 2 genotyping methods among the Legionella strains isolated from public bath water in Toyama Prefecture, Japan Masanori Watahiki, Junko Isobe, Junichi Kanatani, Tomoko Shima, Keiko Kimata and Takeshi Kurata Toyama Institute of Health, 17-1, Nakataikoyama, 939-0363 Imizu, Japan [email protected] Aim: This study is to understand the population structure of Legionella pneumophila (Lp) and Legionella spp. (Ls) in the hot water of public bath settings as a potential source of infection. We have done genotyping using both sequence- based typing (SBT) and pulse-field gel electrophoresis (PFGE). Backgrouds: Lp infections are associated with public bath water, suggesting that transmission of this bacterial species occurs via inhalation and are mainly due to Lp serogroup (SG)1. In Toyama Prefecture, legionellosis has increased during past 3 years, and all incidences were sporadic that those sources of infection were unclear. We have been collecting many Legionella strains isolated from the bath water during the last 3 years. There are a lot of strains with not only SG1 but also other SG or untypable serotype (UT). Methods: The isolated Lp and Ls were subjected to SBT and PFGE. For SBT, 6 or 7 genes (flaA, pilE, asd, mip, mompS, proA, and/or neuA) were sequenced using the European Working Group on Legionella Infections (EWGLI)protocol. PFGE was performed using the standard method. Results: Between 2001 and 2008, we analyzed 575 of Lp and Ls isolated from a number of public bath settings (475 strains) and clinical settings (100 strains). The detected serotypes were 233, 158, 57, 33, 30, and 24 for SG1, UT, SG6, SG4, SG9, and SG5 strains, respectively. PFGE was performed for all isolates, while SBT was performed for 139 of isolates. For almost all the strains, PFGE and SBT results showed a good similarity in the serotype and genotype data. However, 2 strains isolated from the same baths had different serogroups (either SG9 or UT) but identical PFGE pattern. In a bath facility, UT strains were identified as L. Londiniensis or Lp by 16S rDNA sequence analysis. Subsequently, we performed SBT for the Lp UT strain. In 1 bath facility, isolates with UT or SG9 were found to be clonal by PFGE typing and SBT. Interestingly in another bath facility, isolates with SG3 or SG4 were found to be clonal by SBT (6,10,15,6,4,14,11 or 6,6,15,6,4,14,11, respectively) but not by PFGE; only 8 fragments of the 24 fragments were identical between the 2 strains. Discussions: Sources of infection of legionellosis were hardly detemined by PFGE and/or SBT despite of isolating Lp SG1 strains as the largest group from bath water. The resuls of the serotyping and 2 genotyping methods were almost associated with each other. Therfore, these analyzed strains were concluded in the identical clonal groups . However, we found strains showing a clonal group by SBT but not by PFGE. Our finding could be indicated the indirect evidence of significant recombination events among these strains Therefore, we speculated that co-infections with a few Legionella strains, co-amplification in the protozoal cell, followed by recombination between strains are important stages in Legionella population. Conclusion: To control emerging Legionella infections, it may be important to understand the population structure of not only SG1 strains (representatives of clinical isolates) but also strains with other serotype Ls. - 288 - SESSION 11 GENOMICS AND COMPARATIVE GENOMICS - 289 - Session 11: Genomics and Comparative genomics P201 Virulence differences in a large systematic Legionella collection from the Netherlands Jeroen Den Boera, Jacob Bruina and Jurgen H. Helbigb aRegional Public Health Laboratory, Boerhaavelaan 26, 2035 RC Haarlem, Netherlands; bInstitute of Medical Microbiology and Hygiene, Fetscherstr. 74, D-01307 Dresden, Germany [email protected] Following a large outbreak of infection in The Netherlands that involved 188 patients with LD, a national outbreak detection programme (NODP) was started on 1 August 2002 with the aim of providing a short response time between the diagnosis of LD patients and the inspection and sampling of potential sources of infection. To identify potential sources during the incubation period, medical specialists in infectious disease control at the Regional Public Health Services carry out structured interviews (using a questionnaire) with the patient and / or a contact. The interviews focus on tracking each patient’s exposure to potential sources of infection. Certain potential sources mentioned specifically in the questionnaire, e.g., swimming pools and saunas, were included on the basis of previous epidemiological and outbreak studies Following the identification of a potential source, trained laboratory staff from the NODP take water and swab samples, which are then cultured for the presence of Legionella spp. Starting on 1 August 2002, patient isolates were sent by all 62 medical microbiology laboratories in The Netherlands that are involved in the diagnosis and treatment of pneumonia patients to the Regional Public Health Laboratory Kennemerland (Haarlem). All environmental strains and patients Legionella isolates that were collected until 1 August, 2008 were typed using the EWGLI defined AFLP and SBT. They were also typed using a Monoclonal Antibody set, also known as the Dresden Panel. Based on the results of this study, it was concluded that the genotype distribution of Legionella isolates from sporadic LD patients in The Netherlands differs from the genotype distribution of Legionella isolates in the environment. Based on these results it remains to be clarified if the distribution differences are a result of virulence traits of Legionella bacteria. Or, if there are other possible explainations. - 290 - Session 11: Genomics and Comparative genomics P202 Genome sequence of L. pneumophila strain Lorraine and an Environmental isolate: comparative genomics of six L. pneumophila strains Laura Gomez-Valeroa, Christophe Rusnioka, Sophie Jarraudb, Benoit Vacheriec, Zoe Rouyc, Valerie Barbec, Claudine Mediguec,Jer´ omeˆ Etienneb and Carmen Buchriesera aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France; cCEA /DSV /FAR /IG/ Genoscope Laboratoire de Genomique Comparative, 2 rue Gaston Cremieux, 91057 Evry Cedex, France [email protected] Legionella pneumophila is a Gram-negative, intracellular, bacterial pathogen, causing respiratory infections known as Legionnaires’ disease and Pontiac fever. Analysis of four L. pneumophila genomes (strains Paris, Lens, Philadelphia and Corby) had revealed as main features of this species: high genetic diversity, marked plasticity and the presence of many eukaryotic-like proteins many of which have now been shown to interfere in different steps of the infectious cycle by mimicking functions of host proteins. In order to identify additional factors probably involved in human disease and in host pathogen interactions, we determined the sequence of two new strains of L. pneumophila: Lorraine, an emerging strain causing community acquired Legionnaires’ disease in France, England, Wales, Netherlands and Greece and an ”non-pathogenic” L. pneumophila strain isolated over many years from the water system of a French hospital, without causing disease Interestingly, the Lorraine strain is significant cause of Legionnaires disease in France and England but is rarely isolated from the environment. Here we present the results of the genomic comparison of these six different L. pneumophila strains with the aim to determine the common L. pneumophila core and the specific traits associated to each strain and to their associated virulence properties. - 291 - Session 11: Genomics and Comparative genomics P203 The Genomic Sequence of the Soil Dwelling Opportunistic Pathogen Legionella longbeachae Natalia Kozaka, Meghan Bussb, Michael Fracec, Dhwani Govilc, Claressa Lucasa, Tatiana Travisa, Melissa Olsen- Rasmussenc, Robert Bensona and Barry Fieldsa aCenters for Disease Control and Prevention, 1600 Clifton Rd NE, Mailstop G03, Atlanta, GA 30333, United States of America; bEmory University, Department of Pediatrics Hematology/Oncology, 2015 Uppergate Drive, Atlanta, GA 30322, United States of America; cCenters for Disease Control and Prevention, 1600 Clifton Rd NE, G36, Atlanta, GA 30333, United States of America [email protected] Legionella longbeachae is responsible for up to half the cases of legionellosis in Australia and may be under-reported in the rest of the world due to the lack of L. longbeachae-specific diagnostic tests. L. longbeachae infections display distinctive differences in intracellular trafficking, caspase-1 activation, and host specificity compared to L. pneumophila, yet these species have an indistinguishable clinical presentation. Unlike other legionellae, which inhabit fresh water systems worldwide, L. longbeachae is found predominantly in moist potting soil. In this study, we sequenced and annotated the genome of an L. longbeachae clinical isolate, D4968, from Oregon, US, and compared it to the published genomes of L. pneumophila. The sequencing was performed using 454 technology. Automatic annotation was conducted by J. Craig Venter Institute (JCVI) and manual curation of genes was done using the Manatee interface developed by JCVI. The study revealed that the L. longbeachae genome consists of over 4 million base pairs (Mb), larger than L. pneumophila genomes by about 0.5 Mb. The gene order substantially differs in these two species. Genes encoding structural components of type II, type IV Lvh, and type IV Icm/Dot secretion systems are conserved between L. longbeachae and L. pneumophila. In contrast, only 1/3 homologs of L. pneumophila Icm/Dot substrates have been found in the D4968 genome. It was determined that L. longbeachae encodes numerous proteins with eukaryotic motifs and eukaryotic-like proteins unique to this species, including 18 ankyrin repeat-containing proteins, 1 U-box protein, and 3 proteins with a Ras family motif. We predict that in L. longbeachae these proteins are secreted by the Icm/Dot secretion system. In contrast to L. pneumophila, L. longbeachae D4968 genome does not carry flagellar biosynthesis genes, yet contains a chemotaxis operon. The lack of a flagellum explains the failure of L. longbeachae species to activate caspase-1 and trigger pyroptosis in C57B1/6 and BALB/c mouse macrophages. The unique features of L. longbeachae genome may reflect the adaptation of this species to life in soil. - 292 - Session 11: Genomics and Comparative genomics P204 Comparative genomics of the lipopolysaccharide biosynthesis gene cluster of Legionella pneumophila: towards rapid identification tests Nathalie Merault´ a,b, Christophe Rusnioka, Sophie Jarraudc, Elsa Bracheta, Mickael¨ Marina, Christel Cazaleta, Laura Gomez Valeroa, Jean-Louis Gaillardb, Jean-Louis Herrmannb,Jer´ omeˆ Etiennec, Christine Lawrenceb and Carmen Buchriesera aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bEA3647, Universite´ Versailles St Quentin en Yvelines, Laboratoire de Microbiologie, C.H.U. Raymond Poincare,´ 104 Boulevard Raymond Poincare,´ 92380 Garches, France; cCentre National de Ref´ erence´ des Legionella, Laboratoire de Bacteriologie´ INSERM U851, Faculte´ de Medecine,´ IFR128, 7 rue Guillaume Paradin, 69372 Lyon, France [email protected] Among the over 50 Legionella species described, L. pneumophila is responsible for over 90% of the legionellosis cases worldwide. Among the 15 L. pneumophila serogroups, serogroup 1 (Sg1) is implicated in ∼80% of the cases. Prevention of legionellosis needs to concentrate on the elimination of this pathogen from water and aerosol producing systems. Thus rapid and precise detection of Legionella in water systems is of utmost importance for prevention and risk prediction. Given the epidemiological background it is necessary to identify rapidly not only the presence of Legionella in water samples but also the species and Sg, in order to estimate the risk of legionellosis. We showed previously that genes coding for the proteins necessary for lipopolysaccharide (LPS) biosynthesis allow distinguishing between Sgs. Thus we further investigated this genomic region in L. pneumophila of different Sgs by sequencing and comparative analysis of Sg1, 6, 10, 12, 13 and 14. This showed that the LPS cluster of Sg1 is indeed specific for Sg1 and allowed to identify genes specific for Sg1 and Sg 2-14. Primers and probes were designed for real-time quantitative PCR (qPCR) and their specificity was tested in 370 L. pneumophila of all Sgs, 49 Legionella sp. non-pneumophila and 43 non-Legionella sp. frequently present in water according to the French organization for standardisation (AFNOR). We report here the development of a qPCR test specific for L. pneumophila Sg1. Sensitivity and efficiency of the method were compared to a commercial kit (Biorad, Marnes-la-Coquette, France) in samples from hospital hot water supply systems. The new method is as sensitive and efficient as the commercial method, but allows in addition to the detection of L. pneumophila contamination of water circuits the parallel typing of Legionella strains. A large-scale study of the water supply systems of 15 hospitals in the Paris area is in progress to compare the different methods, and to get a better knowledge of the contamination of hospital water supply systems. Furthermore, our newly developed test will allow characterizing the distribution of L. pneumophila Sg1 in the Paris hospitals. - 293 - Session 11: Genomics and Comparative genomics P205 Molecular analysis of virulence genes in Legionella strains from environmental samples Silvano Salarisa, Massimiliano Orsinib, Erica Leonia, Pier Paolo Legnania and Sandra Cristinoa aDepartment of Medicine and Public Health, Hygiene Division, University of Bologna, S.Giacomo 12, 40126 Bologna, Italy; bCenter for Advanced Studies, Research and Development in Sardinia (CRS4), Bioinformatics Division, Parco Scientifico e Tecnologico, POLARIS, Edificio 1, 9010 Pula (CA), Italy [email protected] Legionella pneumophila is one of about 42 species of Legionella genus and the main causative agent of Legionnaire’s disease. The icm/dot genes such as other structural and pathogenic genes (flaA, ralf, mip, etc.) in L.pneumophila are essential for bacterial survival within macrophages during lung infection and are associated with the severe pneumonia. Virulence genes are subjected to selection and faster development than the rest of genome; by means of these characteristics bacteria can elude the host’s response. Since most of the proteins encoded by icm/dot, and above mentioned genes are implicated in the ability of legionella to cause disease in human, it is important to know if these genes are equally carried by different L.pneumophila serogroups or/and other Legionella species which, sometimes have been associated with human disease. To verify this hypothesis and better understand the pathogenic mechanisms, we investigated the presence of four well-characterized genes (icmS, ralF, flaA, mip) involved in L.pneumophila virulence, in environmental isolates belonging to different L.pneumophila serogroups and L.species, comparing the genomic pattern and relating it to different pathogenic features. Isolation of Legionella spp. was performed according to ISO11731:9. Colonies were identified and classified by cultural, biochemical and serological features using commercial antisera for L.pneumophila sg 1 to 14 (Polyclonal latex reagents, Biolife), L.species (latex test, Oxoid; agglutinating sera Biogenetics Diagnostics). Starting from 100 Legionella spp. environmental isolates, we selected 20 L.pneumophila strains representing sg 1,3,6,7,8,9 and 20 L.species strains belonging to L.anisa and L.micdadei. Bacterial DNA was purified using Instagene Matrix kit (BIO-RAD Laboratories). Genes detection was performed by specific primers, designed with bioinformatics tools on four L.pneumophila strains published in the Genbank. Multiple alignments were accomplished using ClustalW2 to identify highly conserved regions. PCR reaction was run using the following thermal profile: 30 cycles of 20s at 92◦C, 20s at 58◦C, 30s at 72◦C. Products were detected by agarose gel electrophoresis and ethidium bromide staining. L.pneumophila strains (all serogroups) showed the presence of all selected genes, while L.micdadei and L.anisa strains were lacking in target regions of one or more fundamental genes of virulence as mip, ralF and flaA. Faint evidences of icmS presence were found in three isolates of L.species, other genes showed differences in bands intensity that can also be speculated.We can hypothesize that genetic differences in L.species usually less implicated in human disease, probably attenuate their virulence or switch off some pathogenic pathways. We are planning to confirm these data by large sequencing of the involved genomic region, to clarify mechanisms of gene lacking or inactivation (as example deletions, insertion or nucleotide substitutions). This approach could contribute to better understand the virulence pathways and help to analyze the real risk correlated to aquatic environment colonization. - 294 - Session 11: Genomics and Comparative genomics P206 Molecular analysis of Legionella non-pneumophila species Michel Ossendrijver, Martien Caspers and Frank Schuren TNO Microbial Genomics, P.O. Box 360, 3700 AJ Zeist, Netherlands [email protected] Although the majority of outbreaks of Legionnaires’disease are caused by L. pneumophila in a small number of cases other Legionella species were shown to be involved. In order to analyze these Legionella non-pneumophila species in more detail we made use of genomotyping technology earlier applied L. pneumophila and a number of other bacterial species. We made genomic representations of L. anisa, L. bozemanii, L. brunensis, L. feeleii, L. longbeachae and L. micdadei and created a microarray covering these species and L. pneumophila. This microarray was analyzed with genomic DNA of a strain collection covering all abovementioned species. The following results were obtained: - All species could be clearly discriminated based on unique patterns - Discrimination of strains within species was shown - There are indications that strains provided as L. longbeachae encompass two separate species This work may lead to the development of a novel diagnostic tool for a broader analysis of Legionella in water samples. Analysis of a larger number of strains belonging to these species would strengthen the currently available results. - 295 - Session 11: Genomics and Comparative genomics P207 Case studies on Next Gen sequencing technologies - experiences and results Dr. Kerstin A. Stangier and Dr. Yadhu Kumar GATC Biotech AG, Jakob-Stadler-Platz 7, 78467 Konstanz, Germany [email protected] Today it is no longer a challenge to produce giant amounts of sequence raw data of prokaryotic or eukaryotic genomes or transcriptomes. With the Next Generation Sequencers, the output performance per machine has been exploding during the past few years. While a classical Sanger sequencing machine is spitting out approximately 1 MB of raw data per day, a Next Generation system produces at least 400 MB a day. Above all, new and improved consumables and reagents are developed and released for the existing Next Gen machines in short time intervals which further considerably increase the amount of sequence data generated in a single run. With the Next Generation sequencing technologies, a wide range of applications are now affordable and within reach. The systems can be used for applications ranging from metagenome and genome sequencing (de novo or re-sequencing) to transcriptome analysis (e. g. cDNA, SAGE), to regulome studies (e.g. ChIP, microRNA). The main difference between the technologies on the market effecting the bioinformatic analysis is the read length. While the classical Sanger system produces reads of 1,000 bp in average, the reads of the Next Gen machines are 250/400 bp (Roche / 454 GS FLX sequencer) or up to 75 bp (Illumina / Solexa Genome Analyzer; Applied Biosystems SOLiD). Having performed many projects during the past years, each with a different set of questions and goals, various issues have emerged: 1. Depending on the goals of the project, different sequencing technologies or combination of technologies should be applied 2. The project goals and technologies used will dictate the analysis routines involving a wide range of rapidly changing bioinformatic tools. We will show in various case studies that in-depth knowledge of the strength and limitations of the Next Generation technologies and of the available analysis tools is the most crucial issue in project design and analysis. - 296 - SESSION 12 FUNCTIONAL GENOMICS - 297 - Session 12: Functional genomics P208 Deciphering Legionella biofilms at the proteomic level Sandra Ahmed-Lecheheba, Arbia Khemirib, Philippe Chanc, David Vaudryc, Jacques Frere` d, Thierry Jouennea and Pascal Cosettea aFRE3101 CNRS - Plateforme proteomique´ de l’IFR23, Universite´ de Rouen, 76821 Mont-Saint-Aignan Cedex, France; bLab. de Bioenerg´ etique´ Cellulaire, CEA / Cadarache, Batˆ 156, 13108 Saint Paul-lez-Durance, France; cEA4310 - Plateforme proteomique´ de l’IFR23, Universite´ de Rouen, 76821 Mont-SAint-Aignan Cedex, France; dUniversity of Poitiers - Laboratoire de Chimie et Microbiologie de l’Eau CNRS UMR 6008, 40 avenue du recteur Pineau, 86022 Poitiers, France [email protected] Legionella is a natural inhabitant of the aquatic medium where it may survive as floating cells, or within adherent bacterial communities, or by invading protozoans. Noticeably, this bacterium is at the origin of different types of lung infectious diseases, which arise after Legionella dissemination from air-cooling towers or water distribution networks. In the latter environments, Legionella biofilm has been largely described as a reservoir. In the light of these assumptions, we have undertaken proteomic investigations in order to better understand at the functional level, the way Legionella adapts in the different lifestyles, with a particular emphasis in the biofilm state. To draw up a reference map in basal conditions, we first assessed the protein expression in floating Legionella. That allowed to obtain 2D gels containing many hundreds of proteins which were further identified when sufficient biological material was present. In parallel, we developed sub-cellular fractionation to address more specifically the proteins present in the bacterial cell wall, namely in the outer membrane or associated to the surface. These compartments are highly important, since they constitute the first bacterial components toward the environment with for instance proteins involved in adhesion, in host interaction or also in resistance mechanisms. In this case, proteins that were never characterized before were visualized in both fractions. The latter constitute good candidates for extended investigations. Afterward, we moved to the characterization of Legionella biofilms. This was done by growing mono-species Legionella biofilms on stainless steel coupons in a dedicated bioreactor. Protein extraction from the collected bacterial biomass allowed 2D gels experiments and highlighted a few proteins with specific or markedly enhanced expression within biofilms. More recently, we performed preliminary assays with multispecies biofilms that will be more representative of what occurs in natural environment. Legionella was taken over from this community by using immune-capture and protein identification is now in progress. - 298 - SESSION 13 GENE REGULATION IN THE ENVIRONMENT AND THE HOST - 299 - Session 13: Gene regulation in the environment and the host P209 Flagella gene regulation in Legionella pneumophila Christiane Albert-Weissenbergera,b, Tobias Sahra,Jorg¨ Hackerc, Klaus Heunerc and Carmen Buchriesera aInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; bInstitut fur¨ Molekulare Infektionsbiologie, Rontgenring¨ 11, 97070 Wurzburg,,¨ Germany; cRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany [email protected] The bacterial pathogen Legionella pneumophila responds to environmental changes by differentiation. At least two forms are well described: replicative bacteria are avirulent; in contrast, transmissive bacteria are virulent and express virulence factors and flagella. With the aim to get a more complete understanding of how flagellar biosynthesis is controlled in L. pneumophila, we evaluated the impact of the regulatory proteins FleQ, FleR, RpoN, and FliA on flagella gene expression. Phenotypic analysis, Western blot and electron microscopy of mutants in the genes coding for FleQ, FleR, RpoN and FliA demonstrated that flagellin expression is strongly repressed in these mutants and that these four mutants are non-flagellated in transmissive phase. Transcriptome analysis of these regulatory mutants as compared to the wild type strain elucidated that FleQ enhances together with RpoN, the expression of 14 of the 46 flagellar genes. Unexpectedly, FleQ but not RpoN increases the transcript level of fliA encoding the sigma 28 factor FliA and subsequently FliA-dependent flagellar genes. Anyhow, although the flagellin encoding gene flaA is directly regulated by FliA the level of the flagellin protein is dependent on both, RpoN and FleQ. Surprisingly, FleR inhibited the expression of at least 14 flagellar genes. Based on these results we propose that early flagellar genes are controlled by FleQ and RpoN together, whereas the expression of the flagellar regulator FliA controlling expression of the late flagellar genes that lead to the complete synthesis of the flagellum - is enhanced by FleQ alone. Thus, in contrast to the commonly accepted view that enhancer binding proteins as FleQ always interact with RpoN to fulfil their regulatory functions, our results strongly indicate that FleQ regulates gene expression RpoN-dependent and RpoN-independent. - 300 - Session 13: Gene regulation in the environment and the host P210 Legionella pneumophila diguanylate cyclases and phosphodiesterases: defining the role of the cyclic-di-GMP signaling network in the life cycle of Legionella pneumophila Assaf Assaf Levia, Marc Folcherb, Urs Jenalb and Howard Shumana aColumbia University Medical Center, Department of Microbiology & Immunology, 701 West 168th Street, New York, NY 10032, United States of America; bBiozentrum, University of Basel, Klingelbergstrasse 50-70, 4056 Basel, Switzerland [email protected] The opportunistic human pathogen Legionella pneumophila , the causative agent of legionellosis, or Legionnaires’ disease, persists in natural and human-made aquatic environments where it is associated with multi-species biofilms and replicates within protozoa. Such biofilms are the source for legionellosis outbreaks. The molecular mechanisms controlling Legionella’s switch from a biofilm-associated to an infectious agent are currently unknown. Recently, the nucletotide-based second messenger, bis-(3’-5’)-cyclic dimeric guanosine monophosphate (c-di-GMP) has been shown to be involved in the regulation of virulence-related factors in several pathogenic bacteria species. It has been proposed that c-di-GMP controls the transition from sessile-to-planktonic life-styles; however numerous other cellular processes that are orchestrated by this small molecule have also been identified. The large number of proteins containing the domains responsible for the c-di-GMP molecule synthesis and hydrolysis (GGDEF and EAL, respectively) encoded in the L.pneumophila genome, and the co-occurrence of these domains with other sensor and response domains, indicate that this second messenger plays a major role in Legionella’s multifaceted biology. A set of 22 L. pneumophila genes, encoding proteins that contain putative domains of diguanylate cyclases (DGCs) and phosphodiesterases (PDEs) was identified in the genome of L. pneumophila Philadelphia-1 strain, and cloned on a medium-copy number plasmid under an inducible promoter. These cloned genes were expressed in two other bacterial species, Salmonella enterica serovar Typhimurium (ATCC 14028) and Vibrio cholerae, where phenotypes associated with increased and reduced levels of intracellular cyclic-di GMP are well-documented. In S. enterica , increased levels of cyclic-di-GMP cause the appearance of the RDAR colony morphotype on Congo-Red agar plates; increased cellulose biosynthesis, and reduced motility. In contrast, increased PDE activity is responsible for the White and Smooth colony morphotype on Congo- Red plates; decreased levels of cellulose biosynthesis, and hyper-motility. In V. cholerae increased levels of cyclic-di-GMP result in hyper-biofilm formation and reduced motility. Thus, expressing the Legionella genes in these bacterial species permitted us to assign a putative DGC or a PDE activity to them, even though many of these proteins contain both GGDEF and EAL domains. We used RP-HPLC to quantify the intracellular levels of cyclic-di-GMP in extracts of L. pneumophila carrying these cloned genes and were able to detect varying levels of this nucleotide in the different strains and confirmed the activities of many of these proteins as DGC or PDE. - 301 - Session 13: Gene regulation in the environment and the host P211 A database dedicated to gene expression data of Legionella pneumophila Sandrine Rousseaua, Tobias Sahrb, Pierre Ventadoura, Carmen Buchrieserb and Ivan Moszera aInstitut Pasteur, Integration´ et Analyse Genomiques,´ 28 rue du Docteur Roux, 75724 Paris Cedex 15, France; bInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France [email protected] DNA chips are a powerful technology to analyse differential gene expression through transcriptome studies. We have successfully used this technique for gene expression studies of Legionella pneumophila and have analyzed different regulatory mutants in different conditions and environments. Data generated through gene expression experiments using microarrays are usually very substantial in volume and difficult to interpret. In order to facilitate experimental and numerical data management and analysis, a database-driven web server, GenoScript, was created, that we have used to handle the L. pneumophila gene expression data. GenoScript is an integrated database for array-based gene expression profiling experiments, designed for collecting, storing and querying transcriptomic data. GenoScript allows the user to enter experiments, associated results (raw and analysed data) and array design, in compliance with the MIAME recommendations. GenoScript is ”project-oriented”, meaning that the interface can be easily customized according to specific features, and that different levels of ownership can be set for experiments. Retrieving data is performed through a user interface that enables both quick simple questions and complex multi-criteria queries. Users can visualize results in connection to classifications like GO, COG, KEGG BRITE or other organism-specific functional classifications. To this end, GenoScript is also dynamically linked to specialized tools, such as GoMiner and Cytoscape. An additional module, GenoScript Publication, provides independent and confidential access to experiments for reviewers. Subsequent submission to the ArrayExpress public repository (EBI) is facilitated by the MAGE-TAB export functionality. We have used the GenoScript database for storing and analysing our multiple gene expression data obtained during different experimental set-ups. We have integrated the biological information, the experimental and numerical methods, and the raw and statistically analysed data, all accessible in this database ( - 302 - Session 13: Gene regulation in the environment and the host P212 Further analysis of FleQ, RpoN and FliA - Regulators of the flagellar regulon of Legionella pneumophila Tino Schulza, Christiane Albert-Weissenbergerb,c, Carmen Buchrieserb and Klaus Heunera aRobert Koch-Institut, PG 26- Infections of the Elderly, Nordufer 20, D-13353 Berlin, Germany; bInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; cInstitut fur¨ Molekulare Infektionsbiologie, Rontgenring¨ 11, 97070 Wurzburg,,¨ Germany [email protected] BACKGROUND: Bacterial flagella are complex machines mediating bacterial motility. Most regulators and genes of the flagellar regulon are yet known. FleQ, FleR and FliA are regulators of the regulatory network of the flagellum. In addition, the expression of virulence factors is genetically linked to the expression of the flagellum in Legionella pneumophila. It was shown that alternative sigma factor FliA is involved in the expression of some virulence traits and that flagellin (FlaA) and FliA are needed for full invasion capacity of of L. pneumophila. METHODS: To get further insight into the regulation cascade at the level of gene and protein expression over time, we analysed selected genes via Reverse Transcription-PCR and via Western Blotting. In addition, we started to identify the transcriptional start site of fliA via RACE. We also began to generate and analyse mutant strains of FliA-dependent genes to further investigate the role of FliA as a virulence factor of L. pneumophila. RESULTS: The expression of different genes concerning the flagellar regulon is growth-dependent and confirm the data of former microarray analysis. The detection of FlaA, FliA and FleQ via specific antibodies approves these results at translational level. Comparative in silico analysis revealed high homology of most flagellar regulon genes and operons within the four known L. pneumophila genomes. CONCLUSION: The structure and overall expression pattern of the flagellum and its regulators seems to be conserved in the different Legionella strains investigated. However, there are also differences. Nevertheless, it is important to understand the regulation of the flagellar cascade to simutaneously learn more about the expression of virulence traits. - 303 - Session 13: Gene regulation in the environment and the host P213 The autoinducer synthase LqsA and cognate sensor kinase LqsS regulate host cell interactions, sedimentation and a genomic island of Legionella pneumophila Andre´ Tiadena, Thomas Spirigb, Tobias Sahrc, Carmen Buchrieserc and Hubert Hilbid aInstitute of Microbiology, ETH Zurich, Wolfgang-Pauli-Strasse 10, 8093 Zurich,¨ Switzerland; bDepartment of Chemistry and Biochemistry, University of California, 607 Charles E. Young Drive East, Los Angeles, CA 90095-1570, United States of America; cInstitut Pasteur, Biologie des Bacteries´ Intracellulaires and CNRS URA 2171, 25-28, Rue du Dr. Roux, 75724 Paris, France; dInstitute of Zoology, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland [email protected] Legionella pneumophila employs a biphasic life cycle to replicate in host cells and spread to new niches. Upon entering the stationary growth phase, the bacteria switch to a transmissive (virulent) state, which involves a complex regulatory network including the lqs gene cluster (lqsA-lqsR-hdeD-lqsS). The lqsA and lqsS genes are conserved in several bacterial species, including Vibrio cholerae, where cqsA and cqsS constitute the Cqs quorum sensing system. LqsR is a putative response regulator that promotes host-pathogen interactions and represses replication of L. pneumophila. The autoinducer synthase LqsA catalyzes the production of the diffusible signaling molecule 3-hydroxypentadecan-4-one (LAI-1) that is presumably recognized by the cognate sensor kinase LqsS. Here, we analyzed L. pneumophila strains lacking lqsA or lqsS. Compared to wild-type L. pneumophila, the ∆lqsS strain was more salt-resistant and impaired for the Icm/Dot type IV secretion system-dependent uptake by phagocytes. In addition, L. pneumophila strains lacking lqsS, lqsR or the alternative sigma factor rpoS displayed a novel phenotype, namely a retarded sedimentation and the production of an extracellular matrix. Deletion of lqsA moderately reduced the uptake of L. pneumophila by phagocytes, and the defect was complemented by expressing lqsA in trans. Unexpectedly, the overexpression of lqsA also restored the virulence and sedimentation defects of L. pneumophila strains lacking lqsS or lqsR, but not the phenotypes of strains lacking rpoS or icmT. These results suggest that the LqsA product LAI-1 can bypass the signaling pathways involving LqsS or LqsR and also signals through a system not encoded by the lqs gene cluster. A transcriptome analysis of the ∆lqsA and ∆lqsS mutant strains revealed that under the conditions tested lqsA regulated only few genes, whereas lqsS regulated the expression of close to 300 genes. These include more than 60 genes clustered in a 130 kb high plasticity genomic island, which is flanked by putative DNA-mobilizing genes and encodes multiple metal ion efflux systems. Upon overexpression of lqsA, a cluster of 18 genes in the genomic island was also upregulated, suggesting that LqsA and LqsS constitute a common signaling pathway. The current study lays the foundation for further analysis of cell-cell communication and signaling via diffusible small molecules in L. pneumophila. - 304 - Session 13: Gene regulation in the environment and the host P214 An assessment of the prevalence of Legionella spp. and effectiveness of thermal disinfection in eradicating the bacteria from water distribution system in a single hospital of Lublin. Małgorzata Wojtowicz´ , Agnieszka Sikora, Agnieszka Magrys´ and Maria Kozioł-Montewka Medical University of Lublin, ul. Chodzki´ 1, 20-093 Lublin, Poland [email protected] Introduction: Monitoring water systems for Legionella spp. has become a focus for hospitals because they represent ideal locations for Legionnaires’ disease transmission: at-risk individuals are present in large numbers; plumbing systems are frequently old and complex and water temperatures are often reduced to prevent scalding of patients. To reduce the likelihood of Legionnaires’ disease transmission in health care facilities, a strategy focusing on proper maintenance of water systems and elimination of recognized hazards is required. The aim of the work was to study occurrence of Legionella pneumophila in water distribution system of a single hospital as well as eradication efficacy by using thermal decontamination method. Materials and methods: Forty two hot water samples (1000 ml each) were collected form six sampling locations. The water samples were collected from proximal, distal and selected intermediate sites of the water system. The samples were analyzed a standard method based on filtration procedure and culture of bacteria on selective media. The species and/or serogroups were determined by a commercially available latex agglutination test kit. Results: All water samples tested for the presence of Legionella spp. gave positive results. The numbers of legionellae detected from 2.8x102 CFU/100 ml to 3.3x103 CFU/100 ml. All of the samples contained L.pneumophila SGs 2-14. The hospital water distribution system was regarded as colonized by Legionella. The major intervention included thermal eradication of bacteria by increase in the hot water temperature in the heater up to 70oC for 2 months. As a result, the number of Legionella spp decreased to the level of 4.9x102 CFU/100 ml. The next water examination was performed a month later. The number of Legionella pneumophila CFU decreased to the level of 52 CFU/100 ml. During the last examination, no legionellae were detected in water samples. Conclusions: 1. Legionella pneumophila SGs 2-14 colonized hospital hot water distribution system in the amounts exceeding acceptable norms, 2. Thermal disinfection significantly decreased the number of Legionella in examined water system 3. thermal disinfection is a long lasting but yielding good results process 4. the hot water temperature in draw off points should not be lower than 55oC. - 305 - P215 Using Flow Cytometry to Monitor the Risk of Legionellosis in Bath Water Toshitsugu Taguri1, Yasunori Oda2, Kanji Sugiyama3, Shinji Izumiyama4 and Fumiaki Kura4 1 Health Depatment, Nagasaki Prefectural Institute for ER and Public Health, 2-1306-11 Ikeda, Omura 856-0026, Japan, Phone: +81 0957 48 7560, FAX: +81 0957 48 7570, e- mail: [email protected], Web: http://www.pref.nagasaki.jp/kanhoken/ 2 Sysmex Corporation 3 Shizuoka Institute of Environment and Hygiene 4 National Institute of Infectious Diseases Maintaining clean water in public bathing facilities is necessary in order to prevent water borne diseases, and thereby protects the bathing facility industry’s reputation for health and safety. It is for that reason that water quality is constantly checked for contaminants, and chlorine is added whenever levels of contamination exceed the guidelines established by local and national governments. In the case of Legionella, however, the effectiveness of 2+ 2+ chlorination is greatly reduced by the presence in the water of Fe , Mn , NH3, skin debris and/or biofilms; and therefore it is necessary to add more chlorine whenever such presence occurs. In the present detection system, we use a flow cytometer to measure the total number of bacteria (both intact and degraded) within 2 minutes per sample by detecting typical patterns of light scattering and fluorescence (derived from DNA/RNA stained with BactQuick Dye), and thereby above a threshold of 3,000 of total bacteria consistently indicates the occurrence of Legionella in the bathing system tested. First, 4 bath waters of hot springs with inocula of 10 species of bacterial strains were fully chlorinated to provide the specific scattergram in FCM and their inactivation was assessed by using Live/Dead bacterial fluorescence assay and plate counting. Second, in a laboratory model of a hot tub, it was demonstrated that this specific criteria could monitor the growth curve of naturally-occurring heterotrophic bacteria with one and two days-delayed growth of Amoeba and Legionella, respectively, and also the killing curve of these bacteria by chlorination. Finally, “positive” or “negative” samples determined by the FCM criteria in samples from different 149 hot tubs were significantly correlated with positive or negative for Legionella in samples (detection limit, 10 CFU/100mL) with 95% sensitivity and 84% specificity. As a result, the FCM assay can be used for Legionella control at bathing facilities, especially those where the effectiveness of chlorine is reduced. ACKNOWLEDGEMENT This work was supported by grants from H19-20 Nagasaki Prefectural Specific Research Project and Health and Labour Sciences Research Grant (H19-kenki-015 to T. E.). P216 DETECTION of Legionella pneumophila in surface water and in sewage using a validated quantitative PCR method Bart Wullings, Ronald Italiaander and Dick van der Kooij KWR Water Cycle Research Institute, Groningenhaven 7, Nieuwegein 3430 BB, The Netherlands, Phone: #31 306069748, FAX: #31 306061165, e-mail: [email protected] A quantitative detection method (QPCR) for Legionella pneumophila was developed using real-time PCR. The method includes newly developed PCR primers and a taqman probe targeting the mip gene. The recovery of the DNA-isolation and possible PCR-inhibition is quantified by an internal control. This internal DNA standard is added to each analyzed water sample after filtration and isolated simultaneously with the sample DNA. In a multiplex real- time PCR, both the L. pneumophila DNA and the standard DNA is quantified. The method is validated for a large range of water types including drinking water, surface water and water from cooling towers and accepted as Dutch standard method (NEN 6254). Limited information is available of the presence of L. pneumophila in surface and in sewage. In the summer of 2008 34 surface water samples from rivers and lakes in the Netherlands were analyzed with the QPCR method. The recovery of the internal control was in most samples well above the minimal requirement of 20 % showing the robustness of the method. L. pneumophila was detected in only 8 of the analyzed water samples. The presence of L. pneumophila in the surface water samples could be related to sewage effluent close to the water sampling site. The presence of L. pneumophila in the influent and effluent of 32 sewage treatment plans was analyzed. L. pneumophila was detected in 47 % of the influent samples in a concentration ranging 103 – 108 mip gene copies/l. In 65% of the treated sewage samples L. pneumophila was detected in approximately equal concentrations. No significant removal was observed during treatment showing the ability of L. pneumophila to survive the treatment process. Neither, the concentration of E. coli in the waste water nor the percentage industrial waste in the influent showed a significant relation to the concentration of L. pneumophila. The high concentrations of L. pneumophila in certain sewage types can pose a health risk when aerosols are formed during sewage treatment. Reduction of the L. pneumophila load of sewage treatment plants requires identification of the source(s) of this organism present in the influent. P217 OCCURRENCE and genetic diversity of uncultured Legionella spp. in two unchlorinated groundwater supplies Bart Wullings1, Geo Bakker2 and Dick van der Kooij1 1 KWR Watercycle Research Institute, Groningenhaven 7, Nieuwegein 3430 BB, The Netherlands, Phone: #31 30 6069748, FAX: #31 30 6061165, e-mail: [email protected] 2 Vitens Water Technology, Leeuwarden, The Netherlands Uncultured Legionella spp. are common members of the microbial communities in drinking water from surface and groundwater treated at temperatures of below 15 ºC. The identified Legionella 16S rRNA gene sequences revealed a large diversity of yet-undescribed and unknown Legionella spp. in treated water. The concentration of natural organic matter (NOM) in treated water varies extensively between different types of drinking water. Limited information is available about the influence of NOM on the concentration and diversity of Legionella spp., including L. pneumophila in drinking water in The Netherlands. Two drinking water supplies were selected treating aerobic groundwater with a low NOM concentration (DOC < 0.5 ppm) and anaerobic groundwater with a high concentration of NOM (DOC 8 ppm), both at a temperature of about 10 ºC. Chlorination was not applied during treatment nor in the treated water. Legionella spp. concentrations were analyzed using quantitative PCR and identified by 16S rRNA gene sequence analysis in water samples taken from raw groundwater, during treatment and in the distribution system. Legionella spp. were detected in low concentration of 102 16S rRNA gene copies/l in water samples from three out of four aerobic groundwater wells. In the anaerobic groundwater, Legionella was not detected, but a four-log increase was observed after aeration and rapid sand filtration of the water, indicating that Legionella spp. multiplied in the filter bed. Legionella spp. were detected in all water samples from both supplies collected after each treatment step and from the distribution system. The anaerobic-groundwater supply however, showed significantly higher Legionella concentrations than the aerobic-groundwater supply. The Legionella spp. were identified by sequence analysis of the 16S rRNA gene fragment. The Legionella spp. richness in the clone libraries from the water samples treating aerobic groundwater was, in comparison to quantitative data, significantly higher (96 versus 43 OTU’s based on 97% sequence identity)than in the anaerobic-groundwater supply. Sequences related to L. pneumophila (>97%) were only incidentally observed. The Legionella populations of the two supplies are clearly separated in two distinct clusters based on the phylogenetic distance in the phylogenetic tree as fraction of the branch length. AUTHOR INDEX - 306 - Aase, A...... 161 Becerril, R...... 171 Abu Kwaik, Y...... 76, 128 Beddoe, T...... 155 Accorsi, M...... 241 Begic, G...... 160 Achene, L...... 274 Belyi, Y...... 27 Adam, P...... 156 Bender, J...... 143, 156 Adepoju, Y...... 77 Benson, R...... 292 Aebersold, R...... 169 Bentham, R...... 176, 183 Afshar, B...... 118 Benz, R...... 230 Ahmed-Lecheheb, S...... 298 Berge,M.´ ...... 37 Aili, M...... 154 Berger, F...... 181, 268 Airaksinen, P...... 198, 237 Berjeaud, J.-M...... 228, 230, 244, 262 Aitken, M...... 232 Berk, S.G...... 38, 43, 184 Akhter, A...... 159 Bernard, K...... 286 Aktories, K...... 27 Berne, F...... 216 Al Khodor, S...... 128 Bernier, M...... 224 Al-Hadithi, H...... 180 Beron, R...... 250 Al-Khodor, S...... 76 Berthelot, N...... 207, 216, 217 Al-Quadan, T...... 76 Berthelot, P...... 181, 268 Albert-Weissenberger, C...... 71, 72, 300, 303 Bettelli, R...... 218 Albrechtsen, H.-J...... 236 Bianchi, A...... 109 Alexander, N...... 11, 52, 243 Binet, M...... 197, 200, 201 Alexeev, J...... 212 Birteksoz,¨ S...... 106 Allan, D.S...... 38, 142 Bitkina, V.V...... 263, 287 Allegra, S...... 181, 268 Blaschitz, M...... 125 Allestam, G...... 17, 239 Bleeker, S...... 88, 89, 219 Almahmoud, I...... 80 Boccia, S...... 241 Almal, A...... 67 Bognar,´ C...... 214 Alvarez,´ I...... 81 Bolstad, K...... 124, 163 Alyapkina, Y.S...... 58, 212 Bonfrate, N...... 113, 192, 193 Amemura-Maekawa, J...... 188, 278, 283 Boninti, F...... 241 Amer, A...... 159 Borella, P...... 40, 245 Amore, R...... 241 Bouchier, C...... 61 Anacarso, I...... 245 Bouteleux, C...... 201 Angel, E...... 172 Bouton, S...... 215 Anne,´ J...... 157 Boyd, D...... 68 Arruga, M.V...... 81 Boyette, S...... 15 Arslan, H...... 86, 87, 271, 280 Brachet, E...... 293 Arslan-Aydogdu,˘ E.O.¨ ...... 234, 235 Brandsema, P...... 88 Ashford, G...... 239 Brassinga, A.K...... 166 Assaf Levi, A...... 301 Broich, M...... 143 Atlan, D...... 153 Brossier, F...... 116 Atsma, A...... 253, 254 Brown, A...... 110, 111 Aurass, P...... 146 Bruggemann,¨ H...... 72, 151 Azap, O.K...... 271 Bruhn, H...... 230 Baba, T...... 213 Bruin, J...... 88, 89, 185, 186, 219, 290 Baggiani, A...... 187 Bryan, A...... 22 Bakker, G...... 178 Buathong, R...... 220 Baladron, B...... 99, 108, 285 Buchrieser, C. . . 5, 23, 61, 64, 71, 72, 151, 262, 291, 293, Banerji, S...... 152 300, 302, 303, 304 Banga, S...... 22 Buonopane, G...... 218 Bangsborg, J.M...... 82, 103 Buss, M...... 292 Bannert, N...... 146 Cabodi, D...... 113, 192, 193 Barbe, V...... 291 Campese, C...... 101, 221, 276 Barbuti, S...... 205 Camponovo, L...... 225 Barna, Z...... 214 Cano, R...... 285 Barricarte, A...... 99 Capra, A.M...... 225 Bartfeld, S...... 152 Carratala,´ M.A...... 204 Baudart-Lenfant, J...... 206 Carrillo, J.A...... 111 Baume, M...... 215 Carter, L...... 62 Bayle, S...... 182 Casas, I...... 9, 97, 98 Bayo, J...... 204 Casati, S...... 185, 186, 189 - 307 - Case, C...... 30 Demina, Y...... 58 Casini, B...... 187 Den Boer, J...... 67, 88, 89, 185, 219, 290 Caspers, M...... 67, 254, 295 Dente, G...... 251 Castaldo, V...... 218 Dervins-Ravault, D...... 23 Castilla, J...... 99 Di Leo, F...... 251 Castillo, F.J...... 171 Di Marino, O...... 225 Catalan,´ V...... 266 Ditommaso, S...... 195, 227 Catarino, J...... 91 Dogru˘ oz,¨ N...... 196, 226, 233 Caugant, D.A...... 124 Doric, M...... 160 Caughley, B...... 199, 240 Doublet, P...... 130 Cazalet, C...... 61, 72, 293 Douglas, N...... 282 Cencetti, S...... 245 Drasar, V...... 83, 190, 191 Cevahir-Oz,¨ G...... 234 Drevinek, M...... 191 Chan, P...... 298 Droguet, J...... 250 Chan, S.H...... 167 Dubois, G...... 250 Chandrakumar, M...... 239 Dubourg, K...... 37 Chang, B...... 188, 278, 283 Dubow, M...... 197 Chaperon, G...... 247 Duncan, C...... 145, 281 Charpentier, X...... 73, 130 Duperrier, S...... 151 Chasqueira, M.-J...... 91, 279 Dupuy, M...... 216 Chatfield, C...... 144 Duran,´ E...... 171 Che, D...... 101, 221, 276 Dusserre, E...... 208 Cheape, G...... 231, 232 Edelstein, P.H...... 8 Chen, J...... 15 Edwards, G...... 110 Chidiac, C...... 101, 276 Egging, B...... 42 Choate, B...... 184 Ehricht, R...... 117 Chong, A...... 142 Eisenreich, W...... 147 Christensen, J.J...... 104 Eitel, J...... 32 Cianciotto, N...... 19, 134, 137, 138, 140, 144 Ejenguele, G...... 272 Collignon, A...... 269, 270 El Kanouni, F...... 97, 98 Consonni, M...... 109 Elgngihy, N...... 281 Conza, L...... 186, 189 Elola, C...... 108 Cook, A...... 90 Elverdal, P...... 84, 85, 105 Coppola, A...... 251 Endo, T...... 283 Coscolla,M.´ ...... 51 Engelhardt, S...... 168 Cosette, P...... 298 Engelmann, S...... 34, 63, 177 Costa, J...... 59 Erdogan, A...... 86, 87 C¸otuk, A...... 196, 226, 233 Erdogan, H...... 86, 87, 271, 280 Coulon, C...... 269, 270 Esperet, D...... 259 Cramp, G...... 282 Etienne, J. . 50, 61, 123, 182, 208, 249, 250, 276, 291, 293 Crespi, S...... 57, 222 Euser, S...... 88, 89, 219 Cristino, S...... 294 Ewann, F...... 148 Croize, J...... 92 Exner, M...... 47 Cupp, M...... 166 Eylert, E...... 147 Curran, E...... 231 Faber, C...... 230 D’Antonio, G...... 251 Falge, M...... 230 Da Costa, M...... 59 Fan,S...... 15 Dadkhah, N...... 104 Farhat, M...... 260 Daguier, N...... 259 Farone, A...... 184 Dalla Torre, F...... 241 Farone, M...... 43, 184 David, F...... 217 Faulkner, G...... 38 De Jager, P...... 164 Feliciano, J...... 91 De Jong, B...... 239 Ferguson, C...... 90 De Wit, B...... 164 Ferhat, M...... 153 Deacon, R...... 199, 240 Fernandes, T...... 91 Declerck, P...... 223 Fernandez Morera, E...... 129 Del Nord, P...... 241 Ferreira, T...... 262 Delattre, J.-M...... 224 Ferrer, J...... 222 Delgado, J.M...... 120 Fields, B...... 3, 15, 52, 62, 243, 292 Delgado-Viscogliosi, P...... 224 Finsel, I...... 169 Demarie, V...... 192 Flieger, A...... 21, 143, 146, 152, 156 - 308 - Fogle, P...... 62 Gorman, J...... 90 Folcher, M...... 301 Gormlie, D...... 231 Fontana, S...... 112, 274 Gosselin, F...... 172 Fontvieille, D...... 174 Gourabathini, P...... 148 Foret,ˆ C...... 260 Govil, D...... 292 Frace, M...... 292 Graham, F...... 10, 282 Fraile, D...... 120 Grattard, F...... 181, 268 Fraisse, P.-O...... 215 Gruas, C...... 81 Francius, G...... 172 Grubesic, T...... 160 Franco, I...... 127 Gruner,¨ I...... 125 Franzin, L...... 113, 192, 193 Guarini, P...... 215 Frenkiel, H...... 215 Gubbay, J...... 281 Frere,` J...... 260, 298 Guijarro Rodriguez, J.L...... 248, 252 Frese, F...... 177 Guna, R...... 204 Fresi, P...... 193 Gunderson, J...... 184 Freval, A...... 217 Gut,W...... 96 Frezza, G...... 245 Guyard, C...... 145, 281 Friesen, J...... 184 Ha, T.-L...... 200, 203, 209, 229 Fritzsønn, E...... 124 Haas, A...... 34, 63 Fry, N...... 114, 118, 190 Habyarimana, F...... 76, 128 Fuertes, A...... 120 Hacker, J...... 5, 300 Furtado, C...... 91 Hajmirbaba, E...... 92 Gabarre, J...... 116 Hakii, C...... 175 Gaboriaud, F...... 172 Hall, I...... 284 Gaia, V...... 16, 47, 185, 186, 189 Hanshoawarakul, W...... 220 Gaillard, J.-L...... 293 Hargrave, I...... 84 Galka, F...... 34, 177 Harrison, T...... 14, 53, 114, 118, 239 Galli, M.G...... 109 Harte, D...... 10, 199, 282 Garc´ıa, Concepcion´ ...... 171 Hartemann, P...... 47 Garc´ıa, Cristina ...... 81 Hartland, E...... 23, 66, 154, 155 Garc´ıa-Cenoz, M...... 99 Hasenberger, P...... 125 Garc´ıa-Nunez,˜ M...... 9, 98, 194 Hasimoglu, R...... 271 Garduno, R.A...... 38, 142 Hawn,T...... 29 Garner, C...... 184 Hayashi, T...... 135 Garrec, N...... 216 Hechard, Y...... 216, 228, 230, 262 Garrelly, L...... 182, 215 Hecker, M...... 63 Garrett, N...... 84 Heidtman, M...... 77 Garusi, F...... 225 Helbig, J.H...... 129, 162, 173, 278, 290 Garusi, G...... 225 Herbelin, P...... 216 Gendron, C...... 184 Herrmann, J.-L...... 293 Gentile, M...... 195, 227 Herrmann, V...... 147 Gerard,´ A...... 121 Hervet, E...... 130 Gestin, B...... 92 Herwaldt, L...... 49 Giacomuzzi, M...... 195, 227 Heuner, K. 5, 34, 63, 71, 117, 133, 143, 147, 156, 300, 303 Giao,˜ M.S...... 170 Hicks, L...... 11, 52 Gil, V...... 108 Higa, F...... 156 Gilbert, C...... 153 Hilbi, H...... 35, 168, 169, 304 Gimeno, C...... 204 Hoffman, Paul S...... 25 Ginevra, C...... 50, 123, 172, 208 Hoffman, Paul S...... 148, 150, 166 Gintsburg, A.L...... 287 Høiby, A...... 161 Girardin, N...... 228 Holland, G...... 146 Giria, J...... 91 Hood, J...... 231, 232 Gjørup, I...... 82 Hori, J.I...... 149 Gobin, I...... 160 Horta, C.V...... 149 Goksay,¨ D...... 196, 233 Horton, K...... 257 Gomes, C...... 91 Horvath,´ J.K...... 214 Gomez Valero, L...... 293 Host, H...... 215 Gomez-Lus,´ R...... 171 Hu, B...... 275 Gomez-Valero, L...... 61, 64, 291 Huber, P...... 255 Gonzalez,´ L...... 120 Iamsirithaworn, S...... 220 Gonzalez-Candelas,´ F...... 51 Iatta, R...... 205 - 309 - Ichinose, M...... 283 Kozak, N...... 62, 292 Ihle, Ø...... 161 Kozioł-Montewka, M...... 305 Ilhan Sungur, E...... 196, 233 Krogfelt, K.A...... 236 Imai, Y...... 135 Krøjgaard, L.H...... 236 Indra, A...... 125 Kuhn,¨ J...... 129 Inoue, N...... 213 Kumar, D.Y...... 296 Irisarre, F...... 99 Kunda, M.S...... 263, 287 Isberg, R...... 68, 77 Kura, F...... 188, 278, 283 Isobe, J...... 278, 288 Kurata, T...... 288 Ivanov, S...... 131 Kusnetsov, J...... 198, 237 Iversen, O.-J...... 119 La Mura, S...... 238 Iwasaki, Y...... 135 Labanowski, J...... 262 Jacobs, T...... 184 Lacombe, C...... 230, 244, 262 Jakubek, D...... 197 Lai, S...... 53, 239 Jameson-Lee, M...... 150 Lakadamyali, H...... 86 Jamieson, A...... 30 Landelle, C...... 249 Jandos, A.-M...... 207 Lang, C...... 152 Jarlier, V...... 116 Lara, C...... 81 Jarraud, S. 50, 61, 101, 123, 172, 182, 208, 215, 221, 249, Lau, R...... 199, 240 250, 276, 291, 293 Laurent, F...... 207 Jenal, U...... 301 Laurenti, P...... 241 Jeong, K.C...... 132 Lautner, M...... 133, 147 Johnson, A...... 239 Lawrence, C...... 293 Johnson, R...... 15 Lazzaroni, J.-C...... 153 Jones, G...... 114 Le Cann, P...... 121 Joppolo, C.M...... 238 Le-Brun, M...... 197, 200, 201 Jordan,´ L...... 204 Leach, S...... 284 Jørgensen, C.S...... 85 Lebaron, P...... 206 Jørgensen, N.D...... 82 Leblon, G...... 197 Joseph, C...... 4, 45, 53, 100, 239, 264 Lechat, P...... 64 Jouenne, T...... 298 LecsO-Bornet,¨ M...... 116 Jules, M...... 23, 72, 151 Lee, H.K...... 202 Kad´ ar,´ M...... 214 Lee, J.V...... 47, 53, 239 Kafatos, G...... 264 Lee, R...... 15 Kalia, A...... 76 Legnani, P.P...... 294 Kamantigue, E...... 15 Leitner, G...... 125 Kanatani, J...... 288 Lemming, L...... 103 Kaneko, N...... 278 Lemon, D...... 284 Kanestrøm, A...... 163 Leoni, E...... 294 Kang, Y.H...... 202 Lerat, I...... 116 Kanneganti, T...... 159 Leslie, J...... 62 Karpova, T.I...... 58, 212, 263, 287 Levet, M...... 130 Kato, N...... 175 Li,J...... 15 Kauppinen, A...... 198 Lima, C...... 38 Kawano, K...... 278 Lin, Y...... 242 Kay, E...... 80, 123 Lindberg, E.H...... 163 Keevil, C.W...... 170 Lindh, J...... 17 Kese, D...... 115 Lindsay, D...... 110, 111 Khemiri, A...... 298 Lindsten, T...... 30 Kim, W.-S...... 202 Lippmann, J...... 32 Kimata, K...... 288 Lobez,´ S...... 171 Kimiran-Erdem, A...... 234, 235 Loftus, B...... 69 Kimmitt, P...... 118 Lohan, A...... 69 Kinchen, J...... 166 Loisy-Hamon, F...... 121 Kingham, S...... 10 Lomma, M...... 23, 61, 72 Kjærgaard, M...... 93 Lorenz, U...... 156 Klaine, S...... 246 Low, D.E...... 145, 281 Kogoj, R...... 115 Lowe, M...... 43 Koike, H...... 135 Lucas, C...... 52, 243, 292 Kolberg, J...... 161 Lucibello, T...... 251 Kopilovic, B...... 115 Luck,¨ C...... 13, 47, 117 - 310 - Luck, P.C...... 210, 263 Nakajima, H...... 278 Lundstrom,¨ J...... 17 Napoli, C...... 205 Lunin, V.G...... 263 Nascimento, M...... 91, 279 Luo, Z.-Q...... 22 Nasrallah, G...... 142 Magliocca, M...... 218 Nasu, M...... 213, 265 Magrys,´ A...... 305 Ndayo Wouafo, M...... 272 Maier, H...... 230 Newton, H...... 23, 154 Maine, C...... 221 Nguyen, T.L...... 139 Marchand, A...... 244 Nicolle, M.-C...... 249 Marchegiano, P...... 245 Nielsen, S.S...... 93 Marchesi, I...... 245 Nogareda, F...... 285 Marin, M...... 293 Nogueira, C...... 30 Marino, M...... 241 Nor, A...... 119 Marques, T...... 91, 279 Norden, I...... 261 Marsico, T...... 205 Normand, P...... 123 Martin, C...... 285 Novokshonova, I.V...... 212 Mart´ınez, L...... 266 Nukina, M...... 278 Massis, L.M...... 149 Nunez,˜ R...... 9, 97, 98 Mateu, L...... 9, 97, 98, 194 Nzouankeu, A...... 272 Mathieu, L...... 172, 203, 229 O’Connor, T...... 68, 77 Maurice-Blanc, C...... 174 Oberlechner, A...... 255 Maurin, M...... 80, 92, 123 Oberti, S...... 207, 216, 217 Mazoua, S...... 216 Oca, M...... 171 Mazure, C...... 208, 249 Ocete, M.D...... 204 McCoy-Simandle, K...... 134 Ofte, B.W...... 119 McDonnell, G...... 269, 270 Ohno, A...... 175 McNealy, T...... 246 Olesen, H.V...... 105 Mead, A...... 231, 232 Oliver, I...... 264 Medigue, C...... 64, 291 Ollevier, F...... 223 Mehdi, J...... 180 Olsen, C.W...... 93, 94, 95 Meier, R...... 168 Olsen-Rasmussen, M...... 292 Melada, S...... 225 Opitz, B...... 32 Mentasti, M...... 114, 118 Orsini, M...... 294 Mentula, S...... 237 Ossendrijver, M...... 253, 254, 295 Merault,´ N...... 293 Ott, C...... 133 Merchat, M...... 38, 247 Ottaviani, M...... 274 Messi, P...... 245 Pagano, M...... 251 Meyer, T.F...... 152 Palepou-Foxley, C...... 118 Miettinen, I...... 198 Pancer, K...... 96, 102, 273 Minarro˜ Del Moral, R...... 248, 252 Pangon, B...... 47 Mineo, D...... 112, 274 Pardo, T...... 120 Mira, E...... 204 Parthuisot, N...... 206 Miyake, M...... 135 Pearce, M...... 137 Moletta-Denat, M...... 260 Pecastaings,´ S...... 37 Molmeret, M...... 208 Pedro-Botet, M.L...... 9, 97, 98, 194 Monecke, S...... 117 Pelaz, C...... 99, 108, 171, 285 Montagna, M.T...... 205 Pereira, M.D.S.F...... 149 Montijn, R...... 67 Petrosino, A...... 251 Moore, M...... 52 Peyron, C...... 247 Morel, Y...... 209 Phin, N...... 264 Morelot-Panzini, C...... 116 Pinci, F...... 112, 274 Morgantetti, G.F...... 149 Pinon, A...... 207 Moroder, L...... 255 Pires-Cronenberger, S...... 101, 276 Moszer, I...... 64, 302 Pitera,` L.A...... 238 Mueller, C.A...... 136 Pitkanen,¨ T...... 198 Muller,¨ C...... 129 Plasencia, A...... 55 Munafo,` E...... 241 Pless, B...... 146 Murray, L.E...... 142 Poirier, R...... 276 Murtula,´ R...... 266 Polcar, R...... 190 Nagelkerke, N...... 219 Pourcel, C...... 121, 281 Naik, F...... 53 Powrie, I...... 231 - 311 - Pozzetto, B...... 181, 268 Saleh, A...... 62 Prammer, W...... 125 Salvo, S...... 171 Prashar, A...... 145 Samsom, F...... 23 Prendergast, J...... 69 Sanli Yurudu, N.O...... 233, 235, 261 Price, C...... 76, 128 Sansom, F.M...... 155 Privitera, G...... 187 Santic, M...... 76, 128 Quaranta, G...... 241 Savva, D...... 81 Ragaz, C...... 168, 169 Scaturro, M...... 112, 274 Ragull, S...... 194 Schmid, D...... 125 Rallo, A...... 204 Schmidt, A...... 169 Rastew, E...... 152 Schneeberger, P...... 164 Ratcliff, R...... 56 Schneider, D...... 80, 123 Raymond, M...... 258 Schousboe, M...... 282 Reichardt, K...... 162 Schuelein, R...... 154 Reimer, A...... 286 Schulz, T...... 71, 303 Remer, K...... 156 Schumacher, C.-J...... 173 Rentenaar, R...... 164 Schunder, E...... 143, 156 Rey-Joly, C...... 9, 97, 98 Schur, C...... 250 Reyrolle, M...... 208, 215, 249, 250 Schuren, F...... 67, 253, 254, 295 Ricci, M.L...... 47, 112, 274 Sedgwick, J...... 239 Ricciardi, G...... 241 Seeber, M...... 255 Ricketts, K...... 45, 100, 264 Sezzatini, R...... 241 Riedmaier, P...... 155 Shelton, B...... 62 Riffard, S...... 181, 268 Shen, X...... 22 Riley,T...... 90 Shevchuk, O...... 34, 63 Riveroll, A...... 142 Shih, H.-Y...... 242 Robert, N...... 9, 97 Shim, J.I...... 202 Robine, E...... 203, 209, 229, 260 Shima, T...... 288 Robl, B...... 125 Shin, S...... 30, 32 Rodier, M.-H...... 216 Shohdy, N...... 127 Rodrigues, L...... 279 Shrapnel, R...... 256, 257 Rodr´ıguez-De La Rosa, I...... 111 Shuman, H...... 36, 73, 127, 136, 301 Roig, B...... 182 Sifri, C...... 166 Rojas, A...... 120 Sikora, A...... 305 Rojas, J...... 120 Silveira, T...... 31 Roques, C...... 37 Simola, L...... 237 Ross, K...... 176 Simonet, J...... 172, 203 Rossi, A.M...... 218, 251 Simonsen, Ø...... 163 Rossi, P...... 255 Skovsted, I.C...... 84 Rouil, L...... 209 Slickers, P...... 117 Rousseau, S...... 302 Slimani, S...... 208 Rouy, Z...... 64, 291 Sloan, J...... 154 Roy, C...... 30, 32, 75, 131 Sobral, D...... 121 Rubio Garcia, A...... 248, 252 Solignac, L...... 224 Rubio, C...... 171 Sopena, N...... 9, 97, 98 Ruggenini Moiraghi, A...... 227 Soreau, S...... 200, 216 Ruijs, H...... 88 Soria, E...... 266 Rummelhard, M...... 92 Spirig, T...... 304 Rusniok, C...... 61, 151, 291, 293 Spitaler, G...... 255 Russo, G...... 218 Stahl, M...... 27 Russo, M...... 241 Stanco, M...... 218 Russo, R...... 218 Stangier, D.K.A...... 296 Ruutu, P...... 237 Steinert, M...... 5, 34, 63, 156, 177, 230 Rydzewski, K...... 143, 146 Stenico, A...... 255 Sabria,` M...... 7, 9, 47, 97, 98, 194 Stewart, C...... 138 Sadler-Reeves, L...... 239 Stojak, A...... 246 Sadretdinova, O.V...... 212, 263 Stout, J.E...... 46 Saels, V...... 157 Stuart, J...... 264 Sahr, T...... 23, 71, 72, 262, 300, 302, 304 Study, G...... 101 Salaris, S...... 294 Stypulkowska-Misiurewicz, H...... 96, 102 Salazar, A...... 204 Sugiyama, K...... 188 - 312 - Surman-Lee, S...... 47, 53 Van Buynder, P...... 90 Susa, M...... 160 Van Der Hoek, W...... 89 Sutherland, M...... 139 Van Der Kooij, D...... 42, 178 Suttorp, N...... 32 Van Der Meer, W...... 67 Suzuki-Hashimoto, A...... 283 Van Laak, V...... 32 Swanson, M...... 22, 26 Vanhems, P...... 101, 249, 276 Sykes, K...... 282 Vanysacker, L...... 223 Szax, A...... 214 Vargha, M...... 214 Sze, C.C...... 167 Vaudry, D...... 298 Szotkowski, T...... 83 Ventadour, P...... 302 Tada, Y...... 278 Vepsal¨ ainen,¨ A...... 198 Taguri, T...... 188 Verdon, J...... 230, 244, 262 Tang, P...... 145, 281 Vergnaud, G...... 121 Tao, L...... 275 Vergnes, M...... 123 Taran, A...... 122 Ver´ıssimo, A...... 59 Tarnaud, E...... 209 Veschetti, E...... 274 Tartakovskiy, I.S...... 58, 263, 287 Vialette, M...... 207 Tartakovsky, I.S...... 212 Vianney, A...... 130 Taylor, G.A...... 32 Viboud, S...... 174 Taylor, M...... 176, 183 Vidal, A...... 207, 217 Taylor, T...... 52, 243 Vincent, C...... 139 Tchipeva, D...... 118 Vitko, S...... 148 Temezhnikova, N...... 122 Vogel, J...... 20, 132, 139 Terashita, D...... 52 Voronina, O.L...... 58, 263, 287 Terebiznik, M...... 145 Vranckx, L...... 157 Tesauro, M...... 109 Vu,T...... 184 Thiery, L...... 259 Vuckovic, D...... 160 Thioulouse, J...... 123 Wai, S.N...... 34, 177 Thomas, V...... 269, 270 Wallensten, A...... 264 Thomas, W...... 256, 257 Wallet, F...... 216 Thompson, C...... 30 Wang, M...... 105 Tiaden, A...... 304 Watahiki, M...... 288 Tiago, I...... 59 Watanabe, H...... 188, 278, 283 Tiefenau, J...... 177 Watanabe, Y...... 278 Tijet, N...... 281 Wedege, E...... 124, 163 Todorova, V...... 258 Weinbreck, P...... 101 Tognet, F...... 209 Weinstein, P...... 90 Tonolla, M...... 16 Weissenmayer, B...... 69 Torii, M...... 265 Wever,P...... 164 Torracca, F...... 187 Wewalka, G...... 125 Torvinen, E...... 198 Wewers, M...... 159 Touron-Bodilis, A...... 206 Whiley, H...... 183 Tran Minh, N.N...... 237 White, J...... 43 Travis, T...... 15, 52, 292 White, P...... 10 Treilles, M...... 259 Wilkinson, P...... 100 Trobonjaca, Z...... 160 Willeford, W...... 43 Trouilhe,´ M.-C...... 260 Wills, G...... 231, 232 Tseng, V...... 139 Witherell, L...... 52 Turan, H...... 271 Wojtowicz,´ M...... 305 Turetgen,¨ I...... 233, 261 Worzel, B...... 67 Turmeau, C...... 209 Wullings, B...... 42, 178 Turnaturi, C...... 241 Xiao, W...... 15 Uebayashi, Y...... 265 Xu,L...... 22 Uldum, S.A...... 84, 85, 93, 94, 95, 103, 104, 105, 236 Xu,R...... 15 Underwood, A...... 114 Xu,W...... 15 Ungchusak, K...... 220 Yılmaz, A...... 86 Unger, C...... 210 Yadav,R...... 45 Urwyler, S...... 169 Yamagauchi, N...... 265 Vacherie, B...... 291 Yamaguchi, K...... 175 Valentini, P...... 187 Yamaguchi, N...... 213 Valster, R...... 42, 178 Yanez,˜ M.A...... 266 - 313 - Yang, K...... 15 Yzerman, E...... 67, 88, 89, 185, 219 Yang,P...... 15 Zamboni, D.S...... 31, 149 Yip, E...... 140 Zeybek, Z...... 106, 226, 235 Yoshida, S.-I...... 41 Zhao, Z...... 275 Yu, J.-Y...... 202 Zidane, N...... 61 Yu,V.L...... 46 Zotti, C...... 195, 227 Yuzbas¸ıo¨ glu,˘ E...... 234 - 314 - SUMMARY - 315 - Oral presentations Tuesday, October 13th, 2009 Opening Session 6:00 pm Environmental approaches to the prevention of legionellosis 3 Barry Fields 6:30 pm Legionnaires’ disease in Europe 1995 - 2008: Trends and Challenges 4 Carol Joseph 7:00 pm Legionella: A Model Organism for Genomics and Cellular Microbiology 5 Jorg¨ Hacker, Klaus Heuner, Michael Steinert and Carmen Buchrieser Wednesday, October 14th, 2009 Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment 8:30 am Clinical aspects of Legionnaires’Disease 7 Miquel Sabria` 9:00 am Antimicrobial therapy for Legionnaires’ disease, still in need of 8 improvement Paul H. Edelstein 9:30 am Legionella pneumonia (LP) in immunosuppressed (IS) patients. 9 Maria Luisa Pedro-Botet, Raquel Nunez,˜ Nieves Sopena, Lourdes Mateu, Irma Casas, Neus Robert, Marian Garc´ıa-Nunez,˜ Celestino Rey-Joly and Miquel Sabria` 9:45 am Changing Epidemiologic Trends of Legionellosis in New Zealand, 1979 - 10 2008 Frances Graham, Simon Kingham, Paul White and David Harte 10:00 am Travel-Associated Legionnaires’ Disease Surveillance in the United 11 States, 2005-2008 Nicole Alexander and Lauri Hicks Session 2: Clinical aspects a) detection and subtyping 10:45 am Microbiological diagnostic methods for the detection of Legionella 13 infections Christian Luck¨ 11:15 am Legionella pneumophila DNA sequence-based typing: progress, pitfalls 14 and perspectives. Timothy Harrison 11:45 am Evaluation of a fully integrated point-of-testing system for the detection 15 of viable Legionella pneumophila Patrick Yang, Barry Fields, Tatiana Travis, Randy Johnson, Richard Lee, Edgar Kamantigue, Sara Fan, Scott Boyette, Rong Xu, Jing Chen, Weiqing Xu, Weimin Xiao, Kechao Yang and Jie Li 12:00 am MALDI-TOF MS based protein mass fingerprinting: a rapid and reliable 16 method for the identification of Legionella spp. Valeria Gaia and Mauro Tonolla 12:15 am An improved and rapid sequence-based method for typing of Legionella 17 pneumophila. Johan Lindh, Jonas Lundstrom¨ and Gorel¨ Allestam - 316 - Session 3: Host - Microbe interactions: a) secretion systems and their substrates 3:30 pm Type II protein secretion, siderophores, and pigment and what they mean 19 for Legionella ecology and pathogenesis Nicholas Cianciotto 3:55 pm The Legionella Dot/Icm type IVB secretion system. 20 Joseph Vogel 4:20 pm Legionella pneumophila phospholipases and their mode of secretion 21 Antje Flieger 4:45 pm Inhibition of Host Vacuolar H+-ATPase Activity by a Legionella 22 pneumophila effector Li Xu, Xihui Shen, Andrew Bryan, Simran Banga, Michele Swanson and Zhao-Qing Luo 5:00 pm Legionella pneumophila infections involve bacterial F-box proteins 23 Mariella Lomma, Delphine Dervins-Ravault, Hayley Newton, Fiona Sam- som, Matthieu Jules, Tobias Sahr, Elizabeth Hartland and Carmen Buchrieser Session 4: Host - Microbe interactions: b) metabolism and virulence 5:45 pm The Legionella pneumophila Developmental Cycle: Role in Survival and 25 Pathogenesis Paul S. Hoffman 6:15 pm Legionella pneumophila virulence as a response to starvation 26 Michele Swanson 6:45 pm Cytotoxic glucosyltransferases of Legionella pneumophila. 27 Yury Belyi, Michael Stahl and Klaus Aktories Thursday, October 15th, 2009 Session 5: Immunology and host susceptibility 8:45 am Genetic variation and the pulmonary innate immune response to 29 Legionella Tom Hawn 9:15 am Dendritic cells rapidly undergo apoptosis to limit intracellular replica- 30 tion of Legionella pneumophila Catarina Nogueira, Tullia Lindsten, Amanda Jamieson, Christopher Case, Sunny Shin, Craig Thompson and Craig Roy 9:30 am Pore formation is an Nlrc4/Ipaf inflammasome-dependent host cell 31 response against virulent Legionella species that express flagellin Tatiana Silveira and Dario S. Zamboni 9:45 am Type I IFNs mediate innate intracellular defense to Legionella pneu- 32 mophila Juliane Lippmann, Vincent Van Laak, Julia Eitel, Sunny Shin, Gregory A Taylor, Norbert Suttorp, Craig Roy and Bastian Opitz Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms 10:30 am Proteomic and functional characterization of outer membrane vesicles 34 and Legionella-containing phagosomes Michael Steinert, Olga Shevchuk, Frank Galka, Albert Haas, Klaus Heuner, Sun Nyunt Wai and Susanne Engelmann 11:00 am Signals produced and exploited by Legionella 35 Hubert Hilbi - 317 - 11:30 am Host determinants of Legionella infection 36 Howard Shuman 12:00 am A new method of producing Legionella pneumophila biofilms in a 37 minimal medium Sophie Pecastaings´ , Mathieu Berge,´ Karine Dubourg and Christine Roques 12:15 am Mature infectious forms (MIFs) and MIF-pellets as factors in the 38 transmission of Legionnaires’ disease Rafael A. Garduno, Celia Lima, Gary Faulkner, Sharon G. Berk, Michele Merchat and David S. Allan Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence 3:00 pm Risk factors for Legionella contamination and persistence in potential 40 sources of infection Paola Borella 3:30 pm Legionella physiology in the environment 41 Shin-Ichi Yoshida 4:00 pm Detection of hosts for Legionella pneumophila in freshwater by 42 enrichment of free-living protozoa in a biofilm-batch system Rinske Valster, Bastiaan Egging, Bart Wullings and Dick Van Der Kooij 4:15 pm Long-term survival of novel amoeba-associated bacteria compared with 43 Legionella pneumophila under conditions of hydration and desiccation Sharon G. Berk, Mary Farone, Wesley Willeford, James White and Michelle Lowe Session 8: Legionella prevention and control 4:50 pm Travel associated Legionnaires’ disease in Europe: Reporting and 45 response Kate Ricketts, Rekha Yadav and Carol Joseph 5:20 pm Legionella and the Conundrum of Low Risk Buildings: a Very Modest 46 Proposal Janet E. Stout and Victor L. Yu 5:50 pm An international trial of quantitative PCR for monitoring Legionella in 47 artificial water systems John V. Lee, Martin Exner, Valeria Gaia, Phillipe Hartemann, Christian Luck,¨ Beatrice´ Pangon, Maria Luisa Ricci, Miquel Sabria` and Susanne Surman-Lee Friday, October 16th, 2009 Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals 8:30 am Decontamination of Hospital Water Supplies: A Review of the Literature 49 and the University of Iowa’s Experience Loreen Herwaldt 9:00 am Nosocomial infections, Legionella strain characterization and host- 50 related risk factors Christophe Ginevra, Sophie Jarraud and Jer´ omeˆ Etienne 9:30 am Multiple Legionella strains infection detected by sequence analysis 51 directly from respiratory samples Mireia Coscolla´ and Fernando Gonzalez-Candelas´ - 318 - 9:45 am The impact of monochloramine introduction on Legionella colonization 52 in a hospital potable water system. Lauri Hicks, Tatiana Travis, Linden Witherell, Nicole Alexander, Thomas Taylor, Dawn Terashita, Matthew Moore, Claressa Lucas and Barry Fields 10:00 am Lessons learnt from investigating hospital-acquired legionellosis and the 53 remedial actions in England. Susanne Surman-Lee, John V. Lee, Sandra Lai, Falguni Naik, Carol Joseph and Timothy Harrison Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy 10:45 am Managing public health crises: from Legionellosis to pandemic influenza 55 Antoni Plasencia 11:15 am Ecological diversity within Legionella 56 Rod Ratcliff 11:45 am Legionellosis Prevention Guidelines: Better, but Not Well 57 Sebastian Crespi 12:00 am From large community outbreak in Verhnaya Pyshma to effective 58 prevention of legionellosis in Russia. Igor S. Tartakovskiy, Yulia Demina, Olga L. Voronina, Yulia S. Alyapkina and Tatiana I. Karpova 12:15 am Molecular evolution of dotA gene in Legionella pneumophila: contribu- 59 tion of natural environmental isolates. Joana Costa, Igor Tiago, Milton Da Costa and Antonio´ Ver´ıssimo Session 11: Genomics and Comparative genomics 3:30 pm Legionella pneumophila and Legionella longbeachae pathogenesis: new 61 insights gained form comparative and functional genomics Christel Cazalet, Laura Gomez-Valero, Mariella Lomma, Christophe Rus- niok, Nora Zidane, Christiane Bouchier, Sophie Jarraud, Jer´ omeˆ Etienne and Carmen Buchrieser 4:00 pm Polymorphic loci in Legionella pneumophila serogroup 1 62 Brian Shelton, Paul Fogle, Barry Fields, Natalia Kozak, Laurel Carter, Amgad Saleh and John Leslie 4:15 pm Proteomic analysis of a Legionella-containing phagosome in Dic- 63 tyostelium Olga Shevchuk, Susanne Engelmann, Michael Hecker, Albert Haas, Klaus Heuner and Michael Steinert 4:30 pm Genome re-annotation and web resources for Legionella pneumophila 64 species Laura Gomez-Valero, Pierre Lechat, Zoe Rouy, Claudine Medigue, Carmen Buchrieser and Ivan Moszer Session 12: Functional genomics 5:15 pm Hidden protein treasures revealed in the Legionella genome 66 Elizabeth Hartland 5:45 pm Epidemiological genome analysis of a large Dutch Legionella pneu- 67 mophila strain collection identifies five markers highly correlated with pathogenic strains Ed Yzerman, Jeroen Den Boer, Martien Caspers, Arpit Almal, Bill Worzel, Walter Van Der Meer, Roy Montijn and Frank Schuren - 319 - 6:00 pm Solving functional redundancy amongst effectors of the bacterial 68 pathogen, Legionella pneumophila Tamara O’Connor, Dana Boyd and Ralph Isberg 6:15 pm Analysis of the transcriptome of Legionella pneumophila under infection 69 conditions using RNA.seq Barbara Weissenmayer, James Prendergast, Amanda Lohan and Brendan Lof- tus Saturday, October 17th, 2009 Session 13: Gene regulation in the environment and the host 9:00 am The Flagellar Regulon of Legionella pneumophila and the Expression of 71 Virulence Traits Christiane Albert-Weissenberger, Tino Schulz, Tobias Sahr, Carmen Buchrieser and Klaus Heuner 9:30 am Two small ncRNAs jointly govern virulence and transmission in 72 Legionella pneumophila Tobias Sahr, Holger Bruggemann,¨ Matthieu Jules, Christel Cazalet, Mariella Lomma, Christiane Albert-Weissenberger and Carmen Buchrieser 9:45 am Parasexual behavior in response to genotoxic stress in Legionella 73 pneumophila Xavier Charpentier and Howard Shuman Session 14: Cell biology - protozoa - macrophages 10:45 am Establishment of a vacuole that supports Legionella pneumophila 75 replication. Craig Roy 11:15 am Exploitation of conserved eukaryotic pathways by L. pneumophila 76 Chris Price, Souhaila Al-Khodor, Tasneem Al-Quadan, Marina Santic, Fabien Habyarimana, Awdhesh Kalia and Yousef Abu Kwaik 11:45 am Doubling up as a way to combat resilience 77 Ralph Isberg, Tamara O’Connor, Yewande Adepoju and Matthew Heidtman - 320 - Poster presentations Session 1: Epidemiology and Clinical aspects a) disease, diagnosis and treatment P1 Mutational paths towards increased fluoroquinolone resistance in Legionella 80 pneumophila Iyad Almahmoud, Elisabeth Kay, Dominique Schneider and Max Maurin P2 Evaluation of a fluorescence-based method for the rapid detection of Legionella 81 in environmental water samples Cristina Gruas, Isidro Alvarez,´ Carlos Lara, Cristina Garc´ıa, Demetris Savva and M. Victoria Arruga P3 Legionella infection - a survey in the Greater Copenhagen area from 2000-2007 82 Nicolai D. Jørgensen, Jette M. Bangsborg and Ida Gjørup P4 Atypical Legionella maceachernii Infection in a Severly Immunocompromised 83 Patient. Vladimir Drasar and Tomas Szotkowski P5 Early ”point-of-care” diagnosis of Legionella infection by detection of IgM 84 with Lateral flow test Pernille Elverdal, Ian C. Skovsted, Nina Garrett, Iain Hargrave and Søren A. Uldum P6 In-house ELISA for detection of IgM and IgG to Legionella pneumophila 85 serogroup 1, 3 and 6 as a routine test Pernille Elverdal, Charlotte S. Jørgensen and Søren A. Uldum P7 Travel associated legionnaires disease: Clinical features of 17 cases 86 Haluk Erdogan, Askin Erdogan, Huseyin Lakadamyali, Aynur Yılmaz and Hande Arslan P8 Legionnaire’s disease presenting with travellers diarrhea and resulting acute 87 respiratory distress syndrome: A case report Haluk Erdogan, Askin Erdogan and Hande Arslan P9 Legionella Source Identification in the Netherlands: 2007-2008 88 Sjoerd Euser, Petra Brandsema, Helma Ruijs, Ed Yzerman, Jacob Bruin, Sacha Bleeker and Jeroen Den Boer P10 Saunas and Legionnaires’ disease in the Netherlands 89 Sjoerd Euser, Wim Van Der Hoek, Ed Yzerman, Jacob Bruin, Sacha Bleeker and Jeroen Den Boer P11 Variations in Legionella seroprevalence across Australia - a national serology 90 study Jessica Gorman, Paul Van Buynder, Chantal Ferguson, Angus Cook, Thomas Riley and Philip Weinstein P12 Epidemiological Surveillance of Legionnaires’ Disease in Portugal, 2004-2008 91 Teresa Fernandes, Carlos Gomes, Judite Catarino, Cristina Furtado, Maria- Jesus Chasqueira, Marta Nascimento, Joao˜ Feliciano, Jose´ Giria and Teresa Marques P13 Predominance of L. pneumophila serogroup 1 ST-23 in Grenoble, France. 92 Elahe Hajmirbaba, Marilyne Rummelhard, Brieuc Gestin, Jacques Croize and Max Maurin P14 Occurrence of Legionella non-pneumophila species in Denmark. 93 Christina Wiid Olsen, Mette Kjærgaard, Sanne Søgaard Nielsen and Søren A. Uldum - 321 - P15 An evaluation of the sensitivity of The Xpect Legionella test - an 94 immunochromatic test for detection of Legionella pneumophila antigen in human urine samples. Christina Wiid Olsen and Søren A. Uldum P16 Legionella urinary antigen test as a routine diagnostic assay, one year 95 experiences from Denmark Christina Wiid Olsen and Søren A. Uldum P17 Cross-reactions in ELISA IgM tests for Legionella pneumophila and Bordetella 96 pertussis. Katarzyna Pancer, Wlodzimierz Gut and Hanna Stypulkowska-Misiurewicz P18 Trends observed in Legionnaires’disease (LD) in a hospital of Calatonia 97 (Spain) (1983-2008) Irma Casas, Raquel Nunez,˜ Maria Luisa Pedro-Botet, Lourdes Mateu, Nieves Sopena, Neus Robert, Fatima El Kanouni, Celestino Rey-Joly and Miquel Sabria` P19 Has the Legionella urinary antigen test changed Legionella pneumonia? 98 Lourdes Mateu, Maria Luisa Pedro-Botet, Raquel Nunez,˜ Nieves Sopena, Fatima El Kanouni, Irma Casas, Marian Garc´ıa-Nunez,˜ Celestino Rey-Joly and Miquel Sabria` P20 Recurrent Infection Due to Legionella pneumophila serogroup 1 in an 99 immunocompromised patient Carmen Pelaz, Manuel Garc´ıa-Cenoz, Beatriz Baladron, Aurelio Barricarte, Jesus´ Castilla and Fatima´ Irisarre P21 Wet cooling systems as a source of sporadic Legionnaires’ disease: a 100 geographical analysis of data for England and Wales, 1996 to 2006 Kate Ricketts, Carol Joseph and Paul Wilkinson P22 Factors Associated with in-Hospital Mortality of Legionnaires’ Disease (LD): 101 A Prospective Multicenter Study of 595 Patients (pts) in France. Christian Chidiac, Silene Pires-Cronenberger, Pierre Weinbreck, Didier Che, Christine Campese, Sophie Jarraud, Philippe Vanhems and Group Study P23 Legionellosis in Poland 1997-2009. The Epidemiology diffent from EU 102 Countries Hanna Stypulkowska-Misiurewicz and Katarzyna Pancer P24 Report of two cases each with two separate episodes of Legionnaires’ disease 103 Søren A. Uldum, Jette M. Bangsborg and Lars Lemming P25 Isolation of Legionella bozemanii from a lymph node in a healthy young woman 104 Søren A. Uldum, Naser Dadkhah and Jens Jørgen Christensen P26 A case of recurrent meningitis associated with positive Legionella urinary 105 antigen test and seroconversion to Legionella pneumophila Søren A. Uldum, Hanne Vebert Olesen, Mikala Wang and Pernille Elverdal P27 Post-antibiotic effect of various antibiotics on Legionella pneumophila strains 106 isolated from environmental samples Seher Birteksoz¨ and Zuhal Zeybek Session 2: Clinical aspects a) detection and subtyping P28 Evaluation of a PCR assay to detect Legionella pneumophila in respiratory 108 samples Beatriz Baladron, Virginia Gil, Consuelo Elola and Carmen Pelaz P29 Occurrence of Legionella pneumophila sequence types circulating in clinical 109 and environmental areas of Northern Italy and comparison with other regions of the world Annalisa Bianchi, Marina Tesauro, Michela Consonni and Maria Gabriella Galli - 322 - P30 Legionella longbeachae Sg 1 infections linked to Potting Compost 110 Diane Lindsay, Alistair Brown and Giles Edwards P31 Evaluation of an oligochromatographic test for Legionella pneumophila 111 detection in respiratory samples Diane Lindsay, Alistair Brown, Jose A Carrillo and I Rodr´ıguez-De La Rosa P32 Molecular typing of Legionella Pneumophila strains isolated in Italy from 1987 112 to 2008: preliminary results Stefano Fontana, Maria Luisa Ricci, Desiree´ Mineo, Federica Pinci and Maria Scaturro P33 Detection of Legionella DNA by Real-Time PCR in Respiratory Samples. 113 Laura Franzin, Daniela Cabodi and Nicoletta Bonfrate P34 Use of automated online tools in External Quality Assurance programs for 114 sequence-based typing uncovers both strengths and weaknesses in technique and proficiency Norman Fry, Massimo Mentasti, Anthony Underwood, Garan Jones and Timo- thy Harrison P35 Implication of sequence - based typing method for epidemiological investiga- 115 tion of Legionella pneumophila infections Darja Kese, Rok Kogoj and Boris Kopilovic P36 Community acquired pneumonia due to Legionella wadsworthii . 116 Marylin LecsO-Bornet¨ , Florence Brossier, Isabelle Lerat, Jean Gabarre, Ca- pucine Morelot-Panzini and Vincent Jarlier P37 DNA microarray-based genotyping of Legionella pneumophila serogroup 1 117 Christian Luck¨ , Stefan Monecke, Klaus Heuner, Peter Slickers and Ralf Ehricht P38 Utility of a Legionella pneumophila real-time PCR assessed using respiratory 118 and serum samples from proven cases of Legionnaires’ disease. Massimo Mentasti, Norman Fry, Draga Tchipeva, Baharak Afshar, Chan- tal Palepou-Foxley, Patrick Kimmitt and Timothy Harrison P39 Detection of Legionella pneumophila using real-time PCR in the presence of 119 other Legionella species Anne Nor, Bente Wallervand Ofte and Ole-Jan Iversen P40 Evaluation of a new immunochromatographic rapid test for the identification 120 of legionella from cultures. Teresa Pardo, Dori Fraile, Almudena Rojas, Luis Gonzalez,´ Jose´ M. Delgado, Antonio Fuertes and Jose´ Rojas P41 Novel automated MLVA typing assay for a rapid and high-throughput 121 Legionella pneumophila genotyping tool Daniel Sobral, Christine Pourcel, Fabienne Loisy-Hamon, Gilles Vergnaud, Anne Gerard´ and Pierre Le Cann P42 Sensitivity and specificity of the detection of Legionella pneumophila using 122 PCR assays targeting three different genes Nadezhda Temezhnikova and Alexandr Taran P43 Genomic markers of Legionella pneumophila Paris strains 123 Mike Vergnes, Christophe Ginevra, Max Maurin, Philippe Normand, Jean Thioulouse, Jer´ omeˆ Etienne, Sophie Jarraud, Elisabeth Kay and Dominique Schneider P44 Characterization of Legionella pneumophila isolates in Norway by sequence 124 typing and monoclonal sero- and subgroups. Elisabeth Wedege, Karin Bolstad, Elisabeth Fritzsønn and Dominique A. Caugant - 323 - P45 Three sporadic cases of L. longbeachae Legionnaires Disease in Austria, 2008 125 and 2009 Daniela Schmid, Alexander Indra, Marion Blaschitz, Barbara Robl, Petra Hasen- berger, Ingrid Gruner,¨ Wolfgang Prammer, Gerhard Leitner and Gunther¨ Wewalka Session 3: Host - Microbe interactions: a) secretion systems and their substrates P46 The mechanism of action of VipA, a Legionella pneumophila actin-binding 127 effector Irina Franco, Nadim Shohdy and Howard Shuman P47 Molecular characterization of the Dot/Icm-translocated AnkH and AnkJ 128 eukaryotic-like effectors of Legionella pneumophila Fabien Habyarimana, Souhaila Al Khodor, Chris Price, Marina Santic and Yousef Abu Kwaik P48 Influence of regulatory elements of Legionella pneumophila on expression of 129 lipopolysaccharides in liquid cultures Johanna Kuhn,¨ Cacilia¨ Muller,¨ Esteban Fernandez Morera and Jurgen H. Helbig P49 Comparative analysis of the role of protein-kinases in Legionella pneu- 130 mophilavirulence Eva Hervet, Xavier Charpentier, Melanie´ Levet, Anne Vianney and Patricia Doublet P50 Host Prenylation of Legionella CaaX Box Proteins 131 Stanimir Ivanov and Craig Roy P51 The role of the secreted effector protein SidJ in virulence of Legionella 132 pneumophila Kwang Cheol Jeong and Joseph Vogel P52 The Trb/Tra conjugation/type IVA secretion system of Legionella pneumophila 133 Corby Monika Lautner, Christine Ott and Klaus Heuner P53 The Role of Type II Secretion during Lung Infection by Legionella 134 pneumophila Kessler McCoy-Simandle and Nicholas Cianciotto P54 Legionella pneumophila PmiA contributes to infectivity to macrophages and 135 protozoa by stabilizing DotA, a component of the Icm/Dot typeIV secretion apparatus. Masaki Miyake, Yasunori Iwasaki, Tsuyoshi Hayashi, Hitomi Koike and Yasuyuki Imai P55 Characterization of the type IVB secretion system coupling protein IcmO/DotL 136 of Legionella pneumophila Catherine A. Mueller and Howard Shuman P56 The Legionella pneumophila type II secretion system secretes an endoglu- 137 canase, a eukaryotic-like protease, and a novel effector that is required for virulence. Meghan Pearce and Nicholas Cianciotto P57 Sliding motility by Legionella pneumophila 138 Catherine Stewart and Nicholas Cianciotto P58 Characterization of the IcmS/IcmW-DotL interaction in type IVB substrate 139 secretion Molly Sutherland, Carr Vincent, Victor Tseng, T. Linh Nguyen and Joseph Vogel P59 Cytochromes c3 and c4 and TolC promote the production and secretion of the 140 Legionella pneumophila siderophore Emily Yip and Nicholas Cianciotto - 324 - Session 4: Host - Microbe interactions: b) metabolism and virulence P60 Legionella pneumophila requires polyamines for intracellular growth - 142 Potential role of the L. pneumophila chaperonin in modulating polyamine levels in host cells. Gheyath Nasrallah, Angela Riveroll, Audrey Chong, Lois E. Murray, Rafael A. Garduno and David S. Allan P61 Characterization of L. pneumophila PlaB, a member of a novel phospholipase 143 A family and its potential role in virulence Jennifer Bender, Kerstin Rydzewski, Markus Broich, Eva Schunder, Klaus Heuner and Antje Flieger P62 Legionella pneumophila outer membrane protein LbtU promotes both 144 siderophore utilization and pathogenesis Christa Chatfield and Nicholas Cianciotto P63 LHBP1, a novel immunogenic gylcosaminoglycan binding protein of Legionella 145 pneumophila, is produced during legionellosis and contributes to alveolar epithelial cells adhesion. Carla Duncan, Akriti Prashar, Patrick Tang, Mauricio Terebiznik, Donald E Low and Cyril Guyard P64 The bdhA-patD Operon as a virulence determinant, revealed by a novel large- 146 scale aproach for virulence-attenuated Legionella pneumophila mutants Philipp Aurass, Birgit Pless, Kerstin Rydzewski, Gudrun Holland, Norbert Bannert and Antje Flieger P65 Investigations into the metabolism of Legionella pneumophila 147 Vroni Herrmann, Eva Eylert, Wolfgang Eisenreich, Monika Lautner and Klaus He- uner P66 Genetic analysis and function of redundant ompS genes in Legionella 148 pneumophila Paul S. Hoffman, Serhiy Vitko, Poornima Gourabathini and Fanny Ewann P67 Legionella pneumophila flagellum in the fateful journey from amoeba to 149 macrophages. Juliana I. Hori, Giuliano F. Morgantetti, Catarina V.Horta, Marcelo De S. F. Pereira, Liliana M. Massis and Dario S. Zamboni P68 The essential, membrane-bound oxidoreductase Com1 is required for growth 150 and infectivity of Legionella pneumophila. Max Jameson-Lee and Paul S. Hoffman P69 Metabolic reconstruction of the human pathogen Legionella pneumophila 151 Matthieu Jules, Christophe Rusniok, Sandra Duperrier, Holger Bruggemann¨ and Carmen Buchrieser P70 Virulence properties of Legionella pneumophila GDSL lipolytic enzymes 152 Christina Lang, Elena Rastew, Sangeeta Banerji, Sina Bartfeld, Thomas F. Meyer and Antje Flieger P71 Expression of Legionella pneumophila efflux pump genes during different 153 physiological states Mourad Ferhat, Daniele` Atlan, Jean-Claude Lazzaroni and Christophe Gilbert P72 Interaction between COPI and the Legionella pneumophila protein LpnE. 154 Hayley Newton, Ralf Schuelein, Margareta Aili, Joan Sloan and Elizabeth Hartland P73 Structural investigation into the Legionella CD39 NTPDase Lpg1905 155 Patrice Riedmaier, Fiona M Sansom, Travis Beddoe and Elizabeth Hartland - 325 - P74 Characterization of the major phospholipase A/ lysophospholipase A activity 156 of Legionella pneumophila Eva Schunder, Patrick Adam, Futoshi Higa, Katharina Remer, Udo Lorenz, Michael Steinert, Jennifer Bender, Antje Flieger and Klaus Heuner P75 Increased expression of the plasminogen activator Lpa of Legionella 157 pneumophila in low-magnesium medium Leen Vranckx, Veerle Saels and Jozef Anne´ Session 5: Immunology and host susceptibility P76 How can an excutioner caspase control Legionella pneumophila infection 159 Anwari Akhter, Thirumala Kanneganti, Mark Wewers and Amal Amer P77 The early immune response in Legionella longbeachae serogroup 1 mice 160 infection Ivana Gobin, Zlatko Trobonjaca, Darinka Vuckovic, Gabrijela Begic, Tiana Grubesic, Miljenko Doric and Milorad Susa P78 Polyreactivity of a mAb against human erytrocyte glycophorin A with the 161 major outer membrane protein from Legionella pneumophilia Jan Kolberg, Ø Ihle, Arne Høiby and Audun Aase P79 Phase variable changes and intracellular delivery of LPS components of 162 Legionella pneumophila serogroup 1 strain Corby and its MAb 3/1-negative mutant Katja Reichardt and Jurgen H. Helbig P80 Comparative immunological studies of sera from patients hospitalized during 163 a large outbreak of Legionnaires’ Disease. Elisabeth Wedege, Karin Bolstad, Anita Kanestrøm, Eva H. Lindberg and Øystein Simonsen P81 Cytomegalovirus latency does not influence clinical presentation and outcome 164 of infection with Legionella pneumophila serogroup 1 Rob Rentenaar, Peter De Jager, Peter Schneeberger, Bart De Wit and Peter Wever Session 6: Microbe - Environment interactions: a) cell biology, protozoa, biofilms P82 A missing link in Legionella host-parasite interactions 166 Ann Karen Brassinga, Jason Kinchen, Meghan Cupp, Paul S. Hoffman and Costi Sifri P83 In vitro analysis of interaction of Acanthamoeba castellanii with Legionella 167 pneumophila of the filamentous morphological form Sock Hoai Chan and Chun Chau Sze P84 Modulation of host cell phosphoinositide metabolism by Legionella pneu- 168 mophila Sabrina Engelhardt, Roger Meier, Curdin Ragaz and Hubert Hilbi P85 Proteome analysis of Legionella vacuoles isolated by immuno-magnetic 169 separation Ivo Finsel, Simon Urwyler, Curdin Ragaz, Alexander Schmidt, Ruedi Aebersold and Hubert Hilbi P86 Copper indirectly controls the proliferation of Legionella pneumophila in 170 drinking water heterotrophic biofilms Maria Salome´ Giao˜ and C. William Keevil - 326 - P87 In vitro antagonistic activity among clinical and environmental strains of 171 Legionella spp, strains showing inter and intraspecific differences Rafael Gomez-Lus´ , Carmen Pelaz, Carmen Rubio, Raquel Becerril, Francisco Javier Castillo, Estrella Duran,´ Concepcion´ Garc´ıa, Silvia Lobez,´ Mercedes Oca and Soledad Salvo P88 Investigating Legionella pneumophila physicochemical surface properties to 172 explore the intimate interactions with host cells. Florence Gosselin, Fabien Gaboriaud, Julien Simonet, Edith Angel, Gregory´ Fran- cius, Christophe Ginevra, Sophie Jarraud and Laurence Mathieu P89 Large particles of Legionella pneumophila formed inside Acanthamoeba 173 castellanii have a complex spherical LPS-architecture that does not depend on special serotypes Clemens-Johannes Schumacher and Jurgen H. Helbig P90 Influence of temperature and flow velocity on the proportion of Legionella 174 pneumophila in natural biofilms Cecile´ Maurice-Blanc, Sylvie Viboud and Dominique Fontvieille P91 Interrelationship between Legionella pneumophila and free-living amoeba at 175 low temperature Akira Ohno, Naoyuki Kato, Chikako Hakii and Keizo Yamaguchi P92 Spatial Arrangement of Legionella Colonies in Lab-Scale Model Cooling 176 Tower Systems Michael Taylor, Kirstin Ross and Richard Bentham P93 Destructive enzyme activities in the secretome of Legionella pneumophila 177 Jana Tiefenau, Frederike Frese, Frank Galka, Sun Nyunt Wai, Susanne Engelmann and Michael Steinert P94 Diversity and identity of free-living protozoa in unchlorinated drinking water 178 Rinske Valster, Bart Wullings, Geo Bakker and Dick Van Der Kooij Session 7: Microbe - Environment interactions: b) detection in natural and artificial reservoirs - survival, persistence P95 Bacterial Colonization and Incidence of Legionella in Cooling Towers of 180 Southern General Company for Fertilizer Production in Basrah City Hadeel Al-Hadithi and Jundi Mehdi P96 Immunomagnetic capture enhance Legionella pneumophila serogroup 1 181 detection and recovery from environmental samples. Severine´ Allegra, Franc¸oise Berger, Philippe Berthelot, Florence Grattard, Bruno Pozzetto and Serge Riffard P97 Use of DGGE to study and follow Legionella population in environmental 182 waters Sandrine Bayle, Benoit Roig, Sophie Jarraud, Jer´ omeˆ Etienne and Laurent Garrelly P98 A comparison of culture methods and fluorescent in situ hybridization (FISH) 183 for detecting Legionella in commercial potting mixes Harriet Whiley, Michael Taylor and Richard Bentham P99 Occurrence of Legionella-like amoebal pathogens and other amoeba- 184 associated microorganisms in cooling towers and a municipal water system Mary Farone, Sharon G. Berk, Tu Vu, Jeremy Friesen, Christopher Garner, Tim Jacobs, Brian Choate, Celia Gendron, John Gunderson and Anthony Farone P100 Detection of Legionella spp. in potting soil from different sources 185 Jacob Bruin, Simona Casati, Jeroen Den Boer, Valeria Gaia and Ed Yzerman P101 Compost facilities as alternative reservoirs of Legionella spp. 186 Simona Casati, Jacob Bruin, Lisa Conza and Valeria Gaia - 327 - P102 Chlorine regulation of virulence genes expression in enviromental Legionella 187 pneumophila isolates Beatrice Casini, Paola Valentini, Francesca Torracca, Angelo Baggiani and Gaetano Privitera P103 Detection and quantification of viable Legionella cells from environmental 188 water samples by combined use of ethidium monoazide and real-time PCR Bin Chang, Toshitsugu Taguri, Kanji Sugiyama, Junko Amemura-Maekawa, Fumiaki Kura and Haruo Watanabe P104 Free-living amoebae and Legionella in bioaerosols from composting facilities 189 of southern Switzerland Lisa Conza, Simona Casati and Valeria Gaia P105 Legionella Colonization of the Czech Environment and the Associated Health 190 Risk : A Ten Year Study. Vladimir Drasar, Radomir Polcar and Norman Fry P106 Rapid Identification of Legionella Species using Whole Cell MALDI-TOF Mass 191 Spectrometry Michal Drevinek and Vladimir Drasar P107 Usefulness of Real-Time PCR for Detection of Difficult Growing Legionella 192 spp. from a Water Supply. Laura Franzin, Daniela Cabodi, Nicoletta Bonfrate and Valerio Demarie P108 Legionella and Amoeba from Water Samples 193 Laura Franzin, Daniela Cabodi, Nicoletta Bonfrate and Paola Fresi P109 Prevalence and degree of Legionella colonization in cooling towers and 194 associated variables. Marian Garc´ıa-Nunez˜ , Sonia Ragull, Maria Luisa Pedro-Botet, Lourdes Mateu and Miquel Sabria` P110 Evaluation of the usefulness of a new direct immunofluorescence assay 195 (ScanVIT-Legionella TM) for monitoring hospital water systems contaminated with Legionella spp. Monica Giacomuzzi, Savina Ditommaso, Marino Gentile and Carla Zotti P111 Research of Legionella contamination in dental units’ waters and aerosols 196 samples Ays¸ın C¸otuk, Duygu Goksay¨ , Nihal Dogru˘ oz¨ and Esra Ilhan Sungur P112 A Legionella typing method for ecological studies: Infrequent Restriction Site 197 Polymerase Chain Reaction (IRS PCR) Delphine Jakubek, Matthieu Le-Brun, Gerard Leblon, Michael Dubow and Marie Binet P113 Occurrence of legionella and other pathogenic microbes in communal waste 198 water treatment plants with low water temperature Jaana Kusnetsov, Piia Airaksinen, Tarja Pitkanen,¨ Ari Kauppinen, Asko Vepsal¨ ainen,¨ Ilkka Miettinen and Eila Torvinen P114 Prevalence of Legionella strains from cooling towers and laboratory diagnosed 199 legionellosis cases in New Zealand Robert Lau, Brian Caughley, David Harte and Rob Deacon P115 Impact of water quality on the transfer and the survival of aerosolised 200 Legionella . Matthieu Le-Brun, Thi-Lan Ha, Sylvie Soreau and Marie Binet P116 Metrological evaluation of a system of collection of biological aerosols in a 201 cooling tower of a french nuclear power plant. Matthieu Le-Brun, Marie Binet and Celine´ Bouteleux - 328 - P117 Distribution of Legionella species from the environmental source of public 202 facilities in Korea Hae Kyung Lee, Jeong Im Shim, Woo-Sik Kim, Jae-Yon Yu and Yeon Ho Kang P118 Improvement of airborne Legionella survival in presence of organics and 203 minerals in waters Julien Simonet, Thi-Lan Ha, Enric Robine and Laurence Mathieu P119 Opportunity catch-study on Legionellosis surveillance. Support legionellosis 204 surveillance for Microbiology. A collaborative effort Ma Antonia Carratala,´ Juan Bayo, Antonio Salazar, M◦ Dolores Ocete, Concepcion´ Gimeno, Remedios Guna, Lorena Jordan,´ Ana Rallo and Enrique Mira P120 ”Legionella spp on sea”: preliminary results of a surveillance program on the 205 Italian Military Boats. Maria Teresa Montagna, Christian Napoli, Roberta Iatta, Teresa Marsico and Salvatore Barbuti P121 Dynamics of Legionellae in a Mediterranean coastal river subjected to different 206 pollutions Nathalie Parthuisot, Philippe Lebaron, Aurelie´ Touron-Bodilis and Julia Baudart- Lenfant P122 Recovery of Legionella from aerosol sampling 207 Anthony Pinon, Anne-Marie Jandos, Nelsie Berthelot, Alain Vidal, Franck Laurent, Sandrine Oberti and Michele` Vialette P123 PCR-quantification of viable Legionella in water samples: development of a 208 propidium monoazide pretreatment of bacteria directly applied on membrane filters Sami Slimani, Christophe Ginevra, Maelle Molmeret, Eric Dusserre, Celine´ Mazure, Monique Reyrolle, Jer´ omeˆ Etienne and Sophie Jarraud P124 Correlating numerical tools and experimental measurements for airborne 209 microorganisms concentrations : application to the spreading of Legionella pneumophila from cooling towers. Eric Tarnaud, Thi-Lan Ha, Enric Robine, Frederic Tognet, Cyrille Turmeau, Yannick Morel and Laurence Rouil P125 Susceptibility of extra- and intracellular Legionellae to silver and a method to 210 detect the concentration of microbicidal active silver in water Catharina Unger and Paul Ch. Luck Session 8: Legionella prevention and control P126 Quality and quantity detection of Legionella in environmental water samples 212 Yulia S. Alyapkina, Igor S. Tartakovsky, Tatiana I. Karpova, Oksana V. Sadretdi- nova, Jakov Alexeev and Irina V. Novokshonova P127 Rapid and accurate enumeration of active Legionella pneumophila in aquatic 213 environments by microcolony method Takashi Baba, Naoko Inoue, Nobuyasu Yamaguchi and Masao Nasu P128 Prevalence of Legionella in different aquatic environments in Hungary 214 Zsofia´ Barna, Csaba Bognar,´ Judit Krisztina Horvath,´ Mihaly´ Kad´ ar,´ Anita Szax and Marta´ Vargha P129 Development and certification of a standard reference material (SRM) for 215 Legionella detection and quantification by quantitative PCR in environmental samples. Maud Baume, Laurent Garrelly, Philippe Guarini, Hel´ ene` Host, Hel´ ene` Frenkiel, Sebastien´ Bouton, Pierre-Olivier Fraisse, Monique Reyrolle and Sophie Jarraud - 329 - P130 Impact of oxidizing disinfectants against co-culture of Legionella pneumophila 216 and Acanthamoebae Mathieu Dupuy, Stephane´ Mazoua, Florence Berne, Nathalie Garrec, Pasca- line Herbelin, Sandrine Oberti, Marie-Hel´ ene` Rodier, Sylvie Soreau, France Wallet, Yann Hechard and Nelsie Berthelot P131 Legionella growth and thermal treatment in copper domestic hot water 217 network Fabienne David, Alain Vidal, Nelsie Berthelot, Armelle Freval and Sandrine Oberti P132 Correlation between early diagnosis, clinical expiry and respiratory insuffi- 218 ciency in patients with pneumonia caused by Legionella Roberto Bettelli, Raffaele Russo, Giuseppe Russo, Vincenzo Castaldo, Mario Magliocca, Marilena Stanco, Giampaolo Buonopane and Anna Maria Rossi P133 Source or coincidence? Factors contributing to source identification of 219 Legionella infections Sacha Bleeker, Ed Yzerman, Sjoerd Euser, Jacob Bruin, Nico Nagelkerke and Jeroen Den Boer P134 An Environmental Investigations in a Response to the Notification of 220 Travel-associated Legionnaires’ Disease from European Working Group for Legionella Infection Network and World Health Organization, Thailand, 2006- 2007 Rome Buathong, Wanna Hanshoawarakul, Sopon Iamsirithaworn and Kum- nuan Ungchusak P135 Progress in the surveillance and control of Legionella infection in France, 1998- 221 2008 Christine Campese, Catherine Maine, Sophie Jarraud and Didier Che P136 Legionella Concentrations In Hotel Hot Water Systems Not Associated With 222 Cases Of Legionnaire’s Disease: a Power Law Distribution? Sebastian Crespi and Juan Ferrer P137 The use of power ultrasound as an effective pretreatment step during 223 hypochlorite disinfection of Legionella pneumophila-contaminated drinking water systems Priscilla Declerck, Louise Vanysacker and Frans Ollevier P138 v-PCR : an innovative method for quantification of viable Legionella in water 224 samples Pilar Delgado-Viscogliosi, Matthieu Bernier, Lydie Solignac and Jean-Marie Delat- tre P139 The Water Safety Plan Applied for the Control of Legionella Infections in an 225 Italian Hospital Oscar Di Marino, Aldo Maria Capra, Lorenza Camponovo, Stefano Melada, Francesco Garusi and Gianfranco Garusi P140 Antimicrobial Activities of Some Bacillus Strains Isolated from Tap Water 226 against Legionella pneumophila Strains Nihal Dogru˘ oz¨ , Zuhal Zeybek and Ays¸ın C¸otuk P141 Effective environmental sampling strategies for monitoring Legionella spp. 227 contamination in hot water systems. Monica Giacomuzzi, Savina Ditommaso, Marino Gentile, Angela Ruggenini Moiraghi and Carla Zotti P142 Expression of recombinant warnericin RK in Escherichia coli 228 Nicolas Girardin, Jean-Marc Berjeaud and Yann Hechard P143 What strategy for detecting airborne Legionella pneumophila? 229 Thi-Lan Ha, Laurence Mathieu and Enric Robine - 330 - P144 Detergent-like activity of warnericin RK, a specific anti-Legionella peptide 230 Julien Verdon, Mirjam Falge, Helke Maier, Heike Bruhn, Christian Lacombe, Jean- Marc Berjeaud, Michael Steinert, Cornelius Faber, Roland Benz and Yann Hechard P145 A novel system for the control of legionella in the water system of the bone 231 marrow transplant unit at Glasgow Royal Infirmary - nine years experience. John Hood, Gordon Cheape, David Gormlie, Alex Mead, Ian Powrie, Gordon Wills and Evonne Curran P146 16 months experience of legionella control/water quality in a new cancer centre 232 (including a bone marrow transplant unit): a comparison of instantaneous water heaters versus a conventional domestic hot/cold water system. John Hood, Gordon Cheape, Mel Aitken, Alex Mead and Gordon Wills P147 Monochloramine Efficacy on Biofilms in a Model Water Distribution Pipe Rig 233 Nazmiye Ozlem Sanli Yurudu, Esra Ilhan Sungur, Irfan Turetgen,¨ Nihal Dogru˘ oz,¨ Ays¸ın C¸otuk and Duygu Goksay¨ P148 Antibacterial and Hemolytic Activities of Amsonia orientalis (Syn. Rhazya 234 orientalis) Decne. against Legionella pneumophila strains Ayten Kimiran-Erdem,Gul¨ Cevahir-Oz,¨ Elif Ozlem¨ Arslan-Aydogdu˘ and Elif Yuzbas¸ıo¨ glu˘ P149 The effect of various biocides on survival of different Legionella pneumophila 235 strains Ayten Kimiran-Erdem, Nazmiye Ozlem Sanli Yurudu, Elif Ozlem¨ Arslan-Aydogdu˘ and Zuhal Zeybek P150 Legionnaires’ disease associated with a new residential area - risk factors and 236 remedial actions Louise Hjelmar Krøjgaard, Karen Angelika Krogfelt, Hans-Jørgen Albrechtsen and Søren A. Uldum P151 Occurrence and prevention of legionella in water systems of a ferry ship after 237 two Legionnaires’ disease cases Jaana Kusnetsov, Lotta Simola, Silja Mentula, Petri Ruutu, Piia Airaksinen and Nhu Nguyen Tran Minh P152 Design, Operation and Maintenance of Building Services to Minimize the Risks 238 of Legionellosis: a new European Guidebook for Practitioners Sergio La Mura, Cesare Maria Joppolo and Luca Alberto Pitera` P153 Lessons from an outbreak of Legionnaires’ disease on a cruise ship 239 Sandra Lai, John V. Lee, Lorraine Sadler-Reeves, Timothy Harrison, Carol Joseph, Allan Johnson, Gorel¨ Allestam, Birgitta De Jong, Gillian Ashford, James Sedgwick and Mathibalasingham Chandrakumar P154 Preventing Legionella growth and controlling water quality in public spa pools 240 Robert Lau, Brian Caughley and Rob Deacon P155 Legionella risk assessment and control on board train: effectiveness of 241 environmental surveillance and bacteria genotyping Patrizia Laurenti, Romina Sezzatini, Gianluigi Quaranta, Stefania Boccia, Rosarita Amore, Cinzia Turnaturi, Francesco Dalla Torre, Federica Boninti, Marta Marino, Elio Munafo,` Pasquale Del Nord, Mario Russo, Massimo Accorsi and Gualtiero Ricciardi P156 A controlled evaluation of deadlegs in Legionella colonization in a model 242 plumbing system Yusen Lin and Hsiu-Yun Shih P157 Accuracy of Legionella isolation by US Labs in the ELITE Program pilot study 243 Claressa Lucas, Thomas Taylor, Nicole Alexander and Barry Fields - 331 - P158 Identification and purification of anti-Legionella peptides produced by 244 staphylococcal strains Adrienne Marchand, Julien Verdon, Christian Lacombe and Jean-Marc Berjeaud P159 Control of Legionella Colonization and Effects on Biofilm in a Hospital Water 245 System Treated With Monochloramine Isabella Marchesi, Patrizia Messi, Immacolata Anacarso, Stefano Cencetti, Patrizia Marchegiano, Giuseppina Frezza and Paola Borella P160 Gold nanoparticle interactions and impact upon a common biofilm source: 246 Legionella pneumophila Amber Stojak, Stephen Klaine and Tamara McNealy P161 Relevance of indicators for legionella monitoring in cooling systems 247 Michele Merchat, Corinne Peyron and Gilles Chaperon P162 Optimal maintenance for outdoor ornamental fountains to avoid potential 248 dissemination of Legionella aerosols Rosa Minarro˜ Del Moral, Ana Rubio Garcia and Jose Luis Guijarro Rodriguez P163 The hospital tap waters: a risk of Legionella contaminated aerosol inhalation 249 Monique Reyrolle, Caroline Landelle, Celine´ Mazure, Marie-Christine Nicolle, Sophie Jarraud, Jer´ omeˆ Etienne and Philippe Vanhems P164 Prevention of Legionella and Pseudomonas colonisation for hot and cold 250 hospital water systems: a one year experience of Waterclean°R system Monique Reyrolle, Rodolphe Beron, Jerome Droguet, Cindy Schur, Georges Dubois, Sophie Jarraud and Jer´ omeˆ Etienne P165 Legionella in Campania: report of the year 2008 251 Anna Maria Rossi, Trofimena Lucibello, Mariangela Pagano, Francesca Di Leo, Antonio Coppola, Antonio Petrosino, Giacomo Dente and Giuseppe D’Antonio P166 Evaluation of Legionella Prevention and Control Measures at high risk 252 facilities in the Municipality of Cordoba´ - Spain, 2005-2008 Ana Rubio Garcia, Rosa Minarro˜ Del Moral and Jose Luis Guijarro Rodriguez P167 The Legionella chip: practical experience with this novel analytical tool to 253 analyze Legionella positive water samples Michel Ossendrijver, Adrie Atsma and Frank Schuren P168 Recognizing mixtures of Legionella strains 254 Adrie Atsma, Michel Ossendrijver, Martien Caspers and Frank Schuren P169 Application of ribotyping for the identification of the infection source for a 255 legionellosis case Alberta Stenico, Margit Seeber, Paul Huber, Armin Oberlechner, Gabriele Spitaler, Ludwig Moroder and Patrizia Rossi P170 A Risk Management Approach for Preventing or Controlling Health Risks 256 from Legionella bacteria William Thomas and Roger Shrapnel P171 The Role of Corporate Standards and Best Practice Guidance for Managing 257 Health Risks from Legionella bacteria in Multinational Organisations William Thomas, Roger Shrapnel and Kerry Horton P172 Legionella risk monitoring in a hospital water network through control of 258 bacterial proliferation using a new generation of Quantitative ATP-metry technology Marc Raymond and Veliana Todorova P173 Evaluation of automated nucleic acids extraction system for Legionella 259 pneumophila detection in water samples Michael Treilles, Delphine Esperet, Nadege Daguier and Laurent Thiery - 332 - P174 Development of a real-scale test rig for Legionella elimination in biofilms in hot 260 water distribution systems: thermal and chemical treatment evaluation Maha Farhat, Marie-Cecile´ Trouilhe´, Marina Moletta-Denat, Enric Robine, Christophe Foretˆ and Jacques Frere` P175 Comparison of the SANIPACKING°R Cooling Tower Fill Material against 261 Standard Polypropylene Fill Material in Recirculating Model Water System Irfan Turetgen¨ , Nazmiye Ozlem Sanli Yurudu and Imke Norden P176 Membrane composition is involved in high sensitivity to warnericin RK peptide 262 Julien Verdon,Jer´ omeˆ Labanowski, Tobias Sahr, Christian Lacombe, Thierry Fer- reira, Carmen Buchrieser, Jean-Marc Berjeaud and Yann Hechard P177 Distribution of Legionella pneumophila serogroups and monoclonal antibody 263 subgroups in environmental isolates from Russia (2005-2009) Olga L. Voronina, Marina S. Kunda, Vera V. Bitkina, Vladimir G. Lunin, Oksana V. Sadretdinova, Tatiana I. Karpova, Paul Ch. Luck and Igor S. Tartakovskiy P178 The use of windscreen wiper fluid without added chemicals in cars and 264 commercial vehicles: A newly identified risk factor for Legionnaires’ disease Anders Wallensten, Isabel Oliver, Kate Ricketts, George Kafatos, Nick Phin, James Stuart and Carol Joseph P179 Rapid detection of Legionella pneumophila in aquatic environment by using a 265 microfluidic device Nobuyasu Yamagauchi, Yuko Uebayashi, Masashi Torii and Masao Nasu P180 Quantification of Viable Legionella pneumophila Cells using Propidium 266 Monoazide combined with real-time PCR M. Adela Yanez˜ , Lorena Mart´ınez, Elena Soria, Raquel Murtula´ and Vi- cente Catalan´ Session 9: Epidemiology: a) nosocomial infections and Legionella control in hospitals P181 Usefulness of flow cytometry to assess the effect of heat treatment on Legionella 268 in water systems. Severine´ Allegra, Franc¸oise Berger, Philippe Berthelot, Florence Grattard, Bruno Pozzetto and Serge Riffard P182 Resistance of Acanthamoeba spp. cysts to disinfection treatments 269 Celine Coulon, Anne Collignon, Gerald McDonnell and Vincent Thomas P183 Resistance of Chlamydia-like to disinfection treatments and survival on 270 surfaces Celine Coulon, Anne Collignon, Gerald McDonnell and Vincent Thomas P184 Colonization of Legionella species in hospital water system in Turkey 271 Haluk Erdogan, Hale Turan, Riza Hasimoglu, Ozlem Kurt Azap and Hande Arslan P185 Preliminary Data on Legionella detection in Water Distribution Systems in 272 Cameroon Marguerite Ndayo Wouafo, Ariane Nzouankeu and Guy Ejenguele P186 Sequence-based typing of Legionella pneumophila sg 6 strains isolated from 273 hospital hot water systems in Poland. Katarzyna Pancer + P187 In-vitro study of H2O2 - Ag activity against Legionella and evaluation of its 274 efficacy in hospital waterborne infection control Maria Luisa Ricci, Stefano Fontana, Laura Achene, Maria Scaturro, Federica Pinci, Desiree´ Mineo, Massimo Ottaviani and Enrico Veschetti - 333 - P188 Contamination of legionella spp. and amoeba of water distribution systems in 275 3 hospitals of Shanghai, China Lili Tao, Bijie Hu and Zhaoyan Zhao P189 Prospective Study of Legionnaires’ Disease Mortality Factors in France: 276 Description of 21 Nososomial Cases, (April 2006 - June 2007) Silene Pires-Cronenberger, Christian Chidiac, Christine Campese, Didier Che, Sophie Jarraud, Jer´ omeˆ Etienne, Roland Poirier and Philippe Vanhems Session 10: Epidemiology: b) outbreaks and population genetics and taxonomy P190 Distribution of serogroups, sequence types, and monoclonal antibody 278 subgroups among Legionella pneumophila isolates from patients in Japan Junko Amemura-Maekawa, Fumiaki Kura, Jurgen H. Helbig, Bin Chang, Noriko Kaneko, Yuko Watanabe, Junko Isobe, Masafumi Nukina, Hiroshi Naka- jima, Kimiko Kawano, Yuki Tada and Haruo Watanabe P191 Sequence types of Portuguese clinical isolates of Legionella pneumophila and 279 their correlation with geographic distribution and infection origin Maria-Jesus Chasqueira,Lucia´ Rodrigues, Marta Nascimento and Teresa Marques P192 A new hotel opened and wellcome Legionella: what’s wrong? 280 Haluk Erdogan and Hande Arslan P193 First population-based molecular epidemiology analysis of non-serogroup 1 281 Legionella pneumophila clinical isolates. Naglaa Elgngihy, Christine Pourcel, Patrick Tang, Nathalie Tijet, Carla Duncan, Jonathan Gubbay, Donald E Low and Cyril Guyard P194 An Outbreak of Pontiac Fever Due to Legionella longbeachae Serogroup 2 282 Found in Potting Mix in a Horticultural Nursery in New Zealand Geoffrey Cramp, David Harte, Nicholas Douglas, Frances Graham, Mona Schous- boe and Kate Sykes P195 Two groups of Legionella anisa isolates of environmental origin in Japan 283 Fumiaki Kura, Junko Amemura-Maekawa, Bin Chang, Atsuko Suzuki-Hashimoto, Masayuki Ichinose, Takuro Endo and Haruo Watanabe P196 Development of a UK Legionnaires disease outbreak toolkit 284 David Lemon, Ian Hall and Steve Leach P197 Description of clusters of Legionnaires’ disease associated with Spanish re- 285 offender accommodation sites in the period 1987 to 2009 Rosa Cano, Francisco Nogareda, Beatriz Baladron, Carmen Martin and Carmen Pelaz P198 Characterization of a potentially novel Legionella species isolated from human 286 clinical material Aleisha Reimer and Kathryn Bernard P199 The state of cooling towers Legionellae population in Verkhnyaya Pyshma after 287 outbreak 2007. Olga L. Voronina, Marina S. Kunda, Vera V. Bitkina, Tatiana I. Karpova, Igor S. Tartakovskiy and Alexandr L. Gintsburg P200 Genetic relationships analyzed by 2 genotyping methods among the Legionella 288 strains isolated from public bath water in Toyama Prefecture, Japan Masanori Watahiki, Junko Isobe, Junichi Kanatani, Tomoko Shima, Keiko Kimata and Takeshi Kurata - 334 - Session 11: Genomics and Comparative genomics P201 Virulence differences in a large systematic Legionella collection from the 290 Netherlands Jeroen Den Boer, Jacob Bruin and Jurgen H. Helbig P202 Genome sequence of L. pneumophila strain Lorraine and an Environmental 291 isolate: comparative genomics of six L. pneumophila strains Laura Gomez-Valero, Christophe Rusniok, Sophie Jarraud, Benoit Vacherie, Zoe Rouy, Valerie Barbe, Claudine Medigue, Jer´ omeˆ Etienne and Car- men Buchrieser P203 The Genomic Sequence of the Soil Dwelling Opportunistic Pathogen 292 Legionella longbeachae Natalia Kozak, Meghan Buss, Michael Frace, Dhwani Govil, Claressa Lucas, Tatiana Travis, Melissa Olsen-Rasmussen, Robert Benson and Barry Fields P204 Comparative genomics of the lipopolysaccharide biosynthesis gene cluster of 293 Legionella pneumophila: towards rapid identification tests Nathalie Merault´ , Christophe Rusniok, Sophie Jarraud, Elsa Brachet, Mickael¨ Marin, Christel Cazalet, Laura Gomez Valero, Jean-Louis Gaillard, Jean-Louis Herrmann, Jer´ omeˆ Etienne, Christine Lawrence and Carmen Buchrieser P205 Molecular analysis of virulence genes in Legionella strains from environmental 294 samples Silvano Salaris, Massimiliano Orsini, Erica Leoni, Pier Paolo Legnani and Sandra Cristino P206 Molecular analysis of Legionella non-pneumophila species 295 Michel Ossendrijver, Martien Caspers and Frank Schuren P207 Case studies on Next Gen sequencing technologies - experiences and results 296 Dr. Kerstin A. Stangier and Dr. Yadhu Kumar Session 12: Functional genomics P208 Deciphering Legionella biofilms at the proteomic level 298 Sandra Ahmed-Lecheheb, Arbia Khemiri, Philippe Chan, David Vaudry, Jacques Frere,` Thierry Jouenne and Pascal Cosette Session 13: Gene regulation in the environment and the host P209 Flagella gene regulation in Legionella pneumophila 300 Christiane Albert-Weissenberger, Tobias Sahr, Jorg¨ Hacker, Klaus Heuner and Carmen Buchrieser P210 Legionella pneumophila diguanylate cyclases and phosphodiesterases: defining 301 the role of the cyclic-di-GMP signaling network in the life cycle of Legionella pneumophila Assaf Assaf Levi, Marc Folcher, Urs Jenal and Howard Shuman P211 A database dedicated to gene expression data of Legionella pneumophila 302 Sandrine Rousseau, Tobias Sahr, Pierre Ventadour, Carmen Buchrieser and Ivan Moszer P212 Further analysis of FleQ, RpoN and FliA - Regulators of the flagellar regulon 303 of Legionella pneumophila Tino Schulz, Christiane Albert-Weissenberger, Carmen Buchrieser and Klaus He- uner P213 The autoinducer synthase LqsA and cognate sensor kinase LqsS regulate 304 host cell interactions, sedimentation and a genomic island of Legionella pneumophila Andre´ Tiaden, Thomas Spirig, Tobias Sahr, Carmen Buchrieser and Hubert Hilbi - 335 - P214 An assessment of the prevalence of Legionella spp. and effectiveness of thermal 305 disinfection in eradicating the bacteria from water distribution system in a single hospital of Lublin. Małgorzata Wojtowicz´ , Agnieszka Sikora, Agnieszka Magrys´ and Maria Kozioł- Montewka - 336 - COMPANY PROFILES OF THE SPONSORS - 337 - DESINFECCIONES ALCORA s.a. Laboratory service, Control of Legionella, Manufacture of metal ionization Mission: Control of Legionella Activity: Study of installations in buildings. Water analysis and treatment. Manufacture of metal ionization. Team: Our team is made up of experienced Engineers, Microbiologists, Chemists, and technicians specialised in water treatment. Address: Desinfecciones Alcora s.a. Calle : F, Oeste nº 98 P. I. Malpica 50016 Zaragoza (Spain) Tel : 976291019 Fax : 976291332 Web : www.alcora.es E-Mail : [email protected] ______Contact: Luis Antonio Sánchez Guillén C.E.O. [email protected] Alcora was founded in 1978, and as of 1986 has been a pioneer in treatments for controlling Legionella in Spain. Between 1996 and 2005 the company was hired by the health authorities in several Autonomous Communities in Spain as consultants in the investigation of outbreaks of legionella in urban communities. From the time the company was first set up, our work has been developed in our modern laboratories in Madrid and Zaragoza, where thousands of samples are processed every year. Our manufacturing division also has a production line of ionizers for treating water infected by legionella bacteria. Références Alcora is a leader in technical service for Spas in Spain, including the creation of lines of special treatments for hot springs, where the organoleptic factors and specific conditions of this kind of waters are respected. APPLIED BIOSYSTEMS Instrumentation Activity: Applied Biosystems sert depuis 25 ans la science de la vie en industrie et recherche en développant et commercialisant des instruments, consommables, logiciels et des services. Nos clients utilisent les outils d'Applied Biosystems pour analyser l'ADN et l'ARN, les petites molécules et les protéines pour aider à la découverte scientifique de nouveaux médicaments. Les produits d'Applied Biosystems servent aussi en dehors de la science de la Address : vie à travers des marchés appliqués comme la police scientifique, la biosécurité et la qualité et la sécurité en alimentaire et pour Applied Biosystems l'environnement. Les instruments et consommables d'Applied 25, avenue de la baltique Biosystems sont des outils qui permettent de manipuler l'ADN en le multipliant ou en déterminant sa structure et son code. 91943 Courtaboeuf cedex Ces outils permettent de détecter ou quantifier la présence de Tel :+33 (0)1.69.59.85.85 micro-organismes dangereux. Ils sont aussi utilisés pour identifier des micro-organismes sur les surfaces, l'environnement autour Fax : +33 (0)1.69.59.85.00 d'un produit et pour le produit lui-même. Les secteurs d'activité Web : comme la production stérile pharmaceutique, le monde hospitalier www.appliedbiosystems.com ou l'agro-alimentaire sont des utilisateurs de ces nouvelles technologies pour le Contrôle Qualité de routine. ______Contact: Annika Thorenson Ingénieur Commercial Tel : 01.69.59.85.85 [email protected] AQUA-TOOLS Microbiological monitoring of water using an innovative quantitative tool of ATP- metry technology. Innovative filtration modules for shower and tape water to remove Legionella and other bacteria. Mission : Legionella Risk Monitoring in sanitary water networks through control of bacterial proliferation using a new generation of Quantitative ATP-metry technology, also used as indicator of surface cleanliness and cleaning efficiency for Hand hygiene, Environmental hygiene, Reusable medical instruments and devices Production & sales of Innovative Hollow fiber modules offering safety of shower & tape water. Address : Activity : An innovative company, serving european & MEA market, Route de Renault dedicated to microbiological monitoring of water and surface Hygiene, using a new generation of Quantitative ATP-metry 78410 Flins sur Seine -France technology. Our dedicated products allow a real-time monitoring Tel : +33 1 30 95 79 50 of water quality and immediate detection of microbial proliferation. Proactive approach for decreasing sanitary risks as Legionella Fax : +33 1 30 95 54 55 risks and improving water networks becomes possible to Web : www.aqua-tools.com implement whatever the installation. React and measure immediately the efficiency of actions – this is where Aqua-tools E-Mail : [email protected] products bring the major benefit. TM ______Aqua-tools is launching Aqua-ray Corp , with an innovative hollow fiber membrane module for shower and tape water, a Contact: filtration module to remove Legionella and other bacteria during Marc RAYMOND ,CEO months. Team: Microbiology specialists Water and Waste water sanitation engineers International Experts and Consultants in Management of Sanitary risks linked to microbial contamination Business Development Engineers Sales and Marketing Managers References : upon request Examples : France, Europe, Middle East Africa : Large Hospitals and Research Institutes, large buildings like European Parliament of Strasbourg, Public buildings and Spas North american market covered by LuminUltra Technologies. BIOMERIEUX SA BioMerieux, leader mondiale en microbiologie, développe, produit et commercialise des systèmes composés d’instruments, de réactifs, de logiciels et de service . Mission : Contribuer à l’amélioration de la santé publique dans le monde , grâce au diagnostic in vitro . Offrir des solutions de diagnostic qui déterminent l’origine d’une maladie ou d’une contamination pour améliorer la santé des patients et assurer la sécurité des consommateurs . Adresse Activité : Chemin de l’Orme Deux domaines d’intervention : le diagnostic clinique et la microbiologie industrielle . 69 280 Marcy l’Etoile Ses produits sont utilisés dans le diagnostic des maladies ______infectieuses et pour la détection de micro-organismes dans les Contact(s) : produits agroalimentaires, pharmaceutiques , cosmétiques et l’eau . Tout renseignement et demande d’information Equipes : www.biomerieux.com bioMérieux a été créé en 1963 . > Présence dans 150 pays au travers de 39 filiales et d’un large réseau de distributeurs ¾ Un effectif important de 6000 collaborateurs ¾ 17 sites de productions ¾ 13 sites de R&D ¾ 2 centres de formation ( France et US ) destinés aux formations clients et personnels bioMérieux - assurent environ 300 sessions /an de formation pour plus de 1500 stagiaires . BIO-RAD Activity : Bio-Rad Laboratories, Inc. (NYSE: BIO and BIOb) has remained at the center of scientific discovery for more than 50 years manufacturing and distributing a broad range of products for clinical diagnostics and life science research markets. The company is renowned worldwide among hospitals, universities, major research institutions, as well Address : as biotechnology and pharmaceutical companies for its commitment to quality and customer service. Founded in 2000 Alfred Nobel Drive, Hercules, 1952, Bio-Rad is headquartered in Hercules, California. California 94547, USA The company employs over 6,300 people and serves more Tel: 1 510 741 1000 than 85,000 research and industry customers worldwide Fax: 1 510 741 5800 through its global network of operations. Web : www.bio-rad.com/diagnostics ______Bio-Rad offers a wide range of tests for quality indicators Contact : or pathogens detection in food and water samples Angélique Duverger (drinking water, bathing water, waste water ...). These Environmental Microbiology diagnoses are performed with traditional and Product Manager, Europe standardized culture media or with alternative Bio-Rad Laboratories technologies more innovative and faster: chromogenic Tel: + 33 1 47 95 81 19 substrates (RAPID’ method), real-time gene amplification Fax: + 33 1 47 95 62 24 (iQ-Check™ Real-Time PCR) or impedance (XplOrer64™ E-mail: Angelique.duverger@bio- System, CheckN'Safe™ kits). rad.com Bio-Rad tests for water testing allow the detection and / or quantification of the Total Flora, E. coli, Coliform, Enteroccocci, Legionella, Pseudomonas aeruginosa, Staphyloccocci, Clostridia and Salmonella. Team: Food Science Division GATC Biotech SARL Service de séquençage d’ADN, séquençage de nouvelle génération, services de bioinformatique et logiciels d’analyse séquentielle – en matière de séquençage d’ADN et de bioinformatique, nous sommes leader incontesté en Europe GATC Biotech est un prestataire leader en services de séquençage et logiciels de Bio-informatique pour l´industrie et les laboratoires académiques. GATC offre des solutions complètes de séquençage, depuis la préparation de l´echantillon jusqu´au séquençage à haut débit et à la bioinformatique. Nous utilise trois technologies de séquençage de pointe : Adresse : ABI 3730XL, Roche GS FLX et Illumina Genome Analyzer GATC Biotech SARL Campus Scientifique de Luminy Bat. CCIMP – Case 922 Equipes 13288 Marseille Cedex 9 Le siège social de la société à Constance, sur le lac, est Tel : +33 (0) 4 91 82 84 88 également un facteur motivant aussi bien qu’une source Fax : +33 (0) 4 42 01 12 00 d’inspiration pour les presque 100 collaborateurs que compte Web : www.gatc-biotech.com entre temps l’équipe GATC. E-Mail : customerservice@gatc- biotech.com ______Contact : Zélie Dubreucq Scientifique Sales Represantative GIRPI - Aliaxis group Mission / Mission : Fabrication de canalisations en CPVC (System’O® : HTA® & HTA- F®) Manufacturer of CPVC pipework systems ( System’O®: HTA® & HTA-F) Activité / Activity : Réseaux d’eau chaude et froide sanitaire pour les Etablissements GIRPI - Rue Robert Ancel – BP36 de santé ou Recevant du Public (Hôpitaux, Hôtels, Spa…) 76700 HARFLEUR - FRANCE Domestic hot & cold water services for Buildings Receiving the Public (Hospitals, Hotels, Spas…) Tel : +33232796000 Fax : +33232796027 Equipes / Teams : Web : www.girpi.com Une équipe de prescription formée au risque Légionelle & E-Mail : [email protected] Pseudomonas sur les aspects réglementaires, conception et traitements. ______Notre bureau d’étude d’exécution réalise sur les plans des réseaux d’eau en intégrant les contraintes liées aux chocs thermiques. Notre service assistance technique peut suivre les démarrages de chantiers. Notre service formation intervient sur les conférences et assure les formations destinées aux services techniques, bureaux d’ingénierie, hygiénistes, plombiers… A specification team, trained to dealing with various aspects Alexandre Potier – Technical Specification Manager – 00 33 2 32 79 60 08 of the legionella risk: laws & guidelines, network design and [email protected] water treatment. A technical project study team, trained to dealing with thermal shock implications on pipework execution drawings. A technical assistance team, trained to dealing with induction training interventions on building sites. A training team, intervening on conferences and seminars meant for new build & maintenance technical specialists, consulting engineers, hygienists… Frédéric Omar – Export Area Manager Tel 00 33 2 32 79 60 06 [email protected] Plus de 500 références de réseaux d’eau sanitaire dans des Etablissements de santé en Europe et Afrique. Nombreuses références dans l’hôtellerie. More than 500 healthcare building references for domestic hot & cold water services throughout Europe and all over the world. Références References INVERNESS MEDICAL Activité : Inverness Medical assure et développe la commercialisation de produits dédiés au diagnostic médical autour de 5 divisions : la cardiologie (insuffisance cardiaque aiguë, syndromes coronariens aigus, …), les maladies infectieuses (legionelle, strepto. Pneumoniae, clostridium, HIV, RSV, Adresse : grippe, malaria, …), l’oncologie, la toxicologie et la santé de la femme. Tel : +33 (0)1 46 23 63 63 Fax : +33 (0)1 46 26 34 94 Web : www.invernessmedical.fr E-Mail : [email protected] ______Contact(s) : Mr Stéphane COISNE, Directeur des Ventes et Marketing Mr Ludovic BAL, Directeur Commercial Marques: Binax Now, Clearview, Determine, Clondiag, Orgenics Immunocomb, Osteomark, Surestep, TechLab, Testpack, Triage Biosite, Cholestech, Panbio, Matritech. Références LEGYON Mission: The new standard in Legionella identification. Legyon has been established by Vitens and TNO to successfully introduce the new standard in Legionella Address: detection and identification Postbus/Box 1090 8200 BB Lelystad. Activity: Netherlands Legyon is a diagnostic company offering now; ______The first diagnostic test in the world able to: Contact(s): • distinguish between pathogenic and non-pathogenic Legionella strains(1) but also…… • confirming Legionella species • confirming Legionella pneumophila serotype 1 or serotype 2-14(2) An unique single day test providing all required actionable information to support management in a cost effective Legionella Rik Thijssen (Director Business battle. Development) Maarten Offinga [CEO] Scientific Meeting Friday 16th 17.45 pm Frank Schuren Epidemiological genome analysis of a large Dutch Legionella pneumophila strain collection identifies five markers highly correlated with pathogenic strains Références LIQUITECH, Inc Mission : For 18 years, the liquitech ionization system has been recognizer worlwide as the leading « green » non-toxic solution proven to prevent waterbone pathogens in the healthcare and hospitality industry environments. Everyday some of the most prestigious hospitals and hotels around the world are using our products as a proactive approach to Address : safety. 421 Eisenhower Lane South Lombard, IL 60148 USA Team: Tel : 1 630 693 0500 Steve Schira, President, Liquitech, Inc. Fax : 1 630 693 0505 Guy Heijen, Dir. Marketing LN, Hatenboer-Water Web : www.liquitech.net Xander Terpstra, Director, HydroVRTX BV Chris Derksen, Hatenboer-Water E-Mail : [email protected] OXOID Thermo Fisher Scientific Fournisseur en réactifs de microbiologie Mission : Fournisseur en réactifs de microbiologie Activité : Développement et fabrication de milieux de culture et de tests de diagnostics dans les domaines de la microbiologie alimentaire, pharmaceutique et clinique. Adresse : 6 route de Paisy – BP13 69671 DARDILLY cedex Tel : +33 (0)4 72 52 33 70 Fax : +33 (0)4 78 66 03 76 Web : www.oxoid.fr E-Mail : [email protected] ______Contact(s) : Yann LE BIHAN, chef de produits industrie Marie-Françoise MARQUET, chef de produits clinique Christophe BONGAY, directeur commercial Industrie alimentaire et prestataires : Silliker, Eurofins, Bongrain, Institut Pasteur de Lille, Lilano, Bigard, Nestlé… Industrie pharmaceutique : Beaufour Ipsen, Lilly France, Merial, Novartis, Pierre Fabre, Sanofi-Aventis, GSK… Hôpitaux : AP-HP, AP Marseille, HCL, CHU Bordeaux, CHU Nice, CHU Besancon, CHR Reims… Références PALL MEDICAL Mission: Lead with innovative technologies and service Activity : Delivering water safety products which exceed customer expectation through intelligence, value & performance Address: Team: Pall Medical Catherine Whapham Walton Road Bertrand Coissac Farlington Isabelle Vezole Portsmouth Eric Samuels Stefan Rother PO6 1TD Fabrice Le Gendre Tel : +44 (0)2392 303452 Fax : +44 (0)2392 303324 Web : pall.com/healthcarewater E-Mail : [email protected] ______Contact: Dr Catherine Whapham European Marketing Manager +44 (0)7808 912401 [email protected] m Pall-AquasafeTM Water Filters are indicated for the removal of bacteria, protozoa, fungi and particles from the water supply and provide a barrier to waterborne infection. Benefits include instant and cost effective protection against Legionella spp., high volume throughput, compatible with thermal and chemical systemic treatments, easy installation and an engineering solution and adjunct for water management practices. Pall GeneDisc® Technology enables quantification of Legionella within 3 hours. Rapid reliable results enable infection control experts and water network managers to make quick and effective decisions on appropriate water disinfection measures. Benefits of this latest generation of PCR Références technology include real time detection of contamination, minimal operator intervention, simultaneous read out for Legionella spp. and L. pneumophila for a broad approach to risk assessment and elimination of manual tracking/recording of test results ROCHE DIAGNOSTICS Mission: La division Roche Applied Science de la société Roche Diagnostics à pour mission de fournir des réactifs et des systèmes innovants pour la recherche et le diagnostic expérimental. Address: 2, av du Vercors Activity: B.P.59 Roche Applied Sciences commercialise des outils (systèmes, plate formes automatisées...) et des réactifs pour la recherche 38242 Meylan Cedex notamment en santé humaine et plus particulièrement en Biologie Moléculaire. Le LightCycler est devenu une référence dans la quantification absolue ou relative des acides nucléiques (PCR temps réél). Il est maintenant la référence pour le dépistage de variants par Fusion Haute Résolution (HRM). Phone: + 33(0)4 76 76 30 00 Roche Applied Science apporte également de nouvelles Fax: + 33 (0)4 76 76 30 01 solutions haut débit, que ce soit en séquençage de nouvelle génération (GS Flex) en extraction d’acides nucléiques Website: https://www.roche-applied- (MagnaPure 96) qu’en dépistage ou quantification science.com/ (LightCycler 1536). Nous apportons maintenant de nouveaux outils dans la détection génomique par notre activité Puces Roche Nimblegen, ainsi que de nouveaux outils en Biologie E-Mail: [email protected] Cellulaire (xCELLigence) permettant un suivi en temps réel et en ligne de la croissance cellulaire. Team: Sales Representative : Nicolas Richard Sales Manager : Sylvain Kanamori Marketing Manager : Nathaël Menras Product Manager Genomic systems : Nicolas Brunet Product Manager Core & Cell Analysis : Claire Vigneron Marketing Support : Nathalie Ruggieri & Carole Donne-Goussé Contact Institut Pasteur Nicolas Richard Customer Service Center : Hatim Jawhari, Anne Hamm, Nicola Viebig Sales Rep. [email protected] Phone: +33(0)6 19 41 10 13 +33(0)4 76 76 30 15 E-mail:[email protected]