International Journal of Microbiology

Microbial Ecology and Global Health

Guest Editors: Max Teplitski, Jorge H. Leitão, and Shlomo Sela Microbial Ecology and Global Health International Journal of Microbiology

Microbial Ecology and Global Health

Guest Editors: Max Teplitski, Jorge H. Leitao,˜ and Shlomo Sela Copyright © 2011 Hindawi Publishing Corporation. All rights reserved.

This is a special issue published in volume 2011 of “International Journal of Microbiology.” All articles are open access articles distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, pro- vided the original work is properly cited. International Journal of Microbiology

Editorial Board

Vasco Azevedo, Brazil Marco Gobbetti, Italy Carla Pruzzo, Italy Arvind A. Bhagwat, USA Robert P. Gunsalus, USA R. M. Roop, USA Dulal Borthakur, USA Akira Hiraishi, Japan Kenneth S. Rosenthal, USA Todd R. Callaway, USA Po-Ren Hsueh, Taiwan Isabel Sa-Correia,´ Portugal Michael L. Chikindas, USA Jeroen Hugenholtz, The NetherlandsWilliam M. Shafer, USA P. Patrick Cleary, USA Barbara H. Iglewski, USA Kaarina Sivonen, Finland Luca Simone Cocolin, Italy Vijay K. Juneja, USA J. Glenn Songer, USA Peter Coloe, Australia Thomas L. Kieft, USA A. J. M. Stams, The Netherlands Giuseppe Comi, Italy Sandra Macfarlane, UK David C. Straus, USA Gregory M. Cook, New Zealand MichaelJ.McInerney,USA John Tagg, New Zealand Michael A. Cotta, USA Susana Merino, Spain Effie Tsakalidou, Greece Daniele Daffonchio, Italy Timothy A. Mietzner, USA Michael M. Tunney, Norway EduardoDei-Cas,France Hugh W. Morgan, New Zealand Mumtaz Virji, UK J. Wiegel, USA Joseph Falkinham, USA Ingolf Figved Nes, Norway Marcel H. Zwietering, The Netherlands Paula J. Fedorka-Cray, USA James D. Oliver, USA Arsenio M. Fialho, Portugal Toni L. Poole, USA Contents

Microbial Ecology and Global Health, Max Teplitski, Jorge H. Leitao,andShlomoSela˜ Volume 2011, Article ID 564701, 2 pages

Megacities as Sources for Pathogenic Bacteria in Rivers and Their Fate Downstream, Wolf-Rainer Abraham Volume 2011, Article ID 798292, 13 pages

Defining Established and Emerging Microbial Risks in the Aquatic Environment: Current Knowledge, Implications, and Outlooks,NeilJ.Rowan Volume 2011, Article ID 462832, 15 pages

Prominent Human Health Impacts from Several Marine Microbes: History, Ecology, and Public Health Implications,P.K.Bienfang,S.V.DeFelice,E.A.Laws,L.E.Brand,R.R.Bidigare,S.Christensen, H. Trapido-Rosenthal, T. K. Hemscheidt, D. J. McGillicuddy Jr., D. M. Anderson, H. M. Solo-Gabriele, A.B.Boehm,andL.C.Backer Volume 2011, Article ID 152815, 15 pages

Burkholderia cepacia Complex: Emerging Multihost Pathogens Equipped with a Wide Range of Virulence Factors and Determinants,Sılvia´ A. Sousa, Christian G. Ramos, and Jorge H. Leitao˜ Volume 2011, Article ID 607575, 9 pages

Molecular Detection of Persistent Francisella tularensis Subspecies holarctica in Natural Waters, T. Broman, J. Thelaus, A.-C. Andersson, S. Backman,¨ P. Wikstrom,¨ E. Larsson, M. Granberg, L. Karlsson, E. Back,¨ H. Eliasson, R. Mattsson, A. Sjostedt,¨ and M. Forsman Volume 2011, Article ID 851946, 10 pages

Gene Expression during Survival of Escherichia coli O157:H7 in Soil and Water,AshleyD.Duffitt, Robert T. Reber, Andrew Whipple, and Christian Chauret Volume 2011, Article ID 340506, 12 pages

Quantification of Persistence of Escherichia coli O157:H7 in Contrasting Soils,A.MarkIbekwe, Sharon K. Papiernik, Catherine M. Grieve, and Ching-Hong Yang Volume 2011, Article ID 421379, 11 pages

Adhesion of Pathogenic Bacteria to Food Contact Surfaces: Influence of pH of Culture, Akier Assanta Mafu, Corinne Plumety, Louise Deschenes,ˆ and Jacques Goulet Volume 2011, Article ID 972494, 10 pages

Assessment of Bacterial Antibiotic Resistance Transfer in the Gut, Susanne Schjørring and Karen A. Krogfelt Volume 2011, Article ID 312956, 10 pages

Parenteral Antibiotics Reduce Bifidobacteria Colonization and Diversity in Neonates,Seamus´ Hussey, Rebecca Wall, Emma Gruffman, Lisa O’Sullivan, C. Anthony Ryan, Brendan Murphy, Gerald Fitzgerald, Catherine Stanton, and R. Paul Ross Volume 2011, Article ID 130574, 6 pages Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 564701, 2 pages doi:10.1155/2011/564701

Editorial Microbial Ecology and Global Health

Max Teplitski,1 Jorge H. Leitao,˜ 2 and Shlomo Sela3

1 Soil and Water Sciences Department, Genetics Institute, University of Florida-IFAS, Gainesville, FL 32611, USA 2 Instituto de Biotecnologia e Biogenharia, Centro de Engenharia Biologica´ e Qu´ımica, Instituto Superior T´ecnico, Portugal 3 Microbial Food Safety Research Laboratory, Department of Food Quality and Safety, Agricultural Research Organization (ARO), The Volcani Center, Bet Dagan, Israel

Correspondence should be addressed to Max Teplitski, maxtep@ufl.edu

Received 20 January 2011; Accepted 20 January 2011

Copyright © 2011 Max Teplitski et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This special issue of the International Journal of Microbiol- vision of the pressing research needs for science-based ogy focuses on global health. More specifically, our aim is solutions for the water quality and sanitation crisis in the to establish whether there is a role for microbial ecology in world’s urban population centers. global health. “Global health” is often defined as the “health A critical review by N. J. Rowan, builds a case for the need of populations” a problem that transcends geographic to adapt risk assessment and management framework to boundaries and exists at the intersection of demography, accommodate viable but nonculturable (VBNC) pathogens. economics, epidemiology, political economy, and sociology. N. J. Rowan’s review contains a thoughtful analysis of Conspicuously absent from this definition is microbial ecol- the methods for the detection and quantification of viable ogy. Is there, after all, a role for microorganisms in health of but non-culturable (VBNC) pathogens and Cryptosporidium populations or is “global health” defined strictly by its human oocysts, estimating virulence potential of the pathogens in dimensions? We argue that the analyses of the factors that led the VBNC state and developing risk assessment models that to the emergence and reemergence of pathogens, expansion take into consideration viable but non-culturable as well as of vector-transmitted diseases into the new areas of the drug-resistant pathogens. N. J. Rowan also offers a critical planet, and recent zoonotic outbreaks highlight an important analysis of the efficacy of new water sanitation technologies role for microbial ecology in the health of human popula- and the potential impacts of their byproducts on water tions. Original research and review articles within this special quality. issue contribute to the scientific scaffolding, upon which the Issues associated with the usefulness of indicator organ- risk assessment and management framework is to be built. isms arise again in the review by Bienfang et al. The first Nowhere on this planet are issues of microbial and part of the review by P. K. Bienfang et al. contains a human ecology more important than in large urban slums. thorough survey of marine dinoflagellates and cyanobacteria There, humans and microbes do not only intersect, they that produce toxic secondary metabolites, and comments on collide, and the consequences of this collision are often the hypotheses which link production of toxins with various catastrophic. A review by Abraham traces pathogenic, often environmental factors. The second part of their review is drug-resistant microbes from inadequate sanitation facilities dedicated to infectious marine microorganisms. This discus- in megapolises through rivers toward downstream con- sion is framed by the five dilemmas associated with the use sumers of tainted waters. The review addresses persistence of of indicator organisms to estimate the presence of infectious pathogens in rivers running through the world’s major cities, (introduced and autochthonous) pathogens in seawater. interactions of waterborne pathogens with highly diverse Autochthonous opportunistic pathogens take the center aquatic microbiota, their survival in favorable environmental stage in the review by S. A. Sousa et al. and the environmental niches, in biofilms, and in association with amoeba. The survey of T. Broman et al. Over a 3-year period, T. author then comments on the utility of indicator organisms Broman et al. tested nearly six hundred water and sediment in assessing quality of waters in megacities and offers his samples and almost one hundred wild rodents for the 2 International Journal of Microbiology presence of endemic strains of Francisella tularensis. This mous infamy and thus diminish the role of ecology of study demonstrates that culturable F. tularensis is widely common soil and water microorganisms in the health of distributed in the environment. It tests the utility of certain human populations. Each year throughout the past century, genetic markers for the discrimination between potentially humans were more and more assertive in defining our global pathogenic and nonpathogenic environmental isolates of environment. We rarely, however, reflect on the question of Francisella.S.A.Sousaetal.provideathoroughreviewof how global climate change, population dynamics, methods the epidemiological, comparative, and, functional genomics of food production, crowding in megacities, changes in studies carried out with environmental and pathogenic life styles, and dietary habits affect the ecology of planet’s members of the Burkholderia cepacia complex. The latter part microbiota, the microbiota that shaped this planet, affected of the review by S. A. Sousa et al. contains a critical analysis global health and often defined the course of human of the methods that are currently available to the researchers civilizations. for the identification of virulence determinants in isolates of B. cepacia and for the determination of mechanisms Max Teplitski which these pathogens use to proliferate in the different Jorge H. Leitao˜ environments and infect its vertebrate and invertebrate hosts. Shlomo Sela One of the most disturbing realizations to which sci- entists and lay people came within the last decade is that human pathogens, like enterovirulent E. coli and Salmonella can persist in soils and in association with plants for extended periods of time. Consumption of fruits and vegetables, which harbor these human pathogens has already resulted in dozens of large international outbreaks of gastroenteritis. Not surprisingly, therefore, there has been an explosion in research aimed at understanding behavior of Salmonella and enterovirulent E. coli outside of their human or animal hosts. Research articles by A. D. Duffitetal.andA.M. Ibekwe et al. further contribute to our understanding of the ecology of enterohemorrhagic E. coli O157:H7 in soils and water by defining gene expression associated with the persistence in these environments, in the presence (or absence) of fumigants or native soil microbiota. A. A. Mafu et al. correlate attachment of Salmonella and other human pathogens to abiotic surfaces with culture conditions and surface properties, thus furthering our understanding of persistence of human pathogens outside of their hosts. A review by S. Schjørring and K. A. Krogfelt and a rese- arch article by S. S. Hussey et al. address the impact of antibiotics on human gut-associated microbiota. The study by S.S. Hussey et al. demonstrates that a short-term par- enteral antibiotic administration to human neonates can affect the composition and diversity of the gut associated microbiota.ThefocusofS.SchjørringandK.A.Krogfelt’s review on antibiotic-resistance gene transfer between native gut microbiota and enteric pathogens echoes the reviews by R. Abraham and N. J. Rowan, which appear earlier in the issue, although ecological scales of these three reviews are decidedly different. S. Schjørring and K. A. Krogfelt also tackle the issue of antibiotic resistance in the food production environment, further contributing to our understanding of the ecology of human enteric pathogens outside of their animal hosts. Microorganisms are abundant on this planet. Microbial processes like photosynthesis and nitrogen fixation continue to shape this planet. The Plague, Irish Potato Famine, cholera—no textbook is complete without these events that shaped human history. Our history textbooks, however, never mention by name the microorganisms that reshaped continents and almost halted civilizations. In trying to understand these events, we exile their culprits into anony- Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 798292, 13 pages doi:10.1155/2011/798292

Review Article Megacities as Sources for Pathogenic Bacteria in Rivers and Their Fate Downstream

Wolf-Rainer Abraham

Helmholtz Center for Infection Research, Chemical Microbiology, Inhoffenstrasse 7, 38124 Braunschweig, Germany

Correspondence should be addressed to Wolf-Rainer Abraham, [email protected]

Received 14 May 2010; Revised 9 July 2010; Accepted 19 July 2010

Academic Editor: Jorge H. Leitao

Copyright © 2011 Wolf-Rainer Abraham. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Poor sanitation, poor treatments of waste water, as well as catastrophic floods introduce pathogenic bacteria into rivers, infecting and killing many people. The goal of clean water for everyone has to be achieved with a still growing human population and their rapid concentration in large cities, often megacities. How long introduced pathogens survive in rivers and what their niches are remain poorly known but essential to control water-borne diseases in megacities. Biofilms are often niches for various pathogens because they possess high resistances against environmental stress. They also facilitate gene transfers of antibiotic resistance genes which become an increasing health problem. Beside biofilms, amoebae are carriers of pathogenic bacteria and niches for their survival. An overview about our current understanding of the fate and niches of pathogens in rivers, the multitude of microbial community interactions, and the impact of severe flooding, a prerequisite to control pathogens in polluted rivers, is given.

1. Introduction and health problems caused by such a concentration of humans [11]. Megacities are very dynamic because people A multitude of human activities is usually connected with from rural areas or small cities migrate into megacities with severe impacts on the environment which also includes the hope of a better life. Many of them settle in undeveloped human settlements [1]. The growth of human population areas with insufficient sanitation standards, worsening the over the last decades and their concentration in large cities already existing problems. Usually the development of [2] contribute to the deterioration of water quality due to megacities is hardly controlled and informal settlements intensifications in the industrial processes, domestic sewage within the city lacking any sanitation and clean water are discharge as well as agricultural chemicals and eroded soils the rule [12]. Their waste is washed into nearby rivers which [3]. Urban populations have exploded worldwide over the is the case in all developing countries. Due to the high last 50 years [4]. Today about 50% of the global population demand for water in a city, the river water is often used are living in urban areas [5], placing one-third of their repeatedly and fed into water works before it leaves the city inhabitants into slums [6], and creating huge challenges to [13]. Such an intensive use of the water resources requires their environment and sanitation [7]. In many countries, careful monitoring of the water quality in all water bodies to the rapid development in the last century was not equally exclude risks for human health. This is especially the case for followed by equivalent measures to protect the environment. rivers in megacities. To monitor their impact and to assess Most cities on this planet are located close to rivers which the ecological consequences, a set of physicochemical and serve as transport routes and water supplies [8]. Too often bacteriological parameters (e.g., turbidity, pH, conductivity, these rivers are also used as dump sites for waste water and suspended solids, alkalinity, potassium, sodium, calcium, sewage (Figure 1). The percentage of households with piped magnesium, chloride, nitrate, phosphate, sulphate, chemical or well water nearby or with flush toilets generally decline oxygen demand, 5-day biochemical oxygen demand (BOD5), with city size [9]. Megacities, cities with more than 10 million dissolved oxygen, total coliforms, Escherichia coli, and total inhabitants [10], are textbook examples for environmental heterotrophic bacteria) is analysed usually according to the 2 International Journal of Microbiology

2000 1800 1600 1400 1200 1000 800 600 Precitiptation (mm) 400 200 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Figure 2: Annual precipitation 1986 - 2008 in the Metropolitan Region of Sao˜ Paulo; data taken from the CETESB report 2009 [16]. −1 Figure 1: Pinheiros River in Sao˜ Paulo. The Pinheiros River The mean annual precipitation 1879 - 2008 is 1410 mm a .A as the Tiete River in Sao Paulo passes through the inner part tendency of decreasing rainfall is evident. of the city and is heavily polluted. Pinheiros River is largely anaerobic and heavy methane formation can be observed. In 2008 further down of this place at station PINH 04900 a mean conductivity of 480 μScm−1, 18.67 mg l−1 ammonium, 55.7 mg l−1 before it leaves the city. This is done by pumping water BOD5 and 1,100,000 thermotolerant coliforms per 100 ml have from the two rivers into two large reservoirs, Represa been measured; data taken from the CETESB report 2009 [16]. Guapiranga and Represa Billings. This recycling of waste water becomes increasingly important because of an always growing demand for water in the city. The pressure on the water resources is worsened by decreasing precipitations over Standard Method for the Examination of Water and Wastew- the last decades probably due to climate change (Figure 2). ater [14]. The application of such a procedure revealed a Other cities rely on a steady water supply from mountains, severe impact of urban activities dependent on the quality either due to large catchment areas or melt water from of water treatment on the trophic status of a river [15]. glaciers. This is the situation in Santiago de Chile where the It is also important to elucidate what the fate of water supply for the Maipo and the Mapocho Rivers comes pathogens in the river is, and how fast they are cleared from the Andes, especially the glaciers of the Maipo Vulcan. after leaving the city. The classical way to do this is the Estimations of the effect of global warming predict melting of quantification of colony forming units (cfus) on different these glaciers within the next few decades, depriving Santiago selective agars. Although the selective media used do not de Chile the constant source for its booming needs for water only select for pathogenic but also for related bacteria, which [24]. occur in the environment as well but are not pathogenic, Due to combustion of fossil fuel in combination with their viable cell counts are an important parameter in such heavy deforestation of large areas, the amount of carbon studies [17].This approach only detects bacteria which are dioxide in the atmosphere is increasing, causing a global able to grow and form cfus, and not many other bacteria, climate change [25]. The local effects of this change vary including pathogens known to be difficult to grow [18], those but in general it can be assumed that extreme weather able to form the viable-but-not-culturable state, and others situations will be more frequent [26]. If these floods affect that do not grow at all on any known media [19]. To include also sewage plants, high numbers of pathogens are swept these bacteria and to achieve a complete overview over all into the flooded areas and into the rivers. Low sanitation bacteria present in a habitat, culture-independent methods, standard and incomplete sewage treatment, characteristic usually based on the 16S ribosomal RNA or its gene, have for many cities in poorer countries, are not the only been developed [20]. source of pathogens in rivers, since severe flooding events Not only megacities are under constant changes, the present an important entry for pathogens into rivers even in environment is changing as well [21]. Over the last decade, industrialized countries. These floods after extensive rainfalls a steady increase in global temperature caused by an increase are expected to increase as the result of global warming. in carbon dioxide in the atmosphere has been reported. Extreme precipitation in August 2002 led to the flooding The resulting local climate change differs for individual of large areas of the Elbe River in Germany and the Czech megacities but provokes generally more extreme weather Republic. This flood impacted also the local sewage systems situations, for example, severe floods and longer droughts leading to a release of untreated water into the river. After [22]. The increasing fluctuations in water availability worsen the retreat of the flood, mud enriched by an unknown degree the situation for megacities where the demand for water with facultative pathogenic microorganisms remained. High is rising sharply [23]. Due to the high and still increasing bacterial cell counts were observed in the cellars of the demand for water in Sao˜ Paulo, the water of the Tieteˆ flooded houses, the playgrounds and the streets, forming and Pinheiros Rivers is repeatedly fed into water works a pathogenic reservoir. This is an especially important risk International Journal of Microbiology 3 factor in situations where many persons try to prevent and 1800 to repair flood damages. The high cell counts were not 1600 observed in open water and in wells implying that mud is a 1400 special niche for the survival of pathogenic bacteria [27]. In 1200 megacities, these floods in connection with the insufficient 1000 separation of sewage from the river water potentiate the problem. Heavy rainfalls rather often cause flooding of the 800 TieteRiverinSˆ ao˜ Paulo. This is the case when the TieteRiverˆ 600 cfu per 100 mL Sao Paulo Salto Tiete suddenly receives large water volumes from its tributaries 400 such as the Aricanduva River or the Pinheiros River, which 200 unload thousands of cubic meters in few minutes. The 0 resulting flood wave rapidly raises the TieteRivertoanewˆ 0 50 100 150 200 250 300 350 400 450 500 level. The river than floods first the lateral areas along its Distance from Biritiba-Mirin (SP-088) (km) bank. However, if this is not sufficient, it floods the highways at both sides of the river, severely hampering the trafficin Figure 3: Mean load of thermotolerant coliforms in 2008 along the Tiete River [cfu per 100 mL]. The distance is given in river- Sao˜ Paulo, and transporting large numbers of pathogens in km starting from the first sampling site west of Biritiba-Mirim, Sao˜ the densely populated areas along the river. Paulo, and the locations of main cities are shown. The onset of With the still growing human population, the increasing input of sewage from the city between km 80 and 120 can clearly demand for food and the water needed to produce it, and the be seen in the coliforms load in the river. Downstream from this on-going concentration of humans in large city, the problems area the number of coliforms declines considerably but tributaries, of sanitation will grow as well [28]. The results of both for example, Pinheiros River, or cities in the outskirts of Sao˜ Paulo, extremes, severe droughts and severe floods, are catastrophic for example, Carapicu´ıba, also dump waste in the river which can be to the fragile ecology of megacities. Both situations can seen by the coliforms number between river-km 130–160. Further cause a bloom of pathogens in the water bodies leading down, the number of coliforms declines rapidly but the river is to severe infections of many citizens [29]. Although the still heavily polluted as judged by the much slower decline in the majority of these diseases is caused by “classical” water- biochemical oxygen demand (BOD5). As can be seen from the related pathogens, newly-recognized pathogens are being graph, the cities of Salto and Tiete do not contribute significantly to the thermotolerant coliforms load of Rio Tiete. Data taken from identified that present important additional challenges. New the CETESB report 2009 [16]. agents of disease were discovered, many have reemerged after long periods of inactivity, and others are expanding with the climate change into areas where they have not previously been reported [30]. Assessing the load of pathogens in the water of megacities, understanding the fate of human Pinheiros, Brazil. For the city of Sao˜ Paulo, Brazil, the Tieteˆ pathogens in the environment and their antibiotic resistances River is an important water reservoir; however, especially [31], and identification of niches for their survival will in Sao˜ Paulo with its estimated 25 million inhabitants, the contribute the knowledge base for their control and an water of the Tieteˆ River is heavily loaded with untreated improved management of disasters. waste of all types and it is assumed that the sewage of several million persons is washed without any treatment into the river. Water samples of Tieteˆ River taken at several places 2. Most Rivers in Large Cities Are Polluted and from Sao˜ Paulo to Salto, about 100 km downstream, revealed Harbour Pathogenic Bacteria high loads of pathogens in Sao˜ Paulo, including aggressive pathogens like Escherichia coli O157:H7, Shigella flexneri,and A large number of bacteria, viruses, fungi, protists, and Shigella boydii [33]. Downstream of Sao˜ Paulo both these animalia have been identified to be pathogenic for humans pathogens disappeared rather fast 30 km below Sao˜ Paulo and the majority is water-borne and a study from 2001 and the overall bacterial load also decreased considerably compiled 1415 pathogens [32]. Some of the most important (Figure 3)[34]. This finding indicate that the survival of ones are listed in Table 1. The main source for pathogenic pathogens in river water even in the subtropical climate is bacteria in rivers is sewage. Even for small settlements, rather short but there may be niches including zoonoses pathogens in river water can be a problem if sewage is where they survive longer and pose a long lasting risk for incompletely treated or not treated at all. This problem human health [35]. is especially potentiated for megacities with more than 10 The situation is somewhat different in rivers where million inhabitants. People from rural areas migrate into not a clearly localized source for pathogens exists, but the these cities hoping to get a better income and settle in cheap bacteria are introduced into the stream from a number of housings often devoid of any planning and public control. cities. This is the case for the Ganges River in India. From Most of these wild settlements lack proper sanitation. Varanasi, a city with more than one million inhabitants, an Especially during periods of heavy rain, the sewage including estimated 200 million litres per day of untreated human the faecal sewage is transported into the river, posing a severe sewage is discharged into the Ganges River and faecal danger for health. This was shown in a study of pathogenic coliform counts up to 108 per 100 mL have been observed. bacteria and their antibiotic resistances in the rivers Tieteandˆ The water-borne and enteric disease incidence, including 4 International Journal of Microbiology

Table 1: Important agents of water-borne diseases. acute gastrointestinal disease, cholera, dysentery, hepatitis- A, and typhoid, was estimated to be about 66% during Agent Disease a year. Significant associations were found between water- Bacteria borne disease occurrence and the use of the river for bathing, Vibrio cholerae Cholera, diarrhea, cramps laundry, washing eating utensils, and brushing teeth. Thirty- Vibrio vulnificus, V. three cases of cholera were identified among families exposed alginolyticus, V. Diarrhea, nausea, cramps to washing clothing or bathing in the Ganges compared to parahaemolyticus no cholera cases in unexposed families (104 families studied, Escherichia coli STEC Diarrhea, feces with blood, time period 1 year) [36]. A study on Enterococcus along the etc. vomiting (shigellosis) Ganges River revealed that the number of Enterococcus cells increased along the stream as well as its diversity. Significant Salmonella typhi Fever, diarrhea, delirium antibiotic resistances were observed among the isolates Botulism, respiratory Chlostridium botulinum including vancomycin resistance [37]. This corroborated the failure view that sewage from many cities along the river contributed Pontiac fever, Legionares’ Legionella pneumophila tremendously to the load of pathogens in the water. As disease, pneumonia expected from the high faecal coliform cell counts, a number Meningitis, jaundice, renal Leptospira spp. of pathogens are present in the river water. One of them is failure, head ache Escherichia coli serotype O157:H7, an important pathogen River blindness when of humans [38], causing hemorrhagic colitis and hemolytic- Wolbachia pipientis released from Onchocerca uremic syndrome [39]. It has been calculated that 50 cells volvulus of this serotype can start an infection in humans [40]. The Virus detection of potentially pathogenic O157:H7 bacteria in the Pneumonia, croup, river is alarming due to high risks for visiting pilgrims which Adenovirus bronchitis routinely use the river for religious bathing. Many poorer Jaundice, fatigue, fever, residents along the Ganges River use its water daily for Hepatitis A virus diarrhea bathing, washing laundry, and for cooking [41]. In this river, Poliomyelitis, headache, not only the serotype O157:H7 but also other highly virulent Poliovirus fever, spastic paralysis E. coli strains have been detected [42]. Respiratory infection, Another source for pathogens is manure [43]. To deter- Polyomavirus leukoencephalopathy mine the transport of pathogens from fields into the water, concentrations of human health-related microorganisms Norovirus Vomiting, nausea, cramps in runoff from agricultural plots treated with fresh and Protozoa aged cattle manure and swine slurry were determined. It Entamoeba histolytica Diarrhea, fatigue, fever was shown that large microbial loads could be released Flu-like symptoms, via heavy precipitation events that produce runoffsfrom Cryptosporidium parvum diarrhea, nausea livestock manure-applied agricultural fields and could have a Giardia lamblia Diarrhea significant impact on water bodies within the watershed [44]. From a river in Belgium, the biodiversity of the human Parasites pathogenic bacterium Pseudomonas aeruginosa was anal- Malaria, transmitted by Plasmodium spp. ysed bimonthly over a 1-year period at seven sites evenly Anopheles mosquitoes dispersed. A positive relationship between the extent of Bilharziasis, itching, fever, Schistosoma spp. pollution and the prevalence of P. aeruginosa was found. cough The detected P. aeruginosa community was almost as diverse Nausea, diarrhea, allergic Dracunculus medinensis as the entire global P. aeruginosa population [45] and the reaction river was populated by members of nearly all known clonal Taenia spp. Cysticercosis, loss of weight complexes [46]. With the exception of one multidrug- Diarrhea, liver resistant strain, antibiotic resistance levels were relatively low. Fasciolopsis buski enlargement, cholangitis, These findings illustrate the significance of river water as a jaundice reservoir and source of distribution of potentially pathogenic Abdominal pain, nervous P. aeruginosa strains [47]. Hymenolepis nana manifestation There is a large list of newly emerging or reemerging Echinococcus granulosus Liver enlargement, jaundice pathogens and there is not a clear cut between established, Inflammation, fever, emerging or re-emerging pathogens [48, 49]. In Table 2, Ascaris lumbricoides diarrhea, nausea an attempt has been made to list some of them. Some of Itching, hyperactivity, these pathogens, called re-emerging pathogens, were well Enterobius vermicularis insomnia known for decades, but gained new importance because of newly established aggressive serotypes, for example, Vibrio River blindness, itching, Onchocerca volvulus blindness cholerae O139, several multidrug-resistant pathogens, or new emerging diseases (e.g., AIDS). Fungi are now coming as International Journal of Microbiology 5 well into the limelight as emerging water-borne pathogens Table 2: Some emerging water-borne diseases. [50]. Here the isolation of pathogenic Fusarium [51]and Aspergillus species [52]havebeenreportedfromwater. Agent Disease However, it is still not clear how important water really is for Bacteria their transmission. Vibrio cholerae O139 Diarrhea The number of thermotolerant coliforms is usually Aeromonas spp. Gastroenteritis determined to assess the load of pathogenic bacteria in water. Several studies raised doubts that the mere num- Escherichia coli EHEC Diarrhea ber of E. coli cells or coliform bacteria is sufficient to Yersinia enterocolitica Gastrointestinal infections describe the pathogenic potential of a water system [53]. Campylobacter jejuni Dysentery, high fever, diarrhea Alternatives such as the presence of genetic markers of Wound infections with bad Pseudomonas aeruginosa Bacteroides-Prevotella or pathogenicity factors have been healing, otitis, gastroenteritis proposed [54]. The prevalence and diversity of Salmonella Meningitis, jaundice, renal Leptospira spp. species and their correlation with faecal pollution indicators failure (total coliforms, faecal coliforms, enterococci) and total Mycobacterium spp. Lesions heterotrophic bacteria counts were investigated in several water samples from northern Greek rivers. It was found that Cyanobacteria Neurotoxication the number of Salmonella isolates was higher in summer Fungi than in winter, probably due to the requirement of higher Aspergillus spp. Nosocomial aspergillosis temperatures for the survival of human pathogens [55]. A Virus recent Canadian study revealed a poor relation between the Dengue fever, transmitted by Dengue virus numbers of thermotolerant coliforms and Campylobacter Aedes mosquitoes species and suggested genus-specific monitoring techniques as alternative [56]. Biochemical parameters of different Parvoviruses Gastroenteritis Jaundice, fatigue, fever, isolates from polluted rivers can be used as a fingerprint Hepatitis E virus for a given isolate. Combining these fingerprints allows an diarrhea ff Astrovirus, Calicivirus, assessment [57] and a comparison of di erent water samples Diarrhea, nausea, fever [58]. A biochemical fingerprinting method using enterococci Parvovirus and E. coli has been proposed to provide evidence of septic SARS Fever, lethargy, cough system failure [59]. From the reports, it becomes obvious TT virus (Circoviruses) Hepatitis that there is no group of bacteria which can be cultivated as a Coxsackie B virus Myocarditis detector for all pathogens in a river. It can be concluded that culture-independent methods should be used to monitor the Rotavirus Viral gastroenteritis pathogen load of rivers and that several pathogens should Protozoa be detected simultaneously. Molecular methods allow both Isosporiasis, diarrhea, Isospora belli the detection of unculturable bacteria and the identification abdominal craps of pathogens. This can be done by the detection of specific Toxoplasmosis, fever, muscle Toxoplasma gondii sequences of the 16S rRNA gene or virulence factors or a pain, flu-like syndromes combination of both. It can also include specific regions Blastocystosis, diarrhea, Blastocystis hominis of viruses or ITS regions of the 18S rRNA gene to include nausea, abdominal craps eukaryotes, for example, protozoa, fungi and helminths. One Balantidiasis, diarrhea, Balantidium coli solution can be the use of DNA microarrays tailored for perforation of the colon the specific detection demands [60]. Today there is no gold Microsporidia spp. Diarrhea standard for the detection of all pathogens [61] and gene- based detection methods are still struggling with the problem Cyclospora cayetanensis Nausea, fever, vomiting Headache, fever, nausea, of a cheap and fast way of quantification. Naegleria fowleri pharyngitis Parasites Inflammatory reaction, 3. Bacteria in Rivers Possess Considerable Heterophyes heterophyes Antibiotic Resistances intestinal pain Intestinal pain, nausea, Anisakis simplex The discovery of penicillin by Fleming in 1929 opened diarrhea an entire new way to control bacterial infections [62]. Gnathostoma spp. Fever, vomiting, anorexia The industrial production of penicillin in 1940 and the Abdominal pain, nausea, Angiostrongylus cantonensis subsequent introduction of new antibiotics into medical vomiting, meningitis application saved many lives and raised hopes for permanent Inflammation of the biliary Clonorchis sinensis control of pathogens. However, the continuous and increas- tract, bile adduct carcinoma ing use of antibiotics led to the emergence of pathogenic bacteria resistant to many of these anti-infectiva [63]. During Metagonimus yokogawi Abdominal pain, diarrhea 6 International Journal of Microbiology antibiotic treatment, these resistances are probably generated the respective countries and the occurrence of associated by hypermutating strains [64]. An elevated number of strains resistance was observed. This study indicates that the use of exhibiting high mutation frequencies have recently been antimicrobial agents for growth promotion has been selected reported in the population of many pathogenic bacteria, for resistance to most of these drugs among E. faecium in for example, Pseudomonas aeruginosa in the cystic lung food animals [80]. This has recently been confirmed for [65]. The majority of naturally occurring strong mutators several pathogens by a study of the WHO in Denmark [81]. possessing up to 1000-fold higher than the normal mutation Pathogens with increased resistances are transported rates have an advantage against normal strains for the from the animal via faeces into rivers and groundwater selection of some antibiotic-resistance mutations [66, 67]. [82]. The impact of nontherapeutic use of antibiotics in Horizontal gene transfer is also enhanced in mismatch swine feed on swine manure-impacted water sources has repair defective mutators, facilitating the spread of drug been assessed. The goal of this study was to analyze surface resistance in bacteria. However, hypermutators have a price water and groundwater situated up and down gradient to pay for their fitness which is not the case for weak from a swine facility for antibiotic-resistant enterococci and mutators found in many clinical isolates [68]. Furthermore, other fecal indicators. As expected, the median concentra- the sewage systems are loaded with antibacterials excreted by tions of enterococci, fecal coliforms, and Escherichia coli humans and animals treated for prophylactic and therapeutic were 4- to 33-fold higher in down-gradient versus up- reasons. Several classes of chemotherapeutics have been gradient surface water and groundwater. Higher amounts found in the outlets of sewage treatment plants and can be of erythromycin- and tetracycline-resistant enterococci were detected in rivers [69]. It has been hypothesized that these detected in down-gradient surface waters. Tetracycline- and pharmaceuticals as well contribute to the increasing number clindamycin-resistant enterococci were detected in down- of resistant pathogens[70]. gradient groundwater. These findings demonstrated that Most faecal bacteria from humans released into the water contaminated with swine manure could contribute to environment carry antibiotic resistance genes [71]. Their fate the spread of antibiotic resistance in the environment [83]. and the transfer of antibiotic resistances by gene transfer Plasmids carrying antibiotic resistance genes often to other bacteria are of great concern to human health encode resistance to heavy metals and detergents as well. [72]. Two main mechanisms are involved in the develop- Mercury from dental fillings promotes antibiotic-resistant ment of antibiotic resistance, mutation and acquisition of bacteria in the human mouth [84]butheavymetalsare resistance by horizontal gene transfer. Whereas mutation- also of interest when considering the fate of antibiotic driven resistance usually happens during antibiotic treat- resistances in polluted rivers. The complete genome of the ment, gene transfer-acquired resistance needs a donor of the multidrug-resistant Salmonella enterica serovar Typhi CT18 resistance genes which can be a human-associated bacteria has a large conjugative plasmid that carries 18 genes involved but also an environmental microorganism. The mecha- in resistance to a large number of antimicrobials and heavy nisms for horizontal gene transfer include transformation metals. The plasmid possessed several intact and degen- between different species of bacteria competent for natural erate integrases and transposases. In the chloramphenicol transformation, transduction via viruses, and transfer of resistance cassette, a mercury resistance operon cassette plasmids. Integrons play an important role in horizontal gene is found [85]. An interesting connection between copper- transfer comprising most of the known antibiotic-resistance resistance and antibiotic resistance has been described from gene cassettes [73]. The discovery of genomic islands and Denmark. Copper sulphate is used as a growth-promoting the elucidation of their role in horizontal gene transfer feed supplement for pig production. In 1998, the percentage greatly improved our knowledge of the spread of antibiotic of copper-resistant Enterococcus faecium isolates was found to resistances [74]. Resistance genes are probably moving to be higher from pigs (76%) than those from broilers (34%), plasmids from chromosomes more rapidly than in the past calves (16%), and sheep (<5%), which receive less or no and are aggregating upon plasmids [75]. Goni-Urriza˜ et copper in the feed, and humans (10%). A transferable gene al. found the genetic information for antibiotic resistances tcrB, which confers resistance to copper in enterococci, is mostly in the chromosomes and not on plasmids of isolates located on the same plasmid containing the glycopeptide obtained from a river in Spain [76], contrary to what has and macrolide resistance. The glycopeptides avoparcin was been demonstrated for Aeromonas isolates from rivers [77]. banned in 1995 for growth promotion in Denmark but Another important source of increased resistances is still in use for treatment of sick animals. It seems that against antibiotics is the intensive use of antibiotics in the copper resistance coselects for resistance to macrolides agriculture and fish farming [78], which is regarded as one of and glycopeptides, as genes conferring resistance to them the main reasons for the growing number of multiresistant transfer together and thus are genetically linked. Five out of bacteria [79]. Enterococcus faecium isolated from pigs and five tested glycopeptides resistant E. faecium strains isolated poultry in Denmark, Finland, and Norway were tested for from humans were found to be resistant both to copper and their susceptibility to the antimicrobial agents avilamycin, to macrolides. It seems that these strains and the plasmid avoparcin, bacitracin, flavomycin, monensin, salinomycin, have spread from the porcine reservoir to humans. It can spiramycin, tylosin, and virginiamycin used for growth be assumed that co-selection caused by copper delayed the promotion. Only a limited number of isolates were found decrease of glycopeptide resistance since its ban and it can to be resistant to monensin or salinomycin. In general, an be speculated that this co-selection by copper did not only association between the usage of antimicrobial agents in act in the animals but also in the environment, especially International Journal of Microbiology 7 in the water [86]. This casts a new light on the influence of important role biofilms have as niches for pathogens as it has polluted rivers on the survival of pathogens and antibiotic been shown for Legionella pneumophila and others [105]. resistances [87] and the specific role heavy metals have here To determine the impact bacteria introduced into rivers [88]. in a megacity have on the antibiotic resistances, the antibiotic The pattern of antibiotic resistance of indicator bacteria resistances of isolates obtained from the TieteRiveralongˆ has been used to locate the source of faecal contamination 100 km starting from the city of Sao˜ Paulo, Brazil, were [89] and a classification tree method has been developed determined [35]. The data were compared with those from [90]. Instead of detecting the source of the bacteria, two German rivers. The antibiotic resistances observed in the this approach has been used to locate the source of the Tieteˆ River were generally low and decreased after the major antibiotics. In a study, it was demonstrated that it was input in Sao˜ Paulo to significantly lower levels about 30 not the discharge of a hospital, as assumed, but that of a km downstream. The sensitivity for ampicillin was between pharmaceutical plant that was associated with an increase 33% and 50% for strains from a given site, where the of both single- and multiple-antibiotic resistance among highest sensitivity came from the Sao˜ Paulo site. Gentamycin Acinetobacter species in the sewers [91]. However, not all showed the opposite tendency and the sensitivity increased antibiotic resistances in bacteria are connected with the from the Sao˜ Paulo site (30%) to sites further downstream. medical application of antibiotics. This is confirmed by the Almost all strains tested were sensitive against kanamycin, detection of antibiotic resistances in habitats that are likely only two isolates from Sao˜ Paulo showed resistance against to have been influenced by human activities, for example, this antibiotic. The opposite was the case for novobiocin remote places in the Arctic [92] or the Amazon basin [93], where only one isolate from a site downstream of Sao˜ Paulo the deep terrestrial subsurface [94], or wild rodents [95]. was sensitive. Most isolates were resistant against bacitracin To understand the spread of antibiotic resistances, one has and only 14% sensitive strains were detected. When the to take into account that the pathogens are not isolated in mean resistance of the isolates was determined it was found the river but surrounded by other, nonpathogenic bacteria that each isolate from 100 km downstream of Sao˜ Paulo and a multitude of contaminants which also influence the displayed resistances against 3.57 antibiotics while those spread of resistances [96]. Necessarily, human pathogens from Sao˜ Paulo were resistant against 4.75 antibiotics out were susceptible to antibiotics before the use of these of the seven antibiotics tested. It is interesting to compare drugs for the treatment of infections. Although human these results with strains obtained in Germany from the commensals can provide antibiotic resistance to pathogens, Elbe River showing moderate pollution and the Oker River in most cases, the environmental microbiota is the source with no pollution background. About a quarter of the for the antibiotic resistance genes [97]. The incidence for isolates (24%) from the Elbe River were sensitive against the resistance to ampicillin, chloramphenicol, kanamycin erythromycin or ampicillin and more than half of them nalidixic acid, neomycin, and streptomycin was significantly (53%) could be controlled by gentamycin. From the Elbe higher of native heterotrophic bacteria than for E. coli River, 65% of the isolates could be killed by at least one isolated from several sites along a river in Australia. While of the tested antibiotics and for those from the Oker River multivariate analyses indicated no clear spatial pattern in the number raised to 83%. Comparing the results from the incidence of resistance in native bacteria obtained from Brazil and Germany, differences in antibiotic resistances were clean and from polluted river water, E. coli isolated from found. While the Brazilian isolates were more susceptible clean water samples tended to have a lower incidence of for ampicillin than the German ones, the reverse was resistance than isolates from polluted sites [98]. These results observed for gentamycin [106]. The knowledge about the strongly point against application of antibiotics by humans extent and the origin of antibiotic resistances of pathogens as the only source for the multiresistant strains and support in the environment contributes to our understanding of the view that soil-dwelling bacteria which are exposed to a this phenomenon and can be used for optimal control of myriad of antibiotics evolved at least some of these resistance infections in humans. What we have learned from antibiotic mechanisms [99]. Antibiotic compounds are produced by resistances of environmental bacteria teaches us that there microorganisms in the environment to protect them, but will never be a complete repression of antibiotic resistances also to communicate with other microorganisms in the same but a decent control. To achieve this strategies to control habitat [100]. Therefore, bacteria are used to a rich diversity the emergence and spread of antimicrobial resistance in of antibiotics [101] and have even learned to use them as car- hospitals, infection control should be optimized and focused bon source [102].Environmentalbacteriaareareservoirof on preventing the spread of infections within the health resistance determinants, the resistome that can be mobilized care setting and for antibiotic stewardship minimizing the into the microbial community [103]. This suggests that the emergence of multidrug-resistant organisms by promoting susceptibility of E. coli against antibiotics may not be a good prudent use of antibiotics [107]. bacteriological water quality parameter. Studies carried out in hospital outlets, wastewater treatment, and drinking water distribution systems nearby to the Rhine River, Germany, 4. Pathogens Interact with the Highly Diverse showed that L-lactam-hydrolysing Enterobacteriaceae and Microbial Communities in Polluted Rivers vancomycin-resistant Enterococci could be cultivated from all wastewater biofilms but were found less frequently in Prokaryotes can live in any environment inhabited by higher surface water biofilms [104]. This underlines again the life forms, as well as a variety of inhospitable settings that 8 International Journal of Microbiology any eucaryote would find too hostile [108, 109]. Their or water cooling systems [128] causing severe epidemics, but ability to persist throughout the biosphere is based on their not by consumption of contaminated drinking water. outstanding metabolic versatility and phenotypic plasticity. Biofilms are not only niches for many pathogens. Due Microbes in rivers are diverse and dynamic in composition to their high cell densities, they are also a hot spot for the due to environmental stresses [110] and therefore, the transfer of antibiotic resistances between different bacteria composition of a microbial community in a river has been species. The development of antibiotic resistance in bacteria suggested as an indicator for pollution [111]. In most natural has two components: the selective pressure by exposure environments, association with a surface in a structure to antibiotics and the presence and exchange of resistance known as a biofilm is the prevailing microbial lifestyle genes between different bacteria. Resistance genes can arise [112, 113]. Surface association in rivers is an efficient means from mutations in genes that code for targets of antibiotics, of staying in a favourable microenvironment rather than for proteins involved in their uptake or efflux, or by being swept away by the current. Through attachment, the acquisition of additional genes coding for the detoxification bacteria not only position themselves on a surface, they of antibiotics. Dissemination of resistance is mediated by can form strongly interacting communities and obtain the clonal spread of a particular resistant strain and/or by additional benefit of the phenotypic versatility of their spreading of resistance genes. The latter may involve plasmid neighbours [114] as well. Microbial communities organized transfer, transposition, or dissemination via integrons [129]. in biofilms show a multitude of interactions, including The lateral gene transfer of resistance genes depends on the carbon sharing [115], interspecies communication [116], cell density of the exchanging bacteria. This cell density is and steep physicochemical gradients [117]. They are also especially high in sewage plants with its activated sludge and very well protected against environmental stress factors in biofilms. Therefore, these habitats have been shown to be such as heavy metals, pH shifts, salt stress, or grazing hot spots for the generation of resistant pathogens [130]. [118]. These characteristics make biofilms the preferred Although pathogenic amoebae [131] and nematodes are lifestyle of microorganisms in most habitats [119]. It has known to occur in river waters [132], many of them are also been shown that some bacteria in river biofilms not pathogenic for humans but offer niches for survival of form microcolonies showing significant metal selectivity. pathogens [133]. It has been demonstrated that Legionella, In these microcolonies, the bacterial cells are protected by Salmonella, Shigella, Campylobacter, or Chlamydia species a layer of extracellular polymeric substances against heavy can survive in amoebae where they are much better protected metals like aluminium, iron, or manganese which selectively against disinfecting agents than in the free water [134]. Some adsorbs these ions. This protection mechanism blocks the bacterial pathogens essential require amoebae for survival, diffusion of the metals into the microcolonies keeping the for example, Legionella, Coxiella, or Chlamydia species [135]. concentration of the metals at the surface of bacterial cells at This leads to the wide field of zoonotic diseases where subtoxic levels [120]. The same protection mechanism may pathogens need eucaryotes for survival in the environment also support selection of antibiotic resistances [121]. and/or as shuttle to enter the human body. Biofilms are also niches for several pathogens. Cholera A very interesting finding contributing to our under- is a serious health problem and often regarded as a classical standing of the survival of pathogens in river water has been example for water-borne diseases. The causing bacterium, recently reported. The survival of E. coli O157:H7 in fresh Vibrio cholerae, is associated with epidemic and pandemic water from the Ganges River was compared with the survival cholera. Recently, aquatic biofilms have been identified as in 8-year-old water and in Milli Q water. It was found that niches for the persistence of Vibrio cholerae serotype O1. In the survival time was 3 days in fresh water and 7 days in 8- laboratory microcosms, it was found that cells of V. cholerae year-old water, while in the Milli Q water it was more than O1 were nonculturable in planktonic form, but culturable in 30 days. Survival of E. coli O157:H7 was greater in boiled biofilms after 495 days of incubation and after animal passage water compared with sterile-filtrated water, indicating heat- [122]. This demonstrates that biofilms may act as a reservoir labile compounds influencing the survival of E. coli O157:H7 for V. cholerae between epidemics due to its durable viability. in the river water. These results suggest that Ganges River Cell-cell communication, known as quorum sensing, is an water has certain novel antimicrobial attributes which may essential phenomenon for biofilm formation [123, 124]. In influence the survival of pathogens [136].Thenatureofthe V. cholerae, quorum-sensing was identified to control its underlying mechanism is still unknown and anything from pathogenicity and biofilm formation in an unusual complex antimicrobial peptides to bacteriophages may explain the mechanism composed by three parallel signaling pathways effect. [125] and these pathways certainly respond to many signals produced by other bacteria inhabiting these multispecies biofilms. In the same way, aquatic biofilms may potentially 5. Conclusions act as a reservoir for other pathogens, for example, E. coli O157 [126]. Legionella pneumophila, causing a severe form Polluted rivers harbour microbial communities which are of pneumonia, called Legionnaires’ disease or a milder form, highly diverse and dynamic. However, microbial commu- called Pontiac fever, can be found in parasitizing protozoa, nities in polluted rivers display not only broad functional living in biofilms in river and tap water [127]. The pathogen diversities, but they usually also include bacteria which L. pneumophila is special because it can enter the human are pathogenic for humans and livestock. The occurrence lung through aerosols, for example, from air conditioning of pathogenic bacteria in river water is highly enhanced International Journal of Microbiology 9 near megacities where wastewater treatment is incomplete [3] United Nations Educational, Scientific and Cultural Orga- but also after breakdown of sewage plants, for example, nizations (UNESCO), Water for people—water for life, The after flooding or earthquakes. Although the cell number of United Nations world water development report 1, Barcelona, pathogens declines with the distance to the point source, Spain, 2003. pathogenic bacteria in river water constitute also a poten- [4] D. E. Bloom, D. Canning, and G. Fink, “Urbanization and tial threat because of increasing resistances against many the wealth of nations,” Science, vol. 319, no. 5864, pp. 772– 775, 2008. antibiotics. They adapt to the conditions and environmental [5] C. Dye, “Health and urban living,” Science, vol. 319, no. 5864, changes specific for their habitats and settle in niches where pp. 766–769, 2008. they benefit the most. Due to their organization in consortia [6] UN-HABITAT/(United Nations Human Settlement Pro- and in biofilms, they are able to tolerate conditions which gramme), The Challenge of Slums: Global Report on Human are not tolerable for the free floating cell and they exchange Settlements, Earthscan, London, UK, 2003. genetic information including resistances against several [7] J. Cohen, “Pipe dreams come true,” Science, vol. 319, no. antibiotics. 5864, pp. 745–746, 2008. Many efforts have been undertaken to control pathogens [8] M. J. Paul and J. L. Meyer, “Streams in the urban landscape,” in waste water and to minimize their cell numbers in rivers Annual Review of Ecology and Systematics, vol. 32, pp. 333– and lakes. As has been impressively described in the three 365, 2001. United Nations World Water Development Reports, much [9] United Nations Educational, Scientific and Cultural Orga- has been achieved but much has still to be done. Especially nizations (UNESCO), “Water—a shared responsibility,” The United Nations world water development report 2, Paris, in countries with low income, fast growing populations, and France, 2006. severe water stress, infection rates by water-borne pathogens [10] United Nations (UN), World Urbanization Prospects. The is high. The struggle for clean water for everyone has to be 2003 Revision, United Nations Publications, New York, NY, put into the framework of global warming with its increase USA, 2003. in water stress in many countries and higher frequencies [11] F. Wenzel, F. Bendimerad, and R. Sinha, “Megacities— in extreme weather situations. The results will enhance the megarisks,” Natural Hazards, vol. 42, no. 3, pp. 481–491, outbreak of severe infections of many citizens. Currently, 2007. the majority of these diseases are caused by classical water- [12] M. Batty, “The size, scale, and shape of cities,” Science, vol. related pathogens, however, newly recognized pathogens are 319, no. 5864, pp. 769–771, 2008. being identified presenting important additional risks. To [13] O. Varis, A. K. Biswas, C. Tortajada, and J. Lundqvist, tackle these challenges, not only huge efforts in low-cost “Megacities and water management,” International Journal of Water Resources Development, vol. 22, no. 2, pp. 377–394, technical solutions, socioeconomic support, and governance 2006. activities are needed, but also a deeper knowledge about [14] American Public Health Association (APHA), the Amer- the pathogens is required. We need a close meshed net ican Water Works Association (AWWA), and the Water of monitoring of pathogens in the water of large cities, Environment Federation (WEF), “Standard Methods for the especially megacities, we need low-cost and fast protocols to Examination of Water and Wastewater,” London, UK, 1368 quantify pathogens, preferably at genus-level, and we need to pages, 2005. broaden our understanding of the fate of human pathogens [15] C. A. Almeida, S. Quintar, P. Gonzalez,´ and M. A. Mallea, in the environment and to identify niches for their survival. “Influence of urbanization and tourist activities on the We also need to know how climate changes will influence water quality of the Potrero de los Funes River (San Luis, the survival of pathogens in rivers and the emergence of Argentina),” Environmental Monitoring and Assessment, vol. new pathogens [137] and whether changes in the natural 133, no. 1–3, pp. 459–465, 2007. ecosystems will also lead to changes in the resistance of [16] CETESB (Companhia de Tecnologia de Saneamento Ambi- ental), Secretary of state of the environment, “Annual report human pathogens. This will contribute to the knowledge on the interior water quality of the state of Sao˜ Paulo,” Sao˜ base for the control of human pathogens in megacities and Paulo, Brazil, 2009, http://www.cetesb.sp.gov.br/. to an improved management of disasters. [17] T. M. Straub and D. P. Chandler, “Towards a unified system for detecting waterborne pathogens,” Journal of Microbiolog- ical Methods, vol. 53, no. 2, pp. 185–197, 2003. Acknowledgment [18] J. D. Oliver, “Recent findings on the viable but nonculturable state in pathogenic bacteria,” FEMS Microbiology Reviews, The financial support by the grant “Risk Habitat Megacity” vol. 34, no. 4, pp. 415–425, 2010. of the Helmholtz Association is greatly acknowledged. [19] D. McDougald, S. A. Rice, D. Weichart, and S. Kjelleberg, “Nonculturability: adaptation or debilitation?” FEMS Micro- biology Ecology, vol. 25, no. 1, pp. 1–9, 1998. References [20] M. S. Rappe´ and S. J. Giovannoni, “The uncultured microbial majority,” Annual Review of Microbiology, vol. 57, pp. 369– [1]E.H.Decker,S.Elliott,andF.A.Smith,“Megacitiesandthe 394, 2003. environment,” The Scientific World Journal, vol. 2, pp. 374– [21] N. B. Grimm, S. H. Faeth, N. E. Golubiewski et al., “Global 386, 2002. change and the ecology of cities,” Science, vol. 319, no. 5864, [2] M. R. Montgomery, “The urban transformation of the pp. 756–760, 2008. developing world,” Science, vol. 319, no. 5864, pp. 761–764, [22] M. Sugiyama, H. Shiogama, and S. Emori, “Precipitation 2008. extreme changes exceeding moisture content increases in 10 International Journal of Microbiology

MIROC and IPCC climate models,” Proceedings of the [40]J.Y.Lim,J.W.Yoon,andC.J.Hovde,“Abriefoverviewof National Academy of Sciences of the United States of America, Escherichia coli O157:H7 and its plasmid O157,” Journal of vol. 107, no. 2, pp. 571–575, 2010. Microbiology and Biotechnology, vol. 20, no. 1, pp. 1–10, 2010. [23] World Water Assessment Programme, “Water in a changing [41]S.Hamner,S.C.Broadaway,V.B.Mishraetal.,“Isolation world,” The United Nations world water development report of potentially pathogenic Escherichia coli O157:H7 from the 3, UNESCO, Paris, France/Earthscan, London, UK, 2009. Ganges River,” Applied and Environmental Microbiology, vol. [24] R. Villalba, A. Lara, J. A. Boninsegna et al., “Large-scale 73, no. 7, pp. 2369–2372, 2007. temperature changes across the southern Andes: 20th- [42] S. Ram, P. Vajpayee, and R. Shanker, “Prevalence of multi- century variations in the context of the past 400 years,” antimicrobial-agent resistant shiga toxin and enterotoxin Climatic Change, vol. 59, no. 1-2, pp. 177–232, 2003. producing Escherichia coli in surface waters of river Ganga,” [25] S. Levitus, J. I. Antonov, T. P. Boyer, and C. Stephens, Environmental Science and Technology, vol. 41, no. 21, pp. “Warming of the world oceans,” Science, vol. 287, pp. 2225– 7383–7388, 2007. 2229. [43] J. Venglovsky, N. Sasakova, and I. Placha, “Pathogens and [26] N. A. Keath and R. R. Brown, “Extreme events: being antibiotic residues in animal manures and hygienic and prepared for the pitfalls with progressing sustainable urban ecological risks related to subsequent land application,” water management,” Water Science and Technology, vol. 59, Bioresource Technology, vol. 100, no. 22, pp. 5386–5391, 2009. no. 7, pp. 1271–1280, 2009. [44] J. A. Thurston-Enriquez, J. E. Gilley, and B. Eghball, “Micro- [27] F. C. Cabello, “Heavy use of prophylactic antibiotics in aqua- bial quality of runoff following land application of cattle culture: a growing problem for human and animal health and manure and swine slurry,” Journal of Water and Health, vol. for the environment,” Environmental Microbiology, vol. 8, no. 3, no. 2, pp. 157–171, 2005. 7, pp. 1137–1144, 2006. [45] L. M. C. Hall and S. K. Henderson-Begg, “Hypermutable [28] J. Marshall, “Environmental health: megacity, mega mess,” bacteria isolated from humans—a critical analysis,” Microbi- Nature, vol. 437, no. 7057, pp. 312–314, 2005. ology, vol. 152, no. 9, pp. 2505–2514, 2006. [29] B. D. Slenning, “Global climate change and implications for [46] C. Kiewitz and B. Tummler,¨ “Sequence diversity of Pseu- disease emergence,” Veterinary Patholog y,vol.47,no.1,pp. domonas aeruginosa: impact on population structure and 28–33, 2010. genome evolution,” Journal of Bacteriology, vol. 182, no. 11, [30] WHO, Emerging Issues in Water and Infectious Disease,World pp. 3125–3135, 2000. Health Organization, France, 2003. [47] J.-P. Pirnay, S. Matthijs, H. Colak et al., “Global Pseudomonas [31] WHO, “Overcoming antibiotic resistance,” World Health aeruginosa biodiversity as reflected in a Belgian river,” Organization Report in Infectious Diseases, World Health Environmental Microbiology, vol. 7, no. 7, pp. 969–980, 2005. Organization, Geneva, Switzeland, 2000. [48] P. Karanis, “A review of an emerging waterborne medical [32] L. H. Taylor, S. M. Latham, and M. E.J. Woolhouse, important parasitic protozoan,” Japan, Journal of Protozool- “Risk factors for human disease emergence,” Philosophical ogy, vol. 39, no. 1, pp. 5–18, 2006. Transactions of the Royal Society B, vol. 356, no. 1411, pp. 983–989, 2001. [49] N. Nwachcuku and C. P. Gerba, “Emerging waterborne [33] S. K. Niyogi, “Shigellosis,” Journal of Microbiology, vol. 43, no. pathogens: can we kill them all?” Current Opinion in 2, pp. 133–143, 2005. Biotechnology, vol. 15, no. 3, pp. 175–180, 2004. [34] W.-R. Abraham, A. J. Macedo, L. H. Gomes, and F. C. A. [50] G. M. Merlani and P. Francioli, “Established and emerging Tavares, “Occurrence and resistance of pathogenic bacteria waterborne nosocomial infections,” Current Opinion in Infec- along the Tieteˆ River downstream of Sao˜ Paulo in Brazil,” tious Diseases, vol. 16, no. 4, pp. 343–347, 2003. Clean, vol. 35, no. 4, pp. 339–347, 2007. [51] E. J. Anaissie, R. T. Kuchar, J. H. Rex et al., “Fusariosis [35]K.C.Ho,Y.L.Chow,andJ.T.S.Yau,“Chemicaland associated with pathogenic Fusarium species colonization microbiological qualities of The East River (Dongjiang) of a hospital water systemml: a new paradigm for the water, with particular reference to drinking water supply in epidemiology of opportunistic mold infections,” Clinical Hong Kong,” Chemosphere, vol. 52, no. 9, pp. 1441–1450, Infectious Diseases, vol. 33, no. 11, pp. 1871–1878, 2001. 2003. [52] E. J. Anaissie, S. L. Stratton, M. C. Dignani et al., “Pathogenic [36] S. Hamner, A. Tripathi, R. K. Mishra et al., “The role Aspergillus species recovered from a hospital water systemml: of water use patterns and sewage pollution in incidence a 3-year prospective study,” Clinical Infectious Diseases, vol. of water-borne/enteric diseases along the Ganges River 34, no. 6, pp. 780–789, 2002. in Varanasi, India,” International Journal of Environmental [53] T. M. Straub and D. P. Chandler, “Towards a unified system Health Research, vol. 16, no. 2, pp. 113–132, 2006. for detecting waterborne pathogens,” Journal of Microbiolog- [37] P. Lata, S. Ram, M. Agrawal, and R. Shanker, “Entero- ical Methods, vol. 53, no. 2, pp. 185–197, 2003. cocci in river Ganga surface waters: propensity of species [54] S. Okabe and Y. Shimazu, “Persistence of host-specific distribution, dissemination of antimicrobial-resistance and Bacteroides-Prevotella 16S rRNA genetic markers in environ- virulence-markers among species along landscape,” BMC mental waters: effects of temperature and salinity,” Applied Microbiology, vol. 9, article no. 140, pp. 1–10, 2009. Microbiology and Biotechnology, vol. 76, no. 4, pp. 935–944, [38] P. Kaur, A. Chakraborti, and A. Asea, “Enteroaggregative 2007. Escherichia coli: an emerging enteric food borne pathogen,” [55] M. Arvanitidou, K. Kanellou, and D. G. Vagiona, “Diversity Interdisciplinary Perspectives on Infectious Diseases, vol. 2010, of Salmonella spp. and fungi in northern Greek rivers and Article ID 254159, 10 pages, 2010. their correlation to fecal pollution indicators,” Environmental [39] H. Karch, P. I. Tarr, and M. Bielaszewska, “Enterohaem- Research, vol. 99, no. 2, pp. 278–284, 2005. orrhagic Escherichia coli in human medicine,” International [56] K. St-Pierre, S. Levesque,´ E. Frost, N. Carrier, R. D. Arbeit, Journal of Medical Microbiology, vol. 295, no. 6-7, pp. 405– and S. Michaud, “Thermotolerant coliforms are not a good 418, 2005. surrogate for Campylobacter spp. in environmental water,” International Journal of Microbiology 11

Applied and Environmental Microbiology, vol. 75, no. 21, pp. [72] B. H. Normark and S. Normark, “Evolution and spread of 6736–6744, 2009. antibiotic resistance,” Journal of Internal Medicine, vol. 252, [57] I. Kuhn,¨ G. Allestam, T. A. Stenstrom, and R. Mollby, no. 2, pp. 91–106, 2002. “Biochemical fingerprinting of water coliform bacteria, a [73] D. Mazel, “Integrons: agents of bacterial evolution,” Nature new method for measuring phenotypic diversity and for Reviews Microbiology, vol. 4, no. 8, pp. 608–620, 2006. comparing different bacterial populations,” Applied and [74] M. Juhas, J. R. Van Der Meer, M. Gaillard, R. M. Harding, Environmental Microbiology, vol. 57, no. 11, pp. 3171–3177, D. W. Hood, and D. W. Crook, “Genomic islands: tools 1991. of bacterial horizontal gene transfer and evolution,” FEMS [58] I. Kuhn,¨ G. Allestam, M. Engdahl, and T.-A. Stenstrom,¨ “Bio- Microbiology Reviews, vol. 33, no. 2, pp. 376–393, 2009. chemical fingerprinting of coliform bacterial populations— [75] M. Barlow, “What antimicrobial resistance has taught us comparisons between polluted river water and factory efflu- about horizontal gene transfer,” Methods in Molecular Biol- ents,” Water Science and Technology, vol. 35, no. 11-12, pp. ogy, vol. 532, pp. 397–411, 2009. 343–350, 1997. [76] M. Goni-Urriza,˜ M. Capdepuy, C. Arpin, N. Raymond, [59] W. Ahmed, R. Neller, and M. Katouli, “Evidence of septic and C. Q. Pierre Caumette, “Impact of an urban effluent system failure determined by a bacterial biochemical finger- on antibiotic resistance of riverine Enterobacteriaceae and printing method,” Journal of Applied Microbiology, vol. 98, Aeromonas spp,” Applied and Environmental Microbiology, no. 4, pp. 910–920, 2005. vol. 66, no. 1, pp. 125–132, 2000. [60] D. R. Call, M. K. Borucki, and F. J. Loge, “Detection of [77] G. Rhodes, G. Huys, J. Swings et al., “Distribution of bacterial pathogens in environmental samples using DNA oxytetracycline resistance plasmids between aeromonads microarrays,” Journal of Microbiological Methods, vol. 53, no. in hospital and aquaculture environments: implication of 2, pp. 235–243, 2003. Tn1721 in dissemination of the tetracycline resistance deter- [61] J. R. Stewart, R. J. Gast, R. S. Fujioka et al., “The minant Tet A,” Applied and Environmental Microbiology, vol. coastal environment and human health: microbial indicators, 66, no. 9, pp. 3883–3890, 2000. pathogens, sentinels and reservoirs,” Environmental Health, [78] A. S. Schmidt, M. S. Bruun, I. Dalsgaard, K. Pedersen, and vol. 7, supplement 2, article no. S3, 2008. J. L. Larsen, “Occurrence of antimicrobial resistance in fish- [62] A. Fleming, “On the antibacterial action of cultures of a pathogenic and environmental bacteria associated with four Penicillium, with special reference to their use in the isolation danish rainbow trout farms,” Applied and Environmental of B. influenzæ,” ReviewsofInfectiousDiseases, vol. 10, no. 31, Microbiology, vol. 66, no. 11, pp. 4908–4915, 2000. pp. 226–236, 1929. [79] D. J. Austin, K. G. Kristinsson, and R. M. Anderson, [63] S. B. Levy and B. Marshall, “Antibacterial resistance world- “The relationship between the volume of antimicrobial wide: causes, challenges and responses,” Nature Medicine, vol. consumption in human communities and the frequency of 10, no. 12, pp. S122–S129, 2004. resistance,” Proceedings of the National Academy of Sciences of [64] N. Woodford and M. J. Ellington, “The emergence of the United States of America, vol. 96, no. 3, pp. 1152–1156, antibiotic resistance by mutation,” Clinical Microbiology and 1999. Infection, vol. 13, no. 1, pp. 5–18, 2007. [65] A. Oliver, R. Canton,P.Campo,F.Baquero,andJ.Bl´ azquez,´ [80] F. M. Aarestrup, H. Kruse, E. Tast, A. M. Hammerum, and “High frequency of hypermutable Pseudomonas aeruginosa in L. B. Jensen, “Associations between the use of antimicrobial cystic fibrosis lung infection,” Science, vol. 288, no. 5469, pp. agents for growth promotion and the occurrence of resis- 1251–1253, 2000. tance among Enterococcus faecium from broilers and pigs in [66] J. E. LeClerc, B. Li, W. L. Payne, and T. A. Cebula, “High Denmark, Finland, and Norway,” Microbial Drug Resistance, mutation frequencies among Escherichia coli and Salmonella vol. 6, no. 1, pp. 63–70, 2000. pathogens,” Science, vol. 274, no. 5290, pp. 1208–1211, 1996. [81] D. Ferber, “Antibiotic resistance. WHO advises kicking the [67] M. D. Mac´ıa, D. Blanquer, B. Togores, J. Sauleda, J. L. Perez,´ livestock antibiotic habit,” Science, vol. 301, no. 5636, p. 1027, and A. Oliver, “Hypermutation is a key factor in development 2003. of multiple-antimicrobial resistance in Pseudomonas aerugi- [82] F. Baquero, J.-L. Mart´ınez, and R. Canton,´ “Antibiotics nosa strains causing chronic lung infections,” Antimicrobial and antibiotic resistance in water environments,” Current Agents and Chemotherapy, vol. 49, no. 8, pp. 3382–3386, Opinion in Biotechnology, vol. 19, no. 3, pp. 260–265, 2008. 2005. [83] A. R. Sapkota, F. C. Curriero, K. E. Gibson, and K. J. Schwab, [68] M.-R. Baquero, A. I. Nilsson, M. D. C. Turrientes et “Antibiotic-resistant enterococci and fecal indicators in sur- al., “Polymorphic mutation frequencies in Escherichia coli: face water and groundwater impacted by a concentrated emergence of weak mutators in clinical isolates,” Journal of swine feeding operation,” Environmental Health Perspectives, Bacteriology, vol. 186, no. 16, pp. 5538–5542, 2004. vol. 115, no. 7, pp. 1040–1045, 2007. [69] K. Kummerer,¨ “Antibiotics in the aquatic environment—a [84]A.O.Summers,J.Wireman,M.J.Vimyetal.,“Mercury review—part I,” Chemosphere, vol. 75, no. 4, pp. 417–434, released from dental ’silver’ fillings provokes an increase 2009. in mercury- and antibiotic-resistant bacteria in oral and [70] E.-B. Goh, G. Yim, W. Tsui, J. McClure, M. G. Surette, intestinal floras of primates,” Antimicrobial Agents and and J. Davies, “Transcriptional modulation of bacterial gene Chemotherapy, vol. 37, no. 4, pp. 825–834, 1993. expression by subinhibitory concentrations of antibiotics,” [85] J. Parkhill, G. Dougan, K. D. James et al., “Complete genome Proceedings of the National Academy of Sciences of the United sequence of a multiple drug resistant Salmonella enterica States of America, vol. 99, no. 26, pp. 17025–17030, 2002. serovar Typhi CT18,” Nature, vol. 413, no. 6858, pp. 848–852, [71] H. K. Allen, J. Donato, H. H. Wang, K.A. Cloud-Hansen, 2001. J. Davies, and J. Handelsman, “Call of the wild: antibiotic [86] H. Hasman and F. M. Aarestrup, “Relationship between cop- resistance genes in natural environments,” Nature Reviews per, glycopeptide, and macrolide resistance among Enterococ- Microbiology, vol. 8, no. 4, pp. 251–259, 2010. cus faecium strains isolated from pigs in Denmark between 12 International Journal of Microbiology

1997 and 2003,” Antimicrobial Agents and Chemotherapy, vol. [103] V. M. D’Costa, K. M. McGrann, D. W. Hughes, and G. D. 49, no. 1, pp. 454–456, 2005. Wright, “Sampling the antibiotic resistome,” Science, vol. 311, [87] P. K. Dhakephalkar and B. A. Chopade, “High levels of multi- no. 5759, pp. 374–377, 2006. ple metal resistance and its correlation to antibiotic resistance [104] T. Schwartz, W. Kohnen, B. Jansen, and U. Obst, “Detection in environmental isolates of Acinetobacter,” BioMetals, vol. 7, of antibiotic-resistant bacteria and their resistance genes no. 1, pp. 67–74, 1994. in wastewater, surface water, and drinking water biofilms,” [88] S. Silver and L. T. Phung, “Bacterial heavy metal resistance: FEMS Microbiology Ecology, vol. 43, no. 3, pp. 325–335, 2003. new surprises,” Annual Review of Microbiology, vol. 50, pp. [105] P. Declerck, “Biofilms: the environmental playground of 753–789, 1996. Legionella pneumophila,” Environmental Microbiology, vol. [89] V. J. Harwood, J. Whitlock, and V. Withington, “Classifi- 12, no. 3, pp. 557–566, 2010. cation of antibiotic resistance patterns of indicator bacteria [106] W.-R. Abraham and D. F. Wenderoth, “Fate of facultative by discriminant analysis: use in predicting the source of pathogenic microorganisms during and after the flood of the fecal contamination in subtropical waters,” Applied and Elbe and Mulde rivers in August 2002,” Acta Hydrochimica et Environmental Microbiology, vol. 66, no. 9, pp. 3698–3704, Hydrobiologica, vol. 33, no. 5, pp. 449–454, 2005. 2000. [107] D. J. Anderson and K. S. Kaye, “Controlling antimicrobial [90] B. Price, E. A. Venso, M. F. Frana, J. Greenberg, A. Ware, and resistance in the hospital,” Infectious Disease Clinics of North L. Currey, “Classification tree method for bacterial source America, vol. 23, no. 4, pp. 847–864, 2009. tracking with antibiotic resistance analysis data,” Applied and [108] P. Lopez-Garc´ ´ıa and D. Moreira, “Tracking microbial biodi- Environmental Microbiology, vol. 72, no. 5, pp. 3468–3475, versity through molecular and genomic ecology,” Research in 2006. Microbiology, vol. 159, no. 1, pp. 67–73, 2008. [91] L. Guardabassi, A. Petersen, J. E. Olsen, and A. Dalsgaard, [109] E. F. DeLong and N. R. Pace, “Environmental diversity of “Antibiotic resistance in Acinetobacter spp. isolated from bacteria and archaea,” Systematic Biology,vol.50,no.4,pp. sewers receiving waste effluent from a hospital and a phar- 470–478, 2001. maceutical plant,” Applied and Environmental Microbiology, [110] B. Nogales, MA. M. Aguilo-Ferretjans,´ C. Mart´ın-Cardona, vol. 64, no. 9, pp. 3499–3502, 1998. J. Lalucat, and R. Bosch, “Bacterial diversity, composition [92] H. K. Allen, L. A. Moe, J. Rodbumrer, A. Gaarder, and J. and dynamics in and around recreational coastal areas,” Handelsman, “Functional metagenomics reveals diverse β- Environmental Microbiology, vol. 9, no. 8, pp. 1913–1929, lactamases in a remote Alaskan soil,” ISME Journal, vol. 3, 2007. no. 2, pp. 243–251, 2009. [111] R. Atlas, “Use of microbial diversity measurements to assess [93] A. Bartoloni, L. Pallecchi, H. Rodr´ıguez et al., “Antibiotic environmental stress,” in Current Perspectives in Microbial resistance in a very remote Amazonas community,” Interna- Ecology, M. J. Klug and C. A. Reddy, Eds., pp. 540–545, tional Journal of Antimicrobial Agents, vol. 33, no. 2, pp. 125– American Society for Microbiology, Washington, DC, USA, 129, 2009. 1984. [94] M. G. Brown and D. L. Balkwill, “Antibiotic resistance [112] J. W. Costerton, Z. Lewandowski, D. E. Caldwell, D. R. in bacteria isolated from the deep terrestrial subsurface,” Korber, and H. M. Lappin-Scott, “Microbial biofilms,” Microbial Ecology, vol. 57, no. 3, pp. 484–493, 2009. Annual Review of Microbiology, vol. 49, pp. 711–745, 1995. [95]M.A.Gilliver,M.Bennett,M.Begon,S.M.Hazel,andC.A. [113] L. Hall-Stoodley, J. W. Costerton, and P. Stoodley, “Bacterial Hart, “Antibiotic resistance found in wild rodents,” Nature, biofilms: from the natural environment to infectious dis- vol. 401, no. 6750, pp. 233–234, 1999. eases,” Nature Reviews Microbiology, vol. 2, no. 2, pp. 95–108, [96] J. L. Martinez, “The role of natural environments in the evo- 2004. lution of resistance traits in pathogenic bacteria,” Proceedings [114] M. E. Davey and G. A. O’Toole, “Microbial biofilms: from of the Royal Society B, vol. 276, no. 1667, pp. 2521–2530, 2009. ecology to molecular genetics,” Microbiology and Molecular [97] J. Davies, “Inactivation of antibiotics and the dissemination Biology Reviews, vol. 64, no. 4, pp. 847–867, 2000. of resistance genes,” Science, vol. 264, no. 5157, pp. 375–382, [115] A. T. Nielsen, T. Tolker-Nielsen, K. B. Barken, and S. Molin, 1994. “Role of commensal relationships on the spatial structure [98] P. I. Boon and M. Cattanach, “Antibiotic resistance of of a surface-attached microbial consortium,” Environmental native and faecal bacteria isolated from rivers, reservoirs Microbiology, vol. 2, no. 1, pp. 59–68, 2000. and sewage treatment facilities in Victoria, south-eastern [116] B. Schachter, “Slimy business—the biotechnology of Australia,” Letters in Applied Microbiology, vol. 28, no. 3, pp. biofilms,” Nature Biotechnology, vol. 21, no. 4, pp. 361–365, 164–168, 1999. 2003. [99] A. Alonso, P. Sanchez,´ and J. L. Mart´ınez, “Environmen- [117] L. Hall-Stoodley, J. W. Costerton, and P. Stoodley, “Bacterial tal selection of antibiotic resistance genes,” Environmental biofilms: from the natural environment to infectious dis- Microbiology, vol. 3, no. 1, pp. 1–9, 2001. eases,” Nature Reviews Microbiology, vol. 2, no. 2, pp. 95–108, [100] J. L. Mart´ınez, “Antibiotics and antibiotic resistance genes in 2004. natural environments,” Science, vol. 321, no. 5887, pp. 365– [118] C. Matz, T. Bergfeld, S. A. Rice, and S. Kjelleberg, “Micro- 367, 2008. colonies, quorum sensing and cytotoxicity determine the [101] H. P. Bais, B. Prithiviraj, A. K. Jha, F. M. Ausubel, and J. M. survival of Pseudomonas aeruginosa biofilms exposed to Vivanco, “Mediation of pathogen resistance by exudation of protozoan grazing,” Environmental Microbiology, vol. 6, no. antimicrobials from roots,” Nature, vol. 434, no. 7030, pp. 3, pp. 218–226, 2004. 217–221, 2005. [119] P. Watnick and R. Kolter, “Biofilm, city of microbes,” Journal [102] G. Dantas, M. O. A. Sommer, R. D. Oluwasegun, and G. M. of Bacteriology, vol. 182, no. 10, pp. 2675–2679, 2000. Church, “Bacteria subsisting on antibiotics,” Science, vol. 320, [120] H. Lunsdorf,¨ I. Brummer,¨ K. N. Timmis, and I. no. 5872, pp. 100–103, 2008. Wagner-Dobler,¨ “Metal selectivity of in situ microcolonies in International Journal of Microbiology 13

biofilms of the Elbe river,” JournalofBacteriology, vol. 179, no. 1, pp. 31–40, 1997. [121] A. Hernandez,´ R. P. Mellado, and J. L. Mart´ınez, “Metal accumulation and vanadium-induced multidrug resistance by environmental isolates of Escherichia hermannii and Enterobacter cloacae,” Applied and Environmental Microbiol- ogy, vol. 64, no. 11, pp. 4317–4320, 1998. [122] M. Alam, M. Sultana, G. Balakrish Nair et al., “Viable but nonculturable Vibrio cholerae O1 in biofilms in the aquatic environment and their role in cholera transmission,” Proceedings of the National Academy of Sciences of the United States of America, vol. 104, no. 45, pp. 17801–17806, 2007. [123] W.-R. Abraham, “Controlling pathogenic gram-negative bacteria by interfering with their biofilm formation,” Drug Design Reviews Online, vol. 2, no. 1, pp. 13–33, 2005. [124] W.-R. Abraham, “Controlling the biofilm formation of Gram-positive pathogenic bacteria,” Current Medicinal Chemistry, vol. 13, pp. 1509–1524, 2006. [125] M. B. Miller, K. Skorupski, D. H. Lenz, R. K. Taylor, and B. L. Bassler, “Parallel quorum sensing systems converge to regulate virulence in Vibrio cholerae,” Cell, vol. 110, no. 3, pp. 303–314, 2002. [126] I. R. Cooper, H. D. Taylor, and G. W. Hanlon, “Virulence traits associated with verocytotoxigenic Escherichia coli O157 recovered from freshwater biofilms,” Journal of Applied Microbiology, vol. 102, no. 5, pp. 1293–1299, 2007. [127] M. Steinert, U. Hentschel, and J. Hacker, “Legionella pneu- mophila: an aquatic microbe goes astray,” FEMS Microbiology Reviews, vol. 26, no. 2, pp. 149–162, 2002. [128] D. W. Fraser, T. R. Tsai, and W. Orenstein, “Legionnaires’ disease. Description of an epidemic of pneumonia,” New England Journal of Medicine, vol. 297, no. 22, pp. 1189–1197, 1977. [129] W. Witte, “Ecological impact of antibiotic use in animals on different complex microflora: environment,” International Journal of Antimicrobial Agents, vol. 14, no. 4, pp. 321–325, 2000. [130] J. Davison, “Genetic exchange between bacteria in the environment,” Plasmid, vol. 42, no. 2, pp. 73–91, 1999. [131] M. M. Marshall, D. Naumovitz, Y. Ortega, and C. R. Sterling, “Waterborne protozoan pathogens,” Clinical Microbiology Reviews, vol. 10, no. 1, pp. 67–85, 1997. [132] A. A. Gajadhar, W. B. Scandrett, and L. B. Forbes, “Overview of food- and water-borne zoonotic parasites at the farm level,” OIE Revue Scientifique et Technique,vol.25,no.2,pp. 595–606, 2006. [133] F. Bichai, P. Payment, and B. Barbeau, “Protection of water- borne pathogens by higher organisms in drinking water: a review,” Canadian Journal of Microbiology,vol.54,no.7,pp. 509–524, 2008. [134] C. H. King, E. B. Shotts Jr., R. E. Wooley, and K. G. Porter, “Survival of coliforms and bacterial pathogens within protozoa during chlorination,” Applied and Environmental Microbiology, vol. 54, no. 12, pp. 3023–3033, 1988. [135] F. Lamoth and G. Greub, “Amoebal pathogens as emerging causal agents of pneumonia,” FEMS Microbiology Reviews, vol. 34, no. 3, pp. 260–280, 2010. [136] C. S. Nautiyal, “Self-purificatory ganga water facilitates death of pathogenic Escherichia coli O157:H7,” Current Microbiology, vol. 58, no. 1, pp. 25–29, 2009. [137] L. C. Ivers and E. T. Ryan, “Infectious diseases of severe weather-related and flood-related natural disasters,” Current Opinion in Infectious Diseases, vol. 19, no. 5, pp. 408–414, 2006. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 462832, 15 pages doi:10.1155/2011/462832

Review Article Defining Established and Emerging Microbial Risks in the Aquatic Environment: Current Knowledge, Implications,and Outlooks

Neil J. Rowan

Department of Nursing and Health Science, School of Science, Athlone Institute of Technology, Dublin Road, Athlone, Co. Westmeath, Ireland

Correspondence should be addressed to Neil J. Rowan, [email protected]

Received 22 March 2010; Accepted 27 July 2010

Academic Editor: Max Teplitski

Copyright © 2011 Neil J. Rowan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This timely review primarily addresses important but presently undefined microbial risks to public health and to the natural environment. It specifically focuses on current knowledge, future outlooks and offers some potential alleviation strategies that may reduce or eliminate the risk of problematic microbes in their viable but nonculturable (VBNC) state and Cryptosporidium oocysts in the aquatic environment. As emphasis is placed on water quality, particularly surrounding efficacy of decontamination at the wastewater treatment plant level, this review also touches upon other related emerging issues, namely, the fate and potential ecotoxicological impact of untreated antibiotics and other pharmaceutically active compounds in water. Deciphering best published data has elucidated gaps between science and policy that will help stakeholders work towards the European Union’s Water Framework Directive (2000/60/EC), which provides an ambitious legislative framework for water quality improvements within its region and seeks to restore all water bodies to “good ecological status” by 2015. Future effective risk-based assessment and management, post definition of the plethora of dynamic inter-related factors governing the occurrence, persistence and/or control of these presently undefined hazards in water will also demand exploiting and harnessing tangential advances in allied disciplines such as mathematical and computer modeling that will permit efficient data generation and transparent reporting to be undertaken by well-balanced consortia of stakeholders.

1. Viable But Nonculturable Forms of immediately culturable state) may potentially pose as yet an Waterborne Bacteria undefined risk to public health, which is attested by the fact that there is increasing evidence to support the viewpoint 1.1. Background. Since the introduction of the concept or that stressed cells in this quiescent state may actually be more sublethally injured or viable but nonculturable (VBNC) cells virulent than well-fed laboratory-tamed microorganisms by Byrd and Colwell in the 1980’s [1], there is increasing due to augmented virulence factor expression. Xu et al. [5] evidence for the existence of such a state in microbes, were the first to bring experimental evidence of the existence particularly in the aquatic environment that elicits a myriad of VBNC state in pathogenic bacteria, where they showed of interrelated sub-lethal microbial stresses such as nutrient that E. coli and V. cholera cells that were suspended in starvation and osmotic stress [2, 3](Table1). This is a cause artificial seawater quickly lost their ability to grow on the for concern because of evidence that microbial pathogens in culture media normally used for their detection. such a state may still retain their capacity to cause infections after ingestion by fish, animals, or by humans, despite their 1.2. Definition. According to Oliver [16],abacteriumin inability to grow under conditions employed in laboratory- the VBNC state is defined as “a cell which is metabolically based procedures for determining their presence in water active, which being incapable of undergoing the cellular [4]. Albeit currently unknown in terms of its severity or division required for growth in or on a medium normally scope, it is now generally appreciated that heavily stressed supporting grown of that cell.” Besnard et al. [17]suggest pathogenic microbial species existing in a VBNC (or not that the transition to the VBNC state in L. monocytogenes 2 International Journal of Microbiology

Table 1: Methods used to detect VBNC state in waterborne microorganisms.

Method(s) Employed Reporting author(s)∗ Failure of microbial growth in culture media [5] Use of redox probes to detect microbial respiratory chain activity [6, 7] Incorporation of radio-labelled substrates in culture media [8] Resuscitation in embryo of egg yolk [4] Detection in immunodeficient mice [9] Addition of antioxidants to culture media [10] RNA-based genotypic approaches (16S/23S rRNA, mRNA) [11] cDNA microarrays [2] In situ hybridisation (FISH), microradiography, epi-fluorescence microscopy, flow cytometry [12] Rapid enzyme assays [13] Oligonucleotide probes and tagged green fluorescent protein [14] Microbial quorum sensing [15] ∗ This is a representative list of authors citing use of named methods for detection of VBNC state in waterborne organisms and therefore does not convey all published work in this area.

represents a survival strategy that bacteria can adopt under L. monocytogenes isolated from salmon, patients and the adverse conditions (starvation, salt stress, etc.). VBNC environment. L. monocytogenes were tested for virulence in microorganisms are considered to represent a subpopulation a cell plaque assay and by intraperitoneally inoculation in of cells that are unable to grow in the usual culture media and immunodeficient RAG1 mice. Moreover, Moreno et al. [22] cannot resuscitate by traditional resuscitation techniques, described successions in cellular alterations in Helicobac- but yet remain physically active for several functions such ter pylori NCTC 11637 after inoculation into chlorinated as cellular elongation [18], respiratory chain activity [6, drinking water. They concluded that H. pylori could survive 7, 17], or incorporation of radio-labelled substrates [8]. disinfection practices normally used in drinking water For example, Cappelier and coworkers [4] recently reported treatment in the VBNC form, which would allow them that avirulent VBNC cells of L. monocytogenes incubated to reach final consumption points and, at the same time, in filtered sterilized distilled water need the presence of an enable them to be undetectable by culture methods. Whereas embryo to be recovered in egg yolk and regain virulence after Kastberg et al. [23] recently reported that L. monocytogenes recovery. The VBNC state was observed after a 25 to 47 days cells, whether planktonic or attached, were homogenous with incubation period (concentration of culturable cells less than respect to sensitivity to acidic disinfectants at the single-cell 1 colony forming unit per mL). level.

1.3. VBNC State and Occurrence of Atypical Morphologi- 1.4. Use of Fluorescent Redox and Other Enzymatic Probes. cal Types. As microorganisms are extremely diverse and Direct visualization of actively respiring bacteria is gaining in dynamic, it is not surprising that the many different types popularity amongst research groups investigating this VBNC of microbial species present in the water environment exist state [3, 6, 7]. Researchers have exploited use of metabolic in a number of physiological states that possess different staining to reveal an underestimation in the level of microbial requirements for survival and to sustain growth. Indeed, survival compared to similar samples cultured on traditional the number of waterborne bacteria in which the VBNC agar plates. Recent research in our laboratory has also shown state has been reported has greatly increased, particularly that subpopulations of waterborne pathogens such as E. in recent times that reflect technological advances. For coli and Pseudomonas spp. treated with novel pulsed-power instance, Campylobacter jejuni hasbeenreportedtoexist disinfection technologies (such as use of pulsed-plasma gas- in two different cellular morphotypes, where the atypical discharge technology that will be expanded on later in this coccus-form (currently associated with nongrowing VBNC paper) were capable of reducing the redox dye 5-cyano-2, state) occurs in water under extended nutrient depletion 3-ditolyl tetrazolium chloride (CTC) that is an indicator conditions [19]. A number of different research groups have of electron acceptor function, yet similarly treated samples reported that these atypical culture forms are still capable were unable to form colonies on a variety of laboratory- of infection mice and poultry [20]. Moreover, Rowan and based culture media [6, 7]. This corroborates more recent coworkers [21] recently reported on that different culture research undertaken by Sawaya and coworkers [3]whoused morphotypes of L. monocytogenes generated in water after a combined 4’6-diamidion-2-phenylindole (DAPI) and CTC exposure to novel pulsed-plasma gas-discharge treatment stains to highlight the occurrence of physiologically active can survive internalization by human polymorphonuclear bacteria in river and wastewater treatment plants that were leukocytes. While Lindback¨ et al. [9] reported that the ability much higher than those obtained by plate counting. These to enter into an avirulent VBNC form is widespread among researchers also reported that microscopic viable bacteria International Journal of Microbiology 3 were more chlorine resistant than culturable bacteria. That been determined by amplification and sequence analysis of said, a demonstration of active respiration does not neces- prokaryotic RNA genes with universal prokaryotic primers sarily infer that these stressed bacteria are capable of future and authenticity has been verified by use of in situ probes. growth. As numerous researchers continue to report on Such nonculturable organisms may only be detected by use the use of redox stains for highlighting differences in agar- of such molecular techniques based on probes such as 16S plate counts, it is important that we holistically explore and and 23S rRNAs or on determination of mRNA, either by identify specific microbial cues at the cellular level which quantitative real-time PCR and/or by fluorescent techniques govern the transition to the VBNC state along with exploiting such as in situ hybridization (FISH), microradiography, commensurate advances in media formulations that are epifluorescence microscopy, and flow cytometry [2, 11, tailored for optimal resuscitation of these sublethally stressed 12]. While Fiksdal and Tryland [13]advocateduseof cells (cited in [10, 24]). The latter authors showed that the rapid enzyme assays for monitoring of water quality, which addition of a commercially available antioxidant Oxyrase and may also detect organisms in the injured or VBNC state. a heat-stable autoinducer of growth secreted by enterobac- Garcia-Armisen and Servais [11] showed that the ratio of terial species in response to norepinephrine, resuscitated E. direct viable- (DVC-) FISH count and the culturable count coli,andSalmonella enteric serovar Typhimurium that were increased with decreasing abundance of culturable E. coli stressed by prolonged incubation in water microcosms. in river water, and therefore the slope of the linear-log- log correlation of DVC-FISH versus colony forming unit numbers was less than one. The authors hypothesized that 1.5. VBNC State and Cell Suicide Phenomenon. Advancing the more stressful conditions, such as nutrient deprivation the earlier pioneering work of Dodd et al. [25]and and increase solar stress at low turbidities met in low Aldsworth et al. [26], who previously postulated that self- contaminated environments, were responsible for the larger destruction or “cell suicide” may be attributed to sublethally fraction of VBNC E.coli.Asmentionedpreviouslybymany stressed or damage microbes being incapable of coping research groups [13, 29–33], all field trials plotting log β- with oxidative burst when rapidly growing on nutrient rich D-galactosidase (GALase) activity or log β-D-glucuronidase media, it is important that we also exploit advances in (GLUase) activity versus log culturable target bacteria in toxicology (such as methods exploring cellular apoptosis fresh or marine water have shown regression straight lines and necrosis, cell membrane lipid peroxidation assays, and with a slope less than one, suggesting that enzyme activity nuclear chromatin/comet assays) that will provide valuable calculated per culturable indicator bacteria increases when insights into the possible role of intracellular free radicals in their numbers decrease (e.g., when sewage effluent becomes combination with the direct physical action of the applied diluted in receiving waters). Pure culture studies of E. coli environmental stress on the generation and persistence of have also shown that after exposure to other types of VBNC organisms in water. Indeed, Servais and coworkers extrinsic stress such as chlorination, the GALase activity [12] recently reported that certain environmental factors is less reduced than the direct viable count [29]. While such as nutrient scarcity and solar irradiation lead to a high Zimmerman and coworkers [34] demonstrated that E. coli proportion of VBNC E. coli in freshwater. The latter fecal can be present in higher numbers in recreational water microorganisms are brought into freshwater environments samples using fluorescent antibody direct viable counting mainly through wastewater release, surface runoff, and soil that what are detected with standard culture methods. The leaching. Interestingly, modified guidelines for bathing water latter advances the early landmark study of Bjergbaek and states that enumeration of E. coli will replace total coliforms Roslev [14] who reported on the occurrence and persistence and fecal (also called thermotolerant) coliforms as bacterial of VBNC E. coli in nondisinfected drinking water using indicators of water quality [27]. The latter authors articulated different cultivation dependent methods, fluorescence in situ that the number of E. coli in freshwater is systematically hybridization (FISH) using specific olignucleotide probes, underestimated by traditional culture-based methods such direct viable counts (DVC), and by enumeration of GFP- as multiple tube fermentation and or membrane filtration tagged E. coli (green fluorescent protein, GFP). These studies techniques), which is cause for concern from a public health specifically reemphasises the need for a rapid, accurate, and perspective. On a related theme, numerous researchers precise method for detecting health risks to humans from have also recently reported on the occurrence of “bacterial contaminated water. autophagy” (i.e., microbial adaptation to autophagic micro- Other likely candidate methods for efficient and rapid- bicidal host immune cell defences), which is an important detection of VBNC bacteria in the aquatic and soil envi- cell survival process initiated during nitrogen starvation ronments include RNA-based genotypic approaches. Dunaev conditions [28]. et al. [2] recently reported on the rapid and accurate quantification of VBNC pathogens in biosolids via mon- 1.6. Molecular-Based Techniques for Rapid Detection and itoring and quantifying stress-related genes in Salmonella Classification of VBNC Organisms. Among the minority of spp. using cDNA microarrays combined with quantitative bacteria that have been discovered it is estimated that more reverse transcription polymerase chain reaction (qRT-PCR). than90%areasyetnonculturableasattestedbythefact Quantification of mRNA was correlated to cell viability and that International Committee on Systematic Bacteriology has their ability to grow. Okabe and Shimazu [35] also describe recognised a new category for nonculturable bacteria that it detection of host-specific Bacteroides-Prevotella 16S rRNA named “Candidatus” for which phylogenetic relatedness has genetic markers (total, human, cow- and pig-specific) as 4 International Journal of Microbiology promising alternative indicators for identifying the sources commonly shared cellular mechanisms (and associated gene of fecal pollution in environmental water because of their expression regulators and gene markers) that govern cellular abundance in the feces of warm-blooded animals. The conversion to this VBNC state. Moreover, there is a dearth authors clearly state that detection of aforementioned genetic of knowledge regarding specific underlying molecular and markers mainly reflected the presence of VBNC Bacteroides- associated cellular mechanisms governing transition and Prevotella cells in water, suggesting that seasonal and geo- persistence of waterborne microorganisms in this VBNC graphical variations in persistence of these host-specific state, in addition to obviously establishing what specific Bacteroiides-Prevotella 16S rRNA genetic markers must be environmental conditions or triggers cause these changes considered if used as alternative fecal indicators in envi- in culturable state. Greater studies are also required to ronmental waters. This corroborates emerging studies that investigate the presence and role of this VBNC state in suggest that increases in coliform concentration after STP microbial pathogens that cause disease in the aquatic natural and dewatering processes may be attributed to cells going setting such as in fish.Taking for example, Flavobacterium into VBNC state implying traditional coliform enumeration psychrophilum, the causative agent of rainbow trout fly methods are not sufficient to determine number of viable syndrome and cold water disease in salmonids, where cells. This also reinforces common viewpoint shared by many subpopulations were still culturable after starvation for 300 scientists that a variety of problematic bacteria can enter days in sterilised fresh water. The virulence of starved F. VBNC states as a survival response when exposed to dele- psychrophilum was maintained for at least 7 days after terious environmental stresses such as nutrient starvation, the transfer of the bacterial cells to fresh water [37]. The osmotic stress, and so forth. However, in order to gain VNBC state was earlier reported for this fish pathogen a greater appreciation of environmental and public health by Madetoja and Wiklund [38]whorevealeddifferences risks associated with persistence of microbial pathogens in between enumeration using advanced immunoflourescence different culturable states, it is imperative that we acquire and genetic probes (i.e., nested PCR) compared to that an understanding of the interrelated molecular responses of using traditional agar plate cultivation. While the long- governing tolerance to these applied sub-lethal stressors. term survival cellular responses to salinity, temperature and Such as detailed investigations reported by Brackman et al. starvation have been studies in the eel pathogen Vibrio [15] who recently found that autoinducer- (AI-) 2 quorum- vulnificus for over a decade [39].Tocombatsuchsevere sensing inhibitors affected starvation response and reduce microbial infections affecting fish, many researchers have virulence in several Vibrio species, most likely by interfering advocated water recirculation and good management as the signal transduction pathway at the level of LuxPQ. potential methods to avoid disease outbreaks particularly with F. psychrophilum [40]. Thus, highlighting the impor- tance of augmenting water quality through improvements 1.7. Risk Assessment and Predictive Modeling for Management in wastewater treatment, which should be augmented in of VBNC State Organisms in Water. The aforementioned efficiency and capacity to cater for under-appreciated and for detection methods are not as yet commonly used for routine emerging microbial threats. measurement as they either are not simple enough and/or the equipment is too expensive. Therefore, in terms of enabling 1.9. Addressing Scientific Shortfalls in VBNC to Inform Policy. effective risk-based assessment and environmental manage- To comprehensively address such issues, detailed analysis of ment to address the occurrence and persistence of VBNC the proteome that arise during this morphological transition pathogens in water, critical data must be acquired to convert will enable identification and characterization of key proteins VBNC-phase potential pathogens from unknown to known that are responsible for this VBNC state. In addition, and defined hazards. In terms of satisfying these information tandem use of microarray gene analysis along with real-time gaps, accurate enumeration of total microbial load and real quantitative PCR will help unravel specific gene functions time identification to the various types present in water are and will pin-point over-arching regulatory framework(s). pivotal to hazard identification and characterization, which The aforementioned studies should also be carried out in will contribute significantly to assuring water safety. Regard- parallel with developing improved protocols for resuscitating ing the latter, Smeets et al. [36] recently described improved VBNC organisms, particularly availing of advances made in methods for modelling drinking water safety using a robust fermentation technology such as exploiting chemostat-based quantitative microbial risk assessment (QMRA), where a bioreactor approaches that can simulate and monitor mul- case study monitored Campylobacter data for rapid sand tiple environmental stresses (either applied simultaneously filtration and ozonation processes. This study showed that or sequentially) over extended time periods. Indeed, Kooi currently applied methods do not predict monitored data and coworkers [41] recently exploited use of a chemostat used for validation. Therefore, underestimation in the levels to report on the dynamic behaviour of simple aquatic of problematic microbes in water, and failure to identify the ecosystems with emphasis on nutrient recycling in order presence of pathogenic organisms in representative water to monitor toxicants. Thus, use of the latter provides a samples also pose significant threats to public health. useful vehicle for investigating the effects of deleterious microbial stressors on the structure and functioning of 1.8. Use of Advanced Detection Techniques to Investigate ecosystems. Other related studies that merit attention include Virulence Potential in VBNC State Organisms. Greater infor- the development of more rapid, user-friendly, field-based mation is also required on elucidating the existence of technologies that will reliably and repeatably detect and International Journal of Microbiology 5 quantify various waterborne pathogens that may persist use of conventional decontamination methods failed to in different culture states. Less than four hours was rec- eliminate this enteroparasite in treated water (cited in Garvey ommended as “rapid” by Noble and Weisberg [42]ina et al. [52]). This situation has become a major concern for recent review of rapid detection methods for bacteria in water authorities and consequently, significant modifications recreational waters. Due to plethora of different detection to drinking water regulations have been proposed for the methods currently being developed combined with obvious detection and surveillance of this protozoan in the US and interlaboratory variability in terms of associated operating in other developed countries [50]. protocols, it is imperative that we develop and agree upon a standardized battery of reliable tools for detecting and 2.2. Alternative or Complementary Decontamination Methods. quantifying culturable and nonculturable bacteria so as Development of alternative methods of Cryptosporidium to establish future unified, quantitative-risk management disinfection for water applications (such as ozone and/or protocols for monitoring our aquatic environments. The UV) has been hindered by the uncertainty surrounding latter will greatly facilitate our endeavours to collectively efficacy of using in vitro surrogate viability assays due to comply with emerging environmental policies such as EU’s their overestimation of oocysts survivors posttreatments Water Framework Directive. and the lack of critical data on the preferred use of in vitro cell culture and/or in vivo animal-based infectivity 2. The Waterborne Enteroparasite: assays to determine interrelated factors governing repeatable Cryptosporidium disinfection of oocysts suspended in water [53]. Although recent studies that utilised at least 20 different cell lines 2.1. Background. Cryptosporidium species have emerged over have advocated the preferential use of the human ileoce- the past decades as major waterborne pathogens causing gas- cal adenocarcinoma HCT-8 cell line as an equivalent in troenteritis in humans [43]. The occurrence of the environ- vitro method to that of using the “gold standard” mouse mentally resistant thick-walled oocyst stage of this organism assay for measuring infectivity of Cryptosporidium [44], has become a worldwide concern due to its resistance to there has been limited evidence to date on the combined disinfection with chlorine at concentrations typically applied use of these approaches for assessing critical operational in drinking water treatment plants (2 to 6 mg/L) [44]. Thus, parameters governing pulsed UV light (PUV) as a means the control of Cryptosporidium oocysts remains a major chal- of disinfecting water contaminated with this enteroparasite lenge for drinking water utilities due to fact that the common [53]. This study demonstrated that there is good agreement chemical disinfectants-free and combined chlorine, when between use of in vitro culture-qPCR and the SCID-mouse- used singly, are practically ineffective for inactivating this infectivity assays for evaluating the disinfection efficacy of protozoan under conditions encountered in most treatment pulsed UV for inactivating C. parvum suspended in saline, facilities [45]. Cryptosporidium spp. are found ubiquitously thus reducing the requirement to unnecessarily use animals and the transmission “oocyst-stage” may remain viable for for these particular research studies. However, the authors several months in the aquatic environment. This protozoan recommended that pressing investigations are needed to is transmitted via the faecal-oral route, where consumption comprehensively demonstrate efficacy of using this novel of contaminated drinking water and use of recreational water disinfection UV approach for treating lower concentrations are major sources of infection [46]. Cryptosporidium oocysts of infectious oocysts under dynamic conditions found in are resistant to conventional disinfectants at concentrations WWTPs and in the aquatic environment. Development of a and exposure times commonly used, and their infectious reliable and repeatable method of measuring fluence values doses in humans have been estimated to be as low as 30 from UV-delivered pulses for water disinfection applications oocysts [47]. Indeed, more than 160 waterborne outbreaks is also required, and should take on board critical interrelated of cryptosporidiosis have been reported globally, with the factors identified previously by Bolton and Linden [54]for greatest documentation occurring in the US and the UK continuous low- (LP) and medium-pressure (MP) UV units. [46, 48]. Development of PUV has recently received attention as This situation has become a major concern for water a potentially novel strategy for decontaminating water as it authorities and consequently, significant modifications to offers many benefits including rapid microbial reductions drinking water regulations have been proposed for the and efficiency of energy usage due to underpinning high detection and surveillance of this protozoan in the US and in peak-power dissipation during treatments [55, 56]. Indeed, other developed countries [49]. The new European Drink- use of ultraviolet (UV) light have become widely accepted ing Water Directive advocates that all the state members as alternative methods to chlorination for wastewater should provide drinking water supplies with absence of disinfection [57, 58]. There are also over 2,000 wastewater pathogenic organisms [50]. However problems associated treatment plants worldwide using either LP or MP ultra- with the determination of oocysts viability/infectivity make violet technologies. Recent studies investigating continuous- the establishment of maximum acceptable concentrations use UV lamp technology has demonstrated the effectiveness very difficult, and concentrations of ≥3–30 oocysts/100 of UV in inactivating pathogens in wastewater [57]. Research litres in treated water have been proposed as action levels has also shown that, to ensure permanent inactivation and [51]. Recent contamination of drinking water supplies in prevent the recovery of microorganisms following exposure the west of Ireland have led to a significant number of to UV, a broad polychromatic spectrum of UV wavelengths confirmed cases of cryptosporidiosis, intimating that the is necessary such as doses delivered by MP and PUV 6 International Journal of Microbiology systems. These wavelengths inflict irreparable damage not there are very limited published findings to date that only on cellular DNA, but on other molecules such as holistically investigate critical factors governing the effective enzymes as well. Moreover, numerous studies have also and repeatable destruction of this recalcitrant enteroparasite highlighted limitations of decontamination techniques in drinking water supplies using ozone or other oxidative such as conventional low pressure mercury lamps designed agents. Of those researchers that have reported on this to produce energy at 254 nm (called monochromatic or complex oxidation process, it would appear that using ozone germicidal light) that include microbial repair and the singly or combined with (or without) free chlorine or necessity for lengthy durations of exposure to obtain suitable monochloramine (in addition to other synergistic factors levels of decontamination [59]. More recently, medium- such as disinfection concentration, contact time, dissolved pressure mercury UV lamps have been used because of organic content concentration, pH, and temperature) are of their much higher germicidal UV power per unit length critical importance for the inactivation of treated C. parvum and because of their ability to emit polychromatic light [72, 73]. Despite limited understanding of the dynamic comprising germicidal wavelengths from 200 to 300 nm) complexities underpinning advanced oxidation processes, [54]. This approach kills microorganisms by using ultrashort ozonation is often preferred to chlorination because former duration pulses of an intense broadband emission spectrum leads to smaller concentrations of potentially harmful halo- that is rich in UV-C germicidal light (200 to 280 nm band). genated disinfection by-products. Yargeau and Leclair [74] PUV is produced using techniques that multiply power also recently reported that use of ozone appears to be a manifold by storing electricity in a capacitor over relatively promising technique for degradation of antibiotics, even in long times (fractions of a second) and releasing it in a wastewater. After 4.5 min of ozone treatment, the concen- short time (millionths or thousandths of a second) using tration of sulphamethoxazole was below the HPLC detection sophisticated pulse compression techniques [56, 60, 61]. The limit of 0.6 mgL−1 indicating a degradation efficiency higher emitted flash has a high peak power and usually consists than 99.24%. Intense research has been recently focused of wavelengths from 200 to 1100 nm broad spectrum light on the development of novel pulsed plasma gas discharge enriched with shorter germicidal wavelengths [62, 63]. This technology as a complementary means of treating water technology has received several names in the scientific containing unwanted microbial pathogens and chemicals. literature: pulsed UV light [64–66], high intensity broad- Pulsed plasma gas-discharge (PPGD) technology in- spectrum pulsed light [67], pulsed light [60], intense pulsed volves applying high voltage pulses to gas-injected test liquids light [56], and pulsed white light [68]. Despite the fact resulting in the formation of a plasma that causes free that UV light appears effective for inactivating waterborne radicals such as dissolved ozone and hydrogen peroxide, free enteropathogens, including Cryptosporidium spp. it is recog- electrons, ultraviolet light (UV), acoustic shock waves, and nised that many organisms have mechanisms for repair- electric fields at levels between 10–50 kV/cm to be generated ing light-induced DNA damage. However, Rochelle et al. in the test liquids (Figure 1). Rowan and coworkers [75]has [69] recently reported that Cryptosporidium spp. oocysts been shown that application of PPGD successfully reduces could neither repair UV light-induced damage nor regain unwanted Campylobacter and Salmonella spp. in poultry infectivity under standard conditions used for storage and wash water. PPGD (akin to PUV) is an enabling technology distribution of treated drinking water. This is despite the fact that requires transient generation of high voltage and high that both C. parvum and C. hominis were shown to harbour current that in turn results in the generation of large peak genes encoding UV repair proteins for mechanisms such powers ranging from Megawatts to Terrawatts. Depending as nucleotide excision repair and photolyase enzymes [70]. upon the application (PPGD or PUV), a pulse generator Indeed, irradiated oocysts were unable to regain preirradia- will deliver a large energy level on a single shot basis or tion levels of infectivity, following exposure to broad array alternatively will deliver a modest amount of energy (1–10 J) of potential repair conditions, such as prolonged incubation, at a repetition rate from 10 to 10,000 pulses per second. preinfection excystation triggers, and post-UV holding peri- Thus PPGD offers a radical new approach to energy delivery ods. Otaki et al. [71] reported that adaptive microbial sur- that involves the use of repetitive switching techniques vival (tailing phenomenon) occurs when samples are treated to deliver stored energy in intense ultrashort bursts (85– in high turbidity solutions using continuous UV sources 100 nanoseconds). During each pulse, very high levels of whereas tailing did not occur when similar samples were peak power are generated (10–20 MW), and treatment is treated with pulsed xenon lamp. In terms of potential future achieved using the required number of pulses, which is alleviation strategies for combating C. parvum in water, it is favourable in terms of requirements for new energy efficient likely the combined use MP and PUV technologies would technologies for the fast-approaching post peak oil era. Other offer considerable advantages as the next-generation decon- research groups have also reported on the successful use of tamination bolt-on approach including rapid processing and pulsed high voltage for decontaminating microbial popula- efficiency of oocyst destruction, and should accommodate tions [76]andphenol[77] in liquid solutions. Interestingly, dynamic operational requirements at WWTP level. research in our laboratory has shown that the myriad of Use of ozone is gaining in popularity as an alternative biocidal properties generated by use of PPGD technology or complementary approach for disinfection in drinking has also demonstrated promising results for the removal water facilities worldwide [58]. Some landmark studies of prions on surgical material intimating potentially other have recently reported on the possible efficacy of using healthcare applications for this decontamination approach ozone for destroying Cryptosporidium oocysts [45]. However, [78]. International Journal of Microbiology 7

[81]. Antibiotics in the broader sense are chemotherapeutic agents that inhibit or abolish the growth of microorganisms, and have been used extensively in human and veterinary medicine as well as aquaculture for the purpose of prevent- ing or treating microbial infections. Wise [82]estimated antibiotic consumption worldwide to lie between 100,000 and 200,000 ton per annum. The reader is also directed to the recently published landmark reviews of Kummerer¨ [83, 84] for further detailed information on the possible input, occurrence, fate, and effects of antibiotics in the aquatic environment. The concentrations of antibiotics in municipal sewage and in sewage treatment plants are typically lower by a factor of 100 compared to hospital effluent [85]. While bacterial resistance to antibiotics have also been found in soil [86]. Some studies revealed that many different types Figure 1: Pulsed-plasma gasdischarge decontamination treatment of antibiotics are biodegradable under aerobic or anaerobic of water. conditions, while other related research has reported that certain aquatic microbes can actually use these free active compounds as a sole carbon source. However, research has also shown that the concentrations and activity spectrum 2.3. Potential Hurdles for New Technologies. However, a of compounds found in the environment does not correlate nonoptimized pulsed plasma-gas discharge process may with the presence of resistant bacteria isolated from the same also lead to the formation of undesirable by-products environments. Concern over the development of secondary such as brominated organic compounds and halogenates resistance to commonly-used antibiotics (either through in treated water [79]. Brominated organic compounds are vertical and/or horizontal gene transfer among related and also considered potentially harmful and remain the subject unrelated bacteria) in the environment with potential for of international study in order to elucidate mechanistic adversely affecting aquatic and terrestrial organisms along and kinetic information regarding their formation during with the potential for a cyclic unwanted return to humans ozonation so as to identify effective strategies for their again via consumption of drinking water has stimulated reduction or elimination [80]. Considerable attention should numerous research groups to investigate these important be given to reducing or eliminating the occurrence of issues. Kummerer¨ [83] clearly articulated that the issue of harmful by-products of ozone during this project via acquired resistance was nearly always addressed in pub- adjustments in sparged-gas composition, decreasing pH and lications by describing the presence of antibiotics in the complementary use of pulsed UV technology. Surprisingly, environment, with often mere speculation as to the possible despite increased interest in the development of nonthermal relationship between low concentrations of antibiotics in advanced oxidative processes (such as corona plasma dis- water and emergence of resistance. In truth, the latter is a charges, ozone combined with H O , low/medium pressure 2 2 highly complex process that is as yet not fully understood UV combined with H O , etc.), this author has been unable 2 2 [87]. However, the trend of accumulating and accelerating to source any published reports on possible toxicological resistance to antibiotics is of concern as this is also juxtaposed issues associated with use of these new technologies for by the marked reduction in mankinds’ current arsenal of disinfection applications in healthcare, agri-food or the effective tools for combating this phenomenon. environment. On a related point, our research group has From a review of best published literature it would recently reported on the occurrence of VBNC state for appear that possibly the most significant undefined risk food and water-borne microorganisms generated in liquid in terms of antibiotics are microorganisms harbouring suspensions after separate exposure to PPGD and pulse multiple antibiotic resistance genes such as vancomycin- electric field technologies [6, 7]. resistant enterococci (VRE), methicillin-resistant Staphylo- coccus aureus (MRSA), and multiresistant pseudomonads 3. Antibiotics in the Aquatic Environments that are living in close proximity to each other (such as in biofilms—sewage sludge flocs) as opposed to the free active Although antibiotics have been used for decades, only compounds present at low concentrations in the aquatic recently has an increasing number of studies highlighted environment. Indeed, Davison [88] provided evidence that the lack of understanding and knowledge about persistence antibiotic resistance is already present in natural environ- of antibiotics in the aquatic environment and the potential ments and that it can be exchanged between bacteria for environmental risks that this may pose [81]. Particularly as at least a decade. Therefore, under intense discussion is antibiotics are often poorly degraded or removed in conven- the possibility that nutrient-poor, oxygen-limited and cold tional wastewater treatment plants (WWTPs) which causes aquatic environments exemplified by sewage sludge may be formation of toxic degradation products that may impact selecting for and acting as a reservoir for slowing growing negatively on the aquatic environment and public health resistant organisms. This environment provides for a higher 8 International Journal of Microbiology biodiversity of microbial types and numbers affording a selective pressure for that resistance. That said, recent greater probability for accelerated exchange of antimicrobial findings also suggest that bacteria which have already have drug resistance and virulence determinants in the presence become resistant through the application of antibiotics of other neighbouring microbes deficient in such deleterious will not necessarily have a selective advantage in sewage properties. Indeed, Ohlsen et al. [89] revealed that antibiotics treatment [92, 93]. Interestingly Kummerer¨ [83]reported even at sub-inhibitory concentrations can have an impact on that resistance was found to be high in hospital effluents and cell function and can stimulate expression of quiescent viru- in sewage treatment plants, yet hospital effluents contribute lence factors or potentiate transfer to an antibiotic resistant to less than 1% of the total amount of municipal waste state. The latter is strongly aligned with findings from unre- suggesting that hospitals are not the main source of resistant lated environmental stress studies which previously showed bacteria in municipal sewage. The latter would also suggest that many food and waterborne microbial species can sense that antibiotic usage in the community accounts for the and adapt to related and unrelated sub-lethal environmental main input of resistant bacteria into municipal sewage. stresses (such as osmotic stress or starvation) through com- The question should also be posed as to whether or not plex quorum sensing and respond through an orchestrated multi-drug resistant bacteria that transcend to the quiescent controlled subsequence of preferential gene expression [90]. VBNC state after periods of extended exposure to a nutrient Thus, not alone may the latter confer microbial tolerance to depleted stressful environment (such as in water or in soil) the applied stress through a process of elevating activities in are still capable of gene transfer and whether or not these important genes that regulate key house-keeping functions important molecular determinants remain unaltered and such as osmotic stress response and protection vital proteins stable. Kummerer¨ [84] also reported that the concentration (chaperone), but such environmental tempering may also of antibiotics may be much higher if the active compounds act as stimuli for up-regulating microbial virulence factor are persistent and accumulate, for example, by sorption to expression. The classic example being L. monocytogenes, solid surfaces such sewage sludge, sediments, or soil. where its’ transcriptional activator PrfA up-regulates viru- While most antibiotics tested to date have not been lence factor expression in the presence of a known stress biodegradable under aerobic conditions(Kummerer¨ (e.g., body temperature), yet down-regulates transcription [83, 94, 95]), biodegradability has been poor for most of of pathogenic determinants in the presence of soil-related the compounds investigated in laboratory tests, including carbohydrates. Interestingly, Zupan and Raspor [91] recently some to the β-lactams [96]. Out of 16 antibiotics tested, only reported on the development of an invasion-agar assay for benzyl penicillin (penicillin G) was completely mineralized determining in vitro invasiveness of Saccharomyces cerevisiae, in a combination test [97]. No evidence of biodegradation and showed that this yeast augmented virulence when for tetracycline was observed during a biodegradability exposed to temperatures typical of human fever (37 to test, and sorption was found to be the principal removal 39◦C) yet exhibited strong repression effect on invasion in mechanisms for tetracycline in activated sludge [98]. the presence of salts, anoxia and some preservatives. The Substances extensively applied in fish farming had long aforementioned also highlights the extreme versatility of half-lives in soil and sediment as reported in several investi- microbial species present in complex communities (such as gations [99]. However, some substances were at least partly in biofilms in WWTPs) to rapidly change and adapt when degradable [100]. Maki et al. [101] found that ampicillin, confronted with a sustained external selective pressure. deoxycycline, oxytetracycline, and thiamphenicol were Kummerer¨ [83] stated that there is a dearth of infor- significantly degraded, while josamycin remained at initial mation on the interrelated factors governing the fate and levels. Despite the aforegoing, it has yet to be established effects of antibiotics in the environment (i.e., microbial that permanent exposure of antibiotics in sewage treatment ecology), highlighting that the majority of published studies systems promotes the development of antibiotic resistance to date are limited to single compound approaches. He also and selective effects on bacterial communities. recently postulated that the effects of antibiotics in WWTPs Studies have also revealed that bacteria that are resistant or in the aquatic environment may be under-estimated, to antibiotics are present in surface water. Furthermore, particularly as there is evidence that microbial exposure to antimicrobial resistance has also been found in marine antibiotics from the same group or from different groups bacteria [102] and bacteria living in estuaries or coastal may result in additive effects. For β-lactams it has been waters polluted with sewage water [103]. Other researchers shown that their potency is much higher in the presence of have reported that antibiotic resistance genes (ARGs) can be 5-fluorouracil, a cytotoxic compound also present in sewage found in a region where no selection pressure exists [104]. in concentrations at the mg or μgL−1 range. Use of biocides Whereas Pruden et al. [105] noted that tetracycline ARGs such as triclosan and quaternary ammonium compounds tet(W) and tet(o) were present in treated drinking water used in hospitals and homes may also select for antibiotic and recycled wastewater, suggesting that these are potential resistance in microbial pathogens. Additionally, there is no pathways for the spread of ARGs to and from humans. published information as yet on the potential impact of low High loads of antibiotics in sediments at concentrations levels of other pharmaceutically active compounds (such as potent enough to inhibit the growth of bacteria have been endocrine disrupting chemicals) on development of antibi- reported for aquaculture [83]. While antibiotic-resistant otic resistance as the former also tolerate sewage treatment. bacteria have been detected in drinking water supplies as far There is also limited information on how long bacteria back as the 1980s. In fish farming sector such as aquaculture maintain antibiotic resistance in the absence of continued and mariculture, the widespread use of antibiotics for International Journal of Microbiology 9 treating bacterial diseases has been associated with the and have been found in treated effluent at the ng/L level development of resistance in a range of bacterial pathogens [113, 118]. Whilst intensive conventional-based treatment including Aeromonas hydrophilia, Aeromonas salmonicida, approaches have investigated many interrelated factors in the Edwardsiella tarda, Edwardsiella icttaluri, Vibrio anguillarum, design and operating conditions of WWTPs for reduction Vibrio salmonicida, Pasteurella piscida and Yersinia ruckeri or elimination of EDCs in treated wastewater, none have [106]. There are also considerable gaps in current knowledge reported on their effective removal. Moreover, due to the concerning the possible transfer of chemical contaminants common practice of residual chlorine in drinking water and microbial pathogens (including those harbouring ARGs) distribution, halongated by-products such as haloacetic acids into the food chain through land spreading of some treated and trihalomethanes form that exhibit a range of potential organic municipal and industrial material on agricultural carcinogenic potentials (cited in [119]). The European land used for food production. The reader is directed to commission (COM [1999] 706) acknowledged that there is the comprehensive report produced recently by the Food an urgent need for further scientific research to denature Safety Authority of Ireland that addresses critical issues deconjugated EDCs in the environment. Therefore, it is underpinning these putative concerns [107]. essential that such compounds be efficiently and effectively Other Pharmaceutically Active chemicals. A marked removedfromprocessedwaterdischargedfromWWTPs.On observation from this paper is the pressing need for greater a similar theme, Yamamoto and coworkers [120] recently risk-based assessment and for new and/or improved allevia- reported the persistence of 8 pharmaceuticals with rela- tion strategies for the optimal decontamination of wastewa- tively high ecological risk and high consumption (namely ter at treatment plant level so as to reduce or eliminate the acetaminophen, atenolol, carbamazepine, ibuprofen, ifen- microbial (and their associated metabolite) risks to public prodil, indomethacin,metenamic acid, and propranolol) health. In the context of improving wastewater treatment and using river water. planning for safe drinking supplies, one must also acknowl- Previous researchers have recently shown that use of edge growing international concerns about the release of similar non-thermal corona discharge processes such as the certain chemicals into the aquatic environment that may PPGD technology may be potentially effective at destroy- result in alterations in the reproductive health of humans ing structurally-related organic compounds such as dyes, and wildlife [108–110]. The environmental presence of such phenol and aniline in aqueous solutions [121–125]. Other man-made and naturally occurring compounds, properly nonplasma related studies have shown that the use of UVA referred to as endocrine disrupting chemicals (EDCs), can can also facilitate removal of EDCs including E1,E2,andEE2 mimic or interfere with the binding and action of natural from water by photolysis [123, 126]. Given that the majority hormones, thus disrupting normal physiological processes of known EDCs are phenolic in chemical structure, the ben- [111, 112]. Consumption of water contaminated with EDCs zene rings underpinning these EDCs should be rapidly pho- may also cause reproductive disorders in humans as such todegraded by hydroxyl radical attack during PPGD treat- chemicals have the ability to mimic the function of natural ments. It is worth noting that the use of ozone become widely estrogens as well as disrupting the synthesis and metabolism accepted as alternative methods to chlorination for wastew- of hormones by binding to hormonal receptors (cited in ater disinfection [57, 58], and ozone has traditionally been [108, 110]). Thyroid system-disrupting activity in water applied in drinking water treatment plants for disinfection from municipal domestic sewage treatment plants was also [58]. While other researchers have demonstrated the produc- detected recently [112]. The list of known EDCs is extensive tion of short-lived, high-oxidative, plasmochemical elements and includes natural and synthetic steroid hormones (such in pulsed-plasma-treated test liquids such as water and as 17 β-estradiol(E2), estrone (E1), and 17 α-ethinylestradiol effluent for bacterial decontamination, no study to date has (EE2)), phytoextrogens, pesticides, pharmaceuticals, and reported on the use of corona discharges singly or combined surfactants, all of which have been detected worldwide in with pulsed UV light as a novel approach for the destruction processed water from domestic treatment plants (WWTPs) of established and emerging microbial threats to public at the ng/L level that may cause abnormalities to aquatic health in water or effluents. These dynamic environmental organisms [113, 114]. EDCs were recently detected in fate studies must be carried out in tandem with experiments effluents from WWTPs and from receiving waters at levels that ascertain the likelihood of by-product toxicant forma- exceeding ng/L in Ireland [115]. tion and unwanted persistence in water, and address these Recent studies conducted by Fogarty and McGee [115] associated undefined risks by either adjusting existing and/or have demonstrated that fish habituating downstream of implementing new complementary alleviation strategies. waste-water treatment plants (WWTPs) in the Midlands region in Ireland exhibited delayed spermatogenesis com- pared with fish upstream and intersex (feminization) was 4. Conclusion and Outlooks discovered in roach [115–117]. The scientific community has particularly focused on estrogenic EDCs (i.e., compounds There is a pressing need to generate critical data in gaps interacting with the human estrogen receptor α), which enter highlighted above via provision of funding to establish the environment from a variety of sources including effluent consortia of meaningful stakeholders comprising scien- discharge pipes, agricultural runoff and landfills [117]. In tists, engineers, and end-users in order to facilitate policy particular, such EDCs are considered to be dominant con- makers and implementers (suchas water managers) strive tributors to estrogenic activity in treated water from WWTPs towards meeting ambitious objectives set for the EU’s Water 10 International Journal of Microbiology

Framework Directive that requires member states to attain the predicted environmental concentration of the active a good ecological status for all water bodies by 2015 ingredient (AI) and its predicted no-effect concentration. [127]. While a global response to emerging environmental Hazard is expressed in terms of AI’s persistence, potential problems has been positive with the provision of substantial for bioaccumulation, and ecotoxicity [131]. The combined research funding been made available to the broad and use of monitoring data aligned with development and diverse researcher communities (such as EU FP7 initiatives), application of dynamic pollutant (contaminant) fate models one obvious real challenge that still remains is how does is recommended. Pharmaceutical producers should also mankind harness and channel vast amounts of relevant highlight environmental precaution when designing new sophisticated data that is produced and disseminated from more environmental friendly AIs, and that the environmen- a multiplicity of research groups worldwide into meaningful tal data should be transparent to the general public. In terms streamlined forums that can be shared and understood by of holistic monitoring and prediction, we must also factor all in a timely fashion. Exploiting advances in information in seasonal environmental changes such as atmospheric technology will assist greatly with this challenge and will and oceanic processes that occur in response to increasing drive effective risk-based assessment, evaluation, and man- greenhouse gases to map disease potentiation dynamics as agement of present and emerging problems for the aquatic this will aid development of appropriate strategies for con- environment. trolling microbial risks across a range of human and natural Currently, there is insufficient information available to systems. Indeed, Sedas [132] recently showed that climate reach definitive conclusion on the significance and impact also influences the abundance and ecology of pathogens, and of the VBNC state organisms and antibiotic resistant the links between pathogens and changing ocean conditions, bacteria in the environment which would allow for the including human disease such as cholera. It is recognised that assessment of the potential risk related, for instance, to environmental risk-based management is typically uncertain human health and ecosystem functions. Indeed impact due to different perceptions of the risk problem and our of antibiotics present in the aquatic environment on the limited knowledge about the inter-play of biological, physical frequency of resistance transfer is questionable, with greater and chemical processes underlying these risks [133]. While a concern placed on the input of resistant bacteria into plethora of real-time quantitative microbial detection tools the environment from different sources. This suggests that combined with more rapid and efficient decontamination greater emphasis should be placed on reducing the likelihood approaches are on the horizon, we must be equally prudent of antibiotic resistance occurrence in the first instance at about exhaustively confirming their efficacy such as agreed the point of source by prudent management and careful standards for sensitivity and reliability for detection, and eco- rotation of antibiotic usage in both the hospital and com- friendliness for decontamination. munity settings and by systematic continued monitoring It is clear and apparent that the emergence of such of resistance, which will collectively impact positively on complex problems has heralded a new dawn in research public health and wellbeing. The lack of understanding and innovation (i.e., a marked departure from the solitary and knowledge surrounding a broad range of established PhD student environment), where collection, analysis and and emerging risks to the aquatic environment is apparent. timely dissemination of such vast and meaningful infor- For example, Mena and Gerba [128] recently reviewed best mation can only be effectively addressed using a well- published data for risk assessing the opportunist pathogen managed consortia of networked researchers from various P. aeruginosa in water and concluded that process of complementary disciplines by way of using a plethora of estimating risk is currently significantly constrained because appropriate frontier funding initiatives such those offered of the absence of specific (quantitative) occurrence data for by the EU (http://cordis.europa.eu/fp7/home.html). While Pseudomonads. it is recognised that undertaking such far-reaching cross- In order to gain a greater appreciation and understanding boundary initiatives will enable the potential impact, impli- of the critical environmental fate and associated impact of cations and future proofing of established and emerging such undefined and variable microbial risks, more controlled risks to be managed and catered for properly. Is must laboratory investigations are needed to be undertaken in be equally recognised that effectively managing and har- combination with conducting field studies. There is also nessing the potential of such diverse consortia comprising pressing need to obtain definitive data on proposed risk to academics, industrial partners (SMEs to multinationals), public health and to the aquatic environment combined with policy makers and so forth pose significant logistic and developing appropriate short- and long-term mathematical complex challenges, for example, attempting to holistically and computer models with capacity for monitoring and cater for all stakeholders in a united global society on a predicting ecotoxicological effects of these microbial stres- single theme who have different needs and goals are real sors, particularly under dynamic naturalistic settings akin challenges. Therefore, such important issues must also be to those recently described by Bontje and coworkers [129] managed with a strong over-arching foresight, particularly and Gevaert et al. [130] so as to inform policy makers in the context of embracing and exploiting advances in the and managers. It is quite apparent that proper judgement communication and information technology landscape so of the impact of microbial pathogens, their metabolites, that we can accommodate and provide for the real-time and pharmaceutically active compounds require a thorough flow of knowledge to all stakeholders in order to identify evaluation of their risk and hazard. Regarding the latter, potential synergies, emerging trends (problematic, beneficial where risk is normally expressed as the ratio between or otherwise), and/or opportunities. International Journal of Microbiology 11

Acknowledgments [13] L. Fiksdal and I. Tryland, “Application of rapid enzyme assay techniques for monitoring of microbial water quality,” The author would like to thank Dr. Dominik Kirf for his Current Opinion in Biotechnology, vol. 19, no. 3, pp. 289–294, assistance with taking images of the PPGD system, and to 2008. also acknowledge the Science Foundation Ireland (SFI) for [14] L. A. Bjergbæk and P. Roslev, “Formation of nonculturable research funding support that has contributed to findings Escherichia coli in drinking water,” Journal of Applied Micro- discussed in this paper. biology, vol. 99, no. 5, pp. 1090–1098, 2005. [15] G. Brackman, S. Celen, K. Baruah et al., “AI-2 quorum- sensing inhibitors affect the starvation response and reduce References virulence in several Vibrio species, most likely by interfering with LuxPQ,” Microbiology, vol. 155, no. 12, pp. 4114–4122, [1] J. J. Byrd, H.-S. Xu, and R. R. Colwell, “Viable but noncultur- 2009. able bacteria in drinking water,” Applied and Environmental [16] J. D. Oliver, “Formation of viable but nonculturable cells,” in Microbiology, vol. 57, no. 3, pp. 875–878, 1991. Starvation in Bacteria, S. Kjelleberg, Ed., Plenum Press, New [2] T. Dunaev, S. Alanya, and M. Duran, “Use of RNA-based York, NY, USA, 1993. genotypic approaches for quantification of viable but non- [17] V. Besnard, M. Federighi, and J. M. Cappelier, “Evidence of culturable Salmonella sp. in biosolids,” Water Science and viable but non-culturable state in Listeria monocytogenes by Technology, vol. 58, no. 9, pp. 1823–1828, 2008. direct viable count and CTC-DAPI double staining,” Food [3] K. Sawaya, N. Kaneko, K. Fukushi, and J. Yaguchi, “Behaviors Microbiology, vol. 17, no. 6, pp. 697–704, 2000. of physiologically active bacteria in water environment and chlorine disinfection,” Water Science and Technology, vol. 58, [18] D. B. Roszak and R. R. Colwell, “Metabolic activity of no. 7, pp. 1343–1348, 2008. bacterial cells enumerated by direct viable count,” Applied [4]J.M.Cappelier,V.Besnard,S.M.Roche,P.Velge,and and Environmental Microbiology, vol. 53, no. 12, pp. 2889– M. Federighi, “Avirulent viable but non culturable cells of 2893, 1987. Listeria monocytogenes need the presence of an embryo to [19] A. Reezal, B. McNeil, and J. G. Anderson, “Effect of low- berecoveredineggyolkandregainvirulenceafterrecovery,” osmolality nutrient media on growth and culturability of Veterinary Research, vol. 38, no. 4, pp. 573–583, 2007. Campylobacter species,” Applied and Environmental Microbi- [5]H.S.Xu,N.Robert,F.L.Singleton,R.W.Attwel,D.J. ology, vol. 64, no. 12, pp. 4643–4649, 1998. Grimes, and R. R. Colwell, “Survival and viability of non- [20] J. M. Cappelier, C. Magras, J. L. Jouve, and M. Federighi, culturable Escherichia coli and Vibrio cholera in the estuarine “Recovery of viable but non-culturable Campylobacter jejuni and marine environment,” Microbial Ecology, vol. 8, pp. 313– cells in two animal models,” Food Microbiology, vol. 16, no. 4, 323, 1982. pp. 375–383, 1999. [6]S.Yaqub,J.G.Anderson,S.J.MacGregor,andN.J.Rowan, [21] N. J. Rowan, D. Kirf, and P. Tomkins, “Studies on the “Use of a fluorescent viability stain to assess lethal and sub- susceptibility of different culture morphotypes of Listeria lethal injury in food-borne bacteria exposed to high-intensity monocytogenes to uptake and survival in human polymor- pulsed electric fields,” Letters in Applied Microbiology, vol. 39, phonuclear leukocytes,” FEMS Immunology and Medical no. 3, pp. 246–251, 2004. Microbiology, vol. 57, no. 2, pp. 183–192, 2009. [7] N. J. Rowan, S. Espie, J. Harrower et al., “Evidence of [22] Y. Moreno, P. Piqueres, J. L. Alonso, A. Jimenez,´ A. Gonzalez,´ lethal and sublethal injury in food-borne bacterial pathogens and M. A. Ferrus,´ “Survival and viability of Helicobacter exposed to high-intensity pulsed-plasma gas discharges,” pylori after inoculation into chlorinated drinking water,” Letters in Applied Microbiology, vol. 46, no. 1, pp. 80–86, 2008. Water Research, vol. 41, no. 15, pp. 3490–3496, 2007. [8] D. M. Rollins and R. R. Colwell, “Viable but nonculturable [23] V. G. Kastbjerg, D. S. Nielsen, N. Arneborg, and L. Gram, stage of Campylobacter jejuni and its role in survival in the “Response of Listeria monocytogenes to disinfection stress natural aquatic environment,” Applied and Environmental at the single-cell and population levels as monitored by Microbiology, vol. 52, no. 3, pp. 531–538, 1986. intracellular pH measurements and viable-cell counts,” [9] T. Lindback,M.E.Rottenberg,S.M.Roche,andL.M.¨ Applied and Environmental Microbiology, vol. 75, no. 13, pp. Rørvik, “The ability to enter into an avirulent viable but non- 4550–4556, 2009. culturable (VBNC) form is widespread among Listeria mono- [24] M. Kitajima, Y. Tohya, K. Matsubara, E. Haramoto, E. cytogenes isolates from salmon, patients and environment,” Utagawa, and H. Katayama, “Chlorine inactivation of Veterinary Research, vol. 41, no. 1, pp. 1–8, 2010. human novovirus, marin novovirus and poliovirus in [10] R. Reissbrodt, I. Rienaecker, J. M. Romanova et al., “Resus- drinking water,” Letters in Applied Microbiology, vol. 51, pp. citation of Salmonella enterica serovar typhimurium and 119–121, 2010. enterohemorrhagic Escherichia coli from the viable but non- culturable state by heat-stable enterobacterial autoinducer,” [25] C. E. R. Dodd, R. L. Sharman, S. F. Bloomfield, I. R. Booth, Applied and Environmental Microbiology, vol. 68, no. 10, pp. and G. S. A. B. Stewart, “Inimical processes: bacterial self- 4788–4794, 2002. destruction and sub-lethal injury,” Trends in Food Science [11] T. Garcia-Armisen and P. Servais, “Combining direct viable and Technology, vol. 8, no. 7, pp. 238–241, 1997. counts (DVC) and fluorescent in situ hybridization (FISH) to [26] T. G. Aldsworth, R. L. Sharman, C. E. R. Dodd, and G. enumerate viable Escherichia coli in rivers and wastewaters,” S. A. B. Stewart, “A competitive microflora increases the Water Science and Technology, vol. 50, pp. 271–275, 2004. resistance of Salmonella typhimurium to inimical processes: [12] P. Servais, J. Prats, J. Passerat, and T. Garcia-Armisen, evidence for a suicide response,” Applied and Environmental “Abundance of culturable versus viable Escherichia coli in Microbiology, vol. 64, no. 4, pp. 1323–1327, 1998. freshwater,” Canadian Journal of Microbiology, vol. 55, no. 7, [27] EU, “Directive 2006/7/EC of the European Parliament pp. 905–909, 2009. and of the Council of 15 February 2006 concerning the 12 International Journal of Microbiology

management of bathing water quality and repealing directive [42] R. T. Noble and S. B. Weisberg, “A review of technologies for 76/160/EEC,” Official Journal of the European Communities, rapid detection of bacteria in recreational waters,” Journal of L 64/37, 2006. Water and Health, vol. 3, no. 4, pp. 381–392, 2005. [28] V. Deretic and B. Levine, “Autophagy, immunity, and [43] P. R. Hunter and Q. Syed, “Community surveys of self- microbial adaptations,” Cell Host and Microbe,vol.5,no.6, reported diarrhoea can dramatically overestimate the size of pp. 527–549, 2009. outbreaks of waterborne cryptosporidiosis,” Water Science [29] I. Tryland, M. Pommepuy, and L. Fiksdal, “Effect of and Technology, vol. 43, no. 12, pp. 27–30, 2001. chlorination on β-D-galactosidase activity of sewage bacteria [44] P.A. Rochelle, M. M. Marshall, J. R. Mead et al., “Comparison and Escherichia coli,” Journal of Applied Microbiology, vol. 85, of in vitro cell culture and a mouse assay for measuring no. 1, pp. 51–60, 1998. infectivity of Cryptosporidium parvum,” Applied and Envi- ronmental Microbiology, vol. 68, no. 8, pp. 3809–3817, 2002. [30] A. H. Farnleitner, L. Hocke, C. Beiwl, G. G. Kavka, and [45]B.Corona-Vasquez,A.Samuelson,J.L.Rennecker,andB.J. R. L. Mach, “Hydrolysis of 4-methylumbelliferyl-β-D- Marinas,˜ “Inactivation of Cryptosporidium parvum oocysts glucuronide in differing sample fractions of river waters and with ozone and free chlorine,” Water Research, vol. 36, no. its implication for the detection of fecal pollution,” Water 16, pp. 4053–4063, 2002. Research, vol. 36, no. 4, pp. 975–981, 2002. [46] T. R. Slifko, H. V. Smith, and J. B. Rose, “Emerging parasite [31] I. George, P. Crop, and P. Servais, “Use of β-D-galactosidase β zoonoses associated with water and food,” International Jour- and -D-glucuronidase activities for quantitative detection nal for Parasitology, vol. 30, no. 12-13, pp. 1379–1393, 2000. of total and fecal coliforms in wastewater,” Canadian Journal [47] H. L. DuPont, C. L. Chappell, C. R. Sterling, P. C. of Microbiology, vol. 47, no. 7, pp. 670–675, 2001. Okhuysen, J. B. Rose, and W. Jakubowski, “The infectivity of [32]G.Caruso,E.Crisafi,andM.Mancuso,“Developmentof Cryptosporidium parvum in healthy volunteers,” New England an enzyme assay for rapid assessment of Escherichia coli in Journal of Medicine, vol. 332, no. 13, pp. 855–859, 1995. seawaters,” Journal of Applied Microbiology,vol.93,no.4,pp. [48] G. F. Craun, N. Nwachuku, R. L. Calderon, and M. F. Craun, 548–556, 2002. “Outbreaks in drinking-water systems, 1991–1998,” Journal [33] I. George, A. Anzil, and P. Servais, “Quantification of fecal of Environmental Health, vol. 65, no. 1, pp. 16–23, 2002. coliform inputs to aquatic systems through soil leaching,” [49] R. M. McCuin and J. L. Clancy, “Modifications to United Water Research, vol. 38, no. 3, pp. 611–618, 2004. States Environmental Protection Agency methods 1622 and [34]A.M.Zimmerman,D.M.Rebarchik,A.R.Flowers,J.L. 1623 for detection of Cryptosporidium oocysts and Giardia Williams, and D. J. Grimes, “Escherichia coli detection using cysts in water,” Applied and Environmental Microbiology, vol. mTEC agar and fluorescent antibody direct viable counting 69, no. 1, pp. 267–274, 2003. on coastal recreational water samples,” Letters in Applied [50] “Directive 98/83/CE relative a la qualite des eaux destinees Microbiology, vol. 49, no. 4, pp. 478–483, 2009. a la consommation humaine,” Journal Official no. L L330 de [35] S. Okabe and Y. Shimazu, “Persistence of host-specific L’Union Europeene, Decembre 1998. Bacteroides-Prevotella 16S rRNA genetic markers in [51] C. N. Haas and J. B. Rose, “Developing an action level for environmental waters: effects of temperature and salinity,” Cryptosporidium,” Journal of the American Water Works Applied Microbiology and Biotechnology,vol.76,no.4,pp. Association, vol. 87, pp. 81–84, 1995. 935–944, 2007. [52] M. Garvey, H. Farrell, M. Cormican, and N. Rowan, [36]P.W.M.H.Smeets,Y.J.Dullemont,P.H.A.J.M.VanGelder, “Investigations of the relationship between use of in vitro J. C. Van Dijk, and G. J. Medema, “Improved methods cell culture-quantitative PCR and a mouse-based bioassay ff for modelling drinking water treatment in quantitative for evaluating critical factors a ecting the disinfection microbial risk assessment; a case study of Campylobacter performance of pulsed UV light for treating Cryptosporidium reduction by filtration and ozonation,” Journal of Water and parvum oocysts in saline,” Journal of Microbiological Methods, Health, vol. 6, no. 3, pp. 301–314, 2008. vol. 80, no. 3, pp. 267–273, 2010. [53] A. M. Johnson, K. Linden, K. M. Ciociola, R. De Leon, [37] J. Madetoja, S. Nystedt, and T. Wiklund, “Survival and viru- G. Widmer, and P. A. Rochelle, “UV inactivation of lence of Flavobacterium psychrophilum in water microcosms,” Cryptosporidium hominis as measured in cell culture,” FEMS Microbiology Ecology, vol. 43, no. 2, pp. 217–223, 2003. Applied and Environmental Microbiology,vol.71,no.5,pp. [38] J. Madetoja and T. Wiklund, “Detection of the fish pathogen 2800–2802, 2005. Flavobacterium psychrophilum in water from fish farms,” [54] J. R. Bolton and K. G. Linden, “Standardization of methods Systematic and Applied Microbiology,vol.25,no.2,pp. for fluence (UV Dose) determination in bench-scale UV 259–266, 2002. experiments,” Journal of Environmental Engineering, vol. 129, ff [39] E. Marco-Noales, E. G. Biosca, and C. Amaro, “E ects of no. 3, pp. 209–215, 2003. salinity and temperature on long-term survival of the eel [55] N. Elmnasser, S. Guillou, F. Leroi, N. Orange, A. Bakhrouf, pathogen Vibrio vulnificus biotype 2 (serovar E),” Applied and and M. Federighi, “Pulsed-light system as a novel food Environmental Microbiology, vol. 65, no. 3, pp. 1117–1126, decontamination technology: a review,” Canadian Journal of 1999. Microbiology, vol. 53, no. 7, pp. 813–821, 2007. [40] L. Madsen and I. Dalsgaard, “Water recirculation and good [56] V. M. Gomez-L´ opez,´ P. Ragaert, J. Debevere, and F. management: potential methods to avoid disease outbreaks Devlieghere, “Pulsed light for food decontamination: a with Flavobacterium psychrophilum,” Journal of Fish Diseases, review,” Trends in Food Science and Technology, vol. 18, no. 9, vol. 31, no. 11, pp. 799–810, 2008. pp. 464–473, 2007. [41] B. W. Kooi, D. Bontje, and M. Liebig, “Model analysis of [57] B. F. Kalisvaart, “Re-use of wastewater: preventing the a simple aquatic ecosystems with sublethal toxic effects,” recovery of pathogens by using medium-pressure UV lamp Mathematical Biosciences and Engineering,vol.5,no.4,pp. technology,” Water Science and Technology, vol. 50, no. 6, pp. 771–787, 2008. 337–344, 2004. International Journal of Microbiology 13

[58] L. Meunier, S. Canonica, and U. von Gunten, “Implications [75] N. J. Rowan, S. Espie, J. Harrower, J. G. Anderson, L. of sequential use of UV and ozone for drinking water Marsili, and S. J. MacGregor, “Pulsed-plasma gas-discharge quality,” Water Research, vol. 40, no. 9, pp. 1864–1876, 2006. inactivation of microbial pathogens in chilled poultry [59] T. Bintsis, E. Litopoulou-Tzanetaki, and R. K. Robinson, wash water,” Journal of Food Protection, vol. 70, no. 12, pp. “Existing and potential applications of ultraviolet light in 2805–2810, 2007. the food industry—a critical review,” Journal of the Science of [76] A. M. Anpilov, E. M. Barkhudarov, N. Christofi et al., “Pulsed Food and Agriculture, vol. 80, no. 6, pp. 637–645, 2000. high voltage electric discharge disinfection of microbially [60] N. J. Rowan, S. J. MacGregor, J. G. Anderson, R. A. Fouracre, contaminated liquids,” Letters in Applied Microbiology, vol. L. McIlvaney, and O. Farish, “Pulsed-light inactivation of 35, no. 1, pp. 90–94, 2002. food-related microorganisms,” Applied and Environmental [77]W.F.L.M.Hoeben,E.M.VanVeldhuizen,W.R.Rutgers,C. Microbiology, vol. 65, no. 3, pp. 1312–1315, 1999. A. M. G. Cramers, and G. M. W. Kroesen, “The degradation [61] H. Farrell, M. Garvey, and N. Rowan, “Studies on the of aqueous phenol solutions by pulsed positive corona inactivation of medically important Candida species on agar discharges,” Plasma Sources Science and Technology, vol. 9, surfaces using pulsed light,” FEMS Yeast Research, vol. 9, no. no. 3, pp. 361–369, 2000. 6, pp. 956–966, 2009. [78] N. J. Rowan, J. G. Laffey, and M. Rogers, “Studies on the [62] S.-U. Lee, M. Joung, D.-J. Yang et al., “Pulsed-UV light removal of prions from surgical instruments using novel inactivation of Cryptosporidium parvum,” Parasitology pulsed plasma gas discharge system,” Science Foundation Research, vol. 102, no. 6, pp. 1293–1299, 2008. Ireland (Research Frontier Grant ENMF 368), 2007. [63] S. J. MacGregor, N. J. Rowan, L. McIlvaney, J. G. Anderson, [79] U. Von Gunten, “Ozonation of drinking water—part II: R. A. Fouracre, and O. Farish, “Light inactivation of food- disinfection and by-product formation in presence of related pathogenic bacteria using a pulsed power source,” bromide, iodide or chlorine,” Water Research, vol. 37, no. 7, Letters in Applied Microbiology, vol. 27, no. 2, pp. 67–70, 1998. pp. 1469–1487, 2003. [64]J.G.Anderson,N.J.Rowan,S.J.MacGregor,R.A.Fouracre, [80] F. Hammes, E. Salhi, O. Koster,¨ H.-P. Kaiser, T. Egli, and U. and O. Farish, “Inactivation of food-borne enteropathogenic von Gunten, “Mechanistic and kinetic evaluation of organic bacteria and spoilage fungi using pulsed-light,” IEEE Trans- disinfection by-product and assimilable organic carbon actions on Plasma Science, vol. 28, no. 1, pp. 83–88, 2000. (AOC) formation during the ozonation of drinking water,” Water Research, vol. 40, no. 12, pp. 2275–2286, 2006. [65] R. R. Sharma and A. Demirci, “Inactivation of Escherichia coli [81] E. De Bel, J. Dewulf, B. D. Witte, H. Van Langenhove, and C. O157:H7 on inoculated alfalfa seeds with pulsed ultraviolet Janssen, “Influence of pH on the sonolysis of ciprofloxacin: light and response surface modeling,” Journal of Food Science, biodegradability, ecotoxicity and antibiotic activity of its vol. 68, no. 4, pp. 1448–1453, 2003. degradation products,” Chemosphere, vol. 77, no. 2, pp. [66]T.Wang,S.J.MacGregor,J.G.Anderson,andG.A.Woolsey, 291–295, 2009. “Pulsed ultra-violet inactivation spectrum of Escherichia [82] R. Wise, “Antimicrobial resistance: priorities for action,” coli,” Water Research, vol. 39, no. 13, pp. 2921–2925, 2005. Journal of Antimicrobial Chemotherapy,vol.49,no.4,pp. [67] P. Roberts and A. Hope, “Virus inactivation by high intensity 585–586, 2002. broad spectrum pulsed light,” Journal of Virological Methods, [83] K. Kummerer,¨ “Antibiotics in the aquatic environment—a vol. 110, no. 1, pp. 61–65, 2003. review—part I,” Chemosphere, vol. 75, no. 4, pp. 417–434, [68] D. Marquenie, A. H. Geeraerd, J. Lammertyn et al., 2009. “Combinations of pulsed white light and UV-C or mild [84] K. Kummerer,¨ “Antibiotics in the aquatic environment—a heat treatment to inactivate conidia of Botrytis cinerea review—part II,” Chemosphere, vol. 75, no. 4, pp. 435–441, and Monilia fructigena,” International Journal of Food 2009. Microbiology, vol. 85, no. 1-2, pp. 185–196, 2003. [85] J. L. Caplin, G. W. Hanlon, and H. D. Taylor, “Presence of [69]P.A.Rochelle,D.Fallar,M.M.Marshall,B.A.Montelone, vancomycin and ampicillin-resistant Enterococcus faecium of S. J. Upton, and K. Woods, “Irreversible UV inactivation epidemic clonal complex-17 in wastewaters from the south of Cryptosporidium spp. despite the presence of UV repair coast of England,” Environmental Microbiology, vol. 10, no. genes,” Journal of Eukaryotic Microbiology,vol.51,no.5,pp. 4, pp. 885–892, 2008. 553–562, 2004. [86] H. Schmidt and J. Rombke,¨ “The ecotoxicological effects of [70]P.A.Rochelle,S.J.Upton,B.A.Montelone,andK.Woods, pharmaceuticals (antibiotics and antiparasiticides) in the “The response of Cryptosporidium parvum to UV light,” terrestrial environment—a review,” in Pharmaceuticals in the Trends in Parasitology, vol. 21, no. 2, pp. 81–87, 2005. Environment. Sources, Fate, Effects and Risks,K.Kummerer,¨ [71] M. Otaki, A. Okuda, K. Tajima, T. Iwasaki, S. Kinoshita, and Ed., pp. 285–303, Springer, Berling, Germany, 3rd edition, S. Ohgaki, “Inactivation differences of microorganisms by 2008. low pressure UV and pulsed xenon lamps,” Water Science [87] A. J. Alanis, “Resistance to antibiotics: are we in the post- and Technology, vol. 47, no. 3, pp. 185–190, 2003. antibiotic era?” Archives of Medical Research, vol. 36, no. 6, [72] A. Driedger, E. Staub, U. Pinkernell, B. Marinas,˜ W. Koster,¨ pp. 697–705, 2005. andU.V.Gunten,“InactivationofBacillus subtilis spores and [88] J. Davison, “Genetic exchange between bacteria in the formation of bromate during ozonation,” Water Research, environment,” Plasmid, vol. 42, no. 2, pp. 73–91, 1999. vol. 35, no. 12, pp. 2950–2960, 2001. [89] K. Ohlsen, T. Ternes, G. Werner et al., “Impact of antibiotics [73] J. L. Rennecker, B. J. Marinas,˜ J. H. Owens, and E. W. Rice, on conjugational resistance gene transfer in Staphylococcus “Inactivation of Cryptosporidium parvum oocysts with aureus in sewage,” Environmental Microbiology, vol. 5, no. 8, ozone,” Water Research, vol. 33, no. 11, pp. 2481–2488, 1999. pp. 711–716, 2003. [74] V. Yargeau and C. Leclair, “Potential of ozonation for the [90]J.W.FosterandM.P.Spector,“HowSalmonella survive degradation of antibiotics in wastewater,” Water Science and against the odds,” Annual Review of Microbiology, vol. 49, pp. Technology, vol. 55, no. 12, pp. 321–326, 2007. 145–174, 1995. 14 International Journal of Microbiology

[91] J. Zupan and P. Raspor, “Invasive growth of Saccharomyces [107] Food Safety Authority of Ireland, “Report on food safety cerevisiae depends on environmental triggers: a quantitative implications of land spreading agriculture, municipal and model,” Yeast, vol. 27, no. 4, pp. 217–228, 2010. industrial organic matter on agricultural land used for [92] J. Wiethan, A. Al-Ahmad, A. Henninger, and K. Kummerer,¨ food production in Ireland,” 2008, www.fsai.ie/WorkArea/ “Simulation of the selection pressure of the antibiotics DownloadAsset.aspx?id=8226. ciprofloxacin and ceftazidim in surface water with classical [108] R. M. Sharpe, “Hormones and testis development and methods,” Vom Wasser, vol. 95, pp. 107–118, 2001. the possible adverse effects of environmental chemicals,” [93] A. Al-Ahmad, A. Haiß, J. Unger, A. Brunswick-Tietze, J. Toxicology Letters, vol. 120, no. 1-3, pp. 221–232, 2001. ff Wiethan, and K. Kummerer,¨ “E ects of a realistic mixture [109] EPA, Final Report: Endocrine Disruptors in the Irish Aquatic of antibiotics on resistant and nonresistant sewage sludge Environment, Environmental Protection Agency, Wexford, bacteria in laboratory-scale treatment plants,” Archives of Ireland, 2005. Environmental Contamination and Toxicology, vol. 57, no. 2, [110] M. P. Fernandez, P. M. Campbell, M. G. Ikonomou, and pp. 264–273, 2009. R. H. Devlin, “Assessment of environmental estrogens ¨ [94] S. Gartiser, E. Urich, R. Alexy, and K. Kummerer, “Ultimate and the intersex/sex reversal capacity for Chinook salmon biodegradation and elimination of antibiotics in inherent (Oncorhynchus tshawytscha) in primary and final municipal tests,” Chemosphere, vol. 67, no. 3, pp. 604–613, 2007. wastewater effluents,” Environment International, vol. 33, no. [95] D. Li, M. Yang, J. Hu, L. Ren, Y. Zhang, and K. Li, 3, pp. 391–396, 2007. “Determination and fate of oxytetracycline and related [111] R. M. Sharpe and N. E. Skakkebaek, “Are oestrogens involved compounds in oxytetracycline production wastewater and in falling sperm counts and disorders of the male reproduc- the receiving river,” Environmental Toxicology and Chemistry, tive tract?” Lancet, vol. 341, no. 8857, pp. 1392–1395, 1993. vol. 27, no. 1, pp. 80–86, 2008.  [96] R. Alexy, T. Kumpel,¨ and K. Kummerer,¨ “Assessment of [112] T. Murata and K. Yamauchi, “3,3 ,5-triiodo-l-thyronine-like ffl degradation of 18 antibiotics in the closed bottle test,” activity in e uents from domestic sewage treatment plants Chemosphere, vol. 57, no. 6, pp. 505–512, 2004. detected by in vitro and in vivo bioassays,” Toxicology [97] S. Gartiser, E. Urich, R. Alexy, and K. Kummerer,¨ “Anaerobic and Applied Pharmacology, vol. 226, no. 3, pp. 309–317, inhibition and biodegradation of antibiotics in ISO test 2008. schemes,” Chemosphere, vol. 66, no. 10, pp. 1839–1848, 2007. [113] T. A. Ternes, M. Stumpf, J. Mueller, K. Haberer, R.-D. Wilken, [98]S.Kim,P.Eichhorn,J.N.Jensen,A.S.Weber,andD.S.Aga, and M. Servos, “Behavior and occurrence of estrogens in “Removal of antibiotics in wastewater: effect of hydraulic municipal sewage treatment plants—I. Investigations in and solid retention times on the fate of tetracycline in Germany, Canada and Brazil,” Science of the Total Environ- the activated sludge process,” Environmental Science and ment, vol. 225, no. 1-2, pp. 81–90, 1999. Technology, vol. 39, no. 15, pp. 5816–5823, 2005. [114] L. Lishman, S. A. Smyth, K. Sarafin et al., “Occurrence and [99] H.-T. Lai, Y.-H. Chien, and J.-S. Lin, “Long-term reductions of pharmaceuticals and personal care products transformation of oxolinic acid in water from an eel and estrogens by municipal wastewater treatment plants in pond,” Aquaculture, vol. 275, no. 1–4, pp. 96–101, 2008. Ontario, Canada,” Science of the Total Environment, vol. 367, [100] S. Thiele-Bruhn, “Pharmaceutical antibiotic compounds in no. 2-3, pp. 544–558, 2006. soils—a review,” Journal of Plant Nutrition and Soil Science, [115] A. Fogarty and C. McGee, “Male or female: that is the vol. 166, no. 2, pp. 145–167, 2003. question?” The Irish Scientist, vol. 15, pp. 40–44, 2007. [101] T. Maki, H. Hasegawa, H. Kitami, K. Fumoto, Y. Munekage, [116] S. J. Brennan, The ecotoxicological assessment and microbial and K. Ueda, “Bacterial degradation of antibiotic residues transformation of selected oestrogen mimics, Ph.D. thesis, in marine fish farm sediments of Uranouchi Bay and Athlone Institute of Technology, 2005. phylogenetic analysis of antibiotic-degrading bacteria using [117] S. J. Brennan, C. A. Brougham, J. J. Roche, and A. M. Fogarty, 16S rDNA sequences,” Fisheries Science, vol. 72, no. 4, pp. “Multi-generational effects of four selected environmental 811–820, 2006. oestrogens on Daphnia magna,” Chemosphere, vol. 64, no. 1, [102] F. A. Neela, L. Nonaka, and S. Suzuki, “The diversity of pp. 49–55, 2006. multi-drug resistance profiles in tetracycline-resistant Vibrio [118] F. Protoxkit, Freshwater Toxicity Test with a Ciliate Protozoan. species isolated from coastal sediments and seawater,” Standard Operating Procedure, Creasal, Deinze, Belgium, Journal of Microbiology, vol. 45, no. 1, pp. 64–68, 2007. 1998. [103] A. Kimiran-Erdem, E. O. Arslan, N. O. Sanli Yurudu, Z. Zeybek, N. Dogruoz, and A. Cotuk, “Isolation and [119] A. M. Reid, C. A. Brougham, A. M. Fogarty, and J. J. identification of Enterococci from seawater samples: Roche, “An investigation into possible sources of phthalate assessment of their resistance to antibiotics and heavy contamination in the environmental analytical laboratory,” metals,” Environmental Monitoring and Assessment, vol. 125, International Journal of Environmental Analytical Chemistry, no. 1–3, pp. 219–228, 2007. vol. 87, no. 2, pp. 125–133, 2007. [104] M. Sjolund,¨ J. Bonnedahl, J. Hernandez et al., “Dissemination [120] H. Yamamoto, Y. Nakamura, S. Moriguchi et al., “Persistence of multidrug-resistant bacteria into the Arctic,” Emerging and partitioning of eight selected pharmaceuticals in the Infectious Diseases, vol. 14, pp. 70–72, 2008. aquatic environment: laboratory photolysis, biodegradation, [105] A. Pruden, R. Pei, H. Storteboom, and K. H. Carlson, and sorption experiments,” Water Research, vol. 43, no. 2, “Antibiotic resistance genes as emerging contaminants: pp. 351–362, 2009. studies in northern Colorado,” Environmental Science and [121] D. M. Willberg, P. S. Lang, R. H. Hochemer,¨ A. Kratel, Technology, vol. 40, no. 23, pp. 7445–7450, 2006. and M. R. Hoffmann, “Degradation of 4-chlorophenol, [106] P.H. Serrano, “Responsible use of antibiotics in aquaculture,” 3,4-dichloroaniline, and 2,4,6- trinitrotoluene in an Fisheries Technical Paper 469, Food and Agriculture Organi- electrohydraulic discharge reactor,” Environmental Science zation of the United Nations (FAO), Rome, Italy, 2005. and Technology, vol. 30, no. 8, pp. 2526–2534, 1996. International Journal of Microbiology 15

[122] A. G. Bubnov, V. I. Grinevich, and N. A. Kuvykin, “Phenol degradation features in aqueous solutions upon dielectric- barrier discharge treatment,” High Energy Chemistry, vol. 38, no. 5, pp. 338–343, 2004. [123] Y. J. Liu and X. Z. Jiang, “Phenol degradation by a nonpulsed diaphragm glow discharge in an aqueous solution,” Environmental Science and Technology, vol. 39, no. 21, pp. 8512–8517, 2005. [124] X. Hao, M. Zhou, Q. Xin, and L. Lei, “Pulsed discharge plasma induced Fenton-like reactions for the enhancement of the degradation of 4-chlorophenol in water,” Chemosphere, vol. 66, no. 11, pp. 2185–2192, 2007. [125] M. T. Amin, S. Ryu, and H. Park, “Degradation of phenol and Bisphenol-A using discharged water generating system,” Journal of Water Supply: Research and Technology—AQUA, vol. 56, no. 3, pp. 203–216, 2007. [126] H. M. Coleman, E. J. Routledge, J. P. Sumpter, B. R. Eggins, and J. A. Byrne, “Rapid loss of estrogenicity of steroid estrogensbyUVAphotolysisandphotocatalysisoveran immobilized titanium dioxide catalyst,” Water Research, vol. 38, pp. 3233–3240, 2004. [127] G. Vaes, P. Willems, P. Swartenbroekx, K. Kramer, W. de Lange, and K. Kober, “Scientific-policy interfacing in support of the water framework directive implementation,” Water Science and Technology, vol. 60, pp. 47–54, 2009. [128] K. D. Mena and C. P. Gerba, “Risk assessment of Pseudomonas aeruginosa in water,” Reviews of Environmental Contamination and Toxicology, vol. 201, pp. 71–115, 2009. [129] D. Bontje, B. W. Kooi, M. Liebig, and S. A. L. M. Kooijman, “Modelling long-term ecotoxicological effects on an algal population under dynamic nutrient stress,” Water Research, vol. 43, no. 13, pp. 3292–3300, 2009. [130] V. Gevaert, F. Verdonck, L. Benedetti, W. De Keyser, and B. De Baets, “Evaluating the usefulness of dynamic pollutant fate models for implementing the EU Water Framework Directive,” Chemosphere, vol. 76, no. 1, pp. 27–35, 2009. [131] A. Wennmalm and B. Gunnarsson, “Pharmaceutical man- agement through environmental product labelling in Swe- den,” Environment International, vol. 35, pp. 775–777, 2009. [132] V. T. Sedas, “Influence of environmental factors on the presence of Vibrio cholerae in the marine environment: a climate link,” Journal of Infection in Developing Countries, vol. 1, pp. 224–241, 2007. [133] A. M. J. Ragas, M. A. J. Huijbregts, I. Henning-De Jong, and R. S. Leuven, “Uncertainty in environmental risk assessment: implications for risk-based management of river basins,” Integrated Environmental Assessment and Management, vol. 5, no. 1, pp. 27–37, 2009. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 152815, 15 pages doi:10.1155/2011/152815

Review Article Prominent Human Health Impacts from Several Marine Microbes: History, Ecology, and Public Health Implications

P. K. Bi en fa n g , 1 S. V. DeFelice,1 E. A. Laws,1 L. E. Brand,2 R. R. Bidigare,1 S. Christensen,1 H. Trapido-Rosenthal,1 T. K. Hemscheidt,1 D. J. McGillicuddy Jr.,3 D. M. Anderson,3 H. M. Solo-Gabriele,4 A. B. Boehm,5 andL.C.Backer6

1 Center for Oceans and Human Health, Pacific Research Center for Marine Biomedicine, School of Ocean and Earth Science and Technology, MSB no. 205, University of Hawaii, Honolulu, HI, 96822, USA 2 Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Cswy, Miami, FL 33149, USA 3 Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA 4 Department of Civil, Architectural, and Environmental Engineering, University of Miami, Coral Gables, Florida and University of Miami Center for Oceans and Human Health, Key Biscayne, FL 33124-0630, USA 5 Department of Civil and Environmental Engineering, Stanford University, Stanford, California and University of Hawaii Center for Oceans and Human Health, Honolulu, HI 96822, USA 6 National Center for Environmental Health Centers for Disease Control and Prevention, 47770 Buford Highway NE MS F-46, Chamblee, GA 30341, USA

Correspondence should be addressed to P. K. Bienfang, [email protected]

Received 15 June 2010; Revised 23 July 2010; Accepted 25 July 2010

Academic Editor: Max Teplitski

Copyright © 2011 P. K. Bienfang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This paper overviews several examples of important public health impacts by marine microbes and directs readers to the extensive literature germane to these maladies. These examples include three types of dinoflagellates (Gambierdiscus spp., Karenia brevis, and Alexandrium fundyense), BMAA-producing cyanobacteria, and infectious microbes. The dinoflagellates are responsible for ciguatera fish poisoning, neurotoxic shellfish poisoning, and paralytic shellfish poisoning, respectively, that have plagued coastal populations over time. Research interest on the potential for marine cyanobacteria to contribute BMAA into human food supplies has been derived by BMAA’s discovery in cycad seeds and subsequent implication as the putative cause of amyotrophic lateral sclerosis/parkinsonism dementia complex among the Chamorro people of Guam. Recent UPLC/MS analyses indicate that recent reports that BMAA is prolifically distributed among marine cyanobacteria at high concentrations may be due to analyte misidentification in the analytical protocols being applied for BMAA. Common infectious microbes (including enterovirus, norovirus, Salmonella, Campylobacter, Shigella, Staphylococcus aureus, Cryptosporidium,andGiardia) cause gastrointestinal and skin-related illness. These microbes can be introduced from external human and animal sources, or they can be indigenous to the marine environment.

1. Introduction ways to the microbial organisms that are at the base of marine ecosystems. In 2004, the National Science Founda- The health and welfare of humans residing in the coastal tion and the National Institutes of Environmental Health zone and in island communities are inextricably linked Sciences initiated collaborative funding of four centers to the oceans and its foodwebs. The multitude of rela- for oceans and human health, and in the same year, tionships between human societies and the oceans has the National Oceanic and Atmospheric Administration led to many human dimensions of ocean issues and pro- launched its Oceans and Human Health Initiative. One of cesses. The effects of climate change, pollution, popula- the important accomplishments of these centers has been tion increases, and the myriad of anthropogenic effects the cross-discipline synergistic collaboration of scientists attendant increasing population are all related in various within and between centers. This paper summarizes five 2 International Journal of Microbiology areas of this research focusing on the oceans and human fish contained significant numbers of a dinoflagellate that health. was designated as a new genus and species known as Gambierdiscus toxicus [20]. Since then several new species 2. Gambierdiscus spp. and Ciguatera have been added to the genus [21–23], and recent works Fish Poisoning using refined morphological and molecular sequencing techniques have caused substantial changes to the taxonomy Gambierdiscus spp., a genus of dinoflagellates (division of this genus [24, 25]. Derived from gambiertoxins produced Phyrophyta), produce natural toxins that cause ciguatera fish by the dinoflagellates, ciguatoxin is a polar, lipid-soluble poisoning (CFP) in humans. In contrast to other noxious polyether. The toxin is heat stable, tasteless, odorless, and dinoflagellates known for their dense “red tide” aggre- effective at extremely low (i.e., sub-ppb) concentrations; the gations, Gambierdiscus do not form conspicuous blooms severe analytical challenges presented by these properties that color the water. These dinoflagellates normally grow have been central to the slow progress in detection of epiphytically on various macroalgae in coral reef ecosystems ciguatoxin for prevention and/or research purposes. within the 35◦N–35◦S latitudinal band. The dinoflagellates A CDC report for a period in the 70’s indicted that are consumed by herbivorous fish, beginning processes of reported CFP incidences accounted for 25% of all food- bioaccumulation and biomodification through the reef food borne outbreaks, which was five times the reported incidence web, as the herbivores are consumed by carnivores and for paralytic shellfish poisoning and neurological shellfish ultimately by humans. Several informative reviews on CFP poisoning combined [26]. Kite-Powell [27] concluded that may be found in Bienfang et al. [1], Dickey [2], and Dickey the economic impact from CFP exceeded that from any and Plakas [3] and references cited therein. other form of hazardous algae bloom. Additionally, CFP Ciguatera is a food-borne disease that has affected coastal is associated with societal/public health impacts in island populations and travelers in tropical and subtropical regions communities due to dietary changes in response to concerns throughout the world for centuries. Ship logs as far back as over the quality of local seafood. Chateau-Degat et al. [28] the 16th century mention clinical symptomologies consistent showed correlations of Gambierdiscus abundance and CFP with CFP [4, 5], and during WWII, CFP was a serious with sea surface temperature in the South Pacific. This problem for military troops stationed in many Pacific island has attracted particular concern because warming oceans locals. There are at least 50,000 reported cases of CFP cases would expand the range of Gambierdiscus into higher peryear[6, 7], but due to the high degree of misdiagnosis latitudes where population density is generally greater. and underreporting, it is estimated that the actual frequency Though potentially influenced by improved awareness, the of CFP cases is closer to 500,000 per year [8, 9]. It is recognition of CFP in new geographic areas [29–31]has estimated that >50% of the populations of small islands in been suggested as evidence of an expanded range for CFP. the Caribbean and South Pacific have suffered from CFP; Because of the underreporting artifact, it is difficult to see reviews by Lewis [10], Lange [11], and Fleming et al. ascertain whether CFP incidence is increasing over time, [12, 13]. though expansion of international trade in seafood from Ciguatoxin produces gastrointestinal, neurological, and tropical regions and climatic warming make this a distinct cardiovascular symptoms. These normally develop within probability. 12–24 hours of eating contaminated fish. Gastrointestinal effects may disappear in four days. The normal pro- 3. Karenia brevis and Neurotoxic gression of symptoms is gastrointestinal symptoms (e.g., Shellfish Poisoning diarrhea, abdominal pain, nausea, and vomiting), followed by neurological symptoms (e.g., numbness and tingling Karenia brevis is an unarmored photosynthetic dinoflagellate of hands and feet, dizziness, altered hot/cold perception, that lives primarily in the Gulf of Mexico and produces a muscle aches, low heart rates, and low blood pressure). suite of neurotoxins called “brevetoxins”. When K. brevis Symptoms may persist in some forms for weeks, months, occurs at concentrations above ∼100,000 cells/L, humans or even years [9, 14–16]. Generally, feelings of weakness become aware of its presence of primarily as a result of last ∼1 week, and neurosensory manifestations (e.g., muscle three effects. At such concentrations, brevetoxins can cause: aches, tingling extremities, and thermal reversals) com- (1) fish kills, (2) filter-feeding shellfish to become extremely monly represent the most prolonged discomfort. The key toxic to humans, and (3) respiratory distress, coughing, and pathonuemonic symptom is the neurological malady of eye irritation in humans due to aerosolization. Such impacts reversal of hot/cold sensation. Fortunately, death is rare (i.e., were experienced long before K. brevis was known to be the <0.1%) and is most commonly the result of respiratory causative agent. Spanish explorers as early as the 15th century failure due to cardiovascular shock induced by severe recorded fish kills in the Gulf of Mexico that were probably dehydration. An interesting feature is that CFP intoxication caused by K. brevis. Explorers also noticed that the native does not confer any immunity in its victims, and to the Americans were aware that shellfish could be toxic [32–34]. contrary, its frequency results in heightened sensitivity to Davis [34] was the first to demonstrate that K. brevis was in ciguatoxin. fact the causative organism of fish kills and toxic shellfish. TheetiologyofCFPwasadvancedwhenworkin When first identified, it was named Gymnodinium breve.It the Gambier Islands of French Polynesia by Yasumoto was later renamed Ptychodiscus brevis and is currently known et al. [17–19] revealed that the guts of toxic (herbivorous) as Karenia brevis. International Journal of Microbiology 3

K. brevis is an unarmored dinoflagellate with a flat- blooms of K. brevis could be an important source of nutrients tened shape, approximately 10–15 microns thick and 25–35 as they decompose. A perusal of the data on the spatial microns in diameter. It tends to swim up to the light during and temporal distribution of K. brevis blooms suggests the day and disperse throughout the water column at night that no one hypothesis will provide a simple explanation [35, 36]. It typically reproduces by binary fission every 2–5 for its distribution. A variety of nutrient sources probably days [37]. It is suspected to have a resting stage, perhaps in contribute to varying degrees to blooms of K. brevis. the benthos, but this is not firmly established at the present K. brevis produces a suite of around 12 brevetoxins that time [38]. activate the sodium channel of neurons [51]. At concentra- Under typical nonbloom conditions, K. brevis concen- tions >100,0000 cells/L, blooms of K. brevis can kill many trations are usually <10 cells/L offshore and ∼1000 cells/L marine animals, including fish, turtles, sea birds, manatees, inshore [39]. For reasons that are still not well understood, and dolphins [52]. Because brevetoxins are large lipid- it occasionally forms blooms with concentrations of one to soluble molecules, they tend to accumulate in fatty tissues tens millions of cells per liter. K. brevis has been observed and are not easily broken down or excreted. As a result, filter- sporadically throughout the Gulf of Mexico, but the largest feeding shellfish can accumulate high concentrations in their and most frequent blooms occur along the west coast of tissues. Brevetoxins are heat and acid stable, thus remain Florida [32, 33, 40]. Although not as frequent, blooms also toxic after cooking. Human ingestion of toxic shellfish occur along the coast of Texas and Mexico [32]. About once can lead to a variety of neurological and gastrointestinal a decade, conditions allow the Loop Current in the Gulf symptoms [53], giving the general term “neurotoxic shellfish of Mexico to pick up blooms of K. brevis and transport it poisoning” (NSP). Because of this, government agencies in through the Straits of Florida and along the east coast of Florida monitor for K. brevis blooms and close shellfish beds North America [32, 41]. As a result, blooms of K. brevis have to harvesting at times of blooms [53].Asaresult,NSPisnow been found as far north as North Carolina [41]. Currently, rare, usually the result of illegal or uninformed harvesting of there is no evidence that these blooms along the east coast toxic shellfish. were generated locally, but rather were transported there Because the brevetoxins are lipid soluble, they also from the Gulf of Mexico. have the potential to accumulate and biomagnify up the Along the west coast of Florida, where records have been food chain [54]. As a result, sublethal concentrations of kept for a half century, blooms of K. brevis are statistically K. brevis can still have lethal consequences [40]. Sublethal more likely during the fall months, a month or two after the concentrations of K. brevis that do not kill nevertheless heaviest freshwater runoff from land [40] although they can release brevetoxins that accumulate in organisms; thus occur at any time of the year. Huge year-to-year variations in higher level predators can accumulate high concentrations bloom concentrations and/or duration are evident. In some in their tissues. This may explain why dolphins and manatees years, essentially no blooms are observed, some years have have been found dead with high concentrations of brevetoxin short sporadic blooms, and some years large blooms may in their stomachs and tissues in areas where no obvious persist throughout the entire year. blooms of K. brevis were observed [55]. Naar et al. [56]have Understanding the causes for the spatial and temporal found brevetoxins in the tissues of many fish species many occurrence/variability of K. brevis blooms is an area of months after the occurrence of a K. brevis bloom. These active research. Factors that promote initiation of blooms recent data suggest that brevetoxins may be more widespread maywellbedifferent from those that determine its extent in seafood than previously thought. and/or duration. Physical aggregation due the interaction As an unarmored dinoflagellate, K. brevis is delicate of complex hydrography and the swimming behavior of K. compared to most toxic dinoflagellate species. As a result, brevis is probably also important. Olascoaga et al. [42]have cells can be broken apart by turbulence due to wave argued that areas of low mixing due to certain hydrographic action at the sea surface and along beaches. This results features along the west coast of Florida allow the slow in aerosolization of brevetoxins that may be inhaled by growing K. brevis to accumulate large populations without humans and resulting in respiratory distress, coughing, being dispersed by mixing and dilution rates larger than its and eye irritation in humans [57–60]. Hospital emergency growth rate. room admittances for respiratory distress increase 50% A variety of hypotheses have been proposed for the when blooms of K. brevis occur [61] and is particularly sources of nutrients that would ultimately influence the acute in people with asthma [59, 62]. Beaches along concentrations and spatial extent of K. brevis blooms. Gunter the west coast of Florida are major recreational areas et al. [43], Rounsefell and Nelson [44], Dixon and Steidinger for Florida residents, including many elderly retirees and [45], and Brand and Compton [40] have argued that tourists. When large blooms of K. brevis develop, dead fish nutrients in land runoff could be an important source. Lenes wash up on the beaches and/or brevetoxin aerosolization et al. [46], Walsh and Steidinger [47], and Walsh et al. [48] occurs and the tourism suffers considerable economic loss have argued that iron-rich atmospheric dust stimulation of [63]. nitrogen-fixing Trichodesmium in phosphorus-rich waters is an important nutrient source that can stimulate blooms 4. BMAA-Containing Cyanobacteria of K. brevis. Hu et al. [49] have argued that nutrient-rich groundwater could be important. Vargo et al. [50]have Cyanobacteria have been a part of the human diet pri- argued that, along with other sources, fish that have died in marily in non-Western civilizations for centuries. Today, 4 International Journal of Microbiology cyanobacteria are produced in mass, controlled cultivation Rao et al. [77] have demonstrated that in the presence processes and/or harvested from natural habitats and mar- of 5% CO2,BMAAcausesselectivemotorneuronloss keted as food supplements around the world. Historical and in dissociated mixed spinal cord cultures at concentra- current uses of cyanobacteria and their derivative products tions of ∼30 μM. These investigators also reported that are thoroughly reviewed by [64]. The amino acid β-N- the glutamate receptor antagonist 2,3-Dioxo-6-nitro-1,2,3,4- methylamino-L-alanine (BMAA) is one of a number of bio- tetrahydrobenzo[f]quinoxaline-7-sulfonamide (NBQX) pre- logically active natural compounds produced by cyanobac- vented BMAA-induced death, implicating BMAA in the teria, and its potential importance in species of marine excitotoxic activation of receptors for α-amino-3-hydroxy-5- cyanobacteria has attracted considerable research attention methyl-4-isoxazolepropionic acid (AMPA) and kainic acid, recently. and that BMAA selectively induced the production of BMAA was originally discovered in cycad seeds [67], reactive oxygen species (ROS) in motor neurons. More and later implicated the etiology of Amyotrophic Lateral recently, Lobner et al. [78] demonstrated that BMAA also Sclerosis/Parkinsonism Dementia Complex (ALS/PDC) that functions as an agonist for the NMDA and mGluR5 receptors occurs among the Chamorro people of Guam [68]; see in mouse cortical cell cultures. In addition to AMPA/kainate, TemaNord [69] for an exhaustive review on this subject. NMDA and mGluR5 receptor activation and oxidative stress, BMAA is biomagnified in the Guam ecosystem and occurs BMAA may also induce toxicity by inhibiting the reuptake in the brain tissues of Chamorros who died of ALS/PDC of glutamate. Furthermore, BMAA may be misidentified [70]. Axenic cultures of the endosymbiont cyanobacterium, by transfer RNAs, resulting in its misincorporation into Nostoc sp., isolated from the coralloid roots of the cycad proteins. palm, Cycas micronesica Hill, produce free BMAA at a Since the initial report of the widespread distribution concentration of 0.3 μg g−1. Following up on this work, of BMAA in representatives of all five cyanobacteria mor- Cox et al. [71] examined BMAA production in free- photypes by Cox et al. [71], there have been a number living and symbiotic clones representing the five morpho- of conflicting studies published regarding the detection types of cyanobacteria (cf. Rippka et al. [72].) A wide and quantification of BMAA in cyanobacteria (including range of cyanobacterial strains were screened, including blue green algae nutritional supplements). Eleven studies strains maintained at the University of Dundee in Scot- have confirmed the presence of BMAA in a wide range of land, Stockholm University in Sweden, and the University marine, brackish, and freshwater cyanobacteria via LC/MS, of Hawaii in the United States. Liquid chromatography- LC/MS/MS or GC/MS methods that identify and detect mass spectrometry (LC/MS) and high-performance liquid underivatized [79, 80]andderivatized[75, 81–88] BMAA. By chromatography (HPLC) were employed to identify and comparison, four studies were not able to detect the presence quantify free and protein-associated BMAA for each sample. of BMAA in a wide range of cyanobacteria via HPLC, For free-living cyanobacteria, Cox et al. [71] found that LC/MS, or LC/MS/MS that identify and detect derivatized BMAA was produced by members of all five cyanobacterial [89]andunderivatizedBMAA[90–92]. These disparate morphotypes as well as 95% of the genera and 97% of findings are likely caused by analyte misidentification [91, the strains that were screened. Analysis of Nostoc strains 92] and/or differences in methodological sensitivities [88]. isolated from symbiotic relationships with fungi and host Because of potential coelution artifacts, we recommend plants of broad taxonomic diversity indicated that 73% of that that BMAA identification and quantification be based these strains produced BMAA. The ubiquity of cyanobacteria on the LC/MS analysis of BMAA-specific fragments (m/z in diverse terrestrial and aquatic environments suggests that 88) [91] or AQC-derivatized BMAA (m/z 258) [88]in ingestion of BMAA may occur through even less esoteric order to minimize the possibility of reporting false positive routes, including direct consumption of cyanobacteria or data [88, 91]. Based on this criterion, only four studies at cyanobacterial hosts, bioaccumulation in additional food the time of this writing have confirmed the presence of chains, or exposure to cyanobacteria-contaminated water BMAA in cyanobacteria [79, 80, 87, 88]. Due to the serious supplies.Coxetal.[71] recommended that BMAA concen- implications of BMAA and neurodegenerative disease, it is trations should be monitored in invertebrates, fish, and/or further recommended that NMR analysis be used for the grazing animals used for human consumption that either unequivocal identification of BMAA in biological samples. directly consume cyanobacteria or forage on plants or prey It should be noted that two studies employing LC/MS that may have accumulated cyanobacteria-produced BMAA. analysis of BMAA-specific fragments failed to detect BMAA This conclusion was reinforced by subsequent articles pub- in a wide range of cyanobacteria samples [91, 92]. We lished in the Journal of the American Medical Association conclude that BMAA was either absent or below the limit [73, 74] and Neuropathology and Applied Neurobiology of detection in these samples. Spa´cilˇ et al. [88] recommend [75]. subjecting samples to a pretreatment protocol to both The amino acid, β-N-methylamino-L-alanine (BMAA), remove impurities and to concentrate BMAA prior to LC/MS is an excitotoxic neurotoxin that functions as a glutamate analysis. Finally, we recommend that future studies should agonist. By virtue of its unique structural characteristics, monitor BMAA concentrations using only BMAA-specific BMMA reacts with CO2 atphysiologicalpHtoform LC/MS methods in animals used for human consumption α-andβ-carbamate adducts [65, 66, 76] that are struc- that either directly consume cyanobacteria or forage on turally similar to the neurotransmitter glutamate and its plants or prey that may have accumulated cyanobacteria- selective agonist N-methyl-d-aspartate (NMDA) (Figure 1). produced BMAA. International Journal of Microbiology 5

O O O

−H+ −H+ − + − HN O + − H2N O H2N O + + CH3 NH3 CH3 NH3 CH3 NH 2 β-N-methylamino-L-alanine β-N-methylamino-L-alanine β-N-methylamino-L-alanine (β-deprotonated) (α-deprotonated) (L-BMAA) O CO2 CO2

+ − H2N O O O

NH CH3 −O N O−

−O O + CH3 NH3 α-N-carboxy-β-methylamino-L-alanine β-(N-carboxy-N-methyl)-amino-L-alanine (BMAA-α-NCO2) (BMAA-β-NCO2)

Figure 1: Reaction of the α-andβ-amino groups of BMAA with CO2 to form the corresponding carbamate adducts (after [65, 66]).

5. Alexandrium fundyense and Paralytic with PSP outbreaks, and whether or not effective monitoring Shellfish Poisoning programs are in place to prevent contaminated shellfish from entering commercial markets. However, despite warning Paralytic shellfish poisoning (PSP) has been recognized in signs and other outreach efforts, recreational harvesters still the Pacific Northwest of the United States for centuries [93]. become victims of PSP. For example, in June 2010, five cases Human poisonings have been recorded primarily in North of suspected PSP, including two fatalities, were reported in America, Asia, and Europe [94], but outbreaks have been Anchorage Daily News due to shellfish collected from waters reported worldwide [93, 95, 96]. PSP is caused by eating in Alaska, US. The victims had eaten personally harvested bivalve shellfish (clams, mussels, scallops, etc.) contaminated shellfish and crabs, including those from areas normally with one or more of a group of structurally related congeners avoided because of historically high levels of contamination. of saxitoxin [97]. Filter-feeding fish can sometimes be vectors The causative organism in New England PSP outbreaks for the toxins as well. is Alexandrium fundyense. Although the Bay of Fundy and Saxitoxins are produced by dinoflagellates of the genera northeastern Canadian waters have a long history of PSP, in Gymnodinium [101], Alexandrium [95, 96], and Pyrodinium the U.S., toxicity was restricted to far-eastern Maine (ME) [97, 101]. These toxins act to selectively block the voltage- until 1972, when a massive, visible red tide of A. fundyense gated sodium channel of excitable membranes, thus blocking stretched from ME to Massachusetts (MA), causing toxicity the generation and propagation of action potentials in nerve in some southern areas for the first time. Virtually every axons and skeletal muscle fibers. Mammals, birds, and fish year since 1972, western ME has experienced PSP outbreaks, can be affected by PSP toxins; however, humans are the most and for the first 20 years of that interval, MA did as well. sensitive—the fatal oral dose of saxitoxin is 1–4 mg [102]. That pattern was a direct result of A. fundyense cysts being PSP symptoms begin to occur within 30 minutes to retained in western GOM waters after the 1972 bloom and three hours of eating contaminated seafood. The initial subsequent events [103]. Between 1994 and 2004, toxicity symptoms include paresthesia and numbness around the lips was infrequent in MA and the southern GOM. Then in 2005, and mouth [93]. These sensations then spread to the face and another massive bloom occurred [98], leading to closure of neck. Victims may also experience nausea and vomiting. In shellfish beds from ME to southern MA and 40,000 km2 of moderately severe poisonings, paresthesia progresses to the offshore federal waters as well. Economic losses in 2005 were arms and legs. Victims may experience giddiness, incoherent estimated to be $50 million for the MA shellfish industry speech, and light-headedness. In severe poisonings, death alone. can result from respiratory failure and hypoxia. Historically, A. fundyense has a complex life cycle includes a resting the fatality rate from PSP varies from no deaths in recent cyst, a phase of vegetative growth, sexual reproduction, and outbreaks in the U.S. or Europe to rates of 2%–14% in other re-encystment (Figure 2(a)). Observations indicate several parts of the world [93]. The frequency of mortalities is related salient characteristics of the vegetative cell distributions: to the availability of emergency hospital care, past experience patterns of abundance are gulf-wide in geographic scope; 6 International Journal of Microbiology

3: If conditions remain Vegetative optimal, cells will continue to ME cells divide, reproducing 45 fundy exponentially, 2 to 4 to 8. A 45 N f single cell could produce several Penobscot Bay o hundred cells within a couple of bay weeks. If other single cells E reproduce similarly, then Casco toxicity in shellfish may occur. 44 bay m 20 10 44 00 200 0 1 0m 100 1 Jordan m 3. 300 Basin Gametes 60 J Latitude NH 2 43 00 m 4 and 5: When nutrients 4. N 43 2 are gone, growth stops Cape and gametes are formed. ann Two gametes join to form M 300 300 S 0m one cell, which develops 2: Cysts can germinate only MA 20 into a zygote and then during certain times of the 42 42 Cape 100 m into a cyst. This falls to year with warmer Cod 3 the ocean bottom and G temperatures and 200 60 0m is capable of 00 increased light N 1 germination the stimulating germination. 100 following year. The cyst breaks open, and 2. 41 a swimming cell emerges. −70 −68 −66 71 70 69 68 67 66 Zygote 5. The cell reproduces by simple division within a few days of "hatching." Longitude W 1: Alexandrium resting cysts lay dormant on the ocean floor, buried in sediment. If 1. undisturbed by physical or natural forces, they 0 100 200 300 400 500 600 can stay in this state for years. If oxygen is present, cyst − germination may proceed if conditions are right. Cysts cm 3 (a) (b) (c)

Bay of fundy 45 May 45 June Penobscot bay 44.5 44 Casco bay 44 43.5 x

Latitude 43 43 Latitude 42.5 42 42 41.5

−70 −68 −66 −70 −68 −66 Longitude Longitude

45 July 45 August

44 44 x x

Latitude 43 Latitude 43

42 42

−70 −68 −66 −70 −68 −66 Longitude Longitude

01010010000 10 100 1000 Cells/liter Cells/liter (d)

Figure 2: Upper left: life cycle of A. fundyense. Upper middle: distribution of cysts (number of cysts cm−3) in the upper 3cm of sediment derived from a 1997 survey of the Gulf of Maine [98] and surveys of the Bay of Fundy in 1981 (White and Lewis, 1982), 1982, and 1983 (data provided by Jennifer Martin, DFO). Upper right: schematic of the Maine Coastal Current, reprinted from McGillicuddy et al. [99]. Branch points are located at (1) Penobscot Bay, (2) Cape Ann, and (3) Great South Channel. Seven segments of the current are indicated: (E)astern, (J)ordan, (W)estern, (M)assachusetts, (S)tellwagen, (N)antucket, and (G)eorges Bank. Dashed white lines denote state boundaries of Maine (ME), New Hampshire (NH), and Massachusetts (MA). From Backer and McGillicuddy [100]. Reprinted with permission from Oceanography. International Journal of Microbiology 7 the distributions are associated with the Maine Coastal inoculation and development of the bloom. At this point, it Current, and the center of mass of the distribution is from is not yet possible to model the formation and deposition west to east during the April-to-August growing season of new cysts from these blooms, though work is ongoing in [104]. This latter aspect is particularly notable given the this direction. This is an area where our lack of knowledge is coastal current flows in the opposite direction (Figure 2(b)). evident—the termination of blooms remains poorly under- A model based on the seasonal mean flow that includes stood, and in particular, the relationship between bloom size germination, growth, mortality, and nutrient limitation can and the size of the resulting cyst seedbed is not established, produce simulations that are qualitatively consistent with the nor indeed is it intuitive. For example, some of the largest observations (Figure 2(c);[99]. In general, cells germinated regional blooms (e.g., in 2005 [98]) were followed by very from the major cyst beds in the Bay of Fundy and near low cyst abundance on a regional basis whereas more modest Penobscot and Casco Bays (Figure 2(d))areadvectedin bloom years (e.g., 2007) led to widespread and high density the alongshore direction from east to west in the coastal cyst accumulations {D.M. Anderson, unpub. data}. current. Growth of the vegetative cells is limited primarily Modeling results to date suggest that simulations initi- by temperature from April through June throughout the ated from A. fundyense cyst distributions can capture large- gulf whereas nutrient limitation occurs in July and August scale seasonal patterns in the distribution and abundance of in the western gulf. Thus, the seasonal shift in the center vegetative cells. To the extent that cyst abundance is a first- of mass of cells from west to east can be explained by order predictor of regional bloom magnitude the following changing growth conditions: growth is more rapid in the year in the WGOM (even though the converse is not true), western gulf early in the season due to warmer temperatures that information can be used in a seasonal forecast of PSP whereas growth is more rapid in the eastern gulf later in the on a regional basis. Near-real-time nowcasts and forecasts of season due to severe nutrient limitation in the western gulf harmful algal blooms (HABs) in the Gulf of Maine have been during that time period. Hydrodynamic transport of these run routinely each year since 2006 (2006: http://science.whoi offshore populations to inshore shellfish beds is a key aspect .edu/users/ruoying/Redtide 06/, 2007: http://omgrhe.meas regulating the PSP threat to human health [100, 105, 106]. .ncsu.edu/Redtide/Redtide 07/, 2008: http://omglnx3.meas In the wake of the historic bloom of 2005 in the .ncsu.edu/yli/08forecast/, 2009: http://omglnx3.meas.ncsu western GOM, a suite of models was used to diagnose .edu/yli/09forecast/). During the bloom season, weekly the underlying causes. Anderson et al. [98] described three updates have been made available to more than 150 managers factors to explain the 2005 bloom: (1) high abundance of and other officials and scientists involved with PSP outbreaks resting cysts in fall 2004 that provided a large inoculums, (2) in the northeastern US. Web interfaces provide the latest storms with strong northeast winds that carried toxic cells model simulations, with one-week forecasts driven by mete- towards and along the coast, and (3) abundant fresh water orological predictions. At present, this kind of early warning runoff, providing macro- and micronutrients, a stratified appears to be the most practical approach to mitigating water column, and an alongshore (towards the southwest) the impacts of these blooms, insofar as available direct transport mechanism. These factors were evaluated using intervention strategies are not practical by virtue of the fact a sensitivity analysis that utilized field observations and that even in bloom conditions, A. fundyense is typically a amodelofA. fundyense population dynamics [99, 107, small fraction of the total phytoplankton biomass. 108], coupled to a regional circulation model to hindcast the 2005 bloom [109, 110]. Initial conditions of the three 6. Infectious Microbes sensitivity experiments are identical to the central hindcast (an animated version of the central hindcast is available Waterborne infectious microbes normally include viruses, at http://science.whoi.edu/users/ruoying/Redtide 05/Papers/ bacteria, and protozoa that can be transmitted through avg fields.avi) in all respects except: experiment 1 utilizes recreational exposure to seawater or consumption of seafood. the 1997 cyst map instead of 2004; experiment 2 is forced Other groups of infectious microbes that are less commonly by winds from a more typical year (2004) instead of the considered include the helminthes and yeasts. The infectious strong downwelling-favorable winds of 2005; experiment 3 microbes differ from harmful algal species in that disease uses riverine discharge from a typical year (2004) instead is caused by the growth of the microbes within humans. of the anomalously large discharge of 2005. This sensitivity Thus, exposure to even low levels of infectious microbes analysis suggests that high cyst abundance in the WGOM can cause illness. Upon consumption, inhalation, or contact was the main cause of the 2005 bloom. Wind forcing with the infectious microbe it then multiplies within the was an important regulator, in the form of both episodic gastrointestinal system, respiratory tract, or within exposed bursts of northeast winds and the downwelling-favorable skin resulting in human disease. Harmful algae, on the mean condition, causing onshore advection of offshore other hand, grow outside humans within external water populations. The anomalously high river runoff enhanced bodies and release toxins that cause disease when the alongshore transport near the coast, but had limited impact toxin-contaminated water is ingested or inhaled [62, 111]. on the gulf-wide bloom distribution. Typically, the numbers of infectious microbes that are needed Model initial conditions are dependent on maps of A. to cause disease are low; the precise number depends fundyense cyst abundance obtained on an annual basis. upon the virulence of the particular strain of the microbe Mathematical representations of laboratory-derived germi- [112] and also the immune status of the infected host. nation and growth data are used with these maps to drive the Riskstohumansfrominfectiousmicrobesareevaluated 8 International Journal of Microbiology by considering human exposure to seawater or seafood in in the environment. These studies have generally highlighted addition to the concentration of infectious microbes therein. the importance of pathogens associated with sediments [129] Infectious microbes in seawater can be separated into two and show a positive correlation between the occurrence of groups, those which are introduced from outside sources pathogens and rainfall [130]. More research is clearly needed and those which are indigenous, hereafter referred to as to understand the dynamics of pathogens once release to the allochthonous and autochthonous, respectively [113]. environment. Infectious microbes in coastal waters impact the health Vibrio spp. represent the classic example of indigenous of a large number of people globally. Worldwide, up to 170 or autochthonous infectious microbes [144]. Other notable million enteric and respiratory illnesses attributed to swim- autochthonous infectious microbes include helminthes ming in and consuming shellfish from infectious microbes which are indigenous to fish populations which can be in coastal waters [114]. In the US, 20,300 recreational beach transmitted to humans via ingestion of undercooked fish advisories were reported in 2008 due to contamination with [145] and amoeba which can enter the nasal cavity [118]. fecal bacteria, up from 6,200 in 1999 [115]. In the US, Common disease-causing Vibrio spp. include V. cholerae 33% of shellfish harvesting waters are impaired by infectious (toxigenic and nontoxigenic), V. mimicus, V. parahaemoly t i- microbes [116]. In southern California alone, it is estimated cus, V. vulnificus,andV. alg inoly t icus. V. cholerae and V. that 1.5 million excess enteric illnesses occur per year from mimicus are closely related species, both of which cause swimming in waters with infectious microbes, at a cost of gastroenteritis, V. parahaemoly t icus causes gastroenteritis as $50 million per year [117]. One of the challenges of assessing well as wound infections, V. vulnificus can cause septis and the impact of infectious microbes on human health is that wound infections, and V. alg inoly t icus can cause wound most illnesses associated with these infectious microbes are infections [119].Anumberofstudieshaveexaminedthe self-limiting, so medical advice is not always sought. In environmental factors that control the occurrence of Vibrio addition, identifying the etiologies of the illnesses can be spp. in coastal waters or within shellfish [146–149]. Common challenging even in the most modern diagnostic laboratories. correlates to Vibrio concentrations include salinity and Most of the illnesses are also not reportable, so they are temperature; warmer temperatures typically correlate to not tracked by a central agency. Yoder et al. [118]report higher Vibrio concentrations; salinity and Vibrio densities Cryptosporidium as the most common etiology of freshwater covary, but the direction depends on the organism. Vibrio recreational waterborne illness, and also reports Vibrio spp. have been shown in some cases to adsorb to zooplankton or as an important etiology for seawater-acquired recreational phytoplankton and be associated with sediment [146, 150, waterborne illness. A review of the epidemiology of seafood- 151]. Further information in V. cholerae, V. parahaemoly t icus, associated illness in the United States between 1973 and 2006 and V. vulnificus can be found in excellent reviews on the [119] identifies Vibrio parahaemolyticus as being responsible organisms [152–154]. for the most seafood illnesses (35%), norovirus and hepatitis In order to protect human health and provide warnings A together responsible for the second highest number for unsafe conditions, the presence of infectious microbes of illnesses (32%), and Salmonella and Shigella together in coastal waters is evaluated through the use of “indicator” responsible for the third highest number of illnesses (19%). microbes. Indicator microbes are commensal inhabitants of Introduced or allochthonous infectious microbes can the gastrointestinal tract of humans and are present in large come from human sewage, stormwater, feces of animals, and numbers in fecal releases, especially releases from humans skin from infected humans during bathing. Allochthonous and warm blooded animals [155]. Indicator microbes are infectious microbes (from the virus, bacteria, and protozoan not necessarily pathogenic but are used as a surrogate for groups) include enterovirus, norovirus, Salmonella, Campy- the presence of pathogenic microbes. For marine recreational lobacter, Shigella, Cryptosporidium, Giardia, Legionella sp. waters, enterococci are the indicators recommended by the and Staphylococcus aureus. All of these microbes cause U.S. Environmental Protection Agency [156]. For freshwater gastrointestinal disease with the exception of Legionella sp., recreational waters, both enterococci or E. coli are recom- which causes respiratory disease and S. aureus which causes mended [156]. The use of these microbes as indicators is skin disease. Transmission of helminthes via recreational supported by their correlation to measured, adverse, human contact with seawater is generally limited to developing health outcomes during epidemiology studies focused on countries and includes incidental ingestion of infective eggs illness during exposure to waters impacted by point sources or through contact with contaminated waters for forms that of treated wastewater [157]. For shellfisheries in the US, can penetrate skin [120]. In addition to infectious microbes fecal coliform are recommended to assess risk of exposure from the virus, bacteria, and protozoan groups [121], coastal to pathogens through shellfish consumption [158]. beach sands have also been implicated as a means of Five basic dilemmas are associated with the use of indi- potentially transmitting pathogenic helminthes [122]and cator microbes to establish the safety of recreational waters. yeasts [123]. There are a number of excellent reviews on the First, as enterococci and E. coli are natural inhabitants of the occurrence of allochthonous infectious microbes, particu- digestive tract of humans, the disease endpoint associated larly bacteria, viruses, and protozoa, in coastal waters [124– with these microbes is gastrointestinal disease. Although, 126] and in shellfish [127, 128]. Table 1 provides examples gastrointestinal illness can be transmitted through water use of studies that report allochthonous pathogen occurrence in [159], other types of illnesses can be transmitted during coastal waters. There are a limited number of studies that swimming including ear-, eye-, respiratory-, and skin-related describe the fate and transport of allochthonous pathogens diseases [160–162]. Second, studies have shown that in many International Journal of Microbiology 9

Table 1: Examples of allothchonous human pathogens detected in coastal waters.

Viruses Concentration/Occurrence Reference Present in 9 of 72 1-liter samples using RT-PCR at Avalon Beach, CA∗. Present in 1 of 18, 220-liter enteroviruses [131, 132] samples using culture based methods for brackish water in St. Lucie River Estuary, FL. adenoviruses Present in 15 of 30 250-liter samples using PCR at Silver Beach, MI∗ [133] hepatitis A 105 to 30,771 viral particles/liter using Q-RTPCR at Imperial Beach, CA∗ [134] norovirus 2 of 19 samples in 110-liters using RT-PCR at Key West sites (FL)∗ [135] rotavirus (reovirus) 2 of 19 sites with 2–5 MPN/L at Italian coastline [136] Bacteria Detected in 25 of 192 100–1000 mL Spanish marine recreational water samples using culture based Campylobacter [137] methods Detected in 70%–100% of samples from a lagoon in Brazil using culture-based methods, volume Salmonella [138] assayed not reported 60%–70% of approx. 100 mL seawater samples from Doheny and Avalon Beach, CA using Staphylococcus culture-based methods. 37% of 668, 50 mL seawater samples from Hobie Cat Beach, FL using [139, 140] culture-based methods and confirmation by PCR 2 of 377 E. coli isolates from North Carolina and Southern California coastal waters using combined Pathogenic E. coli [141] culture and PCR methods Shigella 100% of algal mat samples from Lake Michigan near Burns Ditch by PCR [142] Legionella sp. Found in 35 of 72 samples from Lake Pontchartrain with 1 of 72 positive for L. pneumophila [140] Protozoa Cryptosporidium 13.7 ± 1.7 oocysts/L on weekends at Chesapeake Bay beach, MD [143] Giardia 9.1 ± 1.1 cysts/L on weekends at Chesapeake Bay beach, MD [143] ∗ Volumes reported do not account for the fact that a fraction of water sample was used during polymerase chain reaction (PCR), reverse-transcriptase- (RT-) PCR, or quantitative (Q) PCR.

cases indicator microbes do not track pathogenic microbes waters. There is some evidence that there is a human health on a one-to-one basis [163, 164]. Ortega et al. [131]have risk upon exposure to indicators from nonpoint pollution shown that sites subject to sporadic increases in indicator sources [167], but there still exists a great deal of debate on levels are also characterized by detectable levels of pathogens, the topic [132]. Finally, fecal indicator bacteria cannot be although the time that the pathogens are detected do not used to protect individuals from exposure to autochthonous necessarily coincide with the time that the indicator microbe pathogens, like Vibrio. Significant risk of Vibrio infections levels exceed regulatory guideline levels. Abdelzaher et al. could be present when there are no fecal bacteria. [165] found that indicators and pathogens are generally In the future, safety from infectious microbes in recre- elevated during similar environmental conditions (low solar ational waters and in seafood should consider multiple radiation, after rainfall, and during a particular tidal period), lines of evidence where indicator microbe measurements but this correspondence was not always consistent. The third are supplemented with direct measures of a cluster of dilemma associated with the use of indicator microbes is the pathogenic targets that are relevant to the pollution sources time required to measure indicator bacteria using traditional and pathogens affecting the study area [168]. In addition, culture methods. Traditional culture methods require an 18 more work should be done to understand the ecology to 24 hours incubation period before detection of the bacte- of allochthonous infectious microbes in the environment. ria.Asaresult,contaminatedbeachescanremainopenfora Although these organisms are historically viewed as transient significant period of time before levels are known, thereby members of the microbial community of coastal waters, resulting in exposures to human populations. Conversely, the coastal environment may serve as an important niche as contamination tends to be highly variable [166], beaches for these organisms to persist, exchange genetic material can also be closed during times when they are safe. The (for bacteria), and grow. Indeed, some researchers have fourth dilemma is that the use of E. coli and enterococci found allochthonous pathogenic bacteria to be widespread in to assess risk was established using data from epidemiology macroalgae of the Great Lakes [142]. Research is also needed studies conducted at beaches polluted by point sources of to develop more rapid analyses methods for both indicators treated wastewater. At the present time, most point sources and pathogens [169], so that the time between measures of pollution in the developed world are well regulated and and warnings can be reduced. Method detection limits also controlled to minimize human health impacts. Nonpoint need to be improved if direct pathogen presence is to be sources, including urban and agricultural runoff, wildlife considered for future monitoring purposes, as pathogens feces, bather shedding, and other “environmental reservoirs” are typically present in very low numbers [169]. Currently, are leading contributors of E. coli and enterococci to coastal the procedures for measuring pathogens in environmental 10 International Journal of Microbiology waters are time consuming, in part due to detection limit [12] L. E. Fleming, D. Katz, J. A. Bean, and R. Hammond, issues, and thus data on the environmental occurrence of “Epidemiology of seafood poisoning,” in Foodborne Disease infectious microbes is lacking. Handbook, Y. H. Hui, D. Kitts, and P. Standfield, Eds., pp. Ideally, the public should be warned prior to adverse 287–310, 2001. water quality events. Given the current limitations in [13]L.E.Fleming,D.G.Baden,J.A.Bean,R.Weisman,andD. microbial measurements with respect to labor and time, G. Blythe, “Seafood toxin diseases: issues in epidemiology early warning systems will likely rely on models [170, and community outreach,” in Harmful Algae. Zunta de 171] designed to predict health risks based upon read- Galicia and Intergovernmental Oceanographic Commission of ily measureable environmental parameters and impending UNESCO, B. Reguera, J. Blanco, M. L. Fernandez, and T. Wyatt, Eds., pp. 245–248, 1998. environmental conditions. Such models will require direct measures of microbial water quality for calibration and [14] J. Cameron, A. E. Flowers, and M. F. Capra, “Electrophys- iological studies on ciguatera poisoning in man (Part II),” verification purposes. Journal of the Neurological Sciences, vol. 101, no. 1, pp. 93– 97, 1991. Acknowledgments [15] E. Benoit, P. Juzans, A. M. Legrand, and J. Molgo, “Reversal of Pacific ciguatoxin-1B effects on myelinated axons by agents This paper was made possible through the Centers used in ciguatera treatment,” Cybium,vol.24,no.3,pp.33– for Oceans and Human Health (COHH) program of 40, 2000. the National Institute of Environmental Health Sciences [16] M. A. Friedman, L. E. Fleming, M. Fernandez et al., “Ciguat- (National Institutes of Health), and National Science Foun- era fish poisoning: treatment, prevention and management,” dation that provided Grants nos. P50ES012740, OCE04- Marine Drugs, vol. 6, no. 3, pp. 456–479, 2008. 32479, and OCE09-11000 to the University of Hawaii, [17] T. Yasumoto, I. Nakajima, R. Bagnis, and R. Adachi, “Finding Grants nos OCE0432368/0911373 and P50 ES12736 to the of a dinoflagellate as a likely culprit of ciguatera,” Bulletin of University of Miami, and Grants nos. OCE-0430724 and P50 the Japanese Society for the Science of Fish, vol. 43, pp. 1021– ES012742 to the Woods Hole Oceanographic Institution. 1026, 1977. [18] T. Yasumoto, A. Inoue, and R. Bagnis, “Ecological survey References of a toxic dinoflagellate associated with ciguatera,” in Toxic Dinoflagellate Blooms, D. L. Taylor and H. Seliger, Eds., pp. [1] P. K. Bienfang, M. L. Parsons, R. R. Bidigare, E. A. Laws, and 221–224, Elsevier, New York, Ny, USA, 1979. P. D. R. Moeller, “Ciguatera fish poisoning: a synopsis from [19] T. Yasumoto, Y. Oshima, Y. Murakami, I. Nakajima, R. ecology to toxicity,” in Oceans and Human Health: Risks and Bagnis, and Y. Fukuyo, “Toxicity of benthic dinoflagellates,” Remedies from the Sea,P.J.Walsh,S.L.Smith,L.E.Fleming, Bulletin of the Japanese Society for the Science of Fish , vol. 46, H. Solo-Gabriele, and W. H. Gerwick, Eds., pp. 257–270, pp. 327–331, 1980. Elsevier, New York, NY, USA, 2008. [20] R. Adachi and Y. Fukuyo, “The thecal structure of a marine [2] R. W. Dickey, “Ciguatera toxins: chemistry, toxicology and toxic dinoflagellate Gambierdiscus toxicus gen. et sp. nov. detection,” in Seafood and Freshwater Toxins: Pharmacology, collected in a ciguatera endemic area,” Bulletin of the Japanese Physiology, and Detection, L. M. Botana, Ed., pp. 479–500, Society for the Science of Fish , vol. 45, pp. 67–71, 1979. CRC Press, Boca Raton, Fla, USA; Taylor & Francis, London, UK, 2nd edition, 2008. [21] M. A. Faust, “Observation of sand-dwelling toxic dinoflag- ellates (Dinophyceae) from widely differing sites, including [3] R. W. Dickey and S. M. Plakas, “Ciguatera: a public health two new species,” Journal of Phycology, vol. 31, no. 6, pp. 996– perspective,” Toxicon, vol. 56, no. 2, pp. 123–136, 2010. 1003, 1995. [4] P. Helfrich, “Fish poisoning in Hawaii,” Hawaii Medical [22] M. J. Holmes, “Gambierdiscus yasumotoi sp.nov.(Dino- Journal, vol. 22, pp. 361–372, 1964. phyceae), a toxic benthic dinoflagellate from Southeastern [5] B. W. Halstead, Poisonous and Venomous Marine Animals of Asia,” Journal of Phycology, vol. 34, no. 4, pp. 661–668, 1998. the World, vol. 2, U. S. Government Printing Office, 1967. [23] M. Chinain, M. A. Faust, and S. Pauillac, “Morphology and [6] J. P. Quod and J. Turquet, “Ciguatera in Reunion Island (SW molecular analyses of three toxic species of Gambierdiscus Indian ocean): epidemiology and clinical patterns,” Toxicon, (Dinophyceae): G. pacificus, sp. nov., G. australes, sp. nov., vol. 34, no. 7, pp. 779–785, 1996. and G. polynesiensis, sp. nov,” Journal of Phycology, vol. 35, [7] R. J. Lewis, “The changing face of ciguatera,” Toxicon, vol. 39, no. 6, pp. 1282–1296, 1999. no. 1, pp. 97–106, 2001. [24] M. L. Richlen, S. L. Morton, P. H. Barber, and P. S. Lobel, [8] J. Pearn, “Neurology of ciguatera,” Journal of Neurology, vol. “Phylogeography, morphological variation and taxonomy 70, no. 1, pp. 4–8, 2001. of the toxic dinoflagellate Gambierdiscus toxicus (Dino- [9] P. Arena, B. Levin, L. E. Fleming, M. A. Friedman, and D. phyceae),” Harmful Algae, vol. 7, no. 5, pp. 614–629, 2008. Blythe, “A pilot study of the cognitive and psychological [25] R. W. Litaker, M. W. Vandersea, M. A. Faust et al., “Taxonomy correlates of chronic ciguatera poisoning,” Harmful Algae, of Gambierdiscus including four new species, Gambierdiscus vol. 3, no. 1, pp. 51–60, 2004. caribaeus, Gambierdiscus carolinianus, Gambierdiscus car- [10] N. D. Lewis, “Disease and development: coguatera fish penteri and Gambierdiscus ruetzleri (Gonyaulacales, Dino- poisoning,” Social Science and Medicine, vol. 23, no. 10, pp. phyceae),” Phycologia, vol. 48, no. 5, pp. 344–390, 2009. 983–993, 1986. [26] K. A. Steidinger and D. G. Baden, “Toxic marine dinoflag- [11] W. R. Lange, “Ciguatera fish poisoning,” American Family ellates,” in Dinoflagellates, D. L. Spector, Ed., pp. 210–261, Physician, vol. 50, no. 3, pp. 579–584, 1994. Academic Press, Orlando, Fla, USA, 1985. International Journal of Microbiology 11

[27] H. Kite-Powell, “Public health and economic implications of [43]G.Gunter,R.H.Williams,C.C.Davis,andF.G.Walton oceans and human health research,” in Proceedings of Oceans Smith, “Catastrophic mass mortality of marine animals and Human Health Symposium & Workshop, Washington, and coincident phytoplankton bloom on the west coast DC, USA, April 2010. of Florida, November 1946-August 1947,” Ecological Mono- [28] M.-L. Chateau-Degat, M. Chinain, N. Cerf, S. Gingras, B. graphs, vol. 18, pp. 309–324, 1947. Hubert, and E.´ Dewailly, “Seawater temperature, Gambierdis- [44] G. A. Rounsefell and W. R. Nelson, “Red-tide research cus spp. variability and incidence of ciguatera poisoning in summarized to 1964 including an annotated bibliography,” French Polynesia,” Harmful Algae, vol. 4, no. 6, pp. 1053– U.S. Fish and Wildlife. Service Special Scientific Report— 1062, 2005. Fisheries, no. 535, p. 85, 1966. [29] J. L. Perez-Arellano, O. P. Luzardo, A. Perez Brito et al., [45] L. K. Dixon and K. A. Steidinger, “Correlation of Karenia “First report of ciguatera fish poisoning in West-African brevis presence in the eastern Gulf of Mexico and with rainfall coast: a public health concern in the Canary Islands (Spain),” and riverine flow,” in Harmful Algae, K. A. Steidinger, J. H. Emerging Infectious Diseases, vol. 11, pp. 1981–1982, 2005. Landsberg, C. R. Tomas, and G. A. Vargo, Eds., pp. 29–31, [30] T. A. Villareal, S. Hanson, S. Qualia, E. L. E. Jester, H. 2004. R. Granade, and R. W. Dickey, “Petroleum production [46] J. M. Lenes, B. P. Darrow, C. Cattrall et al., “Iron fertilization platforms as sites for the expansion of ciguatera in the and the Trichodesmium response on the West Florida shelf,” northwestern Gulf of Mexico,” Harmful Algae,vol.6,no.2, Limnology and Oceanography, vol. 46, no. 6, pp. 1261–1277, pp. 253–259, 2007. 2001. [31] P. Bienfang, B. Oben, S. Defelice et al., “Ciguatera: the [47] J. J. Walsh and K. A. Steidinger, “Saharan dust and Florida detection of neurotoxins in carnivorous reef fish from the red tides: the cyanophyte connection,” Journal of Geophysical coast of Cameroon, West Africa,” African Journal of Marine Research C: Oceans, vol. 106, no. 6, pp. 11.597–11.612, 2001. Science, vol. 30, no. 3, pp. 533–540, 2008. ff [32] P. A. Tester and K. A. Steidinger, “Gymnodinium breve [48] J. J. Walsh, J. K. Jolli ,B.P.Darrowetal.,“Redtidesin red tide blooms: initiation, transport, and consequences of the Gulf of Mexico: where, when, and why?” Journal of surface circulation,” Limnology and Oceanography, vol. 42, Geophysical Research C: Oceans, vol. 111, no. 11, Article ID no. 5, pp. 1039–1051, 1998. C11003, 2006. [33] K. M. Kusek, G. Vargo, and K. Steidinger, “Gymnodinium [49] C. Hu, F. E. Muller-Karger, and P. W. Swarzenski, “Hurri- breve in the field, in the lab, and in the newspaper— canes, submarine groundwater discharge, and Florida’s red a scientific and journalistic analysis of Florida red tides,” tides,” Geophysical Research Letters, vol. 33, no. 11, Article ID Contributions in Marine Science, vol. 34, pp. 1–229, 1999. L11601, 2006. [34] C. C. Davis, “Gymnodinium brevis sp. nov., a cause of [50] G. A. Vargo, C. A. Heil, K. A. Fanning et al., “Nutrient discolored water and animal mortality in the Gulf of Mexico,” availability in support of Karenia brevis blooms on the Botany Gaz, vol. 109, pp. 358–360, 1948. central West Florida Shelf: what keeps Karenia blooming?” [35] C. A. Heil, Vertical migration of ptychodiscus brevis (Davis) Continental Shelf Research, vol. 28, no. 1, pp. 73–98, 2008. steidinger,M.S.thesis,UniversityofSouthFlorida,St. [51] D. G. Baden, A. J. Bourdelais, H. Jacocks, S. Michelliza, Petersburg, Fla, USA, 1986. and J. Naar, “Natural and derivative brevetoxins: historical [36] D. Kamykowski, E. J. Milligan, and R. E. Reed, “Relationships background, multiplicity, and effects,” Environmental Health between geotaxis/phototaxis and diei vertical migration in Perspectives, vol. 113, no. 5, pp. 621–625, 2005. autotrophic dinoflagellates,” Journal of Plankton Research, [52] J. H. Landsberg, “The effects of harmful algal blooms on vol. 20, no. 9, pp. 1781–1796, 1998. aquatic organisms,” Reviews in Fisheries Science, vol. 10, no. [37] K. A. Steidinger, G. A. Vargo, P. A. Tester, and C. R. 2, pp. 113–390, 2002. Tomas, “Bloom dynamics and physiology of Gymnodinium [53] S. M. Watkins, A. Reich, L. E. Fleming, and R. Hammond, breve with emphasis on the Gulf of Mexic,” in Physiological “Neurotoxic shellfish poisoning,” Marine Drugs, vol. 6, no. 3, Ecology of Harmful Algal Blooms,D.M.Anderson,A.D. pp. 431–455, 2008. ff Cembella, and G. M. Hallegrae , Eds., pp. 133–153, Springer, [54] P. A. Tester, J. T. Turner, and D. Shea, “Vectorial transport of Heidelberg, Germany, 1998. toxins from the dinoflagellate Gymnodinium breve through [38] K. A. Steidinger, “Implications of dinoflagellate life cycles on copepods to fish,” Journal of Plankton Research, vol. 22, no. 1, initiation of Gymnodinium breve red tides,” Environmental pp. 47–61, 2000. Letters, vol. 9, no. 2, pp. 129–139, 1975. [55] L. J. Flewelling, J. P. Naar, J. P. Abbott et al., “Brevetoxicosis: [39] M. Geesey and P. A. Tester, “Gymnodinium breve: ubiquitous red tides and marine mammal mortalities,” Nature, vol. 435, in Gulf of Mexico waters,” in Toxic Phytoplankton Blooms no. 7043, pp. 755–756, 2005. in the Sea, T. J. Smayda and Y. Shimizu, Eds., pp. 251–255, Elsevier, New York, NY, USA, 1993. [56] J. P. Naar, L. J. Flewelling, A. Lenzi et al., “Brevetoxins, [40] L. E. Brand and A. Compton, “Long-term increase in Karenia like ciguatoxins, are potent ichthyotoxic neurotoxins that brevis abundance along the Southwest Florida Coast,” Harm- accumulate in fish,” Toxicon, vol. 50, no. 5, pp. 707–723, 2007. ful Algae, vol. 6, no. 2, pp. 232–252, 2007. [57] L. C. Backer, L. E. Fleming, A. Rowan et al., “Recreational [41]P.A.Tester,R.P.Stumpf,F.M.Vukovich,P.K.Fowler, exposure to aerosolized brevetoxins during Florida red tide and J. T. Turner, “An expatriate red tide bloom: transport, events,” Harmful Algae, vol. 2, no. 1, pp. 19–28, 2003. distribution, and persistence,” Limnology & Oceanography, [58] B. Kirkpatrick, L. E. Fleming, D. Squicciarini et al., “Litera- vol. 36, no. 5, pp. 1053–1061, 1991. ture review of Florida red tide: implications for human health [42] M. J. Olascoaga, I. I. Rypina, M. G. Brown et al., “Persistent effects,” Harmful Algae, vol. 3, no. 2, pp. 99–115, 2004. transport barrier on the West Florida Shelf,” Geophysical [59]L.E.Fleming,L.C.Backer,andD.G.Baden,“Overviewof Research Letters, vol. 33, no. 22, Article ID L22603, 2006. aerosolized Florida red tide toxins: exposures and effects,” 12 International Journal of Microbiology

Environmental Health Perspectives, vol. 113, no. 5, pp. 618– [76] J. H. Weiss and D. W. Choi, “β-N-methylamino-L-alanine 620, 2005. neurotoxicity: requirement for bicarbonate as a cofactor,” [60] W. M. Abraham and D. G. Baden, “Aerosolized Florida red Science, vol. 241, no. 4868, pp. 973–975, 1988. tide toxins and human health effects,” Oceanography, vol. 19, [77] S. D. Rao, S. A. Banack, P. A. Cox, and J. H. Weiss, “BMAA pp. 107–109, 2006. selectively injures motor neurons via AMPA/kainate receptor [61] B. Kirkpatrick, L. E. Fleming, L. C. Backer et al., “Environ- activation,” Experimental Neurology, vol. 201, no. 1, pp. 244– mental exposures to Florida red tides: effects on emergency 252, 2006. β room respiratory diagnoses admissions,” Harmful Algae, vol. [78]D.Lobner,P.M.T.Piana,A.K.Salous,andR.W.Peoples,“ - 5, no. 5, pp. 526–533, 2006. N-methylamino-l-alanine enhances neurotoxicity through [62] L. E. Fleming, B. Kirkpatrick, L. C. Backer et al., “Aerosolized multiple mechanisms,” Neurobiology of Disease, vol. 25, no. red-tide toxins (brevetoxins) and asthma,” Chest, vol. 131, no. 2, pp. 360–366, 2007. 1, pp. 187–194, 2007. [79] R. R. Bidigare, S. J. Christensen, S. B. Wilde, and S. A. Banack, “Cyanobacteria and BMAA: possible linkage with avian [63] F. Alcock, “An assessment of Florida red tide,” Tech. Rep. vacuolar myelinopathy (AVM) in the south-eastern United 1190, Mote Marine Laboratory, Sarasota, Fla, USA, 2007. States,” Amyotrophic Lateral Sclerosis, vol. 10, supplement 2, [64] M. Gantar and Z. Svircev,ˇ “Microalgae and cyanobacteria: pp. 71–73, 2009. food for thought,” Journal of Phycology,vol.44,no.2,pp. [80] E. J. Faassen, F. Gillissen, H. A. J. Zweers, and M. 260–268, 2008. Lrling, “Determination of the neurotoxins BMAA (β- β [65] P. B. Nunn and P. O’Brien, “The interaction of - N-methylamino-L-alanine) and DAB (α-,γ-diaminobutyric 1 N-methylamino-L-alanine with bicarbonate: an H-NMR acid) by LC-MSMS in Dutch urban waters with cyanobac- study,” FEBS Letters, vol. 251, no. 1-2, pp. 31–35, 1989. terial blooms,” Amyotrophic Lateral Sclerosis, vol. 10, supple- [66] T. G. Myers and S. D. Nelson, “Neuroactive carbamate ment 2, pp. 79–84, 2009. adducts of β-N-methylamino-L-alanine and ethylenedi- [81]H.E.Johnson,S.R.King,S.A.Banack,C.Webster, amine. Detection and quantitation under physiological con- W. J. Callanaupa, and P. A. Cox, “Cyanobacteria (Nostoc ditions by 13C NMR,” Journal of Biological Chemistry, vol. commune) used as a dietary item in the Peruvian highlands 265, no. 18, pp. 10193–10195, 1990. produce the neurotoxic amino acid BMAA,” Journal of [67] A. Vega and E. A. Bell, “α-amino-β-methylaminopropionic Ethnopharmacology, vol. 118, no. 1, pp. 159–165, 2008. acid, a new amino acid from seeds of Cycas circinalis,” [82] J. S. Metcalf, S. A. Banack, J. Lindsay, L. F. Morrison, P. A. Phytochemistry, vol. 6, no. 5, pp. 759–762, 1967. Cox,andG.A.Codd,“Co-occurrenceofβ-N-methylamino- [68] P. S. Spencer, P. B. Nunn, and J. Hugon, “Guam amyotrophic l-alanine, a neurotoxic amino acid with other cyanobacterial lateral sclerosis-Parkinsonism-dementia linked to a plant toxins in British waterbodies, 1990–2004,” Environmental excitant neurotoxin,” Science, vol. 237, no. 4814, pp. 517–522, Microbiology, vol. 10, no. 3, pp. 702–708, 2008. 1987. [83] T. A. Caller, J. W. Doolin, J. F. Haney et al., “A cluster of [69] U. Beckman Sundh, C. Andersson, J. Rosen,F.Fonnum,I.´ amyotrophic lateral sclerosis in New Hampshire: a possible Knudsen, and S. Sippola, “Analysis, occurrence, and toxicity role for toxic cyanobacteria blooms,” Amyotrophic Lateral of β-methylaminoalanine (BMAA): a risk for the consumer?” Sclerosis, vol. 10, supplement 2, pp. 101–108, 2009. TemaNord, vol. 516, p. 128, 2007. [84]P.A.Cox,R.Richer,J.S.Metcalf,S.A.Banack,G.A.Codd, [70] P. A. Cox, S. A. Banack, and S. J. Murch, “Biomagnification andW.G.Bradley,“CyanobacteriaandBMAAexposure of cyanobacterial neurotoxins and neurodegenerative disease from desert dust: a possible link to sporadic ALS among among the Chamorro people of Guam,” Proceedings of the Gulf War veterans,” Amyotrophic Lateral Sclerosis, vol. 10, National Academy of Sciences of the United States of America, supplement 2, pp. 109–117, 2009. vol. 100, no. 23, pp. 13380–13383, 2003. [85] D. Craighead, J. S. Metcalf, S. A. Banack, L. Amgalan, H. V. Reynolds, and M. Batmunkh, “Presence of the neurotoxic [71] P. A. Cox, S. A. Banack, S. J. Murch et al., “Diverse taxa amino acids β-N-methylamino-L-alanine (BMAA) and 2,4- of cyanobacteria produce β-N-methylamino-L-alanine, a diamino-butyric acid (DAB) in shallow springs from the neurotoxic amino acid,” Proceedings of the National Academy Gobi Desert,” Amyotrophic Lateral Sclerosis, vol. 10, supple- of Sciences of the United States of America, vol. 102, no. 14, pp. ment 2, pp. 96–117, 2009. 5074–5078, 2005. [86] B. R. Roney, L. Renhui, S. A. Banack, S. Murch, R. Honegger, [72]R.Rippka,J.Deruelles,J.B.Waterbury,M.Herdman,and and P. A. Cox, “Consumption of fa cai Nostoc soup: a R. Y. Stanier, “Generic assignments, strain histories and Potential for BMAA exposure from Nostoc cyanobacteria in properties of pure cultures of cyanobacteria,” Journal of China?” Amyotrophic Lateral Sclerosis, vol. 10, supplement 2, General Microbiology, vol. 111, no. 1, pp. 1–61, 1979. pp. 44–49, 2009. [73] B. M. Kuehn, “Environmental neurotoxin may pose health [87] S. Jonasson, J. Eriksson, L. Berntzon et al., “Transfer of a threat,” Journal of the American Medical Association, vol. 293, cyanobacterial neurotoxin within a temperate aquatic ecosys- no. 20, pp. 2460–2462, 2005. tem suggests pathways for human exposure,” Proceedings [74] P. G. Ince and G. A. Codd, “Return of the cycad hypothesis— of the National Academy of Sciences of the United States of does the amyotrophic lateral sclerosis/parkinsonism demen- America, vol. 107, no. 20, pp. 9252–9257, 2010. tia complex (ALS/PDC) of Guam have new implications [88] Z. Spacil,´ J. Eriksson, S. Jonasson, U. Rasmussen, L. L. Ilag, for global health?” Neuropathology and Applied Neurobiology, and B. Bergman, “Analytical protocol for identification of vol. 31, no. 4, pp. 345–353, 2005. BMAA and DAB in biological samples,” Analyst, vol. 135, no. [75] M. Esterhuizen and T. G. Downing, “β-N-methylamino- 1, pp. 127–132, 2010. l-alanine (BMAA) in novel South African cyanobacterial [89] P. M. Scott, B. Niedzwiadek, D. F. K. Rawnben, and P.- isolates,” Ecotoxicology and Environmental Safety, vol. 71, no. Y. Lau, “Liquid chromatographic determination of the 2, pp. 309–313, 2008. cyanobacterial toxin β-N-methylamino-l-alanine in algae International Journal of Microbiology 13

food supplements, freshwater fish, and bottled water,” Journal [106] R. M. Luerssen, A. C. Thomas, and J. Hurst, “Rela- of Food Protection, vol. 72, no. 8, pp. 1769–1773, 2009. tionships between satellite-measured thermal features and [90]T.Kubo,N.Kato,K.Hosoya,andK.Kaya,“Effective Alexandrium-imposed toxicity in the Gulf of Maine,” Deep- determination method for a cyanobacterial neurotoxin, β- Sea Research II, vol. 52, no. 19–21, pp. 2656–2673, 2005. N-methylamino-l-alanine,” Toxicon, vol. 51, no. 7, pp. 1264– [107] C. A. Stock, D. J. McGillicuddy, A. R. Solow, and D. 1268, 2008. M. Anderson, “Evaluating hypotheses for the initiation [91] J. Rosen´ and K.-E. Hellenas,¨ “Determination of the neu- and development of Alexandrium fundyense blooms in the rotoxin BMAA (β-N-methylamino-L-alanine) in cycad seed western Gulf of Maine using a coupled physical-biological and cyanobacteria by LC-MS/MS (liquid chromatography model,” Deep-Sea Research II, vol. 52, no. 19-21, pp. 2715– tandem mass spectrometry),” Analyst, vol. 133, no. 12, pp. 2744, 2005. 1785–1789, 2008. [108] C. A. Stock, D. J. McGillicuddy Jr., D. M. Anderson, A. M. [92] T. Kruger,B.M¨ onch,S.Oppenh¨ auser,¨ and B. Luckas, “LC- Solow, and R. P. Signell, “Blooms of the toxic dinoflagellate MS/MS determination of the isomeric neurotoxins BMAA Alexandrium fundyense in the western Gulf of Maine in 1993 (β-N-methylamino-l-alanine) and DAB (2,4-diaminobutyric and 1994: a comparative modeling study,” Continental Shelf acid) in cyanobacteria and seeds of Cycas revoluta and Research, vol. 27, no. 19, pp. 2486–2512, 2007. Lathyrus latifolius,” Toxicon, vol. 55, no. 2-3, pp. 547–557, [109] R. He and D. J. McGillicuddy Jr., “Historic 2005 toxic bloom 2010. of Alexandrium fundyense in the west Gulf of Maine: 1. In situ [93] C. Y. Kao, “Paralytic shellfish poisoning,” in Algal Toxins in observations of coastal hydrography and circulation,” Journal Seafood and Drinking Water,I.R.Falconer,Ed.,pp.75–86, of Geophysical Research C: Oceans, vol. 113, no. 7, Article ID Academic Press, London, UK, 1993. C07039, 2008. [94] A. Prakash, J. Medcof, and A. Tennant, Paralytic Shellfish [110] R. He, D. J. McGillicuddy Jr., B. A. Keafer, and D. M. Ander- Poisoning in Eastern Canada, Fisheries Research Board of son, “Historic 2005 toxic bloom of Alexandrium fundyense in Canada, no. 177, Fisheries Research Board of Canada, the western Gulf of Maine: 2. Coupled biophysical numerical Ottawa, Canada, 1971. modeling,” Journal of Geophysical Research C: Oceans, vol. [95] D. M. Anderson, D. M. Kulis, Y.-Z. Qi, L. Zheng, S. Lu, and 113, no. 7, Article ID C07040, 2008. Y.-T. Lin, “Paralytic shellfish poisoning in Southern China,” [111] J. Zaias, L. C. Backer, and L. E. Fleming, “Harmful algal Toxicon, vol. 34, no. 5, pp. 579–590, 1996. blooms (HABs),” in Human-Animal Medicine: A Clinical [96] M. E. E. Popkiss, D. A. Horstman, and D. Harpur, “Paralytic Guide to Toxins, Zoonoses, and Other Shared Health Risks,P. shellfish poisoning. A report of 17 cases in Cape Town,” South Rabinowitz and L. Conti, Eds., pp. 91–104, Elsevier Science, African Medical Journal, vol. 55, no. 25, pp. 1017–1023, 1979. New York, NY, USA, 2010. [97]B.W.HalsteadandE.Shantz,Paralytic Shellfish Poisoning, [112] C. N. Haas, J. B. Rose, and C. P.Gerba, Quantitative Microbial WHO Offset Publication, no. 79, World Health Organiza- Risk Assessment, John Wiley & Sons, New York, NY, USA, tion, Geneva, Switzerland, 1984. 1999. [98] D. M. Anderson, B. A. Keafer, D. J. McGillicuddy et al., [113] L. C. Backer, J. K. Kish, H. M. Solo-Gabriele, and L. E. “Initial observations of the 2005 Alexandrium fundyense Fleming, “Naturally-occurring water pollutants,” in Water bloom in southern New England: general patterns and and Sanitation Related Diseases and the Environment: Chal- mechanisms,” Deep-Sea Research II, vol. 52, no. 19-21, pp. lenges, Interventions and Preventive Measures, J. Selendy, Ed., 2856–2876, 2005. Horizon International, Yale University, 2010. [99] D. J. McGillicuddy Jr., D. M. Anderson, D. R. Lynch, and D. [114] H. Shuval, “Estimating the global burden of thalassogenic W. Townsend, “Mechanisms regulating large-scale seasonal diseases: human infectious diseases caused by wastewater fluctuations in Alexandrium fundyense populations in the pollution of the marine environment,” Journal of Water and Gulf of Maine: results from a physical-biological model,” Health, vol. 1, no. 2, pp. 53–64, 2003. Deep-Sea Research II, vol. 52, no. 19-21, pp. 2698–2714, 2005. [115] National Research Defense Council (NRDC), “NRDC test- [100] L. C. Backer and D. J. McGillicuddy, “Harmful Algal Blooms: ing the waters,” 2009, http://www.nrdc.org/water/oceans/ at the interface between coastal oceanography and human ttw/titinx.asp. health,” Oceanography, vol. 19, no. 2, pp. 94–106, 2006. [116] U.S. Food and Drug Administration (U.S. FDA), “National [101] D. M. Anderson, J. J. Sullivan, and B. Reguera, “Paralytic Shellfish Registry,” U.S. Food and Drug Administration and shellfish poisoning in northwest Spain: the toxicity of the the Interstate Shellfish Sanitation Conference, The National dinoflagellate Gymnodinium catenatum,” Toxicon, vol. 27, no. Shellfish Register, 1995, http://seaserver.nos.noaa.gov/ 6, pp. 665–674, 1989. projects/95register. [102] D. G. Baden, “Marine food-borne dinoflagellate toxins,” [117] S. Given, L. H. Pendleton, and A. B. Boehm, “Regional International Review of Cytology, vol. 82, pp. 99–150, 1983. public health cost estimates of contaminated coastal waters: a [103] D. M. Anderson and D. Wall, “Potential importance of case study of gastroenteritis at southern California beaches,” benthic cysts of Gonyaulax tamarensis and G. excavata in Environmental Science and Technology, vol. 40, no. 16, pp. initiating toxic dinoflagellate blooms,” Journal of Phycology, 4851–4858, 2006. vol. 14, no. 2, pp. 224–234, 1978. [118] J. S. Yoder, M. C. Hlavsa, G. F. Craun et al., “Surveillance [104] D. W. Townsend, N. R. Pettigrew, and A. C. Thomas, for waterborne disease and outbreaks assoicated with recre- “Offshore blooms of the red tide dinoflagellate, Alexandrium ational water use and other aquatic facility-associated health sp., in the Gulf of Maine,” Continental Shelf Research, vol. 21, events—United States, 2005-2006,” Morbidity and Mortality no. 4, pp. 347–369, 2001. Weekly Report, vol. 58, no. 57/SS-9, pp. 1–38, 2008. [105] P. J. S. Franks and D. M. Anderson, “Toxic phytoplankton [119] M. Iwamoto, T. Ayers, B. E. Mahon, and D. L. Swerdlow, blooms in the southwestern Gulf of Maine: testing hypothe- “Epidemiology of seafood-associated infections in the United ses of physical control using historical data,” Marine Biology, States,” Clinical Microbiology Reviews, vol. 23, no. 2, pp. 399– vol. 112, no. 1, pp. 165–174, 1992. 411, 2010. 14 International Journal of Microbiology

[120] S. Nithiuthai, M. T. Anantaphruti, J. Waikagul, and A. [134] R. M. Gersberg, M. A. Rose, R. Robles-Sikisaka, and A. Gajadhar, “Waterborne zoonotic helminthiases,” Veterinary K. Dhar, “Quantitative detection of hepatitis A virus and Parasitology, vol. 126, no. 1-2, pp. 167–193, 2004. enteroviruses near the United States-Mexico border and [121] C. D. Heaney, E. Sams, S. Wing et al., “Contact with beach correlation with levels of fecal indicator bacteria,” Applied sand among beachgoers and risk of illness,” American Journal and Environmental Microbiology, vol. 72, no. 12, pp. 7438– of Epidemiology, vol. 170, no. 2, pp. 164–172, 2009. 7444, 2006. [122] I. Paquet-Durand, J. Hernandez,´ G. Dolz, J. J. R. Zuniga,˜ [135] D. W. Griffin, C. J. Gibson III, E. K. Lipp, K. Riley, J. H. Paul T. Schnieder, and C. Epe, “Prevalence of Toxocara spp., III,andJ.B.Rose,“Detectionofviralpathogensbyreverse Toxascaris leonina and ancylostomidae in public parks and transcriptase PCR and of microbial indicators by standard beaches in different climate zones of Costa Rica,” Acta methods in the canals of the Florida Keys,” Applied and Tropica, vol. 104, no. 1, pp. 30–37, 2007. Environmental Microbiology, vol. 65, no. 9, pp. 4118–4125, [123] C. Vogel, A. Rogerson, S. Schatz, H. Laubach, A. Tallman, and 1999. J. Fell, “Prevalence of yeasts in beach sand at three bathing [136] F. A. Aulicino, P. Orsini, M. Carere, and A. Mastrantonio, beaches in South Florida,” Water Research,vol.41,no.9,pp. “Bacteriological and virological quality of seawater bathing 1915–1920, 2007. areas along the Tyrrhenian coast,” International Journal of [124] L. C. Backer, J. K. Kish, H. M. Solo-Gabriele, and L. E. Flem- Environmental Health Research, vol. 11, no. 1, pp. 5–11, 2001. ing, “Naturally-occurring water pollutants,” in Water and [137] J. L. Alonso and M. A. Alonso, “Presence of campylobacter Sanitation Related Diseases and the Environment: Challenges, in marine waters of Valencia, Spain,” Water Research, vol. 27, Interventions and Preventive Measures,J.Selendy,Ed.,John no. 10, pp. 1559–1562, 1993. Wiley & Sons, New York, NY, USA, 2010. [138] A. M. Gonzalez, R. Paranhos, and M. S. Lutterbach, “Rela- [125] D. W. Griffin, K. A. Donaldson, J. H. Paul, and J. B. tionships between fecal indicators and pathogenic microor- Rose, “Pathogenic human viruses in coastal waters,” Clinical ganisms in a tropical lagoon in Rio de Janeiro, Brazil,” Microbiology Reviews, vol. 16, no. 1, pp. 129–143, 2003. Environmental Monitoring and Assessment, vol. 164, no. 1–4, pp. 207–219, 2010. [126] S. E. Henrickson, T. Wong, P. Allen, T. Ford, and P. [139] K. D. Goodwin and M. Pobuda, “Performance of R. Epstein, “Marine swimming-related illness: implications CHROMagarTM Staph aureus and CHROMagarTM MRSA for monitoring and environmental policy,” Environmental for detection of Staphylococcus aureus in seawater and beach Health Perspectives, vol. 109, no. 7, pp. 645–650, 2001. sand—comparison of culture, agglutination, and molecular [127] R. Girones, S. Bofill-Mass, M. D. Furones, and C. Rodgers, analyses,” Water Research, vol. 43, no. 19, pp. 4802–4811, “Foodborne infectious diseases and monitoring of marine 2009. food resources,” in Oceans and Human Health: Risks and [140] C. D. Sinigalliano, M. L. Gidley, T. Shibata et al., “Impacts of Remedies from the Sea,P.J.Walsh,S.L.Smith,L.E.Fleming, Hurricanes Katrina and Rita on the microbial landscape of H. Solo-Gabriele, and W. H. Gerwick, Eds., pp. 359–380, the New Orleans area,” Proceedings of the National Academy Elsevier Science, New York, NY, USA, 2008. of Sciences of the United States of America, vol. 104, no. 21, pp. [128] J. W. Santo Domingo and J. Hansel, “Waterborne diseases 9029–9034, 2007. and microbial quality monitoring for recreational water [141] A. L. Lang, Y.-L. Tsai, C. L. Mayer, K. C. Patton, and C. bodies using regulatory methods,” in Oceans and Human J. Palmer, “Multiplex PCR for detection of the heat-labile Health: Risks and Remedies from the Sea,P.J.Walsh,S.L. toxin gene and shiga-like toxin I and II genes in Escherichia Smith, L. E. Fleming, H. Solo-Gabriele, and W. H. Gerwick, coli isolated from natural waters,” Applied and Environmental Eds., pp. 337–357, Elsevier Science, New York, NY, USA, Microbiology, vol. 60, no. 9, pp. 3145–3149, 1994. 2008. [142] S. Ishii, T. Yan, D. A. Shively, M. N. Byappanahalli, R. L. [129] L.-A. H. Krometis, G. W. Characklis, P. N. Drummey, and Whitman, and M. J. Sadowsky, “Cladophora (Chlorophyta) M. D. Sobsey, “Comparison of the presence and partitioning spp. harbor human bacterial pathogens in nearshore water behavior of indicator organisms and Salmonella spp. in an of Lake Michigan,” Applied and Environmental Microbiology, urban watershed,” Journal of Water and Health,vol.8,no.1, vol. 72, no. 7, pp. 4545–4553, 2006. pp. 44–59, 2010. [143] T. K. Graczyk, D. Sunderland, L. Tamang, F. E. Lucy, and P.N. [130] E. K. Lipp, R. Kurz, R. Vincent, C. Rodriguez-Palacios, S. R. Breysse, “Bather density and levels of Cryptosporidium, Gia- ff Farrah,andJ.B.Rose,“Thee ects of seasonal variability and rdia, and pathogenic microsporidian spores in recreational weather on microbial fecal pollution and enteric pathogens in bathing water,” Parasitology Research, vol. 101, no. 6, pp. a subtropical estuary,” Estuaries, vol. 24, no. 2, pp. 266–276, 1729–1731, 2007. 2001. [144] D. P. Keymer, L. H. Lam, and A. B. Boehm, “Biogeographic [131] C. Ortega, H. M. Solo-Gabriele, A. Abdelzaher, M. Wright, Y. patterns in genomic diversity among a large collection of Deng, and L. M. Stark, “Correlations between microbial indi- vibrio cholerae isolates,” Applied and Environmental Micro- cators, pathogens, and environmental factors in a subtropical biology, vol. 75, no. 6, pp. 1658–1666, 2009. Estuary,” Marine Pollution Bulletin, vol. 58, no. 9, pp. 1374– [145] T. Scholz, H. H. Garcia, R. Kuchta, and B. Wicht, “Update 1381, 2009. on the human broad tapeworm (genus diphyllobothrium), [132] A. B. Boehm, N. J. Ashbolt, J. M. Colford Jr. et al., “A sea including clinical relevance,” Clinical Microbiology Reviews, change ahead for recreational water quality criteria,” Journal vol. 22, no. 1, pp. 146–160, 2009. of Water and Health, vol. 7, no. 1, pp. 9–20, 2009. [146] A. C. Wright, R. T. Hill, J. A. Johnson, M.-C. Roghman, [133] M. Wong, L. Kumar, T. M. Jenkins, I. Xagoraraki, M. S. R. R. Colwell, and J. G. Morris Jr., “Distribution of Vibrio Phanikumar, and J. B. Rose, “Evaluation of public health risks vulnificus in the Chesapeake Bay,” Applied and Environmental at recreational beaches in Lake Michigan via detection of Microbiology, vol. 62, no. 2, pp. 717–724, 1996. enteric viruses and a human-specific bacteriological marker,” [147] K. R. O’Neill, S. H. Jones, and D. J. Grimes, “Seasonal Water Research, vol. 43, no. 4, pp. 1137–1149, 2009. incidence of Vibrio vulnificus in the Great Bay estuary of International Journal of Microbiology 15

New Hampshire and Maine,” Applied and Environmental [164] A. B. Boehm, K. M. Yamahara, D. C. Love, B. M. Peterson, K. Microbiology, vol. 58, no. 10, pp. 3257–3262, 1992. Mcneill, and K. L. Nelson, “Covariation and photoinactiva- [148] A. M. Zimmerman, A. DePaola, J. C. Bowers et al., “Variabil- tion of traditional and novel indicator organisms and human ity of total and pathogenic vibrio parahaemolyticus densities viruses at a sewage-impacted marine beach,” Environmental in Northern Gulf of Mexico water and oysters,” Applied and Science and Technology, vol. 43, no. 21, pp. 8046–8052, 2009. Environmental Microbiology, vol. 73, no. 23, pp. 7589–7596, [165] A. M. Abdelzaher, M. E. Wright, C. Ortega et al., “Presence 2007. of pathogens and indicator microbes at a non-point source [149] R. R. Colwell, “Global climate and infectious disease: the subtropical recreational marine beach,” Applied and Environ- cholera paradigm,” Science, vol. 274, no. 5295, pp. 2025– mental Microbiology, vol. 76, no. 3, pp. 724–732, 2010. 2031, 1996. [166] A. B. Boehm, S. B. Grant, J. H. Kim et al., “Decadal [150] M. L. Tamplin, A. L. Gauzens, A. Huq, D. A. Sack, and R. and shorter period variability of surf zone water quality at R. Colwell, “Attachment of Vibrio cholerae serogroup O1 Huntington Beach, California,” Environmental Science and to zooplankton and phytoplankton of Bangladesh waters,” Technology, vol. 36, no. 18, pp. 3885–3892, 2002. Applied and Environmental Microbiology,vol.56,no.6,pp. [167] R. W. Haile, J. S. Witte, M. Gold et al., “The health effects 1977–1980, 1990. of swimming in ocean water contaminated by storm drain [151] D. P. Keymer and A. B. Boehm, “Recombination shapes the runoff,” Epidemiology, vol. 10, no. 4, pp. 355–363, 1999. structure of an environmental Vibrio cholerae population,” [168] A. M. Abdelzaher, H. M. Solo-Gabriele, C. J. Palmer, and T. submitted. M. Scott, “Simultaneous concentration of enterococci and [152] S. W. Joseph, R. R. Colwell, and J. B. Kaper, “Vibrio coliphage from marine waters using a dual layer filtration parahaemolyticus and related halophilic Vibrios,” Critical system,” Journal of Environmental Quality, vol. 38, no. 6, pp. Reviews in Microbiology, vol. 10, no. 1, pp. 77–124, 1982. 2468–2473, 2009. [153] C. Pruzzo, A. Huq, R. R. Colwell, and G. Donelli, “Pathogenic [169] K. D. Goodwin and R. W. Litaker, “Emerging technologies Vibrio species in the marine and estuarine environment,” in for monitoring recreational waters for bacteria and viruses,” Oceans and Health: Pathogens in the Marine Environment,S. in Oceans and Human Health: Risks and Remedies from the Belkin and R. Colwell, Eds., Springer Science, New York, NY, Sea, P. J. Walsh, S. L. Smith, L. E. Fleming, H. Solo-Gabriele, USA, 2006. and W. H. Gerwick, Eds., pp. 381–404, Elsevier Science, New [154] J. D. Oliver, “Vibrio vulnificus,” in Oceans and Health: York, NY, USA, 2008. Pathogens in the Marine Environment,S.BelkinandR. [170] M. B. Nevers and R. L. Whitman, “Nowcast modeling of Colwell, Eds., Springer Science, New York, NY, USA, 2006. Escherichia coli concentrations at multiple urban beaches of [155] R. M. Maier, I. L. Pepper, and C. P. Gerba, Environmental southern Lake Michigan,” Water Research, vol. 39, no. 20, pp. Microbiology, Academic Press, New York, NY, USA, 2nd 5250–5260, 2005. edition, 2008. [171] C. J. Palmer, J. A. Bonilla, T. D. Bonilla et al., “The future [156] U.S. Environmental Protection Agency, Ambient Water Qual- of microbial ocean water quality monitoring,” in Oceans and ity Criteria for Bacteria,Office of Water Regulations and Human Health: Risks and Remedies from the Sea,P.J.Walsh, Standards, Washington, DC, USA, 1986, EPA 440/5-84-002. S. L. Smith, L. E. Fleming, H. Solo-Gabriele, and W. H. [157] T. J. Wade, N. Pai, J. N. S. Eisenberg, and J. M. Colford Jr., “Do Gerwick, Eds., pp. 405–421, Elsevier Science, New York, NY, U.S. Environmental Protection Agency water quality guide- USA, 2008. lines for recreational waters prevent gastrointestinal illness? A systematic review and meta-analysis,” Environmental Health Perspectives, vol. 111, no. 8, pp. 1102–1109, 2003. [158] U.S. Food and Drug Administration, Guide for the Control of Molluscan Shellfish—Section IV, chapter 2, 2007, Growing Areas .01 Total Coliform Standards, http://www.issc.org/. [159] C. P. Gerba, “Assessment of enteric pathogen shedding by bathers during recreational activity and its impact on water quality,” Quantitative Microbiology, vol. 2, no. 1, pp. 55–68, 2000. [160] N. Charoena and R. S. Fujioka, “Assessment of staphylococ- cus bacteria in Hawaii’s marine recreational waters,” Water Science and Technology, vol. 27, no. 3-4, pp. 283–289, 1993. [161] J. M. Fleisher, L. E. Fleming, H. M. Solo-Gabriele et al., “The BEACHES study: health effects and exposures from non- point source microbial contaminants in subtropical recre- ational marine waters,” International Journal of Epidemiology. In press. [162] C. D. Sinigalliano, J. M. Fleisher, M. L. Gidley et al., “Tra- ditional and molecular analyses for fecal indicator bacteria in non-point source subtropical recreational marine waters,” Water Research, vol. 44, no. 13, pp. 3763–3772, 2010. [163] R. T. Noble and J. A. Fuhrman, “Enteroviruses detected by reverse transcriptase polymerase chain reaction from the coastal waters of Santa Monica Bay, California: low correlation to bacterial indicator levels,” Hydrobiologia, vol. 460, pp. 175–184, 2001. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 607575, 9 pages doi:10.1155/2011/607575

Review Article Burkholderia cepacia Complex: Emerging Multihost Pathogens Equipped with a Wide Range of Virulence Factors and Determinants

Sılvia´ A. Sousa, Christian G. Ramos, and Jorge H. Leitao˜

IBB - Institute for Biotechnology and Bioengineering, Centre for Biological and Chemical Engineering, Instituto Superior T´ecnico, Av. Rovisco Pais, 1049-001 Lisboa, Portugal

Correspondence should be addressed to Jorge H. Leitao,˜ [email protected]

Received 11 May 2010; Accepted 2 June 2010

Academic Editor: Max Teplitski

Copyright © 2011 S´ılvia A. Sousa et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The Burkholderia cepacia complex (Bcc) comprises at least 17 closely-related species of the β-proteobacteria subdivision, widely distributed in natural and man-made inhabitats. Bcc bacteria are endowed with an extraordinary metabolic diversity and emerged in the 1980s as life-threatening and difficult-to-treat pathogens among patients suffering from cystic fibrosis. More recently, these bacteria became recognized as a threat to hospitalized patients suffering from other diseases, in particular oncological patients. In the present paper, we review these and other traits of Bcc bacteria, as well as some of the strategies used to identify and validate the virulence factors and determinants used by these bacteria. The identification and characterization of these virulence factors is expected to lead to the design of novel therapeutic strategies to fight the infections caused by these emergent multidrug resistant human pathogens.

1. The Burkholderia cepacia some strains to promote plant growth, bacteria of the Bcc Complex—An Overview have attracted significant commercial interest as biocontrol, bioremediation, and plant-growth promoting agents, mainly Members of the Burkholderia cepacia complex (Bcc) are due to their ability to colonize the rhizosphere of several gram-negative bacteria of the β-proteobacteria subdivision crops of economical interest, like corn, maize, rice, pea, and and include plant, animal, and human pathogens, with a sunflower [2]. However, these bacteria have also emerged as widespread distribution in natural and man-made inhabitats important human pathogens and the risks associated with [1]. These bacteria exhibit an extraordinary metabolic the agricultural uses of Bcc strains remain unclear. There is versatility, allowing their adaptation to a wide range of a general consensus that the large-scale use of organisms of environments. Among the Bcc bacteria, several strains of the Burkholderia genusisimprudentuntilmoreisknown potential environmental application have been identified due about the fate of biocontrol strains after their release in the to their ability to degrade pollutants in water and soils (e.g., environment. The pathogenic mechanisms and traits used crude oils, herbicides, recalcitrant aromatic compounds, and by these bacteria, the clinical outcomes of infected patients, xenobiotics). A summary of Burkholderia strains capable and the interaction of the introduced biocontrol strains with of degrading recalcitrant xenobiotics is available at the environmental and clinical strains need further studies [2]. Biodegradative Strain Database (http://bsd.cme.msu.edu/). Presently, there is some evidence that the environment can Several Bcc strains are also able to produce antifungal com- be a reservoir for the acquisition of novel Bcc infections. pounds and to fix atmospheric nitrogen [2]. Recent evidence For example, the epidemic B. cenocepacia strain PHDC was suggests that members of the Burkholderia genus are ancient recovered both from patients suffering from cystic fibrosis nitrogen-fixing symbionts of Mimosa legumes particularly (CF) in the mid-Atlantic region of USA, as well as from adapted to acidic infertile soils [3]. Due to the ability of agricultural soils [4]. 2 International Journal of Microbiology

2. Bcc as Opportunistic Pathogens in Humans the isolates obtained from CF patients belong to the species B. cenocepacia and B. multivorans [12], the majority of the In the last 30 years, several epidemiological, taxonomic, enviromental isolates belong to the species B. cepacia, B. and molecular biology studies of Bcc strains have been ambifaria, B. cenocepacia,andB. pyrrocinia [2]. carried out by research groups worldwide, mainly due to the A considerable phenotypic variability has been found ability of these strains to cause chronic infections among CF for all the Bcc species [13], even within sequential clinical patients. CF is the most frequent hereditary disorder among isolates of the same strain [20]. This phenotypic variability Caucasians. The disease results from mutations in the cystic difficults the correct identification of Bcc strains by diagnos- fibrosis transmembrane conductance regulator (CFTR), a tic microbiology laboratories [17]. Several phenotypic and cAMP-dependent chloride channel, mainly expressed in the genetic methods have been used for the identification of Bcc apical membrane of epithelial cells [5]. The genetic defect species, including whole-cell protein profile, fatty acid anal- results in multiple organ system impairment, being the respi- ysis, and 16S rRNA and recA gene restriction and sequencing ff ratory tract the most a ected. Chronic pulmonary infections, analysis [2]. However, the genetic methods have proven to although caused by a limited number of bacterial species be the most effective for the correct identification of Bcc (e.g., Pseudomonas aeruginosa,Bcc,Staphylococcus aureus, strains. Nowadays, the multilocus sequence typing scheme Haemophilus influenza,andStenotrophomonas maltophilia), (MLST) is considered the golden standard method for the remain the leading cause of death of these patients [5]. The identification of Bcc species [21]. MLST analysis compares large majority of respiratory infections among CF patients the nucleotide sequence of seven house-keeping genes of Bcc are due to P. aeruginosa strains [6]. Compared to this major and the information obtained for each strain sequence type pathogen, Bcc strains infect a relatively small fraction of CF (ST) is stored in a public database (http://pubmlst.org/bcc/), patients. However, Bcc infections are particularly feared by thus allowing its use worlwide [21]. Recently, this method CF patients and their caregivers since the clinical outcome was redesigned in a nested-PCR MLST format that can be is highly variable and unpredictable, ranging from asymp- used for the accurate identification of Bcc strains directly tomatic carriage to the cepacia syndrome [1]. Additionally, from sputum samples [19]. This approach allowed the in the vast majority of CF patients, pulmonary colonization identification of Bcc strains in 23 sputum samples obtained with Bcc is associated with a worst prognosis, including an from 17 CF patients, of which 8 samples where culture was accelerated decline of the patients’ clinical status and an negative [19]. In addition, the performance of MLST directly increased risk of death [6]. with sputum samples also allowed the identification of Bcc Bcc bacteria are also important pathogens in other strains from CF patients with mixed Bcc infections or co- compromised patients, as is the case of patients suffering infected with P. aeruginosa strains, without the need for from chronic granulomatous disease (CGD) [7]. CGD is a strain isolation [19]. rare hereditary disease that is caused by mutations in the subunits of the NADPH oxidase complex of the phagocytes, resulting in their inability to produce reactive oxygen species 3. Burkholderia cepacia Complex Infections in [7]. Invasive Bcc infections and pneumonia is the second Cystic Fibrosis Patients leading cause of death of CGD patients [7]. There are also some reports of Bcc infections in immunocompromised Bcc bacteria emerged as important CF pathogens during the patients such as cancer and HIV patients, and also among 1980s, when some infected patients exhibited a rapid clinical immunocompetent individuals [8, 9]. In immunocompetent deterioration due to necrotizing pneumonia and sepsis, individuals, Bcc strains have been isolated in cases of resulting in early death [22]. This fatal decline in the patient’s chronic suppurative otitis media, pharyngeal infections, and clinical condition became known as the cepacia syndrome paediatric neck infections [9]. In recent years, an increasing and was not observed for patients infected with any of the number of bacteraemia cases caused by Bcc among non- other CF pathogens. The key determinants associated with CF hospitalized patients have been reported. Most of these the cepacia syndrome are not completely understood, and patients have comorbidities such as chronic hemodialysis, both bacterial and host factors are thought to play important diabetes mellitus, congestive heart failure, and malignancy. roles in determining this dramatic clinical outcome [23, 24]. Among these hospitalized non-CF patients, hemodialysis, Several strains of the species B. multivorans, B. cenocepacia, permanence in intensive care units, use of central venous B. cepacia,andB. dolosa have been shown to be highly catheters, indwelling urinary catheters, and endotracheal transmissible among CF patients through social contact tubes are now recognized as risk factors contributing for [25, 26]. In particular, highly epidemic lineages of the Bcc acquisition. The accumulating reports of nosocomial B. cenocepacia species have been described, including the outbreaks caused by Bcc led to the recognition of these Electrophoretic Type 12 (ET12), the Philadelphia-District bacteria as emergent nosocomial pathogens among non-CF of Columbia (PHDC), and the MidWest epidemic lineages patients, in particular among oncology patients [8]. [27, 28]. These epidemic strains can have an international The Bcc comprises at least seventeen distinct species, impact, as is the case of the highly transmissible ET12 lineage. genetically distinct but phenotypically similar [10, 11] This epidemic lineage spread among individuals with CF (Table 1). Strains from all the Bcc species have been isolated from Canada, UK, and other European countries, being able from CF patients and from the environment, however, their to replace B. multivorans and causing a high mortality due frequency of isolation is uneven [12]. While the majority of to its ability to cause the cepacia syndrome [14, 23, 29]. International Journal of Microbiology 3

Table 1: Burkholderia cepacia complex species and strains with their genome sequences finished or in progress. The Bcc genomes here described are summarized in the Burkholderia genome database (http://www.burkholderia.com/viewAllGenomes.do).

Sources and Relevant Bcc Species Strains sequenced Unfinished genomes References Characteristics Infections in Humans (CF and B. cepacia non-CF); bioremediation and [13] biocontrol agent Infections in Humans (CF and B. multivorans ATCC17616 CGD1, CGD2, CGD2M [13] non-CF) Infections in Humans (CF and J2315, AU1054, HI2424, B. cenocepacia Bu72 [14] non-CF); biocontrol agent MCO-3, PC184 Infections in Humans (CF and B. stabilis [15] non-CF) Infections in Humans (CF and B. vietnamiensis non-CF); bioremediation and G4 [13] biocontrol agent B. dolosa Infections in CF patients AU0158 [16] Infections in Humans (CF and AMMD, MC40-6, MEX-5, B. ambifaria [17] non-CF); biocontrol agent IOP40-10 Infections in CF patients; B. pyrrocinia [18] biocontrol agent Infections in Humans (CF and B. anthina [18] non-CF) B. ubonensis Nosocomial infection Bu [10] B. latens Infections in CF patients [10] Infections in Humans (CF and B. diffusa non-CF), isolated from water and [10] soil Infections in Humans (CF and B. arboris [10] non-CF), environmental sources Infections in Humans (CF and B. seminalis [10] non-CF), environmental sources B. metallica Infections in CF patients [10] Infections in CF patients and in B. contaminans [11, 19] animals B. lata Isolates from forest soil 383 [11]

Due to the easy transmission of highly virulent strains among products. In hospital settings, these pathogens have been CF patients, segregation measures of Bcc-infected patients recovered from tap and distilled water, dialysis machines, have been successfully implemented and led to the reduction nebulisers, catheters, blood gas analysers, thermometers, of the transmission of Bcc strains [30]. ventilator temperature sensors, solutions, and intravenous The prevalence of Bcc species varies geographically, being fluids [9]. B. cenocepacia the most predominant species in CF centers in One of the major problems associated with Bcc infection North America, while B. multivorans is the most common is their intrinsic resistance to most of the clinically avail- speciesinEuropeanCFcenters[31]. However, outbreaks able antimicrobials, including aminoglycosides, quinolones, caused by other Bcc species have occurred worldwide. For polymyxins, and β-lactams [35]. The multiresistance of Bcc instance, in the major Portuguese CF centre, B. cepacia is the bacteria appears to result from various efflux pumps that most prevalent Bcc species. In addition, an outbreak of B. efficiently remove antibiotics from the cell, decreased contact cepacia was reported and associated with the use of nonsterile of antibiotics with the bacterial cell surface due to their ability saline solutions for nasal application [32]. Bcc outbreaks to form biofilms, and changes in the cell envelope that reduce among non-CF populations, mainly due to strains of the the permeability of the membrane to the antibiotic [36]. species B. cenocepacia, B. cepacia, and B. multivorans are Bcc bacteria are also resistant to neutrophil-mediated non- also well documented [33]. Accumulating evidence points oxidative killing and to the antimicrobial peptides produced out contaminated pharmaceuticals, cosmetics, disinfectants, by airway epithelial cells, including lysozyme, lactoferrin, and and preservative products as major sources of Bcc bacteria phospholipase A2 [37]. Therefore, CF patients chronically [9, 34]. This is due to their ability to survive in these infected with Bcc are difficult to treat and, although current 4 International Journal of Microbiology treatment strategies use double or triple antibiotic combi- 5. Strategies to Discover New Bcc Virulence nations to achieve bactericidal activity, they rarely result in Factors and Determinants the eradication of the pathogen, particularly in the case of chronic infection [36]. Different strategies have been designed to identify path- ogenicity-related genes from Bcc bacteria, including the gen- eration of mutant libraries with transposons and plasposons, systematic gene-by-gene inactivation and high-throughput 4. Organization of Bcc Genomes sequencing, as illustrated in Figure 1. Our research group has been using a strategy based on the generation of mutant In 2003, the Wellcome Trust Sanger Institute sequenced the libraries from B. cenocepacia and B. cepacia strains by first genome of a Bcc strain. The strain chosen was the type random mutagenesis with plasposons [41], followed by strain of the ET12 epidemic lineage, the B. cenocepacia strain rescue of the interrupted genes, sequencing and comparison J 2 315 (http: / /www.sanger.ac.uk/Projects/B cenocepacia/). of the nucleotide sequence of the interrupted genes with the Presently, the genomic sequences of 18 strains from 7 Bcc available genome sequences of Bcc strains, combined with species are publicly available (http://pathema.jcvi.org/cgi- the virulence assessment in the Bcc infection models X-CGD bin/Burkholderia/PathemaHomePage.cgi). The genomes of mice and/or Caenorhabditis elegans. A mutant library Bcc bacteria are organized in three circular chromosomal derived from B. cepacia IST408 allowed the identification replicons and one to five megaplasmids, ranging from 6.2 to 8.7 Mbp in size, with a GC content of about 67%. The of the bce-I gene cluster that encodes proteins and enzymes large size and repartition of the genomes of Bcc is thought involved in the biosynthesis of the exopolysaccharide to increase their flexibility to acquire and lose genes. In a (EPS) Cepacian [42]. Cepacian is composed of a branched recent bioinformatics study, Cooper et al. (2010) suggested acetylated heptasaccharide repeat unit with D-glucose, D- that the genes located in secondary chromosomes exhibit rhamnose, D-mannose, D-galactose, and D-glucuronic acid, a weaker codon usage bias than those located in primary in the ratio 1:1:1:3:1, respectively [43]. Several studies chromosomes, being subject to a faster evolutionary rate have shown that Cepacian interfered with the phagocytosis [38]. Several evidences point out that more than 10% of bacteria by human neutrophils and, inhibits neutrophil of the Bcc genomes have been acquired by horizontal chemotaxis, and the production of reactive oxygen species gene transfer, contributing to the genomic plasticity and [44, 45]. The ability to produce this EPS was also associated metabolic diversity of these bacteria. For example, in the case with persistence of infection in the BALB/c and X-CGD of the B. cenocepacia strain J2315, 14 genomic islands, most mice models of infection [44, 46]. Studies performed with probably arisen from horizontal gene transfer, have been cepacian-defective mutants have also shown that cepacian is identified based on their distinct GC content percentage [39]. required for the formation of thick and mature biofilms [47]. The acquisition of genomic islands appears to play a crucial Biofilm formation in vitro is a common trait of Bcc strains role in the evolution of this particular epidemic lineage, and has been associated with the persistence of Bcc infections introducing new functions that promoted survival and [48]. In addition, bacteria of the Bcc growing in biofilms pathogenesis in the CF lung. This is the case of the 31.7 kb have been found to be more resistant to antimicrobials cci pathogenicity island, which appears to be unique to B. than those growing plancktonically. It is also worth to cenocepacia strains [40]. This pathogenicity island encodes mention that in a recent study by Dales et al. (2009), biofilms both virulence and metabolism-associated genes, including formed by Bcc were found to be more resistant to antibiotics the CciIR quorum sensing system, a fatty acid biosynthesis compared to P. aeruginosa biofilms [49]. Remarkably, a operon, transcriptional regulators, and genes involved in the mutant producing about one half of the amount of the metabolism of amino acids [40]. In addition, the genome EPS was recently found to carry a plasposon insertion of B. cenocepacia J2315 contains 79 insertion sequence interrupting a gene encoding the RNA chaperone Hfq [50]. (IS) elements that are most probably involved in genomic Hfq proteins are global regulators of metabolism, acting as rearrangements, replicon fusion, activation/silencing of gene RNA chaperones involved in the riboregulation of target expression, mobilization of DNA, and recruitment of foreign mRNAs by small regulatory noncoding RNAs (sRNAs), genes [39]. Another feature of the genomes of Bcc bacteria is facilitating the interaction with their target mRNAs [51]. the presence of multiple pathways with related functions, and The B. cepacia hfq mutant was shown to be more susceptible gene redundancy due to the occurrence of paralogous genes. to stress conditions, particularly to those that mimicked the Sequencing of several Bcc genomes, followed by com- lung environment of the CF host, indicating that Hfq plays parative genomics, is a powerful tool for the identification a major role in the survival of Bcc bacteria under those stress of virulence-associated genes of Bcc bacteria, including conditions [50]. In addition, the hfq mutants from B. cepacia, new genes encoding proteins with no predicted function. B. dolosa,andB. ambifaria exhibited a reduced ability to In the sequenced Bcc genomes, the percentage of protein colonize and kill the nematode C. elegans, indicating that encoding genes with unknown function varies between 13 Hfq is an important virulence determinant of Bcc bacteria. and 35% (http://img.jgi.doe.gov/cgi-bin/pub/main.cgi). It is In agreement with the roles played by Hfq in other bacteria, quite possible that a significant percentage of these genes Sousa et al. (2010) have also shown that the B. cepacia IST408 of unknown function might be involved, either directly or Hfq is able to bind to sRNAs. Recently, 213 putative sRNAs indirectly, in the pathogenesis of Bcc bacteria. were identified within the genome of B. cenocepacia J2315, International Journal of Microbiology 5

Plasposon random mutagenesis Signature tagged mutagenesis Suppression subtractive High throughput sequencing hybridization DNA Wild-type Wild-type Wild-type

Wild-type Virulent strain Avirulent strain Wild-type Triparental mating Wild-type Wild-type or electroporation Transformation with DNA DNA tagged-plasposons Bacterial Bacterial Digest and hybridize RNA RNA Plating on Selection with selective media antibiotic resistance mark Selection Screening Select for unique with antibiotic for survival in fragments in virulent strain cDNA cDNA resistance mark infection models and generate clone library

Hybridize DNA Screening for reduced virulence in infection models Sequence mutants that failed infection DNA Solexa/illumina sequencing isolation Sequencing 90 100 TTT G C GC GGCG T GCCTTG C G 90 100 Comparative genomics TTT G C GC GGCG T GCCTTG C G Sequencing

90 100 TTT G C GC GGCG T GCCTTG C G BLAST search for homologues Unique genes in infection

Gene identification Figure 1: Strategies used to identify Bcc virulence factors and determinants. based on the combination of comparative genomics and A signature-tagged mutagenesis (STM) strategy was used prediction of their secondary structures [52]. Work in by Hunt et al. to identify putative virulence factors of Bcc [57] progress envisages the identification and characterization of (Figure 1). STM is a comparative hybridization technique sRNAs from Bcc to gain clues on their possible contributions that uses a collection of transposons, each modified by the to virulence. incorporation of a DNA sequence tag [57]. The pool of Another mutant library, derived from the highly epi- mutants is inoculated into a chronic pulmonary infection demic strain B. cenocepacia J2315, allowed the identification animal model and the bacteria recovered after infection are of a gene encoding an acyl carrier protein (ACP) [53]. identified due to the tags. Mutants containing a transposon Bacterial ACPs play a central role in metabolism, being the insertion in genes required for survival will fail to pass donors of the acyl moiety that is required for the biosynthesis through the in vivo selection, thus allowing the identification of fatty acids, phospholipids, endotoxins, glycolipids, and of these genes. This strategy led to the identification of signalling molecules that are necessary for growth and patho- several B. cenocepacia K56-2 genes that were required for genesis [54]. The acp mutant exhibited an increased ability to bacterial survival in a rat model of chronic lung infection, form biofilms in vitro, a more hydrophobic cell surface, and including genes involved in cellular metabolism, global reduced ability to colonize and kill the nematode C. elegans, regulation, DNA replication and repair, cell surface proteins, indicating that ACP protein is a virulence determinant for and polysaccharide production [57]. Bccbacteria[53]. In addition, the amino acid sequence A suppression-subtractive hybridization (SSH) strategy and structural differences between the ACP proteins from was used to identify genes that are unique to the B. cenocepa- bacteria and humans make this protein an attractive target cia and/or to ET12 epidemic lineage strains [58](Figure 1). for the development of novel antimicrobial compounds Recently, a high-throughput sequencing strategy was used [55]. to compare the transcriptional response of clinical and AmutantlibraryfromB. cenocepacia K56-2 was also environmental strains of B. cenocepacia [59](Figure 1). This constructed. The screen of this library for mutants impaired strategy revealed a large number of regulatory differences in their ability to kill the nematode C. elegans allowed the between environmental and clinical strains, which might identification of the regulatory protein Pbr. The pbr mutant result from specific adaptations to each of the different exhibited a pleiotropic phenotype, being unable to produce niches, despite their high degree of DNA sequence similarity. phenazines, exhibited a reduced resistance to stresses such as Genes that encode for molecular chaperones and iron oxidative and osmotic stress, and a reduced ability to survive acquisition proteins were found to be particularly induced prolonged nutrient starvation periods [56]. in the clinical strain [59]. 6 International Journal of Microbiology

All these strategies have allowed the identification of human disease. In fact, contrasting with other pathogens, several genes putatively involved in the virulence of Bcc the pathogenicity of Bcc bacteria does not rely on a single strains. However, the characterization of knockout mutants gene. Accumulating evidences point out that Bcc virulence in these genes is hampered by limited available genetic is polygenic, involving genes related to survival under stress tools and the inherent resistance of Bcc strains to the conditions [50, 76–78]. Nevertheless, the Bcc genome is most common antibiotics used for genetic selection. In equipped with the known crucial genes for colonization this context, some research groups have been developing and initiation of chronic infection in the respiratory tract, molecular tools to genetically manipulate Bcc strains. For which are involved in motility, adhesion, and host tissue instance, Lefebre and Valvano (2002) constructed several damage. Another important feature of some Bcc strains is expression vectors that contain the dhfr gene, encoding the their ability to invade and survive inside eukaryotic cells, dihydrofolate reductase enzyme required for trimethoprim including soil-dwelling amoebae, human macrophages, and resistance, together with either the constitutive promoter of airway epithelial cells [79]. the S7 ribossomal protein gene from Burkholderia sp LB400, or the arabinose-inducible BAD promoter from Escherichia coli [60]. However, the concentration of arabinose required 6. Concluding Remarks for maximal gene expression [2% (w/v) or higher] causes Members of the Bcc have emerged in the last decades as a change in cell volume typical of osmotic stress [61]. important pathogens to human, animals, and plants. The As a consequence, the full complementation of a given pathogenicity of these bacteria is polygenic, and thus involves mutation using these vectors is seldom achieved, limiting its a multitude of known and unknown virulence factors and use [50]. Therefore, those authors have constructed another determinants. Several strategies have been successfully used expression vector, containing the rhamnose-regulated PrhaB by several research groups to reveal novel and unknown promoter of E. coli that allows maximal gene expression at virulence factors and determinants. The knowledge of the low concentrations of rhamnose [61]. Another limitation molecular mechanisms employed by Bcc bacteria for viru- derives from the fact that the B. cenocepacia J2315 strain is lence and pathogenesis is of crucial importance to identify a poor recipient of DNA in transformation and conjugation. new targets for the rational design of novel strategies and/or As an example, transformation of B. cepacia IST408 with molecules to combat Bcc infections, since their resistance 4 ffi plasposon pTnMod was 10 -fold more e cient than B. to most of the clinically-relevant antimicrobials renders the cenocepacia J2315 (S. A. Sousa and J. H. Leitao,˜ unpublished infections untreatable. In order to be regarded as a potential results). Recently, the electroporation procedure for this drug target, a given gene or gene product must be essential ffi strain was modified to increase its transformation e ciency for survival of the pathogen in the host and should be [62]. Factors that contribute to this improvement include conserved in the various strains of the pathogen, while the addition of glycine to the growth medium to weaken presenting little or no conservation in humans. Genome- the thick cell wall, demethylation of transforming DNA by based strategies, including genome sequencing, microarray- extraction from a E. coli dam dcm host strain to escape to based expression technology, and large-scale mutagenesis the J2315 restriction system specific for methylated GATC studies, are expected to contribute, in the near future, for sites, the inclusion of the Ocr protein in the transformation the development of new strategies and/or antimicrobials mixture to act as a decoy to inhibit Type I restriction molecules to fight the devastating and presently difficult-to- endonuclease attack of entering DNA, and the use of treat infections caused by Bcc strains. spermine to reduce the resistance of the B. cenocepacia J2315 strain to several antibiotics [62]. Other strategies used to effectively generate Bcc mutants rely on the lambda red Acknowledgments recombinase system, as proposed by Datsenko and Wanner This work was funded by FEDER and Fundac¸ao˜ para a (2000) [63]. This strategy uses linear DNA transformations Cienciaˆ e Tecnologia (FCT), Portugal, through Contract and has allowed the successful homologous expression of a PTDC/EBB-BIO/098352/2008, and a postdoctoral grant to lipase gene in B. cepacia [64], as well the construction of an S. A. Sousa. A doctoral grant from Fundac¸ao˜ Calouste insertion mutant in B. cenocepacia J2315 BCAL1538 (C. G. Gulbenkian (FCG) to C. G. Ramos is also acknowledged. Ramos, S. A. Sousa, J. H. Leitao,˜ unpublished results). A combination of one or more of the previous approaches has allowed the identification of several potential References virulence factors, including the cable pili and various adhesins [65], flagella [66], a type III and a type VI secretion [1]E.Mahenthiralingam,T.A.Urban,andJ.B.Goldberg,“The systems [67, 68], lipopolysaccharide [69], four types of iron- multifarious, multireplicon Burkholderia cepacia complex,” chelating siderophores (salicylic acid, ornibactin, pyochelin, Nature Reviews Microbiology, vol. 3, no. 2, pp. 144–156, 2005. [2] L. Chiarini, A. Bevivino, C. Dalmastri, S. Tabacchioni, and P. cepaciachelin, and cepabactin) [70], production of extracel- Visca, “Burkholderia cepacia complex species: health hazards lular proteins, like proteases, lipases, and haemolysins [71, and biotechnological potential,” Trends in Microbiology, vol. 72], quorum-sensing systems [73, 74], and others (recently 14, no. 6, pp. 277–286, 2006. reviewed in [75]). However, not all strains produce each [3] C. Bontemps, G. N. Elliott, M. F. Simon et al., “Burkholderia of these virulence factors, and none of these factors has species are ancient symbionts of legumes,” Molecular Ecology, been clearly demonstrated to be a major contributor to vol. 19, no. 1, pp. 44–52, 2010. International Journal of Microbiology 7

[4] J. J. LiPuma, T. Spilker, T. Coenye, and C. F. Gonzalez, “An of Burkholderia cepacia complexinsputumsamplesfrom epidemic Burkholderia cepacia complex strain identified in patients with cystic fibrosis,” Journal of Clinical Microbiology, soil,” Lancet, vol. 359, no. 9322, pp. 2002–2003, 2002. vol. 48, no. 5, pp. 1888–1891, 2010. [5] F. Ratjen and G. Doring,¨ “Cystic fibrosis,” Lancet, vol. 361, no. [20] G. Y. Larsen, T. L. Stull, and J. L. Burns, “Marked phenotypic 9358, pp. 681–689, 2003. variability in Pseudomonas cepacia isolated from a patient with [6] J. B. Lyczak, C. L. Cannon, and G. B. Pier, “Lung infections cystic fibrosis,” Journal of Clinical Microbiology, vol. 31, no. 4, associated with cystic fibrosis,” Clinical Microbiology Reviews, pp. 788–792, 1993. vol. 15, no. 2, pp. 194–222, 2002. [21] A. Baldwin, E. Mahenthiralingam, K. M. Thickett et al., [7] R. B. Johnston Jr., “Clinical aspects of chronic granulomatous “Multilocus sequence typing scheme that provides both disease,” Current Opinion in Hematology, vol. 8, no. 1, pp. 17– species and strain differentiation for the Burkholderia cepacia 22, 2001. complex,” Journal of Clinical Microbiology, vol. 43, no. 9, pp. [8] T. Mann, D. Ben-David, A. Zlotkin et al., “An outbreak of 4665–4673, 2005. Burkholderia cenocepacia bacteremia in immunocompromised [22] A. Isles, I. Maclusky, and M. Corey, “Pseudomonas cepacia oncology patients,” Infection, vol. 38, no. 3, pp. 187–194, 2010. infection in cystic fibrosis: an emerging problem,” Journal of [9] G. Marioni, R. Rinaldi, G. Ottaviano, R. Marchese-Ragona, Pediatrics, vol. 104, no. 2, pp. 206–210, 1984. M. Savastano, and A. Staffieri, “Cervical necrotizing fasciitis: a [23] J. R. W. Goven, P. H. Brown, J. Maddison et al., “Evidence for novel clinical presentation of Burkholderia cepacia infection,” transmission of Pseudomonas cepacia by social contact in cystic Journal of Infection, vol. 53, no. 5, pp. e219–e222, 2006. fibrosis,” Lancet, vol. 342, no. 8862, pp. 15–19, 1993. [10] E. Vanlaere, J. J. LiPuma, A. Baldwin et al., “Burkholderia latens [24] A. M. Jones, M. E. Dodd, J. R. W. Govan et al., “Burkholderia sp. nov., Burkholderia diffusa sp. nov., Burkholderia arboris cenocepacia and Burkholderia multivorans: influence on sur- sp. nov., Burkholderia seminalis sp. nov., and Burkholderia vival in cystic fibrosis,” Thorax, vol. 59, no. 11, pp. 948–951, metallica sp. nov., novel species within the Burkholderia 2004. cepacia complex,” International Journal of Systematic and [25] J. J. Lipuma, S. E. Dasen, D. W. Nielson, R. C. Stern, and Evolutionary Microbiology, vol. 58, no. 7, pp. 1580–1590, 2008. T. L. Stull, “Person-to-person transmission of Pseudomonas [11] E. Vanlaere, A. Baldwin, D. Gevers et al., “Taxon K, a complex cepacia between patients with cystic fibrosis,” Lancet, vol. 336, within the Burkholderia cepacia complex, comprises at least no. 8723, pp. 1094–1096, 1990. two novel species, Burkholderia contaminans sp.nov.and [26] R. Biddick, T. Spilker, A. Martin, and J. J. LiPuma, “Evidence Burkholderia lata sp. nov,” International Journal of Systematic of transmission of Burkholderia cepacia, Burkholderia mul- and Evolutionary Microbiology, vol. 59, no. 1, pp. 102–111, tivorans and Burkholderia dolosa among persons with cystic 2009. fibrosis,” FEMS Microbiology Letters, vol. 228, no. 1, pp. 57–62, [12] E. Mahenthiralingam, A. Baldwin, and C. G. Dowson, 2003. “Burkholderia cepacia complex bacteria: opportunistic [27] W. M. Johnson, S. D. Tyler, and K. R. Rozee, “Linkage pathogens with important natural biology,” Journal of Applied analysis of geographic and clinical clusters in Pseudomonas Microbiology, vol. 104, no. 6, pp. 1539–1551, 2008. cepacia infections by multilocus enzyme electrophoresis and [13] P. Vandamme, B. Holmes, M. Vancanneyt et al., “Occurrence ribotyping,” Journal of Clinical Microbiology,vol.32,no.4,pp. of multiple genomovars of Burkholderia cepacia in cystic 924–930, 1994. fibrosis patients and proposal of Burkholderia multivorans sp. [28] J. S. Chen, K. A. Witzmann, T. Spilker, R. J. Fink, and J. J. nov,” International Journal of Systematic Bacteriology, vol. 47, LiPuma, “Endemicity and inter-city spread of Burkholderia no. 4, pp. 1188–1200, 1997. cepacia genomovar III in cystic fibrosis,” Journal of Pediatrics, vol. 139, no. 5, pp. 643–649, 2001. [14] P. Vandamme, B. Holmes, T. Coenye et al., “Burkholderia [29]D.P.Speert,D.Henry,P.Vandamme,M.Corey,andE. cenocepacia sp. nov.—a new twist to an old story,” Research in Mahenthiralingam, “Epidemiology of Burkholderia cepacia Microbiology, vol. 154, no. 2, pp. 91–96, 2003. complex in patients with cystic fibrosis, Canada,” Emerging [15] P. Vandamme, E. Mahenthiralingam, B. Holmes et al., “Iden- Infectious Diseases, vol. 8, no. 2, pp. 181–187, 2002. tification and population structure of Burkholderia stabilis sp. [30] F. Festini, R. Buzzetti, C. Bassi et al., “Isolation measures for nov. (formerly Burkholderia cepacia genomovar IV),” Journal prevention of infection with respiratory pathogens in cystic of Clinical Microbiology, vol. 38, no. 3, pp. 1042–1047, 2000. fibrosis: a systematic review,” Journal of Hospital Infection, vol. [16] K. Vermis, T. Coenye, J. J. LiPuma, E. Mahenthiralingam, 64, no. 1, pp. 1–6, 2006. H. J. Nelis, and P. Vandamme, “Proposal to accommodate [31] J. R. W. Govan, A. R. Brown, and A. M. Jones, “Evolving Burkholderia cepacia genomovar VI as Burkholderia dolosa epidemiology of Pseudomonas aeruginosa and the Burkholde- sp. nov,” International Journal of Systematic and Evolutionary ria cepacia complex in cystic fibrosis lung infection,” Future Microbiology, vol. 54, no. 3, pp. 689–691, 2004. Microbiology, vol. 2, no. 2, pp. 153–164, 2007. [17] T. Coenye, P. Vandamme, J. R. W. Govan, and J. J. Lipuma, [32] M. V. Cunha, A. Pinto-de-Oliveira, L. Meirinhos-Soares “Taxonomy and identification of the Burkholderia cepacia et al., “Exceptionally high representation of Burkholderia complex,” Journal of Clinical Microbiology, vol. 39, no. 10, pp. cepacia among B. cepacia complex isolates recovered from the 3427–3436, 2001. major Portuguese cystic fibrosis center,” Journal of Clinical [18]P.Vandamme,D.Henry,T.Coenyeetal.,“Burkholderia Microbiology, vol. 45, no. 5, pp. 1628–1633, 2007. anthina sp. nov. and Burkholderia pyrrocinia, two additional [33] R. Reik, T. Spilker, and J. J. LiPuma, “Distribution of Burkholderia cepacia complex bacteria, may confound results Burkholderia cepacia complex species among isolates recov- of new molecular diagnostic tools,” FEMS Immunology and ered from persons with or without cystic fibrosis,” Journal of Medical Microbiology, vol. 33, no. 2, pp. 143–149, 2002. Clinical Microbiology, vol. 43, no. 6, pp. 2926–2928, 2005. [19] P. Drevinek, S. Vosahlikova, K. Dedeckova, O. Cinek, and E. [34] S. G. Geftic, H. Heymann, and F. W. Adair, “Fourteen year Mahenthiralingam, “Direct culture-independent strain typing survival of Pseudomonas cepacia in a salts solution preserved 8 International Journal of Microbiology

with benzalkonium chloride,” Applied and Environmental [48] B.-A.D.Conway,V.Venu,andD.P.Speert,“Biofilmformation Microbiology, vol. 37, no. 3, pp. 505–510, 1979. and acyl homoserine lactone production in the Burkholderia [35] J. H. Leitao,˜ S. A. Sousa, M. V. Cunha et al., “Variation cepacia complex,” Journal of Bacteriology, vol. 184, no. 20, pp. of the antimicrobial susceptibility profiles of Burkholderia 5678–5685, 2002. cepacia complex clonal isolates obtained from chronically [49] L. Dales, W. Ferris, K. Vandemheen, and S. D. Aaron, “Combi- infected cystic fibrosis patients: a five-year survey in the major nation antibiotic susceptibility of biofilm-grown Burkholderia Portuguese treatment center,” European Journal of Clinical cepacia and Pseudomonas aeruginosa isolated from patients Microbiology and Infectious Diseases, vol. 27, no. 11, pp. 1101– with pulmonary exacerbations of cystic fibrosis,” European 1111, 2008. Journal of Clinical Microbiology and Infectious Diseases, vol. 28, [36] A. M. George, P. M. Jones, and P. G. Middleton, “Cystic no. 10, pp. 1275–1279, 2009. fibrosis infections: treatment strategies and prospects,” FEMS [50] S. A. Sousa, C. G. Ramos, L. M. Moreira, and J. H. Leitao,˜ “The Microbiology Letters, vol. 300, no. 2, pp. 153–164, 2009. hfq gene is required for stress resistance and full virulence of [37] R. M. Baird, H. Brown, A. W. Smith, and M. L. Watson, Burkholderia cepacia to the nematode Caenorhabditis elegans,” “Burkholderia cepacia is resistant to the antimicrobial activity Microbiology, vol. 156, no. 3, pp. 896–908, 2010. of airway epithelial cells,” Immunopharmacology, vol. 44, no. 3, [51] P.Valentin-Hansen, M. Eriksen, and C. Udesen, “The bacterial pp. 267–272, 1999. Sm-like protein Hfq: a key player in RNA transactions,” [38]V.S.Cooper,S.H.Vohr,S.C.Wrocklage,andP.J.Hatcher, Molecular Microbiology, vol. 51, no. 6, pp. 1525–1533, 2004. “Why genes evolve faster on secondary chromosomes in [52] T. Coenye, P. Drevinek, E. Mahenthiralingam et al., “Identifi- bacteria,” PLoS Computational Biology, vol. 6, no. 4, Article ID cation of putative noncoding RNA genes in the Burkholderia e1000732, 2010. cenocepacia J2315 genome,” FEMS Microbiology Letters, vol. [39] M. T. G. Holden, H. M. B. Seth-Smith, L. C. Crossman et al., 276, no. 1, pp. 83–92, 2007. “The genome of Burkholderia cenocepacia J2315, an epidemic [53] S. A. Sousa, C. G. Ramos, F. Almeida et al., “Burkholderia pathogen of cystic fibrosis patients,” Journal of Bacteriology, cenocepacia J2315 acyl carrier protein: a potential target for vol. 91, no. 1, pp. 261–277, 2009. antimicrobials’ development?” Microbial Pathogenesis, vol. 45, [40] A. Baldwin, P. A. Sokol, J. Parkhill, and E. Mahenthiralingam, no. 5-6, pp. 331–336, 2008. “The Burkholderia cepacia epidemic strain marker is part [54] D. M. Byers and H. Gong, “Acyl carrier protein: structure- of a novel genomic island encoding both virulence and function relationships in a conserved multifunctional protein metabolism-associated genes in Burkholderia cenocepacia,” family,” Biochemistry and Cell Biology, vol. 85, no. 6, pp. 649– Infection and Immunity, vol. 72, no. 3, pp. 1537–1547, 2004. 662, 2007. [41] J. J. Dennis and G. J. Zylstra, “Plasposons: modular self- [55] S. W. White, J. Zheng, Y.-M. Zhang, and C. O. Rock, “The cloning minitransposon derivatives for rapid genetic analysis structural biology of type II fatty acid biosynthesis,” Annual of gram-negative bacterial genomes,” Applied and Environ- Review of Biochemistry, vol. 74, pp. 791–831, 2005. mental Microbiology, vol. 64, no. 7, pp. 2710–2715, 1998. [56] C. G. Ramos, S. A. Sousa, A. M. Grilo, L. Eberl, and J. [42] L. M. Moreira, P. A. Videira, S. A. Sousa, J. H. Leitao,˜ M. H. Leitao,˜ “The Burkholderia cenocepacia K56-2 pleiotropic V. Cunha, and I. Sa-Correia,´ “Identification and physical regulator Pbr, is required for stress resistance and virulence,” organization of the gene cluster involved in the biosynthesis of Microbial Pathogenesis, vol. 48, no. 5, pp. 168–177, 2010. Burkholderia cepacia complex exopolysaccharide,” Biochemical [57] T. A. Hunt, C. Kooi, P. A. Sokol, and M. A. Valvano, and Biophysical Research Communications, vol. 312, no. 2, pp. “Identification of Burkholderia cenocepacia genes required for 323–333, 2003. bacterial survival in vivo,” Infection and Immunity, vol. 72, no. [43] P. Cescutti, M. Bosco, F. Picotti et al., “Structural study 7, pp. 4010–4022, 2004. of the exopolysaccharide produced by a clinical isolate of [58] S. P. Bernier and P. A. Sokol, “Use of suppression-subtractive Burkholderia cepacia,” Biochemical and Biophysical Research hybridization to identify genes in the Burkholderia cepacia Communications, vol. 273, no. 3, pp. 1088–1094, 2000. complex that are unique to Burkholderia cenocepacia,” Journal of Bacteriology, vol. 187, no. 15, pp. 5278–5291, 2005. [44] B.-A. D. Conway, K. K. Chu, J. Bylund, E. Altman, and D. [59] D. R. Yoder-Himes, P. S. G. Chain, Y. Zhu et al., “Mapping the P. Speert, “Production of exopolysaccharide by Burkholderia Burkholderia cenocepacia niche response via high-throughput cenocepacia results in altered cell-surface interactions and sequencing,” Proceedings of the National Academy of Sciences of altered bacterial clearance in mice,” Journal of Infectious the United States of America, vol. 106, no. 10, pp. 3976–3981, Diseases, vol. 190, no. 5, pp. 957–966, 2004. 2009. [45] J. Bylund, L.-A. Burgess, P. Cescutti, R. K. Ernst, and D. [60] M. D. Lefebre and M. A. Valvano, “Construction and eval- P. Speert, “Exopolysaccharides from Burkholderia cenocepacia uation of plasmid vectors optimized for constitutive and inhibit neutrophil chemotaxis and scavenge reactive oxygen regulated gene expression in Burkholderia cepacia complex species,” Journal of Biological Chemistry, vol. 281, no. 5, pp. isolates,” Applied and Environmental Microbiology, vol. 68, no. 2526–2532, 2006. 12, pp. 5956–5964, 2002. [46] S. A. Sousa, M. Ulrich, A. Bragonzi et al., “Virulence of [61] S. T. Cardona and M. A. Valvano, “An expression vector phox−/− Burkholderia cepacia complex strains in gp91 mice,” containing a rhamnose-inducible promoter provides tightly Cellular Microbiology, vol. 9, no. 12, pp. 2817–2825, 2007. regulated gene expression in Burkholderia cenocepacia,” Plas- [47] M. V. Cunha, S. A. Sousa, J. H. Leitao,L.M.Moreira,P.A.˜ mid, vol. 54, no. 3, pp. 219–228, 2005. Videira, and I. Sa-Correia,´ “Studies on the involvement of [62] N. Dubarry, W. Du, D. Lane, and F. Pasta, “Improved the exopolysaccharide produced by cystic fibrosis-associated electrotransformation and decreased antibiotic resistance of isolates of the Burkholderia cepacia complex in biofilm the cystic fibrosis pathogen Burkholderia cenocepacia strain formation and in persistence of respiratory infections,” Journal J2315,” Applied and Environmental Microbiology, vol. 76, no. of Clinical Microbiology, vol. 42, no. 7, pp. 3052–3058, 2004. 4, pp. 1095–1102, 2010. International Journal of Microbiology 9

[63] K. A. Datsenko and B. L. Wanner, “One-step inactivation macrophages,” Infection and Immunity, vol. 74, no. 10, pp. of chromosomal genes in Escherichia coli K-12 using PCR 5477–5486, 2006. products,” Proceedings of the National Academy of Sciences of [78] M. S. Sald´ıas, J. Lamothe, R. Wu, and M. A. Valvano, the United States of America, vol. 97, no. 12, pp. 6640–6645, “Burkholderia cenocepacia requires the RpoN sigma factor 2000. for biofilm formation and intracellular trafficking within [64] B. Jia, J.-K. Yang, W.-S. Liu, X. Li, and Y.-J. Yan, “Homologous macrophages,” Infection and Immunity,vol.76,no.3,pp. overexpression of a lipase from Burkholderia cepacia using the 1059–1067, 2008. lambda Red recombinase system,” Biotechnology Letters, vol. [79] M. S. Sald´ıas and M. A. Valvano, “Interactions of Burkholderia 32, no. 4, pp. 521–526, 2010. cenocepacia and other Burkholderia cepacia complex bacteria [65] T. A. Urban, J. B. Goldberg, J. F. Forstner, and U. S. Sajjan, with epithelial and phagocytic cells,” Microbiology, vol. 155, “Cable pili and the 22-kilodalton adhesin are required for no. 9, pp. 2809–2817, 2009. Burkholderia cenocepacia binding to and transmigration across the squamous epithelium,” Infection and Immunity, vol. 73, no. 9, pp. 5426–5437, 2005. [66] T. A. Urban, A. Griffith,A.M.Torok,M.E.Smolkin,J. L. Burns, and J. B. Goldberg, “Contribution of Burkholderia cenocepacia flagella to infectivity and inflammation,” Infection and Immunity, vol. 72, no. 9, pp. 5126–5134, 2004. [67] M. Tomich, A. Griffith,C.A.Herfst,J.L.Burns,andC.D. Mohr, “Attenuated virulence of a Burkholderia cepacia type III secretion mutant in a murine model of infection,” Infection and Immunity, vol. 71, no. 3, pp. 1405–1415, 2003. [68]D.F.Aubert,R.S.Flannagan,andM.A.Valvano,“Anovel sensor kinase-response regulator hybrid controls biofilm for- mation and type VI secretion system activity in Burkholderia cenocepacia,” Infection and Immunity, vol. 76, no. 5, pp. 1979– 1991, 2008. [69] A. D. Vinion-Dubiel and J. B. Goldberg, “Lipopolysaccha- ride of Burkholderia cepacia complex,” Journal of Endotoxin Research, vol. 9, no. 4, pp. 201–213, 2003. [70] M. S. Thomas, “Iron acquisition mechanisms of the Burkholderia cepacia complex,” BioMetals,vol.20,no.3-4,pp. 431–452, 2007. [71] C. Kooi, B. Subsin, R. Chen, B. Pohorelic, and P. A. Sokol, “Burkholderia cenocepacia ZmpB is a broad-specificity zinc metalloprotease involved in virulence,” Infection and Immu- nity, vol. 74, no. 7, pp. 4083–4093, 2006. [72]M.L.Hutchison,I.R.Poxton,andJ.R.W.Govan,“Burkholde- ria cepacia produces a hemolysin that is capable of inducing apoptosis and degranulation of mammalian phagocytes,” Infection and Immunity, vol. 66, no. 5, pp. 2033–2039, 1998. [73] P. A. Sokol, R. Malott, K. Riedel, and L. Eberl, “Communica- tion systems in the genus Burkholderia: global regulators and targets for novel antipathogenic drugs,” Future Microbiology, vol. 2, no. 5, pp. 555–563, 2007. [74] Y. Deng, C. Boon, L. Eberl, and L.-H. Zhang, “Differential modulation of Burkholderia cenocepacia virulence and energy metabolism by the quorum-sensing signal BDSF and its synthase,” Journal of Bacteriology, vol. 191, no. 23, pp. 7270– 7278, 2009. [75] J. H. Leitao,S.A.Sousa,A.S.Ferreira,C.G.Ramos,I.N.˜ Silva, and L. M. Moreira, “Pathogenicity, virulence factors, and strategies to fight against Burkholderia cepacia complex pathogens and related species,” Applied Microbiology and Biotechnology, vol. 87, no. 1, pp. 31–40, 2010. [76] R. S. Flannagan, D. Aubert, C. Kooi, P. A. Sokol, and M. A. Valvano, “Burkholderia cenocepacia requires a periplasmic HtrA protease for growth under thermal and osmotic stress and for survival in vivo,” Infection and Immunity, vol. 75, no. 4, pp. 1679–1689, 2007. [77] K. E. Maloney and M. A. Valvano, “The mgtC gene of Burkholderia cenocepacia is required for growth under mag- nesium limitation conditions and intracellular survival in Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 851946, 10 pages doi:10.1155/2011/851946

Research Article Molecular Detection of Persistent Francisella tularensis Subspecies holarctica in Natural Waters

T. Broman, 1 J. Thelaus,1 A.-C. Andersson,1 S. Backman,¨ 1 P. Wi ks t r om,¨ 1 E. Larsson,1 M. Granberg,1 L. Karlsson,1 E. Back,¨ 2 H. Eliasson,2 R. Mattsson,3 A. Sjostedt,¨ 4 and M. Forsman1

1 Department of CBRN Defence and Security, Swedish Defence Research Agency, 901 82 Umea,˚ Sweden 2 Department of Infectious Diseases, Orebro¨ University Hospital, 701 85 Orebro,¨ Sweden 3 National Veterinary Institute, 751 89 Uppsala, Sweden 4 Department of Clinical Microbiology, Umea˚ University, 901 87 Umea,˚ Sweden

Correspondence should be addressed to J. Thelaus, [email protected]

Received 28 May 2010; Accepted 9 July 2010

Academic Editor: Max Teplitski

Copyright © 2011 T. Broman et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Tularemia, caused by the bacterium Francisella tularensis,whereF. tularensis subspecies holarctica has long been the cause of endemic disease in parts of northern Sweden. Despite this, our understanding of the natural life-cycle of the organism is still limited. During three years, we collected surface water samples (n = 341) and sediment samples (n = 245) in two areas in Sweden with endemic tularemia. Real-time PCR screening demonstrated the presence of F. tularenis lpnA sequences in 108 (32%) and 48 (20%) of the samples, respectively. The 16S rRNA sequences from those samples all grouped to the species F. tularensis. Analysis of the FtM19InDel region of lpnA-positive samples from selected sampling points confirmed the presence of F. tularensis subspecies holarctica-specific sequences. These sequences were detected in water sampled during both outbreak and nonoutbreak years. Our results indicate that diverse F. tularensis-like organisms, including F. tularensis subsp. holarctica, persist in natural waters and sediments in the investigated areas with endemic tularemia.

1. Introduction The epizootiology of F. tularensis is complex, involving numerous wildlife species and several potential vectors for Tularemia is a zoonotic disease caused by the bacterium its transmission as a disease-causing agent. Indeed, tularemia Francisella tularensis. At present, four subspecies of F. has been detected in approximately 250 wildlife species, tularensis are suggested [1, 2], two of which are of clinical giving F. tularensis a broader host range than any other importance (subsp. tularensis and holarctica,[2]). Francisella known zoonotic disease-causing organism [11]. Various tularensis subsp. tularensis strains only occur in North bloodsucking arthropods have been found naturally infected America [3, 4]whereasF. tularensis subsp. holarctica with the bacterium, like ticks, tabanid flies, midges, mites, strains are found throughout the Northern Hemisphere fleas, lice, and mosquitoes [12]. Nevertheless, local tularemia [5]. Francisella tularensis is categorized as a category A outbreaks are often patchy, occurring around natural foci in potential bioterrorism agent. Recently, it was established geographically restricted areas, typically in association with that diverse Francisella-like bacteria exist in the environment just one or a few key mammalian and arthropod species. (in soil, seawater, and fish) [6–10]. These Francisella-like Tularemia (caused by F. tularensis subsp. holarctica organisms cluster in various genetic clades together with tick strains) is endemic in areas of northern Sweden and, during endosymbionts, fish pathogens, and bacteria detected in soil the past decade, has emerged in areas of central Sweden and sediment [5]. too. In these areas, it is a local public health threat since it 2 International Journal of Microbiology occurs at a high frequency, especially in late summer and on the knowledge of local physicians about the geographical autumn. The reasons for its geographical distribution and distribution of human tularemia cases (Figure 1). In 2004 seasonal occurrence are unknown. It is generally thought that and 2005, there were ten sampling points in each study naturally infected mosquitoes are the major transmission region. vectors of tularemia in Sweden [13], with the occurrence of naturally infected Aedes cinereus reported as early as 1942 2.2. Small Rodents. Rodents were collected using live-traps [14]. It is still not clear how mosquito vectors acquire the (under ethical permit number C 118/3 issued by the bacteria. Local Ethical Committee on Laboratory Animals in Umea,˚ Francisella tularensis subsp. holarctica is often associated Sweden), baited with a mixture of carrots, potatoes, oatmeal, with water environments like streams, ponds, lakes, and and pieces of apples, from mid-May to mid-September. rivers [15, 16]. Water-borne transmission of tularemia Trapping was performed in Ljusdal (on four occasions) and (subsp. holarctica) has been frequently reported [17–25]. The Orebro¨ (two occasions) during 2003, and in Orebro¨ (five presence of F. tularensis in water and sediments has been occasions) during 2004. At each sampling during 2003, traps proven by its isolation from laboratory animals inoculated were set for five days, and in 2004 traps were set for two with samples [21]. However, the role of natural waters in the tofivedays.Trapswerecheckedevery12hours.Trapped long-term survival of clinically relevant subspecies is not well rodents were anesthetized using halothane and euthanized characterized, as it has not been possible to directly culture through cervical dislocation. Carcasses were kept refrigerated the bacteria from water samples. However, experiments during transportation to a local laboratory, where spleen have shown that F. tularensis subsp. holarctica survive in and liver samples were prepared and deep-frozen (within watercourses, possibly in association with protozoa [26–29]. four hours of euthanization) until further analysis. After The larvae of flood-water mosquitoes significantly prey on thawing, spleen and liver samples were used for F. tularensis the protozoan community [30], and may well be exposed to cultivation and DNA preparation for polymerase chain F. tularensis subsp. holarctica in this way. reaction (PCR) analysis. In the study presented here, we used molecular detection Francisella tularensis wasculturedonmodifiedThayer- techniques to confirm the persistence of F. tularensis subsp. ◦ Martin agar plates [32]at37Cin5%CO2 for six days, holarctica DNA in natural surface waters over a three-year and its growth was confirmed by slide agglutination with a period in two Swedish tularemia regions. Water, sediments, commercial antiserum (Difco Laboratories, Augsburg, Ger- and small rodents were sampled in two regions with some of many). DNA was purified using the guanidine isothiocyanate the highest incidences of tularemia reported in Sweden. method [33]. This was followed by real-time PCR probe- based lpnA assays [27] and typing with multiple locus 2. Materials and Methods variable-number tandem repeat analysis (MLVA) [3]. 2.1. Study Regions. The study was conducted in two regions with reoccurring tularemia in Sweden: Ljusdal and Orebro.¨ 2.3. Water and Sediment Samples. Sampleswerecollected The municipality of Ljusdal (61◦ 500 N16◦50 E), with on several occasions during summer, from mid-May to a population of 19 384 (2005), situated in the county of mid-September, during three consecutive years (on four, Gavleborg¨ (a population of 275 994, 2005), has a history seven and three occasions in Ljusdal and on two, eight and ¨ of tularemia outbreaks dating back to at least the 1930s. three occasions in Orebro, during the years 2003, 2004, and This region is typical of endemic tularemia regions, in which 2005, resp.) (Figure 1). Samples were collected from both outbreaks occur in geographically restricted areas at irregular surface water and sediment. In 2003, sediment samples were intervals. Since 1931, at least 2500 human cases have been collected from two of the sampling points in Ljusdal and ¨ recorded in the county. Data indicate that most patients from all of the sampling points in Orebro. In 2004 and have acquired the infection within or close to the Ljusdal 2005, sediments were sampled from all sampling points. municipality or on a nearby golf course (Figure 1)[31]. The samples were collected as single-grab samples in 100 ml In recent years, the disease has emerged in Orebro¨ plastic tubes. They were refrigerated during transportation ¨ county (59◦ 160 N15◦ 120 E), located 364 km to the laboratory (within 24 and 48 hours for Orebro and south of Ljusdal, with a population of 274 121 Ljusdal samples, resp.). DNA extraction was performed upon − ◦ (2005). Before 2000 only a handful of cases were arrival and the purified DNA was stored at 20 C until reported from the county and limited numbers of cases further analysis. occurred in 2001 and 2002. However, between 2003 and 2005, 229 human cases of tularemia were reported 2.4. DNA Purification and PCR Analysis of Water and (http://www.smi.se/in-english/statistics/tularaemia/) Sediment. Two mL of each water or sediment sample was (Table 1). The tularemia cases have clustered in distinct centrifuged at 16 000 × g for 1 hour, 1.9 mL of the resulting areas, namely: (i) along the west shores of Lake Hjalmaren,¨ supernatant was discarded and DNA was extracted from (ii) close to the city center along River Svartan,˚ (iii) in an the remaining volume using a SoilMaster DNA Extraction area with allotment gardens close to the city center, and (iv) Kit according to the recommendations of the manufacturer around Lake Langen˚ (Figure 1)[31]. for environmental water samples (Epicentre Biotechnologies, For the first year sampling (2003), several sampling Madison, WI, USA). To increase the yield of DNA the points were chosen (26 in Ljusdal and 21 in Orebro),¨ based samples were incubated at 37◦C for ten minutes, without International Journal of Microbiology 3

Orebro¨ Ljusdal N N A N

I B

Lake Hj

River almaren¨ Ljusnan II Golf course

0 1750 3500 7000 0 1500 3000 (meters) 0 250 500 (meters) (kilometers)

Figure 1: Sampling locations (black triangles), in the Ljusdal and the Orebro¨ area. Roman numerals (I, II) and letters (A, B) indicate sampling points selected for detailed analysis. Waterways are represented in white, urban areas are shaded. Each encircled × shows the probable place of disease transmission for patients infected in 1998–2005 (Ljusdal) and 2003-2004 (Orebro)¨ [31].

shaking, after Proteinase K treatment. The resulting DNA (Beckman Coulter Inc., Fullerton, CA, USA) after mixing pellet was resuspended in 60 µLofTEbuffer and either 1 µL of the PCR products from each amplification with frozen and stored or immediately subjected to PCR analysis. standards (from a CEQ DNA size standard kit-400) in sample As negative controls, 2 mL samples of sterile water were loading solution according to the manufacturer’s manual. treated according to the protocol described above. Sample preparation, PCR reaction preparation and thermal cycling 2.5. Sequencing. The lpnA and FtM19InDel PCR amplicons were separated and performed in different rooms. were purified using MicroSpin S-400 HR columns (GE Water and sediment samples were screened for F. Healthcare Bio-Sciences, Uppsala, Sweden), then sequenced tularensis using a real-time PCR probe-based assay using a CEQ8800 Genetic Analysis System and a DTCS Quick (iQFt1F/R) for detection of the F. tularensis-specific Start kit (Beckman Coulter Inc. Fullerton, CA, USA) accord- lpnA sequence, as previously described [27]. To detect ing to the manufacturer’s instructions, with iQFt1F/R and false negative results caused by PCR inhibitory substances, FtM19InDelF/R primers, respectively. Acquired sequences the assay also included an internal control probe [27]. All were deposited with GenBank under accession numbers samples were analyzed in at least triplicate PCR reactions. FJ94649, FJ946492 to FJ946499 (lpnA) and FJ946500 to Samples from selected sampling points (described below) FJ946512 (FtM19InDel). were further subjected to a F. tularensis subsp. holarctica- specific-PCR based on the 30 bp-deletion region FtM19 [3, 4, 34–36], followed by fragment size analysis [34]. Each 2.6. 16S rRNA Cloning and Sequencing. Amplification, reaction consisted of 1 µL template, 1x Amplitaq GOLD direct cloning and subsequent sequencing of 16S rRNA PCR buffer, 40 µM each of the primers FtM19InDelF/R was performed on samples chosen for detailed stud- (WELLRED 5-CCAGTACAAACTCAATT TGGTTATCATC- ies. 16S rRNA Francisella-specific primers Fr153F0.1 (5- 3 and 5-GTTTCAGAATTCATTTTTGTCCGTAA-3), GCCCATTTGAGGGGGATACC-3) and Fr1281R0.1 (5-  2.6 mM MgCl2, 1 M betaine, 0.2 mM dNTP, 0.5 U Amplitaq GGACTAAGAGTACCTTTTTGAGT-3 ) were used as previ- GOLD polymerase, and MilliQ water to give a total volume ously described [6].16SrRNAPCRproductswerepurified of 12.5 µL. An initial denaturation at 94◦Cfor2minutes on SeaKem agarose gels (Cambrex North Brunswick, Inc., was followed by 50 cycles of 94◦C for 30 seconds, 60◦Cfor North Brunswick, NJ, USA) and excised bands were eluted 30 s and 72◦C for 30 seconds, followed by final incubation using GenElute Gel Spin Columns (Sigma-Aldrich, St. Louis, at 72◦C for 5 minutes in a MyCycler thermal cycler (Bio-Rad MO, USA). Products were cloned into the pCRII vector using Laboratories, Hercules, CA). Positive control mixtures a TOPO-TA Cloning Kit according to the protocol recom- using DNA from F. tularensis subsp. holarctica, and negative mended by the manufacturer (Invitrogen Co., Carlsbad, CA, control mixtures without a template, were included in each USA). Fifty clones, representing each PCR reaction, were PCR run. The resulting amplicons were sized by capillary subsequently picked and stored in glycerol at −70◦Cprior electrophoresis using a CEQ 8800 Genetic Analysis System to sequencing. Plasmid DNA was isolated from overnight 4 International Journal of Microbiology cultures using an E.Z.N.A. Plasmid Miniprep Kit (Omega Ljusdal 120 Bio-Tek Inc., Doraville, GA, USA) and sequenced using universal M13F and M13R primers. Detected sequences 45% 100 (∼1150 bp) were deposited with GenBank under accession numbers DQ994171 to DQ994200. 80 21% 30%

2.7. Phylogenetic Analysis of Sequence Data. To evaluate the 60 sequence similarity of the Francisella sequences obtained, ref- erence sequences from GenBank were included in ClustalW 40 alignment, performed using MEGA version 3.1 [37]. For the 13% 17% 16S rRNA sequences, a phylogenetic tree was generated using 20 maximum parsimony analysis and bootstrapping. 0% 0 Sur. Sed. Sur. Sed. Sur. Sed. 3. Results 2003 2004 2005 3.1. Human Cases. During the three-year study period, (a) 19 human tularemia cases were verified in the Ljusdal Orebro¨ area (County Medical Officer, Gavle-Sandviken,¨ Personal 120 ¨ communication) and 229 in Orebro county (Table 1). 105

90 3.2. Small Rodents. During the first year of the study 97 35% 18% rodents (60 in Ljusdal and 37 in Orebro,¨ Table 1)werecaught 75 alive. The presence of F. tularensis in spleen and liver samples 60 of the rodents was investigated by culture and PCR analysis. Two rodents, a water vole (Arvicola terrestris)andayellow- 45 37% 7% necked mouse (Apodemus flavicollis) were infected with F. 30 11% 22% tularensis, as demonstrated by the culture assays and PCR analysis. Both were from the Orebro¨ area, but from different 15 sampling points. Genotyping identified the isolates as F. 0 tularensis subsp. holarctica. Subtyping by MLVA showed that Sur. Sed. Sur. Sed. Sur. Sed. the isolates were distinct, and thus likely contracted from 2003 2004 2005 ff di erent sources (data not shown). The rodent population (b) declined in 2004 and despite intensified sampling only seven individuals was trapped in Orebro,¨ all of which were Figure 2: Stacked bar graph showing numbers of positive (black) Francisella-negative (Table 1). and negative (light grey) samples during the sampling period (2003–2005) in Ljusdal (a) and Orebro¨ (b). The percentages of samples positive for F. tularensis are shown. (sur., water samples; 3.3. Francisella tularensis in Water and Sediment Samples. sed., sediment samples). During the three-year study we collected 341 water surface samples and 245 sediment samples in total. The F. tularensis- specific lpnA sequence was detected in 108 (32%) and 48 we amplified 16S rRNA using the 16S rRNA primers for (20%) samples, respectively, using real-time PCR screening Francisella-like organisms reported by Barns et al. 2005 [6]. (Figure 2). The sequences were detected in samples obtained In total, 30 sequences were obtained, all of which grouped at several sampling points in both study regions and in each exclusively to the subspecies of species F. tularensis in the year. phylogenetic analysis (Figure 4). The lpnA-positive samples from the selected sampling 3.4. Detailed Studies of Selected Sampling Points. Two sam- points were subjected to FtM19InDel fragment size analysis, pling points from Ljusdal (I and II) and Orebro¨ (A and which has been shown to differentiate F. tularensis subsp. B) that were consistently positive in the lpnA assay were holarctica from other F. tularensis subspecies and Francisella retrospectively selected for detailed analysis (Figure 1). A -like bacteria. The fragments amplified corresponded to F. total of 54 samples were analysed from these four locations tularensis subsp.holarctica(100 bp) in 16 of the samples over the three-year sampling period and 24 of the samples and to non-holarctica Francisella-like bacteria (130 bp) in were lpnA-positive (Table 1),anditprovedpossibleto two. Remaining 6 samples were negative presumably because sequence eight of these (Figure 3). The sequences were the FtM19InDel primers are less sensitive than the lpnA compared with published sequences from representatives of primers. The sequences of the 100 bp amplicons (n = all described F. tularensis subspecies and their closest known 12) showed high sequence similarities (95%–100%) to relatives, and found to be 95%–100% similar (Figure 3). those of previously published holarctica strains (Figure 5 In order to further investigate the occurrence of Fran- and Table 1). The 130 bp, full-size amplicons, (n = 2) cisella DNA in water from the selected sampling points, aligned most closely with F. tularensis subsp. mediasiatica International Journal of Microbiology 5

Table 1: Reported human tularemia cases (County Medical Officer, Gavle-Sandviken,¨ personal communication, and Swedish Institute for Infectious Disease Control, SMI, Orebro¨ County) and F. tularensis subsp. holarctica culture-positive rodents in the Ljusdal and Orebro¨ areas during the study period. The total numbers of trapped rodents are shown in parentheses. Results of the molecular analysis of water samples from the four sampling points selected for detailed examination. Positive lpnA assay results indicate the presence of F. tularensis. The length of resulting FtM19InDel sequences indicates the presence of F. tularensis subsp. holarctica or other F. tularensis subspecies (other ssp.). n.d., not detected; n.s., not sampled.

Ljusdal Orebro¨ IIIAB Human cases 1 150 Rodents 0 (60) 2 (37) Water 2003 No. of samples tested (4) (4) (2) (2) lpnA 2 2 n.d. 1 FtM19InDel 1 1 1 holarctica holarctica n.d. holarctica Human cases 0 54 Rodents n.s. 0 (7) Water 2004 No. of samples tested (7) (7) (8) (8) lpnA 2124 FtM19InDel2123 holarctica holarctica holarctica holarctica Human cases 18 25 Rodents n.s. n.s. Water 2005 No. of samples tested (3) (3) (3) (3) lpnA 3331 FtM19InDel 1+1 2 1+1 1 holarctica+otherssp. holarctica holarctica+otherssp. holarctica

624 634 644 654 664 674 684 694 . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . F. t. holarctica LVS TTAGGTGGCTCTG- ATGATGCAAAAGCTTCAGCTAAAGATACTGCTGC - TGCTCAGACAGCTACTACTGAGCAAGC F. t. tularensis Schu S4 (AI) ...... - ...... - ...... F. t. tularensis WY (AII) ...... - ...... - ...... F. t. mediasiatica FSC147 ...... - ...... - ...... F. t. novicida U112 ...... - ...... - ...... Orebro¨ B-03 ...... - ...... - ...... - . . Ljusdal I -03 ...... - ...... - ...... Ljusdal II-03 ...... - ...... - ...... Orebro¨ B-04 C - ...... - ...... T ...... - ...... Orebro¨ A-04 - ...... - ...... - ...... A ...... Orebro¨ A-04 ...... C ...... - ...... Ljusdal II-04 ...... - ...... - ...... Orebro¨ A-05 ...... C ...... T ...... - ...... Dermacentor variabilis . - . T . . . . T . . . . - . . . . . T ...... AC ...... AACAAG . . . CT . CA . CAACAGCC . T . GCT Amblyomma maculatum K - . T . . . . T . . . . - . . . . . T ...... ACT . A ...... AAGAAG . . . CT . CA . CAACAGCC . T . GCT Dermacentor andersonii . - . T . . . . T . . . . - . . . . . T ...... AC . . T ...... AACAAG . . . CT . CA . CAACAGCC . T . GCT Dermacentor hunteri . - . T . . . . T . . . . - . . . . . T ...... AC . . T ...... AACAAG . . . CT . CACCAACAGCC . T . GCT Dermacentor variabilis . - . T . . . . T . . . . - . . . . . T ...... AC ...... T . . . Y . AACAAG . . . CT . CA . CAACGGCC . T . GCT Francisella philomiragia A . . A . CCTACTGCTCAAGAT . . . GCA . CA . . AGCTTCAGATGCAACAGC . . T . . . AG . TT . . . . - - -GAGCAC . . . Francisella piscicida A . . A . CCTGCTGCTCA . GAC . . . GCAACG . . AGCTTCAGATGC . ACAGC . . T . . . AG . TT . . . . . G . AAGCAC . . .

Figure 3: Multiple alignment of lpnA sequences obtained from Ljusdal and Orebro¨ with previously published sequences of Francisella species and subspecies and Francisella-like endosymbionts (FLE). The nucleotide positions 620 to 695 refer to F. t. holarctica LVS (M32059). Reference sequences from GenBank: F. t. holarctica LVS (M32059), F. t. tularensis strain WY96-3418 (CP000608), F. t. tularensis strain Schu S4 (NC 006570), F. t. mediasiatica strain FSC147 (NC 010677), F. t. novicida strain U112 (CP000439), Dermacentor variabilis FLE (AY375420), Amblyomma maculatum FLE (AY375422), Dermacentor andersonii FLE (AY375413), Dermacentor hunteri FLE (AY375417), Dermacentor variabilis FLE (AY375421), F. philomiragia (AY243030) and F. piscicida (DQ825765). 6 International Journal of Microbiology

Water B24 98 Water B33 F. tularensis subsp. novicida CIP 56.12 (AY928396) Water B13 Water B3 F. tularensis subsp. tularensis FSC 054 (AY968224) F. tularensis subsp. holarctica FSC 022 (AY968228) Water A24 Water B19 Water B4 F. tularensis subsp. mediasiatica FSC 147 (AJ698863) Water A4 F. tularensis subsp. holarctica FSC 090 (AJ698864) F. tularensis subsp. mediasiatica FSC 147 (AY968234) F. tularensis subsp. novicida FSC 040 (AY968237) F. tularensis subsp. tularensis FSC 043 (AJ698865) F. tularensis subsp. tularensis SchuS4 (AJ749949) F. tularensis subsp. holarctica LVS (AJ698866) F. tularensis subsp. holarctica UT01-1901-(AY968232) Water B23 86 Water B32 Water B27 94 Water B36 Water B28 74 Water B37 Water B16 Water B26 Water B14 Water B35 Water B39 Water B18 Water B22 99 Water B31 Water B11 Water B1 76 Water B38 Water B29 Water B20 Water B17 Water B12 Water B2 Water B15 Soil 039c (AY968286) Soil 034a (AY968283) Soil 039a (AY968284) Soil 039b (AY968285) Wolbachia persica (M21292) Ornithodorus moubata symbiont (AB001522) 98 Dermacentor variabilis symbiont (AY805305) 97 Dermacentor variabilis symbiont (AY805307) Soil 027a (AY968287) Soil 027c (AY968289) 99 72 Soil 027b (AY968288) Soil 027d (AY968290) Soil 034b (AY968301) 90 Soil 034c (AY968302) F. philomiragia ATCC 25015 (AJ698862) 99 F. philomiragia strain 2669 (AY243027) F. philomiragia ATCC 25017 (AY928395) 76 Soil 039d (AY968300) Soil 005b (AY968294) Soil 015d (AY968298) 97 Soil 015c (AY968296) Soil 013b (AY968299) 83 Soil 015a (AY968292) Soil 005a (AY968291) Soil 013a (AY968295) Soil 005c (AY968297) Soil 015b (AY968293) Soil 045a (AY968303) 99 Soil 045b (AY968304) Soil 045c (AY968305) Figure 4: Phylogenetic analysis based on Francisella 16S rRNA sequences obtained from water samples in Orebro¨ (Water A/B, this study), environmental soil samples [6] and reference sequences from GenBank. The samples are named according to the sampling points, Orebro¨ A and B (Figure 1). Analysis was performed using maximum parsimony. International Journal of Microbiology 7

10 20 30 40 50 60 70 80 90 100 110 . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . F. philomiragia oakridge TATCATTTATT-AGAAATTACATCTCG-CGGATAATTTTAAATTTCT-CATAATTTTCAAATTATTAAATAAATTTCTCATATAAAAATTTACGGATAAAAATGAATTCTGAAACTAA F. t. novicida U112 ...... C. . . . - ...... G. . - ...... - ...... - ...... T . . A ...... C ...... C ...... F. t. tularensis WY (AII) ...... C. . . . - ...... G. . - ...... - ...... - ...... T ...... C ...... C ...... F. t. tularensis Schu S4 (AI) ...... C. . . . - ...... G. . - ...... - ...... - ...... T ...... C ...... C ...... F. t. mediasiatica FSC147 ...... C. . . . - ...... G. . - ...... - ...... - ...... TG. . .G...... C ...... C ...... Orebro¨ A-05 ------. . A ...... G. . - ...... - ...... - ...... TG. . .G...... C ...... C ...... A . . AG. . AC- - Ljusdal I-05 ...... C. . . . - ...... G. . - ...... - ...... - ...... TG. . .G...... C ...... C ...... - - - - ...... G F. t. holarctica LVS ...... C. . . . - ...... G. . - ...... - ...... - . . . . ------C ...... C ...... - - - Ljusdal I-03 ...... C. . . . - ...... G. . - ...... - ...... - . . . . ------C ...... C ...... - ...... - - Ljusdal II-03 C . . . . . C. . . . - . . T ...... G. . - ...... - ...... - . . . . . ------C ...... C ...... A ...... A- - - Orebro¨ B-03 ...... C. . . . - . . T ...... G. . - ...... - ...... - . . . . ------C ...... C ...... A- - - Ljusdal I-04 ------. . . . .G. . - ...... - ...... - . . . . ------C ...... C ...... - - . C . .GAA . - Ljusdal II-04 ...... C. . . . T ...... G. TG...... - . . . - ...... - . . . . ------C .G. . ------Orebro¨ B-04 ------. . - ...... G. . . .GG. . A ...... - ...... A . . . . ------C ...... G...... C ...... - - - Orebro¨ B-04 ------. . . - ...... G. . . . .G. . A ...... - ...... - . . . . ------C ...... C ...... - - - Orebro¨ A-04 ...... C. . . . - ...... G. . - ...... - ...... - . . . . ------C ...... C ...... - - - Ljusdal I-05 ...... C. . . . AGT ...... G. -G...... - ...... - . . . . ------C ...... C ...... CTGAG. . - - - - Ljusdal II-05 ------. . . .G. . . . .G. . - ...... C . . - ...... - . . . . ------C ...... C ...... - - - Orebro¨ B-05 ...... C. . . . - ...... G. . - ...... - ...... - . . . . ------C ...... G- . . . . . C ...... ------Orebro¨ A-05 ...... C. . . . - ...... G. . - ...... - ...... - . . . . ------C ...... C ...... ATCATG. AAC- Figure 5: Multiple alignment of FtM19InDel sequences from Ljusdal and Orebro¨ with previously published sequences of Francisella species and subspecies. In this alignment the F. tularensis subsp. holarctica specific deletion is located from position 52 to 82. Reference sequences from GenBank: F. t. holarctica LVS (M32059), F. t. tularensis strain WY96-3418 (CP000608), F. t. tularensis strain Schu S4 (NC 006570), F. t. mediasiatica strain FSC147 (NC 010677), F. t. novicida strain U112 (CP000439) and F. philomiragia (AY243030).

(Figure 5 and Table 1). Francisella tularensis subsp. holarctica teria, including F. philomiragia. This implies that the water sequences were detected in samples obtained at each of the environments from which we cloned 16S rRNA sequences, four selected sampling points during all three years (Table 1). were more selective for F. tularensis subspecies than the soil and sediment samples analyzed by Barns et al. [6]. 4. Discussion Although related strains F. philomiragia and F. tularensis subsp. novicida can be cultured directly from water [10], this In this study, we used a molecular method to demonstrate the is not currently true for the clinically significant subspecies of occurrence of the clinically relevant subspecies F. tularensis F. tularensis, tularensis and holarctica. Nevertheless, the pres- subsp. holarctica in water and sediment samples from two ence of subspecies holarctica in water and sediments has been tularemia areas in Sweden, during three consecutive years. proven through the isolation of culturable bacteria from lab- Water and sediment samples from the tularemia areas were oratory animals inoculated with water samples [21]. In order screened for the presence of F. tularensis DNA using a to identify F. tularensis subsp. holarctica in water samples we PCR assay to amplify the lpnA gene [33]. This generates developed the FtM19InDel assay. We previously analyzed a aproductfromallfourF. tularensis subspecies, but not total of 688 F. tularensis strains for this marker and found a from other Francisella spp. or Francisella-like endosymbionts 100% correlation between the 30-bp deletion and subspecies (FLE). Although not quantitative, the detection limit of holarctica (unpublished results). Here, we amplified the F. the lpnA assay used here has been estimated to be 103 tularensis subsp. holarctica sequence (Figure 5) in the samples bacteria per mL in natural water samples [27]. Therefore, selected for detailed analysis (i.e., those from four sampling the presence of PCR products from the water and sediment points that yielded samples with consistently positive results samples indicated the presence of F. tularensis in fairly in the initial screen using the lpnA assay, Table 1 and high numbers. The lpnA assay, in contrast to previously Figure 1). On the contrary, the causative agent of human performed animal inoculations [21], is potentially capable of tularemia in North America, F. tularensis subsp. tularensis detecting both pathogenic F. tularensis (i.e., subsp. tularensis (type A), was not detected in environmental samples during and holarctica) and nonpathogenic F. tularensis. This might an ongoing outbreak in the active natural focus on Martha’s have contributed to the high frequency of F. tularensis in our Vineyard (MA, USA) [38]. These findings may reflect dif- samples over the three-year study period (108 positive out of ferences in the environmental stability between F. tularensis 341 water samples analyzed). Since we initially expected low subsp. tularensis and holarctica strains possibly due to differ- frequencies of F. tularensis-positive samples, we investigated ing ecological niches and reservoirs for the two subspecies. a large number of sampling points during the first year of the Using the FtM19InDel assay we also obtained full-length study. However, due to the high detection rate, the number fragments corresponding to non-holarctica F. tularensis sub- of sampling points was reduced in the following two years. species, in samples from both Ljusdal and Orebro¨ (Figure 5 Sequence analysis of 16S rRNA clones amplified from and Table 1). Surprisingly, the sequences of these full-length lpnA-positive samples confirmed that the template organ- InDelFt-M19 fragments showed high similarity to that of F. isms exclusively grouped with the subspecies within species tularensis subsp. mediasiatica. This subspecies occurs as rare F. tularensis. In previously reported environmental study human pathogens in Kazakhstan and Uzbekistan, and has by Barns et al. 2005 [6], in which essentially the same virulence comparable to that of strains of F. tularensis subsp. procedure was used, the targeted bacteria were found to holarctica [39]. However, all clinical isolates originating from consist of a mixture of distantly related Francisella-like bac- the Orebro¨ and Ljusdal regions that we have typed so 8 International Journal of Microbiology far (n = 151), belonged without exception to the subsp. ubiquitous protozoa might be an important environmental holarctica [31]. Therefore, it is highly unlikely that the F. reservoir for the bacterium [26–29]. The aquatic systems tularensis subsp. mediasiatica-like sequences detected in this sampled in this study (Orebro¨ and Ljusdal tularemia areas) region were derived from a human pathogenic clone. Instead, could be characterized as eutrophic systems [46]. In such this finding may reflect the diversity of Francisella and F. systems with high nutrient availability, the bacterial popula- tularensis-like organisms in the environment, as evidenced tion has been shown to be structured by protozoan predation by a growing body of data [6, 10]. pressure [46, 47]. In turn, mosquito larvae, mainly of the Interestingly, we detected F. tularensis subsp. holarctica in species Aedes sticticus and other flood-water mosquitoes, water sampled in Ljusdal during 2004, when no human cases have been shown to exert a significant predatory impact on were recorded in the area. Thus, the presence of F. tularensis a protozoan population in a temporarily flooded wetland subsp. holarctica in water is not necessarily sufficient for [30]. Altogether, this indicates that mosquito larvae may spread of the disease to susceptible hosts. Occurrence of the be exposed to F. tularensis subsp. holarctica in the water bacterium in water during the nonoutbreak year suggests environments investigated here. Accordingly, we identi- that, in addition to the bacterial contamination of water fied F. tularensis DNAinmosquitoesrearedtoadultsin during ongoing outbreaks (from bacteriuria or decomposing the laboratory, from larvae collected in temporary waters carcasses) [16, 40, 41], F. tularensis subsp. holarctica persists in the tularemia area (Orebro)¨ [unpublished, Lundstrom¨ in water between outbreaks. In a recent study, Svensson et al. 2010]. Moreover, Svensson et al. 2009, identified et al. (2009) combined epidemiologic investigations with an association between disease clusters (i.e., locations of high-resolution genotyping of F. tularensis subsp. holarctica tularemia transmission via mosquitoes) and recreational isolates obtained from patients in the same regions, Orebro¨ areas adjacent to water in the Ljusdal and Orebro¨ tularemia and Ljusdal [31]. In line with our results, Svensson et al. areas [31]. As stated above, mosquito bites are the major observed that genetic subpopulations of the bacteria were route of transmission in both study regions [13](Berglund present throughout the tularemia season and persisted over L, personal communication). years [31]. We also detected F. tularensis subsp. holarctica in The natural life-cycle of F. tularensis and the environ- water samples from the same sampling points during three mental reservoir of the bacteria have long been subject to consecutive years, indicating that the bacterium may persist speculation. Our working hypothesis is that F tularensis in water for several years. The intervals between tularemia subsp. holarctica persists in water and/or sediment between outbreaks often span several years, or even decades. Experi- tularemia outbreaks. Data presented here support this ence from rodent models and human outbreaks suggest that hypothesis, although the factors promoting the spread of the there is no healthy chronic carrier stage [42]. Thus, neither bacterium to susceptible hosts remain to be revealed. shedding nor carcass contamination can explain the bacterial persistence between outbreaks. We included analysis of rodents to investigate a potential Acknowledgments correlation between persistence of F. tularensis in water and TheauthorsaregratefultoPerBulow,¨ Lena and Leif rodents. In 2003, 97 rodents were live caught and investigated Silversund for assistance in the sampling and packaging of for the presence of the bacterium. Two of the rodents, both samples for transport from Orebro¨ and Ljusdal. This project caught in Orebro¨ at different sampling points, were positive was supported by grants from the Swedish Research Council by culturing. The obtained isolates were identified as two for Environment, Agricultural Sciences and Spatial Planning distinct F. tularensis subsp. holarctica strains and thus likely (Formas no.#209-2006-1311), the Swedish Armed Forces, contracted from different sources. Due to a drop in rodent and the Swedish Civil Contingencies Agency. population sizes, only seven individuals were trapped in Orebro¨ during 2004, despite extended number of trap nights as compared to 2003. All seven were F. tularensis negative. References However, several water samples were positive for the presence of F. tularensis subspecies holarctica at the same sampling [1]B.Huber,R.Escudero,H.Busse,etal.,“Descriptionof points. Taken together, F. tularensis subsp. holarctica can be Francisella hispaniensis sp. nov., isolated from human blood, found persistent in water also in the absence of infected reclassification of Francisella novicida (Larson et al. 1955) rodents. Moreover, the results show that surveillance of F. Olsufiev et al. 1959 as Francisella tularensis subsp. novicida tularensis in the environment using rodents as sentinels is not comb. nov., and emended description of the genus Fran- cisella,” International Journal of Systematic and Evolutionary reliable over years and between outbreaks. Microbiology, vol. 60, part 8, pp. 1887–1896, 2010. Laboratory experiments have shown that F. tularensis [2]A.B.Sjostedt,¨ D. J. Brenner, N. R. Krieg, J. T. Staley, and G. M. subsp. holarctica can survive in water for months [43]. Garrity, “Family XVII. Francisellaceae, Genus I. Francisella,” However, within days after release in water, the bacterium in Bergey’s Manual of Systematic Bacteriology, vol. 2, pp. 200– enters a viable but nonculturable (VBNC) state [27, 43, 210, Springer, New York, NY, USA, 2005. 44]. Whole-genome sequencing has shown that F. tularensis [3] A. Johansson, J. Farlow, P. Larsson et al., “Worldwide genetic subsp. holarctica has a low metabolic capacity, suggesting that relationships among Francisella tularensis isolates determined it is an obligate host-dependent bacterium [45]. Further, F. by multiple-locus variable-number tandem repeat analysis,” tularensis subsp. holarctica shows enhanced survival when Journal of Bacteriology, vol. 186, no. 17, pp. 5808–5818, co-cultured with certain types of protozoa, indicating that 2004. International Journal of Microbiology 9

[4] J. Farlow, D. M. Wagner, M. Dukerich et al., “Francisella [22] P. Pazdiora, I. Moravkov´ a,´ D. Nocarova,´ M. Velkoborska,´ tularensis in the United States,” Emerging Infectious Diseases, and K. Valeckova,´ “A water-borne epidemic of tularemia in vol. 11, no. 12, pp. 1835–1841, 2005. Chlumcany,” Epidemiol Mikrobiol Imunol, vol. 51, pp. 23–25, [5] P. Keim, A. Johansson, and D. M. Wagner, “Molecular 2002. epidemiology, evolution, and ecology of Francisella,” Annals of [23] R. Reintjes, I. Dedushaj, A. Gjini et al., “Tularemia outbreak the New York Academy of Sciences, vol. 1105, pp. 30–66, 2007. investigation in Kosovo: case control and environmental [6]S.M.Barns,C.C.Grow,R.T.Okinaka,P.Keim,andC.R. studies,” Emerging Infectious Diseases, vol. 8, no. 1, pp. 69–73, Kuske, “Detection of diverse new Francisella-like bacteria in 2002. environmental samples,” Applied and Environmental Microbi- [24] N. I. Tikhenko, V. I. Efremenko, E. I. Omarieva et al., ology, vol. 71, no. 9, pp. 5494–5500, 2005. “Outbreak of tularemia in the Republic of Dagestan,” Zhurnal [7] H. K. Goethert, I. Shani, and S. R. Telford III, “Genotypic Mikrobiologii, Epidemiologii, i Immunobiologii, no. 6, pp. 68– diversity of Francisella tularensis infecting Dermacentor vari- 72, 2001. abilis ticks on Martha’s Vineyard, Massachusetts,” Journal of [25] S. V. Rogutskii, M. M. Khramtsov, A. V. Avchinikov et al., Clinical Microbiology, vol. 42, no. 11, pp. 4968–4973, 2004. “An epidemiological investigation of a tularemia outbreak in [8] J. Mikalsen, A. B. Olsen, T. Tengs, and D. J. Colquhoun, “Fran- Smolensk Province,” Zhurnal Mikrobiologii, Epidemiologii, i cisella philomiragia subsp. noatunensis subsp. nov., isolated Immunobiologii, no. 2, pp. 33–37, 1997. from farmed Atlantic cod (Gadus morhua L.),” International [26] H. Abd, T. Johansson, I. Golovliov, G. Sandstrom,¨ and M. Journal of Systematic and Evolutionary Microbiology, vol. 57, Forsman, “Survival and growth of Francisella tularensis in no. 9, pp. 1960–1965, 2007. Acanthamoeba castellanii,” Applied and Environmental Micro- [9] A. Nylund, K. F. Ottem, K. Watanabe, E. Karlsbakk, and B. biology, vol. 69, no. 1, pp. 600–606, 2003. Krossøy, “Francisella sp. (Family Francisellaceae) causing mor- [27] J. Thelaus, A. Andersson, P. Mathisen, A.-L. Forslund, L. tality in Norwegian cod (Gadus morhua) farming,” Archives of Noppa, and M. Forsman, “Influence of nutrient status and Microbiology, vol. 185, no. 5, pp. 383–392, 2006. grazing pressure on the fate of Francisella tularensis in lake [10] J. M. Petersen, J. Carlson, B. Yockey et al., “Direct isolation water,” FEMS Microbiology Ecology, vol. 67, no. 1, pp. 69–80, of Francisella spp. from environmental samples,” Letters in 2009. Applied Microbiology, vol. 48, no. 6, pp. 663–667, 2009. [28] S. H. El-Etr, J. J. Margolis, D. Monack et al., “Francisella [11] N. G. Olsufiev and T. N. Dunayeva, “Natural focality, epidemi- tularensis type A strains cause the rapid encystment of ology and prophylaxis of tularemia,” Medicina, vol. 9, pp. 11– Acanthamoeba castellanii and survive in amoebal cysts for 18, 1970. three weeks postinfection,” Applied and Environmental Micro- [12]J.M.Petersen,P.S.Mead,andM.E.Schriefer,“Francisella biology, vol. 75, no. 23, pp. 7488–7500, 2009. tularensis: an arthropod-borne pathogen,” Veterinary Research, [29] N. D. Konstantinova, M. I. Kormilitsyna, L. V. Didenko, vol. 40, no. 2, article no. 07, 2009. and I. A. Meshcheriakova, “The ultrastructural character- [13] H. Eliasson and E. Back,¨ “Tularaemia in an emergent area in istics of Francisella tularensis interaction with Tetrahymena Sweden: an analysis of 234 cases in five years,” Scandinavian pyriformis,” Zhurnal Mikrobiologii Epidemiologii i Immuno- Journal of Infectious Diseases, vol. 39, no. 10, pp. 880–889, biologii, no. 1, pp. 10–15, 2000. 2007. [30] O.¨ Ostman,¨ J. O. Lundstrom,¨ and T. Z. Persson Vinnersten, [14] G. Olin, “Occurrence and mode of transmission of tularemia “Effects of mosquito larvae removal with Bacillus thuringiensis in Sweden,” Acta Pathologica et Microbiologica Scandinavica, israelensis (Bti) on natural protozoan communities,” Hydro- vol. 19, pp. 220–247, 1942. biologia, vol. 607, no. 1, pp. 231–235, 2008. [15] C. E. Hopla, “The ecology of tularemia,” Advances in Veteri- [31] K. Svensson, E. Back,¨ H. Eliasson et al., “Landscape epidemi- nary Science and Comparative Medicine, vol. 18, pp. 25–53, ology of tularemia outbreaks in Sweden,” Emerging Infectious 1974. Diseases, vol. 15, no. 12, pp. 1937–1947, 2009. [16] B. P. Berdal, R. Mehl, N. K. Meidell, A.-M. Lorentzen-Styr, and [32] G. Sandstrom, A. Tarnvik, H. Wolf-Watz, and S. Lofgren, O. Scheel, “Field investigations of tularemia in Norway,” FEMS “Antigen from Francisella tularensis: nonidentity between Immunology and Medical Microbiology, vol. 13, no. 3, pp. 191– determinants participating in cell-mediated and humoral 195, 1996. reactions,” Infection and Immunity, vol. 45, no. 1, pp. 101–106, [17]P.Anda,J.S.delPozo,J.M.D.Garc´ıa et al., “Waterborne out- 1984. break of tularemia associated with crayfish fishing,” Emerging [33] A. Sjostedt,¨ U. Eriksson, L. Berglund, and A. Tarnvik,¨ Infectious Diseases, vol. 7, no. 3, pp. 575–582, 2001. “Detection of Francisella tularensis in ulcers of patients with [18] D. Greco, G. Allegrini, and T. Tizzi, “A waterborne tularemia tularemia by PCR,” Journal of Clinical Microbiology, vol. 35, outbreak,” European Journal of Epidemiology, vol. 3, no. 1, pp. no. 5, pp. 1045–1048, 1997. 35–38, 1987. [34] M. Bystrom,S.B¨ ocher,¨ A. Magnusson, J. Prag, and A. Johans- [19] S. Helvaci, S. Gedikoglu,ˇ H. Akalin, and H. B. Oral, “Tularemia son, “Tularemia in Denmark: identification of a Francisella in Bursa, Turkey: 205 cases in ten years,” European Journal of tularensis subsp. holarctica strain by real-time PCR and high- Epidemiology, vol. 16, no. 3, pp. 271–276, 2000. resolution typing by multiple-locus variable-number tandem [20] W. L. Jellison, Tularemia in North America 1930–1974,Univer- repeat analysis,” Journal of Clinical Microbiology, vol. 43, no. sity of Montana, Missoula, Mont, USA, 1974. 10, pp. 5355–5358, 2005. [21] R. R. Parker, E. A. Steinhaus, G. M. Kohls, and W. L. Jelliso, [35] M. P. Dempsey, M. Dobson, C. Zhang et al., “Genomic dele- “Contamination of natural waters and mud with Pasteurella tion marking an emerging subclone of Francisella tularensis tularensis and tularemia in beavers and muskrats in the subsp. holarctica in France and the Iberian Peninsula,” Applied northwestern United States,” Bulletin of the National Institutes and Environmental Microbiology, vol. 73, no. 22, pp. 7465– of Health (U.S.), vol. 193, pp. 1–161, 1951. 7470, 2007. 10 International Journal of Microbiology

[36] P. Larsson, K. Svensson, L. Karlsson et al., “Canonical insertion-deletion markers for rapid DNA typing of Fran- cisella tularensis,” Emerging Infectious Diseases, vol. 13, no. 11, pp. 1725–1732, 2007. [37] K. Tamura, J. Dudley, M. Nei, and S. Kumar, “MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0,” Molecular Biology and Evolution, vol. 24, no. 8, pp. 1596–1599, 2007. [38] Z. L. Berrada and S. R. Telford III, “Diversity of Francisella species in environmental samples from Martha’s Vineyard, Massachusetts,” Microbial Ecology, vol. 59, no. 2, pp. 277–283, 2010. [39] M. Forsman and A. Johansson, “Tularemia (Francisella tularensis),” in Encyclopedia of Bioterrorism Defense,R.F.Pilch and R. A. Zilinskas, Eds., pp. 483–488, John Wiley & Sons, Hoboken, NJ, USA, 2005. [40] A. Willke, M. Meric, R. Grunow et al., “An outbreak of oropharyngeal tularaemia linked to natural spring water,” Journal of Medical Microbiology, vol. 58, no. 1, pp. 112–116, 2009. [41] J. F. Bell and S. J. Stewart, “Chronic shedding tularemia nephritis in rodents: possible relation to occurrence of Fran- cisella tularensis in lotic waters,” Journal of wildlife diseases, vol. 11, no. 3, pp. 421–430, 1975. [42] A. Tarnvik,¨ H.-S. Priebe, and R. Grunow, “Tularaemia in Europe: an epidemiological overview,” Scandinavian Journal of Infectious Diseases, vol. 36, no. 5, pp. 350–355, 2004. [43] M. Forsman, E. W. Henningson, E. Larsson, T. Johansson, and G. Sandstrom,¨ “Francisella tularensis does not manifest viru- lence in viable but non-culturable state,” FEMS Microbiology Ecology, vol. 31, no. 3, pp. 217–224, 2000. [44] I. V. Mironchuk and A. V. Mazepa, “Viability and virulence of Francisella tularensis subsp. Holarctica in water ecosystems (experimental study),” Zhurnal Mikrobiologii Epidemiologii i Immunobiologii, no. 2, pp. 9–13, 2002. [45] P. Larsson, D. Elfsmark, K. Svensson et al., “Molecular evolutionary consequences of niche restriction in Fran- cisella tularensis, a facultative intracellular pathogen,” PLoS Pathogens, vol. 5, no. 6, Article ID e1000472, 2009. [46] J. Thelaus, P. Haecky, M. Forsman, and A. Andersson, “Preda- tion pressure on bacteria increases along aquatic productivity gradients,” Aquatic Microbial Ecology, vol. 52, no. 1, pp. 45–55, 2008. [47] B. Auer, U. Elzer, and H. Arndt, “Comparison of pelagic food webs in lakes along a trophic gradient and with seasonal aspects: influence of resource and predation,” Journal of Plankton Research, vol. 26, no. 6, pp. 697–709, 2004. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 340506, 12 pages doi:10.1155/2011/340506

Research Article Gene Expression during Survival of Escherichia coli O157:H7 in Soil and Water

Ashley D. Duffitt,1, 2 Robert T. Reber,1 Andrew Whipple,1, 3 and Christian Chauret2

1 Earth and Environmental Sciences, Taylor University, Upland, IN 46989, USA 2 Department of Natural, Information, and Mathematical Sciences, IN University Kokomo, 2300 South Washington Street, Kokomo, Indiana 46904-9003, USA 3 Department of Biology, Taylor University, Upland, IN 46989, USA

Correspondence should be addressed to Christian Chauret, [email protected]

Received 15 June 2010; Revised 20 August 2010; Accepted 23 August 2010

Academic Editor: Max Teplitski

Copyright © 2011 Ashley D. Duffitt et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The in vitro survival of Escherichia coli O157:H7 at 15◦C under two experimental conditions (sterile soil and sterile natural water) was examined. DNA microarrays of the entire set of E. coli O157:H7 genes were used to measure the genomic expression patterns after 14 days. Although the populations declined, some E. coli O157:H7 cells survived in sterile stream water up to 234 days and in sterile soil for up to 179 days. Cells incubated in soil microcosms for 14 days expressed genes for antibiotic resistance, biosynthesis, DNA replication and modification, metabolism, phages, transposons, plasmids, pathogenesis and virulence, antibiotic resistance, ribosomal proteins, the stress response, transcription, translation, and transport and binding proteins at significantly higher levels than cells grown in Luria broth. These results suggest that E. coli O157:H7 may develop a different phenotype during transport through the environment. Furthermore, this pathogen may become more resistant to antibiotics making subsequent infections more difficult to treat.

1. Introduction the environment has been linked to runoff contaminated withbovinemanureorbyuseassoilamendment[10]. The Escherichia coli O157:H7 is an enterohemorrhagic strain of contamination of surface and ground water in rural areas of E. coli that produces a powerful shiga-like toxin. It is capable the United States is becoming increasingly more common as of causing bloody stools, hemorrhagic colitis, and hemolytic a result of concentrated animal feeding operations [9]. uremic syndrome [1]. Nearly 75,000 cases of O157:H7 Escherichia coli O157:H7 can survive in varying habitats infection occur every year in the US [2]. Most outbreaks under a wide range of conditions. In the environment, cells have been associated with the consumption of contaminated, are exposed to rapidly changing conditions such as changes undercooked, bovine food products [1]. There also have been in pH, nutrient availability, temperature, oxidative stress, and reports of E. coli O157:H7 outbreaks associated with both osmotic challenge [11]. E. coli O157:H7 pollution of water drinking and recreational water [3–7]. and soil is dependent on the ability of this pathogen to E. coli O157:H7 is ubiquitous on farms where healthy adapt to these changes. However, there is limited information cattle and sheep harbor the pathogen in their gastrointestinal concerning the survival of E. coli O157:H7 in soil and water tracts [8]. As a consequence, farm animal manure is a as growth of bacteria, if any, in these environments is not well source for spreading E. coli O157:H7 into the environment understood or documented. E. coli O157:H7 may respond to and potentially to the human food chain. One of the most adverse conditions in the environment by expressing various common modes by which E. coli O157:H7 is introduced stress response genes that enable survival [12]. The master onto food crops is through contaminated irrigation water regulator of the general stress response is an alternative [9]. In addition, the propagation of this pathogen through sigma factor σ38 (RpoS). This sigma factor may be induced 2 International Journal of Microbiology in response to stresses such as weak acids, starvation, high water, and resuspended in 10 mL of sterile stream water or osmolarity, and high or low temperature [13]. There is 10 mL of sterile deionized water. This suspension was used to evidence that stress responses may enable survival under inoculate the stream water and soil microcosms, respectively. more severe conditions, increase pathogenicity, and enhance The control and water experiments were performed using resistance to secondary stresses [14]. In other words, when E. ten replicates each, four for microarray analysis and six to coli cells are stressed, they become harder to kill and are more monitor survival with replacement. The soil treatment was resistant to starvation and toxic chemicals typically used in prepared with 34 replicates, four for microarray analysis distribution systems such as chlorine. This has significant and 30 to monitor survival without replacement. The water public health implications because E. coli O157:H7 could microcosms were incubated at 15◦C with shaking at 60 rpm, develop a disinfectant-resistant phenotype during transport and the soil microcosms were incubated at 15◦C without to water treatment plants [14]. Therefore, understanding the shaking. Four microcosms, each of soil and water, were effects of stress on gene expression in response to altered removed after 14 days for DNA microarray analysis. Four environmental conditions may be crucial in understanding flasks of the 15◦C LB control were removed after 48 hours for the survival of this organism as it moves from one environ- DNA microarray analysis. Survival was assessed periodically ment to another. using the spread plate method described below. Limited work has been conducted to investigate E. coli O157:H7 survival and functional genomics in the envi- 2.3. Water Microcosms. The stream water used in this ronment. Further research is needed to understand the experiment had a pH of 7.26. Aliquots of water (90 mL) were mechanisms that enable E. coli O157:H7 to survive in placed into 250 mL sterile bottles and autoclaved at 121◦C such a wide variety of environments. This study compared at 15 psi for 15 minutes. Following sterilization, the water genetic expression profiles of Escherichia coli O157:H7 under microcosms were stored at 4◦C until inoculation. Following two environmental conditions (soil and natural water) to inoculation with 8.8 ×108 CFU/mL of E. coli O157:H7 ATCC expression in growth media using DNA microarrays. In 35150, the water microcosms were incubated at 15◦Cwith addition, we investigated the long-term survival of E. coli shaking at 60 rpm. O157:H7 in microcosms simulating these environments. 2.4. Soil Microcosms. The soil used in this study was a 2. Materials and Methods Fox silt loam with a pH of 6.95 [15]. Fox silt loam is a typical agricultural soil in East Central Indiana. It is a 2.1. Soil and Water Collection and Site Description. Soil and well-drained soil with a moderate available water capacity, water were collected from Newby Ditch in June of 2006 medium runoff, and poor filtering capacity [15]. Soil pH within 24 hours of a rainfall event (1.93 cm or 0.76 inches). was determined by placing 20 g of soil into a 50 mL beaker, Newby Ditch is located within the Mississinewa watershed adding 20 mL of deionized water, and stirring for 30 minutes on the Tipton Till Plain of East Central Indiana (USA). This [16]. This suspension was allowed to settle for an hour, and watershed is characterized as a highly disturbed landscape pH was measured with a pH meter. Soil moisture content predominated by row crop agriculture. Environmental sam- was determined using the gravimetric method; soil was ples were placed in sterile bottles and stored in a cooler with weighed, oven dried at 105◦C, and then reweighed until the ice packs for transport to the lab. The samples were processed sample weight was constant [16]. The soil moisture content within 24 hours for use as soil and water microcosms was determined to be 0.323 (32.3%), which is a slightly according to the methods described below. moistsoil[15]. Large rocks and debris were removed from the dried soil, and 100-g aliquots were placed in 250 mL ◦ 2.2. Inoculum and Incubation Conditions. E. coli O157:H7 glass bottles. The soil microcosms were autoclaved at 121 C at 15 psi for 15 minutes Following sterilization, the soil strain ATCC 35150 was used to investigate survival and ◦ differences in gene expression under two different envi- was returned to the drying oven for 24 hours at 105 C ronmental conditions compared to Luria broth (LB) (BD, to remove any residual moisture. Sterile soil, as checked Rockville, MD). For long-term storage, the culture was by plate counting, was stored at room temperature prior maintained in 10% glycerol at −50◦C. In the short term, to inoculation. Soil moisture content was adjusted to field cells were cultured on nutrient agar (BD) slants at 4◦C. conditions by adding 32.3 mL of sterile deionized water . × 8 The E. coli O157:H7 experiment cultures were prepared by [16]. Following inoculation with 8 8 10 CFU/ml of E. coli O157:H7 ATCC 35150, the microcosms were sealed and transferring a loopful of a slant culture into a 100 mL flask ◦ containing 30 mL of LB. The flask was incubated for 24 incubated statically at 15 C. hours at 37◦C without shaking. This culture was used to inoculate the three treatment conditions: 30 mL LB control, 2.5. Colony Counting. The initial (zero-time postinocula- 90 mL sterile stream water microcosms, and 100-g sterile tion) concentration E. coli O157:H7 was determined for each soil microcosms. The 30 mL LB control (100 mL flask) was condition prior to incubation. E. coli O157:H7 concentra- inoculated with 1 mL of the culture and incubated at 15◦C tions were determined for each condition by using the spread for 48 hours with shaking at 60 rpm. The remaining cells plate method. Samples for counting were removed directly were harvested by centrifugation at 10,000 × gfor10minutes from the water microcosms and LB broth. For extraction of at 4◦C (Sorvall RC-5B), washed twice with sterile deionized bacteria from soil, 100 mL 0.1% polyethylene glycol (PEG) International Journal of Microbiology 3 was added to 100 g of soil. This suspension was shaken cRNA was then added to a hybridization solution and (150 rpm) for 15 minutes at 15◦C and then allowed to hybridized to the GeneChip after adding control oligonu- settle for five minutes. The supernatant from this suspension cleotides at 45◦C for 17 hours with constant rotation. The was centrifuged for ten minutes (10,000 × g, 4◦C), and the hybridization mixture was removed, and the GeneChip was resulting supernatant was used to enumerate the cells. Viable washed and stained with phycoerythrin-labeled Streptavidin cell counts were performed in duplicate by serial dilution in using the Affymetrix Fluidics Station. The GeneChip was 0.9% sterile saline and spread plate culturing 0.1 mL onto washed again, incubated with biotinylated antistreptavidin, R2A agar (BD) plates. The plates were incubated at 37◦Cfor and then restained with phycoerythrin-labeled Streptavidin 48 hours. Following incubation, plates with colony counts to amplify the signals. Balanced groups of samples were between 25 and 250 were considered, and duplicate counts handled in parallel to reduce nonrandom error. The arrays were averaged. The average number of colonies was divided were scanned using the dedicated scanner controlled by by the volume or mass of the original solution to estimate the Affymetrix GCOS software. number of CFU/ml and CFU/g, respectively. 2.8. Data Analysis. The microarray expression data were 2.6. Isolation of Total RNA. Total RNA was isolated using generated using Affymetrix GCOS software. The Affymetrix the FastRNA Pro Soil-Direct Kit (Qbiogene, Inc., CA) with Microarray Suite Algorithm was used to analyze the minor modifications (described below) to improve quality hybridization intensity data from GeneChip expression and yield. The kit-supplied Lysing Matrix E tubes were placed probe arrays and to calculate a set of metrics to describe at −10◦C for 2 days prior to RNA extraction to minimize probe set performance. The average intensity of each array heating and RNA degradation. One gram of the sample was normalized by global scaling to a target intensity of 1000. was placed into the kit-supplied Lysing Matrix E tubes for Anaverageexpressionvalueforeachtreatmentgroupwas lysis. Sample lysis was performed using a Mini-BeadBeater- calculated via geometric mean because it is better applied to 1 instrument (Biospec Products, Inc., Bartlesville, OK) for data with large fluctuations. Only probe sets that received 80 seconds at a speed setting of 48 (maximum speed). RNA a “present” call of 75% or greater were considered. The was centrifuged through the kit-supplied Quick-Clean Spin expression values were normalized by log2 transformation Filters to remove residual inhibitors following extraction and [17]. Two treatments were compared by determining the ◦ stored in 100 μLofDEPC-H20at−50 C. RNA quality was log2 ratio of gene expression for the corresponding averaged determined by 1.0% (w/v) agarose gel electrophoresis and by intensities for each treatment. Fold change was calculated spectrophotometric analysis (OD260/OD280)usingaThermo from log2 data such that values for induction range from 1 Scientific Spectronic GENESYS 5 Spectrophotometer. An to 100 while values for repression are restricted to the space OD measurement at 260 nm was used to quantify the RNA between0and1[18]. For example, a value of 2 indicates a yield. Approximately 100 μg of total RNA was obtained per 2-fold upregulation while a value of 0.5 indicates a 2-fold sample. The purified RNA samples were delivered on ice to downregulation for a gene comparing the environmental t Dr. Howard Edenberg’s laboratory at the Indiana University condition to the Luria broth. A -test on the log2 transformed School of Medicine Center for Medical Genomics where the data was performed using Microsoft Excel. Significant gene microarray analysis was performed. selection was performed using the Microsoft Excel filter function to select for genes with greater than or equal to a 2.7. DNA Microarrays, cDNA Preparation, and Hybridization. 2-fold up- or downregulation and a P-value less than.05. DNA microarrays were used to evaluate the genetic expres- sion profiles of E. coli O157:H7 ATCC 35150 maintained 2.9. Functional Groups. Functions of significantly expressed under three environmental conditions. The microarray sys- genes were determined using the Affymetrixs NetAffx tem used for E. coli was the GeneChip E. coli Genome 2.0 Analysis Center (http://www.affymetrix.com/analysis/netaffx/ Array (Affymetrix, Inc., Santa Clara, CA). The microarray index.affx)andEcoCyc(http://ecocyc.org/) database. Func- analysis was performed with four biologically-independent tional group analysis was performed by assigning genes to replicates (with respect to E. coli growth, RNA isolation, sam- one of 13 functional groups. ple preparation, and array hybridization) for each treatment condition. 3. Results The standard protocol for prokaryotic sample and array processing recommended by Affymetrix in their GeneChip 3.1. Survival of E. coli O157:H7 in Sterile Soil Microcosms. Soil Expression Analysis Technical Manual (Affymetrix, Santa microcosms were inoculated with 10 mL of 8.8 × 108 CFU/g Clara, CA) was used. cDNA was synthesized using a T7 of E. coli O157:H7 strain ATCC 35150. Immediately after promoter-dT24 oligonucleotide as a primer with the Invitro- inoculation, the average preincubation concentration of E. gen Life Technologies SuperScrip Choice system. Following coli O157:H7 was 1.8 × 107 CFU/g. Following incubation second-strand cDNA synthesis and incubation with T4 DNA at 15◦C, survival was monitored on a regular basis for polymerase, the products were purified using the Affymetrix 179 days. No significant decrease in the E. coli O157:H7 Cleanup Module. Biotinylated cRNA was made using the concentration was observed during the first 30 days of Affymetrix IVT kit. The cRNA was purified using the Qiagen incubation (Figure 1). The last measurement was taken RNeasy column, quantitated, and then fragmented by incu- on day 179, and the average concentration of cells was bation at high temperature with magnesium. Biotinylated 7.7×107 CFU/g (data not shown). 4 International Journal of Microbiology

1e +9 4

1e +8 2

1e +7 0 fold change 2 1e +6 log −2 Viable bacteria (CFU/g) 1e +5

−4 1e +4 1 1456 0 5 1015202530 Genes Day Figure 3: The log expression ratio of the E. coli O157:H7 genome Figure 1: Survival of E. coli O157:H7 in sterile soil microcosms was plotted for Luria broth versus sterile stream water. Genes more at 15◦C. The average concentrations of E. coli O157:H7 were highly expressed in LB have a negative value, whereas genes more determined by the spread plate method and are shown as CFU per highly expressed in water have a positive value. gram. Error bars represent standard deviations about the means (n = 4). 4

3 1e +9 2

1e +8 1 0 fold change

e 2 1 +7 −1 log −2 1e +6 −3

Viable bacteria (CFU/mL) 1e +5 −4 1 4487 Genes 1e +4 0 5 1015202530Figure 4: The log expression ratio of the E. coli O157:H7 genome Day was plotted for Luria broth versus sterile soil. Genes more highly expressed in LB have a negative value, whereas genes more highly Figure 2: Survival of E. coli O157:H7 in sterile stream water expressed in soil have a positive value. microcosms at 15◦C at 60 rpm. The average concentrations of E. coli O157:H7 were determined by the spread plate method and are shown as CFU per milliliter. Error bars represent standard deviations about the means (n = 4). 3.3. Microarray Analysis. The genomic expression profiles of E. coli O157:H7 ATCC 35150 incubated in sterile soil, water, and LB were evaluated. The log expression ratios of the E. coli O157:H7 genome for cells grown in LB at 15◦Cfor48 3.2. Survival of E. coli O157:H7 in Water Microcosms. Sterile hours with shaking at 60 rpm versus fourteen-day incubation water was inoculated with 10 mL of 8.8 × 108 CFU/mL of in sterile stream water at 15◦C are illustrated in Figure 3.The E. coli O157:H7 strain ATCC 35150. On day 0, the mean whole genome analysis indicated that 705 genes were more concentration of E. coli O157:H7 was 1.1 × 108 CFU/mL. highly expressed in LB (plotted with a negative value) while The microcosms were incubated at 15◦C with gentle shaking 751 genes were more highly expressed in sterile stream water at 60 rpm, and survival was monitored for 234 days. A (plotted with a positive value) (Figure 3). Figure 4 shows decrease in the E. coli O157:H7 concentration below the the log expression ratios of the E. coli O157:H7 genome for postinoculation concentration was not observed until day 3 cells grown in LB at 15◦C for 48 hours versus fourteen- (Figure 2). The population decreased by less than 0.3 log day incubation in sterile soil at 15◦C. The analysis of the by day 28. The final measurement was taken on day 234 whole genome indicated that 2,664 genes were more highly (nearly 8 months following inoculation), and the mean expressed in LB (plotted with a negative value) while 1,823 concentration of cells was 2.98 × 104 CFU/mL (data not genes were more highly expressed in sterile soil (plotted with shown). apositivevalue). International Journal of Microbiology 5

Table 1: Functional groups differentially expressed between growth Table 2: Functional groups differentially expressed between growth in LB and growth in sterile stream water. in LB and growth in sterile soil.

Higher in Higher in Higher in Higher in Functional group Total Functional group Total LB water LB soil Whole genome 38 26 12 Whole genome 397 89 308 Antibiotic resistance 0 00Antibiotic resistance 3 03 Biosynthesis 0 00Biosynthesis 21 120 DNA replication/repair, DNA replication/repair, 2 11 10 28 restriction/modification restriction/modification Metabolism 0 00Metabolism 33 15 18 Pathogenesis and virulence 0 00Pathogenesis and virulence 7 07 Phage, transposon, or Phage, transposon, or 0 00 8 35 plasmid plasmid Ribosomal proteins 1 10Ribosomal proteins 45 045 Signaling and motility 1 10Signaling and motility 2 20 Stress response 3 30Stress response 18 018 Transcription, RNA Transcription, RNA processing, and 4 40processing, and 39 534 degradation degradation Translation and Translation and posttranslational 1 10posttranslational 27 225 modification modification Transport and binding Transport and binding 7 61 48 21 27 proteins proteins Uncategorized 19 910Uncategorized 136 38 98

An analysis of gene ratios with significant expression Tables 3 and 4. For example, seven of the genes for amino levels (≥ 2foldsandP-value < .05) revealed that the majority acid biosynthesis were expressed at significantly higher levels of genes did not differ significantly between conditions. This in soil than in LB (Table 3). There were no differences in was especially true of the cells grown in LB compared with expression of these amino acid biosynthesis genes between the cells incubated in sterile stream water; 26 genes were cells grown in LB and cells incubated in sterile natural more significantly expressed in LB compared to 12 genes water (Table 4). Of the 55 genes that are known to encode in cells incubated in sterile stream water (Table 1). The ribosomal proteins, 45 were expressed at significantly higher comparison of cells incubated in sterile soil compared to cells levels in soil compared to Luria broth (Table 2). Only one growninLByieldedmoredifferences in expression; 89 genes ribosomal protein gene (rpmC) was significantly expressed were expressed at significantly higher levels in LB while 308 in cells grown in LB compared to water (Table 1). genes were more highly expressed in sterile soil (Table 2). The genes responsible for the stress response include A functional group analysis was performed for signif- those that function in temperature shock, acid tolerance, icantly expressed genes in LB versus sterile soil (Table 2) the SOS response, and osmotic challenge. Eighteen stress and LB versus sterile stream water (Table 1). Functional response genes were significantly expressed in cells incubated group analysis of LB versus sterile soil revealed that in sterile soil compared to LB (Table 2). On the other hand, cells incubated in sterile soil expressed more genes for three stress response genes were more significantly expressed antibiotic resistance, biosynthesis, DNA replication/repair in LB compared to cells incubated in sterile natural water and restriction/modification, metabolism, pathogenesis and (Table 1). The rpoS gene is induced in response to entry virulence, phages, transposons, and plasmids, ribosomal into stationary phase and also by stresses such as weak acids, proteins, stress response, transcription, RNA processing, and starvation, osmotic challenge, and temperature changes. The degradation, translation and posttranslational modification, expression of rpoS was significantly elevated in soil (2.68-fold and transport and binding proteins (Table 2). Functional induction) (Table 3). The rpoH heat shock sigma factor group analysis of LB versus sterile stream water revealed 32 (σ32), which regulates the heat shock response, was that cells incubated in sterile stream water expressed more more highly expressed in soil compared to LB (3.19-fold genes of unknown function while cells grown in LB expressed induction) (Table 3). Cells grown in soil expressed heat shock more genes for ribosomal proteins, signaling and motility, genes dnaK and htpX at significantly greater levels (Table 3). stress response, transcription, RNA processing, and degra- In addition, table 3 shows that rseA, an antisigma regulator dation, translation and posttranslational modification, and of the rpoE envelope heat stress system, was induced in transport and binding proteins (Table 1). The nature of these cells incubated in soil (3.29-fold induction). Numerous cold differences in genomic expression is described in detail in shock genes were significantly expressed in cells incubated 6 International Journal of Microbiology

Table 3: Selected genes differentially expressed between growth in LB and sterile soil microcosms.

Logarithmic Function and gene Description P-value ratio (S/C) Amino acid biosynthesis hisG Histidine biosynthesis 2.05 1.04E-03 argB Arginine biosynthesis 3.05 1.60E-04 argC Arginine biosynthesis 3.52 4.13E-04 argG Arginine biosynthesis 3.07 2.56E-06 asnB Asparagine synthetase B 3.41 1.66E-02 thrL Thr operon leader peptide 3.43 1.23E-03 cysK Cysteine biosynthesis 3.53 2.00E-06 Antibiotic resistance marA Multiple antibiotic resistance protein 4.20 2.31E-09 marB Multiple antibiotic resistance protein 4.41 3.25E-05 marR Multiple antibiotic resistance protein 5.16 2.51E-07 DNA replication/repair, restriction/modification priB Primosomal replication protein 2.19 1.23E-03 topA DNA topoisomerase I 2.33 6.56E-08 fis DNA binding protein Fis 2.37 3.72E-05 priA Primosome assembly protein 2.60 4.67E-02 Metabolism aceE Pyruvate dehydrogenase E1 subunit 2.03 5.14E-06 asmA Protein asmA precursor; electron transport 2.21 3.40E-02 lpdA Dihydrolipoamide dehydrogenase; energy metabolism 2.24 1.16E-04 N-ethylmaleimide reductase; central intermediary nemA 2.71 1.56E-05 metabolism glpC Anaerobic glycerol-3-phosphate dehydrogenase subunit C 2.81 7.83E-03 yfhO Cysteine metabolism; amino acid metabolism 3.25 3.64E-05 icdA Isocitrate dehydrogenase; TCA cycle metabolism 3.48 9.13E-05 adhC Alcohol dehydrogenase class III; energy metabolism 4.19 9.70E-05 yibO Phosphoglyceromutase; carbohydrate metabolism 6.21 8.10E-03 ttdB L(+)-tartrate dehydrase; energy metabolism 0.44 2.22E-06 Formate hydrogenlyase subunit 7; mitochondrial electron hycG 0.44 3.00E-06 transport Carbon-phosphorus lyase complex subunit; central phnH 0.45 2.51E-04 intermediary metabolism mhpF Acetaldehyde dehydrogenase; amino acid metabolism 0.46 1.45E-06 Carbon-phosphorus lyase complex subunit; central phnJ 0.46 5.55E-05 intermediary metabolism Hydrogenase 4 Fe-S subunit formate hydrogenlyase, hycF 0.47 2.96E-05 complex iron-sulfur protein eutI Phosphate acetyltransferase 0.47 8.02E-04 2,4-dienoyl-CoA reductase (NADPH), NADH and ygjL 0.49 4.90E-01 FMN-linked Pathogenesis and virulence vacB Ribonuclease R, exoribonuclease R, RNase R 2.26 1.44E-05 tolA Colicin production 2.26 2.51E-05 Colanic acid biosynthesis; resistance to acid stress, wcaL 2.51 1.32E-02 desiccation, and thermal stress ygeO Hypothetical protein 2.66 1.04E-02 Phage, transposon, or plasmid International Journal of Microbiology 7

Table 3: Continued. Logarithmic Function and gene Description P-value ratio (S/C) sieB Rac prophage; phage superinfection exclusion protein 0.43 5.93E-03 Lar Rac prophage; restriction alleviation protein 0.43 9.19E-03 ydaE Rac prophage; conserved protein 0.44 2.28E-03 ydaQ Rac prophase; conserved protein 0.45 2.22E-03 ydaC Rac prophage predicted protein 0.50 8.85E-03 ymfP E14 prophage; conserved protein 4.46 1.87E-02 mprA Translational repressor mprA; plasmid related function 3.53 1.78E-05 Ribosomal proteins rpsU 30S ribosomal protein S21 2.25 1.12E-04 rpsT 30S ribosomal protein S20 2.35 4.46E-04 rplY 50S ribosomal protein L25 2.35 2.21E-03 rpmF 50S ribosomal protein L32 2.43 1.16E-03 rpmH 50S ribosomal protein L34 2.52 3.30E-05 rplK 50S ribosomal protein L11 2.67 9.18E-04 rpsR 30S ribosomal protein S18 2.87 7.21E-05 rplA 50S ribosomal protein L1 2.87 9.28E-05 rpmD 50S ribosomal protein L30 3.17 3.14E-05 rpsJ 30S ribosomal protein S10 3.24 6.19E-04 rpmE 50S ribosomal protein L31 3.32 4.52E-05 rplW 50S ribosomal protein L23 3.53 1.57E-04 rpsH 30S ribosomal protein S8 4.34 3.27E-04 rplF 50S ribosomal protein L6 4.57 1.39E-04 rplX 50S ribosomal protein L24 4.66 2.51E-05 rpsN 30S ribosomal protein S14 5.68 6.06E-05 Stress response cspE Cold shock protein E 2.02 5.08E-03 yfiA Cold shock protein, associated with 30S ribosomal subunit 2.10 1.23E-03 htpX Heat shock protein 2.10 2.02E-04 sulA SOS cell division inhibitor 2.17 1.69E-05 ymcE Cold shock gene 2.33 1.49E-04 dnaK Heat shock protein; molecular chaperone 2.48 3.21E-04 recA Recombinase A; SOS response 2.59 8.09E-04 Response to organic acid stress and acetate induced acid rpoS 2.68 6.83E-04 tolerance; regulatory function ahpC Alkyl hydroperoxide reductase C22 protein; oxidative stress 2.91 9.08E-06 dead Cold shock DEAD box protein A 3.04 1.18E-05 rpoH RNA polymerase sigma factor; heat response 3.19 5.63E-06 Spy Envelope stress induced periplasmic protein 3.54 6.83E-08 Osmotic adaptation; Osmotically inducible lipoprotein B osmB 4.77 1.81E-04 precursor cspG Cold shock protein 5.04 2.93E-05 cpxP Envelope stress response 7.08 1.17E-05 cspA Cold shock protein cspA, major cold shock protein 8.59 6.77E-05 Transcription, RNA processing, and degradation rpoB DNA directed RNA polymerase beta subunit; transcription 2.12 1.54E-04 8 International Journal of Microbiology

Table 3: Continued. Logarithmic Function and gene Description P-value ratio (S/C) rpoD Hypothetical protein, RNA polymerase sigma factor 2.20 1.79E-03 Pnp Polyribonucleotide, nucleotidyltransferase; RNA processing 2.21 1.58E-03 rpnA Ribonuclease P 2.31 4.01E-05 Rne Ribonuclease E, fused ribonuclease E: endoribonuclease 2.72 1.33E-03 DNA-directed RNA polymerase alpha subunit; rpoA 2.92 1.72E-04 transcription nusA Transcription elongation factor NusA 3.16 1.15E-04 Rho Transcription termination factor Rho 3.34 5.92E-05 Translation and posttranslational modification Regulatory protein, DNA binding dual transcriptional soxS 2.44 2.47E-04 regulator Elongation factor Tu, protein chain elongation factor tufA 2.61 8.28E-05 (EF-Tu) infA Translation initiation factor IF-1 3.09 4.37E-05 infC Translation initiation factor IF-3 3.31 3.27E-05 fusA Elongation factor EF-2 3.36 7.54E-05 infB Translation initiation factor IF-2 4.13 5.84E-04 Transport and binding proteins phnO Phosphonate transport, N-acetyltransferase activity 0.46 1.40E-06 PTS system, trehalose-specific IIBC component; transport treB 0.47 3.04E-04 of small molecules malK Maltose/maltodextrin transport 0.48 1.49E-05 thiQ Thiamine transport 0.49 1.48E-04 fepB Iron-enterobactin transporter subunit 0.49 7.73E-05 cusB Copper efflux system protein 2.02 4.55E-06 Phosphate transport, sodium dependent phosphate yjbB 2.07 5.77E-03 transporter livJ Leucine/isoleucine/valine transporter subunit 2.22 3.16E-04 oppA Oligopeptide transporter subunit 2.32 2.98E-04 lolE Lipoprotein releasing system, transmembrane protein lolE 2.32 1.54E-02 prlA Preprotein translocase; protein transport 2.46 3.03E-05 glnH Glutamine ABC transporter, periplasmic-binding protein 2.50 5.52E-05 artP Arginine transport 2.52 1.33E-05 fepD Ferric enterobactin transport system 3.25 3.01E-02 fepE Ferric enterobactin transport protein 3.85 5.61E-03 in soil compared to cells grown in LB: cspA, cspE, cspG, the vacB gene was expressed (2.26-fold induction) (Table 3). ymcE, deaD, yfiA (Table 3). Only one cold shock gene (cspC) tolA, a gene involved in colicin production, was significantly was expressed at significantly lower levels in cells incubated expressed in cells incubated in soil compared to cells grown in sterile water compared to cells grown in LB (0.48- in LB (2.26-fold induction) (Table 3). Also, the soxS gene fold repression) (Table 4). Two genes involved in the SOS was more highly expressed in cells incubated in sterile soil response were significantly expressed in cells grown in soil microcosms (2.44-fold induction) (Table 3). Finally, three compared to LB: recA and sulA. This regulatory network is antibiotic resistance genes (marR, marA, and marB)were induced by DNA damage or interference with DNA repli- expressed at significantly higher levels in cells incubated in cation. The osmotically inducible gene osmB was expressed sterilesoilmicrocosmscomparedtoLB(Table 2). at significantly higher levels in cells incubated in soil (4.77- fold induction). osmB encodes an outer membrane protein of 4. Discussion unknown function. Seven genes that aid in pathogenesis and virulence were significantly expressed in cells incubated in E. coli O157:H7 may encounter conditions that are less soil compared to those grown in LB (Table 2). In particular, than optimal for growth in soil and water and must International Journal of Microbiology 9

Table 4: Selected genes differentially expressed between growth in LB and sterile water microcosms.

Logarithmic Function and gene Description P-value ratio (W/C) DNA replication/repair, restriction/modification DNA ycbY restriction-modification 2.38 4.13E-02 system; DNA methylation Membrane ompA Outer membrane protein 0.42 4.62E-04 Outer membrane protein ompX X; integral to outer 0.43 1.09E-04 membrane Metabolism Protein yfiD, pyruvate yfiD 0.48 3.74E-03 formate lyase subunit Ribosomal proteins Protein biosynthesis, structural constituent of rpmC ribosome, intracellular 0.48 3.25E-04 ribosome, ribonucleoprotein complex Regulatory RNA gcvB Regulatory sRNA 0.23 1.00E-03 csrC Regulatory RNA 0.39 4.40E-03 vmicF Regulatory sRNA 0.41 1.99E-03 ryhA Unknown RNA 0.46 8.79E-03 Stress response cspC Cold-shock stress protein 0.48 1.41E-04 DNA protection during dps 0.49 3.75E-04 starvation conditions Transcription, RNA processing, and degradation Integration host factor alpha subunit; DNA himA 0.49 9.13E-03 recombination and transcription regulation Transport and binding proteins Outer membrane protein F ompF precursor; ion transport, 0.46 1.21E-03 porin activity Putrescine-binding, potF periplasmic protein 2.71 3.22E-02 precursor adapt to these conditions in order to survive. Various plants [14]. Therefore, this study investigated the survival stress response mechanisms allow this pathogen to adapt to and genetic expression profiles of E. coli O157:H7 in sterile sublethal environmental conditions. Extended exposure to soil and sterile natural water. Our results indicate that E. coli these stresses enables E. coli O157:H7 to survive under more O157:H7 can persist for long periods of time in sterile soil severe conditions, increases its pathogenesis, and enhances and sterile stream water. In addition, we found that E.˜coli its resistance to chemicals typically used in water distribution O157:H7 exhibits differential gene expression profiles in systems [14]. This has significant public health implications sterile soil and sterile stream water compared to cells freshly because E. coli O157:H7 could develop a disinfectant- grown in LB. This survival does not account for the possible resistant phenotype during transport to water treatment effects of competition with other bacteria or interactions 10 International Journal of Microbiology with predatory protozoa. Under natural conditions, where when compared to cells cultured in Luria broth. In addition, predators and other bacteria are present, a net die off of the expression of 18 genes involved in the stress response was E. coli O157:H7 would likely occur. It is also possible that the more highly expressed in cells from soil. These genes regulate environmental persistence of E. coli O157:H7 cells initially cellular response to cold shock, heat shock, acid tolerance, grown in fecal extracts could be different. osmotic challenge, and the SOS response [24]. This indicates Microarray analysis revealed that cells incubated in sterile that the soil environment stressed these cells, and they turned soil for 14 days remain very active. In fact, 308 genes were on genes to cope with sublethal environmental conditions. found to be more highly expressed in these cells compared The heat shock response is a protective mechanism to to cells grown in LB. A functional group analysis revealed cope with heat-induced damage to proteins; however, there that the majority of these genes were involved in amino acid is evidence suggesting that these genes are also induced in biosynthesis, DNA replication and repair, pathogenesis and response to acidic conditions [17], SOS-inducing treatments virulence, the stress response, ribosomal proteins, antibiotic [25], and sublethal exposure to chlorine [26]. Most heat resistance, transcription, and translation. On the other hand, shock proteins act as molecular chaperones that bind to and microarray analysis of cells placed in sterile stream water stabilize unfolded proteins and promote protein refolding for 14 days revealed that only 12 genes were more highly and proper assembly [27]. This is the case with the product expressed in this condition. The majority of these genes of the dnaK gene. The dnaK gene product has been shown are uncategorized and of unknown function. There was a to regulate other heat shock proteins, such as htpX,and marked difference in the expression of ribosomal protein and play a major role in digesting irreversibly heat damaged translation genes. Typically, faster-growing cells synthesize polypeptides [28]. In addition to heath shock proteins, proteins more rapidly and contain more ribosomes [19, 20]. numerous cold shock genes were significantly expressed in Tao et al. [18] studied the gene expression of E. coli K12 in cells incubated in soil compared to cells grown in LB: cspA, response to nutrient limitation. These researchers found that cspE, cspG, ymcE, deaD, yfiA. These genes protect the cell 42 ribosomal protein genes were expressed at significantly during sublethal environmental temperatures. CspA is the higher levels in cells grown under high nutrient conditions. major cold shock protein of pathogenic E. coli. It functions The present study, on the other hand, revealed that 45 as an RNA chaperone and facilitates translation at low ribosomal protein genes were more highly expressed in cells temperatures [29]. A specific sigma factor has not yet been incubated in sterile soil compared to cells grown in LB. The identified in the case of the cold shock response [27]. exception to growth-rate-dependent regulation of ribosome Two genes involved in the SOS response were signif- number occurs at very low growth rates [21]. When E. coli icantly expressed in cells grown in soil compared to LB: cells adjust to a slow growth rate from a fast one, RNA recA and sulA. This regulatory network is induced by DNA accumulation is attenuated for a short time until the RNA damage or interference with DNA replication. The RecA content is reduced to that characteristic of cells grown at the protein functions as a positive control for SOS regulation, slower rate [22]. is required for all homologous recombination in E. coli,and It is thought that the same mechanism that functions catalyzes synapsis and strand exchange between homolo- during amino acid starvation also functions during growth gous molecules [30]. The sulA gene product functions as rate transitions. In fact, the continued accumulation of RNA an inducible inhibitor of septation [31]. When cells are in cells under partial amino acid starvation has been shown exposed to SOS-inducing environments, they will continue to be accompanied by a continued synthesis in ribosomal to elongate but fail to septate and thus form filaments. proteins [23]. This could account for the higher expression Several genes responsible for the pathogenesis and vir- of genes for ribosomal proteins observed in the present ulence of E. coli O157:H7 were significantly expressed in study as genes encoding the enzymes needed for amino acid cells from sterile soil microcosms. The vacB gene, which is biosynthesis were more highly expressed in cells incubated in required for the full expression of the virulence phenotype sterile soil microcosms. In E. coli, there are 97 known genes in E. coli [32], was highly expressed in cells incubated in responsible for encoding the enzymes needed for amino acid soil. Moreover, a gene involved in colonic acid biosynthesis, biosynthesis [18]. Previous results [18] indicate that these wcaL, was more highly expressed in soil compared to LB. genes are induced for growth in low nutrient environments wcaL is the last gene of the colanic acid gene cluster as appears to be the case in the present study. Growth [33]. Colanic acid forms a protective capsule around the conditions that lead to a decreased rate of ribosome synthesis bacterial cell surface and plays a role in pathogenesis [34]. typically result in an excess of ribosomal proteins, and their Danese et al. [35] demonstrated that colonic acid synthesis is transcript levels are higher in faster growing cells. upregulated in biofilms and is not synthesized in planktonic The regulatory mechanism that controls the general cells under normal laboratory conditions. This may account stress response is the RpoS sigma factor (σ38) and is encoded for the differences in colonic acid gene expression observed by the rpoS gene [24]. An early adaptation in cells exposed between the cells incubated in sterile soil microcosms (in to environmental stresses involves the expression of rpoS. which biofilms are likely) compared to the LB control. A This gene, which controls the expression of more than 50 gene involved in colicin production, tolA, was significantly proteins, is induced in response to entry into stationary phase expressed at a higher level in cells incubated in soil compared and also by stresses such as weak acids, starvation, osmotic to cells grown in luria broth. Colicins are antibacterial pro- challenge, and temperature changes [13]. The expression of teins produced by some strains of E. coli that kill competing rpoS in cells in sterile soil microcosms was 2.68-fold higher strains of bacteria by inhibiting energy metabolism, protein International Journal of Microbiology 11 synthesis, or DNA synthesis [36]. Colicins are also known to Escherichia coli O157:H7 infection: evidence of prolonged to increase bacterial resistance to host defense. In addition, contamination of a fresh water lake,” Epidemiology and three genes (marR, marA, and marB) responsible for multiple Infection, vol. 119, no. 1, pp. 1–8, 1997. antibiotic resistance were more highly expressed in sterile [5] S. J. Olsen, G. Miller, T. Breuer et al., “A waterborne outbreak soil. The multiple antibiotic resistance (mar)locusinE. coli is of Escherichia coli O157:H7 infections and hemolytic uremic composed of two operons (marC and marRAB). Expression syndrome: implications for rural water systems,” Emerging Infectious Diseases, vol. 8, no. 4, pp. 370–375, 2002. of the marRAB operon protects E. coli against numerous [6] M. Samadpour, J. Stewart, K. Steingart et al., “Laboratory antibiotics [37]. Moreover, the elevated expression of the soxS investigation of an E. coli O157:H7 outbreak associated with gene product has been associated with the multiple antibiotic swimming in Battle Ground Lake, Vancouver, Washington,” resistance (mar) phenotype [37]. The collective expression Journal of Environmental Health, vol. 64, no. 10, pp. 16–25, of these genes and the genes involved in the general stress 2002. response may contribute to bacterial survival and virulence [7] D. J. Bopp, B. D. Sauders, A. L. Waring et al., “Detection, iso- during infection. In fact, there is evidence that antibiotic lation, and molecular subtyping of Escherichia coli O157:H7 treatment increases the development of hemolytic uremic and Campylobacter jejuni associated with a large waterborne syndrome (HUS) in children with E. coli O157:H7 infection outbreak,” Journal of Clinical Microbiology,vol.41,no.1,pp. [38]. 174–180, 2003. In conclusion, Affymetrix GeneChip E. coli Genome [8] P. A. Chapman, C. A. Siddons, D. J. Wright, P. Norman, J. Fox, Arrays were used to demonstrate that E. coli O157:H7 cells andE.Crick,“Cattleasapossiblesourceofverocytotoxin- ◦ producing Escherichia coli O157 infections in man,” Epidemi- placed in sterile soil and water microcosms at 15 Cfor14 ology and Infection, vol. 111, no. 3, pp. 439–447, 1993. days exhibit differential gene expression compared to cells ◦ [9] A. M. Ibekwe, P. M. Watt, P. J. Shouse, and C. M. Grieve, “Fate grown in LB at 15 C for 48 hours. The cells incubated of Escherichia coli O157:H7 in irrigation water on soils and in sterile soil microcosms were undoubtedly stressed and plants as validated by culture method and real-time PCR,” therefore in a different physiological state than cells grown Canadian Journal of Microbiology, vol. 50, no. 12, pp. 1007– ◦ in LB at 15 C for 48 hours. These cells exhibit a phenotype 1014, 2004. that may lead to stress-associated disinfection resistance, [10] I. T. Kudva, K. Blanch, and C. J. Hovde, “Analysis of Escherichia increased pathogenesis, and virulence. This has important coli O157:H7 survival in ovine or bovine manure and manure implications in water treatment and public health because slurry,” Applied and Environmental Microbiology, vol. 64, no. 9, surface and ground waters are the source for municipal pp. 3166–3174, 1998. drinking water. Further research on the mechanisms and [11] M. J. Farrell and S. E. Finkel, “The growth advantage in stationary-phase phenotype conferred by rpoS mutations is regulation of the stress response of E. coli O157:H7 is dependent on the pH and nutrient environment,” Journal of needed to prevent potential risk of disease. It is also possible ff Bacteriology, vol. 185, no. 24, pp. 7044–7052, 2003. that the genetics expression could be di erent in nonsterile [12] K. W. Arnold and C. W. Kaspar, “Starvation- and stationary- environments, and this needs to be investigated. phase-induced acid tolerance in Escherichia coli O157:H7,” Applied and Environmental Microbiology,vol.61,no.5,pp. 2037–2039, 1995. Acknowledgments [13] R. Lange and R. Hengge-Aronis, “The cellular concentration of the σ(S) subunit of RNA polymerase in Escherichia coli The microarray studies were carried out using the facilities is controlled at the levels of transcription, translation, and of the Center for Medical Genomics at Indiana University protein stability,” Genes and Development, vol. 8, no. 13, pp. School of Medicine. The Center for Medical Genomics is 1600–1612, 1994. supported in part by the Indiana Genomics Initiative at [14]J.T.Lisle,S.C.Broadaway,A.M.Prescott,B.H.Pyle, Indiana University (INGENE, which is supported in part by C. Fricker, and G. A. Mcfeters, “Effects of starvation on the Lilly Endowment, Inc.). In addition, the authors thank physiological activity and chlorine disinfection resistance Dr. Michael Guebert at Taylor University for assistance in in Escherichia coli O157:H7,” Applied and Environmental selecting the sample site. Microbiology, vol. 64, no. 12, pp. 4658–4662, 1998. [15] E. L. Jensen, R. J. Braun, K. S. Rich, S. W. Tardy, and C. F. Walker, Soil Survey of Grant County, Indiana,Department References of Agriculture, Soil Conservation Service, Purdue University Agricultural Experiment Station, West Lafayette, Ind, USA, [1]T.G.Boyce,D.L.Swerdlow,andP.M.Griffin, “Escherichia 1988. coli O157:H7 and the hemolytic-uremic syndrome,” The New [16] C. E. Marshal, B. T. Shaw, and K. D. Jacob, Methods of England Journal of Medicine, vol. 333, no. 6, pp. 364–368, 1995. Soil Analysis Part 2: Chemical and Microbiological Properties, [2] N. T. Perna, G. Plunkett, V. Burland, et al., “Genome sequence American Society of Agronomy, Madison, Wis, USA, 1965. of enterohaemorrhagic Escherichia coli O157:H7,” Nature, vol. [17] L. M. Maurer, E. Yohannes, S. S. Bondurant, M. Radmacher, 409, no. 6819, pp. 529–533, 2001. and J. L. Slonczewski, “pH regulates genes for flagellar motility, [3] D. L. Swerdlow, B. A. Woodruff, R. C. Brady et al., “A catabolism, and oxidative stress in Escherichia coli K-12,” waterborne outbreak in Missouri of Escherichia coli O157:H7 Journal of Bacteriology, vol. 187, no. 1, pp. 304–319, 2005. associated with bloody diarrhea and death,” Annals of Internal [18] H. Tao, C. Bausch, C. Richmond, F. R. Blattner, and T. Conway, Medicine, vol. 117, no. 10, pp. 812–819, 1992. “Functional genomics: expression analysis of Escherichia coli [4] D. Ackman, S. Marks, P. Mack, M. Caldwell, T. Root, and growing on minimal and rich media,” Journal of Bacteriology, G. Birkhead, “Swimming-associated haemorrhagic colitis due vol. 181, no. 20, pp. 6425–6440, 1999. 12 International Journal of Microbiology

[19] P. S. Tabor, K. Ohwada, and R. R. Colwell, “Filterable marine K-12 biofilm architecture,” JournalofBacteriology, vol. 182, no. bacteria found in the deep sea: distribution, taxonomy, and 12, pp. 3593–3596, 2000. response to starvation,” Microbial Ecology, vol. 7, no. 1, pp. 67– [36]D.L.Davies,F.R.Falkiner,andK.G.Hardy,“ColicinV 83, 1981. production by clinical isolates of Escherichia coli,” Infection and [20] J. Keener and M. Nomura, “Regulation of ribosome synthesis,” Immunity, vol. 31, no. 2, pp. 574–579, 1981. in Escherichia coli and Salmonella: Cellular and Molecular [37]L.Gambino,S.J.Gracheck,andP.F.Miller,“Overexpression Biology, F. C. Neidharde, R. Curtiss III, J. L. Ingraham, et al., of the marA positive regulator is sufficient to confer multiple Eds., pp. 1417–1431, ASM Press, Washington, DC, USA, 2nd antibiotic resistance in Escherichia coli,” Journal of Bacteriology, edition, 1996. vol. 175, no. 10, pp. 2888–2894, 1993. [21] P. P. Dennis and H. Bremer, “Differential rate of ribosomal [38] C. S. Wong, S. Jelacic, R. L. Habeeb, S. L. Watkins, and P. protein synthesis in Escherichia coli,” Journal of Molecular I. Tarr, “The risk of the hemolytic-uremic syndrome after Biology, vol. 84, no. 3, pp. 407–422, 1974. antibiotic treatment of Escherichia coli O157:H7 infections,” [22] R. M. Winslow, “A consequence of the rel gene during a The New England Journal of Medicine, vol. 342, no. 26, pp. glucose to lactate downshift in Escherichia coli. The rates of 1930–1936, 2000. ribonucleic acid synthesis,” The Journal of Biological Chem- istry, vol. 246, no. 15, pp. 4872–4877, 1971. [23] P. P. Dennis and M. Nomura, “Regulation of the expression of ribosomal protein genes in Escherichia coli,” Journal of Molecular Biology, vol. 97, no. 1, pp. 61–76, 1975. [24] R. Hengge-Aronis, “Survival of hunger and stress: the role of rpoS in early stationary phase gene regulation in E. coli,” Cell, vol. 72, no. 2, pp. 165–168, 1993. [25] J. H. Krueger and G. C. Walker, “groEL and dnaK genes of Escherichia coli are induced by UV irradiation and nalidixic acid in an htpR-dependent manner,” Proceedings of the National Academy of Sciences of the United States of America, vol. 81, no. 5, pp. 1499–1503, 1984. [26] S. Dukan, S. Dadon, D. R. Smulski, and S. Belkin, “Hypochlor- ous acid activates the heat shock and soxRS systems of Escherichia coli,” Applied and Environmental Microbiology, vol. 62, no. 11, pp. 4003–4008, 1996. [27]H.J.Chung,W.Bang,andM.A.Drake,“Stressresponseof Escherichia coli,” Comprehensive Reviews in Food Science and Food Safety, vol. 5, no. 3, pp. 52–64, 2006. [28] J. G. Morris, “Bacterial shock responses,” Endeavour, vol. 17, no. 1, pp. 2–6, 1993. [29] W. Jiang, Y. Hou, and M. Inouye, “CspA, the major cold-shock protein of Escherichia coli, is an RNA chaperone,” The Journal of Biological Chemistry, vol. 272, no. 1, pp. 196–202, 1997. [30] S. C. Kowalczykowski, “Biochemistry of genetic recombina- tion: energetics and mechanism of DNA strand exchange,” Annual Review of Biophysics and Biophysical Chemistry, vol. 20, pp. 539–575, 1991. [31] J. M. Schoemaker, R. C. Gayda, and A. Markovitz, “Regulation of cell division in Escherichia coli: SOS induction and cellular location of the SulA protein, a key to lon-associated filamen- tation and death,” Journal of Bacteriology, vol. 158, no. 2, pp. 551–561, 1984. [32] Z.-F. Cheng, Y. Zuo, Z. Li, K. E. Rudd, and M. P. Deutscher, “The vacB gene required for virulence in Shigella flexneri and Escherichia coli encodes the exoribonuclease RNase R,” The Journal of Biological Chemistry, vol. 273, no. 23, pp. 14077– 14080, 1998. [33] L. Wang, S. Huskic, A. Cisterne, D. Rothemund, and P. R. Reeves, “The O-antigen gene cluster of Escherichia coli O55:H7 and identification of a new UDP-GlcNAc C4 epimerase gene,” Journal of Bacteriology, vol. 184, no. 10, pp. 2620–2625, 2002. [34] G. D. Christensen, W. A. Simpson, A. L. Bisno, and E. H. Beachey, “Adherence of slime-producing strains of Staphylo- coccus epidermidis to smooth surfaces,” Infection and Immu- nity, vol. 37, no. 1, pp. 318–326, 1982. [35] P. N. Danese, L. A. Pratt, and R. Kolter, “Exopolysaccharide production is required for development of Escherichia coli Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 421379, 11 pages doi:10.1155/2011/421379

Research Article Quantification of Persistence of Escherichia coli O157:H7 in Contrasting Soils

A. Mark Ibekwe,1 Sharon K. Papiernik,2 Catherine M. Grieve,1 and Ching-Hong Yang3

1 USDA-ARS, U.S. Salinity Laboratory, 450 W-Big springs Roael, Riverside, CA 92507, USA 2 USDA-ARS, North Central Soil Conservation Research Laboratory, Morris, MN 56267, USA 3 Department of Biological Sciences, University of Wisconsin-Milwaukee, WI 53211, USA

Correspondence should be addressed to A. Mark Ibekwe, [email protected]

Received 16 May 2010; Revised 29 June 2010; Accepted 9 July 2010

Academic Editor: Max Teplitski

Copyright © 2011 A. Mark Ibekwe et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Persistence of Escherichia coli (E. coli) O157:H7 in the environment is a major concern to vegetable and fruit growers where farms and livestock production are in close proximity. The objectives were to determine the effects of preplant fumigation treatment on the survival of E. coli O157:H7 in two soils and the effects of indigenous bacterial populations on the survival of this pathogen. Real-time PCR and plate counts were used to quantify the survival of E. coli O157:H7 in two contrasting soils after fumigation with methyl bromide (MeBr) and methyl iodide (MeI). Ten days after fumigation, E. coli O157:H7 counts were significantly lower (P = .0001) in fumigated soils than in the non-fumigated. Direct comparison between MeBr and MeI within each soil indicated that these two fumigants showed similar impacts on E. coli O157:H7 survival. Microbial species diversity as determined by DGGE was significantly higher in clay soil than sandy soil and this resulted in higher initial decline in population in clay soil than in sandy soil. This study shows that if soil is contaminated with E. coli O157:H7, fumigation alone may not eliminate the pathogen, but may cause decrease in microbial diversity which may enhance the survival of the pathogen.

1. Introduction (use of uncomposted manures) and planting and growing (use of contaminated irrigation water and animal manures Appropriate management of farm waste such as manure is and manure from animals grazing locally or nearby) [6–8]. critical in controlling the spread of pathogens such as E. coli Prevention of preharvest contamination of fresh produce is O157:H7 to vegetable crops. In most management schemes, an essential part of systems approach focused on interven- fumigants are used for the control of plant pathogens, tions designed to achieve delivery of microbiological safe nematodes, and weeds before high-value cash crops such produce to consumers [9]. Suppression of human pathogens as strawberry and tomato are planted. Outbreaks of E. in agricultural soils and the subsequent prevention of spread coli O157:H7 infections historically have been associated into the food chain by contamination of produce must be with consumption of undercooked ground beef; however, realized by the adoption of best management practices. many recent outbreaks have resulted from consumption In the absence of known phytopathogens, many crops of contaminated raw vegetables, including lettuce [1–3]. have exhibited an increased growth response when planted Although many pathogens have been associated with fresh into soil that had been fumigated with MeBr [10] at the produce, E. coli O157:H7 is of particular concern because recommended application rate. One of the likely reasons ingestion of relatively few cells can cause illness [4].E.coli for this observation may be that fumigation altered the O157:H7 can survive for 60 to 120 days in water and in soil, microbial composition of the soil, either enhancing beneficial and under dry and acidic conditions [5]. colonizers or reducing populations of deleterious rhizo- The steps in the production chain that have the greatest sphere colonizers. Fumigation is to control plant pathogens potential for pathogen contamination are soil preparation such as nematodes, soil-borne diseases, and weeds. The 2 International Journal of Microbiology immediate impact of fumigation may be the reduction of following inoculums concentrations in triplicate: 101,103, certain bacterial species in the soil and the development 104,106,108, and 0 for both autoclaved and unautoclaved of new communities after a few weeks [11, 12]. Some soils.Serial dilution was made with 10 g portion of each soil fumigants may be toxic to some microbes, and this may for the enumeration of bacteria. Both the clay and sandy enhance the selection of microbes that may be beneficial soils were treated the same. Soils were mixed for 5 min to plants [11, 12]. Due to the increased focus on food with sterile specula to homogeneously distribute the E. coli safety related to fresh produce, there are more studies O157:H7, and covered with foil and incubated at 20◦ C in the using real-time PCR to quantify pathogens such as E. coli growth chamber for the duration of the experiment. E. coli O157:H7 in the environment [13, 14]. The main objectives O157:H7 population was determined by plating on Tryptic of this study were to determine the effects of soil microbial soy agar (TSA; Becton Dickinson) plates containing 100 μg diversity and fumigation on the survival of E. coli O157:H7 of ampicillin ml−1 (TSA-A) at days 0 (inoculation), 1, 3, 5, in soils contaminated with the pathogen. To accomplish 10, 20, 30, 40, 50, and 60. The GFP-labeled E. coli O157:H7 these objectives, a preliminary study was conducted to colonies were counted under an UV light. determine the survival of the pathogen in both autoclaved and nonautoclaved soils at different concentrations of the pathogen. For our main objectives, both plate count and real- 2.3. Growth Chamber Experiment after Fumigation. The time PCR approaches were used to determine the survival of fumigant methyl iodide (MeI, iodomethane, >99% purity) E. coli O157:H7 in the two soils. was purchased from Chem Service (West Chester, PA) and methyl bromide (MeBr >99% purity) was obtained from Great Lakes Chemical Company (West Lafayette, IN). Plastic 2. Materials and Methods trays (58.2 × 43.2 × 18.5 cm) were filled with approximately 40 kg of soil. The soils were irrigated with approximately 2.1. Bacterial Strain and Growth Conditions. E. coli O157:H7 2.2 ×108E. coli O157:H7. Bacteria were inoculated into the (pGFP) strain 72 was kindly provided by Dr. Pina Fratamico irrigation lines with a Cole-Parmer HPLC pump (Cole- of USDA-ARS [15]. This strain produces Shiga-like toxin Parmer, Chicago, Illinois) and delivered through PVC pipes Stx1andStx2 and the pGFP expressing green fluorescent to each tray with five surface drip lines. The five drip lines protein (GFP) and ampicilin resistance. E. coli O157:H7 6 −1 ◦ delivered about 10 CFU g E. coli O157:H7 at time zero. was cultured at 37 C overnight in modified Tryptic Soy Soil samples were collected on the day of inoculation for broth (mTSB)(Difco Laboratories Inc., Cockeysville, MD) − community analysis, E. coli O157:H7(pGFP) concentration supplemented with 100 μg of ampicillin ml 1 (Sigma, St and heterotrophic plate counts. After the initial sample Louis, MO). Cells were harvested by centrifugation at 3500 X collection, trays were manually covered with a virtually g for 10 min and resuspended in phosphate buffered saline impermeable plastic film; 0.038 mm Hytibar film (Klerk (PBS) (Fisher Scientific, Pittsburgh, PA) to a concentration − Plastics, Belgium) and fumigants were applied. of ∼108 CFU ml 1. Bacterial strains (except strain 72) used to Fumigant rates and application methods were selected determine the specificity and sensitivity of PCR assays were according to the recommended field application rate for obtained from the National Animal Disease Center (Ames, each chemical by California Department of Pesticide Reg- IA) and were cultured on Luria-Bertani (LB) broth agar and ◦ ulation (http://www.cdpr.ca.gov/). To avoid the emission of Sorbitol MacConkey (SMAC) agar plates at 37 C. fumigants to the growth chamber, syringes were used to inject fumigant (MeBr-gas and MeI-liquid) into the trays; 2.2. Effect of Inoculum Density on the Survival of E. coli injection ports covered immediately with duct tape and left O157:H7 in Soils. Clay soil (willows silty clay, saline-alkaline) in the growth chamber for 10 days. After 10 days, trays were and sandy soil (dello loamy sand) were collected from Mystic moved outside and the Hytibar film was removed. Trays Lake dry bed and the Santa Ana River bed, respectively, and remained outside in an area covered with barb wires, opened treated as described by Ibekwe and Grieve [16]. The clay soil and aerated for 2 days before they were moved back to the had a bulk density of 1.51 Mg m−3 with 3.7% sand, 49.1% growth chamber for the continuation of the experiment. At silt, and 47.2% clay. The sandy soil has a bulk density of this point, a total of 14 days has elapsed since fumigation, 1.67 Mg m−3 with 99.1% sand, 0.20% silt, and 0.70% clay. and soil samples were collected for E. coli O157:H7(pGFP) The moisture content of the clay soil was 4.02 % and that of concentration, bacterial diversity, and heterotrophic plate sandy soil was 5.32%, before both were increased to about counts. 12% at the start of the experiment. The pH of sand was 6.85 Soils were irrigated with 50% Hoagland solution [17]in and that of clay was 7.45. Each soil (100 g dry wt) was placed the two growth chambers. The experiment was a completely in 150 ml beakers (18 beakers of each soil type) and the soil randomized design with two replications. The clay soil was was autoclaved for 1 h, cooled for 24 h, and autoclaved again irrigated with distilled water daily and received nutrient beforeuseforthestudy.Eighteenbeakersofeachsoiltype solution weekly. The sandy soil received the above nutrient were also kept unautoclaved for comparison. Inoculums were solution twice daily, because this was a very poor river made in PBS and added to the appropriate soil beaker by bank soil with poor nutrients for plant growth. Soil samples spraying and mixing 10 ml of the culture mixture with a were collected weekly for 5 weeks for E. coli O157:H7, spray bottle (Sprayco, Detroit, Mich.; sanitized by soaking in heterotrophic plate counts and for total bacterial DNA 70% ethanol) on the surface of 100 g of soil to obtain the extraction. The samples were collected in separate sterile International Journal of Microbiology 3 petri dishes or collection bags. Negative controls were of 10X TaqMan buffer (PE Applied Biosystems, Foster City, collected first with sterile spatula. Soil was transferred to CA), 0.3 μMofeachprimer,0.1μMofprobe,and3.5mM ziplock bags and 10 g sample was used for serial dilution. of MgCl2. Genomic DNA purified from E. coli O157:H7 was E. coli/pGFP colonies were enumerated under a hand- used as a template for the positive control and no template held Spectroline ultraviolet lamp (Spectronics Corporation, for negative control. PCR was performed using the following Westbury, N.Y). cycle conditions: denaturation at 95◦C for 10 min, 50 cycles of 94◦C for 20 s, 55◦C for 30 s, 72◦Cfor40s,followedby ◦ ◦ 2.4. DNA Extraction, PCR Amplification, and DGGE Analysis. a 5 min extension at 72 Candaholdat4C. Standard Community DNA was extracted from 0.5 g soil with the Ultra curves generated from plotting the threshold cycle (CT ) Clean Soil DNA Kit (MoBio Laboratories, Solana Beach, CA) versus log10 of starting DNA quantities (pg) were used for according to the manufacturer’s protocol and stored at −20◦ determining the detection limit of the assay. Optimization of C. A 236-bp DNA fragment in the V3 region of the small the multiplex assay was done as previously discussed [21, 22]. subunit ribosomal RNA genes of eubacteria was amplified by Amplification efficiency (E) was estimated by using the slope −1/slope using primer set PRBA338f and PRUN518r [18]. Ready-To- of the standard curve and the formula: E = (10 )−1. Go PCR beads (Amersham Pharmacia Biotech, Piscataway, Reaction with 100% efficiency generated a slope of −3.32. NJ) and 5 pmol of primers in a total volume of 25 ml were used in the PCR reaction. PCR amplifications were done ◦ 2.6. Data Analysis. E. coli O157:H7 concentrations were under the following conditions: 92 C for 2 min; 30 cycles of converted to log CFU g−1 for regression analysis. Statistical 92◦C for 1 min, 55◦C for 30 s, 77◦Cfor1minfollowedbya ◦ analyses were done with the general linear model (GLM) pro- final extension at 72 C for 6 min. DGGE was performed with cedure of the Statistical Analysis System [19]. The population 8% (wt/vol) acrylamide gels containing a linear chemical data were log transformed to obtain a normal distribution gradient ranging from 30% to 70% denaturant with 100% of the data. Comparisons between pairs of treatment means defined as 7 M urea and 40% formamide. Gels were run at any date were accomplished with the Tukey’s test. The for 3 h at 200 V with the Dcode Universal Mutation System log-transform data of E. coli O157:H7 population size of all (Bio-Rad Laboratories, Hercules, CA). DNA was visualized individual samples were plotted over time after inoculation, after ethidium bromide staining by UV transillumination and analyzed by regression analysis [19]. Plate counts and and photographed with a Polaroid camera. DNA fingerprints real time PCR data were transformed to Log10 values and obtained from the 16S rRNA banding patterns on the DGGE survival curves were obtained by plotting the logarithm gels were photographed and digitized using ImageMaster ofsurvivorsagainstthetreatmenttime.Thesurvivaldata Labscan (Amersham-Pharmacia Biotech, Uppsala, Sweden) were fitted to a biphasic model as proposed by Coroller et andanalyzed[11]. The comparison of diversity was done by al. [23, 24] with the Geeraerd and Van Impe inactivation using a one-way analysis of variance, and Tukey HSD test model-fitting tool (GInaFiT) as shown in (2)and(3)andas for post hoc analysis [19]. Diversity was calculated by using described by Franz et al. [25]: the Shannon index of diversity (H)tocomparechangesin diversity of microbial communities within all treatments at   N p p 0 −(t/δ ) +α −(t/δ ) each time [20] by using the following function: N(t) = α 10 1 +10 2 ,(2)  1+10   H =− Pi log Pi,(1) f α = log10 − f ,(3) when Pi = ni/N, ni is the height of peak, and N is the sum 1 of all peak heights in the curve.

where N is the number of survivors, N0 is the inoculums 2.5. Primer and Probe Design for Real-Time PCR. Genomic size; t is the time; p is the shape parameter, when p>1 DNA Was Isolated from pure culture of E. coli O157:H7, a convex curve is observed; when p<1aconcavecurve grownfor12hat37◦CandextractedwiththeQiagen is observed, when p = 1 a linear curve is observed. tissue kit (QIAamp DNA Mini Kit; Valencia, CA). DNA The scale parameter, δ, represents the time needed for first extracted from O157:H7 was used for the construction of decimal reduction. f ,varyingfrom0to1,isthefraction standard curve and for the determination of detection limits of subpopulation 1 in the population. Another parameter, α, of the E. coli by real-time PCR. Total bacterial DNA was varying from negative infinity to positive infinity, is obtained extracted from soil with the Ultra Clean Soil DNA Kit by logit transformation of f as shown in equation 2. The (MoBio Laboratories, Solana Beach, CA) as stated above strong correlation between the scale (δ) and the shape (p) and stored at −20◦C. Primers and probes used for the parameters makes it possible for the double Weibull model detection and quantification of the stx1, stx2, and the eae to fit most of the shapes of deactivation curves. Additionally, gene in E. coli O157:H7 were as described [21, 22]. Real- when δ1 = δ2, the double Weibull model can be simplified time, quantitative PCR was performed with the iCycler iQ into a single Weibull model, and the survival curve can (Bio-Rad, Hercules, CA) as described by Ibekwe et al. [21]. be described by only three parameters. A very important Briefly, PCR was performed in a total volume of 50 μlvolume and useful parameter, time to detection limit (Td) can also containing 200 μMofdNTPs,2μl of genomic DNA from be calculated when using GInaFiT to fit the experimental each concentration, 2.5 U of AmpliTaq Gold polymerase, 5 μl survival data. 4 International Journal of Microbiology

10 10 Sand Clay 8 8

soil) soil) 6 1 1 − 6 − 4 4 2

log 10 N (CFU g 2 log 10 N (CFU g 0

0 010203040 50 60 010203040 50 60 Time (days) Time (days) Unautoclave soil 104 Autoclave soil 104 Unautoclave soil 104 Autoclave soil 104 Unautoclave soil 108 Autoclave soil 108 Unautoclave soil 108 Autoclave soil 108 (a) (b)

600 120 Sand Clay

500 ) 100 10

400 80

300 60

200 40 Parameter estimates

100 Parameter estimates (log 20

0 0 αδ1 P log 10 (NO) δ2 αδ1 P log 10 (NO) δ2 Modeling parameters Modeling parameters

Unauto 104 Auto 104 Unauto 104 Auto 104 Unauto 108 Auto 108 Unauto 108 Auto 108 (c) (d)

Figure 1: Survival of E. coli O157:H7 in autoclaved and unautoclaved (a) sand and (b) clay soils after 60 days incubation in a growth chamber at 20◦C. E. coli O157: H7 was enumerated on TSA-ampicilin plates from soil with different concentrations of pathogen. Each point represents the average of triplicate sampling, and the bars at each point indicate survival coefficient obtained at each sampling point. All were log transformed as reported in the text. Double Weibull model parameters of E. coli O157:H7 in autoclave and nonautoclave sandy and clay soils (c, d)

3. Results there was an increased in population by 2.13 log10 CFU −1 4 g in the 10 dilution and an increased of 1.68 log10 CFU 3.1. E. coli O157:H7 Survival in Clay and Sandy Soils. The g−1 in the 108 dilution. Survival curves showed a concave ff e ects of inoculum density on the survival of E. coli O157:H7 curvature in autoclave sandy soil and a convex curvature in in sandy and clay soils was first determined in the two the unautoclaved sandy soil (Figure 1(a)). In the clay soil, soils used for this study. This was done to determine the the shape parameter was different from sandy soil with soils influence of indigenous microorganism on the survival of with inoculums density of 108 showing the convex shape E. coli O157:H7 in autoclaved and unautoclaved soils. Data whereas soils with cells at 104 showed either concave or linear from the survival study showed that within the first 7 days shape. Modeling parameters (alpha (α), delta (δ), and the E. coli O157:H7 populations decreased by ca. 0.24 log10CFU shape parameter-p) were calculated from equation (2)and −1 4 −1 g in the 10 dilution and by 0.67 log10CFU g in the (3) used to explain the inactivation kinetics. More variations 8 10 dilution for the unautoclaved soil (Figures 1(a) and in δ values were observed from different soils (Figures 1(c) 1(b)). The reverse was the case with autoclaved soil where and 1(d)). When the strain was characterized in sandy and International Journal of Microbiology 5

−1 Table 1: Mean comparison by days and methods (log10g ) of survival of E. coli in soil after fumigation.

Day Day Day Day Day Day Day Day Day Day Day Day Day Day TRT 10-PC 23-PC 28-PC 36-PC 50-PC 10-Stx 1 10-Stx 2 10-eae 23-Stx 1 23-Stx 2 23-eae 28-Stx 1 28-Stx 2 28-eae 4.22 Sb1/2e 4.11 a 3.25 a 2.48 a 1.00 a 6.12 ab 5.67 a 7.10 abc 4.71 a 3.34 a 6.12 a 4.81 ab 2.86 ab 5.97 ab ab 3.91 2.00 sb1e 4.07 a 3.22 a 1.00 a 5.72 ab 4.72 ab 6.77 abc 4.21 a 2.57 a 6.00 a 7.11 a 3.90 ab 5.41 ab ab ab 3.39 sI1/2e 3.74 a 1.00 a 1.00 b 1.00 a 6.73 a 5.57 a 7.87 a 5.73 a 3.88 a 6.51 a 4.95 ab 2.85 ab 5.88 ab abc 3.88 sI1e 3.93 a 2.92 a 1.00 b 1.00 a 6.09 ab 4.19 abc 7.09 abc 5.43 a 3.73 a 6.66 a 5.65 ab 3.36 ab 6.81 a ab so0e 4.99 a 3.48 a 4.16 a 2.84 a 3.53 a 6.34 a 4.75 ab 7.41 ab 5.07 a 3.17 a 4.37 a 4.42 ab 4.16 ab 5.29 ab 3.65 cb1/2e 3.96 a 3.21 a 1.00 b 2.18 a 4.09 bc 1.00 e 3.59 bc 4.59 a 2.88 a 3.55 a 3.90 b 1.00 b 2.44 b ab 3.67 cb1e 2.25 a 2.17 a 1.00 b 1.00 a 5.09 abc 3.10 cd 5.98 abc 4.48 a 3.57 a 4.47 a 4.45 ab 3.41 ab 2.44 b ab cI1/2e 1.00 c 1.99 a 1.25 a 1.00 b 1.00 a 4.67 abc 2.45 d 4.30 abc 4.39 a 2.44 a 3.79 a 3.61 b 2.60 ab 3.19 b 4.16 cI1e 4.44 a 0.58 a 1.00 b 1.00 a 5.34 abc 3.98 bc 6.65 abc 4.78 a 3.36 a 4.75 a 4.16 b 1.22 ab 5.02 ab ab 1.41 co0e 3.85 a 3.01 a 2.59 a 3.10 a 3.27 c 2.44 d 3.08 c 5.32 a 2.57 a 6.24 a 4.10 b 5.86 a 3.50 ab bc Means with different letters within each column are significantly different at P ≤ .05 using Tukey’s Studentized Range Test. s =sandy soil; b = methyl bromide, I = methyl Iodide; c = clay soil; 1/2 = 50% recommended application rate; 1 = recommended application rate; e = sample were inoculated with E. coli O157:H7.

clay soils, distinct δ1 and δ2 were observed indicating that no differences in the levels of heterotrophic plate count in the the two subpopulations behaved differently in both soils, two soils during the study period (data not shown). Mean thus the survival data in both soils might not be simplified comparison by days and methods were used to determine into the single Weibull model that can be described by only the impact of fumigants on the survival of E. coli O157:H7 three parameters, α, δ and p. The initial sharp decrease in in the two soils after fumigation (Table 1). Since one of cell numbers in sandy soil (concave shape) might largely be our objectives was to determine the effects of fumigants on attributed to the faster decline of subpopulation as shown E. coli O157:H7 on a weekly basis, direct comparison of with smaller δ1 (Figure 1(c)). However, with the time going, the two fumigants and the control was done using plate the subpopulation with greater δ2 (i.e., the more resistant) count and real-time PCR to quantify the concentrations of dominated the cell population, leading to a slower and E. coli O157:H7. In the growth chamber soil, ten days after steadily decline of the cell concentration as the curves showed fumigation, E. coli O157:H7 was significantly lower (P = little or no decline. After the first 10 days, E. coli O157:H7 .0001) in fumigated soils than the control clay soil at the populations in unautoclaved soil declined considerably more recommended application rate. During the rest of the study, rapidly than in autoclaved soil below the detection limits of there were no significant differences on the effect of the two 102 CFUg−1 soil. After 60 days, the concentration of E. coli fumigants on the pathogen, except on day 36 (P = .046) O157:H7 in the 104 dilution was undetectable by plate count, where the effects varied. Real-time PCR analysis showed that −1 8 and there was a 6.18 log10CFU g reduction for the 10 10 days after fumigation, E. coli O157:H7 concentration in dilution. Survival of pathogen was greater in the sandy soil non-fumigated soils was significantly higher (P = .002) in (P = .05) than clay unautoclaved soil within the first 7 days sandy soil than clay soil. There were no significant differences (Figure 1). E. coli O157:H7 at 108 CFU g−1dilutions survived (P = .56) in pathogen concentration during day 23 when for more than 60 days in both unautoclaved and autoclaved real-time PCR was used for the analysis. The same effect soils used in this study. was observed during day 36 and 50 (data not shown). Direct comparison between MeBr and MeI within each soil showed that neither had significant greater impact on E. coli 3.2. Impact of Fumigants on Survival E. coli O157:H7 in O157:H7. the Growth Chamber. Before the enumeration of E. coli The majority of the survival curves (Figure 2) showed a O157:H7 in the different matrices, background concentra- concave shape, with a relatively fast initial decline followed by tions of heterotrophic bacterial was determined. The initial a slower decline phase. Survival reached the detection limit heterotrophic plate count in soil was 2.1 × 108 CFU g−1.After faster in clay soil than in sandy soil without fumigation using storage at 20◦C in the growth chamber, the total aerobic plate plate counts (Figures 2(a) and 2(b)). When the pathogen was counts decreased steadily from ca. 108 to ca. 106 CFU g−1± exposed to fumigants (MeI), inactivation was faster than in 102 during the experimental period in both soils. There were control, especially with plate count (Figures 2(c) and 2(d)). 6 International Journal of Microbiology

ThesamepatternwasobservedwithMeBr(Figures2(e) and Table 2: Numerical analysis of DGGE bands from growth chamber 2(f)). However, for both control treatments the population soil samples with Shannon index of diversity (H). size did not reach the detection limit (ttd) of 102 CFU g−1 Treatment Week 1 Week 3 Week 4 Week 5 Week 7 during the experiment due to earlier onset of tailing at about E∗ 35 days using real-time PCR. Also, both soils showed that it CB 0.5X+ 1.34 abc 0.88 a 1.03 a 0.13 a 0.58 abc CB 0.5X−E 0.88abcd 0.66 a 0.30 a 0.53 a 0.91 a took less than a day to inactivate the first log10 of microbial population in most of the fumigated samples. CB 1.0X+E 1.56 a 0.55 a 0.65 a 0.37 a 0.70 ab Effects of soil types on the survival E. coli O157:H7 in CB 1.0X−E 1.27 abc 0.33 a 0.82 a 0.71 a 0.30 bc clay and sandy soils after fumigation was model by fitting CI 0.5X +E 1.12 abc 0.95 a 0.85 a 0.38 a 0.70 ab the experimental data into the survival functions (Figure 3). CI 0.5X −E 1.22 abc 0.72 a 0.94 a 0.86 a 0.96 a α δ Similar modeling parameters ( , ,andp) were calculated CI 1.0X+E 1.48ab 0.68 a 0.79 a 0.67 a 0.78 ab when they were inoculated into the same soil (Figures CI 1.0X−E 1.45 ab 0.57 a 0.42 a 0.89 a 0.79 ab 3(a)–3(f)). However, there more little variations in these CO 0+E 1.10 abcd 0.76 a 0.89 a 0.78 a 0.82 ab parameters from the two soils, except the δ values. When the SB 0.5X+E 0.82 abcd 0.96 a 0.49 a 0.39 a 0.001 c pathogen was characterized in sandy and clay soils, distinct −E δ and δ were observed indicating that the two subpopu- SB 0.5X 0.45 bcd 0.98 a 0.35 a 0.47 a 0.58 abc 1 2 E lations behave differently in both soils. The goodness-of-fit SB 1.0X+ 0.30 cd 0.75 a 0.55 a 0.51 a 0.001 c statistics (R2) did not differ significantly between survival SB 1.0X−E 0.27 cd 1.04 a 0.57 a 0.29 a 0.64 ab curves in sandy soil irrespective of fumigants or no fumigant. SI 0.5X +E 0.79 abcd 1.20 a 0.76 a 0.90 a 0.59 ab The same effect was observed in clay soil, indicating that the SI 0.5X −E 0.49 abcd 0.74 a 0.63 a 0.26 a 0.30 bc model is suitable to fit survival curves of E. coli O157:H7 in SI 1.0X+E 0.75 abcd 0.87 a 0.30 a 0.57 a 0.83 ab an array of different soils. SI 1.0X−E 0.001 d 0.73 a 0.49 a 0.72 a 0.45 abc SO 0+E 1.11 abc 0.89 a 0.52 a 0.38 a 0.38 abc 3.3. Influence of Microbial Diversity on E. coli O157:H7 SO 0−E 0.95abcd 0.75 a 0.61 a 0.59 a 0.59 ab Survival. DGGE analysis of 16S rRNA fragments was used Means with different letters within each column are significantly different to examine the effects of MeBr and MeI on soil microbial at P ≤ .05 using Tukey’s Studentized Range Test. ∗C =clay soil; B = methyl = = communities during week 1 to7 after fumigation. The most bromide, I methyl Iodide; S sandy soil; 0.5X, and 1X represents half ff agricultural application rate and recommended agricultural application drastic e ect occurred on the first week of the experiment rate. + E = E. coli O157:H7 applied to the samples and – E not applied to where there was a significant effect (P ≤ .05) of fumigants the sample. as determined by the Shannon-Weaver index of diversity between clay and sandy soil (Table 2). During this period, microbial diversity in the MeBr fumigated treatments was microorganisms was likely a factor in killing E. coli O157:H7 significantly lower (P = .0003) in sandy soil than in clay soil cells in unautoclaved soils used in this study. E. coli O157:H7 at the recommended application rate. The same effect was was inactivated more rapidly in unautoclaved soil than in observed with MeI fumigated soil. Bacterial communities autoclaved soil in all the different dilutions of E. coli O157:H7 were not different at week three (P = .13), week four used as inoculums (Figures 1(a) and 1(b) for 104 and (P = .06), and week five (P = .11). However, at week − 108 CFU g 1). Our study is in agreement with Jiang et al. seven there was a significant (P = .0001) shift in microbial [26] who showed that small numbers of E. coli O157:H7 community structure as determined by diversity index with from an unautoclaved soil were only detected by enrichment all the treatments (Table 2), and the effect was greater in ◦ culture, and survived for longer period of time at 15 C than sandy soil than clay soil. This resulted in the initial higher ◦ 21 C. This suggests that there was a small population of cells survival rate of E. coli O157:H7 in sandy soil compared to that have survived in the soil under different environmental clay soil. Microbial diversity (expressed as Shannon Weaver stress. The long-term survival of this pathogen in the index of diversity H) was positively correlated with survival environment has been reported by many authors [6, 27–31], of E. coli O157:H7 in sandy soils (Figure 4(a); r2 = 0.56, but very little has been done on the survival in fumigated P = .015) and in clay soil using the plate count method soil. We have shown from this study that E. coli O157:H7 can (Figure 4b; r2 = 0.47; P = .019). Using data from real- survive in fumigated soils for over 60 days due to long-term time PCR analysis, survival of E. coli O157:H7 were positively persistence of a small percent of the population. correlated with microbial diversity in both clay and sandy Our study showed that E. coli O157:H7 can survive soils (data not shown). longer in sandy soil than in clay soil during a short term experiment. However, populations persisted longer in clay 4. Discussion than in sandy soil during a long-term study. Our results showed that E. coli O157:H7 survived longer than 60 days Before the start of this experiment, a preliminary study in both soils. Others have reported survival of more than was conducted in autoclaved and unautoclaved soil to 54 days in manure amended soil [28, 32–34] and 34 days determine the influence of indigenous soil microorganisms or more in sandy loam soil amended with cow manure on the survival and growth of E. coli O157:H7 (Figures [26, 35], and over 90 days in clay soils (8, 40). Others 1(a) and 1(b)). The antagonistic effect of indigenous soil reported longer E. coli O157:H7 survival times of between International Journal of Microbiology 7

7 7 Sand Clay 6 6 soil) soil) 1 1

− 5 − 5

4 4

3 3

2 2 log 10 N (CFU g log 10 N (CFU g

1 1

0 0 010203040 50 60 010203040 50 60 Time (days) Time (days) Sandy soil-0X-PC Sandy soil-0X-stx2 Clay soil-0X-PC Clay soil-0X-stx2 Sandy soil-0X-stx1 Sandy soil-0X-eae Clay soil-0X-stx1 Clay soil-0X-eae (a) (b)

7 7 Sand mel Clay mel 6 6 soil) soil) 1 1 − − 5 5

4 4

3 3 2

2 log 10 N (CFU g log 10 N (CFU g

1 1

0 0 010203040 50 60 010203040 50 60 Time (days) Time (days) Sandy soil-1X-PC Sandy soil-1X-stx2 Clay soil-1X-PC Clay soil-1X-stx2 Sandy soil-1X-stx1 Sandy soil-1X-eae Clay soil-1X-stx1 Clay soil-1X-eae (c) (d)

7 7 Sandy MeBr Clay MeBr 6 6 soil) soil) 1 1

− 5 − 5

4 4

3 3

2 2 log 10 N (CFU g log 10 N (CFU g

1 1

0 0 010203040 50 60 010203040 50 60 Time (days) Time (days) Sandy soil-1XI-PC Sandy soil-1XI-stx2 Clay soil-1XI-PC Clay soil-1XI-stx2 Sandy soil-1XI-stx1 Sandy soil-1XI-eae Clay soil-1XI-stx1 Clay soil-1XI-eae (e) (f)

Figure 2: Quantification of E. coli O157:H7 in growth chamber after 60 days in fumigated soils. Fumigant (MeBr) was applied at the normal application rate: non fumigated control: (a) (sandy soil), (b) (clay soil); normal application rate for MeI (c) (sandy soil), (d) (clay soil); normal application rate for Mebr (e) (sandy soil), D (clay soil); E. coli O157:H7 were enumerated from sandy and clay soil by plate count and by real-time PCR (RT-PCR) using stx1, stx2, and the eae genes. 8 International Journal of Microbiology

600 140 Sand 0X Clay 0X

500 120

100 400 80 300 60 200

Parameter estimates Parameter estimates 40

100 20

0 0 αδ1 P log 10 (NO) δ2 αδ1 P log 10 (NO) δ2 Modeling parameters Modeling parameters (a) (b)

800 600 Sand mel 1X Clay mel 1X

500 600 400

400 300

200 Parameter estimates 200 Parameter estimates 100

0 0 αδ1 P log 10 (NO) δ2 αδ1 P log 10 (NO) δ2 Modeling parameters Modeling parameters (c) (d)

500 350 Sand MeBr 1X Clay MeBr 1X 300 400 250

300 200

150 200

Parameter estimates Parameter estimates 100 100 50

0 0 α δ1 P log 10 (NO) δ2 α δ1 P log 10 (NO) δ2 Modeling parameters Modeling parameters Plate counts Stx2 Plate counts Stx2 Stx1 eae Stx1 eae (e) (f)

Figure 3: Double Weibull Model parameters of E. coli O157:H7 in nonfumigated sandy and clay soils (a and b), fumigated with MeI in sandy and clay soils (c and d), and fumigated with MeBr in sandy and clay soils (e and f). International Journal of Microbiology 9

) 7 ) 6 1 1 − −

6 5 CFU g CFU g 10 10 5 4

4 3 O157:H7 (log O157:H7 (log 3 2 E. coli E. coli

2 1

Survival of 1 Survival of 0 00.20.40.6 0.811.21.4 00.20.40.6 0.811.21.4 Diversityindex(H)byDGGE Diversityindex(H)byDGGE (a) (b)

Figure 4: Obseved points (•) and relations (solid lines between survival and microbial diversity in soil. (a) sandy soil-growth chamber; (b) clay soil- growth chamber.

154 and 217 d in soils amended with inoculated compost the population are rapidly eliminated and that the sturdier [27]. This study with longer survival period agrees with survivors remain. our study because both studies relied on inoculating the Longer persistence of E. coli O157:H7 in clay soil may substrate with relatively high densities of the pathogen (>106 be influenced by interaction between soil particles in the CFU g−1). Also, during our preliminary experiment with E. clay particle sizes that provided niches for the pathogen coli O157:H7 with population of lesser than 104 CFU g−1, and moisture/nutrients within the niches. Other factors that survival of the pathogen was less than seven days. This is in contributed to the survival of pathogen in soil were soil agreement with Franz et al. [36] that monitored the fate of microbial diversity. Recently, the effects of E. coli O157:H7 the pathogens in manure-amended soil after they declined was assessed in a loamy sand soil obtained from species-rich to below detection limit within a short period with low and grassland, in which the microbial community composition more realistic levels of pathogens inoculated into manure had been modified by progressively enhanced fumigation (approximately 102 CFU g−1). Therefore, the survival of E. depths [40]. The authors showed that E. coli O157:H7 in the coli O157:H7 in the environment may depend on the initial soils with modified community structures due to fumigation concentration at the beginning of the experiment. was clearly consistent with the hypothesis that within the In this study, we used the double Weibull model, which single selected habitat (soil), which was relatively unaffected is the cumulative form of the underlying distribution of with respect to abiotic conditions like pH, moisture and soil individual inactivation kinetics, and it was a suitable model chemical conditions, microbial community structure was the for describing the decline of E. coli O157:H7. The survival main determinant of the survival of the pathogen. With the curves generated from our study in most cases showed a present study we found that the values of the log reduction convex fitting, indicating changes in biological stress over time and the shape parameter of the double Weibull model time. The model is sufficiently flexible to account for differ- were higher for clay soils compared to sandy soils. This means ent survival patterns and has been previously used to model that with sandy soils the initial rate of decline of E. coli thermal inactivation of Listeria monocytogenes in sucrose O157:H7 was faster than in clay soil. E. coli O157:H7 was solutions of various water activities [37] and the survival more vulnerable to mortality during the first few weeks in the of E. coli O157:H7 in manure amended soil [32]. Franz et sandy soil than in clay soil. Finer-textured (clayey) soils result al. [32]; Van Boekel, [38]; Peleg, [39] discussed the different in prolonged survival of introduced bacteria compared with processes and mechnisms responsible for the different shape coarser-textured (sandy) soils because of higher availability parameters during inactivation of E. coli O157:H7 in soil. of protective pore spaces against feeding by soil fauna like They pointed out that even though the Weibull model is an protozoa [41]. This could explain the faster initial decrease empirical model, it can be linked to physiological properties in E. coli O157:H7 numbers in the sandy soils compared with at population level and that the population is heterogeneous the loamy soils, but survivors are increasingly more sturdy with respect to the stress encountered in the soil. These compared with survival in the clay soils. The implication authors noted that a convex curve would mean that the of long-term survival of this pathogen in the environment remaining cells become increasingly susceptible to stress, may involve the recontamination of the environment after and the cells are therefore subjected to more damages with the initial contamination event from few surviving strains. time. A linear survival curve means that inactivation does E. coli O157:H7 was significantly lower in fumigated soils not depend on time or other biological activities and the than the control at the normal application rate during the concave survival curve means that sensitive members of first 2 weeks of the experiment [42]. However, survival was 10 International Journal of Microbiology significantly affected by soil type, with survival greater in [10] S. Wilhelm and A. O. Paulus, “How soil fumigation benefits sandy soil in the short run than clay soil [43]. However, the California strawberry industry,” Plant Diseases, vol. 64, pp. long-term persistence occurred in clay soil than sandy soil 264–270, 1980. due to properties of clay soil. None of the fumigants showed [11] A. M. Ibekwe, S. K. Papiernik, J. Gan, S. R. Yates, C.-H. Yang, significant higher toxicity effects on the pathogen at the andD.E.Crowley,“ImpactofFumigantsonSoilMicrobial normal application on the long run, and toxicity was higher Communities,” Applied and Environmental Microbiology, vol. 67, no. 7, pp. 3245–3257, 2001. in fumigated soil than nonfumigated soils during the first [12] R. S. Dungan, A. M. Ibekwe, and S. R. Yates, “Effect of two weeks of the study. Therefore, MeBr and MeI have ff propargyl bromide and 1,3-dichloropropene on microbial identical toxicity e ects on E. coli O157:H7 at the normal communities in an organically amended soil,” FEMS Micro- application rate. biology Ecology, vol. 43, no. 1, pp. 75–87, 2003. [13] C. A. Heid, J. Stevens, K. J. Livak, and P. M. Williams, “Real time quantitative PCR,” Genome Research, vol. 6, no. 10, pp. Acknowledgments 986–994, 1996. [14]R.D.Oberst,M.P.Hays,L.K.Bohraetal.,“PCR-based The authors thank Ms Pamela M. Watt for technical assi- DNA amplification and presumptive detection of Escherichia stance especially with real-time PCR and Ms Phyllis Nash coli O157:H7 with an internal fluorogenic probe and the for statistical analysis. Also, Jack Jobes for collecting soil 5’ nuclease (TaqMan) assay,” Applied and Environmental samples from the Santa Ana River bed and Misty Lake. Also, Microbiology, vol. 64, no. 9, pp. 3389–3396, 1998. John Draper for building the irrigation system inside the [15] P. M. Fratamico, M. Y. Deng, T. P. Strobaugh, and S. A. growth chamber. This research was supported by the 206 Palumbo, “Construction and characterization of Escherichia Manure and Byproduct Utilization Project of the USDA- coli O157:H7 strains expressing firefly luciferase and green ARS. Mention of trademark or propriety products in this fluorescent protein and their use in survival studies,” Journal paper does not constitute a guarantee or warranty of the of Food Protection, vol. 60, no. 10, pp. 1167–1173, 1997. property by the USDA and does not imply its approval to [16] A. M. Ibekwe and C. M. Grieve, “Changes in developing plant microbial community structure as affected by contaminated the exclusion of other products that may also be suitable. water,” FEMS Microbiology Ecology, vol. 48, no. 2, pp. 239–248, 2004. References [17] D. T. Hoagland and D. I. Arnon, “The water culture methods for growing plants without soil. University of California [1]M.-L.Ackers,B.E.Mahon,E.Leahyetal.,“Anoutbreakof (Berkeley) Agriculture Experiment Station,” Circ, vol. 237, Escherichia coli O157:H7 infections associated with leaf lettuce 1950. ∅ consumption,” Journal of Infectious Diseases, vol. 177, no. 6, [18] L. vreas,˚ L. Forney, F. L. Daae, and V. Torsvik, “Distribution pp. 1588–1593, 1998. of bacterioplankton in meromictic lake Saelenvannet, as [2] M. P. Doyle, “Reducing foodborne disease,” Food Technology, determined by denaturing gradient gel electrophoresis of vol. 54, pp. 130–133, 2000. PCR-amplified gene fragments coding for 16S rRNA,” Applied and Environmental Microbiology, vol. 63, no. 9, pp. 3367–3373, [3]E.D.Hilborn,J.H.Mermin,P.A.Msharetal.,“Amultistate 1997. outbreak of Escherichia coli O157:H7 infections associated [19] SAS Institute Inc., “SAS/STAT User’s Guide,” Release 8.2. Cary, with consumption of mesclun lettuce,” Archives of Internal NC, 2001. Medicine, vol. 159, no. 15, pp. 1758–1764, 1999. [20] C. E. Shannon and W. Weaver, The Mathematical Theory of [4] R. L. Buchanan and M. P. Doyle, “Foodborne disease signif- Communication, University of Illinois Press, Urbana, Ill, USA, icance of Escherichia coli O157:H7 and other enterohemor- 1963. rhagic E. coli,” Food Technology, vol. 51, no. 10, pp. 69–76, [21] A.M.Ibekwe,P.M.Watt,C.M.Grieve,V.K.Sharma,andS.R. 1997. Lyons, “Multiplex fluorogenic real-time PCR for detection and [5] Centers for Disease Control and Prevention, National Center quantification of Escherichia coli O157:H7 in dairy wastewater for Infectious Diseases, and Division of Bacterial and Mycotic wetlands,” Applied and Environmental Microbiology, vol. 68, Diseases, “Escherichia coli O157:H7:general information,” no. 10, pp. 4853–4862, 2002. September 2003, http://www.cdc.gov/. [22] V. K. Sharma, “Detection and quantitation of enterohem- [6]M.Islam,M.P.Doyle,S.C.Phatak,P.Millner,andX.Jiang, orrhagic Escherichia coli O157, O111, and O26 in beef and “Persistence of enterohemorrhagic Escherichia coli O157:H7 in bovine feces by real-time polymerase chain reaction,” Journal soil and on leaf lettuce and parsley grown in fields treated with of Food Protection, vol. 65, no. 9, pp. 1371–1380, 2002. contaminated manure composts or irrigation water,” Journal [23] L. Coroller, I. Leguerinel, E. Mettler, N. Savy, and P. Mafart, of Food Protection, vol. 67, no. 7, pp. 1365–1370, 2004. “General model, based on two mixed weibull distributions [7] D. R. Fenlon, I. D. Ogden, A. Vinten, and I. Svoboda, “The ofbacterial resistance, for describing various shapes of inac- fate of Escherichia coli and E. coli O157 in cattle slurry tivation curves,” Applied and Environmental Microbiology, vol. after application to land,” Journal of Applied Microbiology 72, no. 10, pp. 6493–6502, 2006. Symposium Supplement, vol. 88, no. 29, pp. 149s–156s, 2000. [24] I. Albert and P. Mafart, “A modified Weibull model for bac- [8] J. V. Gagliardi and J. S. Karns, “Persistence of Escherichia terial inactivation,” International Journal of Food Microbiology, coli O157:H7 in soil and on plant roots,” Environmental vol. 100, no. 1–3, pp. 197–211, 2005. Microbiology, vol. 4, no. 2, pp. 89–96, 2002. [25] E. Franz, M. M. Klerks, O. J. De Vos, A. J. Termorshuizen, [9] L. R. Beuchat, “Pathogenic microorganisms associated with and A. H. C. Van Bruggen, “Prevalence of Shiga toxin- fresh produce,” Journal of Food Protection,vol.59,no.2,pp. producing Escherichia coli stx1, stx2, eaeA,andrfbE genes 204–216, 1996. and survival of E. coli O157:H7 in manure from organic and International Journal of Microbiology 11

low-input conventional dairy farms,” Applied and Environ- [40]J.D.vanElsas,P.Hill,A.Chronkovˇ a´ et al., “Survival of mental Microbiology, vol. 73, no. 7, pp. 2180–2190, 2007. genetically marked Escherichia coli O157:H7 in soil as affected [26] X. Jiang, J. Morgan, and M. P. Doyle, “Fate of Escherichia coli by soil microbial community shifts,” ISME Journal, vol. 1, no. O157:H7 in manure-amended soil,” Applied and Environmen- 3, pp. 204–214, 2007. tal Microbiology, vol. 68, no. 5, pp. 2605–2609, 2002. [41] J. A. van Veen, L. S. van Overbeek, and J. D. van Elsas, [27] M. Islam, J. Morgan, M. P. Doyle, and X. Jiang, “Fate of “Fate and activity of microorganisms introduced into soil,” Escherichia coli O157:H7 in manure compost-amended soil Microbiology and Molecular Biology Reviews, vol. 61, no. 2, pp. and on carrots and onions grown in an environmentally 121–135, 1997. controlled growth chamber,” Journal of Food Protection, vol. [42] A. M. Ibekwe, C. M. Grieve, and C.-H. Yang, “Survival of 67, no. 3, pp. 574–578, 2004. Escherichia coli O157:H7 in soil and on lettuce after soil [28]A.M.Ibekwe,P.M.Watt,P.J.Shouse,andC.M.Grieve,“Fate fumigation,” Canadian Journal of Microbiology, vol. 53, no. 5, of Escherichia coli O157:H7 in irrigation water on soils and pp. 623–635, 2007. plants as validated by culture method and real-time PCR,” [43] A. M. Ibekwe, P. J. Shouse, and C. M. Grieve, “Quantification Canadian Journal of Microbiology, vol. 50, no. 12, pp. 1007– of survival of Escherichia coli O157:H7 on plants affected by 1014, 2004. contaminated irrigation water,” Engineering in Life Sciences, [29] A. P. Williams, L. M. Avery, K. Killham, and D. L. Jones, vol. 6, no. 6, pp. 566–572, 2006. “Persistence of Escherichia coli O157 on farm surfaces under different environmental conditions,” Journal of Applied Micro- biology, vol. 98, no. 5, pp. 1075–1083, 2005. [30]L.M.Avery,K.Killham,andD.L.Jones,“SurvivalofE. coli O157:H7 in organic wastes destined for land application,” Journal of Applied Microbiology, vol. 98, no. 4, pp. 814–822, 2005. [31] J. M. Ritchie, G. R. Campbell, J. Shepherd et al., “A stable bioluminescent construct of Escherichia coli O157:H7 for hazard assessments of long-term survival in the environment,” Applied and Environmental Microbiology,vol.69,no.6,pp. 3359–3367, 2003. [32] E. Franz, A. V. Semenov, A. J. Termorshuizen, O. J. De Vos, J. G.Bokhorst,andA.H.C.vanBruggen,“Manure-amended soil characteristics affecting the survival of E. coli O157:H7 in 36 Dutch soils,” Environmental Microbiology,vol.10,no.2,pp. 313–327, 2008. [33] A. V. Semenov, E. Franz, L. Van Overbeek, A. J. Termor- shuizen, and A. H. C. van Bruggen, “Estimating the stability of Escherichia coli O157:H7 survival in manure-amended soils with different management histories,” Environmental Microbiology, vol. 10, no. 6, pp. 1450–1459, 2008. [34]J.M.Schouten,E.A.M.Graat,K.Frankena,A.W.vanDe Giessen, W. K. van Der Zwaluw, and M. C. M. De Jong, “A longitudinal study of Escherichia coli O157 in cattle of a Dutch dairy farm and in the farm environment,” Veterinary Microbiology, vol. 107, no. 3-4, pp. 193–204, 2005. [35] A. Maule, “Survival of verocytotoxigenic Escherichia coli O157 in soil, water and on surfaces,” Journal of Applied Microbiology Symposium Supplement, vol. 88, no. 29, pp. 71S–78S, 2000. [36] E. Franz, A. D. van Diepeningen, O. J. De Vos, and A. H. C. van Bruggen, “Effects of cattle feeding regimen and soil management type on the fate of Escherichia coli O157:H7 and Salmonella enterica serovar typhimurium in manure, manure-amended soil, and lettuce,” Applied and Environmen- tal Microbiology, vol. 71, no. 10, pp. 6165–6174, 2005. [37] A. Fernandez,´ M. Lopez,´ A. Bernardo, S. Condon,´ and J. Raso, “Modelling thermal inactivation of Listeria monocytogenes in sucrose solutions of various water activities,” Food Microbiol- ogy, vol. 24, no. 4, pp. 372–379, 2007. [38] M. A. J. S. Van Boekel, “On the use of the Weibull model to describe thermal inactivation of microbial vegetative cells,” International Journal of Food Microbiology,vol.74,no.1-2,pp. 139–159, 2002. [39] M. Peleg, “Microbial survival curves: interpretation, math- ematical modeling, and utilization,” Comments Theory of Biology, vol. 8, pp. 357–387, 2003. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 972494, 10 pages doi:10.1155/2011/972494

Research Article Adhesion of Pathogenic Bacteria to Food Contact Surfaces: Influence of pH of Culture

AkierAssantaMafu,1, 2 Corinne Plumety,1 Louise Deschenes,ˆ 1 and Jacques Goulet2

1 Food Research and Development Centre, Agri-Food and Agriculture Canada, 3600 Casavant Boulevard-West, St-Hyacinthe, QC, Canada J2S 1A2 2 Department of Science and Nutrition, Laval University, Quebec, QC, Canada G1K 7P4

Correspondence should be addressed to Akier Assanta Mafu, [email protected]

Received 26 May 2010; Revised 1 September 2010; Accepted 15 September 2010

Academic Editor: Jorge H. Leitao

Copyright © 2011 Akier Assanta Mafu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The adhesion of Aeromonas hydrophila, Escherichia coli O157:H7, Salmonella Enteritidis, and Staphylococcus aureus to hydrophobic and hydrophilic surfaces in cultures with different pHs (6, 7, and 8) was studied. The results indicated that the type of material had no effect on the attachment capacity of microorganisms, while environmental pH influenced the adhesion of A. hydrophila, E. coli, and S. aureus to both solid substrates. The attachment of S. Enteritidis (P>.05) was not affected by the type of substrate or the culture pH, whereas E. coli displayed the weakest affinity for both polystyrene and glass surfaces. No correlation was established between the physicochemical properties of the materials, or the bacterial and the rate of bacterial adhesion, except for S. aureus. Photomicrographs have shown that surfaces were contaminated by small clusters of S. Enteritidis while S. aureus invaded the food contact surfaces in the form of small chains or cell aggregates.

1. Introduction bacterial adhesion. The effect of corrosion on solid materials must also be considered since it can lead to the formation and In food processing plants, residues of all kinds chemical, bio- expansion of cavities and grooves [8]. This in turn provides logical, organic, or inorganic inevitably accumulate on the breeding sites for microorganisms, thereby compromising surfaces of equipments in contact with food [1]. Attachment the efficacy of cleaning and disinfection procedures. The of undesirable microorganisms to these surfaces is a source surface characteristics of the microorganisms themselves and of concern, since this can result in product contamination the various environmental conditions encountered in agri- leading to serious economic and health problems [2–4]. food industries (organic materials, pH, temperature, water In fact, this microbial contamination has two components: activity, etc.) also influence microbial attachment to inert first, the saprophytic flora responsible for food spoilage surfaces [2, 9–12]. and second, the pathogenic flora, which cause infections Once they have adhered to inert surfaces, the microor- in humans and animals. To adversely affect the sensory, ganisms may exhibit a greater degree of resistance to the physical, and chemical qualities of food, a large population chemical or natural cleaning and disinfecting agents used in of spoilage-causing microorganisms is required, while in the the agri-food industries compared to bacteria in suspension case of food pathogens it only takes a few cells to affect [13, 14]. product safety and cause food poisoning. The potential for attachment and development of In the phenomenon of bacterial adhesion to inert sur- microorganisms on inert surfaces as well as the resis- faces, the physicochemical properties (hydrophobicity and tance of the resulting sessile cells has been and continues charges) and substrates or surface topography are playing to be extensively studied [15–19]. Although an under- important roles [5–7].Jointssuchasvalvesandanyother standing of the parameters that govern the adhesion of difficult-to-reach spaces are the most favourable areas to these bacteria to solid surfaces could help developing new 2 International Journal of Microbiology prevention procedures at the initial stages of microbial 2.4. Rate of Adhesion of the Pathogens to Inert Surfaces. To adsorption, there are still too many unknown factors con- recover the sessile cells, two coupons of each material were cerning the adhesion capacity of the main food pathogens removed from the rods using a sterile clamp and rinsed [16]. twice in tubes containing 10 mL of saline (one tube per The objective of this study was to find out the adhe- rinse, carefully rotating the tubes) in order to eliminate sion capacity of pathogens such as Aeromonas hydrophila, the cells that had not adhered. The substrates were then Escherichia coli O157:H7, Salmonella Enteritidis, and Staphy- placed in a tube containing 10 mL of sterile phosphate buffer lococcus aureus on two commonly used materials in food (Oxoid Ltd, Basingstoke, Hampshire, England) and all of the processing plants (polystyrene and glass). The influence of adhered bacterial cells were detached in a sonication bath culture medium pH on the rate of adhesion by these agents (VWR International Inc., West Chester, PA) for 10 min. The at the interfaces was also simultaneously evaluated. tubes were vortexed for 30 s before the microbial counts were performed. After preparation of serial dilutions, the bacterial counts were determined by plating on TSA (tryptic soy agar; 2. Material and Methods Difco Laboratories, Detroit, MI) incubated at 37◦Cfor24h. 2.1. Bacterial Strains, Media, and Culture Conditions. For The relative adhesion (%) was estimated using the following this study, Aeromonas hydrophila ATCC 7966, Escherichia coli formula: O157:H7 ATCC 35150, Salmonella Enteritidis E1347, and Staphylococcus aureus ATCC 29213 were selected. Cryotubes Relative adhesion (%) of these strains, stored at −80◦C in TSB-YE (tryptic soy broth = / × . supplemented with 1% yeast extract; Difco Laboratories, (adhered bacteria initial concentration) 100 Detroit, MI) containing 20% glycerol (Difco Laboratories, (1) Detroit, MI), were thawed and the bacterial cultures were revived by two successive precultures in 10 mL of TSB-YE ◦ 2.5. Scanning Electron Microscopy (SEM) Observation. In (1% v/v) and then incubated for 24 h at 37 C. preliminary work, the two types of noncontaminated sterile The harvested bacteria were washed three times and materials were gold coated and observed under a scanning resuspended in buffers at pH 6.0, 7.0, or 8.0. A total viable electron microscope in order to characterize the microstruc- count was performed for each culture and the total CFUs ture of the substrates. determined using tryptic soy agar (TSA; Difco Laboratories, After two saline rinses, the contaminated coupons were 8 9 Detroit, MI) were between 4 × 10 and 2 × 10 CFU/mL. fixed by immersion in 5 mL of 2.5% glutaraldehyde (v/v) in a 0.1 M sodium cacodylate buffer (pH 7.3) and left at 2.2. Selection of Test Surfaces and Preconditioning Procedures. room temperature for 2 h. The glutaraldehyde was then Polystyrene (hydrophobic) and glass (hydrophilic) substrates removed using a Pasteur pipette and the substrates were were selected for the adhesion tests. Polystyrene weighing rinsed four times by immersing them for 15 min in 0.1 M dishes (no. 25433; VWR International Inc., West Chester, sodium cacodylate buffer (pH 7.3). PA) were used to obtain 5-cm2 coupons and 5-cm2 glass Dehydration was performed through an ascending series coupons were cut from microscope slides (no. 48300; VWR of ethanol (approximately 5 mL) concentrations (30%, 50%, International Inc., West Chester, PA). Prior to physicochem- 70%, and 80%) for 15 min for each concentration, and then ical characterizations and adhesion tests, these substrates three times for 15 min in 100% ethanol. The duplicates were ◦ were soaked for 24 h in sodium hydroxide (1 N), washed preserved in 70% ethanol and stored at 4 C. and rinsed thoroughly eight times with deionized water Dehydration was completed using CO2 in a critical point (Millipore, Billerica, MA). The polystyrene coupons were dryer (Model E3000 CPD, Bio-Rad, Polaron Equipment Ltd., sterilized for 5 min in boiling distilled water, while the glass Watford Hertfordshire, England). The samples were then coupons were directly autoclaved at 121◦Cfor15minin mounted on an aluminum platform and covering with 8 nm bioreactors. of gold using a sputter coater (Cressington 108, Kurt J. Lesker Co., Clairton, PA). The substrates were observed under a scanning electron microscope (Hitachi S300N, Hitachi, 2.3. Surface Contamination. The attachment tests were con- Tokyo, Japan). ducted in sterile bioreactors (BST Model SC60 Suspend Reactor, BioSurface Technologies Corporation, Bozeman, MT). Using a sterile clamp and under a microbiological 2.6. Determination of the Physicochemical Properties of the hood, the sterile coupons were mounted on metal rods Solid and Bacteria Surfaces. For the inert material, sterile (six rods per bioreactor) in pairs, separated by a stainless dried substrates were positioned on a microscope stage steel nut. To ensure a sufficient attachment of the bacterial for contact angle measurements. A microsyringe (Chro- cells, the cultures of the pathogens in the bioreactors were matographic Specialities Inc., Brockville, Ontario, Canada) left in contact with the inert surfaces for 24 h at ambient was used to place a drop (1 μL) of each wetting agent— temperature (20 ± 2◦C) under low agitation (90 rpm). Each bidistilled water (Barnstead Fistreem GlassStill, England), experiment (bacterium-culture pH-surface type combina- formamide, and α-bromonaphthalene (Aldrich Chemical tion) was repeated three times and the means were used for Co., Inc., Milwaukee, WI)—on the solid surfaces. The the statistical analyses. contact angles formed by these liquids were determined International Journal of Microbiology 3 using a goniometer coupled with a 100 × telescope (Gaertner of the pathogens were evaluated in term of free energy of Scientific Corp., Chicago, Ill.). aggregation using the van der Mei et al. [21] approach (4) In each case, six measurements were taken with each   2 liquid on each substrate to determine the surface charge of ΔGmwm =−4 γmLW − γwLW the substrates [19] as well as their hydrophilic-hydrophobic characteristics. The total surface energy (γTOT) of the solid     − − substrates, the Lifshitz-van der Waals contribution to surface − 4 γm+ × γm + γw+ × γw (4) LW AB energies (γ ) and the Lewis acid-base bonds (γ )were     calculated using an extension of the Young-Dupreequation´ − − − γm+ × γw − γm × γw+ , [20]:

LW − + LW − + where (γm , γm, γm)and(γw , γm, γm) are, respectively, the      Lifshitz-van der Waals dispersion, electron donor (base) and T LS LW + − (1+cosθ) × γL = 2 γS × γL + γS × γL electron acceptor (acid) components of the microorganisms studied and those of the water. A preference for the aqueous

  medium, a characteristic of hydrophilic cell surfaces, is + − (2) + γL × γS , demonstrated by a ΔGmwm value > 0, while hydrophobic organisms, which tend to agglomerate in aqueous suspen-   TOT LW − LW AB Gmwm < . γ = γs × 2 γs+ × γs = γs × γs , sions, are characterized by a Δ 0

2.7. Statistical Analyses. The data were analysed using where θ represents the angle formed by the wetting liquid the GLM procedure of SAS software Version 8.0 (SAS T Institute Inc., Cary, NC). An analysis of variance in on the substrate and γL , the surface tension of the wetting LW − + LW − + the form of a factorial experiment (bacterium ∗ culture liquid. (γL , γL , γL )and(γS , γS , γS ) are, respectively, the ∗ Lifshitz-van der Waals dispersion, electron donor (basic) and pH surface type) in complete blocks was chosen for electron acceptor (acid) components of the wetting liquid adhesion tests. When the pH-pathogen interaction terms and inert surface. were significant, a one-way analysis of variance (culture Once these parameters have been determined, for both pH) was performed for each of the four bacteria. The ff the solid substrates (s) and for each bacterium (b)inan significant di erences for the free energies of adhesion aqueous medium (l), the free energy of microbial adhesion, were detected using a one-way analysis of variance. The ΔG (41), was estimated according to the following for- Duncan multiple ranges test was also used to separate adh P = . mula: means. A confidence level of 05 was chosen during analyses.

  2   2 LW LW LW LW 3. Results ΔGadh = γs − γb − γb − γl 3.1. Surfaces Characterization. Photomicrographs of uncon-   2 − γLW − γLW taminated polystyrene and glass obtained using SEM are s l shown in Figure 1. The substrates microstructure revealed     that glass was a smooth surface (Figure 1(a)), while + − − − +2 γl γs + γb − γl (3) polystyrene had irregularities, with tiny bumps and hollows (Figure 1(b)).     − + − The results obtained for the surface physicochemical + γ γs+ + γ − γ b b l properties of the two substrates, determined from contact

    angle measurement with three solvents, are detailed in + − − + − γs × γb − γs × γb . Table 1. The surfaces were characterized by a similar Lifshitz- van der Waals (γLW ) dispersion component and by very low values for the electron acceptor parameter (γ+). The electron donor (γ−) and Lewis acid base capacities (γAB) In theory, the energy balance is favourable to bacterial showed considerable variability from one type of material TOT adhesion if ΔGadh < 0 and unfavourable if ΔGadh > 0. to the other. The total surface energy (γ )measured Bacteria physicochemical properties were determined for for the polystyrene was lower than for glass. Water drops bacterial cultures grown at pHs 6, 7, and 8. The contact angle spread over the glass surface, demonstrating its hydrophilic measurements were performed on bacterial lawns deposited character, consistent with the high value of the electron on filter paper as described by Mafu et al. [19] and the total donor parameter. Polystyrene was characterized by extremely surface energy (γTOT) of the bacteria and the Lifshitz-van weak electrical properties. In addition, the angles formed by LW der Waals (γS ) contribution to surface energies as well as the water on the specimen were very obtuse compared with AB the Lewis acid-base bonds (γS ) were determined with the those obtained on glass, enabling us to conclude that the help of Young-Dupre´ relationship (2). The surface properties surface is hydrophobic. 4 International Journal of Microbiology

SE 14.09 WD 9.1 mm 5.00 kV x4.5 k 10 μm SE 16.01 WD 9.5 mm 5.00 kV x2.5 k 20 μm

(a) (b)

Figure 1: Microstructure of noncontaminated glass (a) and polystyrene (b) substrates observed under a scanning electron microscope.

40 The effect of each parameter therefore had to be studied independently. The type of substrate did not influence the 35 rate of attachment of pathogens (P>.05), however the ) 2

− effect of pH of the culture varied depending on the type of m · 30 microorganism considered.

(mJ E. coli displayed the lowest relative adhesion, regardless 25 of environmental pH (Table 3). The affinity of E. coli for the mwm

G solid samples was significantly influenced by pH (P<.05). Δ 20 A. hydrophila on solid surfaces exhibited the same profile as E. coli (Table 3), however, Aeromonas cell had greater 15 affinity for the two substrates than the enterohemorrhagic 45 48 51 54 57 60 bacterium, especially in acid and neutral media, with the Relative adhesion (%) number of attached cells ranging from 45% to 53% (P>.05) Figure 2: Relationship between the free energy of aggregation and from 39% to 42% (P>.05), respectively. The adhesion (Gmwm) and the mean rate of adhesion of Staphylococcus aureus to capacity of A. hydrophila to inert surfaces was lower after the surfaces studied (r2 = 0.94). being cultivated in an alkaline medium (P<.05). For E. coli and A. hydrophila, neutral pH was the best condition for attachment of these two pathogens, regardless of the type of 3.2. Prediction of Pathogen Adhesion to Both Surfaces. The surface. Type of substrate and pH had no effect on the adsorption free energy of adhesion of the pathogens on bare surfaces, P>. . determined by combining the surface characteristics of the of S. Enteritidis to surfaces ( 05) Thedegreeof selected substrates (Table 1) and the surface properties of the contamination of polystyrene and glass by this enterobacteria bacteria (Table 4), are shown in Table 2.TheΔG values was considerably higher than that observed for the other adh bacteria at pH 8 (P<.05). At pH 6 and 7, S. Enteritidis had of polystyrene were negative for all microorganism—culture P>. . pH—surface combinations, predicting that adhesion would been attached to the polymer and glass equally ( 05) be thermodynamically favourable for the polymer. This Also, after the contamination period, adhesion of S. aureus to the two types of coupons (Table 3)waslower approach predicts that conditions would be unfavourable to P<. . bacterial attachment in the presence of glass, a hydrophilic compared to the other experimental conditions ( 05) surface, since the free energy of adhesion was positive in all The number of organisms attached to both substrates cases. Theoretically, A. hydrophila and E. coli should exhibit decreased by elevating pH. At pH 8, the number of detached P<. . S. aureus cellswasidenticalonbothpolystyreneandglass weaker adsorption on both substrates at pH 6 ( 05) For P>. . S. aureus, the rate of adhered bacteria should be highest in the surfaces ( 05) case of hydrophobic surface at pH 7 (P<.05) and lowest in P<. . the case of hydrophilic surface at pH 8 ( 05) For the two 3.4. Influence of the Surface Physicochemical Properties of substrates studied, pH of the culture affected the theoretical ffi P>. . Pathogenic Bacteria on Their Rate of Adhesion. Figure 2 a nity of S. Enteritidis ( 05) demonstrates the mean relative adhesion rate of each bacterium as a function of the component of their sur- 3.3. Influence of the Inert Surface and pH of the Culture on faces. In fact, since the solid surfaces had no effect on the Adhesion of Pathogenic Bacteria. At pH 6, 7, and 8, no the concentration of adhered cells (P>.05), the mean interactions were determined between pH of culture and between the relative adhesion to polystyrene and to glass adhesion capability for individual bacterium (P>.05). could be calculated in order to obtain an overall adhesion International Journal of Microbiology 5

SE 15.38 WD 9.6 mm 5.00 kV x2.5 k 20 μm SE 13.48 WD 9.0 mm 5.00 kV x2.5 k 20 μm

(a) (b)

SE 10-oct-03 000000 WD 10.3 mm 5.00 kV x2.5 k 20 μm SE 10-oct-03 000000 WD 9.8 mm 5.00 kV x2.5 k 20 μm

(c) (d)

SE 10-oct-03 000000 WD 9.9 mm 5.00 kV x2.5 k 20 μm SE 10-oct-03 000000 WD 9.8 mm 5.00 kV x2.5 k 20 μm

(e) (f)

Figure 3: Attachment of Salmonella enteritidis to polystyrene ((a), (c), and (e)) and to glass ((b), (d), and (f)) at pH 6 ((a) and (b)), pH 7 ((c) and (d)) and pH 8 ((e) and (f)).

Table 1: Contact angles (◦) and surface energies (mJ · m−2) of solid surfaces.

◦ 1 ◦ 1 ◦ 1 LW + − AB TOT Surface θw( ) θF ( ) θα−B( ) γS γS γS γS γ Polystyrene 95.5 ± 1.2 59.3 ± 2.4 9.8 ± 1.3 43.8 0.2 0.0 0.0 43.8 Glass 13.6 ± 1.7 12.9 ± 2.1 17.0 ± 0.9 42.5 0.8 53.4 12.8 55.3

1θw, θF ,andθα−B are on average the angles formed by water, formamide, and α-bromonaphthalene, respectively, LW + − B TOT where γS , γS , γS , γS ,andγ are, respectively, the Lifshitz-van der Waals dispersion component, electron acceptor (acid) and electron donor (basic) parameters, Lewis acid-base bonds, and the total surface energy of the solid substrates. rate. Only an high correlation coefficient (r2 = 0.94) 3.5. Scanning Electron Microscopy Observations. On both between the free energy of aggregation of S. aureus and its types of surfaces, porous and nonporous, a very low number mean rate of adhesion to the substrates has been found of E. coli and dispersed cells of A. hydrophila were observed (Figure 2). The greater the hydrophilic character displayed (data not shown). That is why no photomicrographs of by this bacterium, the lower its tendency to attach to inert these microorganisms have been provided. S. Enteritidis cells surfaces. were generally isolated or formed small clusters, regardless 6 International Journal of Microbiology

SE 16.27 WD 9.6 mm 5.00 kV x2.5 k 20 μm SE 11.40 WD 9.1 mm 5.00 kV x2.5 k 20 μm

(a) (b)

μ SE 2-oct-03 000000 WD 9.4 mm 5.00 kV x2.5 k 20 μm SE 10-oct-03 000000 WD 9.8 mm 5.00 kV x2.5 k 20 m

(c) (d)

SE 15.12 WD 9.4 mm 5.00 kV x2.5 k 20 μm SE 13.11 WD 9.1 mm 5.00 kV x5.0 k 10 μm

(e) (f)

Figure 4: Attachment (at different pH levels) of Staphylococcus aureus to polystyrene ((a), (c), and (e)) and to glass ((b), (d), and (f)) at pH 6 ((a) and (b)), pH 7 ((c) and (d)), and pH 8 ((e) and (f)). of the adhesion conditions (Figure 3). Their dispersion was culture pH, and type of surface, were analysed. Polystyrene more uniform at pH 6 than at the other pH values (Figures and glass were chosen as surfaces, respectively, hydrophobic 3(a) and 3(b)). S. aureus invaded the abiotic surfaces in and hydrophilic, for the attachment and biofilm formation the form of single cells, small chains, or cell aggregates tests. (Figure 4). It was more difficult to observe staphylococci on Scanning electron microscopy observation (Figures 3 polystyreneandglassatpH8(Figures4(e) and 4(f)). As and 4) showed that all the pathogens had the capacity to demonstrated by the present results, it becomes clear that adhere to both types of surface. Although the cavities and in all case, no visible extracellular material was observed on distortions of the samples observed under the scanning both food contact surfaces. electron microscope increase the specific contact area for bacteria, the differences in surface irregularities between 4. Discussion both solid surfaces had no observable impact on the relative adhesion of the microorganisms. These results are consistent Bacterial adhesion to surfaces and biofilm formation are with the findings of Mafu et al. [22], who found no complex phenomena influenced by a number of factors. In correlation between the microstructure of the materials and this study, three of these factors, namely, microbial strains, the capacity of bacteria to adhere to surfaces. Also in this International Journal of Microbiology 7

Table 2: Free energy of adhesion (mJ · m−2) of the four pathogens in contact with the inert surfaces for 24 hours. After this to polystyrene (PS) and to glass (GS) as a function of culture pH. contact time, the adhesion stage tends to be irreversible. If the exposure time had been shorter (maximum of four hours, G 1 G 1 Bacterium pH Δ adh PS Δ adh GS which corresponds to irreversible bacterial adsorption), the 6 −35.9 ± 1.4b 18.0 ± 0.2b predictions for the adhesion, based on ΔGadh, might have Aeromonas hydrophila 7 −37.7 ± 1.0b 18.2 ± 0.3b proven valid. In fact, the equation used to predict bacterial − . ± . b . ± . b 8 38 0 0 1 18 0 0 1 adhesion to surfaces (ΔGadh) does not take into consideration 6 −38.8 ± 1.3a 17.3 ± 0.3b the exopolysaccharides that might be produced for the Escherichia coli O157:H7 7 −39.7 ± 0.4a 17.3 ± 0.0b irreversible attachment of microorganisms. 8 −39.0 ± 1.1a 17.3 ± 0.1b In another attempt to explain adhesion detection with G > 6 −40.9 ± 0.5a 17.1 ± 0.4a Δ adh values 0, we endeavoured to determine whether the surface properties of the infectious agents influenced their Salmonella Enteritidis 7 −42.8 ± 0.8a 16.9 ± 0.0a adhesion rate. Thus, the free energy of aggregation of S. 8 −42.8 ± 1.4a 16.9 ± 0.0a aureus could be correlated to the degree of attachment of the − . ± . a . ± . c 6 39 6 0 4 16 1 0 8 cocci (Figure 2). Gilbert et al. [29] also demonstrated that Staphylococcus aureus 7 −43.2 ± 1.1b 17.0 ± 0.1bc the adhesion of Staphylococcus epidermidis to glass, unlike 8 −41.4 ± 0.4ab 18.3 ± 0.0a Escherichia coli, was negatively correlated to the microorgan- 1 ΔGadh PS and ΔGadh GS are the free energies of adhesion of the bacteria to ism’s hydrophilicity and surface electronegativity. However, polystyrene and glass, respectively. the physicochemical characteristics of the inert surfaces and a–c In a column, for each given bacterium, values with the same letter are not of the other three pathogens examined throughout this study significantly different (P>.05). did not shed any light on their adhesion capacity. The explanation may lie in the composition of the suspension medium of the microorganisms, which was not taken into Table 3: Effect of culture pH on the adhesion of pathogenic bacteria account in the estimation of PS and glass free energy. to polystyrene and glass. Indeed, conditioning the substrates with compounds Relative adhesion (%) from the suspension medium can increase or decrease subsequent bacterial adhesion [23, 30, 31]. These ionic Organism pH Solid surfaces substances located at the microorganism-surface interface Polystyrene Glass can change ionic strength which is a critical factor in . . . ± . 6488 ± 0 4508 2 6 adhesion. Some of these elements can reduce the repulsive A. hydrophila 750.6 ± 1.954.2 ± 3.5 force between the bacterium and the solid surfaces [32, 33], 845.0 ± 5.144.6 ± 4.3 due to electrostatic interactions, by masking charges of the 639.4 ± 2.040.7 ± 0.6 same sign. Thus, several stress responses are associated with E. coli O157:H7 742.1 ± 2.442.1 ± 0.5 the appearance of macromolecular agents in the medium. 839.3 ± 2.139.5 ± 1.1 Many microorganisms exposed to acidic and alkaline stress 654.6 ± 5.355.3 ± 2.5 synthesize polysaccharides, peptides, or heat-stable proteins [10, 34], which play an active role in induction of tolerance S. enteritidis 756.7 ± 3.455.1 ± 3.0 . ± . . ± . to pH stress. It should also be noted that some bacteria 86211 2 35431 3 8 produce biosurfactants that inhibit their attachment. For . ± . . ± . 6586 3 2556 3 5 instance, synthesis by S. aureus of a surfactant, the toxin δ, S. aureus 754.4 ± 5.651.1 ± 8.3 inhibits the action of D-alanine [12], limiting cell adhesion 846.61 ± 4.246.71 ± 1.8 to polystyrene [35]. Previous studies have also demonstrated the important roles played by cell organelles and bacterial mobility in transport and adhesion to various types of surfaces. Pseu- study, the attachment of solid surfaces by microorganisms domonas fluorescens strains without flagella exhibited a weak was influenced solely by culture pH, which concurs with the capacity to attach to surfaces and develop a biofilm [7]. work of Husmark and Ronner [23]aswellasthatofHerald Flagella apparently play an essential role during the initial and Zottola [24]. The absence of fimbriae and curli [25, 26] reversible stages of attachment by overcoming the repulsive might explain the low contamination of E. coli observed in forces. However, the nonmotile bacterium S. aureus adhered this study as well as in other experiments conducted on the to the tested substrates at pH 6 in greater numbers than contamination of polymers by serotype O157 [24, 27, 28]. E. coli O157:H7 and A. hydrophila (Table 3), which are The data from these experiments are not in agreement motile microorganisms. The presence of flagella is apparently with the predictions deduced from the free energies of adhe- more important when the velocity of the environmental sion (ΔGadh) of pathogens on polystyrene (Table 2). These liquid is high [4, 36].McClaineandFord[37] demonstrated energy characteristics demonstrated that conditions for that rotation of the flagellum caused detachment of fixed adsorption of all microorganisms were thermodynamically motile bacteria when flow rates were low (0.02 mL·min−1), favourable on polystyrene (ΔGadh < 0) and unfavourable on while their adhesion was strengthened at high flow rates −1 glass (ΔGadh > 0) surfaces. The bacterial cultures remained (2 mL·min ). 8 International Journal of Microbiology

Table 4: Surface tension and hydrophobicity (mJ·m−2) of the four pathogens determined from contact angle measurement.

γLW γ+ γ− γAB γTOT G1 Organism pH S S S S Δ mwm 635.3 ± 1.0a 1.9 ± 0.2b 53.0 ± 0.4a 20.0 ± 1.0b 55.3 ± 0.1a 26.3 ± 0.2b A. hydrophila 735.4 ± 0.7a 1.5 ± 0.1b 54.1 ± 0.3a 18.2 ± 0.8b 53.6 ± 0.1a 28.7 ± 0.4a 835.8 ± 0.4a 1.7 ± 0.0b 53.6 ± 0.0a 18.1 ± 0.0b 53.9 ± 0.4a 27.8 ± 0.3ab 637.4 ± 0.1bc 1.5 ± 0.2a 52.3 ± 1.3a 17.9 ± 1.3a 55.3 ± 1.2a 24.9 ± 0.8c E. coli 0157:H7 738.5 ± 0.1a 1.4 ± 0.1a 54.0 ± 0.1a 17.3 ± 0.3a 55.8 ± 0.3a 26.2 ± 0.0b 838.2 ± 0.6ab 1.6 ± 0.2a 53.5 ± 0.4a 17.9 ± 1.0a 56.0 ± 0.3a 25, 6 ± 0.4bc 636.1 ± 0.7d 1.7 ± 0.0a 49.8 ± 0.8b 18.3 ± 0.3a 54.4 ± 0.5a 23.0 ± 1.4a S. enteritidis 738.1 ± 0.0ab 1.0 ± 0.1a 51.1 ± 0.2ab 14.0 ± 0.7a 52.1 ± 0.7a 25.1 ± 0.7a 838.6 ± 0.3a 1.0 ± 0.2a 52.0 ± 0.3a 14.2 ± 1.3a 52.8 ± 1.0a 25.6 ± 0.9a 637.5 ± 0.3b 1.6 ± 0.1a 46.0 ± 3.5b 17.0 ± 0.3a 54.5 ± 0.0a 17.8 ± 4.4c S. aureus 739.7 ± 0.1a 0.9 ± 0.2b 54.6 ± 1.3a 14.2 ± 1.1b 53.8 ± 1.0a 27.7 ± 2.1ab 836.5 ± 0.1b 1.0 ± 0.1b 54.7 ± 0.5a 15.0 ± 0.4b 51.6 ± 0.3b 33.4 ± 0.8a 1 LW + − B TOT Free energy of aggregation of the microorganisms in water where γS , γS , γS , γS ,andγ are, respectively, the Lifshitz-van der Waals contribution energies, electron acceptor (acid), and electron donor (basic) components of the wetting agent and bacterial lawn, Lewis acid-base bonds and the total surface energy of the bacteria. a–d In a column, for a given bacterium, values with the same letter are not significantly different (P>.05).

Although the irreversible attachment of microorganisms (between two production cycles, for example), since they is often associated with the production of exopolysaccha- can effectively detach and eliminate the adhered bacteria rides, no extracellular matrix was detected by scanning when they have not yet formed a biofilm, thereby preventing electron microscopy, regardless of experimental conditions. contamination of raw materials and finished products and This indicates that the presence of polymers is not necessary protecting consumer health. for permanent adhesion of microorganisms to a surface. The visible filaments attached to the substrates may be dependent on the suspension medium. Maximum polysaccharide pro- Acknowledgments duction is generally associated with an environment deficient The authors would like to thank Claude Laberge for the in essential nutrients, such as carbon, nitrogen, calcium, and statistical analyses, Marie-Josee´ Lemay, Marie-Josee´ Gagne,´ iron [26, 38]. Contact time and environmental temperature Avila Croisetiere,` Franc¸ois Saint-Germain, and Ali Taherian are other factors that influence the formation of a three- for their technical assistance and scientific advices. dimensional network [39, 40]. However, these appendices are crucial for biofilm consolidation and their resistance to environmental stresses [41]. References Several molecules from the culture media can contribute [1] G. I. Loeb and R. A. Neihof, “Marine conditioning films,” to produce a conditioning film at the surface of the substrates in Applied Chemistry at Protein Interfaces,R.E.Baier,Ed., [42]. These molecules include proteins, which are block vol. 145 of Advances Chemical Series, pp. 319–335, American copolymers presenting hydrophobic as well as hydrophilic Chemical Society, 1975. sections. It is thus possible for such molecules to adhere [2] B. Carpentier and O. Cerf, “Biofilms and their consequences, to hydrophilic substrates such as glass, resulting in an with particular reference to hygiene in the food industry,” hydrophobization of the surface. This phenomena would Journal of Applied Bacteriology, vol. 75, no. 6, pp. 499–511, lead bacteria presenting the same adhesion trends on both 1993. PS and hydrophobically coated glass. [3] G. A. McFeters, M. J. Bazin, J. D. Bryers et al., “Biofilm development and its consequences,” in Microbial Adhesion and Aggregation, K. C. Marshall, Ed., pp. 109–124, Springer, Berlin, 5. Conclusion Germany, 1984. [4] R. F. Mueller, “Bacterial transport and colonization in low The results of this work indicate that pathogens could adapt nutrient environments,” Water Research, vol. 30, no. 11, pp. to various pH levels of cultural media and adhere indif- 2681–2690, 1996. ferently to inert polystyrene and glass surfaces, regardless [5] C. Faille, C. Jullien, F. Fontaine, M.-N. Bellon-Fontaine, C. of their hydrophilic or hydrophobic nature which leads to Slomianny, and T. Benezech, “Adhesion of Bacillus spores and Escherichia coli cells to inert surfaces: Role of surface biofilm formation and increases the possibility of resis- hydrophobicity,” Canadian Journal of Microbiology, vol. 48, no. tance to sanitizing agents. Therefore, to minimize surfaces 8, pp. 728–738, 2002. contaminations, one should be aware of the influence of [6]C.Jullien,T.Ben´ ezech,B.Carpentier,V.Lebret,andC.´ different environmental conditions which affects survival of Faille, “Identification of surface characteristics relevant to the bacteria in order to reduce their incidence in food systems. hygienic status of stainless steel for the food industry,” Journal The operations must be carried out promptly and regularly of Food Engineering, vol. 56, no. 1, pp. 77–87, 2003. International Journal of Microbiology 9

[7]D.R.Korber,J.R.Lawrence,B.Sutton,andD.E.Caldwell, and rubber surfaces after short contact times,” Journal of Food “Effect of laminar flow velocity on the kinetics of surface Protection, vol. 53, no. 9, pp. 742–746, 1990. recolonization by Mot+ and Mot-Pseudomonas fluorescens,” [23] U. Husmark and U. Ronner, “Forces involved in adhesion Microbial Ecology, vol. 18, no. 1, pp. 1–19, 1989. of Bacillus cereus spores to solid surfaces under different [8] J-C. Cartonne´ and R. Tournier, “Corrosion et anticorrosion environmental conditions,” Journal of Applied Bacteriology, des materiaux´ metalliques´ utilises´ dans les equipements´ et les vol. 69, no. 4, pp. 557–562, 1990. ateliers bio-industriels,” in Nettoyage, D´esinfection et Hygi`ene [24] P. Herald and E. A Zottola, “Attachment of Listeria monocyto- dans les Bio-Industries, J.-Y. Leveau, M. Bouix, J.-Y. Leveau, genes to stainless steel surfaces at various temperatures and pH and M. Bouix, Eds., p. 84, Tec & Doc, Paris, France, 1999. values,” Journal of Food Science, vol. 53, no. 5, pp. 1549–1552, [9] M. Fletcher, “The effects of culture concentration and 1988. age, time, and temperature on bacterial attachment to [25] A. L. Cookson, W. A. Cooley, and M. J. Woodward, “The polystyrene,” Canadian Journal of Microbiology,vol.23,no.1, role of type 1 and curli fimbriae of Shiga toxin-producing pp. 1–6, 1977. Escherichia coli in adherence to abiotic surfaces,” International [10] J. W. Foster and H. K. Hall, “Adaptive acidification tolerance Journal of Medical Microbiology, vol. 292, no. 3-4, pp. 195–205, response of Salmonella typhimurium,” Journal of Bacteriology, 2002. vol. 172, no. 2, pp. 771–778, 1990. [26] A. C. L. Wong, “Influence of culture conditions on biofilm [11] I. Giovannacci, G. Ermel, G. Salvat, J. L. Vendeuvre, and formation by Escherichia coli O157:H7,” International Journal M. N. Bellon-Fontaine, “Physicochemical surface properties of Food Microbiology, vol. 26, no. 2, pp. 147–164, 1995. of five Listeria monocytogenes strains from a pork-processing [27]G.Harkes,H.C.vanderMei,P.G.Rouxhet,J.Dankert,H.J. environment in relation to serotypes, genotypes and growth Busscher, and J. Feijen, “Physicochemical characterization of temperature,” Journal of Applied Microbiology,vol.88,no.6, Escherichia coli—a comparison with gram-positive bacteria,” pp. 992–1000, 2000. Cell Biochemistry and Biophysics, vol. 20, no. 1, pp. 17–32, [12] M. Gross, S. E. Cramton, F. Gotz,¨ and A. Peschel, “Key role of 1992. teichoic acid net charge in Staphylococcus aureus colonization [28] M. A. Assanta, D. Roy, M.-J. Lemay, and D. Montpetit, of artificial surfaces,” Infection and Immunity,vol.69,no.5, “Evidence for Escherichia coli O157:H7 attachment to water pp. 3423–3426, 2001. distribution pipe materials by scanning electron microscopy,” [13] H. Anwar, M. K. Dasgupta, and J. W. Costerton, “Testing the Journal of Food Protection, vol. 65, no. 12, pp. 1970–1975, 2002. susceptibility of bacteria in biofilms to antibacterial agents,” [29] P. Gilbert, D. J. Evans, E. Evans, I. G. Duguid, and M. R. W. Antimicrobial Agents and Chemotherapy, vol. 34, no. 11, pp. Brown, “Surface characteristics and adhesion of Escherichia 2043–2046, 1990. coli and Staphylococcus epidermidis,” Journal of Applied Bacte- [14] H. Park, Y.-C. Hung, and C. Kim, “Effectiveness of electrolyzed riology, vol. 71, no. 1, pp. 72–77, 1991. water as a sanitizer for treating different surfaces,” Journal of Food Protection, vol. 65, no. 8, pp. 1276–1280, 2002. [30] B. E. Brooker, “Surface coat transformation and capsule [15] P. Chavant, B. Martinie, T. Meylheuc, M.-N. Bellon-Fontaine, formation by Leuconostoc mesenteroides NCDO 523 in the and M. Hebraud, “Listeria monocytogenes LO28: surface presence of sucrose,” Archives of Microbiology, vol. 111, no. 1-2, physicochemical properties and ability to form biofilms pp. 99–104, 1976. at different temperatures and growth phases,” Applied and [31]S.K.HoodandE.A.Zottola,“Adherencetostainlesssteel Environmental Microbiology, vol. 68, no. 2, pp. 728–737, 2002. by foodborne microorganisms during growth in model food [16] CDC. 2002, “Preliminary foodnet data on the incidence of systems,” International Journal of Food Microbiology, vol. 37, foodborne illnesses—Selected sites, United States,” Morbidity no. 2-3, pp. 145–153, 1997. and Mortality Weekly Report, vol. 51, no. 15, pp. 325–329, [32] M.-N. Bellon-Fontaine, N. Mozes, H. C. van der Mei et al., 2002. “A comparison of thermodynamic approaches to predict the [17] A. A. Mafu, M. Pitre, and S. Sirois, “Real-time PCR as a adhesion of dairy microorganisms to solid substrata,” Cell tool for detection of pathogenic bacteria on contaminated Biophysics, vol. 17, no. 1, pp. 93–106, 1990. food contact surfaces by using a single enrichment medium,” [33] R. Briandet, T. Meylheuc, C. Maher, and M. N. Bellon- Journal of Food Protection, vol. 72, no. 6, pp. 1310–1314, 2009. Fontaine, “Listeria monocytogenes Scott A: Cell surface charge, [18] A. A. Mafu, D. Roy, and K. Machika, “Efficiency of disinfecting hydrophobicity, and electron donor and acceptor charac- agents to destroy Listeria monocytogenes, Yersinia enterocolitica teristics under different environmental growth conditions,” and Staphylococcus aureus on a contaminated surface,” Dairy, Applied and Environmental Microbiology, vol. 65, no. 12, pp. Food and Environmental Sanitation, vol. 16, no. 7, pp. 426–430, 5328–5333, 1999. 1996. [34] R. J. Rowbury, N. H. Hussain, and M. Goodson, “Extracellular [19]A.A.Mafu,D.Roy,J.Goulet,andL.Savoie,“Characterization proteins and other components as obligate intermediates in of physicochemical forces involved in adhesion of Listeria the induction of a range of acid tolerance and sensitisation monocytogenes to surfaces,” Applied and Environmental Micro- responses in Escherichia coli,” FEMS Microbiology Letters, vol. biology, vol. 57, no. 7, pp. 1969–1973, 1991. 166, no. 2, pp. 283–288, 1998. [20] C. J. Van Oss, M. K. Chaudhury, and R. J. Good, “Interfacial [35] C. Vuong, H. L. Saenz, F. Gotz, and M. Otto, “Impact of the Lifshitz-van der Waals and polar interactions in macroscopic agr quorum-sensing system on adherence to polystyrene in systems,” Chemical Reviews, vol. 88, no. 6, pp. 927–941, 1988. Staphylococcus aureus,” Journal of Infectious Diseases, vol. 182, [21] H. C. van der Mei, R. Bos, and H. J. Busscher, “A reference no. 6, pp. 1688–1693, 2000. guide to microbial cell surface hydrophobicity based on [36] D. R. Korber, J. R. Lawrence, and D. E. Caldwell, “Effect of contact angles,” Colloids and Surfaces B, vol. 11, no. 4, pp. 213– motility on surface colonization and reproductive success of 221, 1998. Pseudomonas fluorescens in dual-dilution continuous culture [22]A.A.Mafu,D.Roy,J.Goulet,andP.Magny,“Attachmentof and batch culture systems,” Applied and Environmental Micro- Listeria monocytogenes to stainless steel, glass, polypropylene, biology, vol. 60, no. 5, pp. 1421–1429, 1994. 10 International Journal of Microbiology

[37] J. W. McClaine and R. M. Ford, “Characterizing the adhesion of motile and nonmotile Escherichia coli to a glass surface using a parallel-plate flow chamber,” Biotechnology and Bio- engineering, vol. 78, no. 2, pp. 179–189, 2002. [38] C. M. Brown, D. C. Ellwood, and J. R. Hunter, “Growth of bacteria at surfaces: Influence of nutrient limitation,” FEMS Microbioogy Letters, vol. 1, no. 3, pp. 163–166, 1977. [39] M. F. A. Bal’a, I. D. Jamilah, and D. L. Marshall, “Attachment of Aeromonas hydrophila to stainless steel surfaces,” Dairy Food and Environmental Sanitation, vol. 18, no. 10, pp. 1240–1247, 1998. [40] M. A. Assanta, D. Roy, M.-J. Lemay, and D. Montpetit, “Attach- ment of Arcobacter butzleri, a new waterborne pathogen, to water distribution pipe surfaces,” JournalofFoodProtection, vol. 65, no. 8, pp. 1240–1247, 2002. [41] C.-T. Huang, F. P. Yu, G. A. McFeters, and P. S. Stewart, “Nonuniform spatial patterns of respiratory activity within biofilms during disinfection,” Applied and Environmental Microbiology, vol. 61, no. 6, pp. 2252–2256, 1995. [42] S. Sardin, J.-J. Morrier, G. Benay, and O. Barsotti, “In vitro streptococcal adherence on prosthetic and implant materials. Interactions with physicochemical surface properties,” Journal of Oral Rehabilitation, vol. 31, no. 2, pp. 140–148, 2004. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 312956, 10 pages doi:10.1155/2011/312956

Review Article Assessment of Bacterial Antibiotic Resistance Transfer in the Gut

Susanne Schjørring and Karen A. Krogfelt

Department of Microbiological Surveillance and Research, Statens Serum Institut, Artillerivej 5, 2300 Copenhagen S, Denmark

Correspondence should be addressed to Susanne Schjørring, [email protected]

Received 16 June 2010; Revised 21 November 2010; Accepted 9 December 2010

Academic Editor: Jorge H. Leitao

Copyright © 2011 S. Schjørring and K. A. Krogfelt. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

We assessed horizontal gene transfer between bacteria in the gastrointestinal (GI) tract. During the last decades, the emergence of antibiotic resistant strains and treatment failures of bacterial infections have increased the public awareness of antibiotic usage. The use of broad spectrum antibiotics creates a selective pressure on the bacterial flora, thus increasing the emergence of multiresistant bacteria, which results in a vicious circle of treatments and emergence of new antibiotic resistant bacteria. The human gastrointestinal tract is a massive reservoir of bacteria with a potential for both receiving and transferring antibiotic resistance genes. The increased use of fermented food products and probiotics, as food supplements and health promoting products containing massive amounts of bacteria acting as either donors and/or recipients of antibiotic resistance genes in the human GI tract, also contributes to the emergence of antibiotic resistant strains. This paper deals with the assessment of antibiotic resistance gene transfer occurring in the gut.

1. Emergence of Antibiotic Resistance A few countries within the European Union (EU) have acted on the new research regarding the suspicious use of The introduction of antibiotics after World War I resulted in AGP [6]. These countries were Sweden in 1986, Norway a dramatic decrease of numbers of deaths due to bacterial in 1995, and Denmark in 1998-1999 [7, 8]. Despite a infections. Today, antibiotics have lost their status as the significant decrease in bacterial antibiotic resistance levels “miracle drug” [1, 2] and “treatment failure” is a new in the countries banning growth promoter products, four and often seen situation [1–5]. The increase of antibiotic different AGPs were in use till January 2006, at which point resistance is to be blamed for this medical emergency. the EU commission initiated the ban of all AGP [9]. The sustainability of antibiotic resistance is partly due to Numerous factors influence the development of antibi- selection of already resistant bacteria that become the new otic resistance misuse being the obvious factor. The use of dominant population in the environment. Furthermore, antibiotics is influenced by level of knowledge, expectations, antibiotic usage urges bacteria sensitive to antibiotics to choice of prescription, patient behaviour, economics and become resistant in order to survive. Survival mechanisms health system [10]. Patient-related factors often include include the acquisition of antibiotic resistance genes from inappropriate antibiotic use, like self-medication or inad- other bacteria/phages (horizontal gene transfer or trans- equate doses despite the prescription text. The prevalence duction), mutations in specific genes, and alteration of the of self-medication in Europe was investigated in 2006 [11]. bacterial surface. Thus continuous usage and accumulation It was concluded that the levels of self-medication were of antibiotics in the environment has resulted in the increase higher in Eastern and Southern Europe than in Northern of antibiotic resistant bacteria not only in Europe but and Western Europe. Similarly, Northern countries as well also worldwide. The relationship between antibiotics used as the Netherlands had the lowest frequency of antibiotic as antimicrobial growth promoters (AGPs) in production consumption and the lowest level of resistant bacteria [11, animals and the development of resistant bacteria in food 12]. The prescription system for drugs is also important. products has been related to human food born infections In developing countries, antibiotics can be purchased in with resistant strains. This was not easy to acknowledge. single doses, which increase the risk of the antibiotic 2 International Journal of Microbiology treatment being terminated before clinical success. In some they produce as defence mechanism against other microor- countries, antibiotics can be purchased over the counter and ganisms. Similarities among the genes and resistance mech- prescription is not even necessary, which will also contribute anisms found in the antibiotic producers and in the human to the rate of incomplete treatments and self-medication. pathogenic bacteria suggest that the producer bacteria are Advertising by television, radio, print media, or internet the pool of origin of antibiotic resistance genes [29, 30]. allows pharmaceutical companies to target a specific brand During antibiotic treatment, all bacteria in the human/ toward physicians as well as the general public. All of the animal body are exposed to selective pressure of the above-mentioned factors can contribute to a rise in the resis- antibiotic. Consequently, the GI is highly exposed, especially tance level [11]. However, further development of antibiotic during oral therapy. This results in the selection of naturally resistance might be avoided by acquiring knowledge on the resistant strains carrying an important genetic pool that mechanisms of bacterial antibiotic resistance. Furthermore, might be capable of transferring antibiotic resistance genes regulatory agencies can set up guidelines and measures, to other strains present in the human intestine. Moreover, in order to use antibiotics adequately. For example, some resistant food contaminants that originate from animals countries found that advertising against inappropriate use and are consumed by humans, can also act as a gene pool of antibiotics in national campaigns can reduce the total (donors) of antibiotic resistance genes [31]. amount used due to awareness and proper information of In general, becoming resistance towards antibiotics has the public [13]. During the last decades, there has been an been associated with a biological fitness cost. The cost increased focus on persistent bacterial biofilm formation on weakens the bacteria’s ability to multiply and survive within a medical devices, implants, and environmental biofilms [14, host [32]. The connection between resistance and decreased 15]. Interestingly, it has been shown that biofilms were hot fitness has stimulated the idea that a reduction in the use spot for horizontal gene transfer [16–18]. Thus, promoting of antibiotics would lead to a reduction in the frequency development of antibiotic resistance in bacteria. of resistant bacteria through natural selection [33]. Further- Changes in living standards have resulted in a large, more, cross-resistance is of importance. Both in vitro and in ageing human population, and in increased usage of antibi- vivo studies have illustrated that “compensatory evolution otics. Intensive and long-term hospitalisations due to new can stabilise resistant bacterial populations in the absence advances in medicine often result in new infections (hospital- of antibiotics by making them as fit as susceptible clones” acquired infections) that are expensive to control and [33]. In addition, resistant bacteria can alleviate the cost difficult to eradicate. These occur worldwide due to failures of resistance by acquiring additional fitness-compensatory in simple infection control, such as inadequate hand hygiene mutations [34–36]. The importance of environmental con- and changing of gloves [19]. Increased usage of broad ditions affecting the fitness costs was also shown [37, 38]. spectrum antibiotics in order to avoid treatment failure Mutations that have occurred in clinical isolates are seen to created a vicious circle in the hospitals, as the use of broad compensate for fitness cost in order to stabilise the resistant spectrum antibiotics influenced the level of multiresistant pathogens in the population [29, 39]. Still, this reversibility bacteria and their presence [20]. in clinical settings is expected to be slow or nonexistent [33]. Resistance genes from both Gram-positive and Gram- 2. The Gastrointestinal Tract as negative pathogenic bacteria have revealed almost identical a Reservoir of Antibiotic Resistance Genes: sequences, suggesting that transfer of antibiotic resistance Recipients or Donors genes across genera has occurred. Furthermore, it is sug- gested that transfer events have occurred recently and are The human intestine is a complex ecosystem with a large evolutionary recent events due to high sequence identity species diversity, of at least 400 different bacterial species [29]. It is also suggested that a gene flux occurs in nature from [21, 22]. The density varies through the different parts of Gram-positive cocci, (Enterococci/Streptococci) to Gram- the GI tract from 104 bacteria/mL in the stomach to 1012 negative bacteria [29] with genes coding for streptogramins bacteria/g faeces in the distal part of the colon [23, 24]. being described as examples [29]. By classical culture techniques approximately 5–15% of the species present in the GI tract are detected [25]. New 3. Effect of Antibiotics to the estimates by metagenomic approaches of the bacterial flora Gastrointestinal Flora have suggested that the presence of bacterial species might be as high as 1150 [26]. Despite attempts to stop the antibiotic The physiological effect of the bacterial flora of the GI resistance development, the level of resistant bacteria is on tract is described as “Fermentation of non-digestible dietary the rise and the hypothesis that our GI tract acts as a reservoir residues and endogenous mucus: salvage of energy as short- of antibiotic resistance genes is widely accepted [27]. “Could chain fatty acids, production of vitamin K, absorption of the microflora of the human colon, normally considered ions; control of epithelial cell proliferation and differentia- innocuous or beneficial, be playing a more sinister role in tion; development and homoeostasis of the immune system human health as a reservoir for antibiotic resistance genes?” and protection against pathogens (the barrier effect)” [40]. A was the hypothesis set by Salyers and coworkers [28]. symbiotic relation between the bacteria and the host provides Furthermore, it is known that genes responsible for the host with optimal protection. The host provides nutrients antibiotic resistance are present in microorganisms, provid- to bacteria and bacteria repay by providing a colonisation ing them with self-protection to the antibiotic compounds barrier [40]. International Journal of Microbiology 3

When using antibiotics as infection treatment, a distur- the two strains of interest (donor and recipient strains), bance in the flora of the GI tract is created. The barrier thus it is called a diassociated model. The model is a highly is broken and potentially pathogenic bacteria are allowed controlled setup where only few factors in the GI tract to colonise the intestine [40]. Disturbances of the normal vary. However, a very important factor is that the rodents GI flora are also seen after radical changes of the host diet are coprophages (ingest faeces) which leads to uncontrolled or after radiation treatments [41]. The effect of different reinoculation. This can be decreased by using gratings and antibiotics on the indigenous flora was investigated and it individual cages. The model allows the inoculated bacteria to was found that clindamycin, erythromycin, cefoperazone, colonise the GI tract in very high numbers: 109-1010 CFU/g ceftriaxone, and moxalactam have a pronounced influence faeces. Bacteria which are normally transient in conventional on the flora [41, 42]. The most common complication of animals will now colonise the GI tract in high numbers due antibiotic treatment is excess colonisation of the gut by to lack of competition from the indigenous flora [48]. High Clostridium dificile, resulting in severe and sometimes fatal numbers of the two strains increase the chance of cell-to- enteric diseases [43, 44]. cell contact and, therefore, increase the opportunity of gene Furthermore, antibiotics create a selective pressure on the transfer. Thus, the diassociated model is often described as a intestinal flora, risking overgrowth of resistance strains [40]. worst case scenario model. An investigation of the germ-free This, in turn, increases the threat of antibiotic resistant gene mice has revealed that the lymph nodes, spleen, and Peyer’s transfer among the indigenous flora and, at worst, transfer to patches are relatively inactive, thus the germ-free mice are other pathogenic bacteria. The risk of developing antibiotic immune deficient [49]. resistant strains which can be transmitted by patient-to- Another model is the antibiotic-treated mouse model, patient contact and the spreading of resistance genes can be where the very effective colonising barrier of conventional diminished by choosing antibiotics with a minimum effect mice is disturbed by adding streptomycin to the drinking on the GI flora [41]. water [50]. Some of the species colonising the gut are eliminated allowing the new bacteria to colonise. When 4. Model Systems for Studying the administered orally, antibiotics such as ampicillin, clin- Development of Antibiotic Resistance in the damycin, kanamycin, metronidazole, and streptomycin have Gastrointestinal Tract been shown to be useful in this model [51]. Metronidazole was the least effective antibiotic in decreasing the colonising In order to assess the effect of antibiotics on the gastroin- barrier, whereas clindamycin was active against anaerobic testinal flora, a number of models have been developed, both bacteria, kanamycin, and streptomycin against facultative in in vitro models, where a number of conditions can be anaerobic bacteria, and ampicillin had a broader range [51]. controlled as well as in vivo animal models, resembling the Hentges et al. found that streptomycin clears off most of the human/animal host. Enterobacteriaceae—while the total populations of aerobic In vitro conjugation is conducted in liquid media, on and anaerobic bacteria were not affected [52]. Streptomycin agar plates or on filters placed on agar plates. These are increased the level of pH in the GI tracts, enhanced the often the first experiments by which frequency of transfer viscosity of the mucus, making the intestine more susceptible can be observed, since all the parameters can be controlled, to possible infection by pathogenic bacteria [53, 54]. that is, growth media, temperature, conjugation time, A third model is the human microbiota-associated selective pressure, and so forth. In vitro systems trying to rodent (HMA) model where germ-free animals are inocu- mimic the GI tract have been successfully used, such as lated with human faecal bacteria, thus mimicking the human batch fermentors and continuous flow fermentors, to study flora composition in the rodent gut [55]. However, it was the effects of pH on the degradation of nutrients, CO2 shown that changes in the dominant species occasionally production among others [45]. occur following inoculation [56]. The shift in dominant Different in vivo models were used to confirm in vitro species composition may be influenced by the genetic results, and to mimic the complexity of the host environ- differences between humans and mice, but also by the feed ment. These included alfalfa sprouts as a plant model, and intake. The nutrients present in the GI tract can promote a rumen model, and animal models. [46]. Most experiments some bacteria and inhibit others [50]. The dominating strain related to pathogenic bacteria and their frequency of in vivo in the human sample might not be the same as the one gene transfer have been conducted in rodents (especially colonising the animals in high numbers [55, 56]. mice). Rodents are often preferred due to high reproduction, All GI tract animal models have the advantages of short gestation, a minimum of husbandry, and low costs. The small differences in the bacterial flora composition within histological structure of the GI tract has similarities to the the animals due to husbandry standards compared to the human gut, including epithelial layers and mucous secretion variance within humans and their individual flora [57]. [47]. Effect of bacterial colonisation of the gut, including One of the most important points that is worth to discuss gene transfer, expression of virulence characteristics, and concerning the assessment between results obtained from spatial distribution have therefore been investigated in mice. in vitro studies, in vivo diassociated/conventional animal Since the gut has natural resistance (barrier effect) to studies, and extrapolating to humans. The choice of model colonization by foreign bacterial strains, the experiments influences the results and should be addressed carefully. are often performed in germ-free and/or antibiotic-treated In vitro and in vivo transfer experiments are not com- rodents. Germ-free animals are usually inoculated only with parable and do not always provide the same results and 4 International Journal of Microbiology

“frequencies”, similarly different types of in vivo transfer scenario is that antibiotic resistance genes are transferred to experiments are not always comparable. The colonising a pathogenic bacterium which could then lead to treatment barrier and the complexity of the flora in the antibiotic- failure of an infection. Transfer from LAB strains has been treated mice compared to diassociated rats lacking the documented in vitro [62–66], but very few studies have colonising barrier (as they were germ free before bacterial confirmed this antibiotic resistance transfer in vivo and it inoculation) can be considered to be a too simple model, was only observed in the diassociated animal model [62, 64, as it does not depict the complexity of the factors involved 66, 67]. Studying the board-host-range conjugative plasmid in transfer of antibiotic resistance genes in the GI tract. pAMβ1, transfer was observed in vitro from different donors, Transfer in diassociated animals only provides the knowledge for example, Lactobacillus spp. (L. plantarum, L. reuteri, that transfer can occur during colonisation of the GI tract L. fermentum, and L. murinus) to other LAB strains [68– during less realistic circumstances with no indigenous flora 72]. In the diassociated model pAMβ1 has been transferred present and consequently unusually high CFU levels of both from L. reuteri to Enterococcus faecalis [73]andamong the donor and recipient strains. Differences in biochemistry, Lactococcus lactis strains [66], but the transfer among the L. physiology, and immunology have been observed when lactis strain could not be observed in conventional rats [66]. comparing conventional and germ-free animals [58, 59]. The Gram-positive bacteria seem to have a limited influence on influence of these factors on the bacteria interactions and the spread of antibiotic resistance in the GI tract, despite indirectly on the transfer frequency is still unknown. Using the fact that a few multiresistant LAB strains have been the antibiotic-treated mice model with the conventional detected in weaning piglets [74]. Nevertheless, conclusive bacterial flora barrier and an intact immune system has the documentation of transfer in the GI tract from LAB strains is advantage of mimicking the human GI tract better. However, lacking and therefore more studies need to be carried out. a downside of using a model with a complex flora is that gene transfer might occur to other recipient strains than the spe- 6. Antibiotic Resistance Transfer to the cific strain which is being investigated, or that the number of Indigenous Flora in the Gut transconjugants is below the detection limit (10 CFU/g fae- ces). The question to be discussed is what is most important The hypothesis that the indigenous flora can become a in the development of antibiotic resistance—“one” transfer reservoir of antibiotic resistance genes remains to this date event or the level of transconjugants and their spread. neither confirmed nor denied. It is suggested that the Gram- The level of transconjugants as well as their persistence negative part of the flora has an increased prospect to obtain among the indigenous flora in the GI tract reflects the antibiotic resistance genes and might act as a reservoir and influence of the spread and therefore, the antibiotic-treated transfer the resistance gene further to pathogenic bacteria, model seems to give a reliable answer. Nevertheless, transfer which might lead to infections with limited treatment in vivo cannot be calculated from extrapolations of in vitro possibilities [75]. This is especially true within a hospital experiments, and humans and animals are different both in setting, where the antibiotics used are often aimed at the flora composition and metabolism [47]. Gram-negative pathogens, adding selective pressure on the Gram-negative bacteria. The human microbiota associated 5. Lactic Acid Bacteria: Intermediate Hosts of rat model revealed that broad-host range plasmid pAMβ1 Antibiotic Resistance Genes? can be transferred from L. lactis to indigenous Enterococcus spp., whereas no transfer was observed to Lactobacillus spp., One quarter of all food production is estimated to involve Bifidobacterium spp., and Enterobacteriaceae spp. [76]. A microbial fermentation processes by using lactic acid bacte- study using voluntary human subjects in 1974 showed that rial (LAB) strains [60], for example, sausage, ham, cheese, antibiotic resistance transfer occurred between an Escherichia and dairy products. In addition, probiotics have become coli of animal origin and indigenous E. coli in the GI available on the market, containing a single strain or a tract during tetracycline treatment. Transfer was observed combination of strains. In order to obtain health benefits, late in the experiment (36 days after the end of the treat- live bacteria around 1011 CFU/mL should be ingested daily ment), but only with the therapeutic levels of tetracycline [61]. The proposed problem is that probiotic strains and (1000 mg/daily) administrated and not with the low level of starter cultures might contain naturally occurring antibiotic 50 mg/day [77]. The observation of “late transfer” has also resistance genes. From a point of safety, it is necessary been observed in a study using mice, where the indigenous to distinguish between intrinsic and acquired resistance E. coli transconjugant was observed in high numbers 23 days genes and most importantly the transferability of these. The after inoculation of the multiresistant Klebsiella pneumoniae increasing level of consumption of fermented foods (more donor [75]. In addition, vancomycin resistance has been and more products use starter cultures in their production) observed to be transferred between Enterococcus spp. in the and an increasing interest in probiotic products result in GI of humans [78]. Furthermore, the possible transfer of high daily occurrences of antibiotic resistance genes in the GI a plasmid harbouring ampicillin resistance from the food tract. The increased level of ingested Gram-positive bacteria contaminant Salmonella Enteritidis to indigenous E. coli has caused new speculations that these bacteria might also in the GI tract of humans was described, and in vitro contribute to the reservoir of antibiotic resistance genes, experiments confirmed that transfer was possible at high and when the right circumstances are present; these genes frequencies [79]. Transfer of any antibiotic resistance genes could be transferred to the indigenous flora. The worst case is a general concern, yet transfer of Extended spectrum International Journal of Microbiology 5 beta-lactamases (ESBL) resistance genes is a category of product [84], for example, the presence of multiresistant its own, which might result in treatment limitations and Salmonella typhimurium in Carpaccio [85], Shigella sonnei in worst cases of treatment failure [4, 80]. Interestingly, infections caused by imported baby corn [86], E. coli O157 indigenous E. coli strains with a blaCTX−M ESBL gene were infections caused by contaminated lettuce [87], and a recent isolated form pigs. During treatment with cephalosporin’s outbreak caused by Salmonella kedougou [88]. However, (veterinary treatment), the diversity of the indigenous E. coli antibiotic resistant pathogenic strains are also frequently strains receiving the blaCTX−M genes increased, suggesting detected in production animals in Europe, such as MRSA horizontal gene transfer during selective pressure [81]. If in pigs [89], multiresistant Salmonella infantis in broiler meat for consumption is contaminated with the faecal ESBL chicken [90], ESBL producing Salmonella spp. in poultry producing E. coli, and if the meat is not properly processed, [91], and also in food products, such as vancomycin resistant humans can ingest the ESBL producing E. coli and the gene E. faecalis in turkey meat [92], multiresistant E. coli in minced transfer can occur again in the human GI tract. An example beef [93], and tetracycline resistant E. faecalis in poultry meat of selective pressure increasing the possibility of transfer (chicken, turkey and duck), beef, and pork [94]. was described by a clinical case documenting transfer of It has also been demonstrated that antibiotic resistance ACC-1 AmpC (ESBL) from K. pneumoniae to indigenous genes of different Lactic acid bacterial species, for exam- E. coli during antibiotic treatment of 1-year old boy [82]. ple, Lactococcus spp., Streptococcus thermophilus, and Pseu- The child was colonised with a K. pneumoniae harbouring domonas spp., are present in “ready to use” products such as ACC-1 AmpC. The postoperative treatment was imipenem different cheese products, raw milk, meat products like pork and amikacin which lead to a spontaneous mutation that chop, turkey, and beef as well as in mushroom and spinach decreased the sensitivity towards imipenem. Subsequently, [95]. In vitro experiments have shown that L. lactis DNA, the plasmid was transferred to a commensal ampicillin containing the tetracycline resistance gene tetS, was success- resistant E. coli during treatment with cefotaxim due to fully transformed into the oral cariogenic pathogen Strepto- urinary tract infection caused by the E. coli [82]. This is coccus mutans [95]. Although this experiment illustrates the a very important case story illustrating and emphasising transfer of antibiotic resistant genes from a commensal to a the complexity of circumstances and mechanisms during pathogenic bacterium, these events have not been confirmed antibiotic resistance gene transfer in the gut. in vivo. Other studies showed transfer of erythromycin resistance gene ermB from a commensal L. plantarum strain 7. Farm-to-Fork Perspective to E. faecalis in vivo in the diassociated rat model [62]. However, this was not reproducible in the streptomycin- The ecological niches of human, animal, water, and soil treated mice model [62]. Although probiotic and industrial can easily be evaluated as separated small niches, but bacterial cultures were seen to possess the acquired antibiotic this evaluation might not yield the entire truth. Bacteria resistance genes, no transfer in vitro was observed, neither are present in microecological niches, but move between intraspecies nor interspecies [96]. Whereas few natural ecosystems from animals to humans, from humans and resistant LAB were isolated from cheese (raw milk without animals (faeces and manure) to water and soil and return starter cultures), only two L. lactis strains harbouring the to human and animals, through, for example, food (plants conjugative tetM resistance transposon Tn916 were able to or vegetables). In addition, the use of antibiotic treatment transfer to both Lactococcus and Enterococcus [97]. in each small niche (humans, animals or plants) selects Transfer of antibiotic resistance genes from Gram- the resistant strains to become the reservoir of resistance positive to Gram-negative bacteria in vitro is a very rare genes. These antibiotic resistance genes are present and event. Invitrotransfer of a naturally occurring Gram- can be transported within the bacteria from one niche to positive plasmid pIP501 in E. faecalis to E. coli has been another. Figure 1 illustrates the interaction between the described [98]. Another naturally occurring plasmid, pIP823 different reservoirs of bacteria in the food chain [83]. The from Listeria monocytogenes, was only transferred in the nonpathogenic bacteria are widely used as feed additives of presence of the broad host range plasmid, pAMβ1which animal feed, as starter cultures in food preparation, or as mobilised the transfer to both E. faecalis and E. coli [99]. probiotics for humans. Antibiotic treatment of humans (in Nevertheless, sequence analysis documents almost identical the hospitals or in the community), animals, and pest control antibiotic resistance genes and thereby suggests that transfer on plants, creates a selection of resistant bacteria in those occur from Gram-positive to Gram-negative bacteria strains areas. Humans ingest meat and plants which might contain [29]. In addition, the opposite transfer situation is not bacteria with antibiotic resistance genes, but the increasing believed to be possible because of the difference in the gene consumption of fermented food products increases the risk expression systems [29]. of antibiotic resistance genes to occur in the intestine. It is Plasmid transfer across genus from LAB have been a well-known fact that human foodstuffs contain bacteria rarely described and few studies are performed in an in harbouring antibiotic resistance genes, but the main focus vivo model like the diassociated rats [62, 73]. The many has been on pathogen contaminants in foods. Raw food, in vitro studies showed that the Gram-negative flora which is not properly processed before eating, might result harbours the mobilised plasmids and the main focus in infections caused by pathogenic resistant bacterial strains. in antibiotic resistance development should remain on Epidemiologically studies have shown that through trace these. Furthermore, transposons like the Tn916 (harboring back,anoutbreakstraincanberelatedtoaspecificfood tetracycline resistance) were also shown as being transferable 6 International Journal of Microbiology

AN

Hospital Patients Cross contamination

Meat products

Aquatic environment

Surface water

Waste water

Slurry Faeces Production Soil AN animal Community AN

Animal feed Plants Food

AN

Meat products Probiotic

Figure 1: Schematic representation of the complexity of selection/development of antibiotic resistant bacteria in different known reservoirs. The possible routes of transmission throughout the environment of these resistant bacteria are suggested. The reservoirs where antibiotics are applied are also suggested as hot spots for horizontal gene transfer. AN: antibiotic treatment/pest control, adapted from [83].

in vitro between Gram-positive and Gram-negative genus mechanisms that will be observed in the human pathogenic [100]. Studies like this in vitro transfer event should be strains in the future [29]. However, horizontal gene transfer further investigated in in vivo models. does not appear to have homogenised the bacteria. Genetic Transgenic food has been investigated for its influence diversity and a well-defined phylogenetic tree for bacteria on the spread of antibiotic resistance, and the conclusion is are still the rule rather than the exception [102]. that there is very little reason to assume that consumption High level of antibiotic resistance, combined with the of transgenic food or feed additives increases the risk. tendency to treat infections with more broad-spectrum Different environmental factors have an impact on HGT drugs, results in a selection of multiresistant bacteria. Due to events, selective advantage for the bacterial population being import/export of food products and travelling habits, a single the most important [101]. part of the world cannot deal with this threat alone. The optimal solution would be worldwide consensus on the use 8. Concluding Remarks of antibiotics, both in health care and in the veterinary sector. There is no doubt that antibiotic treatment is a necessity, A greatly respected scientist within the antibiotic resistance and the influence on the total GI flora is a matter of sec- area has recently stated that “Evolution of bacteria towards ondary importance. Conversely, antibiotic treatment creates antibiotic resistance is unavoidable as it represents a a great advantage for resistant bacteria which is selected to particular aspect of the general evolution of bacteria. Thus, colonise the intestine and treatment might result in antibiotic at the very best, the only hope we can have in the field of resistance transfer among Gram-negative bacteria, or to the antibiotic resistance is to delay dissemination of resistant indigenous flora or even a pathogen. The ecological effects bacteria or resistance genes” [29]. In addition, it is suggested of antibiotic treatment on the commensal microflora should that identifying more resistance mechanisms in antibiotics be the focus of more studies in the future. Rational use of producing strains might be the solution to predict the antibiotics together with infection control will possibly limit International Journal of Microbiology 7 further spread of multiresistant bacteria, but no matter what [14]N.Høiby,T.Bjarnsholt,M.Givskov,S.Molin,andO.Ciofu, boundaries we set up, transfer of antibiotic resistance genes “Antibiotic resistance of bacterial biofilms,” International among bacteria can and will occur. Journal of Antimicrobial Agents, vol. 35, no. 4, pp. 322–332, 2010. [15]K.D.Xu,G.A.McFeters,andP.S.Stewart,“Biofilm Acknowledgment resistance to antimicrobial agents,” Microbiology, vol. 146, no. 3, pp. 547–549, 2000. This work was partially supported by the European Com- [16] B. B. Christensen, C. Sternberg, J. B. Andersen et al., mission Grant CT-2003-506214 (ACE-ART) under the 6th “Establishment of new genetic traits in a microbial biofilm Framework Programme. community,” Applied and Environmental Microbiology, vol. 64, no. 6, pp. 2247–2255, 1998. [17] S. J. Sørensen, M. Bailey, L. H. Hansen, N. Kroer, and S. References Wuertz, “Studying plasmid horizontal transfer in situ: a critical review,” Nature Reviews Microbiology, vol. 3, no. 9, pp. [1] R. Bud, “Antibiotics: the epitome of a wonder drug,” BMJ, 700–710, 2005. vol. 334, p. s6, 2007. [18] S. Molin and T. Tolker-Nielsen, “Gene transfer occurs [2] I. Andersson, A. C. Terwisscha van Scheltinga, and K. with enhanced efficiency in biofilms and induces enhanced Valegard,˚ “Towards new β-lactam antibiotics,” Cellular and stabilisation of the biofilm structure,” Current Opinion in Molecular Life Sciences, vol. 58, no. 12-13, pp. 1897–1906, Biotechnology, vol. 14, no. 3, pp. 255–261, 2003. 2001. [19] D. Aragon, M. L. Sole, and S. Brown, “Outcomes of an [3] J. J. Ross, M. G. Worthington, S. L. Gorbach et al., “Resistance infection prevention project focusing on hand hygiene and to levofloxacin and failure of treatment of Pneumococcal isolation practices,” AACN Clinical Issues,vol.16,no.2,pp. pneumonia,” New England Journal of Medicine, vol. 347, no. 121–132, 2005. 1, pp. 65–67, 2002. [20] N. Frimodt-Møller and B. Gahrn-Hansen, “Antibiotics in a [4]W.Song,E.S.Moland,N.D.Hanson,J.S.Lewis,J.H. hospital hygienic perspective,” Ugeskrift for Laeger, vol. 169, Jorgensen, and K. S. Thomson, “Failure of cefepime therapy no. 49, pp. 4254–4256, 2007. in treatment of Klebsiella pneumoniae bacteremia,” Journal of [21] G. L. Simon and S. L. Gorbach, “Intestinal flora in health and Clinical Microbiology, vol. 43, no. 9, pp. 4891–4894, 2005. disease,” Gastroenterology, vol. 86, no. 1, pp. 174–193, 1984. [5] D. Talon, M. C. Woronoff-Lemsi, S. Limat et al., “The impact [22] S. P. Borriello, “Microbial flora of the gastrointestinal tract,” of resistance to methicillin in Staphylococcus aureus bac- in Microbial metabolism in the digestive tract, M. J. Hill, Ed., teremia on mortality,” European Journal of Internal Medicine, pp. 2–16, CRC Press, Boca Raton, Fla, USA, 1986. vol. 13, no. 1, pp. 31–36, 2002. [23] E. Isolauri, S. Salminen, and A. C. Ouwehand, “Microbial-gut [6] F. M. Aarestrup and B. Carstensen, “Effect of tylosin used interactions in health and disease,” Probiotics Best Practice & as a growth promoter on the occurrence of macrolide- Research Clinical Gastroenterology , vol. 18, no. 2, pp. 299– resistant enterococci and staphylococci in pigs,” Microbial 313, 2008. Drug Resistance, vol. 4, no. 4, pp. 307–312, 1998. [24] E. G. Zoetendal, C. T. Collier, S. Koike, R. I. Mackie, [7] Anonymous, “DANMAP—Use of antimicrobial agents and and H. R. Gaskins, “Molecular ecological analysis of the occurrence of antimicrobial resistance in bacteria from food gastrointestinal microbiota: a review,” Journal of Nutrition, animals, foods and humans in Denmark,” 2007. vol. 134, no. 2, pp. 465–472, 2004. [8] K. Grave, V. F. Jensen, K. Odensvik, M. Wierup, and M. [25] R. I. Amann, W. Ludwig, and K. H. Schleifer, “Phylogenetic Bangen, “Usage of veterinary therapeutic antimicrobials in identification and in situ detection of individual microbial Denmark, Norway and Sweden following termination of cells without cultivation,” Microbiological Reviews, vol. 59, antimicrobial growth promoter use,” Preventive Veterinary no. 1, pp. 143–169, 1995. Medicine, vol. 75, no. 1-2, pp. 123–132, 2006. [26] J. Qin, R. Li, J. Raes et al., “A human gut microbial gene [9] Anomymous, “European Commission, Opinion of the Sci- catalogue established by metagenomic sequencing,” Nature, entific Panel on Additives and Products or Substances used vol. 464, no. 7285, pp. 59–65, 2010. in Animal Feed on the updating of the criteria used in the [27] E. Grasselli, P. Franc¸ois, M. Gutacker et al., “Evidence assessment of bacteria for resistance to antibiotics of human of horizontal gene transfer between human and animal or veterinary importance Question N◦ EFSA-Q-2004-079,” commensal Escherichia coli strains identified by microarray,” The EFSA Journal, vol. 223, pp. 1–12, 2005. FEMS Immunology and Medical Microbiology, vol. 53, no. 3, [10] C. A.M. McNulty and N. A. Francis, “Optimizing antibiotic pp. 351–358, 2008. prescribing in primary care settings in the UK: findings of a [28] A. A. Salyers, A. Gupta, and Y. Wang, “Human intestinal BSAC multi-disciplinary worksho,” Journal of Antimicrobial bacteria as reservoirs for antibiotic resistance genes,” Trends Chemotherapy, vol. 65, no. 11, pp. 2278–2284, 2010. in Microbiology, vol. 12, no. 9, pp. 412–416, 2004. [11] L. Grigoryan, F. M. Haaijer-Ruskamp, J. G. M. Burgerhof [29] P. Courvalin, “Predictable and unpredictable evolution of et al., “Self-medication with antimicrobial drugs in Europe,” antibiotic resistance,” Journal of Internal Medicine, vol. 264, Emerging Infectious Diseases, vol. 12, no. 3, pp. 452–459, 2006. no. 1, pp. 4–16, 2008. [12] H. Goossens, M. Ferech, R. Vander Stichele, and M. Elseviers, [30] P. Mazodier and J. Davies, “Gene transfer between distantly “Outpatient antibiotic use in Europe and association with related bacteria,” Annual Review of Genetics, vol. 25, pp. 147– resistance: a cross-national database study,” The Lancet, vol. 171, 1991. 365, no. 9459, pp. 579–587, 2005. [31] F. M. Aarestrup, H. Kruse, E. Tast, A. M. Hammerum, and [13] H. Goossens, D. Guillemot, M. Ferech et al., “National L. B. Jensen, “Associations between the use of antimicrobial campaigns to improve antibiotic use,” European Journal of agents for growth promotion and the occurrence of resis- Clinical Pharmacology, vol. 62, no. 5, pp. 373–379, 2006. tance among Enterococcus faecium from broilers and pigs 8 International Journal of Microbiology

in Denmark, Finland, and Norway,” Microbial Drug Resis- Maronpot, Ed., pp. 81–118, Cache River Press, St. Louis, Mo, tance, vol. 6, no. 1, pp. 63–70, 2000. USA, 1999. [32] D. I. Andersson and B. R. Levin, “The biological cost of [48] R. Ducluzeau, “Microbial Interactions in the digestive tract,” antibiotic resistance,” Current Opinion in Microbiology, vol. in The Germ-Free Animal in Biomedical Research,M.Coates 2, no. 5, pp. 489–493, 1999. and B. E. Gustafsson, Eds., vol. 9, pp. 141–154, Laboratory [33] D. I. Andersson and D. Hughes, “Antibiotic resistance and Animal, 1984. its cost: is it possible to reverse resistance?” Nature Reviews [49] M. Pollard and N. Sharon, “Responses of the Peyer’s patches Microbiology, vol. 8, no. 4, pp. 260–271, 2010. in germ-free mice to antigenic stimulation,” Infection and [34] J. Bjorkman,¨ D. Hughes, and D. I. Andersson, “Virulence Immunity, vol. 2, no. 1, pp. 96–100, 1970. of antibiotic-resistant Salmonella typhimurium,” Proceedings [50] R. Freter, “Mechanisms that control the microflora in the of the National Academy of Sciences of the United States of large intestine,” in Human Intestinal Microflora in Health and America, vol. 95, no. 7, pp. 3949–3953, 1998. Disease, D. J. Hentges, Ed., pp. 33–54, 1983. [35] L. L. Marcusson, N. Frimodt-Møller, and D. Hughes, [51] D. J. Hentges, A. J. Stein, S. W. Casey, and J. U. Que, “Interplay in the selection of fluoroquinolone resistance and “Protective role of intestinal flora against infection with bacterial fitness,” PLoS Pathogens, vol. 5, no. 8, Article ID Pseudomonas aeruginosa in mice: influence of antibiotics on e1000541, 2009. colonization resistance,” Infection and Immunity, vol. 47, no. [36] C. Fermer´ and G. Swedberg, “Adaptation to sulfonamide 1, pp. 118–122, 1985. resistance in Neisseria meningitidis may have required [52] D. J. Hentges, “The Influence of streptomycon on coloniza- compensatory changes to retain enzyme function: kinetic tion resistance in mice,” Microecology and Therapy, vol. 14, analysis of dihydropteroate synthases from N. meningitidis pp. 53–62, 1984. expressedinaknockoutmutantofEscherichia coli,” Journal [53] L. K. Poulsen, T. R. Licht, C. Rang, K. A. Krogfelt, and S. of Bacteriology, vol. 179, no. 3, pp. 831–837, 1997. Molin, “Physiological state of Escherichia coli BJ4 growing in [37] J. Bjorkman,¨ I. Nagaev, O. G. Berg, D. Hughes, and D. I. the large intestines of streptomycin-treated mice,” Journal of ff Bacteriology, vol. 177, no. 20, pp. 5840–5845, 1995. Andersson, “E ects of environment on compensatory muta- ff tions to ameliorate costs of antibiotic resistance,” Science, vol. [54] J. U. Que and D. J. Hentges, “E ect of streptomycin adminis- 287, no. 5457, pp. 1479–1482, 2000. tration on colonization resistance to Salmonella typhimurium in mice,” Infection and Immunity, vol. 48, no. 1, pp. 169–174, [38] I. Nagaev, J. Bjorkman,¨ D. I. Andersson, and D. Hughes, 1985. “Biological cost and compensatory evolution in fusidic acid- [55] K. Hirayama, “Ex-germfree mice harboring intestinal micro- resistant Staphylococcus aureus,” Molecular Microbiology, vol. biota derived from other animal species as an experimental 40, no. 2, pp. 433–439, 2001. model for ecology and metabolism of intestinal bacteria,” [39] B. R. Levin, V. Perrot, and N. Walker, “Compensatory Experimental Animals, vol. 48, no. 4, pp. 219–227, 1999. mutations, antibiotic resistance and the population genetics [56] R. Kibe, M. Sakamoto, H. Yokota et al., “Movement and ofadaptiveevolutioninbacteria,”Genetics, vol. 154, no. 3, fixation of intestinal microbiota after administration of pp. 985–997, 2000. human feces to germfree mice,” Applied and Environmental [40] F. Guarner and J. R. Malagelada, “Gut flora in health and Microbiology, vol. 71, no. 6, pp. 3171–3178, 2005. disease,” The Lancet, vol. 361, no. 9356, pp. 512–519, 2003. [57] P. B. Eckburg, E. M. Bik, C. N. Bernstein et al., “Micro- ˚ ff [41] A. Sullivan, “E ect of antimicrobial agents on the ecological biology: diversity of the human intestinal microbial flora,” balance of human microflora,” Lancet Infectious Diseases, vol. Science, vol. 308, no. 5728, pp. 1635–1638, 2005. 1, no. 2, pp. 101–114, 2001. [58] G. W. Tannock, “Microbiota of mucosal surfaces in the gut of [42] C. E. Nord, A. Heimdahl, and L. Kager, “Antimicrobial agents monogastric animals,” in Colonization of Mucosal Surfaces,J. and the human oropharyngeal and intestinal microflora,” P.Nataro,P.S.Cohen,H.L.T.Mobley,andJ.N.Weiser,Eds., Annali dell’Istituto Superiore di Sanita, vol. 22, no. 3, pp. 883– pp. 163–178, ASM Press, Washington, DC, USA, 2005. 892, 1986. [59] H. A. Gordon and L. Pesti, “The gnotobiotic animal as a tool [43] A. Apisarnthanarak and L. M. Mundy, “Death due to in the study of host microbial relationships,” Bacteriological ffi community-associated Clostridium di cile in a woman Reviews, vol. 35, no. 4, pp. 390–429, 1971. receiving prolonged antibiotic therapy for suspected lyme [60] Anomymous and FAO/WHO Consultations and workshops, disease,” Clinical Infectious Diseases, vol. 51, no. 3, pp. 369– “Safety assessment of foods derived from genetically modi- 370, 2010. fied microorganisms,” Report of a Joint FAO/WHO Expert [44] S. M. Wren, N. Ahmed, A. Jamal et al., “Preoperative Consultation on Foods Derived from Biotechnology, pp.1– oral antibiotics in colorectal surgery increase the rate of 29, 2001. Clostridium difficile colitis,” Archives of Surgery, vol. 140, no. [61] R. J. Boyle, R. M. Robins-Browne, and M. L. K. Tang, 8, pp. 752–756, 2005. “Probiotic use in clinical practice: what are the risks?” [45] G. R. Gibson and R. Fuller, “Aspects of in vitro and in vivo American Journal of Clinical Nutrition,vol.83,no.6,pp. research approaches directed toward identifying probiotics 1256–1264, 2006. and prebiotics for human use,” Journal of Nutrition, vol. 130, [62] L. Feld, S. Schjørring, K. Hammer et al., “Selective pressure no. 2, 2000. affects transfer and establishment of a Lactobacillus plan- [46] N. Toomey, A. Monaghan, S. Fanning, and D. Bolton, tarum resistance plasmid in the gastrointestinal environ- “Transfer of antibiotic resistance marker genes between lactic ment,” Journal of Antimicrobial Chemotherapy, vol. 61, no. 4, acid bacteria in model rumen and plant environments,” pp. 845–852, 2008. Applied and Environmental Microbiology, vol. 75, no. 10, pp. [63] D. Gevers, G. Huys, and J. Swings, “in vitro conjugal transfer 3146–3152, 2009. of tetracycline resistance from Lactobacillus isolates to other [47] C. C. Schackelford and M. R. Elwell, “Small and large Gram-positive bacteria,” FEMS Microbiology Letters, vol. 225, intestine, and mesentery,” in Pathology of the Mouse,R.R. no. 1, pp. 125–130, 2003. International Journal of Microbiology 9

[64] L. Jacobsen, A. Wilcks, K. Hammer, G. Huys, D. Gevers, and for Applied Bacteriology symposium series,vol.3,no.0,pp. S. R. Andersen, “Horizontal transfer of tet(M) and erm(B) 241–253, 1974. resistance plasmids from food strains of Lactobacillus plan- [78] C. H. Lester, N. Frimodt-Møller, T. L. Sørensen, D. L. tarum to Enterococcus faecalis JH2-2 in the gastrointestinal Monnet, and A. M. Hammerum, “in vivo transfer of the tract of gnotobiotic rats,” FEMS Microbiology Ecology, vol. 59, vanA resistance gene from an Enterococcus faecium isolate of no. 1, pp. 158–166, 2007. animal origin to an E. faecium isolate of human origin in [65] Y. Sasaki, N. Taketomo, and T. Sasaki, “Factors affecting the intestines of human volunteers,” Antimicrobial Agents and transfer frequency of pAMβ1fromStreptococcus faecalis to Chemotherapy, vol. 50, no. 2, pp. 596–599, 2006. Lactobacillus plantarum,” JournalofBacteriology, vol. 170, no. [79] E. Balis, A. C. Vatopoulos, M. Kanelopoulou et al., “Indi- 12, pp. 5939–5942, 1988. cations of in vivo transfer of an epidemic R plasmid from [66] J. Schlundt, P. Saadbye, B. Lohmann, B. L. Jacobsen, and Salmonella enteritidis to Escherichia coli of the normal human E. M. Nielsen, “Conjugal transfer of plasmid DNA between gut flora,” Journal of Clinical Microbiology,vol.34,no.4,pp. Lactococcus lactis strains and distribution of transconjugants 977–979, 1996. in the digestive tract of gnotobiotic rats,” Microbial Ecology in [80] L. H. Su, C. H. Chiu, C. Chu, M. H. Wang, JU. H. Chia, Health and Disease, vol. 7, no. 2, pp. 59–69, 1994. and T. L. Wu, “in vivo acquisition of ceftriaxone resistance in [67] M. Gruzza, M. Fons, M. F. Ouriet, Y. Duval-Iflah, and Salmonella enterica serotype anatum,” Antimicrobial Agents R. Ducluzeau, “Study of gene transfer in vitro and in the and Chemotherapy, vol. 47, no. 2, pp. 563–567, 2003. digestive tract of gnotobiotic mice from Lactococcus lactis [81] L. M. Cavaco, E. Abatih, F. M. Aarestrup, and L. Guardabassi, strains to various strains belonging to human intestinal “Selection and persistence of CTX-M-producing Escherichia flora,” Microbial Releases, vol. 2, no. 4, pp. 183–189, 1994. coli in the intestinal flora of pigs treated with amoxi- [68] G. W. Tannock, “Conjugal transfer of plasmid pAM beta 1 in cillin, ceftiofur, or cefquinome,” Antimicrobial Agents and Lactobacillus reuteri and between lactobacilli and Enterococ- Chemotherapy, vol. 52, no. 10, pp. 3612–3616, 2008. ff cus faecalis,” Applied and Environmental Microbiology, vol. 53, [82] P. Bidet, B. Burgho er, V. Gautier et al., “in vivo transfer of no. 11, pp. 2693–2695, 1987. plasmid-encoded ACC-1 AmpC from Klebsiella pneumoniae [69] P. S. Cocconcelli, L. Morelli, and M. Vescovo, “Conjugal to Escherichia coli in an infant and selection of impermeabil- transfer of antibiotic resistances from Lactobacillus to Strep- ity to imipenem in K. pneumoniae,” Antimicrobial Agents and tococcus lactis,” Microbiologie Aliments Nutrition, vol. 3, no. 2, Chemotherapy, vol. 49, no. 8, pp. 3562–3565, 2005. pp. 163–165, 1985. [83] W. Witte, “Ecological impact of antibiotic use in animals on different complex microflora: environment,” International [70] M. J. Gasson and F. L. Davies, “Conjugal transfer of the drug Journal of Antimicrobial Agents, vol. 14, no. 4, pp. 321–325, resistance plasmid pAMβ in the lactic streptococci,” FEMS 2000. Microbiology Letters, vol. 7, no. 1, pp. 51–53, 1980. [84] Anonymous, Annual Report on Zoonoses in Denmark 2006, [71]A.W.Shrago,B.M.Chassy,andW.J.Dobrogosz,“Conju- Technical University of Denmark, 2006. gal plasmid transfer (pAMβ1) in Lactobacillus plantarum,” [85] S. Ethelberg, G. Sørensen, B. Kristensen et al., “Outbreak Applied and Environmental Microbiology,vol.52,no.3,pp. with multi-resistant Salmonella Typhimurium DT104 linked 574–576, 1986. to carpaccio, Denmark, 2005,” Epidemiology and Infection, [72] C. A. West and P. J. Warner, “Plasmid profiles and transfer vol. 135, no. 6, pp. 900–907, 2007. of plasmid-encoded antibiotic resistance in Lactobacillus [86] H. C. Lewis, M. Kirk, S. Ethelberg et al., “Outbreaks plantarum,” Applied and Environmental Microbiology, vol. 50, of shigellosis in Denmark and Australia associated with no. 5, pp. 1319–1321, 1985. imported baby corn, August 2007—final summary,” Euro [73] L. Morelli, P. G. Sarra, and V. Bottazzi, “in vivo transfer of Surveillance, vol. 12, no. 10, Article ID E071004, 2007. β pAM 1fromLactobacillus reuteri to Enterococcus faecalis,” [87] A. Soderstr¨ om,¨ A. Lindberg, and Y. Andersson, “EHEC O157 Journal of Applied Bacteriology, vol. 65, no. 5, pp. 371–375, outbreak in Sweden from locally produced lettuce, August- 1988. September 2005,” Euro Surveillance, vol. 10, no. 9, Article ID [74]J.M.Korhonen,Y.Sclivagnotis,andA.V.Wright,“Char- E050922, 2005. acterization of dominant cultivable lactobacilli and their [88] K. E. Emberland, K. Nygard,˚ B. T. Heier et al., “Outbreak antibiotic resistance profiles from faecal samples of weaning of Salmonella Kedougou in Norway associated with salami, piglets,” Journal of Applied Microbiology, vol. 103, no. 6, pp. April-June 2006,” Euro Surveillance, vol. 11, no. 7, Article ID 2496–2503, 2007. E060706, 2006. [75] S. Schjørring, C. Struve, and K. A. Krogfelt, “Transfer of [89] L. Guardabassi, M. Stegger, and R. Skov, “Retrospective antimicrobial resistance plasmids from Klebsiella pneumo- detection of methicillin resistant and susceptible Staphylo- niae to Escherichia coli in the mouse intestine,” Journal of coccus aureus ST398 in Danish slaughter pigs,” Veterinary Antimicrobial Chemotherapy, vol. 62, no. 5, pp. 1086–1093, Microbiology, vol. 122, no. 3-4, pp. 384–386, 2007. 2008. [90] N. Nogr´ ady,´ G. Kardos, A. Bistyak´ et al., “Prevalence and [76]K.Tuohy,M.Davies,P.Rumsby,C.Rumney,M.R.Adams, characterization of Salmonella infantis isolates originating and I. R. Rowland, “Monitoring transfer of recombinant and from different points of the broiler chicken-human food nonrecombinant plasmids between Lactococcus lactis strains chain in Hungary,” International Journal of Food Microbiol- and members of the human gastrointestinal microbiota in ogy, vol. 127, no. 1-2, pp. 162–167, 2008. vivo—Impact of donor cell number and diet,” Journal of [91] H. Hasman, D. Mevius, K. Veldman, I. Olesen, and F. Applied Microbiology, vol. 93, no. 6, pp. 954–964, 2002. M. Aarestrup, “β-Lactamases among extended-spectrum β- [77]G.C.Burton,D.C.Hirsh,D.C.Blenden,andJ.L. lactamase (ESBL)-resistant Salmonella from poultry, poultry Zeigler, “The effects of tetracycline on the establishment of products and human patients in The Netherlands,” Journal Escherichia coli of animal origin, and in vivo transfer of of Antimicrobial Chemotherapy, vol. 56, no. 1, pp. 115–121, antibiotic resistance, in the intestinal tract of man,” Society 2005. 10 International Journal of Microbiology

[92]Y.Agersø,C.H.Lester,L.J.Porsboetal.,“Vancomycin- resistant Enterococcus faecalis isolates from a Danish patient and two healthy human volunteers are possibly related to isolates from imported turkey meat,” Journal of Antimicrobial Chemotherapy, vol. 62, no. 4, pp. 844–845, 2008. [93] A. Dambrosio, V. Lorusso, N. C. Quaglia et al., “Escherichia coli O26 in minced beef: prevalence, characterization and antimicrobial resistance pattern,” International Journal of Food Microbiology, vol. 118, no. 2, pp. 218–222, 2007. [94] A. Wilcks, S. R. Andersen, and T. R. Licht, “Characterization of transferable tetracycline resistance genes in Enterococcus faecalis isolated from raw food,” FEMS Microbiology Letters, vol. 243, no. 1, pp. 15–19, 2005. [95] H. H. Wang, M. Manuzon, M. Lehman et al., “Food commensal microbes as a potentially important avenue intransmitting antibiotic resistance genes,” FEMS Microbiol- ogy Letters, vol. 254, no. 2, pp. 226–231, 2006. [96] I. Klare, C. Konstabel, G. Werner et al., “Antimicrobial susceptibilities of Lactobacillus, Pediococcus and Lactococ- cus human isolates and cultures intended for probiotic or nutritional use,” Journal of Antimicrobial Chemotherapy, vol. 59, no. 5, pp. 900–912, 2007. [97] A. B. Florez,M.S.Ammor,andB.Mayo,“Identification´ of tet(M) in two Lactococcus lactis strains isolated from a Spanish traditional starter-free cheese made of raw milk and conjugative transfer of tetracycline resistance to lactococci and enterococci,” International Journal of Food Microbiology, vol. 121, no. 2, pp. 189–194, 2008. [98] B. Kurenbach, C. Bohn, J. Prabhu, M. Abudukerim, U. Szewzyk, and E. Grohmann, “Intergeneric transfer of the Enterococcus faecalis plasmid pIP501 to Escherichia coli and Streptomyces lividans and sequence analysis of its tra region,” Plasmid, vol. 50, no. 1, pp. 86–93, 2003. [99] E. Charpentier, G. Gerbaud, and P. Courvalin, “Conjugative mobilization of the rolling-circle plasmid pIP823 from Listeria monocytogenes BM4293 among gram-positive and gram-negative bacteria,” Journal of Bacteriology, vol. 181, no. 11, pp. 3368–3374, 1999. [100] J. Bertram, M. Stratz, and P. Durre, “Natural transfer of conjugative transposon Tn916 between gram-positive and gram-negative bacteria,” Journal of Bacteriology, vol. 173, no. 2, pp. 443–448, 1991. [101] G. Van Den Eede, H. Aarts, H. J. Buhk et al., “The relevance of gene transfer to the safety of food and feed derived from genetically modified (GM) plants,” Food and Chemical Toxicology, vol. 42, no. 7, pp. 1127–1156, 2004. [102] P. Wiener, S. Egan, A. S. Huddleston, and E. M. H. Welling- ton, “Evidence for the transfer of antibiotic-resistance genes in soil populations of streptomycetes,” Molecular Ecology, vol. 7, pp. 1205–1216, 1998. Hindawi Publishing Corporation International Journal of Microbiology Volume 2011, Article ID 130574, 6 pages doi:10.1155/2011/130574

Research Article Parenteral Antibiotics Reduce Bifidobacteria Colonization and Diversity in Neonates

Seamus´ Hussey,1, 2 Rebecca Wall,3 Emma Gruffman,4 Lisa O’Sullivan,4 C. Anthony Ryan,1 Brendan Murphy,1 Gerald Fitzgerald,3, 5 Catherine Stanton,3, 4 and R. Paul Ross3, 4

1 Department of Paediatrics and Child Health, University College Cork, Cork, Ireland 2 Division of Gastroenterology, Hepatology and Nutrition, The Hospital for Sick Children, Toronto, ON, Canada M5G 1X8 3 Alimentary Pharmabiotic Centre, Cork, Ireland 4 Department of Biotechnology, Teagasc Moorepark Food Research Centre, Fermoy, Cork, Ireland 5 Department of Microbiology, University College Cork, Cork, Ireland

Correspondence should be addressed to R. Paul Ross, [email protected]

Received 28 April 2010; Accepted 18 June 2010

Academic Editor: Max Teplitski

Copyright © 2011 Seamus´ Hussey et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

We investigated the impact of parenteral antibiotic treatment in the early neonatal period on the evolution of bifidobacteria in the newborn. Nine babies treated with intravenous ampicillin/gentamicin in the first week of life and nine controls (no antibiotic treatment) were studied. Denaturing gradient gel electrophoresis was used to investigate the composition of Bifidobacterium in stool samples taken at four and eight weeks. Bifidobacteria were detected in all control infants at both four and eight weeks, while only six of nine antibiotic-treated infants had detectable bifidobacteria at four weeks and eight of nine at eight weeks. Moreover, stool samples of controls showed greater diversity of Bifidobacterium spp. compared with antibiotic-treated infants. In conclusion, short-term parenteral antibiotic treatment of neonates causes a disturbance in the expected colonization pattern of bifidobacteria in the first months of life. Further studies are required to probiotic determine if supplementation is necessary in this patient group.

1. Introduction Bifidobacterium spp. have several reported health- promoting effects, including inhibition of growth of harmful Up to 10% of newborn infants require admission to the bacteria, stimulation of the immune system, alleviation of neonatal intensive care unit (NICU) for ongoing medical constipation, and prevention of intestinal infections [8– care [1–3]. Many of these infants require treatment with 13]. Species commonly associated with humans include parenteral antibiotics for variable periods of time. Antibiotic Bifidobacterium infantis, B. longum, B. bifidum, B. breve, administration is known to perturb the composition of the B. catenulatum, B. pseudocatenulatum, B. angulatum, B. intestinal microbiota, resulting in suppression of anaerobic gallicum, and B. adolescensis [8], of which the first four are bacteria (with the exception of clostridia, which remain prevalent in infants [14]. at detectable levels) and increased numbers of potentially Suppression of the bifidobacteria population as occurred pathogenic bacteria such as Klebsiella, Enterobacter, Citrobac- following antibiotic exposure in children [5, 15, 16]mayhave ter, and Pseudomonas [4, 5]. Since the pioneering microbiota negative effects on host well-being and can be associated with of infancy probably create perpetual habitats for themselves, higher susceptibility to enteropathogenic bacterial infection defining the life-long composition of the gut microbiota [17]. However, the impact, if any, of parenteral antibiotics and consequently contributing to host health and well-being prescribed shortly following birth on the developing bifi- [6, 7], the initial colonization of the infant gastrointestinal dobacterial population has not been addressed. The aim tract is of significant importance. of this study was therefore to determine the impact of 2 International Journal of Microbiology

Table 1: Description of infant samples.

Sample Sex∗ Feeding∗∗ Antibiotic treatment (days) Type of Antibiotics† Mode of delivery A M B + (9) Amp. + Gent. Caesarean section B‡ M F + B + (5) Amp. + Gent. Caesarean section C‡ M B + (2) Amp. + Gent. Caesarean section D M F + (2) Amp. + Gent. Vaginal delivery E F F + (5) Amp. + Gent. Caesarean section F F B + (2) Amp. + Gent. Vaginal delivery G F B + (2) Amp. + Gent. Vaginal delivery H M F + (2) Amp. + Gent. Caesarean section I M F + (2) Amp. + Gent. Vaginal delivery J M F — — Vaginal delivery K M F — — Vaginal delivery L F F — — Vaginal delivery M F F — — Vaginal delivery N M B — — Vaginal delivery O F B — — Vaginal delivery P M B — — Vaginal delivery Q M F — — Vaginal delivery R F F — — Vaginal delivery ∗M, male; F, female.∗∗B, breast-feeding; F, formula feeding.†Amp. = ampicillin, Gent. = gentamicin. ‡ twins. intravenous antibiotic treatment in the neonatal period on [18] amplified a 1,417 bp fragment of the bifidobacterial the evolution of bifidobacterial colonization over time. 16S rRNA gene. PCR volumes of 50 µL contained 5 µLof 10× PCR buffer, 5 µLofMgCl2 (2.5 mM), 8 µLofdNTPs µ µ 2. Material and Methods (0,2 mM each), 1 L of each primer (5 pmol), 0.5 LofTaq polymerase (5 U/µL), 28.5 µLofsterileMilli-Qwater,and 2.1. Subjects. Faecal material was sampled, using swabs, from 1 µL of DNA solution. PCR-reagents were obtained from 18 infants, nine infants previously treated with antibiotics Bioline (Taunton, MA, USA). The following PCR program ◦ and nine healthy controls (Table 1). Approval for the study was used: initial denaturation at 94 C for 5 min; 3 cycles ◦ ◦ protocol was obtained from the Clinical Research Ethics of denaturation at 94 C for 45 s, annealing at 57 Cfor Committee of the Cork Teaching Hospitals, Cork, Ireland. 2 minutes, and extension at 72◦C for 1 min; 30 cycles of Inclusion criteria were defined as full or near term infants denaturation at 94◦C for 20 s, annealing at 57◦C for 1 min, requiring parenteral antibiotic administration within the and extension at 72◦C for 1 min; and final extension at first 48 hours of life. Control subjects were eligible if they 72◦C for 5 min followed by cooling to 4◦C. The presence were full or near term infants who were otherwise well and of PCR products was verified on a 1.5% (w/v) agarose did not require admission to the NICU. Infants placed on gel. In order to eliminate the remaining oligonucleotides oral antibiotics, those kept nil by mouth or requiring surgery, and original template DNA, purification of the amplicons infants with congenital anomalies, or those born before term was performed by use of the QIAquick PCR purification were excluded from the study. The indications for antibiotic kit (Qiagen) according to the manufacturer’s instructions. treatment were determined clinically, based on symptoms A second PCR was performed using the amplicons of the and/or signs of suspected sepsis, and independent of the first PCR as template DNA. The primer set used (F357- study investigators. Written informed consent was obtained GC and 518R) (5GC-clamp-GCCTACGGAGGCAGCAG- from parents of all infants. Stool samples were taken by the 3 and 5-ATTACCGCGGCTGCTGG-3, resp.) amplifies same clinician at four and eight weeks after birth and stored the V3 region of the bacterial 16S rRNA gene [19]. The ◦ at −20 C pending analysis. forward primer contained a GC clamp (5-CGCCCGCCG- CGCGCGGCGGGCGGGGCGGGGGCACGGGGG-3)to 2.2. PCR -DGGE Analysis. BacterialDNAwasextractedfrom facilitate separation of the amplicons in a DGGE gel. PCR swabs using a QIAamp DNA stool minikit (Qiagen, Hilden, volumes of 50 µL contained 5 µLof10× PCR buffer, 2 µLof Germany) by following the manufacturer’s instructions MgCl2 (1 mM), 8 µL of dNTPs (0.2 mM each), 2 µLofeach (lysis temperature, 95◦C). To investigate the bifidobacterial primer (5 µM), 0.5 µL of Taq polymerase (5 U/µL), 29.5 µ population in the samples, PCR was performed as a nested L of sterile Milli-Q water, and 1 µL of 10-fold diluted DNA approach. The first PCR applied primers Im26-f (5-GAT- solution (obtained from the first PCR). The following PCR TCTGGCTCAGGATGAACG-3)andIm3-r(5-CGGGTG- program was used: initial denaturation at 94◦C for 5 min, 30 CTICCCCACTTTCATG-3) described by Kaufmann et al. cycles of denaturation at 94◦C for 20 s, annealing at 58◦Cfor International Journal of Microbiology 3

D EFGHI JKLMNOPQ R

B. breve B. breve

B. longum B. longum

B. infantis B. infantis B. bifidum B. bifidum

B. adolescentis B. adolescentis

(a) (b)

Figure 1: DGGE of bifidobacterial PCR-products (V3-region) from stool samples taken at four weeks of age from infants treated with antibiotics D–I (a) and controls J–R (b). The mobility of the PCR products obtained in DGGE was compared to the PCR pattern of reference strains obtained with the same primer set.

45 s, and extension at 72◦C for 1 min, and final extension data are summarized in Table 1. While the gender distri- at 72◦C for 7 min, following by cooling to 4◦C. PCR bution and feeding history was similar in both groups, products were analyzed on DGGE gels based on the protocol the antibiotic treated infants were more likely to be born by Temmerman et al. [10]. A Dcode universal mutation by caesarean section (5/9 infants) and all control infants detection system (Bio-Rad Laboratories, Hercules, CA, USA) were born by spontaneous vaginal delivery. Ampicillin and was used, utilizing 16 cm by 16 cm by 1 mm gels. Eight gentamicin were the only antibiotics used and the median percent polyacrylamide gels were prepared and run with 1× duration of therapy was 2 days (range 2–9 days). TAE buffer diluted from 50× TAE buffer (2 M Tris base, 1 M glacial acetic acid, and 50 mM EDTA). The denaturing 3.1. Development of the Bifidobacterium Population at Four gradient was formed with two 8% acrylamide (acrylamide- Weeks of Age. In order to investigate the population of bis/acrylamide 40%) stock solutions (Severn Biotech Ltd, bifidobacteria in the samples, the Bifidobacterium-specific Worcestershire, UK). A 100% denaturing solution contained primers Im-3 and Im-26 were used. Amplicons obtained 40% (v/v) formamide and 7.0 M urea. The gels were poured using these primers served as template DNA for the V3 from the top by using a gradient maker (CBS Scientific primer combination V3R and V3F, during a second PCR Linear Gradient Maker, Bio-Rad Laboratories) and a pump step. The mobility of the PCR products obtained by these (Bio-Rad Laboratories) and gradients of 50% to 70% primers in DGGE was compared with the PCR pattern were used for the separation of the generated amplicons. of bifidobacteria reference strains. Because of the high Before polymerization of the denaturing gel (24 mL gradient G+C content of bifidobacteria, the conventional 35% to volume), a 6 mL stacking gel without denaturing chemicals 70% denaturing gradient was replaced with a 50% to 70% was added, and the appropriate comb was subsequently denaturing gradient. inserted. PCR amplicons were separated by electrophoresis at At four weeks of age, species of bifidobacteria were a constant voltage of 60 V in a 0.5 × TAE buffer at a constant ◦ detected in six out of nine infants treated with antibiotics temperature of 60 C for 16 h. Gels were stained in ethidium (subjects A–I) and in all infants in the control group (subjects bromide for 30 min, allowing digital capturing of the DGGE J–R). The samples from the antibiotic treated infants exhib- band profiles under UV light. ited a lower number of bands in the DGGE gel (Figure 1(a)) compared with controls (Figure 1(b))atfourweeksofage, ff 2.3. Reference Ladder. Amixtureoffivedi erent species of representing a less diverse population of bifidobacteria at bifidobacteria which are the most commonly isolated species this age. Five of six infants in the antibiotic-treated group of bifidobacteria from the infant gastrointestinal tract (GIT) showed a single dominant band, which corresponded to B. [14, 20, 21] was used to create a reference ladder in order longum (Figure 1(a)). All five infants had been treated with to enable a visual comparison of the bands. DNA extraction intravenous ampicillin/gentamicin for two days and there from the reference strains, Bifidobacterium breve, B. bifidum, were both breast fed (subject F, G) and formula fed (subject B. infantis, B. adolescentis, and B. longum, was performed D, H, I) infants within this group. The sample lacking a band ff according to the method described by Ho man and Winston in the position of B. longum was obtained from an infant that [22]. had received antibiotics for five days (subject E). According to the band pattern of this subject, some Bifidobacterium 3. Results species were able to colonize this infant despite antibiotic treatment for five days; that is, the lower band appears to be Nine subjects were recruited to both control and antibiotic- B. bifidum, while the upper band is not comparable to any treated groups in this study. Their pertinent demographic of the species within the reference ladder. All samples from 4 International Journal of Microbiology

BCDE F GH I JK LMNOPQR

B. breve B. breve B. longum B. longum

B. infantis B. infantis B. bifidum B. bifidum B. adolescentis B. adolescentis

(a) (b)

Figure 2: DGGE of bifidobacterial PCR-products (V3-region) from stool samples taken at eight weeks of age from infants treated with antibiotics B–I (a) and controls J–R (b). The mobility of the PCR products obtained in DGGE was compared to the PCR pattern of reference strains obtained with the same primer set.

the formula fed controls (K, L, M, Q, R) exhibited a band in (present in 15 out of 17 infants harbouring bifidobacteria) the same position as B. longum at four weeks of age, except (Figure 2). for one infant (subject J) (Figure 1(b)). In contrast, samples from the breast fed infants in the control group (N, O, P) 4. Discussion showed bands in the positions of B. infantis and B. bifidum. These two species were only found in one of the formula Bifidobacterium and Lactobacillus species are considered fed infants. While all vaginally born infants in the antibiotic- among the most important beneficial bacteria in the human treated group harboured bifidobacteria by four weeks of age, GIT and their presence in large numbers in the microbiota two out of five infants born by caesarean section harboured is regarded as beneficial [23, 24]. Suppression of these bifidobacteria (B. bifidum and B. longum) at this time point. beneficial components could facilitate growth of potentially pathogenic bacteria such as Klebsiella and Clostridium and have consequences for the well-being of the host. This study 3.2. Development of the Bifidobacterium Population at Eight demonstrates that even brief parenteral antibiotic treatment Weeks of Age. At eight weeks of age, eight out of nine alters the pattern of bifidobacterial evolution over time. antibiotic-treated infants and all controls had detectable Infants treated with antibiotics showed a reduction in both levels of bifidobacteria in their stools. Interestingly, the only colonisation and diversity of bifidobacteria compared with sample in which bifidobacteria was not detected was from an controls. While control and antibiotic-treated groups were infant that had received antibiotics for nine days (subject A) similar by eight weeks of age, a number of control subjects (the other infants were treated with antibiotics for 2–5 days). continued to display a more diverse population of bifidobac- Between four and eight weeks, an increase in bifidobacteria teria. Mackie et al. [25] reported that the colonisation of diversity in both groups was observed as evidenced by the bifidobacteria in infants is normally well established by one increased number of bands (Figure 2). Nonetheless, while week of age. It is tempting to speculate that the observed both groups were similar at eight weeks, samples from difference in detectable levels of bifidobacteria between the antibiotic-treated infants continued to display a less diverse antibiotic-treated infants and controls in our study reflects population of bifidobacteria when compared with controls the early antibiotic exposure. Favier et al. [26] previously (Figure 2). The samples obtained from the breast-fed infants described a significant difference in the microbiota of a single in the control group had in general a more diverse population infant receiving antibiotics compared to four nonantibiotic of bifidobacteria compared with formula-fed infants (7- treated infants, using similar methodology that is, PCR- 8 bands compared with 2–5 bands, resp.). However, no DGGE and Bifidobacterium-specific primers as used in our differences between the samples from the breast-fed and study. Bifidobacterium spp. were absent in the antibiotic- formula-fed infants in the antibiotic-treated groups were treated infant at all time points assessed, from five days to observed. Subject G exhibited a decrease in diversity from five months of age. However, in that study, the antibiotic- four to eight weeks of age. Moreover, the band with highest treated subject received prolonged antibiotic therapy with intensity at four weeks, identified as B. longum,wasreplaced both parenteral and oral agents (thirteen days of parenteral by another dominant strain, identified as B. breve,after coamoxiclav followed by oral cotrimoxazole) [26]. In con- eight weeks. Four of the five infants in the antibiotic-treated trast, the majority of participants in our study had relatively group born by caesarean section harboured bifidobacteria by brief periods of intravenous antibiotic administration and week eight. As found at week four, all vaginally born infants yet differences were still observed compared with controls. harboured bifidobacteria at eight weeks of age, with a higher B. longum was the dominant species in both the diversity observed compared with four weeks of age. Overall, antibiotic-treated infants and in the control group. This B. longum appeared to be the most commonly detected finding is consistent with the study by Gronlund¨ et al. Bifidobacterium species in the infants at eight weeks of life [27] which found B. longum to be the most common International Journal of Microbiology 5

Bifidobacterium species present in infants. B. longum was also Acknowledgments the most prevalent species in the antibiotic-treated infants. This result correlates with a recent finding by Matt¨ oetal.¨ This work was funded in part by Science Foundation Ireland [14], demonstrating that B. longum has a higher minimum (SFI) funds, Enterprise Ireland, and the Irish Government inhibitory concentration (MIC) for ampicillin compared under the National Development Plan 2000-2006. with other Bifidobacterium species. Thus, it appears that the antibiotic treatment itself may have selected for this species. References One infant showed a decrease in numbers of species from three species at four weeks of age to one species at eight [1] M. McDonald, A. Moloney, T. A. Clarke, and T. G. Matthews, weeks of age. The species that showed the highest intensity “Blood cultures and antibiotic use in a neonatal intensive care in the DGGE gel after four weeks was replaced by another unit,” Irish Journal of Medical Science, vol. 161, no. 1, pp. 3–4, dominant species after eight weeks. These results are in 1992. agreement with a study by Favier et al. [26] in which the [2] A. R. Spitzer, S. Kirkby, and M. Kornhauser, “Practice variation initially dominant bifidobacteria in two infants were replaced in suspected neonatal sepsis: a costly problem in neonatal following the emergence of new species within the study time intensive care,” Journal of Perinatology, vol. 25, no. 4, pp. 265– period. 269, 2005. Although less than half of the infants in this study were [3] R. Carbajal, A. Rousset, C. Danan et al., “Epidemiology and treatment of painful procedures in neonates in intensive care breast-fed, our study confirms previous reports that breast- units,” JournaloftheAmericanMedicalAssociation, vol. 300, feeding is positively associated with a more diverse bifi- no. 1, pp. 60–70, 2008. dobacterial population compared with formula-fed infants [4] I. Adlerberth, “Establishment of a normal intestinal microflora [16, 28, 29]. Bifidobacterium species are considered among in the newborn infant,” in Probiotics, Other Nutritional Factors the most important beneficial bacteria in the human GIT and Intestinal Microflora, L. A. Hanson and R. H. Yolken, Eds., [18, 19, 24]. A higher density of beneficial bacteria in Nestle` Nutrition Workshop Series, pp. 63–78, Lippincott- infants may confer a degree of protection against childhood Raven, Philadelphia, Pa, USA, 1999. diseases. For example, bifidobacterial colonisation has been [5] R. Bennet, M. Eriksson, and C. E. Nord, “The fecal microflora associated with a lower risk of atopy [21, 30, 31]. In a of 1-3-month-old infants during treatment with eight oral prospective study, children with lower numbers of fecal antibiotics,” Infection, vol. 30, no. 3, pp. 158–160, 2002. bifidobacteria and increased numbers of clostridia from the [6] J. Ballongue, “Bifidobacteria and probiotic action,” in Lactic firstweeksoflifeweremorelikelytobecomesensitizedto Acid Bacteria, Microbiology and Functional Aspects,S.Salmi- common allergens [23]. In addition, it has been reported nen and A. von Wright, Eds., pp. 519–587, Marcel Dekkers, that differences in the gut microbiota during infancy may New York, NY, USA, 2nd edition, 1998. precede the onset of obesity. In this study by Kalliomaki¨ et al. [7] O. C. Thompson-Chagoyan,´ J. Maldonado, and A. Gil, [32], the numbers of Bifidobacterium spp. were higher during “Colonization and impact of disease and other factors on intestinal microbiota,” Digestive Diseases and Sciences, vol. 52, infancy in children who remained at normal weight than no. 9, pp. 2069–2077, 2007. those in children who became overweight. Factors such as [8] R. M. Satokari, E. E. Vaughan, A. D. L. Akkermans, M. Saarela, breast-feeding, vaginal delivery, and the absence of antibiotic and W. M. De Vos, “Bifidobacterial diversity in human feces exposure would promote microbial diversity with higher detected by genus-specific PCR and denaturing gradient gel numbers of bifidobacteria. electrophoresis,” Applied and Environmental Microbiology, vol. In conclusion, the present study demonstrates that short- 67, no. 2, pp. 504–513, 2001. term parenteral antibiotic treatment of neonates causes a [9] J.-E. Germond, O. Mamin, and B. Mollet, “Species specific disturbance in the expected colonization pattern of bifi- identification of nine human Bifidobacterium spp. in feces,” dobacteria in the first months of life. Whether a delay in Systematic and Applied Microbiology, vol. 25, no. 4, pp. 536– bifidobacterial colonisation could adversely influence health 543, 2002. status, either in infancy or later life remains unknown. [10]R.Temmerman,L.Masco,T.Vanhoutte,G.Huys,and However, adequately powered prospective studies that doc- J. Swings, “Development and validation of a nested-PCR- ument changes in the microbiota and incorporate long- denaturing gradient gel electrophoresis method for taxonomic characterization of bifidobacterial communities,” Applied and term health status outcomes are needed to further our Environmental Microbiology, vol. 69, no. 11, pp. 380–385, understanding of the sequelae, if any, of early neonatal 2003. antibiotic administration. Until such data are available, [11] P. Marteau and F. Shanahan, “Basic aspects and pharmacology supplementation with bifidobacteria in this patient group of probiotics: an overview of pharmacokinetics, mechanisms cannot be justified. of action and side-effects,” Bailliere’s Best Practice and Research in Clinical Gastroenterology, vol. 17, no. 5, pp. 725–740, 2003. Abbreviations [12] A. C. Brown and A. Valiere, “Probiotics and medical nutrition therapy,” Nutrition in Clinical Care, vol. 7, no. 2, pp. 56–68, DGGE: Denaturing gradient gel electrophoresis 2004. GIT: Gastrointestinal tract [13] E. Vlkova,´ J. Nevoral, B. Jencikova et al., “Detection of MIC: Minimum inhibitory concentration infant faecal bifidobacteria by enzymatic methods,” Journal of NICU: Neonatal intensive care unit. Microbiological Methods, vol. 60, no. 3, pp. 365–373, 2005. 6 International Journal of Microbiology

[14] J. Matt¨ o,¨ A. H. A. M. van Hoek, K. J. Domig et al., detection methods,” Journal of Pediatric Gastroenterology and “Susceptibility of human and probiotic Bifidobacterium spp. Nutrition, vol. 30, no. 1, pp. 61–67, 2000. to selected antibiotics as determined by the Etest method,” [29] K. C. Mountzouris, A. L. McCartney, and G. R. Gibson, International Dairy Journal, vol. 17, no. 9, pp. 1123–1131, “Intestinal microflora of human infants and current trends for 2007. its nutritional modulation,” British Journal of Nutrition, vol. [15] I. H. Gewolb, R. S. Schwalbe, V. L. Taciak, T. S. Harrison, and 87, no. 5, pp. 405–420, 2002. P. Panigrahi, “Stool microflora in extremely low birthweight [30] B. Bjorkst¨ en,´ E. Sepp, K. Julge, T. Voor, and M. Mikelsaar, infants,” Archives of Disease in Childhood,vol.80,no.3,pp. “Allergy development and the intestinal microflora during the F167–F173, 1999. first year of life,” Journal of Allergy and Clinical Immunology, [16] J. Penders, C. Thijs, C. Vink et al., “Factors influencing the vol. 108, no. 4, pp. 516–520, 2001. ff composition of the intestinal microbiota in early infancy,” [31] S. Watanabe, Y. Narisawa, S. Arase et al., “Di erences in Pediatrics, vol. 118, no. 2, pp. 511–521, 2006. fecal microflora between patients with atopic dermatitis and [17] L. S. Collier-Hyams and A. S. Neish, “Innate immune healthy control subjects,” Journal of Allergy and Clinical relationship between commensal flora and the mammalian Immunology, vol. 111, no. 3, pp. 587–591, 2003. intestinal epithelium,” Cellular and Molecular Life Sciences, vol. [32] M. Kalliomaki,¨ M. C. Collado, S. Salminen, and E. Isolauri, ff 62, no. 12, pp. 1339–1348, 2005. “Early di erences in fecal microbiota composition in children [18] P. Kaufmann, A. Pfefferkorn,M.Teuber,andL.Meile,“Identi- may predict overweight,” American Journal of Clinical Nutri- fication and quantification of Bifidobacterium species isolated tion, vol. 87, no. 3, pp. 534–538, 2008. from food with genus-specific 16S rRNA-targeted probes by colony hybridization and PCR,” Applied and Environmental Microbiology, vol. 63, no. 4, pp. 1268–1273, 1997. [19] G. Muyzer, “DGGE/TGGE a method for identifying genes from natural ecosystems,” Current Opinion in Microbiology, vol. 2, no. 3, pp. 317–322, 1999. [20] T. Matsuki, K. Watanabe, R. Tanaka, M. Fukuda, and H. Oyaizu, “Distribution of bifidobacterial species in human intestinal microflora examined with 16S rRNA-gene-targeted species-specific primers,” Applied and Environmental Microbi- ology, vol. 65, no. 10, pp. 4506–4512, 1999. [21] E. Malinen, J. Matt¨ o,¨ M. Salmitie, M. Alander, M. Saarela, and A. Palva, “PCR-ELISA: II: analysis of Bifidobacterium popula- tions in human faecal samples from a consumption trial with Bifidobacterium lactis Bb-12 and a galacto-oligosaccharide preparation,” Systematic and Applied Microbiology, vol. 25, no. 2, pp. 249–258, 2002. [22] C. S. Hoffman and F. Winston, “A ten-minute DNA prepara- tion from yeast efficiently releases automomous plasmids for transformation of Escherichia coli,” Gene,vol.57,no.2-3,pp. 267–272, 1987. [23] M. Kalliomaki,¨ P. Kirjavainen, E. Eerola, P. Kero, S. Salminen, and E. Isolauri, “Distinct patterns of neonatal gut microflora in infants in whom atopy was and was not developing,” Journal of Allergy and Clinical Immunology, vol. 107, no. 1, pp. 129– 134, 2001. [24] C.-C. Chen and W. A. Walker, “Probiotics and prebiotics: role in clinical disease states,” Advances in Pediatrics, vol. 52, pp. 77–113, 2005. [25] R. I. Mackie, A. Sghir, and H. R. Gaskins, “Developmental microbial ecology of the neonatal gastrointestinal tract,” American Journal of Clinical Nutrition, vol. 69, no. 5, pp. 1035– 1045, 1999. [26]C.F.Favier,W.M.DeVos,andA.D.L.Akkermans, “Development of bacterial and bifidobacterial communities in feces of newborn babies,” Anaerobe, vol. 9, no. 5, pp. 219–229, 2003. [27] M.-M. Gronlund,¨ M. Gueimonde, K. Laitinen et al., “Maternal breast-milk and intestinal bifidobacteria guide the compo- sitional development of the Bifidobacterium microbiota in infants at risk of allergic disease,” Clinical and Experimental Allergy, vol. 37, no. 12, pp. 1764–1772, 2007. [28] H. J. M. Harmsen, A. C. M. Wildeboer-Veloo, G. C. Raangs et al., “Analysis of intestinal flora development in breast-fed and formula-fed infants by using molecular identification and