Genetic and Immunological Characterization of the Major Outer Membrane Protein (OmpS) of Legionella pneumophila.
by
Risini D. Weeratna
Submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy
at
Dalhousie University Halifax, Nova Scotia August, 19 95
©Copyright by Risini D. Weeratna, 1995 I I
National Library Bibliotheque nationale 1*1 of Canada du Canada Acquisitions and Direction des acquisitions et Bibliographic Services Branch des services bibliographiques
395 Wellington Street 395, rue Wellington Ottawa, Ontario Ottawa (Ontario) K1A0N4 K1A0N4 Your Me Voire reference
Our tile Notre reference
The author has granted an L'auteur a accorde une licence irrevocable non-exclusive licence irrevocable et non exclusive allowing the National Library of permettant a la Bibliotheque Canada to reproduce, loan, nationale du Canada de distribute or sell copies of reproduire, prefer, distribuer ou his/her thesis by any means and vendre des copies de sa these in any form or format, making de quelque in an i ere et sous this thesis available to interested quelque forme que ce soit pour persons. mettre des exemplaires de cette these a la disposition des personnes interessees.
The author retains ownership of L'auteur conserve la propriete du the copyright in his/her thesis. droit d'auteur qui protege sa Neither the thesis nor substantial these. Ni la these ni des extraits extracts from it may be printed or substantiels de celle-ci ne otherwise reproduced without doivent etre imprimes ou his/her permission. autrement reproduits sans son autorisation.
ISBN 0-612-15868-3
Canada I I
Name (?tsm i J>hctwtvM^f. Ukgyalfgi Dissertation Abstracts international is arranged by broad, general subject categories Pleats select the on-ject which most nearly describes the con'ent of your dissertation Enter the corresponding four-digit code in the spaces ovided MiC'C'Wlofly ^MB UMI SUBJECT TERM WWCTCOM
Subject Categories THE HUMANITIES AND SOCIAl SCIENCES
COMMUNICATION: AND THE ARTS PsychcAwy 0525 PHILOSOPHY, RELIGION AND Ancient 0579 Architecture 0729 Reading 0535 THEOLOGY Medieval 0581 Art History 0377 Religious 0527 Modern 0182 Philosophy 0422 Cinema 0900 Sciences 0714 Religion Black 0328 Donee 0378 Secondary 0533 African 0331 General 0318 Social Sciences 0534 Asia, Australia and Oceania 0332 Fine Arts 03 J7 Biblical Studies 0321 Information Science 0723 0334 Sociology of 0340 Clergy 0319 Canadian Journalism 0391 Special 0529 History of 0320 European 0335 Library Science 0399 Teacher Training 0530 0336 Philosophy of 0322 latin American Technology 0710 Mais Comrmmicationi 0708 Theology 0469 Middle Eastern 0333 MUSK 0413 Tests arKiMeosurerrtents 0288 United States 0337 Speech Communication 0459 Vocational 0747 SOCIAL SCIENCES History of Science 0585 Theater 0465 American Studies 0323 Law 0398 LANGUAGE, LITERATURE AND Anmropology Political Sc ence EDUCATION General 0615 LINGUISTICS Archoeofogy 0324 International Law and General 0515 Languag» Cultural 0326 Administration 0514 Relations 0616 General 0679 Physical 0327 Adult and Continuing 0516 Public Administration 0617 Ancient 0289 Business Administration 0814 Agricultural 0517 Linguistics 0290 General 0310 Recreation Art 0273 Modern 0291 Accounting 0272 Sociol Work 0452 Bilingual and Multicultural 0282 Sociology Literature Banking 0770 Business 0688 General 0401 General 0626 Management 0454 0627 Community College 0275 Classical 0294 Marketing 0338 Criminology and Penology Curriculum and Instruction 0727 Comparative 0295 Canadian Studies 0385 Demography 0938 Early Childhood 0518 Medieval 0297 Economics Ethnic and Racial Studies 0631 Elementary 0524 Mcdem 0298 General 0501 Individual and Family Finance 0277 African 0316 Agricultural 0503 Studies 0628 Guidance and Counseling 0519 American 0591 Commerce Business 0505 Industrial and Labor Health 0630 0629 Asian 0305 Finance 0508 Relations Higher 0745 Canadian [English) 0352 History 0509 Public and Social Welfare 0630 Kstoryof 0520 Canadian (French) 0355 Labor 0510 Social Structure and Home Economics 0278 English 0593 Theory 0511 Development 0700 Industrial 0521 Germanic 03' 1 Folklore 0358 Theory and Methods 0344 language and Literature 0279 Latin American 0312 Geography 0366 Transportation 0709 Mathematics 0280 Urban and Regional Planning Middle Eastern 0315 Gerontology 0351 0999 Music 0522 Romance 0313 History Women s Studies 0453 Philosophy of 0998 Slavic and East European 0314 General 0578 Physical 0523
THE SCIENCES AND ENGINEERING
BIOLOGICAL SCIENCES Geodesy 0370 Spe»ch Pathology 0460 Eng neermg Agriculture Geology 0372 Toxicology 0383 General 0537 General 0473 Geophysics 0373 Home Economics 0386 Aerospace 0538 Agronomy 0285 Hydrology 0388 Agricultural 0539 Animal Culture and Mineralogy 0411 PHYSICAL SCIENCES Automotive 0540 Nutrition 0475 Paleobotany 0345 0541 Pure Sciences Biomedical Animal Pathology 0476 Paleoecology 0426 0542 Chemistry Chem cal Food Science and Paleontology 0418 Civil 0543 General 0485 Technology 0359 Paleozoology 0985 Electronics and Electrical 0544 Agnculti -al 0749 Forestry ana Wildlife 0478 Parynoloav 0427 Heat and Thermodynamics 0348 Analytical 0486 Plant Culture 0479 Physical Geography 0368 Hydraulic 0545 Biochemistry 0487 Plant Pathology 0480 Physical Oceanography 0415 Industrial 0546 Inorganic 0488 Plant Physiology 0817 Marine 0547 Nuclear 0738 Range Management 0777 HEALTH AND ENVIRONMENTAL Materials Science 0794 Organic 0490 Wood Technology 0746 SCIENCES Mechanical 0548 Biology Pharmaceutical 0491 0743 Environmental Sciences 0768 Metallurgy General 0306 Physical 0494 0551 Health Sciences Mining Anatomy 0287 Polymer 0495 0552 General 0566 Nuclear B ostatistics 0308 Radiation 0754 0549 Audiology 0300 Packaging Botany 0309 Mathematics 0405 Petroleum 0765 Chemotherapy 0992 Physics 0379 Sanitary and Municipal 0554 Dentistry 0567 •oentral 0605 0790 Ecology 0329 System Science Education 0350 Acoustics 0986 0428 Entomology 0353 Geotechnalogy 0369 Hospital Management 0769 Astronomy and 0776 Genetics Human Development 0758 Operations Research Limnology 0793 Astrophysics 0606 0795 Immunology 0982 Plashes Technology Microbiolcgy 0410 Atmospheric Science 0608 0994 Medicine and Surgery 0564 Textile Technology Molecular 0307 Atomic 0748 Neuroscience 0317 Mental Health 0347 Electronics and Electricity 0607 Nursing 0569 Elementary Particles and PSYCHOLOGY Oceanography 0416 Genera! 062. Nutrition 0570 High Energy 0798 Physiology 0433 Behavioral 0384 Obstetrics and Gynecology 0380 Fluidand Plasma 0759 Radiation 0821 Clinical 0622 0778 Occupational Health ana Molecular 0609 Veterinary Scienre Developmental 0620 Zoology 0472 Therapy 0354 Nuclear 0610 Ophthalmology 038' Experimental 0623 Biophysics Optics 0752 Industr al 0624 Pathology 0571 Radiation 0756 General 0786 Personality 0625 Pharmacology 0419 Solid State 0611 MeHicol 0760 Physiological 0989 Pharmacy 0572 Stah-hcs 0463 Psychobiology 0349 EARTH SCIENCE' Physical therapy 0382 0632 Public Health 0573 Applied Sciences Psychometrics Biogeorhemistry 0425 Applied Mechanics 0346 Social 0451 0996 Rad ology 0574 Geochemistry Recreation 0575 Computer Science 0984 TABLE OF CONTENTS
LITERATURE REVIEW , 1
Legionella; morphology and growth 3
The environment 5
Interaction with other microbes 7
Route of infection 9
Clinical manifestations 11
Laboratory diagnosis 13
The disease process , . 14
Host defence against L. pneumophila infection 21
Role of phagocytosis 23
Cell mediated immunity in Legionnaires' disease 25
Immunodominant antigens of L. pneumophila 31
Genetics of Legionella virulence 33
Identifying genes by mutagenesis 35
Identifying genes by complementation of
avirulent mutants 41
Salt resistance and Legionella virulence 45
Cloning genes using functional properties
of the gene product 46
iv i
Zinc metalloprotease 46
Legiolysin 51
Cloning antigens expressed on the cell surface . 52
Macrophage infectivity potentiating
protein 53
Peptidoglycan associated protein 55
Heat shock protein 55
Taj or outer membrane protein 5 8
Other proteins implicated in L. pneumophila
virulence 5 8
Phospholipase C 59
Phosphatase 60
Protein Kinase 61
Outer membrane porin proteins of gram negative
bacteria 62
Major outer membrane protein of L. pneumophila 65
Coordinate regulation of virulence 69
Statement of research objectives 79
MATERIAL AND METHODS 81
Bacterial strains and cloning vectors 81
Extraction of chromosomal DNA 8 5
v Southern blot hybridization assays 87
Colony Southern hybridization 90
RNA extraction 92
Northern blot hybridization asssys 94
Cloning and sequencing of ompS 96
Primer extension reaction 97
Extraction of plasmid DNA 101
Transformation of plasmid DNA into bacterial cells . 103
a: Preparation of competent cells 103
b: Transformation 104
Selective radiolabelling of OmpS expressed in
E. coli JF626 104
Sodium chloride challenge and preparation of
L. pneumophila cell extracts 107
Gel mobility shift assays 108
Southwestern blot assays 112
ompS promoter:lacZ fusions 115
(3-galactosidase assays 117
DNA sequencing 117
Double stranded DNA sequencing 119
Single stranded DNA sequencing 119
a: Preparing replicative forms (RF) of
vi single stranded M13 phage 121
b: Cloning into RF DNA and preparation of
ss template DNA for sequencing 123
Sequencing gel electrophoresis 125
DNAsel footprinting assay 129
Attempt to clone the gene coding for the DNA
binding protein 129
a: Conventional genetic approach 13 3
b: Purification of DNA binding protein ... 138
Immunological characterization of OmpS 13 8
Guinea pigs 13 9
Skin testing 140
Lymphocyte proliferation assays 141
Human Studies 142
Other immunological procedures 143
Guinea pig vaccination studies 146
RESULTS 147
Cloning and sequence analysis of ompS 151
Sequence analysis and primer extension studies 156
Sequence relationship of ompS in legionellae 161
Selective radiolabelling of OmpS 168
vii I
Allelic exchange mutagenesis of ompS 171
Regulation of ompS in L. pneumophila 175
Gel mobility shift assays 185
Effect of other salts on the mobility
shift of ompS DNA 188
Identification of DNA binding site of OmpT .... 183
Effect of sodium chloride on OmpT 196
Southwestern blot assays 200
DNAsel protection assay 208
ompS promoter:lacZ fusions 212
Cloning and regulation of ompT 218
a: Genetic approach 219
b: Purification of OmpT 225
Immunological characterization of OmpS 231
Guinea pig studies 231
Cellular immune responses of immune guinea
pigs to L. pneumophila protein antigens... 231
Immune responses to OmpS 236
The \ nccine trial 23 8
Cellular immune responses to Legionella
antigens in humans 246
DISCUSSION 252
viii
I I • I
Genetic characterization of ompS 252
Cloning and sequencing of oiTipS and primer
extension analysis 252
Expression of ompS in E. coli 256
Mobility shift and southwestern blot assays ... 259
Identification of OmpT binding sites on ompS
DNA 264
Cloning the gene coding for the DNA binding
protein OmpT 2 65
Concluding remarks 278
Immunological characterization 281
APPENDIX 289
M9 minimal medium 289
Triethylamineacetate buff v. - 2 91
Publications 292
BIBLIOGRAPHY 312
ix LIST OF FIGURES
Figure 1. Restriction endonuclease maps of H151
and H246 in pBluescript 149
Figure 2. Primer extension analysis of the
transcription start of ompS 152
Figure 3. Sequence of ompS promoter region .... 154
Figure 4. Southern blot analysis of EcoRI (A)
and Hindlll (B) restricted DNA from selected
Legionella species and serogroups of
L. pneumophila 158
Figure 5. Schematic map of PT7-5, an expression
vector using T7 RNA polymerase 164
Figure 6. Expression of OmpS using the T7 RNA
polymerase/promoter system 166
Figure 7. Allelic exchange mutagenesis of ompS.. 169
Figure 8. Northern blot of L. pneumophila Svir
RNA 173
Figure 9. 285 bp ompS promoter amplicon
generated by polymerase chain reaction 178
Figure 10. Gel retardation of ompS promoter DNA i. !
using cell extracts from L. pneumophila Svir
(V) and Avir (M), unchallenged or challenged
with NaCl 180
Figure 11. Binding competition assay 183
Figure 12. Gel retardation of ompS promoter DNA
using cell extracts from L. pneumophila Svir
(V) strains, unchallenged (-) or challenged
with either NaCl or KC1 186
Figure 13. Gel retardation to identify the
region of binding of OmpT to ompS DNA 19 0
Figure 14. Schematic map to illustrate the
location of oligonucleotide primers F7,
F8, and R6 192
Figure 15. Using truncated ompS promoter DNA
amplicons in mobility shift assays to
identify the DNA binding site of OmpT 194
Figure 16. Kinetics of the NaCl effect on OmpT . 197
Figure 17. Southwestern blot assays using
virulent L. pneumophila cell extracts 2 03
Figure 18. Southwestern blot assays using
virulent (V) and avirulent (M)
L. pneumophila cell extracts unchallenged or
XI challenged with 0 . 85% NaCl 205
Figure 19. DNAsel footprinting of ompS promoter
DNA to identify the OmpT binding site(s) ... 210
Figure 20.Construction of ompS promoter:lacZ
fusion 214
Figure 21. Gel retardation of ompS promoter DNA
by E. coli cell extracts . . . . , 223
Figure 22. Protein elusion profile of
L. pneumophila cell extracts purified through
a DE-52 ion exchange chromatography 227
Figure 23. Identification of the OmpT fraction
from DE-52 column eluates using mobility
shift assays 229
Figure 24. Survival of vaccinated guinea pigs
infected \;ith 2.5 X LD50 dose of L.
pneumophila administered by the
intratracheal route 241
Figure 25. Weights of vaccinated guinea pigs
challenged with 2.5 X LDS0 dose of L.
pneumophila administered by the
intratracheal route 243
Figure 26. Scatter graph of human LPRs to L.
xii pneumophila antigens 247
Figure 27. Proposed model for regulation of ompS 270
xiii I
LIST OF TABLES
Table 1. Bacterial strains and cloning vectors
used in this study 82
Table 2. ompS promoter activity in L. pneumophila . 217
Table 3. LVgalactosidase activity of possible
ompT clones 222
Table 4. DTH reactions to purified protein antigens
in guinea pigs surviving L. pneumophila
infection 234
Table 5. LPRs of guinea pigs surviving legionellosis 23 5
Table 6. LPR and antibody titers of OmpS-vaccinated
guinea pigs to OmpS and Hsp60 antigens ... 237
Table 7. ELISA results for immunized guinea pigs
prior to infection 245
Table 8. LPRs of incubated (I) Vs unincubated (UI)
lymphocytes from a heart transplant
patient 5 months after the onset of
infection 250
xiv I
ABSTRACT
L. pneumophila, a facultative intracellular bacterium, is the causative agent of Legionnaires' disease. Its natural habitat is aquatic. When the bacterium is inhaled by a susceptible individual, it is phagocytosed by alveolar macrophages in which it can grow and multiply. OmpS is one of the most abundant proteins synthesised by this bacterium. It is known to help the bacterium to become intracellular. When virulent (V) cells are intracellular, synthesis of OmpS is decreased. The same phenomenon is seen when V bacteria are suspended in a medium containing high levels of sodium chloride. One of the goals of this study was to investigate the regulation of ompS. The results reveal that there is a transcription factor (OmpT) involved in the regulation of ompS. OmpT itself appears to be regulated by sodium chloride levels. When the bacterium is in a high salt environment (human lungs as opposed to fresh water), there is a decrease in OmpT levels resulting in a decrease in ompS transcription. OmpT is a highly unstable 15 kDa protein which is constantly synthesised by the bacterium. In avirule ;: bacteria, OmpT is not regulated by sodium chloride levels. This study has also looked at the potential of OmpS to elicit a cellular immune response against L. pneumophila. The results show that OmpS is able to induce protective immunity in guinea pigs against Legionnaires' disease. In addition, humans surviving legionellosis also demonstrate a strong cellular immune response to OmpS, suggesting that OmpS would be a good candidate for developing a vaccine against L. pneumophila infections.
xv i
ABBREVIATIONS
APS Ammonium persulfate
BCYE Buffered charcoal yeast extract
BYE Buffered yeast extract
CFU Colony forming units
DEPC Diethyl pyrocarbonate
DTH Delayed type hypersensitive responses
DTT Dithiothreitol
IFN-Y Interferon-gamma
IL Interleukin
LAK Lymphokine-activated killer cells
LPR Lymphocyte proliferative responses
MHC Major histocompatibility complex
Mip Macrophage infectivity potentiating protein
MOI Multiplicity of infection
MOMP Major outer membrane protein
NK Natural killer cells
PBS Phosphate buffered saline
PMN Polymorphonuclear leukocytes
PMSF Para-methyl sulfonyl fluoride
Ppl Peptidoglycan associated protein
xvi I
SDS-PAGE Sodium dodecyl sulphate-polyacrylamide gel
electrophoresis
TEMED Tetramethylethylenediamine
TNF-a Tumoi necrosis factor-alpha
TNF-|3 Tumor necrosis factor-beta ACKNOWLEDGEMENTS
I would like to extend my deepest gratitude to my supervisor, Dr Paul S. Hoffman for his guidance, support and encouragement throughout my career as a Ph.D. student. I also wish to express my gratitude to my supervisory committee Drs.
Gregory Bezanson, Scott Halperin, David Hoskin and Thomas J.
Marrie for their support and guidance. A special thanks go to all the members of the Hoffman Laboratory, past and present, for their support, encouragement and all the good times we have had. I also wish to thank all the members of the
Department of Microbiology & Immunology for their support and kindness throughout the past four years.
A special thanks also go to my husband, Nalaka for all his love, support and patience without which the past few years would have been dreadful. I thank my daughter, Shalini for the great joy she has brought into my life. Last but not least I would like to thank my parents for their love, and encouragement which helped me to get through many obstacles in life.
xviii I I
I also acknowledge the financial support of the Faculty
of Graduate Studies, Nova Scotia Medical Research Foundation
and the Izaak Walton Killam Memoria], Scholarship fund.
xix kJ
LITERATURE REVIEW
In 1976, while public health officials anxiously awaited the swine influenza that killed 2 0 million people in
1918, no one was prepared for the epidemic of different etiology that struck the Pennsylvania chapter of the
American Legion that summer.
In July of 1976, the Bellevue-Stratford hotel in
Philadelphia hosted the 58th annual convention of the
Pennsylvania chapter cf the American Legion. Over 450 0 people attended this convention. Following the convention, nearly 200 of the participants fell sick with an acute respiratory illness to which 3 8 deaths were attributed. At this point no one was aware of the causative agent for this respiratory illness. However, this mysterious epidemic resembling viral pneumonia caused enormous concern amongst the scientific community and considerable public interest resulted from the sensationalistic media coverage. For example, the cover of the August 16th issue of the News Week magazine carried the headline "mystery of the killer fever".
1 M
2
With a tremendous amount of effort put forth by the
Centers for Disease Control (CDC), and with the help of the
Pennsylvania health department, in January of 1977, the causative agent for this mystery killer was identified and the bacterium was named Legionella pneumophila in memory of the American Legionnaires who had died (McDade et. al.,
1977) . A retrospective serological analysis of pneumonias of unknown etiology indicated that this was not really a newly emerging pathogen, but that members of the genus Legionella were responsible for sporadic cases of pneumonia as far back as 25 years earlier (McDade et. al., 1979). After 1976, outbreaks of Legionnaires' disease began to appear world wide; some of which occurred in Burlington, Vermont (Broome et. al. 1979), Los Angeles, California (Haley et. al.,
1979), Memphis, Tennessee (Dondero et. al., 1980), Madrid,
Spain (Bouza and Rodriguez-Creixems, 1984), and Stafford,
England (Rashed et. al., 1986). The first identified case of
Legionnaires' disease in Canada was reported in a man from
New Brunswick (Bennett, 1978). A study carried out in Nova
Scotia showed that 2% of all community-acquired pneumonia cases treated at the Victoria General Hospital in Halifax were caused by L. pneumophila (Marrie et. al., 1989). L.
I 3 pneumophila is now recognized as a significant nosocomial pathogen in hospitals throughout the world (Blatt et. al.,
1993) .
A considerable amount of information has been gathered about this bacterium; yet, detailed information about the pathogenesis of Legionella still remains a mystery.
Legionella'. Morphology and Growth :
Legionella sp. are Gram negative, strictly aerobic, non-sporulating rods that are 0.3-0.9 jxm in width and 2-20 jitm or more in length (Brenner et. al. , 1979) . Except for L. oakridgenesis, L. brunesis, L. Cincinnati ensis, and L. longbeachae serogroup 1, other species are motile by one, or two polar or lateral flagella (Chandler et. al, 1980;
Rodgers et. al., 1980; Bornstein et. al., 1991). These bacteria are chemoorganotropic, utilizing amino acids as their carbon and energy sources. The amino acids are catabolized via the Krebs cycle while the necessary sugars are synthesized by the gluconeogenic enzymes of the Embden-
Myerhof-Parnas pathway (Keen and Hoffman, 1984; George et. al., 1980) . Carbohydrates are neither fermented nor oxidized 4 by Legionellae. Also, they do not reduce nitrate or degrade urea and are weakly positive for the oxidase test (Brenner et. al., 1979).
Under laboratory conditions, Legionellae are rather fastidious organisms and thus do not grow on standard bacteriological media. Mueller-Hinton agar supplemented with
Iso VitaleX was the first culture medium to be used to grow
Legionella in the laboratory (Feeley et. al., 1978). However the subsequently developed, charcoal yeast extract agar, buffered to pH 6.9, and supplemented with L-cysteine and ferric pyrophosphate is the primary medium used for isolation and culture of Legionella (Feeley et. al., 1979).
Addition of ACES (2-{(2-amino-2-oxoethyl)-amino} ethanesulfonic acid) buffer (Pasculle et. al., 1980) and alpha- ketoglutarate (Edelstein 1981) to charcoal yeast extract medium have further improved the growth of these bacteria. Hoffman et. al. (1983) demonstrated that the principle role of charcoal in this medium is to scavenge toxic oxygen radicals that are produced when yeast extract agar is exposed to light. Subsequently, Pine et. al. (1986) showed that a-keto acids have a similar role. However, even 5 on the medium of choice(buffered charcoal yeast extract agar), the growth of Legionellae is relatively slow, taking approximately 4-7 days at 37°C for visible colonies to appear.
The Environment :
Paradoxically, the natural habitat of Legionella is aquatic, where the nutrient levels and the general growth conditions are expected to be poor. They have been isolated from both natural aquatic environments, i.e. rivers, lakes, streams and thermally polluted waters(Nguyen et. al., 1991) as well as from cooling towers, hot-water systems, whirlpool spa baths, clinical humidifiers in respiratory equipment, and a supermarket vegetable sprayer (Lee and West, 1991). In an outbreak of Legionnaires' disease in Australia where L. longbeachae was identified to be the causauire agent, this organism was isolated from potting soil compost used by the patients, who were all avid gardeners (Steel et. al. , 1990) .
In the aquatic environment, Leoionella have been isolated from water temperatures rai. ^ng from 5.7°C to 63"C.
However, significantly more strains have been isolated from . I
6 waters of 36°C to 63°C than from waters of 0°C to 36°C
(Fliermans et. al., 1981). Contrary to the temperature effect, in this study, Fliermans et. al. have not observed any correlation between the incidence of legionellae and other physical properties of water such as conductivity, pH, dissolved oxygen, turbidity, etc. Investigating a disease of unknown etiology among workers exposed to lakes and streams in the Mount St. Helens blast zone in the USA, Tison et. al., (1983) reported the highest counts of Legionella from lakes receiving water from hydrothermal seeps. They also reported the isolation of a new Legionella species; L. sainthelensi as well as L. pneumophila serogroups 4 and 6 from thermally heated lakes within the blast zone.
In a study, looking at the growth of L. pneumophila in potable water systems, Niedveld et. al. (1986) reported that the growth was stimulated by plumbing fixtures containing rubber compounds that did not contain thiuram. In a similar study, a variety of metals were shown to be toxic to legionellae in water but some, such as iron and zinc, were found to be stimulatory for their growth (States et. al,
1985) . - I •Mi
7
Legionellae have also been found to be more resistant
to chlorine than enteric bacteria. In a study by Kuchta et.
al. (1983), 99% of E. coli were shown to be killed by less
than one minute exposure to 0.1 mg of free chlorine residual
at 21°C, pH 7.6 whereas the death of 99% of L. pneumophila
required 40 min under the same conditions.
Interactions with other microbes :
The resistance of Legionella to these harsh
environmental conditions may be explained by the observation
that these bacteria can maintain a commensal association
with other aquatic microorganisms. The first evidence for
symbiosis between Legionella species and other organisms was
presented by Tison et. al. (1980) where they isolated L.
pneumophila from an algal mat in thermally polluted water. A
similar study shows that, removal of algae from culture
would result in a decrease in number of viable Legionella sp
(Bohach and Snyder, 1983). Rowbotham (1980) was the first to
report that Legionellae can infect amoebal trophozoites and
then become incorporated in their cysts. Legionella also
have been shown to infect, and depending on the conditions,
to coexist for prolonged periods in the species of Acanthamoeba, Hartmanella, Valkampfia and Naegleria
(Rowbotham, 1980, 1986; Tyndal and Domingue, 1982; Anand et. al., 1983; Wadowsky et. al., 1988). Apart from the amoebae,
Legionella have also been shown to infect the ciliated protozoans, Tetrahymena and Cyclidium (Fields et. al., 1984;
Barbaree et. al., 1986; King et. al., 1988). Studies by several investigators (Wadowsky and Yee, 1983; Stout et. al,
1985) have shown that in complex media lacking the necessary ingredients for Legionella growth such as cysteine and iron salts, L. pneumophila forms satellite colonies around some common aquatic bacteria including strains of Flavobacterium,
Pseudomonas, Alcaligenes and Acinetobacter. However, no one has so far been able to demonstrate that other bacteria can support the growth of Legionella in simple media or water.
Many genera of amoebae have been shown to form cysts that can withstand environmental temperature extremes
(Kilvington, 1989), chlorination (King et al., 1988), desiccation and other adverse conditions. Therefore
ingestion by protozoa offers a possible mechanism by which
Legionella can persist through adverse conditions such as
extreme temperatures and chlorination (Dowling et. al, 1992) . Studies by Kuchta et. al. (1984) have shown a six- nine fold increase in chlorine resistance in L. pneumophila grown in low-nutrient tap water as opposed to those grown on agar medium. Also another study by Kilvington and
Price(1990) showed that L. pneumophila living intracellularly in amoebal cysts of Acanthamoeba polyphaga can withstand concentrations of free chlorine more than 100 times greater than the level that would kill free living L. pneumophila. Therefore the present day studies suggest that
Legionella are ingested by free-living aquatic amoebae. Once ingested Legionella multiply within the endocytic vacuoles of these amoebae, thus making ^hese amoebae an amplification source for Legionella in the environment (Dowling et. al.,
1992). Furthermore it is possible that amoebae containing vesicles loaded with Legionella are present in the aerosol drifts from contaminated aquatic environments and thus serve as vehicles of transmission of infectious legionellae to humans (Dowling et. al, 1992).
Route of infection :
The available evidence to date suggests that the mode of transmission of Legionella to humans occurs by
I 10 aerosolization, aspiration or by direct instillation into the lung during respiratory tract manipulation (Nguyen et. al., 1991). The airborne spread of Legionella was strongly evident in the Pontiac fever outbreak that occurred in a building of the county health department in Pontiac,
Michigan in 1967 (McDade et. al, 1977). In this instance, the central air conditioning of the building was found to be contaminated by aerosols from an evaporative condenser.
Aerosolization experiments carried out at the Centers for
Disease Control in Atlanta, following the Pontiac outbreak, showed that the guinea pigs exposed to an aerosol of evaporative condenser water developed bronchopneumonia. In the same experiments, animals exposed to the same water that had been either filter sterilized or autoclaved showed no illness (Kaufmann et. al, 1981). During an epidemiological analysis of a cluster of Legionella infections that occurred in a renal transplant unit at the Radcliffe Infirmary in
Oxford, L. pneumophila was isolated from the shower heads and the shower aerosols were assumed to be the means of dissemination of the bacterium (Tobin et. al., 1980).
Further, dissemination of Legionella bacteria by aerosolization has been associated with humidifiers (Woo et. 11
al., 1986) and nebulizers (Mastro et. al., 1991). Other than
this, the rinsing of respiratory apparatus and tubing used
in ventilation machines with tap water contaminated with L. pneumophila has been shovm to cause infection as a result of
direct instillation of bacteria into the lung (Woo et. al.,
1986). Lowry et al (1991) have reported instances where patients developed sternal wound infections with Legionella
following cardiac surgery. In these cases patients were
infected with either L. dumoffi or L. pneumophila as a
result of direct contamination of the sternal wounds by use
of contaminated tap water for bathing of patients or dressing changes.
Clinical Manifestations :
Two distinct clinical syndromes have been associated with Legionella infections; the non-pneumonic, Pontiac
fever, and the pneumonic, Legionnaires' disease (Winn,
1988) .
Pontiac fever is an acute, self-limiting flu-like
illness with an incubation period of 48 hours and an attack
rate greater than 90% (Kaufmann et. al, 1981; Nguyen et. 12 al., 1991). Despite the development of a nonproductive cough, no pneumonia is seen in patients with Pontiac fever.
The predominant symptoms of this disease are malaise, myalgia, fever, chills and headache. The patients usually recover fully within a few days with no antibiotic therapy
(Nguyen et. al., 1991).
In contrast, Legionnaires' disease results in an acute purulent pneumonia with possible extrapulmonary manifestations such as sinusitis, perirectal abscess, pericarditis, endocarditis, pyelonephritis, peritonitis and pancreatitis (Nguyen et. al., 1991). The incubation period
for Legionnaires' disease can range from 2 to 10 days with disease symptoms including malaise, myalgia, anorexia,
shaking chills, and bradycardia. Patients are often febrile with temperatures in excess of 40°C. Gastrointestinal manifestations such as nausea, vomiting and watery diarrhea have also been reported in patients with Legionnaires'
disease. The most common neurological symptom associated with Legionnaires' disease is the change in mental status of
the patients. However, other manifestations such as ataxia 13 and encephalopathy have also been reported (Nguyen et. al.,
1991).
The chest radiographs of patients suffering from
Legionnaires' disease often reveal patchy interstitial infiltrates at an initial stage. The extensions of these infiltrates could later result in consolidation in one or more lobes. Lung abscesses are frequently observed in autopsy specimens from fatal cases of Legionnaires' disease
(Winn and Myerowitz, 1981).
Laboratory Diagnosis :
The single most important test for Legionnaires' disease still remains the culture of the organism on selective media. In order to increase the chances of isolation of the organism from sputum samples, multiple selective media are used in the laboratory. The culture medium is made more selective by the addition of antimicrobial agents to suppress the competing microflora of the sputum. Apart from that, the respiratory tract specimens are pre-treated with acid in order to enrich for Legionella bacteria. Legionella are more acid tolerant than most other I
14 bacteria found in this environment. Also, certain dyes are added to the culture media to enhance the visibility of
Legionella colonies since Legionella can take up these dyes preferentially (Vickers et. al., 1981).
In addition to culturing the organism, serological and molecular diagnostic techniques are also used for laboratory diagnosis of L. pneumophila.
The Disease Process :
As mentioned earlier, infection of human lungs by L. pneumophila is thought to occur either by aerosolization or by direct deposition of the bacteria during treatment regimens for patients requiring respiratory tract manipulations. There has so far been no evidence of
Legionella colonizing the oropharynx, which confirms direct inhalation as possible mode of entry of Legionella into human lung.
Upon entry into the lung, L. pneumophila encounters its host, the alveolar macrophage. The mechanism by which the bacteria attach and gain entry into the host cells is as yet 15 a controversial issue. According to Payne and Horwitz
(1987), the attachment of the bacteria to alveolar macrophages is mediated by the complement receptors CRl and
CR3. Such complement mediated phagocytosis has been demonstrated for other intracellular pathogens such as
Mycobacterium tuberculosis and Leishmania donovani (Payne et. al., 1987, Blackwell et. al., 1985). As demonstrated by
Bellinger-Kawahara and Horwitz (1990), L. pneumophila binds the complement components C3b and C3bi to its major outer membrane protein, OmpS, and gains entry into the alveolar macrophages via the complement receptors, CRl and CR3 located on the surface of these cells. However, studies by the same group of investigators have shown that even in the presence of non-immune serum, pre-treatment of the monocytes with monoclonal antibodies raised against the complement receptors CRl and CR3 fails to completely inhibit the binding of L. pneumophila. Under these conditions, the inhibition of binding was only 75% (Payne and Horwitz,
1987). Contradicting the findings by Payne and Horwitz,
Hushman and Johnson (1992) demonstrated that complement exhibited no enhancement in adherence of L. pneumophila to monocytes in the absence of specific antibodies. According 16 to their data, using either 20% of normal human serum or treating the organisms with normal guinea pig complement serum showed no increase in adherence of L. pneumophila to guinea pig peritoneal macrophages, U937 human monocyte cell line or the J774 mouse macrophage cell line. Therefore these data lead to the hypothesis that in the early stages of infection, prior to the development of a specific humoral response, L. pneumophila may be phagocytosed by complement fixation. However, later on in the infection, specific antibodies play a significant role in the uptake of L. pneumophila into alveolar macrophages. However, it is interesting to note that HeLa cells which lack both Fc and complement receptors are invaded by L. pneumophila. In a more recent report, Gibson et. al. (1994) proposed an opsonin-independent mode of attachment of L. pneumophila to host cells. According to this report L. pneumophila bears an as yet uncharacterized protein adhesin that is responsible for the initial interaction between the bacterium and its host cell in the lung.
A 24 kDa surface protein of L. pneumophila designated as Mip (Macrophage infectivity potentiating protein) was I
17
identified in 1989 by Cianciotto et. al, as being important
for the infectivity of the macrophages by the bacterium. The exact mechanism of action of this protein has not yet been elucidated. However this protein has not been implicated as an adhesion molecule of L. pneumophila. Looking at the phenotypic modulation of L. pneumophila following the entry into macrophages, Abu Kwaik et. al. (1993) have shown that the expression of at least 35 proteins are induced while 32 others are repressed during the exponential growth of the bacteria in U937 cells. In a subsequent study, Abu Kwaik and
Engleberg (1994) identified a 19 kDa L. pneumophila protein that is induced by macrophage infection and a variety of other stress responses. Thus this protein was named the global stress protein.
In unravelling the specific interactions between L. pneumophila and its host cell, studies have generally concentrated on the role of the bacteria. Studies dealing with this interaction from the host's point of view are now starting to elucidate the alveolar macrophage's responses to invasion by L. pneumophila. One such report by Yamamoto et. al. (1992) demonstrated the phosphorylation of a 76 kDa 18 protein in alveolar macrophages in response to the attachment of L. pneumophila. This phosphorylation is selectively associated with infection by virulent strains.
Avirulent strains of L. pneumophila, other gram negative bacteria and macrophage-activating substances such as lipopolysaccharides, and phorbol esters failed to induce phosphorylation of this protein. Therefore, this study suggests that phosphorylation might be associated with a specific cell surface molecule that is present only in virulent L. pneumophila.
Following the initial attachment, L. pneumophila is readily phagocytosed by alveolar macrophages, frequently but not exclusively by a unique process called coiling phagocytosis (Horwitz, 1984). Coiling phagocytosis has also been shown in other microorganisms such as L. donovani
(Chang, 1979). As described by Horwitz (1984), in coiling phagocytosis, a single pseudopod extends out from the phagocyte and coils around the bacterium engulfing it in
coiled vesicle. Treatment of the bacteria with anti-L. pneumophila antibodies abolishes the coiling process and
results in conventional phagocytosis in which phagocytic L I
19 pseudopods symmetrically surround the bacterium until the tips meet, fuse and enclose the organism within a membrane- bound vesicle (Horwitz. 1984).
Approximately 4-8 hours post infection of L. pneumophila into alveolar macrophages, a unique and a complex sequence of cytoplasmic events take place in which host cell ribosomes, smooth vesicles and mitochondria are recruited to the phagosome membrane encasing the bacteria
(Horwitz, 1983b). The bacteria abrogate the acidification of the phagosome and the fusion of the phagosome with primary or secondary lysosomes (Horwitz, 1983c, 1984). The mechanism by which they abrogate phagolysosomal fusion is not known.
By evading the microbicidal action of the macrophage, L. pneumophila is able to multiply within the phagosome, resulting in the eventual lysis of the host cell.
Mycobacterium tuberculosis, M. microti, Chlamydia psittaci, and Toxoplasma gondii are examples of other intracellular pathogens that are capable of abrogating phagolysosomal fusion (Amstrong and D'Arcy, 1971, Lowrie et. al., 1975,
Friis, 1972, Jones and Hirsch, 1972). According to Horwitz
(1984), formalin-killed L. pneumophila are also phagocytosed 20 by alveolar macrophages by coiling phagocytosis. However, the complex cytoplasmic sequelae that follows the infection is absent in this case and the phagosome fuses with the lysosome. In a similar study, Horwitz (1987) has characterized an avirulent mutant of L. pneumophila (25D) obtained by serial passage of a wild-type strain on Mueller-
Hinton agar for ics intracellular survival. This mutant was phagocytozed into human blood monocytes by coiling phagocytosis. However, the mutant neither formed a distinctive phagosome lined with ribosomes nor inhibited phagolysosomal fusion. The mutant was not capable of intracellular multiplication within the monocytes.
Therefore the intracellular fate of L. pneumophila appears to be influenced by factors other that those that determine its mode of entry (Horwitz, 1984).
A recent study, analyzing the avirulent L. pneumophila mutant 25D (which enters macrophages in a manner similar to the wild type yet has lost the ability to recruit host cell organelles and to multiply within the phagosome) has characterized an 8.2 kb region of DNA that complements this mutation (Berger and Isberg, 1993; Berger et. al., 1994). I
21
This genetic locus has been named dot (defect in organelle
trafficking). A more extensive discussion of this genetic
locus appears later under Genetics of Virulence.
As described by Mekalanos (1992), disease is simply a manifestation of the complex interactions between the host
and the parasite. In these interactions, the parasites' priority would be on survival and multiplication while the hosts' priority would be on eradication of the intruder.
Therefore in talking about disease, it is important to discuss both; how does the host defend against the infection by L. pneumophila and what helps L. pneumophila to counter
these defences and establish its niche within the host.
Host defence against L. pneumophila infection :
The incidence of Legionnaires' disease depends on the degree of contamination, the susceptibility of the patient and the intensity of exposure (Nguyen et. al., 1991). The major risk factor for Legionnaires' disease is immunosuppression (Skerrett et. al., 1989; Nguyen et. al.,
1991). This includes old age, excessive cigarette smoking, alcohol abuse, and pharmacologically-induced 22 immunosuppression following anti-cancer treatment,
transplant surgery, etc. Legionnaires' disease among people
with an intact immune systems is uncommon.
Usually, healthy individuals resist microbial
infections by many different mechanisms. Some of these
mechanisms are present prior to exposure to the invading
microorganism. These defence strategies are called natural
or innate immunity and they are neither enhanced by prior
exposure the infectious agent nor discriminate against
different types of intruders. On the other hand, there are
other defence mechanisms that are stimulated by prior
exposure to particular foreign substances. This is called
acquired or specific immunity. The foreign substances that
stimulate such immune responses are called antigens and the
immune responses which develop are exquisitely specific for
the particular antigen and are enhanced by each encounter
with this specific antigen. Specific immune responses are
classified into two types, humoral and cell-mediated,
depending on the components of the immune system mediating
the response. Humoral immunity is mediated by antibodies;
proteins that can recognize and eliminate a specific
I 23 antigen. Humoral immunity can be transferred temporarily to an unimmunized individual by a simple transfusion of the cell-free portion of the blood; plasma or serum. In contrast, cell mediated immunity can not be transferred to a naive individual through serum. Specific cells of the immune system called lymphocytes are the mediators of this type of specific immunity.
Role of Phagocytes:
The principle mechanism of natural immunity against intracellular pathogens such as Legionella is phagocytosis
(Dannenberg, 1989; Kaufmann and Reddehase, 1989; Kaufmann,
1990) . Two types of phagocytes with the ability to kill microorganisms; hence named professional phagocytes, make an important contribution to human defence against bacterial infections. Such professional phagocytes are polymorphonuclear leukocytes (PMNs, or neutrophils) and monocytes. Monocytes differentiate into macrophages upon leaving the blood stream and entering tissues (Pearsall and
Weiser, 1970). However, most intracellular pathogens have developed very shrewd strategies to overcome killing by phagocytosis. As mentioned earlier, a number of 24 intracellular pathogens including Legionella have developed ways to resist acidification of the phagosome and phagolysosomal fusion (Moulder, 1985). Therefore they are not exposed to the potent lytic enzymes sequestered in the
lysosomes. Others such as Listeria possess cytolytic enzymes
(listeriolysin) that are used to lyse the phagosome bearing
the pathogen, thus releasing it to the cytosol where it is protected from all host defences (Geoffroy et. al., 1987;
Kuhn et. al., 1988; Portnoy et. al., 1988). Yet other
intracellular pathogens such as Yersinia pestis (Straley and
Harmon, 1984), Salmonella typhimurium (Carroll et. al.,
1979), Coxiella burnetti (Akporiaye et. al., 1983) have
developed ways to resist the harmful acidic environments
created as a consequence of phagolysosomal fusion.
Besides phagocytic killing, monocytes and macrophages
have other important roles in the defence against infection.
One is the release of cytokines such as tumor necrosis
factor alpha (TNF-a) (Adams and Hamilton, 1984) . Cytokines
are biologically active substances secreted by particular
cells. The cytokines can stimulate cells in an autocrine
or/and paracrine fashion. Another function of monocytes and I
25
macrophages is the induction of specific or acquired
immunity. Macrophages are instrumental in the induction of
both humoral (antibody mediated) and cell-mediated immunity
(Adams and Hamilton, 1984).
Cell mediated immunity in Legionnaires' disease:
The predominant form of protective specific immunity
against L. pneumophila is cell-mediated (Horwitz, 1983;
Winn, 1983; Breiman and Horwitz, 1987). Although there is a
humoral response during infection, antibodies provide little
defence against L. pneumophila. As described earlier, the
development of specific antibodies could be deleterious to
the host since these could facilitate the entry of the
bacterium to alveolar macrophages (Hushman and Johnson,
1992) .
Horwitz and co-workers have shown that activated human
monocytes and alveolar macrophages, including those
activated by gamma-interferon (IFN-y) inhibit the
intracellular multiplication of L. pneumophila (Bhardwaj et.
al, 1986) . Virtually all bacterial pathogens need iron for
their growth. However iron is not a freely available 26 commodity in eucaryotic cells. Iron is stored tightly bound to iron binding compounds such as transferrin and lactoferrin or in iron storage proteins such as ferritin.
L. pneumophila has a very high metabolic requirement for iron as demonstrated by its inability to grow on culture media lacking iron supplements (Feeley et. al., 1978).
However, unlike other gram negative bacteria, L. pneumophila does not produce soluble iron-chelating compounds, called siderophores, that would allow it to scavenge iron from its host (Reeves et. al, 1983). Therefore it is imperative that this organism grows in an environment that provides an iron source for its growth. According to Byrd and Horwitz (1989), the ribosome studded phagosome is one such environment in mononuclear phagocytes. IFN-y treatment of monocytes appears to result in the down-regulation of transferrin receptors on the surface of the macrophages and leads to depletion of the labile iron pool (Byrd, and Horwitz, 1989). Therefore, activated monocytes inhibit L. pneumophila intracellular multiplication by limiting the availability of iron. Adding
iron back into the culture restores the ability to grow
intracellularly (Byrd and Horwitz, 198 9). 27
Furthermore, work by Summersgill et., al. (1992) have demonstrated that treatment of murine macrophages with IFN-y induces the expression of nitric oxide synthase, an enzyme responsible for generating nitric oxide (NO) radicals. These
NO radicals in turn can bind iron to form intracellular iron-nitrosyl complexes which are eventually lost from macrophages. Therefore, the end result of the generation of
NO radicals would be again, the depletion of intracellular iron within macrophages. Apart from this, NO has been shown to have a microbicidal effect against a number of pathogens, i.e. Toxoplasma gondii, Leishmania sp., and Francisella tularensis (Summersgill et. al, 1992). L. pneumophila is known to be sensitive to toxic oxygen metabolites produced by phagocytic cells (Lochner et.al., 1983). Therefore it is possible that the generation of NO by IFN-y treated macrophages is inhibitory for the intracellular growth of L. pneumophila.
A more recent study by Widen et. al. (1993) showed that
L. pneumophila has the ability to induce Interleukin 1 (IL-
1) production by macrophages. This study also shows that opsonization of this bacterium by specific antibody enhances 28
the IL-l inducing potential of the organism. A similar study
has shown that L. pneumophila has the ability to induce TNF-
a production by peritoneal macrophages (Blanchard et. al.,
1987). Therefore, according to these studies, two of the most important pyrogens, IL-1 and TNF-a are produced by macrophages in response to interaction with L. pneumophila which may result in the high fever seen in patients with
legionellosis. Furthermore, IL-1 and TNF are involved in tl. Besides activated macrophages and PMNs, several investigators have shown that T lymphocytes (Friedman, 1988), natural killer (NK) cells (Blanchard et. al., 1988), and lymphokine-activated killer (LAK) cells (Blanchard et. al., 1987) also take part in host defence against L. pneumophila infection. Considering the present literature, it is possible to put forward the following model to explain the possible I I 29 immunological events that would take place in the host in response to L. pneumophila infection. Upon infection, L. pneumophila grow and multiply within the macrophage resulting in eventual lysis of the host cell. Certain L. pneumophila antigens would be presented on the surface of the macrophage in the context of the major histocompatibility complex class II (MHCII) molecules. Others L. pneumophila antigens, such as Hsp60, have been shown to be exposed to the cytoplasm of the macrophage (P.S. Hoffman, personal communication) and thus could be presented in the context of MHC class I molecules. Presentation of antigens in the context of MHC class I results in the activation of CD8+ cytotoxic T cells (Parnes, 1989). Activation of CD8+ cells appears to require several signals. One signal is the interaction of the T cell receptor on the CD8+ cell with the antigen- MHC class I complex. Another vital signal involves interleukin 2 (IL-2), which is produced mainly by activated CD4+ T helper cells, binding to IL-2 receptors on the CD8+ T cells. Activated CD8+ T cells will then go on to kill any target cell presenting the specific antigen that caused their activation. 30 Antigens that are presented on MHC class II molecules may lead to the development of a humoral response (initiated by Th2 cells) or a cell-mediated response (initiated by Thl cells), depending on the class of the responding T cells. Some of the cytokines produced by activated Thl cells are IL-2, IFN-Y; and TNF-(3. On the other hand, activated Th2 cells produce IL-4, IL-5 and IL-10, among others (Kaufmann, 1991). As mentioned earlier, cell-mediated immunity plays the main role in providing protective immunity against L. pneumophila infections. IL-2 plays a pivotal role in inducing cellular immunity. IL-2 secreted by activated T cells stimulates T cells in an autocrine fashion and also other cells of the immune system. IL-2 can stimulate NK cells to become lymphokine activated killer (LAK) cells. Both NK and LAK cells have the capability to lyse infected macrophages (Herberman et. al., 1986). IL-2 will also activate macrophages (Adams and Hamilton, 1984). Activated macrophages have been shown to possess an increased ability to kill L. pneumophila (Horwitz and Silverstein, 1981). In addition, activated macrophages produce a whole array of cytokines including IL-1, TNF-a, and IFN-a (Adams and Hamilton, 1984). The production of IFN-y by activated n 1 * 31 macrophages is inhibitory to the intracellular growth of L. pneumophila (Byrd and Horwitz, 1989). Furthermore studies have also shown the production of TNF-a (Blanchard et. al., 1987) and IL-1 (Widen et. al., 1993) by macrophages in response to infection by L. pneumophila. Immunodominant antigens of L. pneumophila: The search for L. pneumophila antigens involved in protection against infection has been undertaken by many groups of investigators. As a result of these attempts, in 1991 Blander and coworkers reported that vaccination of guinea pigs with a 38 kDa Zn metalloprotease (major secretory protein, MSP) conferred protection against lethal challenge. However, the same group of investigators reported that aproteolytic strains of L. pneumophila are both virulent and immunoprotective suggesting that other factors may be involved in the generation of protective cell- mediated immunity (Blander and Horwitz, 1991a). In this study, membrane preparations of L. pneumophila devoid of the MSP were also found to be effective in protecting guinea pigs against lethal challenge (Blander and Horwitz, 1991b). The major proteins present in L. pneumophila membrane I 32 preparations are the major outer membrane protein (OmpS), the macrophage infectivity potentiating protein (Mip), peptidoglycan associated protein (Ppl), flagellin proteins and the non-covalently associated heat shock protein (Hsp60) (Weeratna et.al, 1994). Heat shock proteins are a group of evolutionarily conserved proteins produced by both prokaryotic and eukaryotic cells in response to a variety of insults such as sudden temperature fluctuations (Kaufmann, 199 0b). These proteins have been shown to play an important role in immune responses to bacterial and parasitic pathogens (Shinnick, 1991). Convalescent sera from patients with culture-confirmed cases of Legionnaires' disease have been shown to contain antibodies against Hsp60 (Sampson et. al, 1986; Hoffman et. al. 1990). A study by Blander and Horwitz (1993) suggested that Hsp60 (major cytoplasmic membrane protein) of L. pneumophila induces cell-mediated protective immunity against infection by this bacterium. However, preliminary investigations by Edelstein et. al. (1991) contradicted these findings. The major outer membrane protein OmpS, which is the major focus of this present study is one of the most 33 abundant proteins produced by L. pneumophila. Several preliminary studies have reported this protein to protect guinea pigs against lethal challenge by L. pneumophila (Quinn et. al., 1987; Tartakovskii et. al., 1990). The present investigation deals partly with the immunological characterization of OmpS. In this endeavour, attempts have been made to compare the immunogenicity of OmpS against other proteins that have already been characterized as immunogens of L. pneumophila; viz: MSP and Hsp60. OmpS was also evaluated as a potential candidate for developing a vaccine against Legionnaires' disease. Genetics of Legionella. Virulence: Every intracellular bacterium has developed mechanisms to gain entry into host cells, to evade host defences, to grow and multiply within the host cell utilizing the limited space and nutrients, and to disseminate into new hosts. Any bacterial products that help them to successfully complete their parasitic life cycle would be considered as virulence factors (factors facilitating virulence). As in many intracellular pathogens, virulence of L. pneumophila is 34 considered to be multifactorial (Engleberg, 19 93). Hence there may be many L. pneumophila proteins associated with virulence viz: the major outer membrane protein (OmpS), macrophage infectivity potentiating protein (Mip), Zn metalloprotease, heat shock protein (Hsp60) the genetic loci icm and dot and the gene product of hel. However, the mechanisms by which these potential virulence factors contribute to the pathogenesis of L. pneumophila and how they are regulated throughout the pathogen's life cycle is not very clear. To date the best studied phenotypic marker that identifies avirulent mutants of L. pneumophila from virulent strains is the avirulent strains ability to grow on culture media containing sodium chloride levels in excess of 0.6% (Catrenich and Johnson, 1989). The virulent strains cannot grow on media containing high levels of sodium chloride (i.e., 0.85%; the physiological concentration). However, the molecular basis for this sodium chloride tolerance in avirulent mutants and its significance is not yet established. Although endemic plasmids are found in several I. 35 Legionella species, they have not been linked to virulence (Brown et. al., 1982). Since the 1980's, several investigators have attempted to use genetic approaches to understand the molecular basis of L. pneumophila pathogenesis. Some of the distinct approaches so far include: 1. attempting to identify genes linked with virulence through mutagenesis; 2. direct cloning of potential virulence-factor genes such as genes coding for toxins, surface proteins and other extracelluar products; and 3. attempting to restore virulent phenotypes through complementation and thus trying to identify factors necessary for virulence. The following is a brief discussion of some of these approaches. Identifying genes by mutagenesis: The three main methods for the introduction of foreign DNA into bacteria are through transformation, conjugation and transduction. Transduction has not been an option for Legionella due to the lack of bacteriophages that infect this bacterium. Transformation of DNA into L. pneumophila can be accomplished successfully by electroporation (Marra 36 et. al. 1992). Conjugative transfer of plasmid DNA between E. coli and L. pneumophila has also been facilitated by the isolation of restriction minus mutants of L. pneumophila (Cianciotto et. al. 1988; Marra and Shuman, 1989). Several investigators have attempted to use transposon mutagenesis as a tool for studying L. pneumophila virulence. In 1985, Dreyfus and Iglewski demonstrated the conjugative transfer of broad-host-range Pseudomonas antibiotic-resistance plasmids from E. coli into a number of Legionella species,- viz: L. pneumophila, L. micdadei, L. longbeachae. They also demonstrated the recombinational exchange of chromosomal traits by reversing a thymidine auxotroph of L. pneumophila into a thymidine prototroph using a plasmid-mediated chromosome mobilization from a prototroph donor strain. Chen et. al. (19 84) introduced the transposable element Tn5 into several Legionella species on a self-transmissible Pseudomonas plasmid, but failed to demonstrate the transposition of the Tn5 element into the L. pneumophila chromosome. However, Keen et. al. (1985) were successful in demonstrating the transposition of Tn5 into the L. pneumophila chromosome. In this study, Tn5 was first 37 introduced into L. pneumophila on a plasmid vector (pRK34 0), that is temperature sensitive for plasmid maintenance. Subsequently, the transposition was achieved by growing the bacteria at the non-permissive temperature. However, the frequency of these transposition events was too low to be useful in mutagenesis experiments (Keen et. al., 1985; Mara and Shuman, 1992) . In 1987, Mintz and Shuman reported efficient transposition of the bacteriophage Mu into multiple sites within the L. pneumophila genome. However, unlike in other gram negative bacteria, bacteriophage Mu is not adsorbed to Legionella and cannot complete its life cycle and produce active phage particles in L. pneumophila. Taking advantage of the efficient transposition of Mu into L. pneumophila genome, Albano et. al. (1992) designed the transposon MudPhoA by cloning the phoA gene from E. coli and the kanamycin resistance gene (neo) from Tn5 into MudII4041. MudPhoA was then used by these investigators to generate fusions in genes coding for secreted proteins. However the MudphoA system too has several draw backs. For example, at higher temperatures (42°C) there is a derepression of Mu cts(62) which leads to transposition. Therefore in animal experiments, febrile temperatures would induce secondary 38 transposition events that would limit the viability of L. pneumophila strains carrying the MudPhoA. Circumventing the problems related with Tn5 and Mu systems, Wiater et. al. (1994) developed TnS03dIIIacZ for transposon mutagenesis of L. pneumophila. This transposon has several features that make it a more efficient tool for mutagenesis compared to the ones used before; the high frequency of transposition, the gene for KmR which is a selectable marker in L. pneumophila, a truncated lacZ gene in frame with the end of Tn903 which could act as a reporter for both transposition events and gene expression if a transposition event occurs in frame with an expressed gene, and the placement of the Tn903 transposase gene out side of Tn303dIIlacZ which enables the stable maintenance of the insertional event. Although the ColEl-based Tn903dIIlacZ delivery plasmid pLAW330 can replicate in L. pneumophila, it is quickly lost without proper antibiotic selection (Chloramphenicol). In addition, the simple transposition of Tn503dIIlacZ results in the degradation of the vector. All these features combined, make Tn903dIIlacZ an efficient tool for mutagenesis of L. pneumophila. However, some of the undesirable features of this system may include the non- I 39 random transposition of Tn.903 and vector integration into the Legionella chromosome (Waiter et. al., 1994; Pope et. al., 1994). In 1994, Pope et. al. reported the use of mini- TnlO for random mutagenesis of L. pneumophila. Similar to Tn503dIIlacZ, mini-TnlO also caries the kanamycin resistance marker and the transposase gene is placed out side of the transposon element. However, in this system, the randomness of transposition has been enhanced by altering the target specificity of the transposase. In addition, the chromosomal rearrangement around the site of insertion has been minimized by shortening the inverted repeats within the transposon (Pope et. al., 1994). These investigators have designed two delivery vehicles for mini-TnlO. Both are ColEl replicons which are known to replicate in L. pneumophila (Kleckner et. al., 1991). However they have used the counterselectable markers sacB which encode for levansucrase which is lethal to wild-type L. pneumophila grown in the presence of 5% sucrose and rpsL which encodes the ribosomal S12 protein and confers streptomycin sensitivity in streptomycin resistant L. pneumophila. Mutagenesis with mini-Tn2 0 cloned in these counterselectable vectors are 40 reported to yield random and stable mutants of L. pneumophila at a high frequency. Using the Tn903dIIlacZ transposon as a tool Wiater et. al. (1994) identified several single insertional events. Out of these, one mutant appeared to be defective in brown pigment production (Pig-) which is a characteristic feature of Legionellae. Pig phenotype appears to be growth phase regulated in that it is turned on during the onset of the stationary phase. Pigment production is also increased during macrophage infection, although it is not required for growth within or killing of the macrophages. According to the authors, pig is a yet uncharacterized gene different from Ily(encoding for brown pigment with fluorescence). Shuttle mutagenesis employing X::TnphoA-oriT was used to obtained several mutants of L. pneumophila that showed ;>50% decrease in host cell cytopathicity (Arroyo et. al. , 1994). Sequencing of L. penumophila chromosomal DNA adjacent to X: -.TnphoA-oriT insertion revealed a gene designated hel for hemolysin expression in Legionella. The inferred amino acid sequence for this gene product shows homology to E. 41 coli TolC, Bordetella pertussis CyaE and the Alcaligens eutropus proteins CzcC, and CnrC. However, the involvement of hel in L. pneumophila virulence is not yet clear. Identifying genes by complementation of avirulent mutants: Horwitz in 1987 isolated an avirulent mutant of L. pneumophila (called 25D) by passage on Mueller-Hinton Agar which was defective in a number of phenotypes; i.e. formation of the ribosome studded specialized phagosome, inhibition of phagolysosomal fusion, intracellular multiplication and causing infection and death in the guinea pig model. By electroporating a L. pneumophila cosmid library into the mutant 25D, Marra et. al. (1992) were successful in identifying a single genetic locus (icm) that complemented all the defects in this mutant, thus restoring the wild type virulent phenotype. In a subsequent study, Berger and Isberg (1993) isolated three types of L. pneumophila mutants defective for intracellular growth by transposon mutagenesis using Tnlacl (Chow and Berg 1988) followed by intracellular thymineless 42 death enrichment. The thymineless death enrichment protocol is based upon the observation that L. pneumophila thymine auxotrophs have a greater than 10-fold decrease in their ability to multiply within cells without the addition of thymine or thymidine (Mintz et. al. 1988). Therefore, thymine requiring avirulent mutants would survive thymine starvation better than the thymine requiring wild type strains due to their inability to multiply within the -intracellular milieu. The three types of mutants isolated were; class I : defective for host cell ribosome recruitment but competent for inhibition of phagolysosomal fusion, Class II: defective for both ribosome recruitment and inhibition of phagolysosomal fusion (similar to mutant 2 5D isolated by Marra et al. in 1992), Class III: wild type for both ribosome recruitment and inhibition of phagolysosomal fusion. By transforming a L. pneumophila genomic library constructed in a low copy number plasmid into Class I and Class II mutants, these investigators were successful in identifying a genetic locus that complements both classes of mutants. This genetic locus was named dot for defect in organelle trafficking. Transposon mutagenesis of the L. pneumophila chromosomal dot locus resulted in a mutant 43 defective for intracellular growth with a phenotype resembling class II mutants. Physical mapping and Southern blot analysis revealed partial overlapping of the dot and icm loci, suggesting the possibility that they are located adjacent to each other in the L. pneumophila chromosome. The DNA sequence of both the icm and dot loci were reported recently (Brand et. al., 1994; Berger et. al., 1994). According to these reports the 4 kb icm locus that lies 5' to dot consists of four genes; icmWXY and Z. The 8.2 kb dot locus consists of 5 open reading frames; dotA, sbpA, orf5, dlpA and orf3. These two loci appear to be divergently transcribed and, most likely, coordinately regulated (Berger et al. 1994). The specific functions of any of these genes have not yet been elucidated and none of the icm or dot gene products resemble known prokaryotic or eukaryotic proteins. However, the predicted protein product for dotA appears to possess a highly hydrophobic C terminus that may interact with the lipid bilayer and influence membrane fusion events (Berger et al., 1994). The N terminal region of dplA predicted product has high homology to a number of dehydrogenases and thus named dplA for dehydrogenase-like protein. P 44 The work done on icm and dot loci have shed new insight on the virulence of L. pneumophila. These studies suggest that both the association of host cell organelles with the phagosome membrane and inhibition of phagolysosomal fusion are needed for the proper intracellular growth and multiplication of L. pneumophila. Further studies on the functional roles of these gene products may reveal the exact mechanisms of action of these gene products in pathogenesis. In an attempt to find out whether there are any other genes or genetic loci involved in the intracellular growth and cell killing by L. pneumophila, Sadosky et al. (1993) analyzed 4536 independently derived TnS03dIIlacZ insertion mutants of L. pneumophila. Out of these, 55 , itants exhibited varying degrees of impotencies in their ability to multiply within and kill human macrophages. Out of these 55 mutants, 9 mutations were located either in icm or dot genes. This suggests that the remaining 46 mutations must lie within other L. pneumophila genes which are also essential to maintain the intracellular lifestyle of L. pneumophila. fe 45 Salt resistance and Legionella virulence: As mentioned earlier, salt tolerance is the only phenotypic marker that has been linked to conversion of virulent L. pneumophila strains to avirulence. Salt resistant avirulent mutants can be obtained by either passaging virulent strains on Mueller-Hinton medium or repeated passage on non-selective BCYE medium (McDade and Shephard, 1979; Catrenich and Johnson, 1988). Interestingly, in the study by Sadosky et. al. (1993), where they created 4,536 independent Tn903dIIlacZ insertion mutants of L. pneumophila, out of which 55 were defective in macrophage killing (Mak-), almost all of the Mak- mutants were also salt tolerant. The insertion mutants with Mak+ phenotype remained salt sensitive. Further, the introduction of the Tn903dIIlacZ mutation from Mak- strains into a salt sensitive virulent strain converted it to a salt tolerant phenotype while the introduction of a Tn903dIIlacZ mutation from a Mak+ salt sensitive strain did not alter its salt sensitive phenotype. This study also reported that although some of the Mak- mutants were complemented by the icm locus, they retained their salt tolerant phenotype. All these data might suggest that either the genes conferring salt 46 sensitivity and macrophage killing ability are closely linked or coordinately regulated. Cloning genes using the functional properties of the gene product: Direct cloning of genes by taking advantage of the functional properties, immunogenicity or the cellular location of the gene product has been the most successful approach for L. pneumophila so far. Zinc Metalloprotease The legionellae have been shown to produce a number of proteolytic enzymes. Some of these remain bound to the organism and others are secreted to the extracelluar environment (Dowling et. al., 1992) . Out of these, one of the most well characterized proteases of L. pneumophila is the Zinc metalloprotease. As the name implies, its activity depends on the presence of several metal ions including Zn2+ and there is one molecule of Zn2+ per molecule of enzyme (Dreyfus and Iglewski, 1986). This protease has various other designations such as the extracelluar protease, tissue destructive protease, phenylalanineaminopepdidase, major 47 secretory protein and cytolysin. The 38 kDa protein exhibits a broad range of activity and has the ability to digest collagen, casein, gelatine (Conlan et. al., 1986) and some serum proteins such as the serine protease inhibitors aa- antichymotrypsin (Mueller, 1980) and ai-antitypsin (Conlan et. al., 1988). The purified protein can impose cytopathic effects on CHO cells and hemolytic activity on canine and guinea pig erythrocytes (Keen and Hoffman, 1989). The purified protease has also been shown to elicit lesions in guinea pig lungs which are indistinguishable from those produced during infection with L. pneumophila (Baskerville et. al. 1986; Conlan et. al., 1986). Immunogold and in vivo activity studies by Williams et. al. (1987) have demonstrated that L. pneumophila produces the protease upon infection of guinea pig alveolar macrophages. The gene encoding this protease named proA (Quinn and Tompkins) or msp (Szeto and Shuman) was cloned independently by two groups of investigators (Quinn and Tompkins, 1989; Szeto and Shuman, 1990) using a similar strategy; i.e. by screening a L. pneumophila cosmid library expressed in E. coli for hydrolysis of casein. According to amino acid I 48 sequence data and competitive inhibitor studies, this protease belongs to the family of bacterial neutral proteases which also include the elastase of Pseudomonas aeruginosa and thermolysin of Bacillus thermoproteolyticus (Black et al. 1990; Moffat et. al., 1994). However, the L. pneumophila protease is somewhat unique from other bacterial proteases in that it contains three phenotypes: protease, cytotoxin, and a hemolysin (Keen and Hoffman, 1989; Quinn and Tompkins, 1989). The gene coding for this protease contains a single open reading frame of 1,629 bp which is much larger in size than necessary for coding a protein of 38 kDa (Black et. al., 1990). Further studies by Moffat et al (1994) have demonstrated that this protease is subjected to post-translational cleavage that result in a mature protein of 38 kDa. The cloning and sequencing studies of the Zn metalloprotease has revealed that in L. pneumophila, the native protease is secreted to the environment whereas in E. coli it is accumulated in the periplasm (Moffat et. al., 1994). This leads to an interesting speculation that L. pneumophila utilizes a unique transport mechanism for export of the protease that is absent in E. coli. 49 The role of protease in the virulence of L. pneumophila is still controversial. Szeto and Shuman in 1990 created a proA::Tn9 mutant by allelic exchange. This mutant was still able to hydrolyse casein, however, to a much lesser extent than the wild type. According to a quantitative measurement of proteolytic activity using the casein-fluorescein isothiocyanate (FITC-casein) assay, the mutant contained „ ?°- of the wild type protease activity (detected in cell supernatants concentrated by an amicon microconcentrator). In immunoblot assays, the proA mutant showed no full length protease, although it did appear to produce two smaller proteins that cross reacted with anti-protease polyclonal antibodies. According to these investigators, this mutant is just as capable as the wild type of intracellular growth and cell killing in the macrophage-like cell line HL-60. Also Blander et al. (1990) demonstrated that when guinea pigs were challenged with protease negative mutant and wild type L. pneumophila, both strains multiplied in the lungs at comparable rates and produced similar pathology of the lungs. Therefore these studies indicate that this protease is not required for the intracellular infection, multiplication or the lethality in vivo. However, in a i 50 subsequent study Moffat et. al. (1994), were able to completely inactivate the proA gene by inserting a kanamycin-resistance cassette. The resulting proA mutant of L. pneumophila was completely devoid of any proteolytic activity as shown by hide powder azure assay, SDS PAGE and immunoblotting. In this study, Moffat et al. (1994) showed that the proA mutant of L. pneumophila was not defective in intracellular growth within Acanthamoeba or macrophages. However, this mutant was attenuated for virulence in guinea pig infection studies. The guinea pigs infected with the proA mutant strain survived longer than those infected with the isogenic parent strain (3.15 days versus 4.65 days). Looking at the histopathology of the animals, the ones infected with the wild type strain had a characteristic acute inflammatory response with large numbers of neutrophil infiltrations, necrosis and edema. Contrary to this, animals infected with the proA mutant strain were able to mobilize a macrophage response, thus reducing the number of viable bacteria per gram of lung tissue. In other studies, protease has been shown to impair phagocytic and natural killer cell functions (Dowling et. 4 ? ^l3l-* 51 al., 1992), and purified protease at a non-cytotoxic concentration has been shown to inhibit neutrophil and monocyte chemotaxis (Rechnitzer et. al. , 1989). Furthermore, work by Blander and Horwitz (1989) has demonstrated that this protein acts as an immunogen in developing a cell- mediated immunity against Legionella infections in guinea pigs. Legiolysin: Recently, another L. pneumophila gene coding for an extracellular enzyme termed legiolysin has been cloned and sequenced based on a functional assay; the ability to lyse canine erythrocytes (Wintermeyer et. al., 1991, 1994). Legiolysin is a 39 kDa protein which is responsible for producing a brown pigment and a fluorescence activity which has been previously observed by other investigators (Baine and Rasheed, 19 79; Pine et.al., 1984; Berg et. al., 1985). When cloned in E. coli, lly confers the ability to lyse human, canine and sheep erythrocytes. However, hemolytic activity attributed to legiolysin has not yet been demonstrated in L. pneumophila. The specific involvement of legiolysin in virulence of L. pneumophila if any is not yet I 52 known. A more recent study by Steinert et. al. (1995) reveal that legiolysin negative mutants of L. pneumophila shows a marked decrease in their resistance to ordinary light. These mutants do not show any impairment in their ability to grow within the amoebae, H. vermiformis. This may suggest that legiolysin plays a critical role in the extracellular phase of L. pneumophila growth rather than in its pathogenicity. Cloning Antigens expressed on the cell surface: A number of L. pneumophila genes have been cloned and sequenced by expressing them in an E. coli genetic background (using L. pneumophila cosmid libraries) and screening with antibodies directed against either purified antigen or the whole cells of L. pneumophila. Pioneering work on this was carried out by Engleberg et al. in 1984 when they identified E. coli clones expressing cell surface antigens of 19, 24, 6 6 and 6 8 kDa that reacted with L. pneumophila antibodies. Out of these both 19 and 24 kDa proteins encoding for the peptidoglycan associated protein (Ppl) and macrophage infectivity potentiating protein (Mip) respectively have already been cloned and sequenced (Ppl: 53 Ludwig et al., 1991; Engleberg et al., 1991 Mip: Cianciotto et al., 1989). Macrophage Infectivity Potentiating Protein (Mip): The 24 kDa Mip protein is a highly basic protein that has been shown to potentiate the early survival of L. pneumophila in macrophages; thus named accordingly. As mentioned earlier, the gene coding for Mip was identified by screening a clone bank of L. pneumophila expressed in E. coli with anti-L. pneumophila anti serum (Engleberg et. al., 1984; Cianciotto et. al., 1989). Disruption of this gene by allelic exchange mutagenesis produced L. pneumophila mutants lacking the 24 kDa protein (Cianciotto et al., 1989). These mutants appeared to be less capable of surviving the early stages of infection of U937 cells and human alveolar macrophages compared to their isogenic parents. In these studies, the mip- mutant was found to be 80 fold less virulent than the wild type strain. The re-introduction of a functional mip gene in to the mutant restored the wild type phenotype (Cianciotto et. al., 1989). Furthermore, Cianciotto and Fields (1992) have demonstrated that mip mutants are also defective in their ability to infect two 54 different types of protozoans, Tetrahymena pyriformis and Hartmanel la vermiformis. A 21 kDa Mip-like protein showing high homology to the Legionella Mip protein has been identified in another intracellular pathogen, Chlamydia trachomatis (Lundemose et. al., 1991, 1992). Furthermore, studies have revealed a striking similarity among the carboxyl terminal ends of the Legionella and Chlamydia Mip proteins and the FK506 binding protein of Neurospora crassa, the human analog (HsFKBP) and a hypothetical protein derived from a cryptic Neisseria meningitidis DNA sequence with homology to HsFKBP (Bangsborg et. al., 1991). These FKBPs have been shown to possess peptidyl-prolyl-cis-trans-isomerase activity and this activity is strongly inhibited by the immunosuppressive drug FK506. These peptidyl-prolyl-cis-trans-isomerases also known as rotamases, function in catalyzing slow protein folding reactions (Tropschug et. al., 1990). Two rotamases; one periplasmic and one cytoplasmic catalyzing the refolding of thermally denatured proteins have been identified from E. coli (Compton et. al., 1992). A report by Fischer et. al. (1992) reveals that L. pneumophila Mip too contains rotamase 55 activity and that this activity is inhibited by FK506. However, the exact mechanism of action of Mip in L. pneumophila and its precise cellular location has not yet been elucidated. Peptidoglycan Associated Protein (Ppl): This is, as mentioned earlier, the 19 kDa surface expressed L. pneumophila protein designated Ppl (Ludwig et. al., 1991) or Pal (Engleberg et. al., 1991). The gene coding for this prot in has been cloned and sequenced (Ludwig et. al., 1991; Engleberg et. al., 1991). Based on the predicted amino acid sequence this protein shows high homology to the peptidoglycan-associated lipoproteins of E. coli and Hemophilus influenzae (Engleberg et. al., 1991). Also DNA sequence comparison data of the genes between these 3 bacteria reveal this protein to be a lipoprotein. Heat shock protein; Hsp60: This is yet another protein for which the encoding gene was cloned (Hoffman et al., 1989) and sequenced (Hoffman et al., 1990) based on its abundant expression in L. pneumophila. The 60 kDa HSP,[E. coli GroEL-like heat shock 56 protein) is the second most abundant protein synthesised by L. pneumophila and is one of the prominent proteins synthesised when the bacteria are intracellular (Hoffman et.al.1990; Abu Kwaik, 1993). Among others, Hsp60 is one of the proteins recognized by western blot assays in convalescent sera from patients with confirmed cases of Legionnaires' disease (Sampson et. al., 1986). The gene coding for the 60 kDa heat shock protein, htpB, together with another gene, htpA, coding for a 16 kDa protein, forms an operon. At least in E. coli this operon appears to be under the control of a heat shock promoter. However, the mode of regulation of this gene in L. pneumophila remains elusive. According to southern blot hybridization assays, this gene appears to be highly conserved among the genus Legionella (Hoffman et. al., 1989). The deduced amino acid sequence for the 60 kDa protein shows 76% homology with the 65 kDa HSP of mycobacteria and at least 85% homology if not higher with the E. coli GroEL and the Coxiella burnetii HtpB proteins. Interestingly, the htpAB operon fails to complement groEL and GroES temperature sensitive mutations in E. coli suggesting that despite their high evolutionary 57 relationships, they are assigned species-specific role in the respective organisms (Hoffman et. al. 1989). It is now a well established fact that heat shock proteins function as molecular chaperons by binding to and preventing renaturation of proteins destined for export or import from one cellular organelle to another. In addition to this, they are also involved in the assembly of oligomeric proteins as well as in the renaturation of denatured proteins during various insults to the cell such as heat (Rothman, 1989). Also, HSPs have been demonstrated to be major T-cell antigens for most pathogenic microorganisms, especially the intracellular parasites (Young et. al., 1988). The involvement of Hsp60 in L. pneumophila pathogenesis is still not clear. However the fact that this is one of the prominent proteins synthesised by virulent bacteria upon infection of HeLa cells, and that avirulent mutants produced decreased amounts of Hsp60 upon being intracellular compared to virulent strains (Hoffman et. al., 1990) suggest a possible involvement of Hsp60 in pathogenesis. Apart from 58 this, work by Blander and Horwitz (1993) suggests this protein to be a protective antigen against Legionnaires' disease in guinea pigs. Major Outer Membrane Protein: The major Outer Membrane protein OmpS, the main focus of the current work, is one of the major proteins synthesised by L. pneumophila. This protein has been purified and characterized by a number of investigators (Hindahl and Iglewski, 1984; Gabay et al., 1985; Butler et al., 1985; Butler and Hoffman, 1990; Hoffman et. al., 1992). The gene coding for this protein was cloned using reverse genetics; i.e.: utilizing degenerate oligonucleotides designed according to the N-terminal sequence of the protein. Cloning and sequencing of ompS was done as a part of this current study and hence will be discussed in a later section of this thesis. Other proteins implicated in L. pneumophila virulence: Apart from the above mentioned proteins, there are a few not well characterized proteins that have been implicated in L. pneumophila virulence. For most of them, I 59 their involvement in L. pneumophila virulence is based merely on the involvement of the homolog proteins in the virulence of some other known bacterial pathogens. A brief discussion of some of these proteins follows. Phospholipase C: With the exception of L. micdadei, all other Legionella sp. tested appear to possess phospholipase activity (Dowling et. al., 1992). In 1988 Baine purified and characterized the phospholipase C of L. pneumophila. It was shown to have molecular weight of 45-50 kDa with a maximum activity observed at a pH range of 8-9. This enzyme was shown to hydrolase phosphotidylcholine which is an important constituent of the eukaryotic cell membrane. Therefore it is possible that the cytolytic action of phospholipase C damages both the inflammatory cells and the lung tissue. Apart from that, phosphotidylcholine is a major component of pulmonary surfactant and thus, the action of this enzyme could impair the pulmonary gas exchange. I 60 Phosphatase One of the pivotal stages of signal transduction in eukaryotic cells is accomplished by protein phosphorylation and dephosphorylation. Pathogenic microbes have therefore evolved mechanisms to block these steps. L. pneumophila is no exception to this and studies by Thrope and Miller (1981) have demonstrated, that 6 serogroups of L. pneumophila tested produced an acid phosphatase. Furthermore, Saha et. al. (1985) have purified 2 phosphatases from L. micdadei which have the ability to dephosphorylate phophatidylinositol biphosphate and inositol triphosphate which are components of the phospholipid signalling pathway. Therefore, it would be safe to assume that these phosphatases have a role in influencing signal transduction in phagocytes (Belyi, 1993). Contradictory to this however, Kim et. al. (1995) recently created site directed mutant in the major alkaline phosphatase (pho) gene of L. pneumophila and showed that these mutants were not significantly attenuated in their ability to infect U937 cells compared to their wild-type counterparts. J 61 Protein Kinase: Preliminary studies by Belyi et. al. have detected the presence of a 40 kDa thermolabile protein kinase in L. pneumophila cell lysates. The biological significance of this protein has yet to be elucidated. However studies by Saha et. al. (1988) have revealed 2 protein kinases of L. micdadei one of which has the ability to phosphorylate, among other substrates, phosphotidylinositol. Apart from these a 54 kDa cytoplasmic membrane protein with ADP-ribosylating activity has been identified from L. pneumophila (Belyi et. al., 1991). Such ADP- ribosylating toxins have been identified and well characterized from bacterial pathogens such as Bordetella pertussis and Vibrio cholera. These toxins have been shown to involved in the inactivation of G-proteins which are a major constituent of the signal transduction pathways of eukaryotes. L. pneumophila also possess enzymes such as peroxidase and superoxide dismutase which may function to neutralize the microbicidal effects of phagocytic cells (Pine et. al., 1984; Steinman, 1992; Sadosky et. al., 1994). 62 As mentioned earlier, L. pneumophila requires high levels of iron for growth. In order to satisfy this requirement in the intracellular milieu, L. pneumophila possesses a not yet fully characterized iron reductase (Reeves et. al., 1983; Johnson et. al., 1991). Recently, Hickey and Cianciotto (1994) have cloned and sequenced a fur gene in L. pneumophila which codes for a 15 kDa protein with homology to E. coli, Yersinia pestis and Vibrio sp. Fur proteins. Outer membrane porin proteins of gram negative bacteria: The outer membrane of gram negative bacteria forr. the interface between the outside and the inside of the cell. Thus it determines the import and export of substances in and out of the cell as well as various cellular interactions with the external environment. A group of proteins that are largely responsible for the permeability of the gram negative outer membrane are called the porins. Porins form water-filled channels across the outer membrane and these channels are of two types; non specific and specific (Hancock, 1987; Nikaido and Vaara, • I 63 1985). Non-specific channels mediate the passive penetration of ions and small hydrophilic nutrient molecules. These porin channels do not appear to contain any specific ligand binding sites (Hancock, 1991; Nikaido, 1992). Most of the well studied porin proteins such as OmpC, OmpF of E. coli and Salmonella typhimurium belong to this class of porins (Nikaido, 1994) . The specific porin channels contain specific ligand binding sites and thus promote the diffusion of these specific molecules (Hancock, 1991; Nikaido, 1992) . The glucose channel OprB, basic amino acid channel OprD, phosphate-channel OprP, pyrophosphate channel OprO of P. aeruginosa and the maltodextrin-specific channel LamB, nucleoside channel Tsx of E. coli fall into this category (Nikaido, 1994). Despite their ubiquitous presence among gram negative bacteria, these porin proteins share many common physical properties. Most porins consist of tight trimeric structures with monomer molecular weights between 30-48 kDa (Hancock, 1991). With respect to secondary structure, these porin proteins consist of a large number of |3-sheet structures that zig-zag across the outer membrane (Hancock, 1991). The three dimensional structures of a number of poiins have been elucidated so far by the use of 64 electron and X-ray crystallography (Jap et. al., 1991; Weiss et al., 1991; Weiss and Schulz, 1992). The structure of the Rhodobacter capsulatus porin protein was the first to be resolved by Weiss et al (1991). This was soon followed by the resolution of E. coli OmpF and PhoE porin structures by Cowan et. al. (1992). Although there is very little sequence homology between the R. capsulatus and E. coli porins, their protein folding patterns exhibit remarkable similarities. In every case, the polypeptide chain of each subunit traverses the membrane 16 times as antiparallel (3-strands, forming a P-barrel structure surrounding a large channel. Each subunit produces a channel and thus the trimers consist of 3 channels. Since every other amino acid side chain of the polypeptide strands faces the hydrophilic channel, the p- barrel spans the membrane without having stretches of hydrophobic residues (Nikaido, 1993, 1994). There are hydrophilic loops connecting the membrane spanning P-sheets. These loops are shorter on the periplasmic but are often longer on the outside. In all three porins,- R. capsulatus porin, OmpF, and PhoE porins of E. coli; the third loop folds inwardly into the lumen of the p-barrel thus narrowing the pore of the porin ("eyelet"). This leads to a channel r 65 with a wide entrance and an exit and a narrow central constriction. Such channels are ideally suited to exclude large solute molecules while allowing entry of small nutrient molecules (Nikaido, 1994). Apart from the classical trimeric porins, there are monomeric porin proteins that allow very slow entry of small nutrient molecules into the cell in a non-specific fashion. The major porin protein of P. aeruginosa, OprF and the E. coli outer membrane protein OmpA fall into this category (Nikaido, 1994). The Major Outer Membrane Protein of L. pneumophila: According to ultrastructure studies, L. pneumophila possesses a typical gram negative-type cell envelope with a cytoplasmic membrane, a peptidoglycan layer and an outer membrane (Flesher et. al. , 1979; Rodgers and Davey, 1982) . The L. pneumophila peptidoglycan layer appears to be resistant to hydrolysis by lysozyme; lysozyme treatment following an alkaline treatment which removes any protease resistant proteins associated with the peptidoglycan layer failed to disrupt the integrity of the peptidoglycan 66 sacculus (Amano and Williams, 1983). The peptidoglycan layer contains muramic acid, glucosamine and meso-diaminopimelic acid with 80 to 90% of the diaminopimelic acid residues cross-linked. Such high cross-linking in the peptidoglycan layer is rare among gram negatives. Gram negative bacteria generally exhibit 30-40% cross-linking compared to gram positives such as Staphylococcus which may exhibit as high as 90% cross-linking (Amano and Williams, 1983). In their work, Amano and Williams (1983b)categorized the peptidoglycan associated proteins of L. pneumophila into 3 categories; 1. non-covalently associated proteins 2. tightly bound proteins which may be covalently linked through disulfide bonds 3. covalently linked. In a subsequent analysis of the outer membrane preparations, Ehret et. al. (1984) discovered a 29 kDa major outer membrane protein /MOMP) that is present in abundance in all serogroups of L. pneumophila. This work was later confirmed by Hindahl and Iglewski (1984), further suggesting that this protein exists as a large aggregate stabilized by disulfide linkage. This protein contains cysteine residues as demonstrated by (35S)labelling studies and disulfide cross-linking as demonstrated by labelling free thiol groups of the reduced I l I 67 complex with iodoacetamide. Only strong reducing agents can dissociate the protein complex into monomers (Butler et. al., 1985). Under non-reducing conditions, L. pneumophila outer membrane preparations appear to contain peptidoglycan. Treatment of the non-reduced outer membrane preparations with lysozyme results in the appearance of a 95 kDa protein complex which dissociates in to 2 8.5 kDa subunits upon reduction (Butler et. al., 1985). Work by Butler and Hoffman in 1990 identified a 31 kDa peptidoglycan bound protein that is cross-linked to the MOMP protein via interchain disulfide bonds. This work resolved the nature of the cross-linking of the peptidoglycan layer of L. pneumophila. Subsequent work by the same group of investigators demonstrated that this 31 kDa protein (anchor protein) is really a 28 kDa MOMP subunit with a piece of peptidoglycan attached to it (Hoffman et. al., 1992a). One of several methods to demonstrate the pore forming ability of porin proteins is to reconstitute the protein in lipid vesicles, fuse the vesicles with planar black lipid membrane and look for the ability of the protein to form channels in the lipid bilayer. By following this method 68 Gabay et. al. (1985) demonstrated that the L. pneumophila MOMP is a porin forming cation selective channels through the outer membrane. Therefore, in summary, the current knowledge regarding the structure of MOMP is that it is a trimeric porin protein consisting of 28 kDa subunits held together by interchain disulfide bonds. One subunit (the 31 kDa anchor protein) helps to anchor the protein complex on to the peptidoglycan sacculus. The gene coding for the MOMP (ompS) has been cloned and sequenced (Hoffman et. al., 1992b). This work was done as a part of this current study and hence will be discussed in the latter sections of the thesis. MOMP was one of the first protein to be implicated in L. pneumophila virulence. As mentioned earlier, Bellinger- Kawahara and Horwitz (1990) demonstrated that MOMP facilitates internalization of L. pneumophila to host macrophages. This is thought to be mediated by the binding of the complement fragments C3b and C3bi to MOMP followed by the binding of the complement:MOMP complex to the complement receptors CRl and CR3 located on the surface of macrophages. 69 Coordinate regulation of virulence: Coordinate regulation of virulence determinants meet.' ^ted by environmental stimuli have been demonstrated in many microbial pathogens. For example, osmolarity has been noted as an environmental signal controlling gene expression in many microorganisms (Mekalanos, 1992). Osmoregulation of porin proteins OmpF and OmpC of E. coli is one extensively studied system. OmpF and OmpC are the two most abundant proteins in the E. coli outer membrane facilitating the uptake of many low molecular weight nutrients. OmpF porin has a larger pore size compared to OmpC. Enterics living in the mammalian gut have to withstand the damaging effects of powerful detergents such as bile salts. As a part of their survival strategy, they have developed ways to down regulate the expression of OmpF, the porin with the larger size pore while in the gut, and up regulate it while the bacteria are outside the host facilitating more efficient nutrient uptake from the nutrient poor environment. This differential regulation of OmpF and OmpC is mediated by a two component regulatory system, a hallmark of bacterial gene regulation (Hall and 70 Silhavey, 1981; Aiba et. al., 1989; Stock et. al., 1989). The main components of this type of a regulatory system are the environmental sensor protein which senses the signals and transduces it to the second component, and the response regulator protein. This signal transduction is mediated mostly by phosphorylation/dephosphorylation of the response regulator (McCleary et. al. , 1993). The response regulator protein in turn regulates the expression of the respecti\v- gene either directly or indirectly by effecting other transcription factors. In OmpF/OmpC system, EnvZ, is the environmental sensor protein (Comeau et al., 1985) . EnvZ belongs to the histidine protein kinase response regulator superfamily. It is an inner membrane protein of 50.3 kDa with two membrane spanning sequences in its aminoterminus (Comeau et al., 1985) . This protein has the ability to phosphorylate a conserved histidine residue (His-243) in response to environmental stimuli. This phosphate residue is then transferred to the response regulator protein OmpR. OmpR is a 27 kDa cytoplasmic protein and is phosphorylated at the aminoterminus by EnvZ (Aiba et. al. , 1989 . The DNA binding capabilities of OmpR are located on the carboxyterminus and the DNA binding is facilitated by 71 phospohylation of this protein (Aiba et al., 1989). The regulation of OmpC and OmpF by EnvZ/OmpR system has been studied extensively both at genetic and biochemical levels. The differential regulation of these porin proteins is dependent on the nature and the distribution of the OmpR-P binding sites located upstream of ompC and ompF promoters (Slauch and Silhavey, 1989; Stock et al., 1989). The ompF promoter contains a high affinity binding site for OmpR-P. Under low osmolarity conditions, the cellular concentration of OmpR-P is low, thus allowing the preferential binding of OmpR-P to ompF promoter. This leads to a preferential synthesis of OmpF under low osmolarity conditions. When in high osmolarity conditions, there is an abundance of OmpR-P in the cell due to enhanced activity of EnvZ. Thus ompR-P binds to the ompC promoter leading to increased expression of OmpC. It has also been shown that under high osmolarity conditions, OmpR-P binds to low affinity binding sites located upstream of the ompF promoter, leading to the formation of a nucleoprotein complex which leads to the repression of ompF expression (Slauch and Silhavey, 1991). In addition to this, another gene, designated micF is involved in the regulation of OmpC:OmpF ratio. micF is I I 72 located rijacent to ompC and is divergently transcribed (Mizuno et. a., 1984). micF codes for a small RNA product that is complementary to the 5' end of ompF mRNA. Binding of OmpR-P to the ompC, micF intergenic sequence induces the transcription of ompC as well as micF. The resulting micF RNA products bind to any ompF mRNA present in the cell and thus hinders the translation process (Mizuno et. a., 1984). Apart from these regulatory mechanisms, the current literature reveal that there are a number of other global regulators such as the leucine responsive regulatory protein (Ernsting et. al., 1993) and the integration host factor (Craig and Nash, 1984) involved in the regulation of OmpC:OmpF ratio in E. coli. This is an extremely simplified version of a vastly complex regulatory system. Apart from osmolarity, OmpF and OmpC levels are also regulated by other environmental conditions such as temperature, redox potential of the growth medium etc and these regulations entail extremely intricate mechanisms. In Vibrio cholerae, the expression of a number of virulence determinants such as the cholera toxin and Tcp pili (Toxin coregulated pili) is regulated by osmolarity, r i I 73 temperature, pH, and the availability of certain amino acids (Miller et al., 1987; Dirita et. al., 1990; Dirita, 1992; Mekalanos, 1992). In this system, the 3 main proteins involved in controlling virulence gene expression are ToxR, ToxS and ToxT. ToxR was identified as it was shown to activate a cholera toxin (ctx): lacZ gene fusion in E. coli when expressed in trans from a recombinant plasmid (Miller and Mekalanos, 1984). ToxR is a 32 kDa transmembrane protein spanning the cytoplasmic membrane with the N terminal portion exposed to the cytoplasm and the C terminal exposed to the periplasm. The N terminal cytoplasmic portion has the ability to bind DNA as illustrated by the transcription activation of ctxAB (gene encoding cholera toxin) by ToxR by directly binding to the DNA repeat element TTTTGAT which is present in 3-8 copies immediately upstream of the ctx promoter (Miller et.al., 1987). This DNA binding domain of ToxR shows homology to OmpR and other response regulator members of the two component regulatory systems (Miller et al., 1987). ToxS is a 19 kDa periplasmic protein. The prevailing hypothesis for the gene regulation by ToxR/ToxS system is based on the fact that ToxR has to be dimerized in order to bind DNA and, ToxS is involved in ensuring the '5Sr P I 74 proper dimerization of ToxR. This is true when ToxR is present in the cell membrane at normal levels. Overexpression of ToxR has been shown to diminished the need for the presence of ToxS, probably due to spontaneous dimerization of ToxR under these conditions (Miller et. al. , 1989; DiRita and Mekalanos, 1991). The third regulatory protein ToxT, is also a transcriptional activator; belonging to the AraC family of transactivators (DiRita and Mekalanos, 1991; Higgins et al., 1992). The ToxT expression is found to be activated by ToxR, and ToxT in turn activates a number of other genes such as the toxin corregulated pilin genes; tcp (DiRita et. al., 1991). The gene coding for ToxT is located within the tcp gene cluster. This raises the possibility that ToxT was originally involved in the regulation of tcp genes, and later gained a more global regulatory role. In addition, ToxT can activate a number of other ToxR dependent genes such as aldA (coding for aldehyde dehydrogenase), tagA (a toxin activated gene for which a function is not yet known), tcpA, and tcpJ in the absence of ToxR (DiRita et. al., 1991) . Although the expression of these genes are dependent on ToxR, they cannot be activated by ToxR alone (DiRita et. al. 1991). This may suggest that the role of Vi 75 ToxR in these gene expression is indirect; by activating expression of toxT. Again, this is an extremely brief summary of the regulatory systems in V. cholerae. A wealth of information has been gathered regarding this system and much is yet to be r'vealed. Apart from the above discussed systems, osmoregulation is seen in the expression of virulence associated alginate capsule synthesis in Pseudomonas aeruginosa (Deretic et. al., 1991), expression of invasion genes in Salmonella typhimurium (Galan and Curtiss, 1990), etc. The alginate capsule synthesis in P. aeruginosa is regulated by a two component regulatory system involving the transcriptional activator protein AlgR which is thought to be activated through phosphorylation by a yet unidentified environmental sensor protein (Deretic et. al., 1989). In contrast to this, the inv genes {inv A, B, C, and D) of 5. cyphimurium, involved in invasion of host cells are regulated by changes in DNA supercoiling brought about by the changes in osmolarity (Galan and Curtiss, 1990). There is significant evidence to show that the level of DNA supercoiling in bacterial cells varies in response to environmental I I 76 conditions such as osmolarity, temperature, nutrient levels, and oxygen levels (Balke and Gralla, 1987; Higgine et. al., 1988; Dorman et. al., 1988) and that changes in DNA supercoiling play an important role in the regulation of gene expression (Hulton et. al., 1990). Current studies are starting to reveal that histone-like proteins of bacteria are involved in these regulatory mechanisms (Hulton et. al., 1990; Schmid, 1990). In B. pertussis, the two component regulatory system, BvgS and BvgA, is involved in regulating virulence gene expression in response to environmental stimuli such as 2 temperature, nicotinic acid and S04 " (Lacey, I960; Schneider and Parker, 1982, Armstrong and Parker, 1986; Melton and Weiss, 1989). A number of virulence genes such as those coding for pertussis toxin, adenylate cyclase toxin, hemolysin, dermonecrotic toxin, filamentous hemagglutinin and fimbriae are under positive control of the bvg locus (Melton and Weiss, 1989). BvgS, showing homology to the histidine protein kinase family, acts as the environmental sensor, and transduces the signal to the response regulator BvgA by phosphorylation. Phosphorylated BvgA in turn 1 I I 77 activates some virulence genes such as fha (coding for filamentous hemagglutinin) by binding to the DNA. However there are many other genes such as the pertussis toxin genes that arf not directly activated by BvgA. Huh and Weiss (1991) were able to demonstrate the presence of a 23 kDa protein that binds to the promoter region of the cya gene coding for adenylate cyclase. This raises the possibility that the role of BvgA in the regulation of ptx and cya genes is indirect, involving a more complex regulatory cascade as present in V. cholerae; Bvg activating the 23 kDa protein which in turn activate ptx and cya genes (Huh and Weiss, 1991). In Yersinia, virulence is regulated by temperature and Ca21". Similarly there are many more examples of environmental regulation of virulence in pathogenic bacteria. Looking at the immense evolutionary adaptations among microbes, it is not hard to imagine that pathogenic bacteria, continually moving through the animal/plant hosts and the environment would regulate their gene expression to suit their life style. In this light, therefore it is almost certain that Legionella being a fresh water bacterium with 78 the capability of infecting humans if given the chance, would have its own schemes and strategies to regulate gene expression accordingly. A recent study revealed that the expression of ompS is differentially regulated in virulent and avirulent strains of L. pneumophila (Fernandez, 1992). According to this study, when virulent, but not avirulent, strains are intracellular or suspended in the tissue culture medium DMEM (a medium containing high levels of sodium chloride sO.85%) the expression of ompS is down regulated. The avirulent mutants however appear to be blind to the environmental changes and show no differential expression of ompS. Previous studies have shown that the avirulent mutants of L. pneumophila are resistant to sodium chloride and can grow on media containing high levels of sodium chloride (aO.65%), whereas the virulent strains are sensitive sodium chloride and can not grow on media containing sodium chloride levels higher than 0.65% (Catrenich and Johnson, 1989). A study by Arakawa et. al. (1992) indicated that there was no correlation between Legionella virulence and sodium chloride tolerance. Nevertheless, it is interesting to note a "I 1 ill I I 79 correlation between regulation of ompS and sodium chloride levels. L. pneumophila, being an aquatic organism, is most probably exposed to very low levels of sodium chloride in its natural habitat. In contrast, when it enters the alveolar macrophages, it is subjected to physiological levels of sodium chloride (0.85%). Therefore, it is tempting to speculate that sodium chloride is an environmental signal mediating the regulation of virulence in L. pneumophila. Statement of research objectives: This study is an attempt to characterize the major outer membrane protein, a potential virulence factor of L. pneumophila both from a genetic and an immunological standpoint. The genetic characterization entails the sequencing of the gene coding for this protein, ompS and studying its regulation in L. pneumophila. It being a potential virulence factor, revealing the mode of regulation of ompS may lead to the unravelling of the more global scheme of regulation of virulence in L. pneumophila. The immunological characterization of OmpS(MOMP) focuses on evaluating this protein's importance as an immunogen in L. pneumophila infections both in humans and in the animal I MI JJ i i 80 model, gui.nea pigs. The study also evaluates the possible candidacy of OmpS in the development of a vaccine against Legionnaires' disease. As written by N.C. Engleberg (1993), "The story of Legionella pathogenesis is not a murder mystery in which all of the plot elements and the evidence are derived from and eventually lead back to the one guilty party. Instead, it is like a Russian novel, with too many characters to remember and a dizzying array of interwoven plots and themes". i I "U| I I I MATERIALS AND METHODS Bacterial Strains and Cloning Vectors The bacterial strains and the cloning vectors used in this study are described in Table 1. Bacterial stock cultures were maintained at -70°C in nutrient broth containing 10% DMSO. Streptomycin-resistant Legionella pneumophila Philadelphia 1 (serogroup 1) (Svir) and the isogenic avirulent mutant strain (Avir) were obtained from the Center for Infectious Diseases, Centers for Disease Control, Atlanta, GA, USA (Hoffman et. al., 1989, 1990). L. pneumophila isolate number 2064 was obtained from the Victoria General Hospital in Halifax, NS and the isogenic avirulent mutant 2064M was created by serially passaging the mutant strain 2064 five times on Mueller-Hinton agar supplemented with hemoglobin and Isovitalex (Fernandez, 1992). The restriction-minus mutant strain of L. pneumophila AA107 was obtained from Dr. Cary Engleberg at University of Michigan Medical School, Ann Arbor, Michigan. 81 82 Table 1. Bacterial strains and cloning vectors used in this s tudy. Strain or Vector Description Reference E. coli JF626 A(lac pro) thi rpsL supE Ron Taylor endA sbcB15 hsdR4 (F' traD36 proAB lacl* lacZM15) BL21 F- ompT hsdSB (rB-mB-) dcm Tabor, gal (DE3) 1990. IJE3 92 supF supE hsdR galK trpR Silhavy et metB lacY tonA al., 1984. MC4100 F' araD139 A (argF-lac)U169 Miller rpsL150 relAl flhD5301 deoCl et. al. , ptsF25 rbsX- 1989. MC4101 MC4100 recAl Miller et. al., 1989. MC4102 MC4101 with Legionella DNA This study cloned in pBluescript Legionella pneumophila Svir StrepR derivative of CDC, USA. Philadelphia 1, serogroup 1. Avir isogenic avirulent mutant CDC, USA. of Svir 2064 Serogroup 1 (Oxford) Fernandez, Human isolate. 1992. 2064M isogenic avirulent mutant Fernandez, of 2064. 1992. • 83 Table 1 ctd. Strain or Vector Description Reference AA107 Strep", res" derivative N.C. of AA100. Engleberg. Plasmids pRS551 bla kan T14 EcoRI smal Simons et. BamEI lacZYA+ al., 1987. PR1S12 pRS551: -.ompS promoter This dtudy pUT mini-Tn5 bla kan; delivery plasmid Lorenzo et for mini-Tn5 Kml al., 1990. pT7-5 pBR332 derived expression Tabor, vector containing T7 1990 promoter and bla pTOMPSl pT7-5 with ompS ORF This study pTOPMS2 pT7-5 with truncated This study ompS ORF pTLP6 CmR rpLs cos' oriT OriV Arroyo et. al., 1993. R pBluescript M13 phagemid Am T7, T3 Strategene promoters, lad, lacZ LaJolla,CA Bacteriophage ARS45 bla-'lacZYA imm21 ind+ Simons et. Kan al., 1987. XRIS12 ARS45::ompS This study M13 tgl30 ss DNA sequencing vectors Kieny et. M13 tgl31 al., 1983. M13 mGPl-2 M13 vector containing T7 Tabor, RNA polymerase gene under 1990. the control of lac promoter. I I 84 All the L. pneumophila strains grown at 3 7°C were maintained on either buffered charcoal-yeast extract (ECYE) agar or in buffered yeast extract (BYE) broth (per 1 liter of distilled water: 10 g yeast extra-'t, 1 g ACES buffer, 1 g a-keto glutaric acid, for agar also add 1.5 g activated charcoal and 1.5% bacto agar: adjust the pH to 6.9 with 6N potassium hydroxide, autoclave and add 4% L-cysteine, 2.5% ferric pyrophosphate). When appropriate, kanamycin (ICN Biochemicals) was added to the media at a concentration of 50 ixg/ml and streptomycin at 200 /xg/rr.l. Broth cultures of L. pneumophila were inoculated from 24 h old BCYE culture. The inoculated broth cultures were incubated statically in flasks for 1-2 hours at 37°C and then transferred to a gyrorotatory shaker and shaken (approximately at 150-200 rpm) for an additional 15-20 h. The Escherichia coli strains were grown on either LB agar or LB broth (per 1 liter of distilled water: 10 g yeast extract, 10 g tryptone, 5 g sodium chloride; for agar also add 1.5% bacto agar) with appropriate antibiotic selection (kanamycin, 50 /xg/ml and ampicillin, 100 /xg/ml) . E. coli strains used in the selective protein radiolabelling I I 85 experiments were grown on the M9 minimal medium supplemented with 0.005% each of isoleucine, leucine, valine, threonine, serine, glutamine, glutamate and aspartate and 0.001 M thiamine. When appropriate, 40 /tl of 100 mM 5-bromo-4chloro- 3-indolyl-B-D-galactopyronidase (X-gal) and/or 40 /xl of 100 mM isopropyl-p-D-galactoside (IPTG) were spread on culture plates. Extraction of Chromosomal DNA Bacterial cultures were grown overnight at 3 7°C in liquid media. Cells were collected by ce.itrifuging 15 ml of culture at 7000 X g for 10 min. The cell pellets were suspended in 0.8 ml of TE buffer, pH 8 (10 mM Tris.Cl pH 8.0, 1 mM EDTA pH 8.0) and transferred to 1.5 ml microfuge tubes (Sarstedt, Germany) . 100 /xl of 20 mg/ml lysozyme (Sigma Chemical Co. St. Louis, MO) were added to the cell suspension and incubated at room temperature for 10 min. Following this incubation, 10 /xl of 20% Sodium Dodecyl Sulfate (SDS) (Sigma Chemical Co.) and 100 /xl of 10 mg/ml Proteinase K (Sigma Chemical Co.) were added and the suspensions were further incubated on a rotating platform for 1 h at 37°C. Then, buffered phenol(made by extracting .' m 86 distilled phenol once with 100 mM Tris pH 8.0 and then with 10 mM Tris pH 8.0 or until the pH of the phenol was close to I r»utral) was added to the top of the microfuge tubes and mixed well by rotating the tubes. The tubes were then incubated for 1 h at 37°C on a rotating platform. Following the incubation, the tubes were centrifuged for 10 min at 7000 X g and the aqueous phase was transferred to a fresh microfuge tube. Then, Chloroform:Isoamyl alcohol(mixed at 24:1 ratio) was added to fill the tubes, contents were mixed well by rotating the tubes by hand for approximately 2 min and the tubes were centrifuged for 5 min at 70 0 0 X g. The aqueous phase was transferred to a fresh microfuge tube, and 10% of its volume was complemented with 3 M Sodium Acetate pH 4.8. The contents in the tubes were mixed well by hand. Then 2.5 X volume 95% ethanol was added to the tubes, mixed well by hand and the DNA was allowed to precipitate at room temperature for 5 min. Following the 5 min incubation, the precipitated DNA was pulled out by using a glass rod and suspended in 0.5 ml of TE pH 8.0. The contaminating RNA in the samples was removed by adding 3 /xl of 10 mg/ml RNase (Boehringer Mannheim Biochemicals, Indianapolis, IN) and incubating for 1 h at 37°C. Then, the DNA was precipitated V I I 87 again by adding 10% 3 M Sodium Acetate and 2.5 X volume of 95% Ethanol and incubating the tubes for 5 min at room temperature. The precipitated DNA was pulled out using a glass rod and washed immediately by gently squirting 95% ethanol. The DNA was then re-suspended in 0 25 ml TE and stored at 4°C. Southern Blot Hybridization Assays These assays were performed as described by Southern (1975) and outlined by Sambrook et. al. (1989). First, the chromosomal DNA samples were digested by using the appropriate restriction enzymes according to the manufacture's instructions and separated by 0.85% agarose gel electrophoresis. The agarose gels were made by dissolving 0.85 g of agarose (SeaKem, FMA BioProducts, Rockland, ME) in 100 ml of 1 X TAE buffer (per 1 liter of distilled water: 242 g Tris base, 57.2 ml glacial acetic acid and 100 ml 0.5 M EDTA pH 8.0). The DNA samples were mixed with 2 /xl of 10 X TAE gel loading buffer (3 0% glycerol and 0.1% bromophenol blue in 10 X TAE buffer) loaded on to the gel and electrophoresed for about 3 hours at 8 0 V. The DNA was then transferred to Nylon membranes (Nytran, I I i 88 Schleicher & Schuell, Keene, NH), by using a gel transfer apparatus (VacuGene XL, Pharmacia) using the protocol outlined by the manufacturer. First, the depurination solution (0.2 N hydrochloric acid) was passed through the gel for 15 min and then the DNA was transferred using 1 M sodium hydroxide for 3 0 min. The membrane was then washed briefly in 2 X SSC (20 X SSC: per 1 litre of distilled water; 175.3 g sodium chloride, 88.2 g of sodium citrate, pH adjusted to 7.0 with 10 N sodium hydroxide) and air dried for about 3 0 min. The air dried membrane was placed between two sheets of 3M Whatman paper and baked under vacuum for 2 h at 8 0°C. The baked membrane was then soaked in 2 X SSC, sealed in a plastic bag (Micro-Seal, Dazey Corporation, Industrial Airport, KS) and incubated overnight in 2 0 ml of Pre-hybridization buffer (12 -xL formamide, 7.2 ml 20 X SSC, 480 /xl of 50X Denhardts solution [(1% bovine serum albumin, 1% polyvinylpyrrolidone, and 1% Ficoll)],4S /xl 0.5 M EDTA, 240 /xl 10% SDS, and 160 /xl of 10 mg/ml single stranded Salmon sperm DNA. Care was taken to remove all trapped air bubbl ; from the plastic bag before sealing it. The radiolabelled DNA probes were made by polymerase chain reaction (PCR). Oligonucleotide primers were used to amplify IP I I * ! 89 the appropriate DNA sequences by PCR. a-[32pUcTP (65 ,aCi in a 50 /xl reaction mixture) was incorporated into the PCR fragments by decreasing the carrier dCTP molar concentration in the deoxynucleotide triphosphate reaction mixture from 200 to 5 0 /xM. The PCR fragments were purified by using the Ultrafree-MC 30,000 NMWL filter units (Millipore Corporation, Bedford, MA). In this purification process, the PCR products were extracted twice with an equal volume of chloroform :isoamyl alcohol (24:1). Then approximately 350 /xl of TE buffer, pH 8.0 was added to the PCR amplified material, placed in a Ultrafree MC filter unit and centrifuged at 3 000 X g until almost all the liquid had passed through the filter. The filter was then washed well with approximately 200 /xl of TE buffer, pH 8.0 and the liquid remaining on the filter was transferred to a fresh microfuge tube. The PCR radiolabelled DNA probes were stored at -20°C in a plexi glass container. Approximately 8 X 10F cpm of denatured, radiolabelled DNA were added to the pre- hybridisation mixture and the DNA binding was allowed to occur at 37°C, overnight. Then the nylon membranes were washed at different concentrations of SSC containing 0.1% SDS. At the end of each wash, the center and the edges of I i 90 the membrane were monitored for radioactivity using a survey meter. The general washing protocol included the following steps. The filter was first washed with 5 X SSC /0.1% SDS at room temperature for 5 min following a 5 min wash with 5 X SSC/0.1% SDS at 55°C, two 5 min washes with 2 X SSC/0.1% SDS at 55°C, one to two 2 min washes with 0.2 X SSC/0.1% SDS at room temperature and if the membrane still appeared to have background radioactivity, one to two 2 min washes with 0.2 X SSC/0.1% SDS at 55°C. At this point, if the background radioactivity was sufficiently low, the membrane was wrapped in plastic wrap, placed in a cassette containing intensifying screens and exposed to Kodak Xomat AR X-ray film for 12-24 h at -70°C. In some instances temperatures as high as 65°C were used for washing in order to increase stringency. Colony Southern hybridization First, isolated bacterial colonies (in pure culture) were grown on Nitrocellulose (NC) disks placed on solid culture media. Four pieces of Whatman 3M paper were cut to appropriate size and shape to fit onto the bottom of four glass Petri plates. Each piece of filter paper was then 91 saturated with one of the following solutions: 10% SDS, denaturation solution (0.5 M sodium hydroxide, 1.5 M sodium chloride), neutralization solution (1.5 M sodium chloride, 0.5 M Tris.HCl pH 7.4), 2 X SSC. The excess liquid was poured off so that no liquid accumulated on the top of filter papers. Then using forceps, the NC disks were picked and placed colony side up on the filter paper saturated with 10% SDS. The filters were exposed to SDS for 3 min and then transferred in a similar fashion onto the filter paper saturated with the denaturation solution. The NC disks were exposed to the denaturation solution for 5 min and transferred to the neutralization solution. After an exposure for 5 min they were transferred to the filter saturated with 2 X SSC for 5 min. Then the NC disks were placed colony side up on a dry sheet of 3M paper and allowed to air dry at room temperature for at least 3 0 min. The air- dried NC disks were then placed between two sheets of dry 3M paper and baked for 1-2 h at 80°C in a vacuum oven. The baked disks were placed in a tray of 2 X SSC, allowed to become thoroughly wet from beneath and submerged in the same solution for 5 min. Then the disks were transferred to a glass beaker containing 200 ml of Prewash solution (5 X SSC, I 92 0.5% SDS and 1 mM EDTA) and incubated with gentle agitation (to prevent the disks from sticking to one another) for 3 0 min at 50CC. Following the incubation, the colony debris was gently scraped off the disks using Kimwipes soaked in pre wash solution and the disks were placed in pre-hybridization solution. Probing with the radiolabelled DNA and the subsequent washing of the disks were performed as with a regular Southern blot hybridization technique. RNA Extraction Bacterial cultures were grown overnight with aeration in liquid media at 3 7°C. 5 ml of the culture were centrifuged at 7000 X g for 10 min and the cell pellets were placed on ice. The RNA was extracted by the hot SDS/ acid phenol method (Hoffman et al., 1992). All the solutions used in the RNA extraction procedure were made using sterile water treated with 0.1% diethyl pyrocarbonate (DEPC, Sigma Chemical Co.). First, the bacterial cells were lysed by adding 3 ml of boiling TES buffer (50 mM Tris.HCl pH 8.0, 1 mM EDTA, 50 mM sodium chloride) containing 1% SDS, vortexed for 5 sec, and I 93 placed in a boiling water bath for 3 0 sec. Then 3 ml of 65°C acid phenol (pH 5.0) was added to the suspension, vortexed briefly and placed in a 65°C water bath for 5 min. The acid phenol was made by equilibrating double distilled phenol with equal volumes of 500 mM Sodium Acetate followed by 2 to 3 changes of 50 mM sodium acetate pH 4.8. After the 5 min incubation, the tubes were centrifuged at 7000 X g for 10 min and the aqueous layer was carefully removed to a fresh tube leaving behind the contaminating protein and DNA in tht, phenolic phase. The phenolic extraction was repeated once more in the same manner, and then the aqueous layer was extracted once more with an equal volume of chloroform: isoamyl alcohol (24:1) in order to remove the left over phenol. The RNA was then precipitated by adding 10% volume of 3 M sodium acetate and 2.5 X volume ice cold 95% ethanol. After incubating for 30 min at -70°C to maximize the RNA precipitation, the samples were centrifuged at 13,000 X g for 10 min. The supernatants were discarded and the pellet was washed with 70% ethanol. The resulting RNA pellet was vacuum dried and re-suspended in 50 /xl of sterile DEPC treated water. An aliquot of the RNA sample was diluted appropriately and the absorbance at 260 nm was determined. 94 From this, the concentration of the RNA preparation was determined and 20 fxg aliquots of RNA were vacuum dried in sterile microfuge tubes. The vacuum dried RNA samples were stored at -20°C until use. Northern Blot Hybridization Assays These assays were performed as described by Kroczek and Siebert (1990). The RNA was first separated by formaldehyde- agarose gel electrophoresis. In order to prepare the gel, 0.5 g of agarose (SeaKem, FMC BioProducts, Rockland, ME) was dissolved in 36.7 ml of distilled water and 5 ml of 10 X 4- morpholinepropanesulfonic acid (MOPS)/EDTA buffer (per 300 ml: 31.4 g MOPS, 6 ml 0.5 M EDTA, pH adjusted to 7.0 with 10 N sodium hydroxide). Before casting the gel, 8.3 ml of 37% formaldehyde were added to the melted agarose. The gel was allowed to solidify and then pre-electrophoresed in 1 X MOPS/EDTA buffer for 3 0 min at 60 V before loading the RNA samples. Before loading, the RNA samples were re-suspended in 11 /xl of gel loading buffer consisting of 2.2 /xl of Buffer A (294 /xl 10 X MOPS/EDTA, 706 /xl DEPC treated sterile distilled water), 4.8 /xl formaldehyde/formamide (89 /xl formaldehyde, 250 /xl formamide) , 2 /xl gel loading buffer (50 H 95 mg ficoll, 5 mg xylene cyanol, 5 mg bromocresol green, 322 /xl buffer A, 178 /xl 37% formaldehyde, 500 /xl formamide) and 2 /xl of 0.5 mg/ml ethidium bromide. The samples were then heated to 65°C for 10-15 min, quickly quenched on ice and loaded onto the gel. In the same manner, 3-4 /xl of a 0.24- 9.5 kilobases RNA ladder (Gibco/BRL) were treated and loaded onto the gel. The RNA samples were then electrophoresed for 5-6 h at 60 V. The gel was viewed using a UV transilluminator, photographed and then the RNA was transferred onto a nylon membrane (Nytran, Schleicher and Schuell, Keene, NH) using a gel transfer apparatus (VacuGene XL, Pharmacia) according to manufacturer's instructions. For the transfer, first the formaldehyde was removed with DEPC water for 5 min. Next, an alkaline solution (50 mM sodium hydroxide, 10 mM sodium chloride) was passed through the gel for 5 min. The alkaline solution was neutralised by passing through a solution of 0.1 M Tris.HCl for 5 min. Finally, the RNA was transferred using 20 X SSC for 30 min. Following the transfer, the membrane was wrapped in plastic wrap and exposed to UV light at the high setting for 5 min. Following this, the membrane was baked under vacuum for 1-2 h at 80°C. I 96 The probing and washing of the membrane were carried out essentially as described for the Southern blot hybridization assays, except for the composition of the pre- hybridization solution. For Northern blots, the pre- hybridization solution consisted of 2 ml of 50 X Denhardts reagent, 5 ml of 20 X SSC, 0.2 ml of 10% SDS, 0.4 ml of 10 mg/ml single stranded salmon sperm DNA and 12.6 ml of sterile distilled water. Cloning and Sequencing of ompS The cloning of ompS was carried out using reverse genetics. The work pertaining to the cloning and sequencing of ompS has been published by Hoffman et. al. (1992b). A copy of this publication is included in the Appendix. ompS was cloned using degenerate oligonucleotides synthesized from the N-terminal amino acid sequence of OmpS (Hoffman et.al., 1992a, b). In Southern blot hybridization analyses, one oligonucleotide within the pool of degenerate oligonucleotides, (24b), hybridized strongly under high stringency conditions to single sites in the L. pneumophila genomic DNA digested with various restriction enzymes. This I 97 provided information regarding the relative sizes of the DNA fragments containing the ompS DNA. Therefore L. pneumophila genomic DNA was digested with the relevant restriction enzymes, separated by agarose gel electrophoresis, the relevant size DNA fragments were excised and ligated into the cloning vector pBluescript (Stratagene, La Jolla, CA) . These clones were confirmed for the presence of ompS DNA by Southern blot hybridization. The insert DNA from pBluescript was subcloned into M13 vectors and sequenced using the dideoxy chain termination method (Sanger et al., 1977) . The sequencing protocol has been explained in detail in a latter part of the methods section. Primer Extension Reaction The primer extension was performed by hybridizing the end-labelled oligonucleotide primer H25 (5' GTCCATACTGGACC- CATAGTACC3') to L. pneumophila RNA. The end-labelling of H25 was done using T4 polynucleotide kinase (New England Biolabs) and the end-labelled oligonucleotide was purified using a Sep-Pack CL8 cartridge (Millipore Corporation, Milford ,MH). The protocol used for the end-labelling reaction is as follows. 3 /xl of 10 X kinase buffer (0.5 M I I I 98 Tris.HCl pH 7.6, 0.1 M magnesium chloride, 1 mM EDTA), 1 /xl of DTT, 1 /xl of 3 0 mM spermidine, 12.5 /xl of y32P-ATP (125 /tCi) were added to 200 ng of oligonucleotide primer and the reaction volume was adjusted to 28 /xl with sterile distilled water. Two /xl of T4 polynucleotide kinase (20 U) were added to this mixture, mixed well and the tube was incubated at 37°C for 1 h. The kinase reaction mix was then separated by 20% PAGE. The gel mix was prepared by adding 25 g urea, 25 ml 4 0% acrylamide (38 g acrylamide and 2 g bis-acrylamide per 100 ml), 5 ml 1 M Tris borate buffer (10 X TBE, in 1 L of distilled water; 108 g Tris base, 55 g Boric acid and 40 ml 0.5 M EDTA pH 8.0) and 3 5 mg ammonium persulphate. Fifteen /xl of TEMED were added to the mixture immediately prior to casting and the gel was casted in a 18 X 16 cm gel apparatus. The gel was allowed to polymerize and then pre- electrophoresed for 1 h at 300 V prior to loading the samples. Four - five /xl of gel loading buffer (95% formamide, 20 mM EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol) was added to the kinase reaction and the entire sample was loaded onto the gel. The samples were electrophoresed for 3-4 h at 3 00 V. After electrophoresis, the gel was wrapped in plastic wrap and exposed to Kodak I I 99 film for 1 min. The resulting autoradiogram was superimposed on to the gel and the. radioactive band (end-labelled primer) was excised out of the gel and placed in a sterile tube containing 3 ml of Sep-Pack elution buffer (100 mM Tris.HCl pH 8.0, 5 mM EDTA, 0.5 M sodium chloride). The gel slice was crushed and incubated in elution buffer over night at 60DC. During this incubation DNA is eluted from the gel slice into the elution buffer. A Sep-Pack cartridge was prepared for DNA purification by first attaching it to a 10 ml syringe (without the plunger) and passing 10 ml of methanol through the cartridge in a drop wise manner. The cartridge was then washed by passing 10-20 ml of sterile distilled water through the cartridge in the same way as methanol. Next, the DNA solution was drained into the syringe and slowly passed through the Sep-Pack cartridge. The gel slice was rinsed three times with 1 ml of sep-pack elution buffer and the buffer was passed through the cartridge. The cartridge was rinsed with 20 ml of sterile distilled water and the DNA was eluted by passing 3 ml of methanol-TEAAc solution (100 mM Triethylamine-acetate buffer, pH 7.3, 50% methanol; the procedure for the preparation of triethylamine-acetate buffer is given in the Appendix)through the cartridge. The I 100 eluate from the cartridge was collected in 3 microfuge tubes with 1 ml in each tube. The DNA in the eluate was pelleted by drying the liquid in a speed vac concentrator. The DNA pellet from each tube was dissolved in 10 /xl of DEPC treated water, pooled together, and the radioat. '-ivity was measured using a liquid scintillation counter. For the primer extension reaction, 10-60 /xg of RNA was added to 12 X 106 cpm of kinased oligonucleotide primer and the total volume of the mixture was brought up to 2 7 /xl with DEPC treated distilled water. 3 /xl of annealing buffer (100 mM Tris.HCl, 2.5 M potassium chloride) was added to thi.s and the tubes were placed in a thermocycler programmed for annealing at 80°C for 5 min, 65DC for 5 min, 42°C for 10 min and 37°C for 20 min. After this reaction was completed the annealed material was precipitated by adding 70 /xl of DEPC water, 10 /xl of 3 M sodium acetate , 260 /xl of ice cold ethanol and placing the tubes at -2 0°C for 3 0 min. The precipitate was washed with 70% ethanol, pelleted and vacuum dried. The pellet was suspended in 7 /xl of DEPC treated water and then 13 /xl of reverse transcriptase buffer (1 M Tris.HCl, pH 8.3, 1 M magnesium chloride, 1 M DTT, 50 mM 101 each of dATP, dCTP, dGTP, TTP, 2 mg/ml actinomycin D per 1 ml of DEPC treated water) and 200 U of reverse transcriptase enzyme (Gibco/BRL) were added. The mixture was then incubated for 1 h at 42°. Following the incubation, the free RNA was digested by adding 1 /xl of 0.5 M EDTA and 1 /xl of 10 mg/ml RNAse (bovine pancreas; Boehringer Mannheim) and incubating for 3 0 min at 3 7°C. The volume of the reaction was adjusted to 10 0 /xl with TE. The cDNA was purified using phenol:chloroform:isoamyl (1:24:1) extraction followed by ethanol precipitation. The cDNA was then pelleted by centrifugation, vacuum dried and the dried pellet was suspended in 10 /xl of gel loading buffer (95% formamide, 20 mM EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol). The cDNA samples were resolved on a polyacrylamide gel beside an ompS sequencing ladder generated using the oligonucleotide primer H25. The sequencing gel electrophoresis was performed as described later in this section. Extraction of Plasmid DNA Broth cultures were inoculated with a single bacterial colony and incubated over night with shaking (100-2 00 rpm) 102 at 37°C. For miniplasmid preparations, 5 ml of bacterial cul-ure were centrifuged for 10 min at 5000 X g and the pellet was suspended in 200 /xl of Solution 1, containing 50 mM Glucose, 10 mM EDTA and 25 mM Tris.HCl, pH 8.0. The tube was vortexed and placed on ice for 5 min. To each tube, 4 00 /xl of Solution 2, consisting of 0.2 N sodium hydroxide and 1% SDS in water was added, gently vortexed and placed on ice for 5 min. Three hundred /xl of 5 M potassium acetate, pH 4.8 was then added, gently mixed, and incubated for 5 min at - 70°C. The tube was centrifuged at 5000 X g for 15 min and 750 /xl of supernatant was transferred to a fresh tube. The plasmid DNA was precipitated by adding 4.5 ml of ice cold isoproponol to the supernatant and incubating the tubes at -70°C for 5 min. The tube was then centrifuged at 5000 X g for 15 min and the pellet washed with 70% ethanol, vacuum dried, and resuspended in 100 /xl of TE, pH 8.0. The RNA in the sample was digested by adding 1 /xl of 10 mg/ml RNAsel and incubated for 1 h at 37°C. Plasmid DNA was purified by a phenol extraction, a phenol:chloroform:isoamyl (24:24:1) extraction and a chloroform: : -^amyl (24:1) extraction. Then the DNA was precipitated by adding 10% sodium acetate and 2.5 X volume 95% ethanol and incubating at -70°C for 3 0 min. I I • 103 The DNA was then pelleted by centrifugation (10 min), washed twice with 70% ethanol and vacuum dried. The plasmid DNA was then re-suspended in T.E. pH 8.0 and stored at 4°C. Transformation of Plasmid DNA into bacterial cells a: Preparation of competent cells: Appropriate E. coli strains were grown in LB broth overnight at 37°C with aeration. The next day, 50 ml of fresh LB broth were inoculated with 1 ml of the overnight culture and incubated under the same conditions. Once the culture had reached an OD6S0 of 0.3, 4 0 ml were centrifuged at 5000 X g for 2 min and the cells were re-suspended in 20 ml of sterile 0.1 M calcium chlor'.de. The cell suspension was then incubated on ice for 2 0 min. After this incubation, the cells were again centrifuged at 5000 X g for 2 min and re-suspended in one tenth of the original volume (i.e. 4 ml). These competent cells were either used immediately or anytime within a 36 h period. If the cells were not used immediately, they were stored at 4°C. I i 104 b: Transformation: 10-20 /xg of plasmid DNA were added to 100 /xl of competent cells and the cells were incubated on ice for 1 h. When transforming a ligation mixture, the entire ligation mixture was added to 100 /xl of competent cells. When ligating DNA in low melting point agarose, the ligation mixture was first heated to 65°C to melt the agarose. 50 /xl of sterile distilled water was added to the melted agarose and the ligation mixture was cooled on ice. At this point 100 /xl of competent cells were added to the ligation mixture and incubated on ice for 1 h. Following the 1 h incubation, the cells were heat shocked by heating the transformation mixture to 42°C for 3 min and immediately immersing in ice. 400 /xl of LB broth were added to the transformed cell, and allowed to grow for 0.5-1 h without any antibiotic selection. Following this, the cells were plated on appropriate culture media with antibiotic selection. Selective radiolabelling of OmpS expressed in E. coli JF626 A 1.5 kb EcoRl fragment and 1.3 kb EcoRI/Pstl fragment of ompS DNA, recovered from the vector pH151 (figure 1), were cloned into the T7 RNA Polymerase/Promoter vector pT7-5 I I 105 (Tabor, 1990). In order to accomplished this, the ompS DNA and the pT7-5 vector DNA were digested using appropriate restriction enzymes according to manufacture's instructions, and separated by 0.85% agarose (low gelling temperature agarose, SeaPlaque, FMC BioProducts, Rockland, ME) gel electrophoresis for approximately 2 h at 75 V. The portions of the gel containing the appropriate DNA fragments were excised, melted at 65°C, and the DNA recovered. The vector and ompS DNA were ligated at 15°C, overnight using 8 00 U of DNA Ligase (New England Biolabs) and then transformed into competent E. coli JF626 cells. Transformed cells were then plated on LB agar containing 100 /xg/ml ampicillin (ICN Biochemicals, Aurora, OH) and incubated overnight at 3 7°C. Transformants were confirmed by colony Southern hybridization and the true positive clones were chosen for the selective radiolabeling of OmpS. An individual true positive E. coli JF626 (pT7-5 : : ompS) colony was grown in LB broth overnight with aeration at 3 7°C. The overnight culture was then diluted 1:100 in fresh LB broth and incubated at 3 7°C until the culture attained an ODS90 of 0.5. Three ml of this culture were then 106 centrifuged at 5 000 X g and the cell pellet was re-suspended in 5 mi of Mg broth containing the amino acid mix and thiamine (see Appendix for the composition). The culture was incubated for 1-2 h at 3 7°C with gentle shaking (approximately 10 0 rpm). The cells were then infected with the M13 phage mGP-1-2 harboring the T7 RNA polymerase gene at a multiplicity of infection (M.O.I) of 10. Fifty /xl of 0.1 M Isopropyl-p-D-galactoside (IPTG) (ICN Biochemicals) were added to induce expression of T7 during a 30 min incubation. 50 /xl of 20 mg/ml Rifampicin (Sigma Chemical Co.) were added to the culture to inhibit non T7 transcription. After a 30 min incubation, 10 /xCi of 35S labelled Methionine (ICN Biochemicals) were added to 1 ml of culture which was then incubated for 5 min at 3 7°C. The cells were pelleted by centrifugation and all the supernatants were carefully removed. The cell pellet was re- suspended in 120 /xl of Cracking buffer (60 mM Tris.HCl, pH 6.8, 1% 2-mercaptoethanol, 1% SDS, 10% glycerol, and 0.01% bromophenol blue) , and heated to 95°C before loading 4 0 /xl onto a 7.5-15% gradient SDS Polyacrylamide (SDS-PAGE) gel. The SDS-PAGE gel was run at 15 mA constant current until the protein sample entered the running gel and then increased to I 107 30 mA. The gel was run until the dye front reached the bottom of the gel. The gel was then removed from the glass plates stained with Coomassie Brilliant Blue, destained, transferred to a piece of 3M filter paper and dried on a gel drier for 1 h. The dried gel was placed in a cassette and exposed to Kodak Xomat AR X-ray film for approximately 24 h. An autoradiogram was analyzed after overnight exposure to Kodak X-ray film. Sodium chloride challenge and preparation of L. pneumophila cell extracts L. pneumophila cells were harvested from 18 h aerated BYE broth cultures grown at 3 7°C. An equal volume of the culture was challenged with 0.85% sodium chloride for 1 h prior to harvesting. Cells were washed once with 50 mM Tris- HCl (pH 7.5), resuspended in sonication buffer (50 mM Tris- HC1 pH 7.5, 1 mM EDTA, 1 mM dithiothriotol [DTT], 50 mM sodium chloride, 0.1 mM para-methyl sulfonyl fluoride [PMSF]) and sonicated at setting 9 using a sonifier cell disrupter model W140 (Heat systems- Ultrasonics, Inc, Plainview, NY) for 1.5 min with 3, 3 0 sec incubations on ice in between. The sonicates were then ultracentrifuged for 1 h 108 at 100,000 X g. Supernatants were aliquoted and frozen at - 70°C until used. Gel Mobility Shift Assays The promoter and upstream region of ompS were amplified and radiolabelled by polymerase chain reaction (PCR) using two synthetic oligonucleotide primers F6 (ATCTTGAATTCCACT- GTA) and R6 (GATTAGCGGATCCTAACAGAA) (Figure 3). The oligonucleotide primers were synthesised at the Gene Probe Laboratory at Dalhousie University. The PCR amplified DNA was purified by two chloroform:isoamyl alcohol (24:1) extractions and then the unincorporated nucleotides were removed by using a filter unit with a 30,000 Da molecular weight cut off limit (Ultrafree-MC, Millipore Corporation, Bedford, MA). The radioactivity of the DNA was determined using a liquid scintillation counter. In some experiments the PCR amplicon was cleaved into two fragments (172 bp and 113 bp) at a Bell site located at -60 bp position (Figure 2) by using the Bell restriction enzyme. Mobility shift assays (also known as gel retardation assays) were performed in order to investigate the presence 109 of a DNA binding protein that binds to the ompS promoter DNA. Mobility shift assays contained 20-25 /xg of L. pneumophila protein extracts (from high speed supernatants) in binding reaction mixes containing 3 ul of 10 mg/ml ds Salmon sperm DNA, 2 ul of 10 mg/ml BSA, 1 ul of glycerol and 2 ul of 10 x binding buffer (400 mM Tris.HCl pH 7.5, 300 mM magnesium chloride, 1 mM EDTA, 1 mM DTT, 1 M sodium chloride) and 10,000 cpm of radiolabelled ompS promoter DNA. The total volume of the reaction mix was then adjusted to 2 0 /xl with sterile distilled water. The L. pneumophila cell extracts were omitted from the control tube in order to determine the electrophoretic mobility of free DNA with no protein bound. All the reaction tubes were then incubated for 15 min at 37°C. After the incubation, 2 /xl of loading buffer (0.3% glycerol and 0.01% bromophenol blue in 10 X Tris borate buffer) were added to the reaction tubes and the DNA : protein mixture was separated by non-denaturing polyacrylamide gel electrophoresis. The non-denaturing acrylamide gel was prepared by adding 10.7 ml of 30% acrylamide (29% acrylamide and 1% bis-acrylamide) to 27 ml of distilled water, 2 ml of 10 X TBE, and 0.25 ml of ammonium persulphate (APS). The gel mix was de-gassed for 5 110 min, 20 /xl of TEMED was added and immediately cast. Following polymerization the gel was pre-electrophoresed in 0.5 X TBE buffer for 1 h at 20 mA before loading the samples. The samples were then loaded and electrophoresed for 3 h at 2 0 mA. Following that, the gel was removed from the glass plates onto a piece of 3M Whatman paper and dried by vacuum for 1 h at 8 0°C. The dried gel was placed in a cassette containing intensifying screens and exposed to Kodak Xomat AR X-ray film for 6-12 h at -70°C. In competition assays, PCR amplified unlabelled ompS promoter DNA was mixed with the PCR amplified radiolabelled ompS promoter DNA. The unlabelled DNA was added to the reaction mixes at concentrations of 0.1, 1 and 3.5 ug/ml prior to incubation. In addition to this, a radiolabelled internal fragment of the ompS gene was also used in these assays in order to further confirm the specificity of binding of the DNA binding protein to ompS DNA. The internal fragment of the ompS gene was obtained by PCR amplifying ompS DNA using the 2 oligonucleotides 24b and R5 (5' AGCCTTAGATAGGATCCCTAACTTAA 3') which have been made to correspond to 2 internal regions of the ompS gene. I 111 To identify the region of binding of the DNA binding protein to ompS DNA, the 286 bp ompS promoter amplicon was cleaved at the Bell site located at the -60 bp position into two fragments of 114 and 172 bp. These cleaved DNA amplicons were then used in the mobility shift assays. To further characterize the region of binding, two synthetic oligonucleotides F7 (GCTGGCGTATAAATCAATAA) and F8 (CAGCAGATACTGATCAAATC) wero made to the promoter region of ompS. These forward oligonucleotides were used with the R6 reverse oligonucleotide to PCR amplify and radiolabel the ompS promoter region immediately upstream and immediately downstream of the -60 bp position. These DNA amplicons were then used in the mobility shift assays. To study the kinetics of the NaCl effect on the affinity of the DNA binding protein for ompS DNA, L. pneumophila cultures were grown in BYE broth with aeration for 18 h at 37°C. These cultures were then challenged either with 0.85% NaCl or 10 0 ug/ml chloramphenicol for time intervals of 0, 2, 5 and 10 min prior to harvesting the cells and preparing the cell extracts. These cell extracts were then used in the mobility shift assays. I 112 Southwestern Blot Assays L. pneumophila cell extracts were fractionated by 15% SDS polyacrylamide gel electrophoresis. The running gels were prepared by mixing 7.5 ml 30% acrylamide, 1.5 ml Tris pH 8.8, 150 /xl 10% SDS, 50 /xl APS and 3.55 ml distilled water. After degassing this mixture for 5 min, the gel was cast immediately upon the addition of 3 /xl of TEMED. After casting the gel, a layer of water saturated butanol was added to the top of the gel and the gel was allowed to solidify. Subsequently, the butanol overlaying the solidified running gel was poured off and the stacking gel was poured after inserting the combs. The stacking gel consisted of 1.33 ml 30% acrylamide, 0.5 ml Tris pH 6.8, 150 /xl 10% SDS, 6.1 ml distilled water, 50 /xl 15% APS and 5 /xl of TEMED. About 3 0 /xl of gel loading buffer were added to the protein samples and the samples were then boiled for 2 min prior to loading on the gel. The samples were electrophoresed at 15 mA in 1 X Tris glycine buffer (per 4 1: 12 g Tris base, 57 g glycine, 4 g SDS) until the dye front reached the running gel and then at 3 0 mA. The gel was then removed from the glass plates and the proteins were transferred to nitrocellulose membrane (NC, Schleicher and 113 Schuell, Keene, NH) as described by Nelson et. al.(1988). Before transferring the proteins, the gel was placed in transfer buffer and rocked gently for about 15 min. Before use, the nitrocellulose membrane was boiled in distilled water and then immersed in 20 X SSC. The nitrocellulose membrane was then placed on top of the equilibrated gel, taking care not to leave any air bubbles between the gel and the membrane. The gel and the nitrocellulose membrane were then sandwiched between 3M Watman filter paper and sponges previously soaked in transfer buffer, and placed in a Transblot apparatus (BioRad) containing the transfer buffer. The transfer was performed for 90 min at 50 V. Throughout the transfer, the buffer in the transblot apparatus was cooled by running cold tap water through the cooling system. After the transfer the membrane was incubated in binding buffer (10 mM Tris-HCl pH 7.0, 1 mM EDTA, 0.02% BSA. 0.2% Ficoll, 0.2% polyvinyl pyrrolidine and 0.05 M NaCl) for 15 min at room temperature. The membrane was then sealed in a plastic bag (Micro-Seal, Dazey Corporation, Industrial Airport, KS) taking care not to leave any air bubbles, and then incubated with 105 cpm/ml of radiolabelled ompS promoter DNA amplicon for 1 h at room temperature. Following 114 the incubation, the membrane was washed with binding buffer containing increasing concentrations of NaCl (0.05 M - 0.25 M) and analyzed by autoradiography. In order to determine the molecular weight of the DNA binding protein, the protein molecular weight standards (BioRad) were loaded alongside with the protein samples, electrophoresed and then transferred onto Immobilon membrane (millipore corporation). Prior to use, the Immobilon was first wetted for a few seconds in 100% methanol and submerged in distilled water and then dipped in transfer buffer. After transferring the protein, the Immobilon was washed in deionized water for 5 min and stained with 0.1% Coomassie Brilliant Blue R-250 in 50% methanol for 5 min. Then the Immobilon was de-stained using a solution of 10% acetic acid, 50% methanol for 5-10 min at room temperature. The membrane was then rinsed in deionized water for 5-10 min and air dried. The molecular weight of the DNA binding protein was determined by superimposing the stained Immobilon membrane on the autoradiogram obtained by southwestern blot assay. I 115 ompS promoter:lacZ fusions The ompS promoter and upstream region were amplified by PCR using the synthetic oligonucleotides FG and Rs. The amplicons were then restricted with EcoRl and BamEl, ligated into the operon fusion vector pRS551 and transformants were selected for kanamycin and ampicillm resistance. The clones were further confirmed by Southern blot hybridization using a radiolabelled ompS promoter amplicon and by direct sequencing. The plasmid containing the ompS promoter:lacZ fusion (pRIS12) was electroporated into L. pneumophila Svir and Avir. Electroporation was accomplished in a 0.2 mm gap cuvette at 2.5 KV, 200 Q, 25 uF using a BioRad Gene pulser. The transformants were selected for Km resistance. The p-gal production by these transformants was assayed as described by Miller (1972). p-galactosidase assays 1 ml of log phase bacterial cells were pelleted by centrifugation at 7000 X g for 10 min. The pellet was re- suspended in 1 ml of Z-buffer (per 1 liter of distilled water: 16.1 g disodium phosphate(Na2HP04. 7H20), 5.5 g mono sodium phosphate (NaH2P04.H20) , 0.75 g potassium chloride, L 116 0.25 Magnesium Sulfate and 2.7 ml 2-Mercaptoethenol, pH adjusted to 7.0). 50-75 /xl of this cell suspension were used in the assay. The following were added to a clean glass test tubes,- 50-75 /xl of the cell suspension, 500 /xl of Z-buffer, 2 drops of chloroform, 1 drop of 0.1% SDS. The contents were vortexed vigorously and the tubes were placed in a 28°C water bath. To each tube, 0.2 ml of 4 mg/ml ONPG solution were added (ONPG dissolved in 1 M phosphate buffer), mixed well by hand and the tubes were placed back in the 28°C water bath. When the liquid in the tubes turned to pale yellow colour, 0.5 ml of 1 M sodium carbonate was added to each tube to stop the reaction. The time between the addition of ONPG and the termination of the reaction was recorded by using a lab timer. The absorbance at 420 nm and 550 nm for the liquid as well as the absorbance at 600 nm for the cell suspension used were determined and the p- galactosidase levels were calculated using the following equation. P-gal levels (in Miller Units) = 1000 X 0D^P - (1.75 X OD5:0) t X v X OD600 117 t = Time between addition of ONPG and 1 M sodium carbonate solution (time of reaction in minutes). v = Volume of cells used in /xl) . DNA sequencing Based on the dideoxy method described by Sanger (1977), both double stranded DNA sequencing using PCR amplicons and single stranded DNA sequencing using M13 vectors were used. The DNA sequencing was performed using a T7 DNA sequencing kit (Pharmacia) and the sequencing reactions consisted of 3 basic steps; annealing, labelling and termination. The labelling and termination steps were essentially the same for both double stranded and single stranded sequencing protocols and the difference was in the annealing step. Double stranded DNA sequencing Double stranded DNA sequencing was used to analyze the fusion junction in the ompS promoter: lacZ fusion. A 20 base pair oligonucleotide primer (LacRj = 5' TTCCCAGTCACGACGTTGTAA 3') was created in reverse orientation to a lacZ sequence present in the operon fusion vector pRS551. The ompS promoter: lacZ fusion and some of the adjoining sequences • 118 were amplified by PCR using the oligonucleotide primers Fs and lacRx. The PCR amplicon was purified by using 2 chloroform: isoamyl alcohol extractions followed by Gene Clean (BiolOl Inc, La Jolla, CA). 10 ul of purified DNA were used in the sequencing reaction. The ompS:lacZ fusion junction was sequenced from both directions; i.e. using the Fs primer for sequencing in the forward direction and the LacRx primer for sequencing in the reverse direction. The annealing of the primers to template DNA (PCR amplicon) was performed as described by Drebot and Lee (1994) . For the annealing step, 10 /xl of template DNA, 1 /xl of primer DNA, 1 /xl of dimethyl sulfoxide (DMSO) and 2 /xl of annealing buffer were mixed in a sterile microfuge tube and then heated to 94°C for 3-5 min. After heating, the tubes were immediately immersed in dry ice. The contents were then thawed and 3 /xl of dATP labelling mix, 50 /xCi of (a-3ES)dATP (ICN Biochemicals) and 2 /xl of T7 polymerase enzyme (diluted 1:4 in enzyme dilution buffer) were added. The contents were mixed and the tubes were incubated at room temperature for 2-3 min. Meanwhile, 3.5 /xl of each of the four termination mixes were added to separate microfuge tubes and incubated 119 at 3 7°C for 1 min. Then 4.5 /xl of reaction mix were added to each tube containing the termination mix and incubated for further 5 min at 37°C. After the 5 min incubation, the sequencing reaction was stopped by adding 4 /xl of the stop buffer and the tubes were stored at -20°C until used. Single stranded DNA sequencing Single stranded DNA sequencing was used for sequencing potential clones containing the DNA coding for the DNA binding protein and for obtaining a sequencing ladder for DNAsel protection assays. In single stranded sequencing, the DNA to be sequenced was first cloned in to a M13 phage (Tg 130 and Tg 131). The M13 cloning and sequencing handbook (Amersham) was used as guide for cloning DNA into M13 vectors for sequencing. a: Preparing replicative forms (RF) of single stranded M13 phage: Single stranded M13 DNA was transformed into competent E. coli JF626 cells. Preparation of competent cells and the transformation was carried out as described previously. The following procedure was used when plating the transformed 120 cells. 40 /xl of 100 mM IPTG, 40 ul of 2% X-gal in dimethyl formamide, 200 /xl of fresh E. coli JF626 cells were added into a clean test tube. The transformation mix was also added to this after heat shock together with 3 ml of molten H top agar(per 1 liter of distilled water: 10 g bacto tryptone, 8 g sodium chloride and 8 g bacto agar) kept at 42°C. The contents were mixed by rolling the tubes between the palms of the hands and poured immediately onto a pre- warmed LB agar plate. The agar plates were left at room temperature until set and then inverted and incubated overnight in a 37°C incubator with humidity. Following the incubation, single blue colour plaques were vsed to inoculate 1 ml cultures of fresh E. coli JF626 cells obtained by adding 0.1 ml of an overnight E. coli JF626 culture per 10 ml of fresh LB broth. The inoculated cultures were incubated at 3 7°C for 4-5 h with gentle shaking (approximately 100 rpm). The cultures were centrifuged in microfuge tubes for 5 min and 1 ml of the supernatant was added to a flask containing 100 ml of fresh LB broth and 1 ml of overnight E. coli JF626 culture. The flask was incubated for 4-5 h at 37°C with gentle shaking, the cells were pelleted by centrifugation and the RF DNA was prepared I 121 by using the standard plasmid extraction procedure described earlier. b. Cloning into RF DNA and preparation of ss template DNA for sequencing: The RF DNA and the DNA to be sequenced were restricted with appropriate restriction enzymes and ligated together. The ligation mix was then transformed into competent E. coli JF626 cells and plated on top agar as described previously. Upon overnight incubation at 37°C, isolated white plaques (which are the ones containing insert DNA) were picked and used to inoculate tubes containing 1.5 ml of fresh E. coli JF626 culture (obtained by adding 1 ml of overnight culture to 100 ml of fresh LB broth). One blue plaque was also picked to serve as the vector control. The tubes were then incubated at 37"C with shaking for 5 h. The contents were then transferred to microfuge tubes and centrifuged for 5 min. The supernatants were then transferred to fresh microfuge tubes taking care not to pick up any cells. 200 /xl of polyethylene glycol 6000 (PEG)/sodium chloride (20% PEG, 2.5 M sodium chloride, stored at 4°C) were added to the supernatants, mixed well and left to stand at room 122 temperature for 15 min. The tubes were then centrifuged for 2 min, all the supernatant was removed from the tubes and 100 /xl of TE buffer were added to the phage pellet. Then 50 /xl of buffered phenol, pH 8.0 were added to the tubes, vortexed for 15-20 sec and left to stand at room temperature for 15 min. The tubes were then centrifuged for 3 min and the aqueous layer was transferred to a fresh tube. The ss M13 DNA was then precipitated by adding 10% 3 M sodium acetate and 2.5 X volume ice cold 95% ethanol. The DNA pellet was washed with 70% ethanol, vacuum dried and re- suspended in 20 /xl of TE buffer. 2 /xl of the DNA preparation were electrophoresed on 0.85% agarose gel using 1 X TAE buffer and the DNA from white plaques were compared with the DNA from the blue plaque in order to confirm the presence of the insert DNA. 7 /xl of the DNA preparation were used for sequencing. In the annealing step, 7 /xl of template DNA was mixed with 1 /xl of primer DNA and 2 /xl of annealing buffer, and incubated for 2 min at 65°C in an aluminium heating block. After this incubation, the block containing the tubes was removed and I 123 allowed to cool to room temperature on the lab bench. The sequencing reaction was performed as described previously. Sequencing gel electrophoresis A 20 X 40 cm gel apparatus with 20 well shark tooth combs was used for sequencing. The glass plates were cleaned well with soap followed by 95% ethanol. On the smaller plate, a few drops of Rain Away (Wynn's Canada Ltd., Mississauga, ON) was applied evenly using a paper towel. Then the two spacers were placed between the glass plates and the plates were taped together carefully using adhesive tape (3M) . 36 g of urea, 11.25 ml of 40% acrylamide (38% acrylamide, 2% bis-acrylamide), 7.5 ml of 5 X TBE and 10 ml of distilled water were added to a 100 ml of glass beaker and stirred under low heat on magnetic stirrer. After the urea had completely dissolved, the volume of the mixture was adjusted to 75 ml with distilled water, and stirred until ready to use. Then 3 80 /xl of 10% APS and 15 ul of TEMED were added to the mixture and the gel was poured immediately using a 60 ml syringe taking care not to trap any air I 124 bubbles in the gel. The combs were placed in the gel upside down in order to make space for the wells, the glass plates were clamped together using spring clamps and the gel was left to polymerise at an approximately 15-20° angle. After the gel had polymerized, the clamps, tape and combs were removed. The plates were rinsed with water and the gel was mounted on the electrophoresis apparatus. 0.5 " TBE buffer was added to the top and bottom chambers of the gel apparatus and the combs were placed in tne gel. The gel was then pre-electrophoresed for about 1 h at 50 W before loading the samples. The samples were boiled for 5 min, loaded on to the gel and electrophoresed until the bromophenol blue dye front had run off the gel. Then another set of samples was loaded into a new set of wells and electrophoresed. Depending on the length of the sequence to be read, in some experiments a third set was loaded and electrophoresed. Afterwards the buffer was drained off from the upper chamber and the glass plates containing the sequencing gel was removed. Using a plastic wedge (Hoefer Scientific, San Fransisco, CA), the glass plates were carefully separated, leaving the gel on the large plate. A mixture of 10% acetic acid and 10% methanol was sprayed on 125 the gel and left to stand for 5 min at room temperature. The gel was then carefully removed onto a piece of 3M Whatman paper and dried for 1 h at 80°C. The gel was then exposed to Kodak X-ray film and the sequence was read out of the autoradiogram. DNasel Footprinting assay For DNasel footprinting reactions, the radiolabelled ompS promoter DNA was prepared by one of two methods. In the first method, ompS promoter DNA was amplified by PCR using the oligonucleotide primers F5 and R5 and directly used in the labelling reaction. The primer F6 has an EcoRI restriction site at the 5' end and R6 has a BamHI site at the 3' end. Therefore depending on which end of the PCR amplicon needed to be radiolabelled, the amplicon was cleaved with the appropriate restriction enzyme and the cleaved end was radiolabelled using the Klenow enzyme and a-32P labelled nucleotides. After radiolabelling, the labelled amplicon was cleaved at the unlabelled end using the appropriate restriction enzyme. This ensured that only one end of the amplicon was labelled. I 126 In the second method, the PCR amplicon was first cloned into the plasmid vector pBluescript. The cloned insert was cleaved off at one end using restriction digest and this end was radiolabelled. Following radiolabelling, the entire cloned piece was cleaved using the restriction site at the opposite end. In both methods, the labelled DNA was purified using GeneClean (BiolOl Inc. La Jolla, CA) and eluted in TE buffer. For the end labelling reaction, approximately 25 ng of cleaved ompS DNA, 50 uCi each of (a-32P)dATP, (a-32P)dCTP, (a-32P)dTTP (ICN Bio Chemicals), 1 ul of 0.5 mM dGTP and 5 /xl of Klenow reaction buffer were added to a sterile microfuge tube and the volume of the reaction mix was adjusted to 20 /xl with sterile distilled water. Then 5 U of the Klenow enzyme(DNA polymerase 1 large fragment, New England Bio Labs) were added to the reaction mix and the tube was incubated 25 min at room temperature, behind a plexiglass shield to avoid exposure to radiation. After the 25 min incubation, 2 /xl of cold dNTP mix (1.25 mM each) were added to the tube and incubated for further 5 min at room temperature. Then the unincorporated nucleotides were 127 removed by using a 30,000 NMWL filter unit (Ultrafree-MC, Millipore Corporation, Bedford, MA) and the radiolabelled DNA was purified by using Gene Clean. After eluting in TE buffer, the DNA was cleaved with the restriction enzyme corresponding to the restriction site located at the end opposite to the one radiolabelled (i.e. for the promoter DNA radiolabelled at the EcoRl end, the DNA was cleaved with BamHl at this point). The DNA was purified using Gene Clean and stored 4°C in a plexiglass container. Before use, the radioactivity of the DNA was measured by using a liquid scintillation counter and 160,000 cpm were used in the DNA: protein binding reaction. In DNAsel Footprinting, the binding of protein to ompS promoter DNA was carried out as explained for mobility shift assays except for the change in the binding buffer to HEPES buffer (12 mM HEPES-NaOH pH 7.9, 60 mM potassium chloride, 1 mM DTT, 1 mM EDTA, 10 ug BSA, 2 /xg of ds salmon sperm DNA and 12% glycerol) . The DNA:protein binding reactions were set up in several tubes containing 160,000 cpm DNA and increasing concentrations of protein. Two control tubes were set up with no protein added. After the binding reaction, the Mg2+ I 128 and Ca2+ concentrations in the reaction mix were adjusted to 5 mM by adding 2 /xl of 50 mM Calcium chloride/magnesium chloride solution. The DNAsel enzyme (Pharmacia) was diluted in the dilution buffer (10 mM Tris.HCl pH 7.5, 10 mM calcium chloride, 10 mM magnesium chloride and 50% glycerol) recommended by the manufacture's. 0.001 U of DNAsel was added to one control tube containing no protein and the DNA was digested for 3 0 sec at 3 0°C. The DNA was not digested with DNAsel in the other control tube and therefore was used to determine the status of the radiolabelled ompS promoter DNA used in the reaction. 0.005 U of DNAsel was added to the reaction tubes containing protein and the reaction was carried out for 30 sec at 30°C. After digesting with the DNAsel, 0.75 /xg of E. coli tRNA (GIBCO/BRL) was added to the reaction tubes, and the DNA was precipitated by adding 250 /xl of ice cold 0.5 M ammonium acetate in 95% ethanol and incubating for 3 0 min at -70°C. The precipitated DNA was then washed twice with 70% ethanol, and vacuum dried. The dried DNA pellet was resuspended in 5 ul of loading buffer (95% formamide, 20 mM EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol). 1 ul of this material was used to determine the radioactivity of the sample. The samples were either 129 immediately electrophoresed on a sequencing gel or stored at -70°C until use. A sequencing ladder of the ompS promoter region was also run alongside the DNAsel reaction samples. The samples were electrophoresed for 90 min at 50 W and afterwards the gel was dried and exposed to Kodak X-ray film as described previously. Attempt to clone the gene coding for the DNA binding protein Two strategies were used to clone the gene coding for the DNA binding protein OmpT; conventional genetic approach c?nd reverse genetic approach. a: Conventional genetic approach: This approach is based on searching for L. pneumophila genes that would activate the ompS promoter in E. coli. In order to do this, first pRIS12 containing the ompS promoter -.lacZ fusion was transformed into the E. coli strain MC4100 which is recA+ and lac . The transformants were 130 selected by plating the bacteria for isolated colonies on LB agar containing kanamycin and ampicillin. A Kan/Amp resistant colony was grown overnight at 37CC with aeration in LB broth containing 10 mM magnesium sulfate, kanamycin and ampicillin. 0.2 ml of this culture was aliquoted into microfuge tubes to which 10 /xl of A-RS45 phage stock (approximately 10s PFU/ml) were added. The tubes were left to stand for 10 min at 37°C. The phage A-RS45 contains the DNA sequences adjoining the ompS promoter:lacZ fusion and therefore the ompS promoter:lacZ fusion will be transferred to A-RS45 by homologous recombination and the recombinant phage could be selected by their kanamycin resistant phenotype. The following contents were added to sterile test tubes; 1 ml of LB broth, 0.1 ml of sterile 1 M magnesium sulfate, 2 ml of molten top agar. The A-RS45 infected E. coli cells were added to this mixture, mixed and immediately plated on pre-warmed LB/kanamycin agar plates. The plates were then incubated overnight in a 37°C incubator with humidity. After the incubation, 3 ml of LB broth were added to the top the agar surface, the top agar was scraped off aseptically and transferred in to a tube. 100 /xl of chloroform were added to the tubes, vortexed vigorously and I I 131 placed on ice for 5 min. This results in the lysis of bacterial cells, thus releasing the phage in to the liquid. The tubes were then centrifuged at 8 000 X g for 5 min and the supernatants containing recombinant phage(X-RIS12) were collected in fresh tubes and stored at 4°C. The A-RIS12 phage containing the ompS promoter:lacZ fusion were then used to infect the E. coli strain MC4101 which is rec mutant of E. coli MC4100. The E. coli cells were infected with A-RIS12 as described earlier and the cells were plated on LB agar containing kanamycin. The colonies that appeared after overnight incubation at 37°C were confirmed to be containing the ompS promoter: lacZ fusion in the chromosome, by both southern blot hybridization and PCR. The next step was to transform a L. pneumophila genomic library into these E. coli colonies. A L. pneumophila genomic library was created using the plasmid vector pBluescript. First, L. pneumophila chromosomal DNA was partially digested with the restriction enzyme Sau3AI. In order to obtain a partial digestion, 28 /xl of 55 mg/ml L. pneumophila Svir chromosomal DNA was mixed with 41 ul of Sau3Al reaction buffer, 4.1 /xl of 10 mg/ml BSA I 132 and 337 ul of sterile distilled water. 100 /xl of the DNA mix were added to a sterile microfuge tube and marked tube #1. 50 /xl were added to 7 more and they were marked as tubes 2 through 7. 2.75 U of Sau3Al restriction enzyme were added to tube #1, mixed well and 50 /xl volume was transferred to tube #2. Likewise, a serial dilution of the Sau3AI enzyme was performed up to 10"7. The tubes were then incubated at 3 7°C for 3 0 min and the enzyme was then inactivated by heating the tubes to 65°C. 2 /xl of partially digested DNA were then separated by agarose (0.4%) gel electrophoresis. The gel was stained with ethidium bromide, viewed under a UV transilluminator to determine the DNA samples showing partial digestion. The DNA from the tubes containing partially digested DNA was pooled together in one tube and used in the ligation reaction. In preparation for ligation, the vector pBluescript was digested with the restriction enzyme BamHl and the cohesive ends of the cleaved DNA was dephosphorylated using calf intestinal alkaline phosphatase (Boehringer Mannheim Biochemicals, Indianapolis, IN). To do this, pBluescript DNA was first digested with BamHl and then 1 U of calf intestinal alkaline phosphatase was added to the restriction digest and incubated for 30 min at 37°C. After 133 the incubation the phosphatase enzyme was inactivated by adding 1/I0th volume of 500 mM EGTA and heating to 65°C for 45 min. The BamHl digested, de-phosphorylated plasmid vector and the Sau3AI partially digested L. pneumophila chromosomal DNA were ligated together using T4 DNA ligase (New England Biolabs). After overnight ligation at 15°C, the ligation mixture was transformed into competent E. coli MC4101 cells containing the ompS promoter: lacZ fusion in the chromosome. The transformants were plated on LB containing X-gal, IPTG, kanamycin and ampicillin. The colonies showing dark blue colour were selected after incubating the plates overnight at 37°C. Plasmid DNA was purified from these colonies and the cloned L. pneumophila DNA was sequenced using the universal primer from the T7 DNA sequencing kit (Pharmacia). b: Purification of the DNA binding protein: The DNA binding protein was partially purified using a DE-52 (diethylaminoethyl cellulose) pre-swollen microgranular anion exchange (Whatman Bio Systems Ltd, Kent, England) column. About 5 0 g of DE-52 were added to a glass beaker containing 500 ml of 100 mM Tris-HCl pH 7.6 and stirred until an even suspension was formed. The column was 134 made in a sterile 60 ml disposable syringe. The syringe was first plugged with glass wool and 10 ml of 50 mM Tris, pH 7.5 was run through the syringe to get rid of any air bubbles trapped in the glass wool. The DE-52 suspension was then added to the syringe little by little allowing the resin to settle to the bottom. Once the required column length was achieved, a few millilitres of 50 mM Tris, pH 7.5 were added to the top of the resin to let it settle well. After the column had set, about 50 ml of 500 mM Tris, pH 9.0 was run through the column to elute any previously bound proteins as well as to charge the resin. Then 5 0 mM Tris pH 7.5 was run through until the column eluate reached a pH of 7.5. Once this was achieved, 100 ml of the running buffer, Buffer X (50 mM Tris-HCl, pH 7.6, 35 mM potassium chloride, 25 mM ammonium chloride, 1 mM DTT, 1 mM PMSF, dissolved in 95% ethanol and 5 mM EDTA) were passed through the column. L. pneumophila high speed supernatant was diluted 1:3 with buffer X and 15 ml were slowly loaded on to the DE-52 column using a 10 ml pipette, taking care not to disturb the column. Samples of the undiluted and diluted extract were stored at -70°C to be used as positive controls when testing the column elutes. After loading the bacterial extract on to II 135 the column, the spectrophotometric detector was switched on, and the protein elution profiles were monitored by looking for the peaks on the graph. Once all the sample had entered the column, a sufficient volume of buffer X was run through until the initial void protein fraction was completely eluted (monitored by the void peak on the graph). The void sample was collected, then a salt gradient (100 ml each of buffer X containing 3 5 mM potassium chloride and buffer X containing 500 mM potassium chloride) was run through the column to elute the bound protein and the column eluate was collected in 35 drop (approximately 3 ml volumes) volumes using a fraction collector. After the salt gradient had run through, approximately 100 ml of buffer X with 50 0 mM potassium chloride followed by 100 ml of buffer X containing 1 M potassium chloride were run through the column and the elutes were collected in 35 drop fractions as before. All protein fractions were stored at 4°C until further testing. The protein fractions containing the DNA binding protein were identified by mobility shift experiments. The fractions showing a shift with the mobility shift assays were pooled together and used for further 136 purification of the protein. The combined protein samples were first dialysed against 10 00 ml of buffer X using 6- 8,000 m.w. cutoff dialysis membrane tubing (Speccrapor; Spectrum Medical Industries, Los Angeles, CA). The dialysis was performed overnight at 4°C in a walk in cold room. The dialysed protein samples were then concentrated using an Amicon PM10 filter (Amicon Corporation, Danvers, MA). The concentrated protein samples were then tested for DNA binding activity using the mobility shift assay. The protein fractions showing a shift with ompS promoter DNA were pooled and used for further purification of the DNA binding protein using an Avidin (monomeric)-Agarose (Sigma chemical company, St. Louis, MO) column. The protein purification using this column was performed as described by Augustin et. al. (1994) . The column was packed in a disposable P1000 tip. The pipette tip was first plugged with nylon wool and a few drops of binding buffer (20 mM Tris.HCl pH 7.0, 50 mM potassium chloride, 1 mM EDTA, 5 mM DTT and 12% glycerol) were run through to get rid of the trapped air bubbles. 100 /xl each of avidin:agarose beads and binding buffer were mixed together in 1:1 ratio and packed into the blocked pipette tip in order to create a column. The column was then 137 washed twice with 300 /xl of binding buffer, once with 300 ul of binding buffer containing 0.5 mg/ml BSA, twice with 3 00 /xl of elution buffer (20 mM Tris.HCl pH 7.0, 1 M potassium chloride, 5 mM magnesium chloride, 1 mM EDTA, 5 mM DTT and 12% glycerol) and once again with 3 00 /xl of binding buffer. At this point, 3 ug of biotinylated ompS promoter DNA (obtained by PCR amplification using the biotinylated F6 oligonucleotide primer and the reverse R6 primer) were incubated (as described in mobility shift assays) with 15 mg of partially purified DNA binding protein fraction and four such reaction mixes (162 /xl each) were loaded onto the column at a flow rate of 10 /xl/min. The column run off was collected and again loaded onto the column to ensure maximum binding of the cmpS DNA: protein complex to the avidin agarose beads in the column. Then the column was washed once with 300 ul of binding buffer and the elute was collected as void sample #1. Then 300 /xl of elution buffer was added to the column and the eluate was collected in microfuge tubes at one drop per tube. This was repeated once again with another 3 00 /xl of elution buffer. The tubes containing the column eluates were stored at 4°C until further use. Half of the volume of these samples were subjected to SDS PAGE, the 138 proteins were transferred to Immobilon membrane and then stained with Coomassie Brriliant Blue. The other half was used to perform a southwestern blot assay in order to both identify the DNA binding protein and to make sure that the protein still retained the DNA binding capabilities. Immunological characterization of OmpS In these studies, the immune response to OmpS in guinea pigs surviving lethal challenge with virulent L. pneumophila, as well as humans surviving Legionnaires' disease was studied and compared with another major protein antigens of L. pneumophila; Hsp60. The results of these studies were published in two scientific publications (viz: Weeratna et. al, 1993; Weeratna et. al, 1994). Copies are included in the Appendix. Guinea Pigs Male or female Hartley strain Guinea pigs in the weight range of 200-300 g obtained from the Charles River Laboratories, La Salle, Quebec, were used in these studies. The animals were housed at the animal care facility in the Sir Charles Tupper Building, Dalhousie University. I 139 Skin Testing Guinea pigs surviving intraperitoneal infection (i.p.) with virulent L. pneumophila serogroup 1 Philadelphia 1 or, in one study, animals immunized with either 25 /xg of OmpS or 25 /xg of Ovalbumin (OA) were used for skin testing. The animals were first shaved over the hind quarters and then injected at different sites with known amounts of different protein antigens. The proteins used in this study were L. pneumophila metalloprotease (10 ug) , 2 8 kDa OmpS monomers (0.1-10 ug) , L. pneumophila 60 kDa heat shock protein (Hsp60, purified from E. coli pSH16 following heat shock induced over expression of htpAB operons) (0.1-10 /xg) and OA (10 ug) each dissolved in 100 /xl of phosphate-buffered saline (PBS). The protein antigens were previously prepared in the laboratory by other researchers (Keen and Hoffman, 1989; Butler and Hoffman, 1990; Hoffman et. al., 1989). OA was purchased from Sigma Chemical Company (St. Louis, MO) . After injecting these antigens, the diameters of induration and erythema were measured at 24 and 48 h time intervals and recorded in mm . 140 Lymphocyte proliferation assays (LPRs) In guinea pigs, peripheral blood was collected by cardiac puncture. The peripheral blood was first diluted 1:2 in sterile PBS and placed gently on Ficoll-Hypaque (Pharmacia, Piscataway, NJ). The ratio of blood : Ficoll- Hypaque was 1:3. The tubes were then centrifuged at 1500 rpm for 40 min at 18°C. The lymphocyte band was carefully removed to a fresh tube, washed twice with sterile PBS and then suspended in RPMI 1640 medium containing penicillin (100 U/ml) , streptomycin (100 /xg/ml) , 2 mM L- glutamine and 5% fetal calf serum (Flow Laboratories, ICN Biochemicals Canada Ltd., Mississauga, ON). Splenic lymphocytes were collected by homogenizing aseptically removed guinea pig spleens. Both peripheral blood and splenic lymphocytes were adjusted to a final concentration of 2.5 X 106 cells per ml and plated into 96-well U-bottomed tissue culture plates (Linbro: Flow Laboratories). The antigens (OA, OmpS and Hsp60) were suspended in RPMI 164 0 medium and used at 0.1, 1.0, and 10 ug/ml. Concanavalin A (5 to 10 /xg/ml) and Pokeweed mitogen (1 to 5 ug) (Sigma Chemical Co.) were also suspended in RPMI 1640 medium and used as positive controls. Each lymphocyte sample was plated in triplicate. The tissue I 141 culture plates were incubated in a humidified C02 incubator at 3 7°C. At specified time intervals, 0.5 /xCi of (3H)thymidine was added to each well and the plates were incubated for 4-6 h under the same conditions. The cells were then harvested and radioactivity incorporated into DNA was counted by liquid scintillation. The proliferative responses were calculated by subtracting the cpm obtained for the medium control from the cpm obtained for the antigen challenged lymphocytes (A cpm). The proliferative responses between different antigens and between different experiments were compared by calculating the stimulation index (SI). The SI was computed by dividing the radioactivity measured in counts per minute (cpm) in the antigen challenged lymphocytes by the cpm obtained for the medium control. Human Studies In human studies, approximately 40 ml of peripheral blood were collected from humans surviving Legionnaires' disease, as well as those who had no history of Legionnaires' disease. Informed consent was obtained from each person prior to drawing blood. The protocol used for isolating human blood lymphocytes and for analyzing the LPR I 142 was identical to that of guinea pig blood lymphocyte isolation. However, for transplant patients receiving cyclosporin A therapy, the lymphocytes were incubated at 37°C in humidified CO., incubator for 24 h in RPMI 1640 medium with no antigens and washed twice in sterile PBS prior to challenge with the antigens. LPRs were determined for all individuals on day 6 and the SI was computed as described in guinea pig studies. ELISA assays Antibody levels in peripheral blood from guinea pigs and humans were determined by enzyme-linked immunosorbent assay (ELISA) as described by Helsel et. al. (1988). The 96 well flat-bottomed microtitre plates were coated at 4°C over night with 20 /xg of antigen (OmpS, Hsp60 or the negative control OA) diluted in coating butter (15 mM sodium carbonate, 35 mM sodium bicarbonate, and 3 mM sodium azide). The coated plates were washed 3 times with PBS containing 0.1% BSA and 0.05% Tween 20 in order to remove unbound antigens. Then 100 /xl serum (diluted in PBS containing 1% BSA and 0.05% Tween 20) were added to each coated well and the plates were incubated for 1 h at 3 7°C in a humidified 143 incubator, The wells were then washed 3 times with PBS/0.01% BSA/0.05% Tween 20 and 100 /xl of alkaline phosphatase conjugated goat anti-guinea pig IgG (Sigma chemical company), diluted 1:1000 in PBS/0.1% BSA/0.05% Tween 20 were added to each well. The plates were incubated for 1 h at 37°C in a humidified incubator. The wells were washed 3 times as before and the alkaline phosphatase activity was measured by using Sigma 104 phosphatase substrate (Sigma Chemical Co.) following the manufacturer's instructions. The enzymati- reaction was terminated by adding 2 N sodium hydroxide and the absorbance of the reaction mixture was measured using an ELISA reader at 410 nm. All human sera were screened for L. pneumophila serogroup-specific antibody at the Victoria General Hospital, Halifax using the protocols of Wilkinson et. al. (1981). Guinea pig vaccination studies The vaccination trials were performed in collaboration with Dr. Paul H. Edelstein at University of Pennsylvania Medical Center in Philadelphia, Pennsylvania. In these 144 trials 21 animals weighing 200-300 g were randomly assigned to 3 groups with 7 animals in each group. They were vaccinated with 25 ug of 2 8 kDa OmpS, 50 ug of Hsp60 or 50 ug of OA. The protein suspensions for immunization were made by diluting the stock protein solutions in PBS. F^r primary immunization, the suspensions were combined with an equal volume of Freund's complete adjuvant (Sigma Chemical Co.) and injected into the shaved flanks of the animals. Three weeks later, the animals were reimmunized with the same amount of protein combined with Freund's incomplete adjuvant (Sigma Chemical Co.) using the same protocol. Six weeks after the primary immunization, animals were challenged with 2.5 times the 50% lethal dose (LD50) of L. pneumophila serogroup 1 by intratracheal route as described by Edelstein et. al (1984). The size of the actual inoculum was determined by plating the bacteria on BCYE agar. For 10 days post infection, the animals were closely monitored for trachypnea, feeding and level of activity . The weight and rectal temperature of the animals were recorded daily. Animals surviving after 10 days post infection were sacrificed. Necropsy was performed on all animals within 24 f • 145 h after death. Lung tissue was harvested for quantitative culture of bacteria and histological analysis. Blood was collected for serological testing. I RESULTS OmpS, the disulfide-bond cross-linked major outer membrane protein of L. pneumophila, is one of the most abundant proteins produced by this organism. Previous work has shown that this protein facilitates the internalization of the bacteria into monocytes and alveolar macrophages and hence is considered as a potential virulence factor (Payne and Horwitz, 1987). Structurally, this protein is unique among outer membrane proteins of Gram negative bacteria because it contains interchain disulfide bonds and is covalently anchored to the peptidoglycan (Butler and Hoffman, 1990) . The objective of this study was to further characterize OmpS from both a genetic and an immunological view point. The genetic characterization has focused on studying the mode of regulation of ompS in L. pneumophila. The immunological characterization has looked into the importance of OmpS in the development of a protective immune response against L. pneumophila infections both in humans 146 I 147 and in the animal model for Legionnaires' disease, the guinea pig. Cloning and sequence analysis of ampS: As a prelude to the investigation into regulatory mechanisms involved in the expression of ompS, the gene coding for this protein was cloned and sequenced by Hoffman et. al. (1992 b). This gene was cloned using reverse genetics. The following is a brief description of the strategies used in the cloning of ompS. In previous work, Hoffman et. al. (1992 a) had generated degenerate oligonucleotide primers based on the N terminal amino acid sequence of OmpS, GTMGPVWT. These primers were end labelled using T4 polynucleotide kinase and used to probe L. pneumophila Svir chromosomal DNA digested with various restriction enzymes (Hoffman et. al., 1992 a). Within this pool of probes, the oligonucleotide primer 24b (GGTACTATGGGTCCAGTATGGAC) was found to hybridize strongly to L. pneumophila genomic DNA at the highest stringency. Furthermore, this oligonucleotide hybridized to a single i 148 band for each restriction digest of genomic DNA and recognized a 1 kb mRNA transcript by northern blotting. Therefore, in the cloning of ompS, L. pneumophila genomic DNA was digested with the restriction enzymes EcoRl, HindiII and PstI, separated by agarose gel electrophoresis and the DNA fragments corresponding to the molecular range that hybridized with the oligonucleotide probe 24b were excised and ligated into the cloning vector pBluescript (Stratagene, La Jolla, CA). No clones were obtained with ScoRI generated DNA fragments. The resulting clones were screened by colony Southern blotting using end labelled 24b oligonucleotide primer. The DNA from the positive clones 2.9 kb H151 and 0.9 kb HP246, (Figure 1) was subcloned into M13 vectors and the DNA was sequenced using the dideoxy chain-termination method (Sanger et. al., 1977). Both DNA strands were sequenced and the sequences were assembled and analyzed using the Wisconsin Genetics Computer Group sequence analysis programs (Genetics Computer Group Inc., University of Wisconsin Biotechnology Center, Madison, WI). I • Figure 1: Restriction endonuclease maps of clones H151 and HP246 in pBluescript. A 2.9 kb HindiII fragment of L. pneumophila DNA (indicated by black boxes) was cloned into pBluescript. The open reading frame of ompS was localized to the 1.5 kb EcoRI region. Abbreviations: E, EcoRI; H, HindiII; P, Pstl. The map distances are given in base pairs. n I 150 EH E P E H H151 700 200 1300 800 EH E P HP246 i i _i _ 700 200 Figure 1 151 Sequence analysis and Primer extension studies: The complete nucleotide sequence of ompS was published by Hoffman et al in 1992(b). According to sequencing data, ompS has an open reading frame (ORF) of 891 bp which codes for a polypeptide of 297 amino acids. Out of these 297 amino acids, the first 21 amino acids code for a signal sequence involved in protein export. The first amino acid of the processed protein was determined by N-terminal sequencing (Hoffman et. al., 1992a). The primer extension studies performed in the current study revealed that the transcription start site lies 97 bp upstream of the translation start site (Figure 2). A palindromic sequence resembling a rho-independent transcription terminator was located down stream of the termination codon TAA. The -10 (TAATAAAAT) and -35 (TCAATGAG) promoter sequence of ompS did not resemble an E. coli consensus promoter sequence (Figure 3). This may explain the several failed attempts to express this gene in E. coli. It is possible that the promoter sequence of ompS, being different from the consensus E. coli sequence, is not recognized by the E. coli RNA polymerases. 152 Figure 2: Primer extension ai-alysis of the transcription start of ompS. Total RNA was extracted from L. pneumophila Philadelphia 1 serogroup 1 (Svir) cells grown in buffered yeast extract medium. The oligonucleotide H25 which had been end labelled with [32P]ATP was annealed to total RNA (25 /xg) . The primer extension was carried out as described in the methods section The resulting cDNA was suspended in sample buffer and resolved on a polyacrylamide gel beside a sequencing ladder generated f.rom the same oligonucleotide primer with a Hindlll-PstI DNA fragment in M33 as the template. The transcription start is located in a CCC sequence flanked by AT-rich sequences. I $ W) ft, ? I 154 FIGURE 3: Sequence of ompS promoter region. The underlined sequences indicate the 2 oligonucleotides (F6 and R6) used to PCR amplify the ompS promoter region. Smaller bases in superscript indicate the base changes made when making the oligonucleotides in order to create restriction sites for subsequent sub-cloning of the promoter region. The transcription start site is indicated by the arrow. The outlined ATG codon depicts the translation start site. Bold letters indicate the restriction sites EcoRI, BelI and BamHl, in that order. OmpS PROMOTER REGION I C ATCTTGAATC I CTCCACTGTACAGGCTTTTTTC TACCCTTACATACCCCAAAAAAACAGTCTTACGAT AAATTTTGGCGCAAATTAATAGTCATGCACTCCCC CTCATTACTACAAAATTAAAGGAAAAAGCAGACTA CTTGCTGGCGTATAAATCAATAACAGCAGATACTG -60 -35 ATCAAATCTTACATTTAATGTTTAAATC.-. ::> -vi -J -10 — ->mRNA TAAATAACTTTAATT.:. "-. .. TTACCCTTATTAT TTGATGAAGAATCAAATAGATTCTGTTAGA"ATT~C GCTAATCCTAGATGTTTAAAGGAATAACAATA ATAA ATCAGTGGAGAACGGGAT. Figure 3. I 156 Sequence relationship of ompS in legionellae: Previous studies by Butler et. al. (1985, 1990) have demonstrated the presence of disulfide-cross-linked outer membrane proteins sharing common epitopes amongst several species of Legionella. In order to investigate the genetic relatedness between the genes coding these proteins, chromosomal DNA from several selected Legionella species and serogroups were probed with a PCR radiolabelled internal DNA fragment of ompS from L. pneumophila Philadelphia (serogroup 1) . According to the results of these Southern blot assays, at moderate stringency (15% mismatch in the duplex), all Legionella species examined exhibited related sequences (Figure 4a). L. jordanis serogroup 1 showed the weakest signal. According to previous reports (Butler et. al., 1990) , L. jordanis does not express a 28 kDa major outer membrane protein. The strongest signals were observed among the serogroups of L. pneumophila. However, some restriction polymorphism was noted among the different serogroups of L. pneumophila. In addition, weak hybridization signals were seen within serogroups. These may suggest the presence of related genes encoding other outer membrane proteins or the presence of possible cryptic genes (Hoffman et al, 1992) . 157 When the Southern blots were washed at a higher stringency (5% mismatch in the duplex), the hybridization of ompS in other species of Legionella disappeared (Figure 4b). I 158 Figure 4: Southern blot analysis of .EcoRI (A) and Hindlll (B) restricted DNA from selected Legionella species and serogroups of L. pneumophila. Genomic DNA was prepared from L. pneumophila unknown serogroup (lane 1), L. pneumophila serogroup 4 (lane 2), L. jordanis serogroup 1 (lane 3), L. pneumophila serogroup 1 Svir (lane 4), L. micdadei HBBA (lane 5), L. pneumophila serogroup 7 (lane 6), L. pneumophila serogroup 3 (lane 7), L. oakridgensis ORIO (lane 8), L. micdadei Tatlock (lane 9), L. pneumophila serogroup 5 (lane 10), L. pneumophila serogroup 2 Togus-1 (lane 11), and L. pneumophila serogroup 1 Avir (lane 12). In panel A, the hybridization was done at a decreased stringency (15% mismatch in the DNA duplex) to detect relatedness among different species. In panel B, hybridization was done at moderate stringency (5% mismatch in the DNA duplex). The probe used in this assay was generated by PCR using the oligonucleotide primers 24b and R3 (a reverse primer hybridizing to an internal region of ompS) . H 1 2 3 4 5 6 v^^Hr ^KKF #Mfi# q^^F j^H^H 7 8 9 10 11 12 •» IS Figure 4 A I I 160 12 3 4 5 6 ^Off/ft 7 8 9 10 11 12 *• •% 4c # 3ta# Figure 4 B 161 Selective radiolabelling of OmpS: During the cloning of ompS, it was evident that ompS was not expressed from its own promoter in E. coli. Attempts to express ompS under an E. coli promoter were also unsuccessful. This suggested that the expression of OmpS may be detrimental to the viability of E. coli cells. To investigate this possibility, a T7 RNA polymerase/promoter system was used to express OmpS in E. coli. A promoterless 1.5 kb EcoRI fragment and a truncated 1.3 kb Pstl/EcoRI fragment from pH151 (Figure 1) were cloned into the T7 promoter vector pT7-5 (Figure 5). These constructs were named pTOMPSl and pT0MPS2 respectively. Both these constructs could be maintained in E. coli strain JF626. This strain does not contain T7 RNA polymerase and hence the T7 promoter in pT7-5 remains inactive. However, several attempts to transform these constructs into E. coli BL21 containing a chromosomal copy of T7 RNA polymerase gene proved unsuccessful. Therefore, a different strategy involving the introduction of a M13 phage (mGPl-2) was attempted. Bacteriophage mGPl-2 contains the T7 RNA polymerase gene cloned under the lac promoter and thus the expression of this gene could be induced by the addicion of P I I 162 IPTG. In addition, the transcription by bacterial polymerases could be inhibited by the addition of rifampicin. Under these conditions, the protein synthesis by the cloned genes could be monitored by using [35S] - methionine. The expression of ompS by pTOMPSl and pT0MPS2 were monitored using this strategy. The proteins were labelled for 5 min, cells were lysed and the proteins were analyzed by SDS-PAGE and autoradiography. A sample of purified OmpS was also electrophoresed on the same gel. The results of this experiment are shown in Figure 6. According to chis figure, pTOMPSl shows faint expression of OmpS (lane E+) . The band corresponding to OmpS was determined by superimposing the autoradiogram with the original protein gel. pTOMPSl also shows the expression of 2 higher molecular weight proteins. These may be hybrid proteins containing OmpS or unprocessed OmpS. No expression of OmpS was seen in this clone in the absence of mGPl-2 infection (lane E-). Plasmid pT0MPS2 expressed a number of low molecular weight proteins (lane E/P+). These could result from the proteolytic degradation of OmpS. Again, this clone failed to express the low molecular weight protein without mGPl-2 I 163 (lane E/P-). The vector control did not show significant protein expression. H I 164 Figure 5: Schematic map of pT7-5, an expression vector using T7 RNA polymerase. pT7-5 is a derivative of pBR322. It contains the p- lactamase gene (jbla) that encodes resistance to ampicillin. This gene is oriented opposite to the transcription by T7 RNA polymerase promoter cj)10. Therefore, promoter c|>10 influences the expression of genes cloned at the multiple cloning site exclusively. 165 bla Hindi ,Clat Hind 111 .Pstl Sail pT7-5 Xbal BamKI 2404 bp Smal Sac I &»R| origin 010 Pvull Figure 5 166 Figure 6: Expression of OmpS using the T7 RNA polymerase/'promoter system. E. coli JF62S containing either pTOMPSl or pT0MPS2 were infected with mGPl-2. OmpS expression by these clones were monitored as described in the text. Lanes E+ and E/P+; protein expression by pTOMPSl and pT0MPS2 infected with mGPl-2 respectively. Lanes E- and E/P- ; protein expression by pTOMPSl and pT0MPS2 without mGPl-2. Lane V; protein expression by E. coli JF626 containing pT7-5 infected with mGPl-2. The signal corresponding to OmpS is indicated by the arrow. 167 + I a a + i \ \ > -> Figure 6 1 * 168 Allelic exchange mutagenesis of ompS: An attempt was made to mutagenise the chromosomal copy of the ompS gene in L. pneumophila by using allelic exchange mutagenesis. To accomplish this, the ompS gene was corrupted in vitro by cloning a kanamycin cassette into the ompS open reading frame (Figure 7). The corrupted copy of ompS (pTLP6::ompS::kan) was electroporated into L. pneumophila Svir and Avir strains with the hope of corrupting the chromosomal copy of the ompS gene by homologous recombination. The recombinational events were selected by screening for kanamycin and streptomycin resistance and chloramphenicol sensitivity in Svir strains. In Avir strains the selection was for kanamycin resistance and chloramphenicol sensitivity. The electroporation of the vector control (pTLP6) and pTLP6::ompS::kan into L. pneumophila Svir and Avir were successful. However, several attempts to obtain ompS mutants of L. pneumophila were unsuccessful. This may be due to the fact that OmpS plays a vital role in the structural integrity of the bacterial cell. Therefore null mutations in this gene may be lethal to L. pneumophi1 a. 169 Figure 7: Allelic exchange mutagenesis of ompS. The ompS gene (filled rectangle) cloned into pBluescript was disrupted by placing a kanamycin cassette (unfilled rectangle) in its middle. The disrupted ompS gene was then cloned into the cloning vector pTLP6 and electroporated into L. pneumophila. Recombinational events between the chromosomal ompS gene and the mutagenised ompS gene were selected by screening for kanamycin and streptomycin resistance and chloramphenicol sensitivity. Abbreviations: E; EcoRI, P; PstI, H; Hindlll Steps involvir 4 the allelic exchange mutagenesis oiompS. E H E P E H I H151 700 200 1300 800 A 3 kb Hwdlll fragment was cloned into pBluscript vector with the EcoRI site deleted from the multiple cloning site. H H 700 800 The 1.5 kb EcoRI fragment was removed by restriction digestion and re-ligation of the vector , Kanamycin resistance element from mini-Tn5 Km 1 was cloned V into EcoRI site. TT TJ ompS gene disrupted by - - a kanamycin cassette. The disrupted ompS gene was removed by digestion with BamHl and KpriL and ligated to pTLP6. FIGURE 7 171 Regulation of ompS in L. pneumophila : L. pneumophila is a bacterium that is found naturally in fresh water environments. In its natural habitat, L. pneumophila can be found either free living or in association with aquatic amoebae. When inhaled by a susceptible individual, it has the ability to grow and multiply within the intracellular environment of alveolar macrophages and bring about disease. One might predict that L. pneumophila is able to sense environmental changes associated with infection and initiate a virulence response. Previous studies have shown that ompS is down regulated in virulent cells when intracellular or suspended in the tissue culture medium DMEM which contains high levels of sodium chloride relative to fresh water environments. This response is not seen with avirulent strains (Fernandez, 1992). The only phenotypic difference which correlates with avirulence, is the mutant's ability to grow on media containing physiological levels of sodium chloride. The virulent strains cannot grow on media containing sodium chloride in excess of 0.65% (Catrenich and Johnson, 198-). Therefore in order to see whether transcriptional regulation of ompS is affected by sodium chloride levels, a northern 172 blot analysis was performed. As seen in figure 8, total RNA was extracted from strains grown in BYE broth overnight and challenged for one hour with 0.85% sodium chloride (Lane B) and unchallenged (Lane A). RNA was probed with a PCR radiolabelled ompS DNA amplicon generated by using the oligonucleotide primers 24b and R5 (AGCTTAGCTAGGATCCCTAA- CTTAA) ; both primer sequences are located within the ompS ORF. As seen in this northern blot, when L. pneumophila Svir cells were challenged with 0.85% sodium chloride there was a cessation of ompS expression which was not seen in the unchallenged cells. I 173 FIGURE 8: Northern blot of L. pneumophila Svir RNA. L. pneumophila SVir cells were grown in BYE broth with aeration for 18 h at 37°C. After 18 h growth, 0.35% NaCl was added to some cultures and all cultures were incubated for a further 1 h. The cells were then pelleted and total RNA was extracted. The RNA was probed with PCR radiolabelled ompS DNA. Lane A corresponds to RNA extracted from cultures that were not challenged with NaCl. Lane B corresponds to RNA extracted from NaCl challenged cultures. Lane C corresponds to molecular weight markers. The picture of the ethidium bromide stained RNA gel is used to show the equal loading of RNA in all lanes. Lane C in the ethidium bromide stained gel contains RNA molecular weight markers. I m I I 174 ABC B Figure 8 I * i I— 175 Several genes of L. pneumophila have been cloned and sequenced, for example htpAB (Hoffman et al, 1990), mip (Engleberg et. al., 1989), pro (Black et. al., 1990), recA (Zhao and Dreyfus, 1990), ppl (Ludwig et. al., 1991), lly (Wintermeyer et. al., 1994), fur (Hickey and Cianciotto, 1994) all contain an E. coli like promoter sequence and are expressed in E. coli by their respective promoters. However, cloning and sequence analysis of cmpS followed by the primer extension studies revealed that ompS has a unique promoter sequence and that it is not expressed in E. coli from its own promoter. Furthermore, northern blot analysis suggested that ompS expression may be regulated. These results are consistent with a regulatory model that includes a transcription factor or a unique sigma factor. To determine whether a transcription factor was involved in regulation of ompS, gel mobility shift and southwestern blot assays were performed with sodium chloride challenged and unchallenged cells. Gel mobility shift assays: In gel mobility shift assays, labelled DNA is incubated with unlabelled protein extracts and separated by gel 176 electrophoresis. The protein bound DNA migrates more slowly compared to protein free fragments. Therefore by comparing the electrophoretic mobility of protein bound and free DNA, it is possible to identify the presence of proteins binding to a specific DNA sequence. Unless otherwise mentioned, the 285 bp nucleotide sequence spanning the ompS promoter region (Figure 3) was used as the DNA amplicon in this study. Oligonucleotides FE and Rs flanking this region were used to amplify and radiolabel the 285 bp double stranded DNA amplicon (Figure 9). This DNA amplicon was incubated with L. pneumophila cell extracts and separated by non-denaturing PAGE. As seen in Figure 10, cellular extracts from both virulent and avirulent strains of L. pneumophila contained protein(s) that retarded the electrophoretic mobility of the 285 bp amplicon. The retardation by the avirulent extracts seemed to be somewhat higher than that of the virulent cell extracts. The DNA binding protein binding upstream of ompS promoter region is referred to as OmpT. Extracts from virulent cells challenged with 0.85% sodium chloride for 1 h showed a decrease in the shift compared to unchallenged cell extracts. The smearing effect seen in lane V+, in which the I P 1 177 ompS promoter DNA was incubated with sodium chloride challenged virulent cell extracts may suggest two possibilities. One is that sodium chloride reduces the affinity for OmpT to ompS promoter DNA and thus the protein falls off the DNA as the DNA:procein complex migrates through the gel. The other is tha'; sodium chloride affects the stability of OmpT and that in the presence of sodium chloride this protein is rapidly degraded. Sodium chloride challenge appeared to have no effect on the mobility shift of avirulent cell extracts. I 178 Figure 9: 285 b.-" ompS promoter amplicon generated by polymerase chain reaction (PCR): Lane A depicts the BstEI digested A DNA markers. Lane B depicts the 285 bp ompS promoter region that was amplified by PCR using the oligonucleotide primers F6 and R6. The ompS clone, H151 in pBluescript (figure 1) was used as the template DNA. I 179 A B Figure 9 i u I 180 FIGURE 10: Gel retardation of ompS promoter DNA using cell extract.? from L. pneumophila Svir (V) and Avir (M) , unchallenged or challenged with NaCl. L. pneumophila Svir and Avir cells were grown in BYE broth with aeratior for 18 h at 37° C. Following 18 h growth, some cultures were challenged with 0.85% NaCl. Cell extracts were prepared as explained in the methods section. Lane P depictn radiolabelled ompS promoter amplicon. In Lane V., the ompS promoter amplicon was incubated with unchallenged virulent cell extracts, In lane V+, the DNA amplicon was incubated with cell extracts from virulent cells that were challenged with 0.85% NaCl. Similarly, in Lane M., the DNA amplicon was incubated with avirulent mutant cell extracts that were unchallenged and in lane M+, the DNA amplicon was incubated with NaCl challenged mutant cell extracts. I i 181 P V V M M Figure 10 I I I I i 182 Several approaches were used to demonstrate the specificity of binding of this protein(s) to ompS DNA. One was the use of DS salmon sperm DNA and poly dldc m the mobility shift assays as competitor DNA. Another was to radiolabel a segment of internal coding region of ompS and use that in the mobility shift assays. Yet another approach was to use increasing amounts of unlabelled ompS promoter DNA to compete with the binding of 0?.pT to labelled DNA. The results of such a competition assay are shown in Figure 11. As seen in this figure, the radiolabelled ompS promoter DNA was competed cut by using 3.5 jig of unlabelled competitor DNA (lane d). Furthermore, the radiolabelled internal coding region of ompS remained unshifted upon incubation with L. pneumophila cell extracts. These results established the specificity of binding of OmpT to ompS promoter DNA. I ' 183 Figure 11: Binding competition assay.. The radiolabelled ompS promoter DNA and DNA from an internal ceding region of the ompS gene were incubated with virulent unchallenged cell extracts. Increasing concentrations of unlabelled ompS promoter DNA was added to these incubation mixtures. Lane P depicts the amplicons for the promoter region (lower band) and for the internal coding region of the ompS gene, (upper band) . In lane a, these DNA amplicons were incubated with virulent cell extracts. In lane b, c, and d, 0.1, 1, and 3.5 ug/ml of unlabelled ompS promoter DNA was added to the mixtures. I d 184 ^w J^;* fnp fP» Hi Figure 11 185 Effect of other salts on the mobility shift of ompS DNA: To distinguish between a generalized osmolarity effect and specificity for specific salts in the regulation of ompS expression, the effects of other metal salts were examined. Challenge with 0.85% of calcium chloride or magnesium chloride similarly affected the mobility shift in virulent cell extracts (data not shown). However, challenge with these salts appeared to have a deleterious effect on the viability of the cells as judged by colony morphology and microscopic observations. Therefore, it is not clear if the decrease in band shift is really due to an effect on the DNA binding protein or due to death and lysis of cells resulting in degradation of the protein. Calcium chloride and magnesium chloride appeared to have similar toxic effects on avirulent strains. Challenge with D.85% potassium chloride (Figure 12; lane VKC1) had no effect on the binding of OmpT to ompS promoter DNA. This may suggest that the effect of sodium chloride on the binding of OmpT to ompS promoter DNA is not merely an osmotic effect but rather a sodium chloride specific (- ;ect. Neither sodium chloride nor potassium chloride had any effect on the mobility shift noted for the avirulent strain (data not shown). • i i I 186 Figure 12: Gel retardation of ompS promoter DNA using cell extracts from L. pneumophila Svir (V) strains, unchallenged (-) or challenged with either NaCl or KC1. L. pneumophila cells extracts were prepared as explained in Figure 10. Some cultures were challenged with either 0.85% NaCl or KC1 for 1 h prior to preparing the cell extracts. Lane P depicts the radiolabelled ompS promoter amplicon. In lane V., the promoter DNA amplicon was incubated with unchallenged virulent cell extracts. In lane VNaC1, the DNA amplicon was incubated with sodium chloride challenged virulent cell extracts. In lane VKC1, the DNA amplicon was incubated with KCl challenged virulent cell extracts. 187 P V V V NaCl KCI Figure 12 I 188 Identification of DNA binding site of OmpT: The region of ompS promoter DNA used in the mobility shift assays possesses a Bell restriction site at the -60 bp position upstream of the transcription start site. Bell restriction was used to cleave the DNA amplicon into two fragments in the hope of determining by mobility shift, which fragment contained the OmpT binding site. As seen in Figure 13, cleaving the 285 bp ompS promoter DNA at the -60 bp position abolished most of the mobility shift obtained with both virulent and avirulent cell extracts suggesting that this region is important for the binding of OmpT to ompS DNA.(Note: The intensities of the retarded bands have been greatly accentuated by subsequent rephotographing of the original photographs). In order to further define the binding domain of OmpT, two oligonucleotides F7 and F8 were designed. These oligonucleotides (20 bp in length) contain sequences complementary to the regions immediately up stream (F7) and down stream (F8) from the -60 bp site (Figure 14) . Therefore, PCR amplification of ompS DNA using F7 or F8 as the forward primer and R6 as the reverse primer resulted in the truncation of the 5' end of the 285 bp promoter region. These truncated DNA amplicons were not retarded by extracts I I . 189 from virulent L. pneumophila suggesting that the 5' end of the promoter amplicon is important for the binding OmpT to oiripS promoter DNA (Figure 15) . 190 Figure 13: Gel retardation to identify the region of binding of OmpT to ompfS DNA. The 285 bp ompS promoter amplicon was cleaved at the Bell site located at the -60 bp position of the amplicon into two fragments of 114 and 172 bp. The cleaved DNA amplicons were used in the mobility shift assays. Pcut depicts the cleaved DNA fragments. In V lanes, the DNA amplicon was incubated with virulent cell extracts. In M lanes the DNA amplicon was incubated with avirulent mutant cell extracts. In - lanes, unchallenged cell extracts were used and in + lanes sodium chloride cell extracts were used. 191 P V V M M cut Figure 13 I 192 Figure 14: Schematic map to illustrate the location of the oligonucleotide primers F6, F7, F8 and R6. This figure shows the relative positions of the oligonucleotide primers F6, F7, F8 and R6 used to amplify different regions of the ompS promoter region. Abbreviations: Eco; EcoRI, Bam; BamHl, Bel: Bell 193 Eco Bel Bam F6 F7 F8 R6 171 114 Figure 14 ,1 194 Figure 15: Using truncated ompS promoter DNA amplicons in mobility shift assays to identify the DNA binding site of OmpT. The 5' end of the ompS promoter amplicon was deleted using two different oligonucleotides F7 and F8. The PCR amplicon generated using F7 and R6 was deleted for 151 bases from the N terminus while the amplicon formed using F8 and R6 was deleted for 177 bases. The PCR amplicon generated using F6 and R6 yielded the full size ompS promoter amplicon that was used in other mobility shift and southwestern blot assays. For the mobility shift assays, these different promoter amplicons were incubated with virulent L. pneumophila cell extracts and separated by non-denaturing PAGE. Panel F6: mobility shift using the intact ompS promoter amplicon. Panels F7 and F8: truncated promoter amplicons generated using the primers F7 and F8 were used respectively. Lane P depict the ompS promoter DNA amplicon. In lane V, ompS promoter DNA amplicon was incubated with virulent L. pneumophila cell extracts. 195 Figure 15 I 196 Effect of sodium chloride on OmpT: The mobility shift assays carried out up to this point show that the binding of OmpT is affected by sodium chloride. However, they do not provide sufficient data to assess whether this is due to an action on the affinity of OmpT for the ompS promoter DNA or on the actual synthesis of OmpT. Therefore the kinetics of sodium chloride challenge on OmpT binding were studied and compared to the kinetics of chloramphenicol inhibition of OmpT synthesis. Figure 16 shows the results of these studies. There is a sharp decrease in the mobility shift with increasing time post- sodium chloride challenge (Figure 16b). The shift is completely abolished by 10 min post-challenge with 0.85% sodium chloride. Figure 16a shows the results of the chloramphenicol challenge. In these experiments, the virulent L. pneumophila cells were challenged with 10 0 ug of chloramphenicol. The mobility shift was completely abolished by 10 min post-challenge with 100 ug of chloramphenicol. Although not definitive, these data suggest that OmpT has a very short half life in L. pneumophila and that synthesis rather than the affinity of OmpT is regulated by sodium chloride levels. 197 FIGURE 16: Kinetics of the NaCl response on OmpT. L. pneumophila Svir cultures were grown for 18 h in BYE broth with aeration. Some cultures were challenged with either 100 ug/ml Chloramphenicol (A) or C 85% NaCl (B). The cells were harvested at different time intervals and cell extracts were prepared as explained previously. The extracts were incubated with radiolabelled ompS promoter DNA and separated by PAGE. In Panels A and B, lane P depicts the radiolabelled ompS promoter amplicon. In lane V, the DNA amplicon was incubated with unchallenged V cell extracts. In lanes To, T2, T5, T10 of panel A and lanes To, T10, T30 of panel B, cell extracts prepared from cultures challenged for corresponding time intervals with chloramphenicol or NaCl respectively. 198 P V T T 0 2 T5 T10 Figure 16A 199 P V TO T10 T30 B Figure 16B I • 200 Southwestern blot assays: Southwestern blot assays were used to determine the size of OmpT as well as to identify other proteins binding to ompS DNA. In these assays, L. pneumophila cellular proteins were separated by SDS PAGE and then transferred onto a nitrocellulose membrane. The protein transfer conditions were chosen to promote the renaturation since the native conformation is required for protein:DNA binding. The nitrocellulose membrane containing the transferred proteins was then incubated with radiolabelled ompS promoter DNA. The membi ine was washed with buffers containing increasing concentrations of sodium chloride. The stringency of binding of DNA to proteins was increased by increasing the salt concentration of the wash buffer. Figure 17 shows the results of a preliminary southwestern assay carried out using increasing concentrations of virulent L. pneumophila high speed cell supernatants. DNA binding was seen with a protein concentration as low as 5.4 ug/ml (2 ul). For this blot, the final wash contained 0.25 M sodium chloride. The results of the southwestern assay carried out using virulent and I Lei 201 avirulent L. pneumophila cell extracts prepared from cultures, unchallenged and challenged with 0.85% sodium chloride, confirmed the results of the mobility shift assays (Figure 18A). According to these results, there was a decrease in binding of DNA to the protein from sodium chloride challenged virulent cell extracts compared to that of unchallenged cell extracts (lanes V+ and V-). However there was no difference between challenged and unchallenged cell extracts from avirulent cells (lanes M+, M-). To determine the molecular weight of the ompS DNA binding protein, a duplicate set of protein extracts and protein molecular weight standards was subjected to SDS PAGE. Protein from one half of the gel containing the molecular weight standards and the protein extracts was transferred to Immobilon membrane (Millipore corporation) and stained with Coomassie Brilliant Blue as described in the methods section. The protein in the other identical half of the gel was transferred to the nitrocellulose membrane and used for southwestern blot assay. By superimposing the autoradiogram from the southwestern blot assay and the stained Immobilon 202 membrane, the molecular weight of the DNA binding protein was estimated to be approximately 15 kDa (Figure 18 B). 203 Figure 17: Southwestern blot assays using protein extracts from virulent L. pneumophila and ompS promoter DNA. Different concentrations of virulent _. pneumophila cell extracts were subjected to SDS-PAGE and the proteins were transferred to nitrocellulose membrane. The membranes were probed with radiolabeled ompS promoter DNA generated by PCR and the probed membranes were washed with binding buffer containing increasing concentrations of NaCl. The final wash contained 0.25 M NaCl. As seen in this figure, ompS DNA binding was achieved with a minimum of 2 ul of cell extract corresponding to 5.4 ug of protein. n 204 V- 05 5 fl\ \** ^^^Mk, ^^^HB1- ^^^Hk •* ^p^^ ^BHr Figure 17 205 Figure 18: Southwestern blot assay using virulent (V) and avirulent (M) L. pneumophila cell extracts unchallenged or challenged with 0.85% NaCl. In Figure 18A L. pneumophila cell extracts were separated by SDS-PAGE and the protein was transferred to nitrocellulose membrane. The membrane was then probed with ompS promoter DNA generated by PCR. The probed membrane was washed with binding buffer containing increasing concentrations of NaCl. The final wash contained 0.2-0.25 M NaCl. The ability of V cell extracts to bind ompS promoter DNA was reduced when cells are challenged with 0.85% NaCl (V+) as compared to untreated cell extracts (V-). DNA binding in avirulent cell extracts was unaffected by NaCl (M-, M+). In Figure 18B, two identical sets of protein samples containing molecular weight markers(MW), V, L. pneumophila cell extracts (1) and V, L. pneumophila cell extracts partially purified using a DE52 column (2) were subjected to SDS-PAGE. One half of the gel was transferred onto an Immobilon membrane and stained with Coomassie Blue. The other half was used for southwestern blot assay. The molecular weight of OmpT was determined by superimposing the autoradiogram on the stained Immobilon membrane. 206 • Figure 18A 207 A B MW 12 12 <**** , • "flS ' " **e**1*1^* . ^MS .A. i,«$ • •! «# H v 1 '|P- Iff Pf Figure 18B 208 DNAsel protection assay: The DNAsel footprinting technique, developed by Galas and Schmitz (1978), allows the determination of the specific binding sites of proteins to DNA. As the name implies, DNAsel footprinting uses DNAsel enzyme as the cleavage agent of DNA. This assay exploits the property of DNASEI that, on limited digestion of DNA it does not completely cleave the DNA duplex, but randomly nicks one strand. Therefore, when a population of DNA that has been labelled on only one strand has been subjected to limited digestion with DNAsel, it results in a population of randomly nicked DNA fragments that have been nicked only one or a few times on the duplex. When such DNA is separated by polyacrylamide gel electrophoresis a characteristic DNA ladder is obtained. However, when the DNA is complexed with protein(s) prior to its digestion with DNAsel, the DNA sites bound by protein are protected from cleavage. Such protein bound sites can be identified as gaps on the characteristic DNA ladder. These gaps are referred to as DNA footprints. The exact sequence of the DNA footprint (the site of protein(s) binding) can be determined by running a DNA sequencing ladder adjacent to the DNASEI digested DNA fragments. I 209 In this study, seveial attempts were made to obtain DNA footprints after mixing OmpT with ompS DNA and thus determine the exact binding sites of OmpT on ompS DNA. On several attempts, footprinting assays revealed four protected areas, at -60, -160, -170 and -191 bp positions on ompS DNA (Figure 19, lane 6). However, these results were not reproducible at all times. It should be mentioned that under the same DNA binding conditions, and with the same ompS DNA template, OmpT does show a mobility shift. I 210 Figure 19: DNAsel footprinting of ompS promoter DNA to identify the OmpT binding site(s). Approximately 160,000 cpm of 32P-5'-end-labelled ompS promoter DNA fragments were incubated with virulent L. pneumophila cell extracts, digested with DNAsel, and electrophoresed on a sequencing gel as described in materials and methods. Lane 1, no protein was added. In lanes 2-6, 4.4, 8.8, 22.0, 30.8, and 44 fxg of protein respectively were added to the DNA mixture. Protected areas are indicated by square brackets on the left hand side of the figure. The sequence of the protected areas are indicated beside the sequencing ladder on the right hand side of the figure. 211 ACGT 12 3 4 5 6 Figure 19 212 ompS promoter:lacZ fusions: An ompS promoter : lacZ fusion was created using the operon fusion vector pRS551 to facilitate the study of ompS promoter activity in L. pneumophila under different environmental conditions. The 285 bp ompS promoter amplicon was cloned into the multiple cloning site of pRS551, thus placing the lac operon under the control of the ompS promoter. Any residual transcription from upstream promoters would be terminated by the four tandem copies of TI terminators in the vector. The vector contains the genes coding for both ampicillin and kanamycin resistance. Kanamycin resistance is an effective screening tool for Legionella. The fusion construct (pRIS12) was inserted into L. pneumophila, by electroporation and the ompS promoter activity was measured by the levels of (3 galactosidase activity. The results of the (3-galactosidase assays are presented in Table 2. Both L. pneumophila Svir and Avir strains harbouring pRIS12 showed approximately 5 fold increases in 3-galactosidase levels compared to the strains harbouring the vector control, pRS551. When Svir strains with pRIS12 were challenged with 0.85% sodium chloride for 1 h there was a 26% decrease in p-galactosidase levels L 213 compared to unchallenged controls. However, the Avir strain challenged with 0.85% sodium chloride for 1 h showed only a 5% decrease in p-galactosidase activity. Challenge with 0.85% potassium chloride had no significant effect on p- galactosidase activity for either strain. I 214 Figure 20: Construction of the ompS promoter:lacz fusion. An ompS promoter:lacZ fusion was created using the operon fusion vector pRS551. Panel A depicts the fusion cor.^truct pRIS12. Panel B shows a Southern blot assay demonstrating the cloning of ompS promoter into the multiple cloning site of pRS551. In this assay, plasmid DNA from 4 different pRIS12 clones (Lanes A-D) and pRS551 (Lane V) were digested with the restriction enzymes EcoRI and BamHl and probed with radiolabelled ompS promoter DNA. Lanes A-D represent pRIS12 clones. Lane V depict the vector control. 215 Figure 20 A 216 A B C D V Figure 20 B Table 2: ompS Promoter Activity in Legionella pneumophila. Bacterium Plasmid Additive ll-gai % decrease Activity in G-gal (Miller Units) levels with additive Lp. Svir pRS 551 . 98 ±8 _ L.p. Svir pRIS12 . 567 ±13 _ Lp. Svir pRIS12 NaCl 419±8 26 L.p. Svir pRIS12 KCI 550 ±15 3 L.p. Avir pRS551 _ 227 ± 70 - L.p. Avir pRIS12 _ 934 ±3 - Lp. Avir pRIS12 NaCl 888 ± 35 5 Lp. Avir pRIS12 KCI 906 ± 40 3 L l 218 As indicated earlier, the results of the mobility shift and southwestern blot assays showed a decrease in the binding of the DNA binding protein OmpT to ompS promoter DNA when virulent (Svir) but not avirulent (Avir) L. pneumophila strains are challenged with 0.85% sodium chloride. In L. pneumophila Svir harbouring pRIS12, the decrease in the ompS promoter activity in the presence of sodium chloride is reflected as a decrease in 3-galactosidast levels. Thus the results of these p-galactosidase assays provide evidence to suggest that OmpT is a transcriptional activator of ompS, and that the ompS promoter activity is depleted under physiological sodium chloride levels in virulent but not in avirulent L. pneumophila strains. Cloning and sequencing of ompT: Two strategies were used to clone the gene coding for the DNA binding protein OmpT. One was a direct genetic approach and the other was a reverse genetic approach; i.e., to purify the protein, obtain the N-terminal protein sequence, and design appropriate oligonucleotide DNA probes. I I 219 a: Genetic approach: The basis of the direct genetic approach was to place the ompS promoter:lacZ fusion in single copy in a hlac/recA E. coli background. Since the ompS promoter is inactive in E. coli, the colonies will be white or pale blue when plated on LB + X-gal. Then a L. pneumophila genomic library could be introduced into this strain thereby jL-ermitting the screening of colonies for those showing dark blue colour on LB/X-gal. These could be further analyzed for the presence OmpT DNA sequences. By using the genetic approach described in the methods section, a single copy of the ompS promoter -.lacZ fusion was placed in the chromosome of E. coli strain MC4101. Using this E. coli strain harbouring the ompS promoter : lacZ fusion, a L. pneumophila Svir genomic library created in pBluescript was screened for possible ompT clones. Following this protocol, two dark blue E. coli MC4101 colonies harbouring A.-RIS12 in the chromosome and pBluescript with L. pneumophila chromosomal DNA were isolated by screening on LB agar containing Kanamycin, IPTG and X-gal. By (3- galactosidase assays, these clones were shown to produce a two fold increase in 3-galactosidase level over the negative 220 control (Table 3). The restriction analysis of the two clones revealed that both clones had L. pneumophila chromosomal DNA inserts of 1.9 kb. The inserts from both clones had identical restriction maps. Therefore at this point, both clones were considered to be identical and were named E. coli MC4102. Mobility shift assays were conducted using cell extracts from E. coli MC4102, E. coli MC4101::ARIS12 with pBluescript and E. coli MC4101. As seen in Figure 21, lanes 2-4, none of the E. coli strains showed retardation of ompS promoter DNA comparable to L. pneumophila extract (lane 1). Therefore, the results of the mobility shift and p-galactosidase experiment do not show encouraging data to suggest that the clone contains ompT. Further experiments were performed to establish the sequence of the L. pne-imophila DNA purified from E. coli MC4102. The restriction analysis performed on the plasmid isolated from E. coli MC4102 indicated the presence of a 1.9 kb DNA fragment cloned into pBluescript. The cloned DNA was recovered from pBluescript using the restriction enzymes Sacl and EcoRI then re-cloned into M13 phage Tgl30 and Tgl31. Single stranded DNA from these phage was used for sequencing. Sequencing of the 1.9 kb L. pneumophila insert I I 221 DNA did not reveal any open reading frames or sequence homology to any known DNA binding proteins. This approach was abandoned in favor of the reverse genetic approach. 222 Table 3: p-galactosidase activity of possible ompT clones. Bacterial strain used p- galactosidase activity E. coli ME:: XRIS12 109.4 ± 19 Possible ompT clone #1 207.6 + 25 Possible ompT clone #2 202.1 + 8 The P-gal levels presented are the mean values of triplicate assays. u 223 Figure 21: Gel retardation of ompS promoter DNA by E.coli cell extracts. Cell extracts were prepared from E. coli MC4101, E. coli MC4101::ARIS12 containing the pBluescript vector and E. coli MC4102. Preparation of cell extracts was conducted as explained for L. pneumophila. Lane P depicts the ompS promoter DNA amplicon. In lane 1, the DNA amplicon was incubated with virulent L. pneumophila cell extract. In lanes 2-4, the DNA amplicon was incubated with extracts from E. coli MC4101, E. coli MC4101::XRIS12, and E. coli MC4102. None of the E. coli cell extracts show a strong retardation of ompS DNA as seen with L. pneumophila cell extract. However the E. coli extracts did show a weak band shift that was distinct from the shift seen with L. pneumophila extracts (shown by the arrow). 224 1 2 3 «»* Ik Figure 21 225 b: Purification of OmpT: In an attempt to purify OmpT, a L. pneumophila cell extract was first separated on a DE-52 anion exchange column. Figure 22 shows the protein elution profile of this separation. The eluted protein fractions were tested for binding to ompS promoter DNA using a mobility shift assay (Figure 23). According to the results of these assays the DNA binding protein OmpT was eluted approximately at a concentration of 0.5 M potassium chloride. The protein fractions showing a mobility shift were pooled and used for further purification. The protein solution was first de-salted by dialysis against 1 liter of buffer X containing 0.35 M potassium chloride and then concentrated using an amicon filter as described in the methods section. The concentrated protein was then passed through an avidin:agarose column in an attempt to affinity purify OmpT. The ompS promoter DNA was biotinylated by using a biotinylated oligonucleotide. The DNA was then attached to avidin agarose by virtue of its biotinylated end and used for affinity purification of OmpT. Due to the small column used, the protein fractions collected from the I 226 avidin:agarose column were extremely small in volume (approximately 10 ul). The protein in these column eluates was separated by SDS-PAGE, transferred to Immobilon and stained with Coomassie Brilliant Blue. However due to the low protein concentrations in these samples the protein bands were barely visible. Although the stained membrane showed a faint but distinct band at approximately 15 kDa range, there was not enough sample to perform a southwestern blot to confirm whether the band corresponded to OmpT. 1 227 Figure 22: Protein elution profile of L. pneumophila cell extracts purified by passage through a DE-52 column. L. pneumophila cell extracts were prepared as described in the methods section. These extracts were diluted approximately 1:3 with buffer X and loaded onto the DE-52 column. The proteins were eluted using a potassium chloride gradient (35 mM - 500 mM) in Buffer X. After the gradient, 500 mM and 1 M potassium chloride concentrations in Buffer X were used to elute the remaining proteins from the column. The protein elution was monitored using an ISCO UV monitor. The parameters of the monitor used were; OD=l, chart speed, 3 cm/h. o Tl >-s H-*- fa OQ o ci-i CD o3 cz> NJ to 8Z3 I 22 Figure 23: Identification of the OmpT fraction from DE-52 column eluaites using mobility shift assays. OmpT was partially purified using a DE-52 column. Mobility shift assays were used to identify the protein fractions containing OmpT. Lane " depicts the ompS promoter DNA amplicon. In lane VI, the DNA amplicon was incubated with the unfractionated L. pneumophila extracts which were used in the purification experiment and was stored at -70°C. In lane V2, the DNA amplicon was incubated with the same cell extract stored at 4°C. In lane V3, the DNA amplicon was incubated with the 3-fold diluted cell extract that were loaded onto the column. In lanes 1-28, the DNA amplicon was incubated with the protein fractions eluted from DE-52 column. According to this assay, fractions 13-23 contain a protein that binds to ompS DNA amplicon. 1 I . 230 P V1 V2 V3 12 3 4 5 6 7 8 9 10 I P 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 ^IPN* Vfc- *!•• $$& ism *** Figure 23 231 Immunological Characterization of OmpS: Guinea Pig Studies: Legionella infect a number of animal species with varying degrees of severity. The most susceptible and therefore favoured animal model for studying Legionnaires' disease is the guinea pig. Hence, guinea pigs were used as the animal model for studying the immune responses to the two major L. pneumophila antigens; OmpS and Hsp60. The animal studies were performed with the assistance of Mr. M.B. Ripley, Dept. Microbiology & Immunology, Dalhousie University and Ms. Rosemary Kuehn, Victoria General Hospital, Halifax, Nova Scotia. Cellular immune responses of previously infected guinea pigs to L. pneumophila protein antigens: The cellular immune responses of guinea pigs surviving L. pneumophila infections were studied by measuring the delayed type hypersensitivity and lymphocyte proliferation responses (LPR) to the antigens OmpS and Hsp60. LPR assays v/ere conducted with the assistance of Dr. D.W. Hoskin, Dept. Microbiology & Immunology, Dalhousie University. In an initial study, 9 animals that had survived LD50 doses of I 232 environment-, and patient- derived L. pneumophila were skin tested 3 0 days postinfection for cutaneous delayed hypersensitive reactions (DTH) to the purified protein antigens OmpS, Hsp60, and protease of L. pneumophila. The results showed a strong DTH response in these animals to both OmpS (9 ± 1 mm at 24 h, 6+1 mm at 48 h) and Hsp60 (12 ± 2 mm at 24 h, <4 mm at 48 h). The DTH response to L. pneumophila protease was lower compared to the others (6+1 mm at 24 h, 0 mm at 48 h). No response was noted in either the animals skin tested with 10 ug of ovalbumin (OA) or in the uninfected animals tested with these antigens. In a similar study, 4 groups of animals surviving infection with different doses of L. pneumophila were tested for DTH and LPR to OmpS, Hsp60 and OA. The 4 groups of animals were as follows; CI- uninfected controls, SLD (sub lethal dose)- survived an intra peritoneal (i.p.) dose of 104 CPU, LD (lethal dose)- survived an i.p. dose of 107 CFU, HD (high lethal dose)- survived an i.p. dose of 108 CFU. Control animals had no DTH response to any of the antigens tested. Animals in the SLD group showed a very weak DTH response to Hsp60 and no significant DTH response to OA and OmpS. The animals in groups LD and HD which were survivors of high 233 doses of infection showed significant DTH responses both to OmpS and Hsp60 (Table 4). The LPR of the animals was measured by challenging splenic lymphocytes with the test antigens for 144 h (6 days). Concanavalin A (ConA), a known T cell mitogen, and pokeweed mitogen (PWM), a T and B cell mitogen, were used as positive controls. OA was used as the negative control. The assays were conducted in triplicate and the mean values for the LPR are shown in Table 5. Maximum proliferation was observed with 10 ug/ml of OmpS and Hsp60. There were no significant differences in LPR to OmpS and Hsp60 in animals from the SLD group. However, animals surviving higher-dose infections, LD and HD, showed greater LPR i_o OmpS than to Hsp60. The antibody titers of the animals to OmpS, Hsp60 and OA were also determined. None of the animals showed antibody titers greater than 1:64 to OmpS or OA. Animals in the SLD group showed no appreciable antibody titers to Hsp60. However, animals in LD and HD groups exhibited antibody titers to Hsp60 ranging from slightly higher than 1:64 to 1:2,800. There was no correlation between the magnitude of LPR to Hsp60 and the antibody titers. 234 Table 4: DTH reactions to purified Legionella protein antigens in guinea pigs surviving L. pneumophila infection. Mean skin induration diam.(mm) ± SDa Guinea pig Challenge groupb dose (CFU) OA OmpS Hsp60 CI None 0 0 0 SLD 104 1.3+1.8 1.2+1.6 2.3± 2.0 LD 107 0 5.3 + 1.1 11.2 ± 3.4 HD 108 0 7.1 ± 2.3 6.3±1.4 a Skin indurations were measured 48 h after antigen challenge. bCl, control group not exposed to L. pneumophila. 235 Table 5: LPRs of guinea pigs surviving legionellosis'. Group Animal ConA PWM OA OmpS Hsp60 # 5 5 10 10 10 ug/ml ug/ml ug/ml Ug/ml ug/ml Cont. 1 66.7 50.2 <0.1 1.9 1.3 2 8.59 27.7 <0.1 <0.1 1.0 SLD 1 11.0 7.2 <0.1 22.16 15.19 2 23.4 14.1 <0.1 16.2 24.61 3 3.33 2.61 <0.1 38.46 20.31 LD 1 23.5 18.2 <0.1 55.27 24.19 2 1.04 0.72 <0.1 11.6 9.37 3 2.02 1.2 <0.1 11.0 2.47 HD 1 0.64 2.1 <0.1 20.85 13.1 2 1.40 1.37 <0.1 15 .03 3.37 3 2.75 3.71 <0.1 16.1 8.9 a; LPRs to OmpS and Hsp60 of splenic lymphocytes were measured 5 weeks postinfection from guinea pigs surviving sublethal (SLD), low-dose (LD), and high-dose (HD)challenge. The control group (Cont.) was not exposed to L. pneumophila. LPR was measured by challenging splenic lymphocytes with antigens for 6 days and then pulsing with [3H] thymidine for 4 h. LPRs are expressed as (Acpm) X 102. The results are means of triplicate assays. The standard deviation ranged from 10-15% of the mean cpm. I I 236 Immune responses to OmpS: Survival from legionellosis is thought to require a cellular immune response (Horwitz, 1983; Klein et al., 1984; Breiman and Horwitz, 1987). Therefore, to establish the efficacy of OmpS as a vaccine candidate, it was important to determine its ability to stimulate a cellular immune response. To test this, guinea pigs were immunized with 25 ug of OmpS in Freund's incomplete adjuvant, and 5 weeks later they were skin tested to assess DTH responses and their peripheral blood and spleens were collected to assess the LPR. Four guinea pigs were immunized with 25 fig OA and used as controls. Concanavalin A was used as a positive control for LPR assays. Table 6 shows the results of two separate studies. Animals immunized with OmpS developed stronger DTH reactions as well as LPR compared to the OA controls. The reciprocal antibody titers (measured by ELISA) of animals immunized with OmpS protein were 2:500. 237 Table 6: LPR and antibody titers of OmpS-vaccinated guinea pigs to OmpS and Hsp60*. Group Antigen Mean Mean Reciprocal antibody tested (103 Acpm) SI titer OmpS Hsp60 0A1 (2) - <50 <50 ConA 9.3 + 1.3 51.2 OA <0.1 0.9 OmpS <0.1 0.8 OA2 (2) - <50 <50 ConA 3.7 + 1.1 21.0 OA 0.7 + 0.4 4.2 OmpS 0.4 ± 0.2 2.3 OmpSl (2) - 500 <50 ConA 9.1 + 2.3 45.0 OA <0.1 1.0 OmpS 7.2 + 1.2 36.0 OmpS 2 (2) - 500 <50 ConA 6.5 + 2.1 16.7 OA 0.5 ± 0.3 1.3 OmpS 4.7 ± 0.2 11.8 a Guinea pigs were immunized with 25 ug of OmpS in 100 ul of Freund's incomplete adjuvant. The DTH responses were determined 5 weeks postimmunization and the peripheral blood lymphocytes were collected for LPR assays. Antibody titers were determined by ELISA with purified antigens. The Acpm and SI values represent means of triplicate assays. I 238 The vaccine trial: Vaccine trials were performed in collaboration with Dr. Paul H. Edelstein, University of Pennsylvania Medical Center. In this study, seven animals per group were immunized with either OmpS, Hsp60 or OA. The primary immunization was given in Freund's complete adjuvant. Three weeks later, animals were reimmunized with the same amount of protein in Freund's incomplete adjuvant. Six weeks after the primary immunization, animals were challenged with a 2.5 X LD50 dose of L. pneumophila serogroup 1 by the intratracheal route. One animal in the Hsp60 group died 24 h post infection with L. pneumophila due to a polymicrobial neck abscess. Data from this animal were excluded from the analysis. All animals immunized with OmpS survived for the entire 10 day study period. On the contrary, the animals immunized with OA were all dead by day 4 (P=0.0006, Fisher exact test)(Figure 24). All animals in the Hsp60 immunized group were dead by day 5 except for one animal which was sick but recovered and survived the entire study period. This puts the survival rate of animals immunized with Hsp60 at 17% (P=0.005 versus the OmpS group, by Fisher's exact test) . I I 239 Mean baseline weights were similar among the three groups of animals (P > 0.8, one-way analysis of variance) and decreased following infection. Until day 2, the weight loss was somewhat similar among all three groups of animals. However, from day 3 onwards, the OmpS-immunized animals started to gain weight and continued to do so throughout the study period. By the end of the 10 day study period, these animals had exceeded their pre-infection weight. Except for one animal, all animals immunized with Hsp60 continued to lose weight and were dead by day 4. One Hsp60-immunized animal although very sick, managed to survive the infection and started gaining weight by the end of day 5. None of the OA-immunized animals gained weight following infection and all were dead by day four (Figure 25). Quantitative lung cultures demonstrated detectable L. pneumophila in all Hsp60 and OA-immunized animals. However only one of the seven OmpS-immunized animals showed any detectable L. pneumophila in lung cultures (P=0.005, Fisher's exact test). The mean lung concentration of L. pneumophila for OA-immunized animals was 10.1 log10 CFU/g and ranged from 9.5-10.4. For Hsp60-immunized animals the 240 mean lung concentration of L. pneumophila was 8.7 log10 when the value of the single surviving animal (3.3) was excluded (7.8 without excluding the survivor) and ranged from 8.0 - 9.2. According to histopathologic scoring of fixed lung samples, Hsp60-immunized animals showed 75% of consolidated lung tissue compared with only 12% in OmpS- immunized animals (P=0.008, Mann-Whitney two-sample test). All lung samples from OmpS-immunized animals had predominantly mononuclear infiltrates whereas 80% of HSP60- immunized ani rials showed both polymorphonuclear and monocytic infiltrates (data not shown). Antibody titers against OmpS, Hsp60 and OA were determined by ELISA for five of seven animals from each group. The results are presented both as the mean of absorption for 1:400 dilution of the sera and the estimated titer (Table 7). Guinea pigs immunized with Hsp60 and OA showed strong humoral immune responses with antibody titers in excess of 12,800. However OmpS-immunized animals showed very low antibody titers to OmpS. I 241 Figure 24: Survival of vaccinated guinea pigs infected with 2.5 X LD50 dose of L. pneumophila administered by the intratracheal route. Seven animals were vaccinated with OmpS, Hsp60 or OA. Six weeks following the primary immunization, animals were challenged with 2.5 X LD50 dose of virulent L. pneumophila. The animals were monitored for a 10 day period following this challenge. 700 600 en cr CD 500 o MOMP • HSP A OA 400 -lh 0 12 3 4 5 10 DAYS POST-INFECTION Figure 25 I 243 Figure 25: Weights of vaccinated guinea pigs challenged with 2.5 X LD50 dose of L. pneumophila administered by the intratracheal route. Seven animals were vaccinated with OmpS, Hsp60 or OA. Six weeks following the primary immunization, animals were challenged with 2.5 X LD50 dose of virulent L. pneumophila. ^le animals were weighed throughout a 10 day period following this challenge. Means and the standard deviations are shown. I 700 600 CO '-U CD 500 o MOMP • HSP A OA 400 0 2 3 4 5 10 DAYS POST-INFECTION Figure 25 M £• L 245 Table 7: ELISA results for immunized guinea pigs prior to infection. Animal group OD at 1:400 Titer (# of animals) dilution of serum (l/dilution) Hsp60 (5) 1.51 >12,800 OmpS (5) 0.46 400-1,600 OA (2) 1.06 >12,800 246 Cellular immune responses to Legionella antigens in humans: To determine whether humans surviving Legionnaires' disease also develop cellular immune responses to OmpS and Hsp60, LPR to these antigens were measured in a group of patients that had culture confirmed Legionnaires' disease and survived the disease. A control group of healthy individuals who were seronegative for Legionnaires' disease were also tested. Figure 26 shows the results of this study. It was evident throughout this study that there was great variation among individuals in their LPR. These variations were observed even in individuals tested on different days. However, lymphocytes from patients surviving Legionnaires' disease exhibited a significant proliferative response to OmpS challenge (SI=16.6 + 10.2) compared to the controls (SI = 4.4 + 2.6; p = 0.0002, Mann-Whitney two-sample test). The magnitude of difference in proliferative responses to Hsp60 in patients (SI = 7.8 ± 4.5) compared with the controls (SI = 3.0 ± 2.2) was not as large as for OmpS (p s 0.0001, Mann- Whitney two-sample test). 247 Figure 26: Scatter graph of human LPRs to L. pneumophila antigens. Peripheral blood lymphocytes were collected from healthy controls and from patients surviving Legionnaires' disease. LPR assays were conducted as described in the methods section. The stimulation indexes (SI) for each individual represent the mean of triplicate assays. The filled circles represent the mean for each group. The patient group was composed of 10 individuals. Two were heart transplant recipients, 3 had no significant underlying disease, 2 had malignancies, 1 had systemic lupus erythematosus, 1 had chronic obstructive pulmonary disease, and 1 had rheumatoid arthritis. Stimulation Index ro CJ) o -T- Control OmpS O O CD O • O OJ 3 XI IT O o Control Hsp - O CXD OO iaoCD •< TJ —i CD o c 5-^" a> w Patients OmpS OO o 5' <4° a 33 ID (A XI O Patients Hsp o a a o ol (A 03 I 249 A few of the patients in this study had undergone heart transplantation and were on Cyclosporin A therapy to suppress organ rejection. When their lymphocytes were used in lymphocyte proliferation assays, they showed no proliferative responses to any antigens. This is to be expected, considering their immunosuppressed status due to Cyclosporin A therapy. However, when their lymphocytes were incubated in culture media for 24 h and washed in PBS prior to use, they -d show a proliferative response to ConA (positive co ;rol) and OmpS (Table 8). Therefore it was very interesting to note that these patients were able to mount a cellular immune response to OmpS despite the Cyclosporin A therapy. 250 Table 8: LPRs of incubated (I) vs unincubated (Ul) lymphocytes from a heart transplant patient 5 months after the onset of infection. _ Day 2 Day3 Day6 Day 8 Antigen Ul I Ul I Ul I Ul I ConA 2.9 5.4 2.3 7.9 0.7 3.3 0.6 0.6 OmpS 1.1 1.2 1.3 3.3 1.5 14.3 1.5 2.0 Hsp60 0.9 1.4 0.6 1.9 0.7 4.3 1.2 1.4 OA 0.7 1.0 0.70.92.1 ND 3.4 1.1 Stimulation indices (SI) presented are mean values of triplicate assays. 251 All individuals in this study were also tested for their antibody titers to Hsp60 and OmpS by ELISA. With the exception of one heart transplant patient with Legionnaires' disease who exhibited a high titer to Hsp60, none of the patients or controls showed detectable titers to either of the antigens. The patient with high antibody titer against Hsp60 showed no detectable antibody titer to OmpS. According to immunofluorescence data, all patients who recovered from Legionnaires' disease seroconverted to L. pneumophila serogroup 1 LP£! antigen, while controls remained negative. DISCUSSION The objective of this study was to characterize the major outer membrane protein, OmpS of L. pneumophila both from a genetic and from an immunological standpoint. The genetic study focused mainly on elucidating the mode of regulation of the ompS gene in L. pneumophila while the immunologic study centered on determining the immunogenicity of OmpS in humans and guinea pigs surviving Legionnaire's disease. The latter studies assessed the potential of OmpS as a possible vaccine against Legionnaires' disease. Genetic characterization of OmpS: Cloning and sequencing of ompS and primer extension analysis: The gene coding for OmpS was cloned and sequenced in this laboratory (Hoffman et. al., 1992). This gene consists of an 891 bp open reading frame which codes for a polypepti.-3 of 297 amino acids. Out of the 297 amino acids, 21 amino acids form the signal sequence leaving a mature protein of 276 amino acids. The deduced amino acid sequence 252 253 revealed four cysteine residues and an abundance of glycine and aromatic amino acids. Out of the four cysteine residues, two are located in the amino terminal and the other two in the carboxy terminal region. According to protein data base searches, OmpS is not closely related to other porin proteins except for the 69 kDa outer membrane protein (Pertactin) of B. pertussis. It shows a 45% amino acid similarity and a 21% identity with this protein. OmpS also shows 22% identity with the variable repeat region of CRl; a macrophage integrin protein (Hoffman et. al., 1992). Similar no L. pneumophila, the obligate intracellular bacterial pathogen C. trachomatis possesses disulfide cross- linked outer membrane proteins. Out of these, the 40 kDa major outer membrane protein accounts for 60% of the total chlamydial outer membrane protein by weight (Caldwell et. al., 1981). The chlamydiae show a unique life cycle involving metabolically dormant elementary bodies (EBs) and vegetative reticulate bodies (RBs). The EBs have been shown to interact with the host cell surface and promote phagocytosis. After phagocytosis, the EBs transform into the vegetative RBs. The RBs have the capacity to inhibit 254 phagolysosomal fusion and grow within the phagosome. Following several rounds of multiplication, the RBs start to differentiate into EBs which will later be released from the cell. A Chlamydia MOMP, which contains nine cysteine residues that form disulfide cross-links within the protein complex as well as with other outer membrane proteins, plays a vital role in the structural integrity of the EBs which have to survive in the extracelluar environment (Newhall and Jones, 1983; Bavoli et. al., 1984; Hatch et. al., 1984). Also according to a report by Hackstadt et. al. (1985), the reduction of disulfide bonds of MOMP triggers the differentiation of EBs to RBs. Apart from this, the C. trachomatis MOMP is also known to function as a porin (Bavoil et. al., 1984) as well as a polymorphic surface antigen (Caldwell et. al., 1981; Stephens et. al., 1982). More recent studies reveal that MOMP of C. trachomatis is glycosylated and that the glycan portion of the protein is involved in the attachment of the bacterium to HeLa cells (Swanson and Kuo, 1991, 1994). A recent report by Gibson et. al. (19 94) suggest the presence of a glycoprotein-like adhesion molecule on the surface of L. pneumophila. However, so far, glycosylation has not been observed in OmpS. Even 255 though the MOMP of C. trachomatis does not show any sequence similarity with L. pneumophila OmpS, it does have a number of functional similarities; i.e.: porin function, entry into host, and structural integrity to the cell. Although disulfide cross-linked outer membrane proteins are a rare feature among gram negative bacteria, it appears to be a common feature of many members of the genus Legionella (Butler and Hoffman, 1990) . A monoclonal antibody commercially available for diagnosing legionellosis recognizes a common epitope of OmpS shared by all serogroups of L. pneumophila (Gosting et. al., 1984). Southern blot analysis carried out in the current study confirmed that all the L. pneumophila serogroups tested possess highly conserved ompS sequences. However, some restriction polymorphisms exist between the different serogroups. Under moderate stringency conditions, DNA-DNA hybridization was evident with a number of other Legionella species. This agrees with the suggestion of Butler et. al. (1985) that major outer membrane proteins of other Legionella species might also contain genus-common epitopes. Apart "f'nom ompS, there are several other Legionella genes that show genus ^ 256 common sequences; viz: mip (Cianciotto et. al., 1990), htpAB (Hoffman et. al., 1990), ppl (Engleberg et. al., 1986; Ludwig et. al., 1991). In contrast, the gene coding for the zinc metalloprotease {pro) may be restricted to the species L. pneumophila (Quinn and Tompkins, 1989) . Primer-extension studies, performed following the sequencing of ompS, identified a single transcription start site located 97 bp upstream of the translation start site and a unique promoter sequence that differs from a consensus E. coli-like promoter sequence. Other genes of L. pneumophila that have been sequenced so far, viz; pro (Quinn and Tompkins, 1989), mip (Engleberg et. al., 1989), htpAB (Hoffman et. al., 1990), recA (Zhao and Dreyfus, 1990), ppl (Ludwig et. al., 1991), lly (Wintermeyer et. al., 1994), fur (Hickey and Cianciotto, 1994) all appear to possess an E. coli-like promoter sequence and are expressed in E. coli. Expression of ompS in E. coli: The cloning studies of L. pneumophila ompS revealed that this gene is not expressed in E. coli. Deliberate attempts to express ompS in E. coli by cloning the gene 257 under an E. coli promoter proved unsuccessful, suggesting that the expression of ompS is lethal to this bacterium. This was further proved by the fact that this gene could be cloned under a T7 promoter (pTOMPSl, pT0MPS2) and stably maintained in E. coli JF626, a strain that lacks T7 RNA polymerase. However no viable cells were obtained upon the transfer of these clones to E. coli BL21 which is a strain that expresses T7 RNA polymerase. Furthermore, to observe the expression of OmpS in E. coli, a selective radiolabelling protocol involving the T7 promoter system was used. A 1.5 kb DNA fragment containing the entire ompS gene (pTOPMSl) and a 1.3 kb DNA fragment containing only part of ompS (pT0MPS2) were cioned into the T7 vector pT7-5. Then the E. coli JF626 strains harbouring the pT7-5::ompS constructs were infected with a M13 bacteriophage expressing T7 RNA polymerase. The transcription by bacterial RNA polymerases was inhibited by adding rifampicin and protein synthesis was monitored by the addition of [n-35S] Methionine. By this protocol, faint expression of OmpS (a 28 kDa protein on the autoradiogram) was seen in clones containing the entire ompS gene. The faint expression was most likely due to the inhibitory effects on growth of E. 1 I 258 coli expressing OmpS. E. coli clones containing the truncated ompS gene expressed a number of lower molecular weight proteins; probably truncated versions of OmpS or degradation products of OmpS. Generally, the expression of cloned porin proteins, especially from nonenteric pathogens are inhibitory for the growth of E. coli (Barlow et. al, 1987; Gotschlich et. al., 1987; Koehler et. al, 1992; Dascher et. al., 1993). This includes the MOMP of Chlamydia. Studies have shown that the expression of the MOMP of C. trachomatis and C. psittaci are detrimental to E. coli cells (Koehler et. al, 1992; Dascher et. al., 1993). Further to the protein expression studies, Northern blot data also confirm that fact that ompS is not expressed in E. coli. These observations suggest that the ompS promoter is not recognized in E. coli and therefore it is possible that L. pneumophila possess a unique sigma factor or a transcription factor that is involved in the transcription of ompS which is absent in E. coli. In this regard, a recent study has reported that when virulent L. pneumophila are intracellular or suspended in a medium containing high concentrations of sodium chloride, the expression of ompS is down regulated. However, the avirulent mutants do not show this differential I P 259 expression of ompS (Fernandez, 1992). The present study also shows that a 1 hour challenge of virulent L. pneumophila cells with 0.85% sodium chloride leads to a marked decrease in ompS mRNA levels. Interestingly, the only phenotypic difference reported so far between virulent and avirulent L. pneumophila is the avirulent strains ability to grow on media containing high levels (> 0.6%) of sodium chloride. The virulent strains cannot grow on media containing sodium chloride (Catrenich and Johnson, 1989). Therefore, studying the underlying mechanisms associated with sodium chloride tolerance could lead to a better understanding of the regulation of virulence in L. pneumophila. Mobility shift and southwestern blot assays: According to the mobility shift assays, cell extracts from both virulent and avirulent L. pneumophila induce a retardation of ompS promoter DNA. This suggests the presence of a DNA binding protein(s) that can bind to DNA sequences in the ompS promoter region. The results of the competition assay as well as the controls (the use of Salmon sperm DNA, poly dldC and radiolabelled internal coding sequences of ompS) established that the binding of this protein(s) to I I 260 ompS promoter is specific. Compared to extracts of the virulent L. pneumophila strain, those derived from avirulent cells showed a greater retardation of ompS DNA. This could be due to the involvement of different proteins or to the same protein binding in multimeric forms. According to the southwestern blot assays a similar size protein (15 kDa, OmpT) is involved in both virulent and avirulent cells which may suggest the latter explanation. Furthermore, mobility shift and southwestern blot experiments also revealed that there is a decrease in binding of OmpT to ompS DNA in virulent L. pneumophila strains when the cells are challenged with 0.85% sodium chloride prior to obtaining the cell extracts. Mobility shift assays in which the ompS promoter DNA amplicon was .uvubated With sodium chloride challenged virulent cell " . .racts, showed a smearing effect. This may be due either to the loss of binding affinity of OmpT to ompS DNA, or degradation of OmpT upon sodium chloride challenge. Avirulent cells do not show such a decrease in binding of OmpT to ompS DNA upon sodium chloride challenge. Here again, the avirulent cells appear to be blind to the change in 261 sodium chloride levels. Challenge of virulent cells with 0.85% magnesium chloride or 0.85% calcium chloride also effected a decrease in binding of OmpT to ompS promoter DNA. However, the addition of these two salts appeared to have a deleterious effect on the viability of L. pneumophila cells as judged by their colony morphology and microscopic observations. Therefore it is possible that the decrease in shift was due merely to the degradation of OmpT upon cell lysis. Use of 0.5% calcium and magnesium chloride was also toxic to L. pneumophila cells. Concentrations lower than this were not investigated, but perhaps would have been less toxic. The challenge of cells with 0.85% potassium chloride had no effect on OmpT binding to ompS DNA in either the virulent or avirulent strains. This suggests that the effect of sodium chloride on OmpT is not merely an osmotic effect but rather a sodium chloride specific effect. The time course experiments carried out to study the kinetics of the sodium chloride effect on OmpT showed a complete abrogation of binding of OmpT to ompS DNA 10 min post-challenge in virulent strains of L. pneumophila. The same kinetics were seen when the virulent cells were challenged with 100 /xg/ml of chloramphenicol, a protein synthesis inhibitor. This I 262 suggests that OmpT is a highly unstable protein and thus must be constantly synthesised by the bacteria. The smear seen with sodium chloride challenged virulent cell extracts may suggest the rapid degradation of OmpT. Seeing similar kinetics with sodium chloride and chloramphenicol challenge may suggest that sodium chloride affects the synthesis of OmpT rather than its DNA binding ability. ompS promoter : lacZ fusions were used to further confirm the mobility shift results. These fusions were created in the operon fusion vector pRS551. This vector possesses a complete lac operon with a functional ribosome binding site. However, it does not have a promoter sequence and thus is not expressed. Therefore, by cloning the ompS promoter sequence upstream of the lac operon, it is possible to monitor its activity by monitoring the (3- gal levels. The results of the p-galactosidase assays performed using the ompS promoter : lacZ fusions confirmed the results of the mobility shift experiments. When virulent L. pneumophila strains harbouring the ompS promoter : lacZ fusion (pRIS12) were challenged for 1 h with 0.85% sodium chloride, there was a 26% decrease in P-gal levels. The avirulent cells I I 263 harbouring pRIS12 showed only 5% decrease in p-gal levels upon 1 h challenge with 0.85% sodium chloride. Neither of the strains showed a significant decrease in (3-gal levels when challenged with 0.85% potassium chloride for 1 h. On the whole, the avirulent strains showed a higher level of p-gal activity compared to the virulent strains. This could be due to a difference in plasmid copy number between the strains. Also, it is possible that the 26% decrease in (3-gal activity in virulent strains when challenged for 1 h with 0.85% sodium chloride is not a true reflection of the level of decrease in ompS promoter activity. This is true especially if p-gal is a stable protein in L. pneumophila and is not turned over rapidly. In this case the residual protein will mask the decrease in transcript levels and thus the true promoter activity. This is supported by the results of the northern blot analysis, in which sodium chloride challenged virulent L. pneumophila cells show a dramatic decrease in the level of ompS transcripts compared to unchallenged ceils. However it is likely that sodium chloride only decreases the levels of ompS transcription and does not completely abolish it. Since OmpS p]ays a pivotal role in the cell wall integrity of L. pneumophila, it is 264 likely that the cell cannot manage to completely abolish its synthesis. This was further established by the failure of several attempts to create null mutants of ompS in L. pneumophila. However, in the same context, it is possible that a repression of ompS is needed in intracellular growth but not for in vitro growth. Therefore the absolute need for OmpS may be an artifact seen only in cells grown on laboratory media. Identification of OmpT binding sites on ompS promoter DNA: As shown in mobility shift assays, cleaving the ompS promoter DNA amplicon at the -60 bp position significantly decreased the binding of OmpT to ompS DNA. Truncating the 5' region of the ompS promoter amplicon by using the forward primers F7 and F8 also completely abolished the mobility shift. These results suggest that the -60 bp region as well as the region(s) upstream of -60 bp position may be important for the binding of OmpT to ompS DNA. The DNAsel footprinting analysis performed to identify the exact region(s) of binding of OmpT to ompS revealed four protected areas. However, these results were not reproducible at all times. Being a small protein (15 kDa) it is possible that tt 265 OmpT either does not bind tightly enough or protect a large enough region to be seen by a footprinting assay. However, it should be mentioned that under the same DNA binding conditions, and with the same ompS DNA template, OmpT does induce a mobility shift. The same dilemma has been seen with the 23 kDa DNA binding protein of B. pertussis. This protein has been shown to bind to the pertussis toxin and adenylate cyclase promoters by mobility shift assays. However it was not possible to obtain a DNA footprint with this protein (Alison A. Weiss, personal communication). Cloning the gene coding for the DNA binding protein OmpT: Both conventional and reverse genetic approaches were taken to identify the OmpT protein and the gene coding for it. The conventional genetic approach involved the insertion of an ompS promoter : lacZ fusion in single copy into the E. coli chromosome, transforming a L. pneumophila genomic library into these E. coli cells and screening for dark blue colonies by plating the cells on media containing X-gal. Since the ompS promoter is inactive in E. coli, only the clones bearing L. pneumophila DNA that have the capability to induce transcription from ompS promoter will give a dark i • r I ' 266 blue colour. This approach had been used previously to identify transcriptional activators in other bacterial pathogens such as V. cholerae (Miller and Mekalanos, 1984) and B. pertussis (Miller et. al., 1989). Using this protocol, a single E. coli colony showing dark blue colour on X-gal containing media was isolated. This clone appeared to contain a 1.9 kb piece of L. pneumophila DNA (cloned in to pBluescript). By p-gal assays, this clone showed a two fold excess in p-galactosidase levels compared to the negative control. However, it failed to alter the mobility of ompS promoter DNA in mobility shift assays. Furthermore, the sequence analysis of the insert did not reveal any open reading frames or sequence homology to any known DNA binding proteins. Therefore this clone may have contained some anomalous piece of L. pneumophila DNA that had the ability to induce low level of transcription from the ompS promoter. It is difficult to predict whether a more vigorous search for E. coli clones expressing high levels of p-gal activity would have resulted in finding ompT. However, it is possible that ompT is not expressed in E. coli or else, being a foreign protein, is subjected to proteolytic degradation. If ompT is not expressed in E. coli by its own promoter, only a I 267 fusion of the gene wit.i the lac promoter in pBluescript (vector used to obtain a clone bank of L. pneumophila) would have resulted in the expression of this gene. This would be an extremely rare occurrence. Also, if OmpT is a part of a regulatory cascade involving other proteins which are needed to induce expression of ompT, the conventional genetic approach would not be the suitable method to clone ompT. The reverse genetic approach to cloning the gene involved purifying the protein, obtaining the N-terminal amino acid sequence and using degenerate oligonucleotide primers to this sequence to identify the piece of DNA containing ompT. The protein was partially purified using a DE-52 anion exchange column. The protein elution profile indicated that the protein was eluted at around 0.5 M potassium chloride. This may suggest that OmpT is a basic protein. The use of affinity chromatography to further purify OmpT using avidin agarose beads proved to be unsuccessful. However, this protein was purified using a Heparin Sepharose column. The purified protein fractions were separated by 12% SDS PAGE, transferred to Immobilon membrane and stained with Coomassie blue. The band 268 corresponding to OmpT (confirmed to bind ompS DNA by southwestern blot assay) was excised out and sent for N- terminal amino acid sequencing. Obtaining the N-terminal amino acid sequence of this protein will pave the way for the sequencing of the gene coding for OmpT, as well characterization of the regulatory mechanism involved in the regulation of OmpT. In retrospect, the affinity purification using avidin agarose beads also yielded pure OmpT. However, the volume of eluate from this column was not large enough to perform the necessary tests (i.e., mobility shift and southwestern blot assays) needed to confirm the protein as being OmpT. Therefore, if performed on a larger scale, avidin agarose beads may also be a feasible alternative for purifying OmpT. As mentioned earlier in the discussion, OmpT has a very short half life. The extreme sensitivity of OmpT to proteolytic degradation was further apparent in the protein purification experiments. It was crucial that the protein purification steps be performed as quickly as possible after obtaining the cell extracts and that the samples were kept at or below 4°C at all times. During several attempts to purify OmpT, the protein either lost its DNA binding ability or was completely degraded after a few 269 steps of purification. Therefore, the extreme sensitivity to proteolytic degradation is a definite stumbling block in the purification of OmpT. Even in the absence of any detailed knowledge about the OmpT protein and its exact mode of action it is tempting to postulate the following model (Figure 27) based on the data gathered in the present study. In order to sense the changing levels of sodium chloride in the environment L. pneumophila must possess a sensor protein perhaps similar to BvgS or ToxR (Miller et. al., 1987; Melton and Weiss, 1989). It is most likely that in its natural habitats this sensor protein has the capability to sense a number of other environmental effectors besides changes in sodium chloride levels. Upon receiving the relevant signal(s), for example low sodium chloride levels in the fresh water, the sensor protein may activate a transducer protein which in turn activates other transcriptional activators. The results f Low NaCl Levels Sensor Protein Activates Transducer Protein Activates Synthesis of •>" Activates other genes? OmpT RNA Polymerase /—\ ompS DNA ompS mRNA Figure 27: Proposed model for regulation oiompS 271 of this study suggest that the transducer may activate the transcription of ompT rather than activating an inactive form of the protein. Therefore under low sodium chloride levels, or^T is activated and the resulting OmpT protein will in turn initiate ompS transcription. Similarly, under hugh sodium chloride levels such as physiological concentrations (0.85%) observed in the intracellular environments, there is a down regulation of ompS as a result of decreased levels of OmpT. It is also possible to speculate that OmpS with its high degree of disulfide cross- linking serves as a protective layer for L. pneumophila cell against hazardous conditions in the fresh water environment. However, once intracellular, the bacterium may not require such a rigid outer layer. On the contrary, it may need to loosen the outer membrane in order to scavenge nutrients from the intracellular environment. Therefore it is possible that, once intracellular, the bacterium has evolved mechanisms to down regulate the synthesis of OmpS. According to Abu Kwaik et. al. (1993) at least 35 L. pneumophila genes are selectively induced and 32 repressed upon infection of macrophages. According to these authors, I I 272 OmpS is one of the genes that is repressed upon infection of macrophages by L. pneumophila. According to the results of the current study and also the study by Fernandez (1992), avirulent mutants of L. pneumophila are not responsive to the signals associated with being intracellular. This lack of coordinate regulation of gene expression in response to environmental stimuli might be the basis for avirulence. Environmental regulation of virulence determinants of bacteria is a well established phenomenon that is under intense investigation. Some of the parameters that act as cues to signal the entry of the microbe into the host are elevated temperature (the body temperature as opposed to the external environment), osmolarity, pH, oxygen, various ion concentrations, with iron playing a dominant role. Different bacteria utilize different regulatory mechanisms to control their gene expression in response to these stimuli (Mekalanos, 1992). Some examples of regulators include the histidine protein kinase/response regulators, AraC transcriptional activators, LysR family of gene regulators, Lrp regulon, specific sigma factors, and the more recently discovered histone-like proteins. The involvement of 1 1 A ii 273 histidine protein kinase/ response regulators in the regulation of outer membrane proteins OmpF and OmpC of E. coli and S. typhimurium has been well characterized. More recent reports indicate that OmpF/OmpC regulation in E. coli is also under the control of other regulator systems such as Lrp (Leucine-responsive Regulatory protein) (Ernsting et. al., 1992; Ferrario et. al., 1995) and the integration host factor (IHF) protein (Helu et. ai., 1988; Huang et. al., 1990). The Lrp was first identified as the regulator of the leucine regulon of E. coli. However, it is now known to regulate the expression of over 40 genes in E. coli, increasing the expression of some and decreasing the expression of others (Ferrario, 1995). Lrp has a monomeric molecular weight of 18.8 kDa and exists in abundance in the E. coli cell; 3,000 molecules per cell when grown in glucose-based minimal medium (Calvo and Matthews, 1994) . The IHF of E. coli is a 21.8 kDa, basic, DNA binding protein composed of two non-identical subunits of 11.4 and 10.7 kDa (Freundlich et. al., 1992). IHF is considered to be a member of the E. coli histone-like family of proteins due to its ability to compact DNA and its strong amino acid sequence homology with HU (Freundlich et. al., 1992). HU is a highly r - Ml i d \ a I 274 abundant DNA binding protein which is known to play an important role in bacterial chromatin structure and has the ability to alter the topology of DNA and the available supercoils upon interaction with DNA (Rouviere-Yaniv et. al., 1979; Broyles and Pettijon, 1986). Unlike HU however, IHF binding to DNA is sequence specific. Previous literature reports that a single IHF heterodimer binds to a site of approximately 35 bp consisting of a 13 bp consensus sequence WATCAANNNNTTR (Craig and Nash, 1984; Leong et. al, 1985) . Although IHF was originally found to be a required component for site-specific recombination of bacteriophage X, subsequent work revealed that it is also involved in various other physiological activities such as a variety of other site-specific recombinational events, phage packaging and partition, DNA replication, and expression of number of genes (Friedman, 1988). IHF is also involved in the phase variation of type 1 fimbriae of E. coli (Eisenstein et. al., 1987; Dorman and Higgins, 1987). E. coli alternates between Fim+ and Fim- state. Studies have shown that there is a fimbriae switch region consisting of 314 bp immediately upstream of the fimA gene coding for the pilin subunits. This switch region containing the promoter for the fimA gene i I I 275 is present in inverted orientations in the Fim+ a..d Fim- phases. Upstream to this switch region there is a consensus binding site for IHF. It is believed that the IHF binds to this site and in\erts the orientation of the fimbriae switch region thus regulating the Fim phenotype (Eisenstein et. al., 1987; Dorman and Higgins, 1987). Interestingly, there are 3 putative IHF binding sequences in the promoter and upstream region of ompS. Taese sites are located at -80 bp, -60 bp and -35 bp regions. Of these three sites, the site at -35 bp shows 100% identity with the consensus sequence for E. coli IHF binding. This raises the possibility that OmpT may be structurally and/or functionally similar to IHF. In mobility shift assays, E. coli cell extracts appear to show some weak binding to ompS promoter DNA as shown in Figure 21. It is possible that this weak band shift represent IHF binding to ompS sequences. As mentioned earlier, ompS has a unique promoter sequence, different from other known L. pneumophila sequences and the consensus E. coli promoter sequences. Therefore if OmpT is functionally similar to IHF, it is tempting to speculate the following model. The ompS gene has a weak promoter sequence II id 276 and is weakly transcribed in the absence of OmpT (i.e. high sodium chloride levels). It is already established that IHF has the ability to bend DNA upon binding (Goodman and Nash, 198 9). The binding of OmpT (under low sodium chloride levels), results in the bending of ompS DNA, which brings together DNA sequences that were far apart prior to binding of OmpT. This conformational change in DNA better facilitates the binding of RNA polymerase to ompS promoter, resulting in mere efficient transcription of ompS. However it is worth mentioning that to date there are no reports indicating that IHF is directly responsive to environmental regulation. Also for a number of E. coli operons, IHF has been shown to inhibit transcription both in vivo and in vitro (Goosen and Van de Putte, 1995). In some of these cases, IHF inhibits transcription by directly binding to -10, -35 regions of the promoter and in other cases it inhibits transcription indirectly by influencing a regulatory protein (Goosen and Van de Putte, 1995). In the OmpF/OmpC system, IHF is known to inhibit transcription of both these porin proteins. For OmpF, IHF negatively regulates the gene by inhibiting the OmpR-mediated activation of the gene (Ramani et. al., 1992). The ompC gene 277 is transcribed by three promoters. Therefore in this case, IHF brings about negative ragulation by binding around position -180 and inhibiting transcription from two of the promoters PI and P3 (Huang et. al., 1990). Like in the OmpF/OmpC system, it is possible that an IHF-like homolog in L. pneumophila is involved indirectly in the regulation of ompS by interacting with OmpT. The cloning and sequencing of ompT as well as functional analysis of this protein will help to prove/disprove these highly speculative models. Also, if OmpT has any homology to IHF, it is possible that this protein plays a more global role in gene expression in L. pneumophila. The involvement of specific sigma factors in gene regulation has been somewhat of a novel concept. In S. typhimurium, the katF gene coding for a putative sigma factor RpoS has been shown to be required for virulence (Fang et. al., 1992). Mutants in the katF gene show reduced virulence in the mouse model (Fang et. al., 1992). Recent studies show evidence for the involvement of anti-sigma factors in gene regulation, which is even a more novel concept in gene regulation. As the name implies, anti-sigma i 278 factors are thought to bind to the cognate sigma factor and prevent it from associating with the core RNA polymerase (Brown and Hughes, 1995). In the process of controlling the action of sigma factors, the anti-sigma factors themselves are regulated by various environmental stimuli. Some of the examples for the involvement of anti-si^ma factors in gene regulation are, the development of the flagellar organelle in S. typhimurium, control of the general stress response in Bacillus subtilis, and cell-type specific gene expression in sporulation of B. subtilis (Brown and Hughes, 1995). Concluding Remarks: L. pneumophila is a facultative intracellular pathogen that lives naturally in aquatic environments. In its natural habitat, this bacterium is found either free living or more often in association with aquatic amoebae. Therefore this bacterium has an inherent ability to grow within a wide range of eukaryotic cells including protozoans and mammalian cells. With this ability, the bacterium must have also developed ways to sense its environment (free living as opposed to being intracellular; protozoans as opposed to mammalian cells) and regulate its gene expression I a I i . 279 accordingly. It is interesting that so far, there are no reports of avirulent mutants of L. pneumophila being isolated from the environment. Therefore, avirulent mutants appear to be a laboratory artifact. As mentioned before, the only phenotypic difference observed so far between the virulent and avirulent mutants of L. pneumophila is the avirulent mutant's ability to grow on media containing high levels of sodium chloride. The virulent strains cannot grow on media containing sodium chloride levels in excess of 0.65% (Catrenich and Johnson, 1989). So, is the sodium chloride tolerance directly related to the avirulence of these mutants? Work by Sadosky et. al. (1993) has revealed that the genes conferring salt sensitivity and macrophage killing ability are either closely linked or coordinately regulated. Therefore unravelling the basis of sodium chloride tolerance may shed, new light into the virulence regulation of L. pneumophila. The current study has revealed that the ompS gene is regulated by sodium chloride levels. This regulation appears to be indirect; sodium chloride affects the synthesis of a DNA binding protein, OmpT, which in turn regulates ompS. • - A 280 According to the results of this study, in high sodium chloride conditions, the synthesis of OmpT is repressed. OmpT, being a highly unstable protein, is degraded rapidly. In the absence of OmpT, expression of ompS is decreased. This seems like a logical strategy for this bacterium. High sodium chloride levels would be one of the initial environmental stimuli experienced by L. pneumophila upon invading the human lung, the bacterium having originated in a fresh water habitat. Once in the intracellular environment of alveolar macrophages, a highly cross-linked rigid outer membrane may not be an advantage to the bacterium. A report by Abu Kwaik et. al. (1993) has shown that there are at least 35 L. pneumophila genes which are selectively induced upon macrophage infection and at least 32 which are selectively repressed. It would be interesting to find out whether these genes also respond to changes in sodium chloride levels. If so, then sodium chloride and perhaps other as yet unidentified stimuli, are involved in the regulation of virulence gene expression in L. pneumophila. Avirulent mutants may have a defect in this I 281 regulatory pathway which hinders their ability to survive in intracellular environments. Immunological characterization: The immunological aspect of this study focused on the immune responses of humans and guinea pigs to OmpS and Hsp60 proteins of L. pneumophila. The data obtained revealed that guinea pigs surviving high dose challenges of L. pneumophila show strong DTH reactions against Hsp60 and OmpS at 6 weeks post infection. Furthermore, lymphocytes from these animals show strong proliferative responses to these antigens. Peripheral blood lymphocytes from humans surviving Legionnaires' disease also showed strong proliferative responses to OmpS and somewhat lower proliferative responses to Hsp60. This may suggest that certain epitopes of these antigens are involved in antigen presentation and activation of immune responses. Immunization of guinea pigs with OmpS showed dramatic protection against Legionnaires' disease as seen by reduced severity of disease and mortality. Although the OmpS- immunized animals were sick initially after challenge, they 282 recovered by day 3 and started to gain weight. When compared with the control animals, the protective response in OmpS immunized animals was also evident by other parameters such as the extent of pneumonia, number of bacteria per gram of lung tissue and percentage of lung consol '.dation. Animals immunized with Hsp60 showed very poor protection against L. pneumophila infections. Compared to OmpS-immunized animals, the Hsp60-immunized animals showed a higher degree of mortality as well as higher severity of disease, as judged by the severity of symptoms and the length of time taken for recovery. These results contradict a report by Blander and Horwitz (1993) in which they demonstrated that immunization of guinea pigs with Hsp60 offered protection against lethal infection by L. pneumophila. However, in this study 3 immunizations were required and the DTH responses reported were unequivocal. Humans surviving Legionnaires' disease also show a significant peripheral blood lymphocyte proliferation response to OmpS compared to healthy controls. Lymphocyte proliferation to Hsp60 was lower relative to OmpS, but still higher than the healthy controls. These results also I m a i 283 indicate that OmpS is more immunogenic in terms of inducing a cellular immune response than Hsp60. Individuals tested 8 years post recovery from legionellosis, showed a stronger lymphocyte proliferative response to OmpS. Two patients who were still at the convalescent stage showed a dramatic proliferative response to Hsp60 and a much lower response to OmpS. This may suggest that Hsp60 stimulates lymphocytes early on in infection whereas OmpS offer a more long term protection. However, additional studies are needed to clarify this. It was interesting to note that two heart transplant patients surviving Legionnaires' disease were capable of developing cellular immune responses to both OmpS and Hsp60 despite their immunosuppressed status due to Cyclosporin A therapy. The lymphocytes from these patients had to be incubated in the tissue culture medium RPMI 1640 for 24 h and washed twice with PBS prior to antigen challenge. Lymphocytes that were not treated accordingly snowed no proliferative response to any antigens including concanavalin A and poke weed mitogen. The incubation in RPMI 1640, and washing with PBS was therefore necessary to dilute I 284 out the inhibitory effects of Cyclosporin A. Infection is one of the serious complications following transplantation due to the immunosuppressed condition of the patients. Current literature indicate that Legionella infections have increased among heart transplant patients (Hofflin et. al., 1987). The results of the present study suggest the possibility that vaccination of potential transplant patients against Legionnaires' disease will aid in their long term survival. The Chlamydia MOMP has also been implicated as a protective antigen against chlamydial infections by a number of investigators (Lucero and Kuo, 1985; Peeling et. al., 1984; Zhang et. al., 1987). However, in this case, serovar- specific neutralizing antibodies are thought to play a critical role in the protective immune response. Some of the antigenic epitopes on MOMP of C. trachomatis have already been mapped (Villeneuve et. al., 1994). The available literature demonstrates that immunity to legionellosis is dependent on mounting a strong cellular immune response (Horwitz, 1983; Breiman and Horwitz, 1987). r i . I 285 Furthermore, it is believed that antibody responses to surface proteins of L. pneumophila facilitate the entry of the bacterium into host cells either by Fc receptor mediated endocytosis or by complement receptor mediated endocytosis (Husmann and Johnson, 1992; Bellinger-Kawahara and Horwitz, 1990). Therefore the primary defence against L. pneumophila infection may be the destruction of infected macrophages and thus control of the spread of the bacteria. In order to eradicate the infected macrophages, the host needs criteria by which to identify the infected macrophages from the uninfected. Several studies have demonstrated that virulent L. pneumophila abundantly synthesize Hsp60 upon internalization into host cells and that this protein accumulates in the phagosome of the host cell (Abu Kwaik et. al., 1993; Hoffman et. al., 1993; Fernandez et. al., unpublished data). Therefore Hsp60 could be an antigen that is presented early in infection in the context of both MHC I and MHC II and is recognized by both CD4+ and CD8+ cells. Natural killer cells which are thought to be MHC unrestricted in their antigen recognition also have been implicated to play a role in immune response to L. pneumophila infections (Blanchard et. al. , 1988). I 286 OmpS consists of disulphide cross linked subunits which are covalently linked to the peptidoglycan layer and are resistant to proteolytic degradation in its native state (Butler and Hoffman, 1990, Hoffman et. al., 1992). Therefore it likely that the eventual destruction of bacterial cells within the phagosome leads to the presentation of OmpS antigens, probably in the context of MHC II. Its abundance in the bacterial cell membrane, the hydrophobicity and the natural adjuvant activity due to the attached peptidoglycan fragments may play important roles in the immunogenicity of OmpS over other antigens. Although the results of this study demonstrate OmpS to be an effective antigen in eliciting a cellular immune response against Legionella infection, much work has to be done before declaring it an efficacious vaccine against Legionnaires' disease. The LPS of Gram negative bacteria is known to act as a simulator of mononuclear phagocyte and polyclonal activators of B cells. In higher concentrations, LPS is known to cause tissue injury, disseminated intravascular coagulation and shock frequently leading to death (Herman et. al., 1991). Although such complications I I i 287 were not seen in guinea pigs vaccinated with OmpS, direct preparations of OmpS from L. pneumophila would not be safe as vaccine preparations for humans. Therefore it will be necessary to express OmpS in a background devoid of LPS such as in a Gram positive genetic background. Furthermore, it would be necessary to identify the exact epitopes of OmpS involved in mounting the protective immune responses. Knowing the exact epitopes may even facilitate the use of synthetic peptides linked to appropriate carrier molecules as vaccines. Using synthetic polypeptides as vaccines has shown promising results in the attempt to formulate a vaccine against malaria (Braun, 1988; Good et.al, 1988; Playfair et.al., 1990; Good et.al., 1992). Therefore, future work on the immunological characterization of OmpS should focus on the identification and mapping of the linear amino acid epitopes involved in eliciting cellular immune responses. Also it would be advantageous to identify the subsets of T cells proliferating in response to OmpS which would provide a better understanding of the immune responses mounted against 288 this antigen and its role in inducing immunity against L. pneumophila infections. 1 APPENDIX M9 minimal medium: Per 1 liter of cooled sterile deionized water: 200 ml 5 X M9 salts 20% glucose 50 ml amino acid stock solution 5 X M9 satis is made by dissolving the following salts in deinonized water to a final volume of 1 liter: Na2HP04. 7H20 64 g KH2P04 15 g NaCl 2.5 g NH4C1 5.0 g The salt solution is devided into 200 ml aliquotes and sterilized by autoclaving for 15 min at 15 lb/sq. in. on liquid cycle. 289 I 290 Amino acid mixture: combine 0.1% stock of each of the amino acids, filter sterilized and added to the M9 medium at 5% volume. Iso leucine Valine Serine Glutamate Leucine Threonine Glutamine Aspartate 291 Triethylamine (TEA)-acetate Buffer: Add 0.2 moles of triethylamine dropwise to 100 ml of water containing 0.2 moles of acetic acid. The flask should be in an ice bath and constantly stirred. After TEA is added, dilute the solution to 200 ml and adjust the pH with TEA or acetic acid to 7.3. This stock is a 1 M solution that can be diluted to 10 mM in sterile water. I e 292 • Dalhousie Univer Department of Microbiology jfaEIME •^ and Immunology r Charles Tupper Medical Building AUG 0 9 1995 Halifax, Nova Scotia Canada B3H 4H7 (902) 494-3587 JOURNALS DEPARTMENT Fax (902) 494-5125 Ms. Linda Illig Director of Journals Journals Department, American Society for Microbiology, 1325, Massachusetts Ave. NW, Washington DC 20005-4171. August 1,1995. Dear Ms. Illig: I am writing to obtain permission to use the data that I have published in the following articles in the Journal of Bacteriology and Infection and Immunity. I wish to use these data for my Ph.D. theses. 1. Cloning and nucleotide sequence of a gene (ompS) encoding the major outer membrane protein of Legionella pneumophila. J. Bacteriol. (1992) 174:914-920. 2. Human and guinea pig immune responses to Legionella pneumophila protein antigens OmpS and Hsp60. Infect. Immun. (1994) 62:3454-3462. Thank you. Yours sincerely, PERMISSION GRANTED CONTINGENT ON AUTHOR PERMISSION AND APPROPRIATE CREDIT Risini Weeratna. American Society for Microbiology Journals Division I—i ^LiL~_ Data _J__Jll__ I The University of Chicago Press 293 5801 Ellis Avenue, Chit-ago, Illinois 60637-1496 ftione 312/702-6096 Since 1S91 Publishers of Scholarly Books and Joumali FAX 312/702 9756 Rights and Permissions August 11, 1995 Prof Risini Weeratna Dept of Microbiology and Immunology Dalhousie University Halifax, Nova Scotia CANADA B3H 4H7 re your 1 August letter Dear Prof Weeratna The University of Chicago Press is happy to grant you gratis permission to use portions of your essay, "Legionnaires Disease in Cardiac Transplant Patients, from our publication, JOURNAL OF INFECTIOUS DISEASES, in your forthcoming PhD thesis as described in your letter Please give full credit to our publication and to the University of Chicago Press as publisher The acknowledgment should also include the identical copyright notice as it appears in our publication Our reco~ds indicate the correct copyright notice for this material is ® 1993 by The Univetsity of Chicago All rights reserved Best wishes1 Perry'Cartwnght Permissions Editor 294 JOURNAL OF BACTERIOLOGY, Feb. 1992, p. 914-920 Vol. 174, No. 3 0021-9193/92/030914-07S02.00/0 Copyright © 1992, American Society for Microbiology Cloning and Nucleotide Sequence of a Gene (ompS) Encoding the Major Outer Membrane Protein of Legionella pneumophila PAUL S. HOFFMAN,12* MURRAY RIPLEY,1 AND RISINI WEERATNA1'2 Department of Microbiology1 and Division of Infectious Diseases, Department of Medicine,2 Dalhousie University, Halifax, Nova Scotia B3H 4H7, Canada Received 13 August 1991/Accepted 21 November 1991 The major outer membrane protein of Legionella pneumophila is composed of 28- and 31-kDa subunits cross-linked by interchain disulfide bonds. The oligomer is covalently anchored to the underlying peptidoglycan via the 31-kDa subunit. We have cloned the structural gene ompS encoding both proteins. Oligonucleotide probes synthesized from the codons of the N-terminal amino acid sequence of purified 28- and 31-kDa subunits were used to identify cloned sequences. A 2.9-kb Hindlll fragment cloned into pBluescript (clone H1S1) contained the ompS gene. Nucleotide sequence analysis revealed an open reading frame of 891 bp encoding a polypeptide of 297 amino acids. A leader sequence of 21 amino acids was identified, and the mature protein contained 276 amino acids. The deduced amino acid sequence of OmpS matched the experimentally determined amino acid sequence (32 amino acids), with the exception of two cysteine residues. The deduced amino acid sequence was rich in glycine and aromatic amino acids and contained four cysteine residues, two in the amino terminus and two in the carboxy region. Primer extension analysis (total RNA from L. pneumophila) identified the transcription start at 96 to 98 bp upstream of the translation start, but no Escherichia co/i'-like promoter sequences were evident. While an mRNA transcript from clone H151 was detected, no cross-reactive protein was detected by immunoblotting with either monoclonal or polyclonal antibody. Attempts to subclone the gene in the absence of the putative promoter region (i.e., under the control of the lac promoter) proved unsuccessful, possibly because of overproduction lethality in E. coli. The ompS DNA sequence was highly conserved among the serogroups of L. pneumophila, and related species also exhibited homology in Southern blot analysis at a moderately high stringency. Evidence is presented to suggest that this gene may be environmentally regulated in L. pneumophila. Legionella pneumophila and related species are a diverse reports suggest that this antigen may be protective against group of environmental microorganisms capable of produc lethal challenge in the guinea pig model (48). ing severe lobar pneumonia in humans. These facultative The MOMP is novel in that one of the «-dbunits is cova intracellular parasites can invade and colonize a wide range lently bound to peptidoglycan via a peptide bond, probably of eukaryotic host cells, including aquatic amoebae (40, 49, to diaminopimelic acid (5). This subunit is covalently cross- 51), alveolar macrophages (28). and vertebrate cell lines (13, linked to 28-kDa subunits via interchain disulfide bonds (5, 33,36). The bacteria reside in the phagosomes, in which they 27) Progress in assessing the role of this protein in patho abrogate phagolysosomal fusion, a process that does not genesis as well as in characterizing the novel structure of this require de novo protein synthesis (29). These observations putative porin has been hindered by an inability to clone the suggest that preexisting surface proteins may participate in structural gene. A number of laboratories, including our the pathogenesis process One surface protein, named Mip laboratory, have been unsuccessful in cloning this gene in (macrophage invasion potentiator), is a 24-kDa protein ex hibiting a pi of 9.3 (14). Null mutations in mip result in expression vectors The expression of cloned porin genes, attenuated virulence for macrophage cell lines and for guinea particularly from nonentenc pathogens, in Escherichia coli pigs (10,11). The most abundant surface protein, referred to is often inhibitory for growth (2, 19). In some cases, gene as the major outer membrane protein (MOMP), is a porin fusions in vectors such as lambda gtll have been used to composed of subunits of 28 and 31 kDa (5, 6, 17, 20, 21). identify porin sequences in expression libraries (19, 45) Studies by Payne and Horwitz (37) have demonstrated that Such approaches have been used to clone the 40-kDa MOMP the MOMP ponn. binds the C3b and C3bi factors of the gene (ompIL2) from Chlamydia trachomatis (46) complement system, which mediate phagocytosis via the Our approach to cloning the ompS gene involved the use macrophage integnn receptors CRl and CR3 Further, com of oligonucleotide probes generated from amino acid se plement enhances the binding of MOMP-contaming lipo quence information from the 28- and 31-kDa MOMP sub- somes to macrophages (3). Studies by Quinn et al. (38) also units. We found that both subunits had a common N-termi suggest that the MOMP may be directly involved in attach nal amino acid sequence (27) On the basis of this sequence ment in a HeLa cell model and also may be a protective (GTMGPVWT), a series of oligonucleotides were generated, immunogen in the guinea pig model. While human and one of which hybridized to a single site of restricted genomic guinea pig convalescent-phase sera contain little or no DNA, as judged by Southern blot analysis. In the present antibody to the MOMP (42), guinea pigs mount a significant study, we report on the use of this oligonucleotide as a probe cellular immune response to this antigen (23). Preliminary for identifying clones containing the putative genes coding for the L. pneumophila MOMP subunits. On the basis of the nucleotide sequence and the deduced amino acid sequence, we have confirmed that an open reading frame (ORF) of 891 : Corresponding author. bp encodes a polypeptide of 297 amino acids The gene is 914 295 VOL 174 1992 L PNEUMOPHILA MOMP GENE 915 poorly expressed under its own promoter in £ coli and may G-50 as descnbed by Silhavy et al (43) Approximately 8 x be down-regulated during the early stages of invasion 106 cpm was added to the hybridization mixture following (A preliminary account of this work was presented at the denaturation Hybridizations and washings were done at a 1991 American Society for Microbiology general meeting moderately high stnngency (15% mismatch in the duplex) [26]) and at a high stnngency (<5% mismatch in the duplex) as descnbed previously (24) MATERIALS AND METHODS Total RNA was prepared by a hot sodium dodecyl sulfate- phenol procedure (27) Pnmer extension was performed by Bacterial strains, protein sequence, and oligonucleotide hybndizmg probe H25 to approximately 25 p.g of total RNA reagents. A streptomycin-resistant strain (SVir) of L pneu The oligonucleotide probe was end labeled with (32P]ATP mophila Philadelphia 1 (serogroup 1) was used as the source (125 (iCi, NEN-DuPont) and T4 polynucleotide kinase (New of the 28- and 31-kDa proteins and chromosomal DNA as England BioLabs, Inc , Beverly, Mass ) as descnbed previ descnbed elsewhere (24, 27) Other serogroups of L pneu ously (27) The probes were used either directly or after gel mophila and various Legionella species used in this study punfication on a 20% polyacrylamide-8 M urea gel The included L pneumophila serogroups 1, 2, 3, 4 5 and 7, L extension reaction was run at «2°C for 1 h in the presence of micdadei (HEBA and Tatlock), L jordanis serogroup 1 and actinomvcm D Following RNase treatment phenol extrac L oakridgensis ORIO Preparation of genomic DNA from tion and ethanol precipitation the cDNA was taken up in these strains has been descnbed previously (24) Purification loading buffer and 2 to 5 |-1 was loaded onto the sequencing of the 31 and 28 kDa proteins peptide maps and peptide gel The extension product was compared with the simulta sequencing were as descnbed elsewhere (5 27) Ohgonucle neously run sequencing reaction products generated from otides used in this study were synthesized at the Molecular M13 containing a Hindlll-Pstl fragment encoding the 5' Resource Center Department of Microbiology and Immu region of the ompS gene and H25 as a pnmer nology University of Tennessee Memphis, and at the Nucleotide sequence accession number. The nucleotide Molecular Gene Probe Laboratory, Dalhousie University sequence accession number (GenBank) for ompS is M76178 Two oligonucleotides (24b and H25 [complement]) used as probes in this studv had the following respective sequences 5 GGTACTATGGGTCCAGTATGGAC3 and5'GTCCATA RESULTS CTGGACCCATAGTACC3 These sequences were based Cloning strategj. We have assumed, on the basis of on the amino acid sequence GTMGPVWT vanous strategies aimed at cloning the MOMP gene into Cloning and nucleotide sequencing of the ompS gene. Chro expression vectors either that the expression of the gene mosomal DNA from L pneumophila restncted with EcoRI was toxic or that the gene was not expressed in the E coli tfindlll, or a combination of tfmdIII and Pstl was subjected genetic background By use of a DNA hybridisation dp to electrophoresis in an 0 8% low-melting-temperature aga proach the requirement for gene expression could be ehm rose gel (SeaPlaque FMC Bioproducts Rockland Maine) mated and in the case of toxic gene expression unexpressed DNA fragments in molecular weight ranges previously iden fragments of the desired gene might be cloned In a previous tified by Southern blotting were excised and ligated into study, we reported that an oligonucleotide synthesized from pBluescnpt (Stratagene), similarly restncted, and punfied the N-terminal ammo acid sequence GTMGPVWT hybnd from low-melting-temperature agarose The resulting colo ized to single sites in genomic DNA restncted with vanous mes were screened with [32P]ATP end labeled oligonucleo restnction enzymes (27) Since Southern hybndization anal tide H25 (reverse pnmer) as generally described by Sam ysis provided information on the relative sizes of the desired brook et al (41) Restnction enzymes, buffers, and restnction fragments we first attempted to clone these sized procedures were as descnbed by the vanous manufacturers fragments into pBluescnpt We were unsuccessful in obtain Restncted DNAs from clones H151 (2 9-kb Hindlll frag ing clones of a 1 5-kb EcoRI fragment in either pBluescnpt ment) and HP246 (900 bp Hindlll-Pstl fragment) were sub or M13 phage The fact that recombinants were obtained cloned into M13 vectors for sequencing Sequencing was (white versus blue colonies or plaques scored with isopro performed by the dideoxv chain termination method with pyl-p-D-thiogalactopyranoside and 5 bromo 4-chloro 3 in [a5S]dATP (NEN-DuPont Toronto Ontano Canada) and dolyl (3 D galactopyranoside) suggested no technical diffi Sequenase (U S Biochemical Corp Cleveland, Ohio) as culty We then screened clones containing DNA fragments suggested by the manufacturers Both DNA strands were obtained by Hindlll cleavage and by a combination of sequenced, and the sequences were assembled and analyzed Hindlll and Pstl restnction Four colonies from a total of with the Wisconsin Genetics Computer Group sequence 280 white colonies were found strongly positive by Southern analysis programs (Genetics Computer Group, Inc Univer colony blot hybndization Restriction endonuclease analysis sity of Wisconsin Biotechnology Center Madison) (12) of cloned sequences showed that two clones contained a 2 9-kb Hindlll fragment (H151 and H157) and that two Southern and Northern (RNA) blot analyses and primer contained a 900 bp ffmdlll Pstl DNA fragment (HP243 and extension. DNA hybndizations were done essentially as HP246) Restriction maps for H151 and HP246 are presented descnbed by Southern (44) Chromosomal DNAs from the in Fig 1 Interestingly, the 1 5 kb EcoRI fragment was vanous Legionella serogroups and species were restncted contained within the 2 9-kb Hindlll fragment Attempts to with EcoRI or Hindlll The DNA probe was radiolabeled by subclone the EcoRI fragment proved unsuccessful suggest the polymerase chain reaction (PCR) Oligonucleotide pnm ing that an ORF designated ompS, may be within the EcoRI ers 24b and R3 (reverse pnmer hybndizmg to an internal restnction fragment and that the expression of this gene may region of the ORF) were used to amplify an approximately 32 produce a toxic product in E coli While not shown a 500-bp segment of ompS [ P]dCTP (65 u,G in a 50-u.l screening of the E coli clones for antigen expression with reaction mixture) was incorporated into the PCR fragments both monoclonal and polyclonal antibodies reactive with by decreasing the earner dCTP molar concentration in the both the 28 and the 31 kDa subunits was also negative The deoxynucleotide tnphosphate reaction mixture from 200 to possibility that a transcnptional defect might account for the 50 u,M The PCR fragments were punfied through Sephadex 296 916 HOFFMAt' ET AL. J. BACTERIOL. E P 700 I I am H151 20B HP24B FIG. 1. Restriction endonuclease maps of H151 and HP246 in pBluescript. A 2.9-kb Hindlll fragment was cloned into pBluescript. The ORF was localized to the 1.5-kb EcoRI region. Abbreviations: H, Hindlll: P, Pstl; E, EcoRI. The map distances are given in base pairs. The lac promoter is in the left arm of pBluescript and would permit the transcription of ompS. lack of expression was examined by Northern blot analysis most likely due to the poor quality of the amino acid of clone H151. A low level of transcription was observed for sequence obtained for this peptide. Downstream of the H151 probed with an internal PCR-generated DNA fragment termination codon (TAA) is a palindromic sequence resem of the ompS gene (Fig. 2). The molecular size of the H151 bling a rho-independent transcription terminator. An obvi mRNA was similar to that detected in total RNA prepara ous E. co//-like promoter sequence was not identified up tions from L. pneumophila SVir and from an isogenic stream of the translation start. The possibility that the gene protease-deficient strain, PRT8. These results suggest that was pan of an operon was resolved by primer extension the ompS promoter regie"•- is poorly recognized in E. coli and analysis. Primer extension revealed that transcription begins that perhaps a translatici.al defect might also account for a within a CCC sequence (bp 172 to .174 in Fig. 3) that is lack of detectable MOMP. flanked on both sides by an AT-rich sequence and that is 90 Sequence analysis. The nucleotide sequence of structural to 100 bp upstream of the translation start (Fig. 4). gene ompS and flanking sequences and the deduced amino In a previous study, we predicted that the MOMP con acid sequence are depicted in Fig. 3. The ORF begins tained three cysteine residues on the basis of amino acid approximately 800 bp in from the leftward Mndlll site (Fig. composition analysis (5). On the basis of the deduced amino 1) and 270 bp into the depicted sequence. A ribosome acid sequence, four cysteine residues were identified. Two binding domain (TGGAG) is located 9 bases upstream from of these residues are located in the N-terminal region (amino the initiation codon. An 891-bp ORF encodes a polypeptide acids 7 and 16 of the processed polypeptide), and two are in of 297 amino acids, of which the first 21 amino acids are the carboxyl region at amino acids 194 and 197. The poly presumably involved in protein export, since the processed peptide also contains seven methionine residues rather than protein begins at amino acid 22 (PheAla/GlyThrMet . . .) three, as previously reported on the basis of the amino acid (slash indicates processing site). With the exception of two composition analysis. The protein is rich in glycine (11%) cysteine residues, there was good agreement between the and the aromatic amino acids phenylalanine (7.4%), tyrosine deduced and experimentally determined amino acid se (6%), and tryptophan (3.4%). The processed polypeptide quences (32 amino acids), as depicted by the underlined exhibits an acidic pi of 4.59, and the amino acid composi N-terminal region of the sequence in Fig. 3. The N-terminal tion, hydrophilicity characteristics, and a lack of long amino acid sequence of a 19-kDa peptide generated by stretches of alpha- or beta-sheet sequences are typical of cyanogen bromide cleavage of the 31-kDa protein was also porin proteins. A sc'ch of the protein data base revealed located in the deduced sequence. However, the differences that OmpS was not closely related to other porin proteins, noted between this sequence and the deduced sequences are with the exception of a 69-kDa outer membrane protein of Bordetella pertussis which exhibited a similarity of 45% and an identity of 21% (8). The sequence also exhibited 22% A B C D identity with a repeat in the CRl or CR3 integrin, a comple ment receptor of macrophages (30). Distribution of ompS in legionellae. Previous studies in our laboratoiy l?'e demonstrated that several Legionella spe cies express disulfide-cross-linked and peptidoglycan-bound outer membrane proteins (5). Furthermore, on the basis of immunoblot studies with polyclonal anti-Eeg/one/Za MOMP serum, we reported that the MOMPs of the various species may share common epitopes (6). To address the possibility that the genes encoding these proteins from the various •I species might be genetically related, we probed chromo somal digests from selected Legionella species and sero groups with an internal PCR-radiolabeled fragment of the ompS gene. Figure 5A depicts the results of Southern FIG. 2. Northern blot analysis of H151 and L. pneumophila hybridization analysis performed at a moderate stringency ompS mRNAs. Total RNA was extracted as described in the text (15% mismatch in the duplex). At this stringency, all Legion from E. coli HB101 (pBluescript) (lane A), E. coli HB101 (clone ella species examined exhibited related sequences. The H151) (lane B), L. pneumophila SVir (lane C),'and L. pneumophila -akest signal was noted for L. jordanis, which does not PRT8 (protease-deficient avirulent mutant of SVir) (lane D). The mRNA was probed with a PCR-generated "P-labeled DNA frag press a 28-kDa MOMP (5). The strongest signals were ment generated to an internal region of the ompS gene (oligonucle - jerved for the serogroups of L. pneumophila. However, otide primers 24b and R3), The hybridization noted in lane B at a multiple bands were seen with the serogroups, suggesting high molecular weight probably represents contaminating plasmid that there might be either related genes encoding outer sequences of H151. membrane proteins or possibly cryptic genes. At higher w 297 VOL 174, 1992 L PNEUMOPHILA MOMP GENE 917 1 CGAA^AAATmGGCGCAMTTAATAOTCATGCACrCCCCCTCATnCTACAAAATTAAAGCAAAAAGCAGACTACTTGC BO 81 TGGCGTArAAATCAATAACAGCACATACTCArCMATCTrACATTTAATGTTTAMTCAATGAGTTAAArAACTTTAAl T 160 161 TAATAAAATTACCCTTATTATTTGATGAAGAATCAAATACATTC-CTrAGAATTCCCTAATCCnGATCTTTAAACSAAT 210 >aJUU 24] AACAATAArAAArCAGICEA£Ai« GGGATATGTTTAGTTTGAAAAAAACAACTGCGGCTGTATTTGCTCTCGCAAGCAGr 320 MeLPheSefLeuLysty sThrrhrAl Ml aValPheAl a LeuGlyScrSer 321 GCTTTOmCCAGGTACGATCGGTCCAGTATGTACGCCAGGCAATGTGACTGTTCCTTGCGAMGAACCGCATGGGATAr 100 AlJi..»Phi-A;JGlMTt,--'r.f; yPrru,.lfir."--•"•«,!vtsnu..iT>irV.iIP.egv.nlHmT'-M aTrnAanll. N H G A 401 TGGTATTACCGCACTTTATCTGCAGCCAACTTATCATCCTGATTCGGGCTATAAIGGTTTTACIGATGTTG&tGuCTGGA 480 >nivrllThr»l:Hi.iHVrli. iGlnf rnThrTvrA«pAlaA«nTrnGI vTvrAsnGlVPheThrAlpVa IGlvCWTrpA 481 GAAATtGGCATGATGtTGATTTAGAGIGGGATTGGGGCTTTAAArTAGAAGGTTCTTATCACTTCAACACAGGCAAlGAC 560 rqAsnTrpHUAspValAspleuGlurrpAapTrpGlyPheLysLeuGluGlySerTyrHisPheAsnThrGlyAsnAsp 561 ATCAACGTCAATTGGTACCATTTCGATGCCAACACCGATCATTGGGCGTTTGTTGATCAATTGCACGCCTACGATATTAC 640 neAsnValAsnTrpTyrHlsCeuAspAUAsnThrAspHlsTrpAlaPheValAspGlnLeuHlsAUTyrAspIleTh 641 CTGGGATGCAGTGAATGCTGAACTGGGGCAATTTGTTCATTTTAGCGCCAACAAAAAAATGCCCTTCCATGGCGGCGTTC 72n rTrpA3pAlaValAsnAUGluLeuGlyGlnPheValAspphe5erAlaAsnLysl.ysHetArrjPheHisGlyG]yValG 721 AGTATGCCIACATCAAAACAGACGTTAATCGi'CACTTAAACGGCTrCTTTCTAGATAACTTCAATTCCAAATTCAATGGA S00 InTyrAiATyrTleLyaThr^pVal AsnArnHl y, r lAsrirl vPl-i.pHfl fntnpA*r.P^pJ> ^?pf v.Ph.t.hr.ly V F A F N B01 TTTGGtCCCCGTACCGGCCTTGATATGAATIATGTATICGGTAAIGGGTTTGGTATTTATGCAAAATCAGCTGTAGCAAr 880 PheGlyProAi:qThrGlyLeuAapMetA5nIyr«alpheGlyA5nGlyPheGlyIleTyrAlaI.y5Sei:AlaValAlaIl 881 CCTGGTCGGTACAAGCAAATTCGTGGATAACTGIACTGTCTG-GGTTTTTCTTATGGCTCCAAAAArGCAATAGTGCCAG 960 •leuVjlGlyThrSerLyaPheValAipAanCyiThtValCy.GlyPheSerryrGlySerLysAsnAlalleValProG 961 AAGITGAAATGAAGTTAGGAGCAGATTATACCTATGCAATGGCTCAAGGTGATGTGACTCTGGATGTGGGTTACATGTCG 1040 luValGluMetLysLeuGlyAlaAapTyrThtTycAlaHetAUGlnGlyAspValThtLeuAspValGlvTycHetTrp 1041 TTCAATTACTTCAATGCCTTACATAATACTGCTGCCGTGAATGTTGCCTTACGCACTTCTCTCGAAACCGACrrCrcTGC 1120 PhcAapTyrPheAsnAlaLeuHlsAsnThrAUAiaValAsnValGlyUuGlyTh'-SerLeuGluTbcAspPheSerAl 1121 ATCTGGCCCCTACATAGGCTTAAAAIATGTrGGCAAIGTCTAMIAACCCCAAATTATGCTGCTAATCIAGTCATGATCA 1200 aSerGlyProTyrllsGlyUuLysTyrValGlyAsnValEnd 1201 AGCAGCAGIATGCAAAAAACCCGTATTGACATAGCCTGCTAATACGGGTTTTTTTTATTTCTCICTGTTTAATTAAG-T> K A 12B0 12B1 GTTATCCTATCTAAGCTAACTTGTAATTATATGATAATTTCCC -ATGAAATTGAGATAACTTTATGCGAAACAATAATTG 1360 1361 GTCCGTGTTGOGCAACATCACCAATGAAAAA-CACATCCGTG TCTGC-GCGAAAACTATAGTAAACGCAATAAICTGCT 1440 1441 GGTCTGCCTGAtGrATAAITCTTTTCCATTCCAAAGATCG GGGTTATAATGAAACCATCTATCACAAACCGTTTATAA 1520 1521 AAATAAACATGCCrGATGTTCTCGCCAAACCTGATGCAACCCAGATTCCAGATGACGTCTGGCTTGACATTCTTATCAAI 1600 1601 TATCTTCAACCACAAGATATTGIAAACCTGTCTGAAATCAACAAICGCTTGAGCAGACTTTTCAAAACTCAACACACTCA 1680 1681 ATCTGAGCAATTGAACCAAAAAAAAAA FIG 3 Nucleotide sequence and derived ammo acid sequence of ompS The transcription start at bp 173 is indicated bv the arrow and the termination sequence is indicated by the inverted arrows in the region from positions 1201 to 1280 of the sequence The putative nbosome binding site upstream of the translation start is underlined as are the regions of the deduced amino acid sequence determined by sequencing of the 31- and 28-kDa proteins (27) Differences from peptide sequencing are indicated by the single letter formal The positions of the 4 cysteine residues are depicted in boldface type stnngencies tor genomic DNA restncted with Hindlll (5% nity (23, 48). DNA sequence analysis revealed an 891-bp mismatch in the duplex), the hybndizations noted with the ORF encoding a polypeptide of 297 ammo acids. The poly- other Legionella species disappeared (Fig 5B), along With peptide contained a 21-aimno-acid signal sequence, and the the high signal intensity of the multiple hybndizations noted mature protein contained 276 amino acids. The deduced for the serogroups Southern blot analysis of L pneumophila ammo acid sequence of the MOMP exhibited four cysteine serogroup 1 DNA restncted with Hindlll and EcoRI identi- residues and in general was nch in glycine and aromatic fied a single fragment of 1 5 kb, confirming that ompS exists amlno acids with tne excepti0n of cysteine residues at in a single copy (data not shown) positions 7 and 16 of the processed polypeptide, there was good correlation between the deduced amino acid sequence DISCUSSION and that obtained by direct sequencing of the purified We have cloned and sequenced the structural gene encod- M0MP subunits (27) In this study, we confirmed the results ing the 28- and 31-kDa subunits of the L pneumophila of earlier work regarding the number of cysteine residues in MOMP oligomer. The MOMP is the most abundant protein the monomers Two cysteine residues were found in the synthesized by L pneumophila, and recent studies have amino terminus while two were found in the carboxyl region implicated this protein in pathogenesis (3, 37. 38) and immu- of the molecule A search of nucleic ar.id and protein data i 298 918 HOFFMAN ET AL. J. BACTERIOL, A C G T A 1 2 3 4 5 6 B 1 2 3 4 5 6 •5 4 •-•-•• - - mm 7 8 9 10 11 12 7 8 9 10 11 12 T A A A A T T A C C C FIG. 5. Southern blot analysis of EcoRI (A)- and tfmdlll (B)- T restncted DNA from selected Legionella species and serogroups of T L. pneumophila Genomic DNA was prepared from L pneumophila A serogroup 7 (lane 1), L. micdadei HEBA (lane 2), L. pneumophila T serogroup 1 SVir (lane 3), L. jordanis serogroup 1 (lane 4), L T A pneumophila serogroup 4 (lane 5), L. pneumophila (unknown sero T group) (lane 6), L. pneumophila serogroup 1 Avir (lane 7), L T pneumophila serogroup 2 Togus-1 (lane 8), L. pneumophila sero - T group 5 (lane 9), L micdadei Tatlock (lane 10), L. oakridgensis ORIO (lane 11). and L. pneumophila serogroup 3 (lane 12) In panel A. [he hybndization was done at a decreased stnngency (15% mismatch in the DNA duplex) to detect relatedness among the different species The stnngency conditions for the hybndization depicled in panel B was at a 5% mismatch in the DNA duplex The DNA probe used in these expenments was generated by PCR with oligonucleotides 24b and R3 as pnmers, and the amplified PCR fragment was radiolabeled as descnbed in the text. FIG 4 Pnmer extension analvsis of the transcription start of confirmed that all of the serogroups examined in this study ompS Total RNA was isolated from cells grown in buffered yeast contained highly conserved sequences, although some re extract medium for 24 h RNA (25 p.g) was annealed to oligonucle 3: striction polymorphism was evident. Under moderate-strin otide H25 which had been end labeled with [ P]ATP. The pnmer gency conditions, DNA-DNA hybndizations were also was extended with 50 U of avian myeloblastosis virus reverse noted with several other Legionella species. A number of transcnptase. and the cDNA was suspended in sample buffer and resolved on a polyacrylamide gel beside a sequencing ladder gener other genes which contain genus-common sequences have ated from the same oligonucleotide pnmer with a Hindlll-Pstl DNA been described for the legionellae, these include mip (14), fragment in M13 as the template The transcription start is located in htpAB (25, 38), and a gene encoding a 19-kDa peptidoglycan- a CCC sequence flanked by AT-rich sequences. associated protein (15, 22. 32). In contrast, the gene encod ing the cytotoxic metalloprotease is common only to the species L pneumophila (39) Since the MOMP is the most abundant protein synthesized bases revealed that no other proteins showed amino acid by L. pneumophila, information regarding the expression of sequence homology with OmpS. There was no homology the gene is of particular interest. Although primer extension with the Chlamydia sp. 40-kDa MOMP, which contains nine analysis identified the transcription start at 97 bp upstream of cysteine residues and participates in inter- and intramolecu the translation start, no E. co/i-like promoter sequences lar disulfide bonding (46). Interestingly, some homology was were seen in regions 10 (TAATAAAAT) and 35 (TCAAT noted with the 69-kDa outer membrane protein of B. pertus GAG) bp upstream These promoter sequences differ from sis and with the macrophage integnn protein CRl. a receptor those noted for other cloned L. pneumophila genes ex for complement (3, 37). The homology with CRl appeared to pressed in E. coli, including mip (14), htpAB (25), the be in a vanable-repeat region (34). While the observed protease gene (4), and recA (52). An unusual tandem pro similarities might be merely coincidence, it should be noted moter sequence has been reported for the omplL2 gene of C. here that while C3 components of complement bind to the L. trachomatis (47). The omplLl gene is developmental^ pneumophila MOMP, the specific binding site has not been regulated, and the MOMP is the most abundant protein identified. synthesized by C. trachomatis. In contrast, Northern blot Disulfide-cross-linked and peptidoglycan-bound outer analysis of L. pneumophila RNA revealed a single transcnpt membrane proteins are a charactenstic feature of many of approximately 1 kb (27), confirming a single promoter and members of the genus Legionella (5). A commercial mono a monocystromc gene structure. Interestingly, secondary clonal antibody diagnostic test for legionellosis recognizes structure predictions for the mRNA showed a substantial an epitope on the MOMPs shared by all serogroups of L. ability of the 5' region to form loops, which may function to pneumophila (18). Moreover, a study by Butler et al. (6) stabilize the message in L. pneumophila but possibly affect suggested that the MOMPs of other Legionella species might translation in E. coli (9) The ompS gene, like the chlamydial also contain genus-common epitopes Southern blot analysis omplLl gene, may be environmentally regulated. We have • i 299 VOL. 174, 1992 L. PNEUMOPHILA MOMP GENE 919 observed that little radiolabel is .iicorporated into the 28-kDa might lead to new insight into assembly processes external to MOMP subunit of virulent L. pneumophila cells during the the cytoplasmic membrane. It will be important in future early stages of invasion of HeLa cell or L-cell monolayers (1 studies to address the possibility that conformational to 3 h postinfection) (16). In contrast, radiolabel is incorpo changes, perhaps mediated by the mechanisms described rated into stress proteins, a phenomenon that can also be here, might be partially responsible for the phenomenon of reproduced in tissue culture medium in the absence of host avirulence acquired through the selection for high salt toler cells (25). When mRNA levels were monitored, the ompS ance under laboratory conditions. message was decreased relative to the htpAB message. The observation that avirulent isogenic strains of I. pneumophila ACKNOWLEDGMENTS do not regulate MOMP levels or show decreased mRNA levels when cells are placed in tissue culture medium implies This work was supported by Public Health Service grant AI24279 that these cells no longer sense changes in the environment from the National Institute of Allergy and Infectious Diseases to or perhaps no longer produce the necessary regulatory P.S.H., by grants from the Faculty of Medicine and Department of Medicine, Dalhousie University, and by a gram from the Medical factors. Environmentally regulated transcription has been Research Council of Canada. well characterized for ompF-ompC porin genes of E. coli (34) and for the vir {bvg) genes encoding many of the REFERENCES membrane-associated proteins of B, pertussis (1). We are 1. Ariro, B., J. F. Miller, C. Roy, S. Stibitz, D. Monack, S. Falkow. presently constructing gene fusions to begin addressing the R. Gross, and R. Rappuoli. 1989. Sequences required for expres regulatory aspects of ompS gene expression as well as sion of Bordetella pertussis virulence factors share homology putative regulatory differences between virulent and aviru with prokaryotic signal transduction proteins. Proc. Natl. Acad. lent isogenic strains. Sci. USA 86:6671-6675. The L. pneumophila MOMP (putative porin) (17) is unique 2. Barlow, A. K., J. E. Heckles, and I. N. Clarke. 1987. Molecular from the porins of other bacteria in that it is covalently cloning and expression of Neisseria meningitidis class 1 outer bound to peptidoglycan and the subunits are cross-linked via membrane protein in Escherichia coli K-12. Infect. Immun. interchain disulfide bonds (5, 27). On the basis of the 55:2734-2740. 3. Bellinger-Kawahara, C, and M. A. Horwitz. 1990. Complement deduced amino acid sequence, we found two cysteine resi component C3 fixes selectively to the major outer membrane dues in the N-terminal amino acid sequence (positions 7 and protein (MOMP) of Legionella pneumophila and mediates 16 of the processed polypeptide) and two at positions 194 phagocytosis of liposome-MOMP complexes by human mono and 197 in the carboxyl region. The first two cysteine cytes. J. Exp. Med. 172:1201-1210. residues are found in an amino acid sequence of high 4. Black, W. J., F. D. Quinn, and L. S. Tompkins. 1990. Legionella (5-sheet-forming potential, suggesting that at least one of the pneumophila zinc metalloprotease is structurally and function cysteine residues may be within the outer membrane lipid ally homologous to Pseudomonas aeruginosa elasiase. J. Bac- bilayer, while Cie other might be either external or near the tenol. 1.72:2608-2613. outer surface. These cysteine residues are separated by an 5. Butler, C. A., and P. S. HofTman. 1990. Characterization of a essentially hydrophobic stretch of eight amino acids, and the major 31-kilodalton peptidoglycan-bound protein of Legionella pneumophila. J. Bacteriol. 172:2401-2407. cysteine residue at position 16 is followed by the charged 6. Butler, C. A., E. D. Street, T. P. Hatch, and P. S. Hoffman. 1985. amino acids glutamate and arginine. Cysteine residue 194 is Disulfide-bonded outer membrane proteins in the genus Legion preceded by the charged amino acids aspartate and aspara- ella. Infect. Immun. 48:14-18. gine and separated from cysteine residue 197 by two hydro 7. Catrenich, C. E., and W. Johnson. 1989. Characterization of the phobic amino acids. While all cysteine residues could poten selective inhibition of growth of virulent Legionella pneumo tially participate in interchain disulfide bonding among phila bv supplemented Mueller-Hinion medium. Infect. Immun. subunits. the minimum number of bonds that each subunit 57:1862-1864. could participate in and still maintain the trimeric form 8. Charles, I. G., G. Dougan, D. Pickard, S. Chatfleld, M. Smith, P. would be two. It is conceivable that subtle changes in the Novotny, P. Morrissey, and N. F. Fairweather. 1989. Molecular cloning and characterization of protective outer membrane conformation of the putative porin might result from dif proiein P.69 from Bordeiella pertussis. Proc. Natl. Acad. Sci. ferent combinations of disulfide bonding among the cysteine USA 86:3554-3558. residues. Such combinations might also be envisioned to 9. Chen, L.-H., S. A. Emory, A. L. Bricker, P. Bouvet, and J. G. change the pore size or other physical characteristics of the Belasco. 1991. Structure and function of a bacterial mRNA molecule. Variable pore sizes noted for Pseudomonas stabilizer: analysis of the 5' untranslated region of ompA aeruginosa OprF have been attributed to different patterns mRNA. J. Bacteriol. 173:4578-4586. of disulfide bond cross-linkages (35). A prominent difference 10. Cianciotto, N. P., B. 1. Eisenstein, C. H. Mody, and IN. C. between virulent and avirulent isogenic strains of L. pneu Engleberg. 1990. A mutation in the mip gene results in an mophila is in the tolerance of sodium chloride (7). While the attenuation of Legionella pneumophila virulence. J. Infect. Dis. porin molecule has been shown to exhibit an anion prefer 162:121-126. ence (17), changes in conformation could affect the magni 11. Cianciotto, N. P., B. I. Eisenstein, C. H. Mody, G. B. Toews, and N, C. Engleberg. 1989. A Legionella pneumophila gene encod tude of this selectivity. It is perhaps noteworthy that the ing a species-specific surface protein potentiates initiation of amino-terminal cysteine residues are in a region substan intracellular infection. Infect. Immun. 57:1255-1262. tially rich in proline. Recently, the Mip protein has been 12. Devereiix, J., P. Haberli, and O. Smithies. 1984. A comprehen shown to be highly homologous with a class of proteins sive set of sequence analysis programs for the Vax. Nucleic known as peptidylprolyl cis-trans isomerases (50). These Acids Res. 12:387-389. enzymes are capable of changing proline residues from one 13. Dreyfus, L. A. 1987. Virulence associated ingestion of Legion isomer to the other (31). Such changes might affect the ella pneumophila by HeLa cells. Microb. Pathog. 3:45-52. secondary and perhaps, in the case of interchain disulfide 14. Engleberg, N. C, C. Carter, D. R. Weber, N. P. Cianciotto, and bonds, the quaternary structures of the porin. The role of B. I. Eisenstein. 1989. DNA sequence of mip, a Legionella chaperone proteins in porin assembly has only recently been pneumophila gene assoc' ted with macrophage infectivity. In fect. Immun. 57:1263-1270. addressed (25), and continued study of these mechanisms 15. Engleberg, N. C E. Pearlman, D. Dixon, and B. I. Eisenstein. 300 920 HOFFMAN ET AL J BACTERIOL. 1986 Antibodies isolated by using cloned surface antigens the ponn genes. Mol Microbiol 4:1077-1082 recognize antigenicity related components of Legionella pneu 35 Moore, R. A., W. A. Woodruff, and R. E. W. Hancock. 1987 mophila and other Legionella species J. Immunol 136:1415- Antibiotic uptake pathways across the outer membrane of 1417 Pseudomonas aeruginosa Antibiot Chemother. 39:172-181. 16 Fernandez, R. C, S. H. S. Lee, R. Weeratna, M. Ripley, and P. 36 Oldham, L. J., and F. G. Rodgers. 1985 Adhesion, penetration S. Hoffman. Unpublished data. and intracellular replication of Legionella pneumophila an in 17 Gabav, J. E., M. Blake, W. D. Niles, and M. A. Horwitz. 1985 vitro model of pathogenesis. J Gen Microbiol 131:697-706 Purification of Legionella pneumophila major outer membrane 37 Payne, N. R., and M. A. Horwitz. 1987 Phagocytosis of Legion protein and demonstration that it is a porin J. Bactenol ella pneumophila is mediated by human monocyte complement 162.85-91. receptor' J Exp. Med. 166:1377-1389 18 Gosling, L. H.. K. Cabrian, J. C. Sturge, and L. C. Goldstein. Quinn, [. D., C. A. Butler, and P. S. Hoffman. 1987 Charac 1984 Identification of a species-specific antigen in Legionella tenzation and do ung of the disulfide-cross-hnked major outer pneumophila by a monoclonal antibody J Clin Microbiol membrane protein of Legionella pneumophila H116. J. Cell, 20-1031-1035 Biochem 1987:11b 19 Gotschlich, E. C, M. E Seiff, M. S. Blake, and M. Koomev 39 Quinn, F. D., and J„. S. Tompkins. 1989 Analysis of a cloned 1987. Ponn protein of Neisseria gonorrhoeae, cloning and gene sequence of Legionella pneumophila encoding a 38 kD metal structure Proc Natl Acad Sci USA 84:8135-8139 loprotease possessing hemolytic and cytotoxic activities Mol 20 Hindahl. M. S., and B, H. Iglewski 1984 Isolation and charac Microbiol 3:797-805 tenzation of the Legionella pneumophila outer membrane J 40 Rowbotham. L. J. 1980 Preliminary report on the pathogenicity Bacteriol 159-107-113 of Legionella pneumophila for freshwater and soil amoeba. J. 21. Hindahl, M. S., and 3. H. Iglewski. 1986. Outer membrane Clin Pathol 33-1179-1183 proteins from Legionella pneumophila serogroups and other 41 Sambrook, J.. E. F. FriLsch, and T. Maniatis. 1989 Molecular Legionella species Infect Immun 51:94-101 cloning, a laboratory manual, 2nd ed Cold Spring Harbor 22. Hindahl. M. S., and B. H. Iglewski. 1987. Cloning and expres Laboratory Cold Spnng Harbor N Y sion of a common Legionella outer membrane antigen in Esch 42 Sampson, J. S., B. B. Plikaylls, and H. W. Wilkinson 1986 enchia coli Microb Pathog 2.91-99 Immunologic response of patients with legionellosis against 23 Hoffman, P. S. Unpublished data major protein-containing antigens of Legionella pneumophila 24 Hoffman, P S , C. A Butler, and F D. Quinn 1989 Cloning and serogroup 1 as shown by immunoblot analysis J Clin Micro temperature dependent expression in Escherichia coli of d biol 23:92-99 Legionella pneumophila gene coding for a genus common 60- kilodalton anugen Infect Immun 57:1731-1739 43 Silhavy, T J., M. L. Berman, and L W. Enquist. 1984 25 Hoffman, P. S.. L. Houston, and C. A. Butler. 1990 Legionella Experiments with gene fusions Cold Spnng Harbor Labora pneumophila htpAB heat shock operon nucleotide sequence tory, Cold Spnng Harbor. N Y and expression of the 60-kilodalton antigen in L pneumophiia- -»4 Southern, E M. 1975 Detection of specific sequences among infected HeLa cells Infect Immun 58.3380-3387 DNA fragments separated by gel electiophoresis S Mol Biol 26 Hoffman. P. S„ and M. Ripley. 1991 Cloning and charactenza 98-503-517 tion of the MOMP gene of Legionella pneumophila abstr 45 Stephens, R. S., C.-C. Kuo, G. Newport, and N. Agabian. 1985 D 174. p 107 Abstr 91st Gen Meet Am Soc Microbiol 1991 Molecular cloning and expression of Chlamydia trachomatis 27 Hoffman. P S , J. H. Sever, and C. A. Butler. 1991 Molecular major outer membran<" protein antigens in Escherichia coli charactenzation of the 28- and 31-kilodalton subunits of the Infect Immun 47:713-718 Legionella pneumophila major outer membrane protein J 46 Stephens, R. S., G. Mullenbach, R. Sanchrz-Pescador, and N Bactenol 174-908-913 Agabian. 1986 Sequence analysis of the major outer membrane 28 Horwitz, M. A., and S. C. Silverstein 1980 The Legionnaires protein gene from Chlamydia trachomatis serovar L2 J Bac disease bacterium {Legionella pneumophila) multiples mtracel tenol 168.1277-1282 lularly in human monocytes J Clin Invest 66:441-450 47 Stephens, R. S., E. A. Wagar, and U. Edman. 1988 Develop 29 Horwitz, M. A., and S. C Silverstein. 1983 Intracellular multi mental regulation of tandem promoters for the major outer plication of Legionnaires disease bacteria (Legionella pneumo membrane protein gene of Chlamydia trachomatis i Bactenol phila) in human monoevtes is reversibly inhibited by erythro 170 744-750 mycin and nfampm J Clin Invest 71.15-26 48 Tartakovsky, I. S., V. F. Konyukxov, V. V. Pelrosov, and S V. 30 Klickstem, L. B., W. W. Wong, J. A. Smith, J. H. Weis, J. G. Prosorovsky 1990 Protective effect of vanous antigens in Wilson, and D T. Fearon. 1987 Human C3b/C4b receptor expenmental legionellosis abstr 33 European Working Group (CRl) Demonstration of long homologous repeating domains on Legionella Infections that are composed of short consensus repeats charactenstic of 49 Thacker, W. L., H. W. Wilkinson, B. B Plikaytis, A. G C3/C4 binding proteins J Exp Med 165 1095-1112 Steigerwalt, W. R. Mavberry, C. W. Moss, and D. J. Brenner. 31 Liu, J., and C. T. Walsh. 1990 Peptidyl-prolyl cijr-iranj-isom- 1985 Second serogroup of Legionella feeleu strains isolateu erase from Escherichia coli a penplasmic homolog of cyclo- from humans J Clin Microbiol 22:1-4 philm that is not inhibited by cyclosporin A Proc Natl Acad 50 Tropschug, M., E. Wachter, S. Mayer, E. R. Schonbrunner, and Sci USA 87:4028-4032 F. X. Schmid. 1990 Isolation and sequence of an FK506-bindmg 32 Ludwig, B., A. Schmid, R Marre, and J. Hacker. 1991 Cloning, protein from A* crassa which catalyzes protein folding Nature genetic analysis, and nucleotide sequence of a determinant (London) 346.674-677 coding for a 19 kilodalton peptidoglycan-associated protein 51 Wadowsky, R. M., L. J. Butler, M. K. Cook, S. M. Verma, (Ppl) of Legionella pneumophila Infect Immun 59:2515-2521 M A Paul, B S Fields, G. Keleti, J. L. Sykora, and R. B. Yee. 33 Marra, A., M. A. Horwitz, and H. A. Shuman. 1990 The HL-60 1988 Growth-supporting activity for Legionella pneumophila in model for the interaction of human macrophages with the tap water cultures and implication of hartmannellid amoebae as Legionnaires disease bacterium J Immunol 144-2738-2744 growth factors Appl Environ Microbiol 54-2677-2682 34 Mizuno, T., and S. Mizushima. 1990 Signal transduction and 52 Zhao, X , and L A Dreyfus. 1990 Expression and nucleotide gene regulation through the phosphorylation of two regulatory sequence analysis of the Legionella pneumophila recA gene components the molecular basis for the osmotic regulation of FEMS Microbiol Lett 70:227-232 301 iNrrxnoN VNB IMMUNITY Aug 1994 p 3454-3462 Vol 62. No 8 (1019 «567/94/$04 ()()+() Copyrniht " IW4 American Socicly for Microbiology Human and Guinea Pig Immune Responses to Legionella pneumophila Protein Antigens OmpS and Hsp60 RISINI WEERATNA ' DAVID A STAMLER - PAUL H CDCLSTEIN - MURRAY RIPLEY ' THOMAS MARRIE '" DAVID HOSKIN ' \m PAUL S HOFFMAN1 v Dtpauineiit oj Micwbiolocy ami Immunology ' and Department oj Medicine ' Dutsion of Injections Diseases DalhouML Unnersm Halifax Vina Scotm Cunudu B3H 4H7 and Department of Pathologj and Laboratory Medicine and Department of Mediant Uimersin of Pennsxlxama Medical Cento Philadelphia Penns\hatua 19104- Reeeived 10 Februirv 1994/Retumed lor modihcation 12 April Iv94/Accepted 17 Mav 1994 We studied the immune responses of guinea pigs and humans to two Legionella pneumophila antigens. Guinea pigs surviving a lethal intraperitoneal challenge dose of virulent L pneumophila exhibited strong cutaneous delayed tvpe hypersensitivity (DTH) reactions to purified OmpS (28-kDa major outer membrane protein) and Hsp60 (heat shock protein or common antigen) while weak DTH reactions were noted for extracellular protease (major secretory protein [MSP] [ProA]) and no reaction was observed with an ovalbumin (OA) control. Lymphocyte proliferation responses (LPRs) were measured for peripheral blood and spleen lympho cytes from guinea pigs surviving sublethal and lethal challenge doses oft pneumophila Lymphocytes from uninfected animals showed no proliferation to Hsp60 or OmpS while lymphocytes from sublethal!* and lethallv challenged animals exhibited strong proliferative responses to Hsp60 and OmpS. Guinea pigs vaccinated with pun lied OmpS exhibited low antibodv titers and strong DTH and LPRs to OmpS, whereas lymphocytes from animals vaccinated with Hsp60 exhibited weak DTH responses and high antibody titers to Hsp60 All guinea pigs immunized with OmpS survived experimental challenge with L pneumophila (two of two in a pilot studv and seven of seven in trial 2) versus zero of seven OA-immunized controls (P = 0 006 bv Fisher s exact test) In three vaccine trials in which animals yvere vaccinated with Hsp6(), only 1 guinea pig of 15 sunned lethal challenge Peripheral blood lymphocytes (PBLs) from humans with legionellosis showed stronger LPRs to OmpS than PBLs from humans with no history of legionellosis (P = 0 0002 by Mann Whitncv test) PBLs of humans suniving legionellosis exhibited a lower but highlv significant proliferate e response to Hsp60 (P - 0 0001 compared with controls bv Mann Whitney test) These studies indicate that OmpS and H.spfiO are important antigens associated with the development of protective cellular immunity However, as determined in yaceine trial studies in the guinea pig model for legionellosis, the species-specific antigen OmpS proved much more elective than the genus common HspfiO antigen Legionella pneumophila is an opportunistic human p ithogcn MSP protease also conferred protection from lethal challenge ' tat causes an acute and often tatal torm ot pneumonia with L pneumophila (8) The major proteins in membrane pirticularlv in debilitated or immunocompromised patients preparations include the major outer membrane protein The b icteria invade iheol ir macrophages where thev ibro (OmpS) (14 29), Mip (15, 21) Ppl (Pal) (22 40) flagellum gate phagosome Ivsosomc fusion and multiplv (33) Survival proteins (41) and noncovalentlv associated HtpB (Hsp60) (24 trom legionellosis is generally btheved to requnc an active 28) eellul ir immune response (12 3~> 37) Macrophages activated HspbO (HtpB) is a member of the highlv conserved GroEL bv g imma intertcion inhibit the intrace'iul ir replication of L or HSP60 tamilv of heat shock protems"(27 28 30), which tor pneumophila (3) T Ivniphacvtes prcsuiiiablv respond to Legio manv microbial pathogens are immunodominant antigens (35 ntllei inrcclcd macrophages through recoanitun of specific 49 -13) Conv-descent phase sera from culture confirmed cases peptide antigens presented bv major histocompatibility com of legionellosis contain ami Hsp60 antibodies as indicated bv ple\ (MHC) class 1 and II molecules L pneumophila protein immunoblot (24 48) and human lymphocytes proliferate in antigens involved in the cellular immune response have onlv response to this antigen (28) We have recently reporter! that reeentlv been investigated Blander and colleagues demon Ivmphocvtes trom cardiac transplant patients convalescing str ited th it vaccination ot guinea pigs with l 38 KDa extruccl from confirmed legionellosis proliferate in response to OmpS luhr protease (major secrctorv protein [MSP]) stimulated a and to a lesser extent toHsp60(51) While Hsp60 is the most eellul ir immune response which conferred protection from abundant protein svnthesized by intracellular L pneumophila lethal challengt (7 11) However aproteolvtic strains of L onanisms (1, 23 28 30) preliminary studies indicated that pneumophila were lound to be both virulent ind immunopro tective suggesting that other hetors m iv be involved in viccination ot guinea pigs with purified Hsp6() does not protect cell mediated immunity (8) These investigators showed that these jnimals from lethal challenge (20) In contnst several guinea puss vaccinated with a membrane preparation devoid of preliminary studies showed (hat OmpS-vaceinated guinea pigs were protected from lethal challenge (47, 50) OmpS is a unique outer membrane protein found onlv in L pneumophila " Corresponding luthor Mailing address Depirtment of Mierohi strains (14 24 29 31) In the present studs we have charac oloev md Immunoiow Sir Cli irlesTupper Medic ll Building D illiou tenzed the humoral and cellulur immune responses of guinea sie Unnersitv Halifax Novi Scotia Canada B1H 4H7 Phone (902) pigs and hum ins to punhed L pneumophila Hsp6() and OmpS 494 itiTO Fix (902) 494il2a Cleitronie m ul lddress PHOFF antigens These studies show that humans with culture con MAN«t AC DAL CA firmed legionellosis and guinea pigs surviving challenge with L "4s4 I 302 VOL 62, 1994 LEGIONELLA IMMUNE RESPONSES 3455 pneumophila develop cellular immune responses to OmpS and plates were incubated in a humidified C02 incubator at 37°C. HspfiO, but only OmpS proved immunoprotective in the guinea On the indicated days, 0.5 u.Ci of [•1H]thymidine was added to pig model for Legionnaires* disease. wells for 4 or 6 h. Cultures were harvested and counted by liquid scintillation. Each sample was plated in triplicate, and the means and standard deviations were determined. The MATERIALS AND METHODS proliferative response was computed by subtracting the counts Strains and protein purification. L. pneumophila SVir (Phil per minute (cpm) of the medium control from the cpm adelphia 1) was usd for preparation of OmpS monomers and obtained for the antigen-challenged lymphocytes (Acpm). HspfiO (13, 25). Other strains of L. pneumophila used in this When proliferative responses from different experiments were study included patient and environmental isolates belonging to compared, the data were normalized by computing a stimula monoclonal antibody subtypes France. Olda. and Oxford. For tion index (SI). The SI was determined by dividing the cpm ihe guinea pig vaccine studies. L. pneumophila serogroup 1 obtained in'the presence of antigen by the cpm obtained for strain F889 was used (19). All strains were grown on ACES controls receiving no antigen. pV-(2-ucetumido)-2-aminoethanesulfonic acidj-buffered char Human LPRs. Approximately 40 ml of peripheral blood was coal yeast extract (BCYEct) agar medium supplemented with collected from humans who had recovered from culture- a-ketoglutarate as described previously (17). Hsp60 was puri confirmed legionellosis and from human volunteers with no fied from Escherichia coli pSH16 following heat shock-induced history of Legionnaires' disease. Informed consent was ob overexpression of the L. pneumophila htpAB operon as de tained from each person. Peripheral blood was diluted in PBS. scribed previously (27) with the following modifications. Sev and the lymphocyte fraction was recovered following centrifu eral drops of Tween 80 were added to the crude bacterial gation on a Ficoll-Hypaque gradient (Flow Laboratories). extract to solublize HspfiO from the membrane material prior Lymphocytes were adjusted to a final concentration of 2.5 X to ion-exchange chromatography on a DE-52 column. The lO1" cells per ml in RPMI 1640 medium containing penicillin pooled and concentrated Hsp60-rich fraction was subjected to (100 U/ml), streptomycin (100 p-g/ml), 2 mM L-glutamine, and molecular sieve chromatography on a Sephadex G100 column, 5% human AB serum. The lymphocytes were washed twice in and then the pooled Hsp6()-containing material was subjected PBS and introduced into 96-well U-bottomed tissue culture to ultracentrifugation on a sucrose gradient as described by plates (Linbro; F'ovv Laboratories). Antigens and [•'H]thymi- Lecker et al. for GroEL purification (39). OmpS was purified dine vvere added us described above. For transplant patients bv repeated boilinas of L. pneumophila bacteria in 1% sodium receiving cyclosporins A prophylaxis, lymphocytes were incu dodecyl sulfate (SDS) followed by boiling in 2Tc SDS plus 10% bated for 24 h in RPMI 1640 medium prior to antigen (3-mercaptoethanoI (13, 27, 31). From the immunological and challenge. Lymphocyte proliferative responses (LPRs) vvere autoradiographic results and as indicated previously, we esti determined for all individuals on day 6. For comparative mate that there was less than 1 ng of HspfiO per mg of OmpS purposes, an SI was computed for each individual. (13. 14.23.25,31). Immunological procedures. Antibody levels in peripheral Guinea pigs. Either male or female Hartley strain guinea blood from guinea pigs and humans vvere determined by an pigs (Charles River Laboratories) were used in these studies. enzyme-linked immunosorbent assay (ELISA) with purified In one set of experiments, female guinea pigs surviving intra Hsp60 and OmpS as described previously (25). Alkaline- phos- peritoneal (i.p.) infection with virulent .strains of L. pneumo phatase-conjugated anti-guinea pig or human immunoglobulin phila were examined for cellular and humoral immune status at G was used for the detection. All human sera were screened 5 weeks postrecovery. For the animal vaccine trials, male for L. pneumophila serogroup-specific antibody as previously Hartley animals weighing 350 lo 425 g (trial 1) and 200 to 300 described (52). g (trial 2) at the time of primary immunization were used. The Guinea pig vaccination. Two animal vaccination trials were animals were housed and quarantined as described previouslv performed. In trial 1, animals were randomly assigned to be (19). immunized with Hsp60 or crude chicken OA, grade I (Sigma, Skin testing. Guinea pigs were shaved over the hind quarters St. Louis, Mo.), as a control. In trial 2, animals were random and injected intradermally at different sites with purified ized to receive OmpS (28 kDa), Hsp60, or OA. The OA was protein antigens. These antigens included purified exoprotease dissolved in PBS and sequentially filtered through 0.45-u.m and (10 etg), 28-kDa OmpS monomers (0.1 to 10 u-g). HspfiO (0.1 to 0.22-eim (low-protein-binding)-pore-size sterile filters (Milli- 10 u.g), and ovalbumin (OA) (10 (ig). each in 100 |el of pore. Bedford. Mass.). Primary immunizing suspensions were phosphate-buffered saline (PBS). Diameters of induration and made by diluting the stock protein solutions in PBS and erythema were measured at 24 and 48 h after injection of the combining equal volumes of each diluted solution in Freund's antigens. complete adjuvant (Sigma). After the flank had been shaved, Lymphocyte proliferation assays. Peripheral blood was col 50 ceg (Hsp60 and OA groups) or 25 p.g (OmpS group) of lected from guinea pigs by cardiac puncture, and lymphocytes protein in 0.05 ml was injected intradermally into each animal. were separated on u Ficoll-Hypaque gradient. Spleens were Animals vvere reimmunized 3 weeks later with the same aseptically removed and homogenized to prepare a single-cell amount of protein by the same protocol except that Freund's suspension of lymphocytes. Lymphocytes were suspended in incomplete adjuvant (Sigma) was used and 0.10 ml was in RPMI 1640 medium containing penicillin (100 U/ml), strepto jected. One animal from each group in trial 1 was skin tested mycin (100 u.g/ml), 2 mM L-glutamine. and 5% fetal calf serum to assess delayed-type hypersensitivity (DTH). (Flow Laboratories, ICN Biochemicals Canada Ltd.. Missis- Guinea pig challenge. Six weeks following primary immuni sauga. Canada). The lymphocytes were a'djusted to a final 6 zation, each animal was challenged with a fatal dose of L. concentration of 2.5 x 10 cells per ml and plated into 96-well pneumophila delivered by the intratracheal route as described U-bottomed tissue culture plates (Linbro: Flow Laboratories). previously (19). A clinical isolate of L. pneumophila serogroup Antigens (OA, OmpS. and Hsp60) were suspended in RPMI 1 (strain F889) was used to infect animals in both vaccination 1640 medium and used at'0.1, t.0. and 10 cig/ml. Concanavalin trials and was prepared from guinea pig lung homogenate as A (5 to 10 p.g/ml) and pokeweed mitogen (1 to 5 ccg/ml) were described previously (19). On the day of infection, bacteria similarly suspended in RPMI 1640 medium. The mierotiter were diluted in normal saline to deliver approximately 2.5 I 303 3456 WEERATNA ET AL INFFCT IMMUN tunes the 50rr lethal dose (LD,„) to each animal, using TABLE I Skin induration 48 h idler skin lestinu with purified previously described methods (19) The ictual infecting inoc inlmen unong guinea pigs surviving L pneumophila infeclion ulum was determined by viable cell plate counting and was Mem sfun indur^tton diam (mm) i SD 160X l()7CFU per animal in trial 1 and 2 47 X 107 CFU per Guinea pig (. hallenee i,ruup* dose (CPU) animal in (rial 2 Immediately pnoi to infection blood was OA Ompb tfcphll collected from the timnuls in each group tor serologic testing CI None () 0 0 Anim.i's vvere observed for tachypnea feeding, and level of SLD ltr1 13 ± 1 S I2± Ifi 2 3 ± 2 0 7 activity tor I Oduys following infection Daily weights and rectal LD I0 0 s3 i 1 1 112*34 temperatures were recorded lor surviving animals on days I HD 10* I) 7 1*23 63 £ 14 through o and on day 10 and were eompired with bnsehne ( I eontrnl Lroup not exposed lo / / mumoplulI measurements taken betore infection Animals vvere classified us m inbund if they vvere inactive and did not eat or drink. This stati has physiologic correlates that usually include a bodv RESULTS temperature of greater than 41 "C or less than 37 5"C and DTH responses of immune guinea pigs to L pneumophila weight loss exceeding 25r<- ot the prcinlection weight and r antigens In a preliminary studv nine guinea pigs surviving predicts ultimate death with 99 r confidence (US) Moribund lethal challenge with various environmental and patient iso animals were killed hur ,anelv bv i p injection of pentob irbitol > lates of L pneumophila were skin tested 30 davs postinfection sodium (50 mg 'Abbott Laboratories North Chicago III) tor cutaneous DTH reactions to purified proteins (rom L Animals surviving the 10-dav studv period vvere killed similarly pneumophila serogroup 1 We found strong DTH responses in Nccropsvvva performed on all animals within 24 h ot death as these guinea pigs lo both the 28 kDa OmpS (major outer described previously (19) Lung tissue was harvested for quun membrane protein) (9 i 1 mm at 24 h and 6 £ I mm at 48 h) titutivc culture and histology ind blood was collected for and Hsphl)(12 i 2 mm at 24 h and <4 mm at 48 h), while the serological testing response to pure exoproteasc was somewhat lower (6 * 1 mm Lung processing. Quantitative culture ot guinea pig lung was it 24 h ind 0 mm at 48 h) No response was noted in these done by plating 0 1 ml ot serial 10 told dilutions of lung experiments tor animals skin tested with II) p.g of OA per ml as homogenatc onto BCYEa agar as described previously (19) a ncaative control Uninfected immais exhibited no DTH res| mse to these antigens In a related experiment three The CFU were counted and normalized to lung weight All groups ot inimals surviving various challenge dosages ol lutm concentrations of £ pneumophila are expressed as log,,, virulent L pneumophila were examined for DTH and LPR CFU per gram of tissue unless otherwise, stated Because The animals were divided into tour groups of three animals guinea pig lung homogenatc is inhibitory to the growth ot each with group CI being uninfected controls The sublethal Legionella spp on BCYEa medium a 11) ' dilution of the dose (SLD^ group survived an i p dose of 104 Cru whereas origin il homoiicnate was the lowest concentration plated the animals in the lethal dose (LD) and high lethal dose (HD) ; Hence the lower limit ot detection was 10 CFU g of lung groups survived challenge doses of I07 and 10" CFU. respec Lung tissue tor histologic studv was hxed in 10rr neutral tivelv Animals in the SLD group exhibited no significant DTH Formalin at 43C for 7 davs prior to processing In trial 1 lung reactions to OA or OmpS while exhibiting weak DTH re samples trom tour Hsp60 immunized animals and three OA- sponses to Hsp6() (see Table 1) Animals surviving ip cnal- lmmunized animals were studied in trial 2 histopathologic lenge with higher doses ot L pneumophila exhibited DTH scoring ot lung tissue was pcrtormed for samples trom five responses to OmpS and Hsp6() Table 2 depicts the LPR results animals each from the OmpS and Hsp6() groups Except tor for splenic lymphocytes obtained 144 h (6 days) after antigen one animal in trial 2 specimens were selected randomly Fixed challenge Concanavalin A and pokeweed mitogen vvere used lung samples were inspected with a dissecting microscope to as positive (mitogemc) controls and OA was used as a negative control The proliferative responses represent the means ot localize the most consolidated areas tor studv After being three determinations tor each animal Maximum proliferation Stained with hematoxylin and eosin sections vvere reviewed in was observed with 10 u,g of OmpS and Hsp6() per ml There a blinded manner bv l single investigator it low power (4()x) was no significant difference in the magnitude ot the LPR tor and the degree ot consolidation was estimated as a percentage splenic lymphocytes from the SLD group challenged with ot lhe total lung area in each held Additionally qualitative OmpS and HsphO However immais surviving higher-dose assessment ot the predominant inflammatory cell infiltrate was challenges exhibited a greater prolifcr itive response to OmpS made The entire slide was examined for each specimen than to HspfiO Antibody titers to Hsp60 OmpS and OA were Statistical analysis When no growth was observed in quan also determined None of the animals exhibited titer greater titutivc lung culture the lower limit of detection (10: CFU g ot than 1 64 to OmpS or to OA Animals m the SLP jioup lung) was used for statistic il analysis Unless otherwise stated exhibited no ippreciable antibody to Hsp60 However the LD mem values between vaccination groups (gumei pms) or and HD groups exhibited antibodv titers to Hsp60 ranging between human patient and control groups vvere compared bv from slightly greater than 1 64 to I 2 800 No correlation was found between the magnitude ot the proliferative response ot using in unpaired two-tailed Students t test If the standard lymphocytes to Hsp60 and antibodv titer deviations between lhe two groups differed significantly or if the values compared were not expected to have a Gaussian Immune r-.>ponse to OmpS In assessing the efficacy of distribution the Mann Whitnev two sample test was used OmpS as a vaccine candidate we firstdetermine d whether this instead ot Student s t test A two-tailed Fisher s exact test was protein could stimulate a strong cellular immune response used to compare survival rates between vaccination groups Guinea pies vvere immunized with 25 p.g ot OmpS in Freund s Statistical calculations vvere performed with Instat Software incomplete adjuvant Four guinea pigs were immunized with (version I 10 1990 Graphpad San Diego Calif) Unless 25 ccg ot OA After 5 weeks the animals vvere skin tested to otherwise stated values given in the form X ~ Y represent the assess DTH and subsequently peripheral blood and spleens mean ± standard deviation were collected and examined for LPR Table 3 lists the results 304 VOL. 62. 1994 LEGIONELLA IMMUNE RESPONSES 3457 TABLE 2. Guineu pig LPRs" Mean LPR (10-Acpm) Addition'1 COlKtl Control group LD group HD group (u.g,ml) SLD group 1 : 1 2 .1 1 2 3 1 -> 3 ConA 5 f>6.7 8.59 11.0 23.4 3.33 23.5 1.04 •2.02 0,64 1.40 2.75 PWM ^ 50.2 27.7 7.2 14,1 2.61 18.2 0.72 ' 1.2 2.1 1.37 3.71 OA 10 <0.l <0.l <(),! <0.l <().! <().! <0.1 <0,I <0.1 <().l <0.1 OmpS (I.I —' — 2.3 3.13 15.1 11.7 0.25 <0.1 <0.1 <0.1 <0.1 1.0 — — 27,7 8.8 36.6 21.71 2.77 0.48 3.9 1.1 2.1 10 1.9 <0.1 22.16 lfi.2 38.46 55.27 11.6 11.0 20.85 15.03 16.1 Hsphtl 0.1 — — 0.25 3.37 0.25 5.93 0.54 <0,1 0.12 <0.I <0.1 1.0 — — 7.53 6.2 3.16 9.68 3.43 0.81 3.61 <0,1 <0.1 10 1.3 1.0 15.19 24.61 20.31 24.19 9.37 2.47 13.1 3,37 8.9 " Guinea pigs surviving .sublethal (SLD). low-dose (LD), —and ...higc..-dosh e (KD) challenge were examined for cellular immune response al 5 weeks postinfection hy measuring splenic LPRs lo OmpS and Hsplifl. The conlrol group was nol exposed to L fmtltiiwplulti. Cultures were set up in °fi-\vell culture dishes and. following antigen challenge lor (i days, were pulsed with [*H jthynudi ic lor 4 h. The results arc means for triplicate cultures. The standard deviation ranged from 10 lo 15rr of the mean cpm. ''ConA. concanaviilin A: PMW. puke weed mitogen. ' —. not delermined. of these experiments. Guinea pigs immunized with OmpS oped induration and erythema 3 mm in diameter at 48 h when developed stronger DTH reactions than the OA-immunized skin tested with OA; no DTH response was observed with controls (data not presented). The reactions persisted for 72 h. HspbO for this animal. Five animals each were immunized with The following week, LPRs were determined for peripheral Hsp60 or with OA and subsequently challenged with L. blood lymphocytes. OmpS-vaccinated animals developed pneumophila. All 10 animals became sick within 2 days of strong LPR to challenge with 10 p.g of purified OmpS protein intratracheal infection. By day 4 postinfection, the first deaths per ml. The antibody titers (measured by ELISA) of animals were observed, and the survival rates were identical in the receiving OmpS proteins were s500. lmmunoblots made with Hsp60 and OA groups. No animals from cither group survived sera from OmpS-immunized animals at a dilution of 1:100 past day 5 in this trial. Quantitative cultures of lung harvested indicated that antibody to OmpS was present in these animals at necropsy revealed L pneumophila in all cases, and bacterial (data not presented). While the proliferative responses of concentrations vvere significantly higher in the OA control lymphocytes from OmpS-immunized animals to challenge with group than in the Hsp60 group. For the OA animals, the mean HspOO were not determined, OmpS-immunized animals exhib lung concentration of L. pneumophila was 10.0 log,n CFU/g ited no anti-Hsp60 titer greater than that of OA-immunized (range, 9.8 to 10.3) compared with 8.9 logm CFU/g (range. 7.5 controls. to 9.8) for the Hsp60 animals (P = 0.046). At necropsy, Vaccine trial I. Skin testing of the Hsp60-immunized guinea bilateral consolidation of the lungs was noted in all animals pig with the homologous protein revealed induration and and was confirmed histologically in the animals whose lungs erythema 6 mm in diameter at both 24 and 48 h and no were examined. The mean percentage of lung consolidation response to OA. Similarly, the OA-immunized animal devel was 95% in the Hsp60 group and 93TJ in the OA group (P = TABLE 3. Proliferation of lymphocytes from OmpS-vaccinated guinea pigs to different proteins" Mean LPR Reciprocal amihod\ liter Antigen Group (no.) (II)'Acpm) Mean SI lested ± SEM OmpS Hspdll OA1 (2) <50 <50 ConA 9.3 ± 1.3 51.2 OA <0.1 0.9 OmpS I 305 3458 WEERATNA ET AL. INFECT. IMMUN. 700 100 'p., 80 '• 60 '& o OmpS 3 . a Hsp60 CO 40 4 OA -. ao - b---lr.-rj 0 - A '—il—'— 2 3 4 5 10 DAYS POSTINFECTION 12 3 4 5 DAYS POSTINFECTION FIG. I. Survival of vaccinated guinea pigs infected with --2.5 times the I.D<„ of L. pneumophila ndministcrcd by the intrainichcal route FIG. 2. Weights of vaccinated guinea pigs infected with -2.5 times (trial 21. Seven animals each were vaccinated with OmpS or OA and six the LD«,of L pneumophila administe-red by lhe intratracheal route in animals vvere vaccinated with HsphD prior to infection. vaccine trial 2. Seven animals each vvere vaccinated with OmpS or OA and six animals were vaccinated with HsphO prior to infection. Me-ans and standard errors are shown. 0.86. Mann-Whitney two-sample test). In all cases, a mixed cellular infiltrate with macrophages and polymorphonucierr lung for 1 IspbO-immunized guinea pigs (75rc) was considerably leukocytes was noted. Preinfection mean guinea pig weights higher than lliat observed for the OmpS-immunized animals were similar for both immunization groups and fell following (12Cr) (P = 0.008. Mann-Whitney two-sample test). All lung infection. There were no significant differences in guinea pig samples from OmpS-immunized guinea pigs examined had weights or temperatures between the two groups at any time predominantly mononuclear infiltrates, while 80rr of those following infection (P >0.7. on day 3). from the Hsp60-immunized animals examined had both poly Vaccine trial 2. In a previous experiment, two guinea pigs morphonuclear and monocytic infiltrates. immunized with 50 p.g of OmpS per ml in Freund's incomplete Mean hasclinc weights vvere similar among the three groups adjuvant survived lethal challenge with 2 x 10* bacteria given (P '•• 0.8, one-way analysis of variance) and decreased follow by the i.p. route, whereas two animals immunized with OA and ing infection. Mean weights differed significantly after day 4 similarly challenged did not survive. To confirm these findings, postinfection between OmpS-immunized guinea pigs and both seven animals per group were immunized with either OmpS. HspOO-immunized animals (P = 0.01) and OA-immunized Hsphl). or OA prior to experimental intratracheal infection. animals (P = 0.02) (Fig. 2). The OmpS-immunized animals One animal in the Hsphl) group died within 24 h after infection began gaining weight on day 3 and continued to do so with Legionella bacteria, of a polymicrobial neck abscess, and throughout the study period. In contrast, Hsp60-immunized data for this animal were excluded from the analysis. All animals did not gain weight until day 5. and no control animals animals immunized with OmpS survived for the entire study gained weight. By day 10. the survivor in lhe Hsp6() group period, in contrast to the OA control group, of which no regained substantial weight but did not exceed its preinfection animals survived (P = 0.0006. Fisher's exact test) (Fig. 1), and weight. In contrast, six of seven animals in the OmpS-immu the HspbO group, of which only 17^ survived (P = 0.005 versus nized group exceeded their preinfection weight by the end of the OmpS group, by Fisher's exact test). Cumulative survival in this 10-day period. the HspbO-immunizcd group was significantly higher than in the OA control group on day 4 following infection (P = 0.02, The mean temperature curves for the Hsp60 and OmpS Fisher's exact test). As observed in vaccine trial 1. most groups (not shown) were very similar and differed from that for animals in the Hsp6l) and OA groups were sick by day 2. the OA group animals, who became hypothermic on day 3 However, all OmpS-immunized animals remained relatively postinfection. well following infection. Survival at day 5 predicted survival at Antibody titers to Hsp60 and OmpS following immuniza day 10. and ail OmpS;immunizcd animals that were alive at the tion. Antibody titers vvere determined for five of seven animals end of the study were well. in each group. The results of the ELISA study are presented both as the means of absorptions for a 1:400 dilution of the Quantitative lung cultures demonstrated detectable L. pneu sera and as the estimated titer (Table 4). All animals vvere mophila in all HspftO- and OA-immunized animals but in only negative when examined for antibody titers to the antigen one of seven OmpS-immunized guinea pigs [P = 0.005, Fisher's exact test). For the OA group, the mean luiui concen tration of L. pneumophila was 10.1 logm CFU'g (range, 9.5 to 10.4) compared with 7.8 log,,, CFU/g for the HspfiO group (8.7 TABLE 4. ELISA results for immunized guinea pigs prior when the value for the sun'ivor [3.3J was excluded: range, 8.0 to infection to 9.2). These results were essentially consistent with the Immunization group OD.iI 1:40(1 Titer results obtained in vaccine trial 1. These differences were (no. ot animals} dilution of serum (I dilution) highly significant for Hsp60 versus OA (P = 0.001. Mann- Whitney two-sample test).. Hspht) (5) 1.51 > 12.800 OmpS (5) 0.46 400-1.600 The histopathologic scoring of fixed lung samples from trial OA(2) 1.06 > 12.80(1 2 animals revealed that the mean percentage of consolidated 306 VOL, 62. 1994 LEGIONELLA IMMUNE RESPONSES 3459 (Hspol) or OmpS) not used for immunization. Guinea pigs immunized with Hsp60 exhibited a strong humoral response. With a titer in excess of 12,800. Although net presented, the antibody titers for guinea pigs immunized with Hsp60 in trial 1 vvere in excess of 105. Similarly, the results for OA-immunized animals indicated a strong humoral immune response (optical density [OD] = 1.06; titer, 12.800), In contrast, antibody titers against OmpS were low (OD = 0.46), and the antibody titers of these animals ranged from 400 to 1,600. Cellular immune responses to Legionella antigens in hu mans. To determine whether humans surviving legionellosis also develop a cellular immune response to OmpS and Hsp60, we studied patients and volunteers, Figure 3 displays the range of LPRs lo Hsp6() and OmpS in these groups. It was evident throughout this study that there was variation among individ uals in response to these antigens. In general, lymphocytes from patients surviving legionellosis exhibited a significant proliferative response to OmpS (SI = 16.6 ± 10.2) compared with controls (SI = 4.4 ± 2.6; P = 0.0002, Mann-Whitney two-sample test). A somewhat lower proliferative response to Hsp60 was observed with lymphocytes from patients (SI = 7.8 i 4.5) than with those from controls (SI = 3.0 ± 2.2; P s 0,0001. Mann-Whitney two-sample test). The relatively low LPR to HspfiO was also observed in a parallel study of human diabetes patients and controls (SI = 2.2 ± 1; manuscript submitted for publication). We found no increase in LPR to Hsp60 among individuals who were tuberculin skin test posi tive (P — 0.12. Mann-Whitney two-sample test). All individuals in this study were tested for antibody titer against Hsp60 and OmpS. With the exception of one heart transplant patient with Legionnaires' disease who exhibited a high antibody titer to HspOO. none of the patients or controls examined in this study exhibited antibody titers to either protein that were detectable by ELISA. The patient with the high antibody titer against HspfiO exhibited no antibody titer to OmpS. All patients who recovered from legionellosis seroconverted to L. pneumophila serogroup 1 LPS antigen, while controls vvere negative, as determined by immunofluorescence (data not presented). DISCUSSION In this study, we examined the humoral and cellular immune FIG. 3. Scatter graph of human LPRs to L. pneumophila antigens. responses of humans and guinea pigs to purified antigens from Peripheral blood lymphocytes were collected from patients and control L pneumophila. Guinea pigs surviving a high challenge dose of subjects. Details of the assay are described in the text. The Sis for each L. pneumophila had strong DTH reactions against Hsp60 and individual represent the means for three determinations. The horizon OmpS 6 weeks postinfection. Lymphocytes from these animals tal lines depict lhe means for the groups. Control, control subjects who also displayed strong proliferative responses to Hsp60 and had no history of legionellosis and vvere serologically negative for OmpS. suggesting that epitopes from these proteins must play serogroup I antigen. The patient group was composed of 10 individuals a prominent role in antigen presentation associated with with culture-confirmed Legionnaires' disease. Two of the patients had activation of the cellular immune response. Interestingly, had heart transplants, three had no significant underlying disease, two animals infected with fewer bacteria appeared to develop a had malignancies, one had systemic lupus erythematosus, one had chronic obstructive pulmonary disease, and one had rheumatoid stronger cellular immune response to these antigens, as mea arthritis. Hsp, HspfiO. Concanavalin A was used as the positive control, sured by LPR, than animals surviving a higher infectious dose. and OA was used as a negative control. Since .sublethal!/ infected animals were negative by skin test, this suggests that the DTH response may not accurately reflect the immune status of animals exhibiting no clinical symptoms of disease. Other studies have tended to support this observa LPRs. while animals immunized with Hsp60 were poorly tion (6). Peripheral blood lymphocytes from humans surviving protected against lethal challenge and developed a strong legionellosis exhibited strong proliferative responses to OmpS humoral response. In addition to improved survival, the and somewhat lower responses to Hsp60. Finally, dramatic OmpS-immunized animals had other evidence of protection protection was observed in guinea pigs immunized with OmpS compared with controls, including the parameters of extent of and subsequently infected with L. pneumophila, as evidenced pneumonia, bacterial concentrations in the lung, and weight by lack of clinical illness and mortality in this group. The loss. Relative bacterial clearance or multiplication rates for the results of this experiment support the findings of a pilot study different groups are unknown because lung bacterial concen that had suggested an immunoprotective role for OmpS. trations were determined on different days. Similarly, the Animals immunized with OmpS exhibited strong DTH and ability of OmpS immunization to diminish disease is reflected 307 3460 WEERATNA ET AL INFECT IMMUN bv the extent ot lung consolidation following challenge in the to generate a strong cellular immune response in guinea pigs OmpS group the mean percentage of lung consolidation was In addition, previous studies had demonstrated that two im \2c'r This contrasts with observations lor control animals, in mumzations with either MSP protease or L pneumophila which consolidation following experimental infection exceeds membrane material were sufficient tor the development of 90' f (19) These hndings are consistent with those of previous cell-mediated immunity (8, 9) Furthermore, because of the protection studies in which immunization led to improved overwhelming humoral response in HsphO-immunized ani survival, induction ot cellular immunity and reduced bacterial mals we assumed that additional immunizations vvould not conccntrttions in the lung (9, 12) alter the immune response While differences in the challenge Additional information suggests that OmpS may be impor dose ot L pneumophila might account for the observed tdnt in eliciting protective immunity to L pneumophila OmpS differences, it was not possible to compare our challenge dose is the most abundant protein on the surface ot the bacterium (2 5 limes the LD,,, instilled into the trachea) with that used by (14 31) Although not clearly established as a virulence factor Blander and Horwitz (L.Dl0(( dose by aerosol) (10) Neither the OmpS is known to bind complement components C3b and purity of the HspfiO preparations noi their sources appear to C3bi facilitating mononuclear phagocytosis and permitting account tor the reported differences In our studies, recombi cellular entry and intracellular multiplication (2) Tdrtnkovskn nant Hsp&O was purified from E eoh pSH16 Overproduction and colleagues have shown that immunization with a hpopolv ot heat shock proteins by recombinant methods is commonly saccharide (LPS) containing outer membrane preparation ot employed for immunological studies (41 49) However we L pneumophila was protective for guinea pigs challenged with tound that guinea pigs immunized with Hsp60 purified trom L L pneumophila (46 50) In a subsequent studv in which LPS pneumophila also did not survive lethal challenge (20) Our was removed Irom the OmpS preparation no protection from results are consistent with those obtained in the mycobacterial lethal challenge was observed (50) However from previous studies (35 43 44 33) While heut shock proteins of the studies as well as the work of others, it is unlikelv that LPS can GroEL or HSP&O class may well be important cellular immune be completely removed from OmpS (24 26) While it has been antigens in the natural infection it is generally believed that shown that LPS mav be important in permitting the formation vaccination with pure HSP60 protein mav not achieve the same ot teniarv structure in porins that are essential for antibodv result (41 43 44 ^3) The overwhelming humoral response production (42) information on the role ot LPS in the which is not seen in the natural infection ot guinea pigs with L development ot cellular immune responses against outer mem pneumophila is a commonly reported consequence of immu brane proteins has not been evamined In our studies with flow nization with HSP60 proteins (43) cvtometrv we found no evidence for B cell proliferation in It is evident from our studies that humans surviving legio response to either Hsp60 or OmpS for human lymphocytes nellosis develop cellular immune responses to Hsp60 while (data not presented) However since similar studies were not humans with no history of Legionnaires disease exhibit onlv perlormed with lymphocytes from guinea pigs we cannot rule marginal LPRs Moreover these individuals do not display out the possibility that B cell proliferation mav have contnb significant antibodv titers to Hsp60 as measured by ELISA Bv utcd to the observed proliferative responses with OmpS and immunoblot there was no difference in reactivity of patient HspfiO Blander and Horwitz demonstrated that L pneiimo serum over that of controls In a related studv we showed thai plula membrane preparations ostensibly rich in OmpS Hsp60 patients with insulin-dependent diabetes melhtus and other and LPS are able to induce cellular and protective immunity autoimmune disorders exhibited low LPRs to L pneumophila against experimental infection with L pneumophila (8) Thus Hspbl) (SI = 2 2 ± 10 manuscript submitted for publication) our hndings are consistent with earlier work done with less Despite the highlv conserved nature of the HSP60 class of specific antigens and suggest that OmpS and Hsp60 mav be the proteins we found no evidence to suggest that lymphocytes components of membrane material that are most important in trom tuberculin-positive individuals were capable ot prolifer elieitina protective immunity ating in response to L pneumophila Hsp60 protein We have The protection afforded by immunizing animals wilh HspfiO previously shown that the mycobacterial 65 kDa heat shock was much less impressive lhan that observed with OmpS Both protein and IIsp60 have substantial sequence similarity at the trials compared HspfiO immunization with OA immunization amino acid level particularly in regions previously shown to controls and demonstrated benefit onlv in terms of improved dehne T cell epitopes (2S) Lymphocytes from patients surviv bacterial clearance Survival weight loss and lung consolida ing lemonellosis exhibited strong proliferative responses to tion vvere not significantly different between the OA and HspfiO OmpS This was particularly true for individuals tested up to 8 groups These data indicate that Hsp60 immunization provides years posirecovery Two patients may not have had sufficient limited protection against challenge with L pneumophila but time to develop a cellular immune response to OmpS despite this protection is incomplete and does not affect long term exhibiting a response to Hsp6() Additional studies vvould be survival In evaluating proteins for potential vaccine efficacv in equired 'o assess whether HspbO stimulates lymphocytes earlv humans survivil must take precedence over protection meu in infection while OmpS affords long term protection These surcd in terms of bacterial clearance studies further suggest that steroid use may not adversely affect These results are in striking contrast to results recently cellular immunity Interestingly two cardiac transplant pa published bv Blander and Horwitz (10) In their studies guinea tients who survived legionellosis developed cellular immune pigs vaccinated on three occasions with HspfiO vvere protected responses to both Hsp6() and OmpS despite eyclosporme A agunst lethal eh dltnge (10) Moreover these investigators prophylaxis (51) For samples trom patients treated with tound that even low doses ot Hsp6() administered in three eyclosporme A it was necessary to incubate the lvmnhocvtes in injections lfrorded protective immunity Protection was pre RPMI 1640 for 24 h prior to antigen challenge presumably, sumed to be tell mediated based solely on DTH responses, this dilutes the inhibitory effects ot cyclosponne A This raises and no data on Ivmphocvte proliferation or on antibody levels the possibility that v iccination of individuals prior to surgerv were presented In evaluating protein antigens in this studv we migut prove elhcacious in providing long-term protection ot did not consider extending the immunization protocol to transplant patients on eyclosporme maintenance Additional include a third immunization because for OmpS a single studies will be required to tullv assess these preliminary immunization in Freund s incomplete idjuvant was sufficient observations 308 VOL 62, 1994 LEGIONELLA IMMUNE RESPONSES 3461 Early work by Horwitz and coworkers established that 2 Bellinger-Kawahara, C, and M A. Horwitz 1990 Complement survival and immunity to legionellosis are dependent on the component C3 fixes selectively to the major outer membrane mounting of a strong cellular immune response (12, 32) protein (MOMP) ot Legionella pneumophila and mediates phago Additional studies have indicated that neutrophils (16), natural cytosis of liposome-MOMP complexes by human monocytes J killer cells (4), and possibly gamma/delta T lymphocytes may Exp Med 172:1201-1210 3 Bhardwaj, N., T. W. Nash, and M. A Horwitz 1986 Interferon be involved in the complex host immune response to this gamma-activated human monocytes inhibit the intracellular mul intracellular parasite Moreover, it is generally believed that tiplication of Legionella pneumophila J Immunol 17 2662-2669 antibody responses to surface proteins of L pneumophila 4 Blanchard, D. K., H. Friedman, W. E. Stewart II, T. W. Mem, and facilitate phagocytosis by Fc receptor mediated endocytosis J. Y Djeu 1988 Role of gamma interferon in induction of natural (34) or by CRl and CR3 complement-mediated endocytosis killer activity by Legionella pneumophila in vitro and in an exper (2) Essentially, antibodv enhances the internalization of L imental murine infection model. Infect Immun 56 1187-1193 pneumophila but apparently does not lead to destruction of the 5 Blanchard, D K, W E. Stewart, T W Klein, H. Friedman, and bacteria within the phagosomes of macrophages Therefore, J. Y. Djeu. 1987 Cytolytic activity of human peripheral blood killing of infected macrophages mav be a primary route by leukocytes against Legionella pneumophila infected monocytes which the host survives legionellosis In this regard there has characterization of lhe effector cell and augmentation bv interleu- been much work on understanding how infected host cells are km2 J Immunol 139 551-556 targeted tor destruction It is known that virallv infected cells 6 Blander, S. J, R. F. Breiman, and M. A Horwitz 1989 A live often process host HSP60 to the cell surface (38) Host cells avirulent mutant Legionella pneumophila vaccine induces protective immunity against lethal aerosol ch lllenge J Clin Invest 83.810—S15 normally do not express HSP60 on the cell surface; those that 1 7 Blander, S. J., and M. A. Horwitz. 19S9 Vaccmalion with the do are targeted for destruction by CDS " T lymphocytes and major secretory protein of Legionella pneumophila induces cell- other cytotoxic lymphocytes (35, 38) Since virulent L pneu mediated and protective immunity in a guinea pig model ut mophila organisms can colonize and multiply in human mono Legionnaires disease J Exp Med 169 691-703 cytes in vitro it is likely that this process also occurs in vivo 8 Blander, S J.andM A Horwitz. 1991 Vaccination with Legitt Therefore the outcome of this relationship in humans may nella pneumophila membranes induces cell mediated and proiec- depend on the ability of the immune surveillance system to live immunirv in a guinea pig model of Legionnaires disease J identify and destroy infected host cells Antigen presentation Clin Invest 87 1054-1059 bv infected macrophages must plav a major role in determining 9 Blander, S. J., and M. A. Horwitz. 1991 Vaccination with the the outcome of infection In this regard vaccination of guinea major secretory protein of Legionella induces humoral and cell pigs with pure HspfiO may not mimic the natural infection in mediated immune responses and protective immunity across dif ferent serogroups of L pneumophila and different species ot terms ot the route ot antigen presentation It is known that L Legionella J Immunol 147:285-291 pneumophila Hsp60 is abundantly synthesized during intracel 10 Blander, S. J, and M. A Horwitz. 1993 Major cytoplasmic lular infection and immunogold cfectron microscopic obser membrane protein of Legionella pneumophila a genus eomrnon vations indicate that Hsp60 accumulates in the phagosomes ot antigen and member of the Hsp60 familv of heat shock proteins host cells (l, 10 23, 28. 30) Therefore, one would predict that induces protective immunity in a guinea pig model of Legion ample Hsp60 would be present early in infection and available naires disease J Clin Invest 91:717-723 tor antigen processing dnd presentation Antigens presented by 11 Blander, S J,S Szeto, H A Shuman, and M. A Horwitz. 1990 infected cells vvould be recognized bv CD4* and CD8+ T cells An immunoprotective molecule the mjjor secretory protein ot (5) as well as by natural killer cells, the latter have been Legionella pneumophila, is not a virulence taelor in a guinea pig implicated in eradication of Legionella-mfcaed host cells (4) model of Legionnaires disease J Clin Invest 86 817 824 OmpS on lhe other hand, is covalently bound to the pepti 12 Breiman R. F, and M A Horwitz. 1987 Guinea pigs sublethal^ doglvcan sacculus and cross-linked by interchain disulfide infected with aerosolized Legionella pneumophila develop humoral and cell-mediated immune responses and are protected against lethal aerosol bonds (13 31) In the native conformation, the OmpS ponn is challenge A model for stuctving host defense against lung infections resistant to the action of proteases (31). It is likely that the caused bv intracellular pathogens J Exp Med 165.799-811 eventual destruction of the bacteria within phagosomes leads 13 Butler, C. A., and P. S Hoffman 1990 Characterization of a major to antigen processing and presentation, possibly by class II 31-kilodalton peptidoglycan-bound protein of Legionella pneumo MHC molecules The factors which may favor the cellular plnle J Bacteriol 172.2401-2407 immune response to OmpS are not known but might include its 14 Buller. C. V, E. D. Street, T. P. Hatch, and P S Hoffman 1985 hvdrophobicity general resistance to both denaturation and Disulfide-bonded outer membrane proteins in the genus Legw the action of proteases and possibly the natural adjuvant nella Infect Immun 48 14-18 activity ot associated peptidoglycan fragments It will be im 15 Cianciotto, N. P., B. I. Eisenstein, C. Mody, and N. C Engleberg. portant in future studies to identify linear amino acid se 1990 A mutation in the mip gene results in attenuation ot quences of OmpS defining T-cell epitopes Such studies will be Legionella pneumophila virulence J Infect Dis 162 121-126 essential in assessing the efficacy of OmpS as a candidate 16 Dowling, J.N,A.K.Saha,andR H.Glcw 1992 Virulence factors vaccine for use in humans of the family Legionellaceae Microbiol Rev 56.32-60 17 Edelstein, P. H 1985 Legionnaires disease laboratory manual 3rd ed National Technical Information Service Springfield Va 18 Edelstein. P. H. 1993 Unpublished data ACKNOWLEDGMENTS 19 Edelstein, P H., K. Calarco, and V. K Yasui. 1984 Antimicrobial We are grateful lo Zunxuan Chen Jianjun Ren Martha Edelstein therapy of experimentally induced Legionnaires disease in guinea Joel Wcidenleld Rosemary Khuen and Linda Yales lor excellent pigs Am Rev Rcspir Dis 130 849-H56 technical assistance We thank Susan M Logan tor assistance in the 20 Ed'elstein, P, P S. Hoffman, M Edelstein, Z. Chen, and I. Ivmphocvte proliferation studies Weidenfeld 1991 Legionella pneumophila 60 kDa heat shock This work was supported by operating grants from the Medie.il protein in experimental Legionnaires disease Program Abstr Research Council of Canada lo P S H and to T J M 31st Intersci Conf Antimicrob Agents Chemother abstr 433 21 Engleberg, N C, C. Carter, D R. Weber, N. P Cianciotto, and REFERENCES B. I. Eisenstein 1989 DNA sequence of imp a Legionella 1 Abu Kwaik, Y, B I Eisenstein, and N. C Engleberg. 19°^ pneumophila gene associated with macrophage infecnvitv Infect Phenotvpic modulation by / egionella pneumophila upon infection Immun 57:1263-1270 of macrophages Infect Immun 61 1320-1329 22 Engleberg, N C, D C Howe J E Rogers, J Arroyo, and B I. 309 3462 WEERATNA ET AL. INFECT. IMMUN. Eisenstein. 1991. Characterization of a Legionella pneumophila ford, Jr., C. A. Kumamoto, and W. Wickner. 1989. Three pure iiene encodinu a lipoprotein untigcn. Mol. Microbiol. 5:2021-2029, chaperonc proteins of Escherichia coli—SecB, trigger faclor. and 23. Fernandez, £ C, M. Ripley, S. M. Logan. S. H. S. Lee, and P. S. GroEL—form soluble complexes with precursor proteins in vitro, Hoffman. Unpublished data. EMBO J. 8:2703-2709. 24. Gabay, J. E.. and M. A. Horwitz. 1985. Isolation and characteriza- 40. Ludwig, B., A. Schmid, R. Marrc, and j. Hacker, 1991, Cloning, tion of the cytoplasmic and outer membranes of lhe legionnaires' genetic analysis, and nucleotide sequence of a determinant coding disease- bacterium (Legionella pneumophila). J. Exp, Med. 161: for a 19-kiiodalton peptidoglycnn-associated protein (Ppl) of 4(19-422. Legionella pneumophila: Infect. Immun. 59:2515-2521. 25. Helsel, L. ()., VV, F. Bibb. C. A. Butler. P. S. Hoffman, and R. M. 41. Mustafa. A. S„ K. E. A. Lundin, and F. Oftung. 1993. Human T McKinney, 1988. Reeogniiion of u genus-wide antigen of Legio- cells recognize mycobacterial heal shock proteins in the context of nellu bv ii monoclonal antibody. Curr. Microbiol. 16:201-208. multiple HLA-DR molecules: studies with healthy subjecis vacci 26. Hindahl. M. S., and B. 11. Iglewski. 1984. Isolation and character nated with Mvcohiicienum bovis BCG and Mycohaciem.m leprae. ization ul the Legionella pneumophila outer membrane. J. Bacte Infect. Immun. 61:5294-5301. riol. 159:11)7-113. 42. Murminen. M., S. Butcher, I. Idanpaan, H. E. Wahlslrnm. S. 27. Hoffman. P. S„ C. A. Buller, and F. D. Quinn. 1989. Cloning and Multilumen. K. Runeberg-Njman. M. Sarvas, and P. H. Makela. temperature-dependent expression in Escherichia coli of a Legio 1992. The class I outer membrane protein ot Neisseria meningitidis nella pneumophila gene coding for a genus-common 60-kilodalton produced in Bacillus subtilis can give rise to protective immunity. antigen. Infect. Immun. 57:1731-1739. Mol. Microbiol. 6:2499-2506. 2.8, Huffman. P. S.. L, Housttm. and C. A. Buller. 1990. Legionella 43. Ormc. I. M.. P. Anderson, and W. 11. Boom. 1993. T cell response pneumophila hlpAB heat shock operon: nucleotide sequence and lo Mycobacterium tuberculosis. J. Infect. Dis. 167:1481-1497. expression of the Wl-kilodtilton antiuen in L. piieumoplulu-infeeieei HeLa cells, Inlecl. Immun. 58:3380-3387. 44. Orme. I. M.. E. S. Miller. A. D, Roberts, S. K. Furnev, J. P. Griflin. K. M. Dnlios. D. Chi, B. Rivoire. and P. J. Brennan. 1992. T 29. Hoffman. P. S.. M. Ripley, and R. Weeralna. 1992, Cloning and lymphocytes mediating protection and cellularcytolysis during the nucleotide sequence of a gene (ompS) encoding the major outer course of Mycobacterium tuberculosis infection. J. Immunol. 148: membrane prmein ol Legionella pneumophila. J. Bacteriol. 174: 189-196 914-920. 30. Hoffman. P. S„ M. Ripley, R. Weeralna. R. Fernandez. G. 45. OIL M.. P. Mcssncr, J. Hcesemunn, R. Marrc. and J. Hacker. Faulkner, 1). Hoskin. T. J. Marrie. S. M. Logan, and M. A. 1991, Temperature-dependent expression of flagclla in Legionella. Trevors. 1993. Legionella pneumophila surface proteins: gene J. Gen. Microbiol. 137:1955-1961. regulation and cellular immunity, p. "'4-78. In J. M. Barbaree. 46. Prosorovskii, S. V., I. S. Tartakovskii, W. Ehret. and T. A. R. F. Breiman. and A. P. Dufour (ed.). Legionella: current status Semenenko. 1991. Experimental approaches for prophylaxis of and emerging perspectives. American Society for Microbiology, legionellosis (abstract). Proceedings of the nth Meeting nl the Washington.D.C. European Working Group on Legionella Infections, Elsinore. 31. Hoffman. P. S.. J. II. Sever, and C. A. Buller. 1992. Molecular Denmark. characterization of the 28- and 31-kilodalton subunits of the 47. Quinn. K. D.. C. A. Butler, and P. S. Hoffman. 1987. Character Legionella pneumophila major inner membrane protein, J. Bacte ization and cloning of the disulfide-cross-linked major outer riol. 174:9014-913. membrane protein of Legionella pneumophila HI 16. J. Cell. Bio- 32. Hnrwilz. M. A. 1983. Cell-mediated immunity in Legionnaires chem. 115:116, disease. J. Clin. Invest. 71:1686-1697. 48. Sampson. J. S., B. B. Plikaytis, and H. W. Wilkinson. 19Sft, 33. Horwiiz. M. A., and S. C. Silverstein. 1983. Intracellular multipli Immunologic response of patients with legionellosis against major cation n| Legionnaires'disease bacteria (Legionella pneumophila) protein-containing antigens of Legionella pneumophila serogroup in human monocytes is reversibly inhibited by erythromycin and I as shown by immunoblot analysis. J. Clin. Microbiol. 23:92-99. rifampin. J. Clin. Invest. 71:15-26. 49. Shinnick. T. M„ M. H. Vodkin. and J. C. Williams. 1986. The 34. Ilusniann, I.. K.. and W. .Johnson. 1992. Adherence of Legionella 65-kilodalton antigen of Mycobacterium tuberculosis. J, Bacteriol. pneumophila to guinea pig peritoneal macrophages. J774 mouse 169:11)80-1088. macrophages, and undifferentiated U937 human monocytes: role 50. Tartakovskii. I. S.. I. K. Zubashev, and V. F. Koniukhnv. 1990. The of Fc and complement receptors. Infect. Immun. 60:5212-5218. protective properties ul' different Legionella antigens in experimen 35. Kaufmann. 8. 11. E. 1990, Heat shock proteins and lhe immune tal Legionella infection, Zh. Mikrobiol. Epidemiol. Immunobiol. response. Immunol. Today 11:129-136. 5:81-85. 36. Keen, ,M. <".., and P. S. "Hoffman. 1989. Characterization of a 51. Weeratna, R„ T. J. Marrie, S. M. Logan. D. Hoskin, P. S. Huffman. Legionella pneumophila extracellular protease exhibiting hemolytic L. Yates. S. liurbridge, D. Haldane, and O. Bezanson. 1993. and cvtotosic activities. Infect. Immun. 57:732-738. Legionnaires' disease in cardiac transplant patients: a cell-medi 37. Klein. T. W.. H. Friedman, and R. Widen. I9K4. Relative potency ated immune response develops despite cyclosporins therapy. J. of virulent versus avirulent Legionella pneumophila lor induction Infect. Dis. 168:521-522. of cell-mediaie-d immunity. Infect. Immun. 44:753-755. 52. Wilkinson. H. W., D. D. Cruece, and C. V. Broom. 1981. Validation 38. Kuan, T„ X W. Wurttenbcrger. J. DeBruyn, M. E. Munk. D. of Legionella pneumophila indirect immunofluorescence assay with Schncl, and S. II. E. Kaufmann. 1989. T cells against a bacterial epidemic sera. J, Clin. Microbiol. 13:139-146. heal shock protein recognize stressed macrophages. Science 245: 53. Young, D.. R. Lathigra, R. Hendrix, D. Sweelser. and R. A. Young. 1112-1115. 1988. Stress proteins are immune targets in leprosy and tubercu 39. Leckcr. S„ R. l.ill. T. Zieeellinrler. C. Georgopoulos. P. ,J. Bass- losis. Proc. Null. Acad. Sci. USA 85:4267-1270. I 310 JID 1993:168 (August) Correspondence 521 Legionnaires' Disease in Cardiac Transplant Patients: other half were incubated for 24 h in RPMI at 37CC in a humidi A Cell-Mediated Immune Response Develops despite fied C02 incubator. The incubated cells were washed twice with Cyclosporine Therapy PBS before plating and introduced into 96-well U-bottomed tis sue culture plates (Limbro; Flow). Antigens used in this study were MOMP (10 Mg/mL) and hsp60 (10 Vg/mL). As controls, Colleagues—Infection is the single most important complica cells were plated without antigen or with 10 Mg/mL concanava tion during the first 2 years after cardiac transplantation [1,2], lin A, ovalbumin, pokeweed mitogen (all Sigma, St. Louis), or with up to 76% of patients having at least one symptomatic in cell-free L, pneumophila extract. fection [1], The use of cyclosporine as part of the immunosup pression regimen for cardiac transplantation began in 1980 [2]. The plates were placed in a humidified C02 incubator and, on 3 This has led to a decrease in the cases of bacterial pneumonia, the appropriate day, were pulsed with 0.5 /iCi of [ H]thymidine aspergillosis, noncardiosis, bacteremia, and Pneumocystis carinti (ICN Radiochemicals, Mississauga, Canada) for 6 h. The pulsed infection compared with conventional immunosuppression [2], wells were washed, and cells were harvested on filter mats (Ti- However, Legionella infections have increased in incidence from tertek. Flow) using a multisample automated cell harvester 2% (I /49) to 5% (4/77) [2], This is likely a true finding, not just a (Skatron AS; Flow). Filter disks were placed into scintillation reflection of a local outbreak of legionellosis. as many cardiac vials, and radioactivity was measured in a liquid scintillation transplant centers have experienced such infections [3-6]. Dur counter. All determinations were done in triplicate, and the stim ing investigation of 2 male cardiac transplant patients (54 and ulation indices (Sis) were calculated from the means by dividing 53 years old) with severe pneumonia due to Legionella pneumo the counts per minute obtained with lymphocytes in the pres phila serogroup 1, we found that despite cyclosponne immuno ence of antigen by the counts per minute obtained with lympho suppression the lymphocytes clearly recognized L pneumophila cytes in medium alone. antigens Both patients developed an antibody response to Legionella A 28-kDa major outer-membrane protein (MOMP) and heat- surface antigens as measured by IFA (< 1:64-1:512 for patient shock protein hsp60 of L. pneumophila Philadelphia I [7] were 1) Patient 1 did not develop an antibody response to either prepared as previously descnbed [7-10]. Antibody responses to hsp60 or MOMP. Patient 2 developed an antibody titer of L pneumophila surface antigens were determined using an indi 1:16,384 to hsp60. There was no significant antibody titer to rect immunofluorescence assay (IFA), and antibody titers to MOMP MOMP and hsp60 were determined using an ELISA [10]. Cells frcm patient I were examined on five occasions after the Peripheral blood (20-40 mL) was obtained from patients and onset of infection. There was no lymphocyte response to conca volunteers (healthy subjects and 2 renal transplant patients who navalin A over an 8-day penod. indicating that the lymphocytes were receiving cyclosponne), and the lymphocyte fraction was were immunosuppressed as a result of the cyclosponne therapy. recovered after centrifugation on a ficoll-hypaque gradient However, proliferative responses to MOMP and to hsp60 gradu (Row Laboratones, McLean. VA). Lymphocytes were adjusted ally increased and were significant at days 6 and 8 The steady lo a final concentration of 2.5 X 10s cells/mL in RPMI 1640 increase in the SI for these antigens would be consistent with containing penicillin (100 units/mL). streptomycin (100 Mg/ lymphocytes initiating a primary response to the antigen By 4 mL), 10 mAf L-glutamine, and 5^ human AB serum Half of the months, the immune response had shifted to a secondary re cells were plated out immediately as descnbed below, and the sponse, as indicated by a peak in SI on day 6 for MOMP but not hsp60 We considered the possibility that cyclosponne might be in hibiting lymphocytes from responding until the latter days of these expenments. We decided to incubate lymphocytes for 24 Reprints or correspondence. Dr T J. Marne. Room 5014 ACC. 1278 h to permit intracellular cyclosponne levels to decrease before Tower Rd Halifax. NS. B3H 2Y9 Canada. antigen challenge In these experiments, incubated patient lym The Journal of Infectious Disuses 1993.168.521-2 phocytes demonstrated a substantially increased proliferative re © 1993 bv The University of Chicago All ngbts reserved 0022-1899/93/6802-0052S01 00 sponse to L pneumophila antigens compared with unincubated Table 1. Proliferative response of incubated (I) and unincubated (Ul) lymphocytes from patient 1 at 5 months after the onset of infection Stimulation index Day 2 Day 3 Day 6 Dav8 Antigen Ul I Ul 1 Ul 1 Ul I Concanavalin A (5 ^ig/mL) 29 54 23 79 07 33 06 06 Major outer-membrane protein (IO^g/mL) 1 1 1 2 1.3 33 1.5 143 1.5 20 hsp60( 10 |ig/mL) 09 1 4 06 1 9 07 43 1 2 1.4 Ovalbumin (10 ug/m L) 07 1.0 07 09 2 1 ND 34 1 1 NOTE Stimulations indices were computed from means of inplicaie assays ND. noi determined I 4 311 522 Correspondence JID 1993,168 (Augusl) cells Conlrol lymphocytes (lymphocytes from healthy, unin 2 Hofflin JM Polasman I, Baldwin JC Oyer PE Stinson EB Remington fected persons) did not show such a response At 5 months after JS lnlecnous complications in heart transplant recipients receiving infection, patient I was again tested for lymphocyte responsive cyciosponn and corticosteroids Ann Intern Med 1987,106 209-16 ness, and in this experiment, incubated and unincubated lym 3 Copeland J, Wicden M, Fineberg W, Salomon N, Hagger D, Galgiam J phocytes were challenged with antigen Incubated lymphocytes Legionnaires' disease following cardiac transplantation Chest exhibited a greater proliferative response than did the unincu 1981:79 669-71 bated lymphocytes (table I) Again, the maximal response was 4 Fuller J Levinson MM Kline JR. Copeland J Legionnaires' disease observed at day 6 for both MOMP and hsp60 A substantial after heart transplantation An" Thorac Surg 1985 39 308-11 response of the incubated lymphocytes to MOMP was noted on 5 Redd SC Schuster DM Quan ' "elikaytis VD, Spika JS, Cohen ML day 6 (SI = 14 vs 15 for unincubated lymphocytes) Similar Legionellosis in cardiac transplant recipients Results of nationwide results were noted 5 months later (data not presented) survey J Infect Dis 1988 158 651-3 Lymphocytes from patient 2 were first examined on day 14 6 Horbach I Fehrembech FJ Legionellosts in heart transplant recipients after onset of infection While the initial response appeared to Infection 1990 18 361-3 be primary, by 10 weeks after infection, responses in patient 2 7 Hoffman PS, Butler CA Quinn A Cloning and lemperature-oependent had shifted to secondary expression in Escherichia coli of a Legionella pneumophila gene cod mg far a genus-common 60-kilodalton antigen Infect Immun Risini Weeratna, Thomas J. Marrie, Susan M. Logan, 1989 57 1731-9 David Hoskin, Paul S. Hoffman, Linda Yates, 8 Buller CA Hoffman PS Charactenzation of a major 31 kilodalton pep Susan Burbridge, David Haldane, and Gregory Bezanson tidoglycan bound protein of Legionella pneumophila J Bactenol Departments of Microbiology and Medicine Dalhousie Unnersitx and 1990 172 2401-7 Victoria General Hospital Halifax. Canada 9 Laemmli UK Cleavage of structural proteins dunng assembly of the References headofbactenophageT4 Nature 1970 227 680-3 I Dummer JS Bahnson HT Griffith BP Hardesty RL Thompson ME 10 Helsel LO BibbWF ButlerCA Hoffman PS McKinneyRM Recogni Ho M Infections in patients on cyciosponn and prednisone follow tion of a genus wide antigen ofLegionella bv a monoclonal antibody ing cardiac transplantation Transplant Proc 1982,l5(suppl 1)2779-81 Curr Microbiol 1988 16 201-8 ^ BIBLIOGRAPHY Abu Kwaik, Y., B.I. Eisenstein, and N.C. Engleberg. 1993. Phenotypic modulation of Legionella pneumophila upon infection of macrophages. Infec. Immun. 61:1320-1329. Abu Kwaik, Y., and N.C. Engleberg. 1994. Cloning and molecular characterization of a Legionella pneumophila gene induced by intracellular infection and by various in vitro stress conditions. Mol. Microbiol. 13:243-251. Adams, D.O., and T.A. Hamilton. 1984. The cell biology of macrophage activation. Ann. Rev. Immunol. 2:283-318. Aiba, H., F. Nakasai, S. Mizushima, and T. Mizuno. 1989. Phosphorylation of a bacterial activator protein, OmpR, by a protein kinase, EnvZ, results in a stimulation of its DNA binding ability. J. Biochem. 106:5-7. Akporiaye, E.T., J.D. Rowatt, A.A. Aragon, and O.G. Baca. 1983. Lysosomal response of a murine macrophege-like cell line persistently infected with Coxiella burnetii. Infect. Immun. 40:1155-1162. Albano, M.A. , J. Arroyo, B.I. Eisenstein, and N.C. Engleberg. 1992. PhoA gene fusions in Legionella pneumophila generated in vivo using a new transposon, MudphoA. Mol Microbiol. 6:1829-1839. Amanos, K.I., and J.C. Williams. 1983 (a). Peptidoglycan of Legionella pneumophila: appearent resistance to lysozyme hydrolysis correlates with a high degree of peptide cross-linking. J. Bacteriol. 153:520-526. Amanos, K.I., and J.C. Williams. 1983 (b). Partial characterization of peptidoglycan-associated proteins of Legionella pnumophila. J. Biochem. 94:601-606. Aland, CM., A.R. Skinner, A. Malic, and J.B. Kurtz. 1983. Interaction of L. pneumophila and a free-living amoeba (AcanthaiTioe-jba palestinensis) . J. Hygiene. 91:167-178. 312 I 313 Arakawa, M., Y. Hashimoto, T. Ezaki, and E. Yabuuchi. 1992. Virulence of Legionella species does not correlate with their sodium chloride resistance. Med. Microbiol. Lett. 1: 30-37. Armstrong, J.A. and P. Hart D'Arcy. 1971. Response of cultured macrophages to Mycobacterium tuberculosis with observations of fusion of lysosomes with phagosomes. J. Exp. Med. 134:713-740. Armstrong, S., and C. Parker. 1986. Surface proteins of Bordetella pertussis: comparision of virulent and avirulent strains and effects of phenotypic modulation. Infect. Immun. 54:308-314. Arroyo, J., B.I. Eisenstein, and N.C. Engleberg. 1993. Construction of phoA gene fusions in Legionella pneumophila that attenuate intracellular infection. In, J.M. Barbaree, R.F. Breiman, and A.P. Dufour (ed.) Legionella: Current t,catus and emerging perspectives. ASM, Washington DC Arroyo, J., M.C. Hurley, M. Wolf, M.S. McClain, B.I. Eisenstein, and N.C. Engleberg. 1994. Shuttle mutagenesis of Legionella pneumophila: identification of a gene associated with host cell cytopathicity. Infect. Immun. 62:4075-4080. Augustin, L.B., B.A. Jacobson, and J.A. Fuchs. 1994. Escherichia coil Fis and DnaA proteins bind specifically to the nrd promoter region and affect expression of a nrd-lac fusion. J. Bacteriol. 176:378-387. Baine, W.B., and J.K. Rasheed. 1979. Aromatic substrate specificity of browning by cultures of the Legionnaires' disease bacterium. Ann. Intern. Med. 9: 619- 620. Baine, W.B. 1988. A phospholipase C from the Dalla IE strain of Legionella pneumophila serogroup5: purification and characterization of conditions for optimal activity with an artificial substrate. J. Gen. Microbiol. 134:489-498. Balke, V.L., and J.D. Gralla. 1987. Changes in linking number of supercoiled DNA accompany growth transitions in 314 Escherichia coli. J. Bacteriol. 169:4499-4506. Bangsborg, J.M., N.P. Cianciotto, and P. Hendersson. 1991. Nucleotide sequence analysis of the Legionella micdadei mip gene, encoding a 30-kDa analog of the Legionella pneumophila Mip protein. Infect. Immun. 59:3836- 3840. Barbaree, J.M., B.S. Fields, J.C. Feeley, G.W. Gorman, and W.T. Martin. 198 6. Isolation of protozoa from water associated with a legionellosis outbreak and demonstration of intracellular multiplication of Legionella pneumophila. Appl. Environ. Microbiol. 51:422-424. Barlow, A.K., J.E. Heckles, and I.N. Clarke. 1987. Molecular cloning and expression of Neisseria meningitidis class 1 outer membrane protein in Escherichia coli K-12. Infect. Immun. 55:2734-2740. Baskerville, A., J.W. Conlan, L.A.E. Ashworth, and A.B. Dowsett. 1986. Pulmonary damage caused by a protease from Legionella pneumophila. Br. J. Exp. Path. 67:527-536. Bavoil, P., A. Ohlin, and J. Schachter. 1.984. Role of disulfide bonding in outer membrane structure and permeability in Chamydia trachomatis. Infect. Immun. 44:479- 485. Belyi, Y.F., I.S. Tartakovskii, Y.V. Vertiev, and S.V. Prosorovskii. 1991. ADP-ribosyltransferase activity of Legionella pneumophila is stimulated by the presence of macrophage lysates. Biomed. Sci. 2:94-96. Belyi, Y.F. 1993. Intracellular parasitism of Legionella and signalling in eukaryotic cells. FASEB. 7:1011-1015. Bellinger-Kawahara, C, and M.A. Horwitz. 1990. Complemet components C3 fixes selectively to the major outer membrane protein (MOMP) of Legionella pneumophila and mediates phagocytosis of liposome-MOMP complexes by human monocytes. J.Exp. Med. 172:1201-1210. I I 315 Bennett, J.S. 1978. Legionnaires' disease in Canada. Can. Med.Assoc. J. 118:1031-1032. Berg, J.D., J.C. Hoff, P.V. Roberts, and A. Katin. 1985. Growth of Legionella pneumophila in continuous culture. Appl. Environ. Microbiol. 49: 1534-1537. Berger, K.H., and R.R. Isberg. 1993. Two distinct defects in intracellular growth complemented by a single genetic locus in Legionella pneumophila. Mol. Microbiol. 7:7-19. Berger, K.H., J.J. Merriam, and R.R. Isberg. 1994. Altered intracellular targeting properules associated with mutations in the Legionella pneumophila dotA gene. Mol. Microbiol. 14:809-822. Bhardwaj, N., T. Nash, and M.A. Horwitz. 1986. Gamma interferon- activated human monocytes inhibit the intracellular multiplication of Legionella pneumophila. J. Immunol. 137:2662-2664. Black, W.J., F.D. Quinn, and L.S. Tompkins. 1990. Legionella pneumophila zinc metalloprotease is structurally and functionally homologous to Pseudomonas aeruginosa elastase. J. Bacteriol. 172: 2608-2613. Blackwell, J.M., R.A.B. Ezekowitz, M.B. Roberts, J.Y. Channon, R.B. Sim, and S. Gordan. 1985. Macrophage complement and lectin-like receptors bind Leishmania in the absence of serum. J. Exp. Med. 162:324-327. Blanchard, D.K., J.Y. Djeu, T.W. Klein, H. Friedman, and W.E. Stewart. 1987. Induction of tumor necrosis factor by Legionella pneumophila. Infect. Immnun. 55:433-437. Blanchard, D.K., H. Friedman, W.E. Stewart II, T.W. Klein, and J. Djeu. 1988. Role of gamma interferon in induction of natural killer activity by Legionella pneumophila invitro and in an experimental murine infection model. Infec. Immun. 56:1187-1193. Blander, S.J., and M.A. Horwitz. 1989. Vaccination 316 with the major secretory protein of Legionella pneumophila induces cell-mediated and protective immunity in a guinea pig model of Legionnaires' disease. J. Exp. Med. 169: 691- 705. Blander, S.J., L. Szeto, H.A. Shuman, and M.A. Horwitz. 1990. An immunoprotective molecule, the major secretory protein of Legionella pneumophila, is not a virulence factor in a guinea pig model of Legionnaires' disease. J. Clin. Invest. 86:817-824. Blander, S.J., and M.A. Horwitz. 1991a. Vaccintaion with the major secretory protein of Legionella pneumophila induces cell-mediated and protective immunity in a guinea pig model of legionnaires' disease. J. Exp. Med. 169:691- 705. Blander, S.J., and M.A. Horwitz. 1991b. Vaccination with Legionella pneumophila membranes induces cell-mediated and protective immunity in a guinea pig model of Legionnaires' disease. J. Clin Invest. 87:1054-1059. Blander S.J., and M.A. Horwitz. 1993. Major cytoplasmic membrane protein of Legionella pneumophila, a genus common antigen and membar of the hsp 60 family of heat shock proteins, induces a protective immunity in a guinea pig model of legionnaires' disease. J. Clin. Invest. 91:717-723. Blatt, S.P., M.D. Parkinson, E. Pace, P. Hoffman, D. Dolan, P. Lauderdale, R.A. Zajac, and G.P. Melcher. Nosocomial legionellosis: a one-year case control and environmental investigation, p.47-50. In J.M. Barbaree, R.F. Breiman, and A.P. Dufour (ed.), Legionella: Current status and emerging perspectives. 1993. American Society for Microbiology, Washington D.C. Bohach, G.E., and I.S. Snyder. 1983. Characterization of surfaces involved in adherance of Legionella pneumophila to Fischerella species. Infect. Immun. 42:318-325. Bornstein, N., D. Marmet, M.H. Dumaine, M. Surgot, and J. Fleurette. 1991. Detection of flagella in 278 Legionella strains by latex reagent sentisized with antiflagellum I I 317 immunoglobulins. J. Clin. Microbiol. 29:953-960. Bouza, E., and M. Rodriguez-Creixems. 1984. Legionnaires disease in Spain. In Legionella: Proceedings of the 2nd International symposium, p.- 15-17. American Society for Microbiology, Washington D.C. Brand, B.C., A.B. Sadosky, and H.A. Shuman. 1994. The Legionella pneumophila icm locus: a set of genes required for intracellular multiplication in human macrophages. Mol Microbiol. 14:797-808. Braun, R. 1988. Molecular and cellular biology of malaria. Bioessays. 8:194-199. Breiman, R.F., and Horwitz. 1987. Guinea pigs sublethally infected with aerosolized Legionella pneumophila develop humoral and cell mediated immune responses and are protected against lethal aerosol challenge. A model for studying host defense against lung infections caused by intracellular pathogens. J. Exp. Med. 165:799-811. Brenner, D.J., J.C. Feeley, and R.E. Weaver. 1979. Legionellaceae. p. 279-288. In N.R. Kreieg, J.G. Holt (ed), Bergey's manual of systematic bacteriology. Volume 1. 1984. Williams & Wilkins, Baltimore, MD. Broome, C.V., S.A.J. Goings, S.B. Thacker, R.L. Vogt, H.N. Beaty, D.W. Fraser, and the Field Investigating Team. 1979. The Vermont epidemic of Legionnaires' disease. Ann. Intern. Med. 90:573-577. Brown, A., R.M. Vickers, M. Elder, M. Lema, and G.M. Garrity. 1982. Plasmids and surface antigen markers of endemic and epidemic Legionella pneumophila strains. J. Clin. Microbiol. 16:230-235. Brown, K.L., and K.T. Hughes. 1995. The role of anti- sigma factors in gene regulation. Mol. Microbiol. 16:397- 404. Butler, C.A., E.D. Street, T.P. Hatch, and P.S. Hoffman. 1985. Disulfide-bonded outer membrane proteins in 318 the genus Legionella. Infect. Immun. 48:14-18. Butler, C.A., and P.S. Hoffman. 1990. Characterization of a major 31-kilodalton peptidoglycan-bound protein of Legionella pneumophila. J. Bacteriol. 172:2401-2407. Broyles, S.S., and D.E. Pettijohn. 1986. Interaction of the Escherichia coli HU protein with DNA. Evidence for formation of nucleosome-like structures with altered DNA helical pitch. J. Mol. Biol. 187:47-60. Byrd, T.F., and M.A. Horwitz. 1989. Interferon gamma- activated human monocytes downregulate transferrin receptors and inhibit the intracellular multiplication of Legionella pneumophila by limiting the availability of iron. J. Clin. Invest. 83:1457-1465. Caldwell, II.D., J. Kromhout, and J. Schachter. 1981. Purification and partial characterization of the major outer membrane protein of Chlamydia trachomatis. Infect. Immun. 31:1161-1176. Calvo, J.M., and R.G. Matthews. 1994. The leucine- responsive regulatory protein, a global regulator of metabolism in Escherichia coli. Microbiol. Rev. 58:466-490. Cannon, R.B. Sim, and S. Gordan. 1985. Macrophage complemet and lectin-like receptors bind Leishmania in the absence of serum. J. Exp. Med. 162:324. Carroll, M.E.W., P.S. Jackett, V.R. Aber, andD.B. Lowrie. 1979. Phagolysosome formation. Cyclic adenosine 3':5' monophosphate and the fate of Salmonella typhimurium within mouse peritoneal macrophages. J. Gen. Microbiol. 110:421-429. Catrenich, C.E., and W. Johnson. 1988. Virulence conversion of Legionella pneumophila: a one-way phenomenon. Infect. Immun. 56:3121-3125. Catrenich, C.E., and W. Johnson. 198 9. Characterization of the selective inhibition of growth of virulent Legionella pneumophila by Supplemented Mueller- I 319 Hinton medium. Infect. Immun. 57: 1862-1864. •*» Chandler, F.W., B.M. Thomason, and G.A. Hebert. 1980. Flagella on Legionnaires' disease bacteria in the human lung. Ann. Intern. Med. 93:715-716. Chang, K.P. 1979. Leishmania donovani promastigote- macrophage surface interactions in vitro. Exp. Parasitol. 48:175-89. Chen, G.C.C., M. Lema, and A. Brown. 1984. Plasmid transfer into members of the family Legionellaceae. J. Infect. Dis. 150:513-516. Chow, W.Y., and D.E. Berg 1988. TnStacl, a derivative of transposon Tn5 that generates conditional mutations. Proc. Natl. Sci. USA 85:6468-6472. Cianciotto, N.P. J.M. Bangsborg, B.I. Eisenstein, and N.C. Engleberg. 1988. Site-specific mutagenesis of Legionella pneumophila by allelic exchange using counterselectable ColEl vectors. FEMS Microbiol. Lett. 56:203-208. Cianciotto, N.P., B.I. Eisenstein, C.H. Mody, G.B. Toews, and N.C. Engleberg. 1989. A Legionella pneumophila gene encoding a species-specific surface protein potentiates initiation of intracellular infection. Infect. Immun. 57:1255-1262. Cianciotto, N.P., J.M. Bangsborg, B.I. Eisenstein, and N.C. Engleberg. 1990. Identification of mip-like genes in the genus Legionella. Infect. Immun. 58:2912-2918. Cianciotto, N.P., andB.S. Fields. 1992. Legionella pneumophila mip gene potentiates intracellulai infection of protozoa and human macrophages. Proc. Natl. Acad. Sci. USA. 89:5188-5191. Cohen, I.R., and D.B. Young. 1991. Autoimmunity, microbial immunity and the immunological homunculus. Immunol. Today 12:105-110. • I 320 Comeau, D.E., K. Ikenaka, K. Tsung, and M. Imouye. 1985. Primary characterization of the protein products of the Escherichia coli ompB locus: structure and regulation of synthesis of the OmpR and EnvZ proteins. J. Bacteriol. 164:578-584. Compton, G.A., J.M. Davis, J.R. Macdonald,a nd H.P. Bachinger. 1992. Structural and functional characterization of Escherichia coli peptidyl-prolyl-cis-trans isomerase. J. Biochem. 206:927-934. Conlan, J.W., A. Baskerville, and L.A.E. Ashworth. 1986. Separation of Legionella pneumophila prtease and purification of a protease which produces lesions like those of Legionnaires' disease in guinea pig lung. J. Gen. Microbiol. 132:481-487. Cowan, S.W., T. Schirmer, G. Rummel, M. Steiert, R. Ghosh, R.A. Pauptit, J.N. Jansonius, and J.P. Rosenbusch. 1992. Crystal structure explain functional properties of two E. coli porins. Nature 358:727-733. Craig, N. and H. Nash. 1984. E. coli intergration host factor binds to specific sites in DNA. Cell. 39:707-716. Dannnenberg, A.M.Jr. 1989. Immune mechanisms in the pathogenesis of pulmonary tuberculosis. Rev. Infect. Dis. 11:S369-S378. Dascher, C., D. Roll, and P.M. Bavoil. 1993. Expression and translocation of the chlamydial major outer membrane protein in Escherichia coli. Microb. Pathog. 15:455-467. DeretLc, V., R. Dikshit, W.M. Konyecsni, A.M. Chakrabarty, and T.K. Misra. 1989. The algR gene, which regulates mucoidy in Pseudomonas aeruginosa, belongs to a class of environmentally responsive genes. J. Bacteriol. 171:1278-1283. Dirita, V.J., K.M. Peterson, and J.J. Mekalanos. 1990. Regulation of cholera toxin synthesis. In B.H. Iglewski and V.L. Clark (ed.) Molecular basis of bacterial pathogenesis, pp.355-376. Academic Press, San Diego. I 321 DiRita, V.J., and J.J. Mekalanos. 1991. Periplasmic interactions between two membrane regulatory proteins, ToxR and ToxS, results in signal transduction and transcriptional activation. Cell. 64:29-37. DiRita, V.J., and J.J. Mekalanos. 1992. Coordinate expression of virulence genes by ToxR in Vibrio cholerae. Mol. Microbiol. 6:451-458. Dondero, T.J., Jr., R.C. Rendtorff, G.F. Mallison, R.M. Weeks, J.S. Levy, E.W. Wong, and W. Schaffner. 1980. An outbreak of Legionnaires' disease associated with a contaminated air-conditioning cooling tower. N. Engl. J. Med. 302:365-370. Dorman, C.J., and C.F. Higgins. 1987. Fimbriae phase variation in Escherichia colidependence on intergration host factor and homologies with other site specific recombinases. J. Bacteriol. 169:3840-3843. Dorman, C.J., N. NiBhriain, and C.F. Higgins. 1990. DNA supercoiling and environmental regulation of virulence gene expression in Shigella flexneri. Nature 344:789-792. Dowling, J.N., A.K. Saha, and R.H. Glew. Virulence factors of family Legionellaceae. Microbiol. Rev. 56:32-60. Drebot, M.A., and S.H.S. Lee. 1994. Direct sequencing of double-stranded PCR products gel purified by centrifugation through blotting paper. Biotechniques. 17:248-250. Dreyfus, L.A. , and B.H. Iglewski. 1985. Conjugation- mediated genetic exchange in Legionella pneumophila. J. Bacteriol. 161:80-84. Dreyfus, L.A., and B.H. Iglewski. 1986. Purification and characterization of an extracellular protease of Legionella pneumophila. Infect. Immun. 51:736-743. Edelstein, P.H. 1981. Improved semiselective medium for isolation of Legionella pneumophila from contaminated clinical and environmental specimens. J. Clin. Microbiol. ! J 322 14:298-303. Edelsteine, P.H. 1982. Comparative study of selective media for isolation of Legionella pneumophila from potable water. J. Clin. Microbiol. 16:697-699. Edelstein, P.H., P.S. Hoffman, M. Edelstein, Z. Chen, and J. Weidenfeld. 1991. Legionella pneumophila 60 kDa heat shock protein in experimental Legionnaires' disease. Program Abstr. 31st Intersci. Conf. Antimicob. Agents Chemother., abstr. 433. Ehret, W., G. Anding, and G. Ruckdeschel. 1984. Characterization of membrane proteins from various strains and serogroups of Legionella pneumophila and other Legionella species. In, Legionella: Proceedings of the 2nd international symposium (C. Thornsberry, A. Balows, J.C. Feeley, and W.Jakubowski, eds), American Society for Microbiology, Washington, D.C., pp265-268. Eisenstein, B.I., D.S. Sweet, V. Vaughn, and D.I. Friedman. 1987. Intergration host factor is required for the DNA inversion that controls phase variation in Escherichia coli. Proc. Natl. Acad. Sci. USA 84:6506-6510. Engelberg, N.C, D.J. Drutz, and B.I. Eisenstein. 1984. Cloning and expression of Legionella pneumophila antigens in Escherichia coli. Infect. Immun. 44:222-227. Engleberg, N.C, E. Pearlman, D. Dixon, and B.I. Eisenstein. 1986. Antibodies isolated by using cloned surface antigens recognize antigenically related components of Legionella pneumophila and other Legionella species. J. Immunol. 136:1415-1417. Engleberg, N.C, C Carter, D.R. Weber, N.P. Cianciotto, and B.I. Eisenstein. 198 9. DNA sequence of mip, a Legionella pneumophila gene associated with macrophage infectivity. Infect. Immun. 57:1263-1270. Engleberg, N.S., D.C. Howe, J.E. Rodgers, J. Arroyo, and B.I. Eisenstein. 1991. Characterization of a Legionella pneumophila gene encoding a lipoprotein antigen. Mol. 323 Microbiol. 5:2021-2029. Engleberg, N.C. 1993. Genetic studies of Legionella pathogenesis. p. 63-68. In J.M. Barbaree, R.F. Breiman, and A.P. Dufour (ed.), Legionella: Current status and emerging perspectives. American Society for Microbiology, Washington D.C Ernsting, B.R., M.R. Atkinson, A.J. Ninfa, and R.G. Matthews. 1992. Characterization of the regulon controlled by the leucine responsive regulatory protein in Escherichia coli. J. Bacteriol 174:1109-1118. Fang,F.C, S.J. Libby, N.A. Buchmeier, P.C. Lowen, J. Switala, J. Harwood, and D.G. Guiney. 1992. The alternative a factor KatF (RpoS) regulates Salmonella virulence. Proc. Natl. Acad. Sci. USA. 89:11978-11982. Feeley, J.C, G.W. Gorman, R.E. Weaver, D.C Mackel, and H.W. Smith. 1978. Primary isolation media for Legionnaires' disease bacterium. J. Clin. Microbiol. 8:320- 325. Feeley, J.C, R.J. Gibson, G.W. Gorman, N.C Langford, J.K. Rasheed, D.C Mackel, and W.B. Baine. 1979. Charcol yeast extract agar: primary isolation medium for Legionella pneumophila. J. Clin. Microbiol. 10:437-441. Fernandez, R.C. 1992. Investigations into the intracellular pathogenesis of Legionella pneumophila. Ph.D. Thesis, Dept. Microbiology and Immunology, Dalhousie University, Halifax, NS. Ferrario, M., B.R. Ernsting, D.W. Borst, D.E. Wiese II, R.M. Blumenthal, and R.G. Matthews. 1995. The leucine- responsive regulatory protein of Escherichia coli negatively reg1 ^tes transcription of ompC and micF and positively reg ,iates translation of ompF. J. Bacteriol. 177:103-113. Fields, B.S., E.B. Shotts, J.C. Feeley, G.W. Gorman, and W.T. Martin. 1984. Proliferation of Legionella pneumophila as an intracellular parasite of the ciliated protozoan Tetrahymena pyriformis. In Legionella, Proceedings 324 of the 2nd international symposium, ed. C. Thornsberry, A. Barlow, J.C. Feeley, and W. Jakubowski. pp. 327-328. ASM, Washington DC Flesher, A.R., S. Ito, B.J. Mansheim, and D.L. Kasper. 1979. The cell envelope of the Legionnaire's disease bacterium: morphologic and biochemical characteristics. Ann. Intern. Med. 90:628-630. Fliermans, C.W., W.B. Cherry, L.H. Orrison, S.J. Smith, D.L. Tison, and D.H. Pope. 1981. Ecological distribution of Legionella pneumophila. Appl. Environ. Microbiol. 41:9-16. Fischer, G., H. Bang, B. Ludwig, K. Mann and J. Hacker. 1992. Mip protein of Legionella pneumophila exhibits peptidyl-prolyl-cic/trans isomerase (Pplase) activity. Mol. Microbiol. 6:1375-1383. Freundlich, M., N.Ramani, E. Mathew, A. Sirko, and P. Tsui. 1992. The role of intergration host factor in gene expression in Escherichia coli. Mol. Microbiol. 6:2557-2563. Friedman, D.I. 1988. Intergration host factor: a protein for all reasons. Cell. 55:545-554. Friis, R.R. 1972. Interaction of L cells and Chlamydia psittaci:entry of the parasite and host response to its development. J. Bacteriol. 180:706-721. Gabay, J.E., M. Blake, W.D. Niles, and M.A. Horwitz. 1985. Purification of Legionellapneumophila major outer membrane protein and demonstration that it is a porin. 162:85-91. Galan, J.E., and R. Curtiss III. 1990. Expression of Salmonella typhimurium genes required for invasion is regulated by changes in DNA supercoiling. Infect. Immun. 58:1879-1885. Galas, J.G., and A. Schmitz. 1978. DNAse footprinting: a simple method for the detection of protein-DNA binding specificity.. Nucleic acid Res. 5:3157-3170. 325 Geoffroy, C, J.L. Gaillard, J.E. Alouf, and P. Berche. 1987. Purification, characterization, and toxicity of the sulfhydryl-activated hemolysin 0 from Listeria monocytogenes. Infect. Immun. 55:1641-1646. George, J.R., L. Pine, M.W. Reeves,and W.K. Harrell. 1980. Amino acid requirments of Legionella pneumophila. J. Clin. Microbiol. 11:286-291. Gibson, F.C., A.O. Tzianabos, and F.G. Rodgers. 1994. Adherence of Legionella pneumophila to U-937 cells, guinea pig alveolar macrophages, and MRC-5 cells by a novel, complement-independent binding mechanism. Can. J. Microbiol. 40:865-872. Good, M.F., J.A. Berzofsky, and L.H. Miller. 1988. The T cell response to the malaria circumsporozoite protein: an immunological approach to vaccine design. Anriu. Rev. Immunol. 6:663-688. Good, M.F. 1992. A malaria vaccine strategy based on the induction of cellular immunity. Immunol. Today. 13:126- 129. Goodman, S.D., H. Nash. 1989. Functional replacement of a protein-induced bend in a DNA recombination site. Nature 341:251-254. Gosting, L.H., K. Cabrian, J.C. Sturge, and L.C. Goldstein. 1984. Identification of a species-specific antigen in Legionella pneumophila by a monoclonal antibody. J. Clin. Microbiol. 20:1031-1035. Gotschlich, E.C, M.E. Seiff, M.S. Blake, and M. Koomey. 1987. Porin protein of Neisseria gonorrhoeae: cloning and gene structure. Proc. Natl. Acad. Sci. USA. 84:8135-8139. Hackstadt, T., W.J. Todd, and H.D. Caldwell. 1985. Disulfide mediated interactions of the chlamydial major outer membrane protein: role in the differentiation of clamydiae? J. Bacteriol. 161:25-31. 326 Haley, C.E., M.L. Cohen, J. Halter, and R.D. Meyer. 1979. Nosocomial Legionnaires' disease: a continuing common- source epidemic at Wadsworth Medical Center. Ann. Intern. Med. 90:583-586. Hall, M.N., and T.J. Silhavy. 1981. Genetic analysis of the ompB locus in Escherichia coli. Annu. Rev. Genet. 15:91- 142. Hancock, R.E.W. 1987. Role of porins in outer membrane permeability. J. Bacteriol. 169:929-933. Hancock, R.E.W. 1991. Bacterial outer membranes: evolving concepts. ASM news 57:175-182. Hatch, T.P., I. Allen, and J.H. Pearce. 1984. Structural and polypeptide differences between envelopes of infective and reproductive life cycle forms of Chlamydia spp. J„ Bacteriol. 157:13-20. Helsel, L.O., W.F. Bibb, CA. Butler, P.S. Hoffman, and R.M. McKinney. 1988. Recognition of a genus-wide antigen of Legionella by a monoclonal antibody. Curr. Microbiol. 16:201-208. Helu, V. Altered osmoregulation of ompF in intergration host factor mutants of Escherichia coli. 1988. J. Bacteriol. 170:4950-4953. Herberman, R.N., C.W. Reynolds, and J.R. Ortaldo. 1986. Mechanisms of cytotoxicity by natural killer cells. Ann. Rev. Immunol.7:309-337. Herman, A., J.W. Kappler, P. Marrack, and A.M. Pullen. 1991. Superantigens: mechanism of T-cell stimulation and role in immune responses. Annu. Rev. Immunol. 9:745-751. Higgins, C.F., C.J. Dorman, D.A. Stirling, L. Waddell, I.R. Booth, G. May, and E. Bremer. 1988. A physiological role for DNA supercoiling in the osmotic regulation of gene expression in S. typhimurium and E. coli. Cell. 52:569-584. Higgins, D.E., E. Nazareno, and V.J. DiRita. 1992. The t virulence gene activator ToxT from Vijbrzo cholerae is a member of the AraC family of transcriptional activators. J. Bacteriol. 174:6974-6980. Hindahl, M.S., and B.H. Iglewski. 1984. Isolation and characterization of the Legionella pneumophila outer membrane. J. Bacteriol. 159:107-113. Hickey, E., and N. Cianciotto. 1994. Cloning and sequencing of the Legionella pneumophila fur gene. Gene. 143:117-121. Hofflin, J.M., I. Potasman, J.C. Baldwin, P.E. Oyer, E.B. Stinson, and J.S. Remington. 1987. Infectious complications in heart transplant recipients receiving cyclosporin and corticosteroids. Ann. Intern. Med. 106:209- 216. Hoffman, P.S., L. Pine, ans S. Bell. Production of superoxide and hydrogen peroxide in medium used to culture Legionella pneumophila: catalytic decomposition by charcol. Appl. Environ. Microbiol. 45:784-791. Hoffman, P.S., CA. Butler, and F.D. Quinn. 1989. Cloning and temperature-dependent expression in Escherichia coli of a Legionella pneumophila gene coding for a genus common 60 kilodalton antigen. Infect. Immun. 57:1731-1739. Hoffman, P.S., L. Houstan, and CA. Butler. 1990. Legionella pneumophila htpAB heat shock operon:Nucleotide sequence and expression of the 60 kDa antigen in L. pneumophila-infected cells. Infect. Immun. 58:3380-3387. Hoffman, P.S., J.H. Seyer, and CA. Butler. 1992a. Molecular characterization of the 28- and 31-kilodalton subunits of the Legionella pneumophila major outer membrane protein. J. Bacteriol. 174:908-913. Hoffman, P.S., M. Ripley, and R. Weeratna. 1992b. Cloning and nucleotide sequence of a gene {ompS) encoding the major outer membrane protein of Legionella pneumophila. J. Bacteriol. 174:914-920. 328 Hoffman, P.S., M. Ripley, R. Weeratna, R. Fernandez, G. Faulkner, D. Hoskin, T.J. Marrie, S.M. Logan, and M.A. Trevors. 1993. Legionella pneumophila surface proteins: gene regulation and cellular immunity. Pg. 74-78. In J.M. Barbaree, R.F. Breiman, and A.P. Dufour (ed.), Legionell : Current status and emerging perspectives,. ASM, Washington DC. Horwitz, M.A., and S.C. Silverstein. 1981. Activated human monocytes inhibit intracellular multiplication of Legionnaires' disease bacteria. J. Exp. Med. 154:1618-1635. Horwitz, M.A. 1983a. Cell mediated immunity in Legionnaires disease. J. Clin. Inves. 71:1686-1697. Horwitz, M.A. 1983b. Formation of a novel phagosome by the Legionnaires' disease bacterium (Legionella pneumophila) in human monocytes. J. Exp. Med. 158:1319-1331. Horwitz, M.A. 1983c. The Legionnaires' disease bacterium (Legionella pneumophila) inhibits phagosome- lysosome fusion in human monocytes. J. Exp. Med. 158:2108- 2111. Horwitz, M.A. 1984. Phagocytosis of the Legionnaires' disease bacterium (Legionella pneumophila) occurs by a novel mechanism; engulfment within a pseudopod coli. Cell 36:27- 33. Horwitz, M.A. and F.R. Maxfield. 1984. Legionella pneumophila inhibits acidification of its phagosome in human monocytes. J. Cell. Biol. 99:1936-1942. Horwitz, M.A. 1987. Characterization of avirulent mutant. Legionella pneumophila that survive but not multiply within human monocytes. J. Exp. Med. 166:1310-1328. Huang, L., Intergration host factor is a negative effector of invivo and invitro expression of ompC in Escherichia coli. 1990. J. Bacteriol 172:5293-5298. Huh, Y.J., and A.A. Weiss. 1991. A 23-kilodalton protein, distinct from BvgA, expressed by virulent 1 329 Bordetella pertussis binds to the promoter region of vir- regulated toxin genes. Infect. Immun. 59:2389-2395. Hulton, C.S.J., A. Seirafi, J.C. Hinton, J.M. Sidebotham, L. Waddell, G.D. Pavitt, T. Owen-Hughes, A. Spassky, H. Buc, and C.F. Higgins. 1990. Histone-like protein HI (H-NS), DNA supercoiling, and gene expression in bacteria. Cell. 63:631-642. Husmann L.K., and W. Johnson. 1992. Adherence of Legionella pneumophila to Guinea pig peritoneal macrophages, J774 mouse macrophages, and undiffrenciated U93 7 human monocytes; Role of Fc and complement receptors. Infec. Immun. 60:5212-5218. Jap, B.K., P.J. Walian, and K. Gehring. 1991. Structural architecture of an outer membrane channel as determined by electron crystolography. Nature 350:167-170. Johnson, W. , L. Varner, and M. Poch. 1991. Acquisition of iron by Legionella pneumophila: role of iron reductase. Infect. Immun. 59: 2376-2381. Jones, T.C. and J.G. Hirsch. 1972. The interactions between Toxoplasma gondii and the mammalian cells. II. The absence of lysosomal fusion with phagocytic vacuoles containing living parasites. J. Exp. Med. 136:1173-1184. Kaufmann, A., J. MacDade, C Patton, et. al. 1981. Pontiac fever: Isolation of the etiological agent (Legionella pneumophila) and demonstration of its mode of transmission. Amer. J. Epidemiol 114:337-347. Kaufmann, S.H.E., and M.J. Reddehase. 1989. Infection of phagocytic cells. Curr. Opin. Immunol. 2:43-49. Kaufmann, S.H.E. 1990a. Immunity to bacteria. Curr. Opin. Immunol. 2:353-359. Kaufmann, S.H.E. 1990b. Heat shock proteins and the immune response. Immunol. Today. 11:129-136. Kaufmann, S.H.E. 1991. Role of T-cell subsets in 330 bacterial infections. Curr. Opin. Immunol. 3:465-470. Keen, M.G., and P.S. Hoffman. 1984. Metabolic pathways and nitrogen metabolism in Legionella pneumophila. Curr. Microbiol. 11:81-88. Keen, M.G., and P.S. Hoffman. 1985. Broad-host-range plasmid pRK340 delivers Tn5 into the Legionella pneumophila chromosome. J. Bacteriol. 162:1332-1335. Keen, M.G., and P.S. Hoffman. 1989. Characterization of a Legionella pneumophila extracellular protease exhibiting hemolytic and cytotoxic activities. Infect. Immun. 57:732- 738. Kilvington, S. 1989. Moist-heat disinfection of pathogenis Acanthamoeba cysts. Lett. Appl. Microbiol. 9:187- 189. King, C.H., E.B. Shotts. Jr., R.E. Wooley, and K.G. Porter. 1988. Survival of coliforms and bacterial pathogens within protozoa during chlorination. Appl. Environ. Microbiol. 54:3023-3033. King, C.H., and J. Price. 1990. Survival of Legionella pneumophila within cysts of Acanthamoeba polyphaga following chlorine exposure. J. Appl. Bacteriol. 68:519-525. Klein, T.W., H. Friedman, and R. Weiden. 1984. Relative potency of virulent versus avirulent Legionella pneumophila for induction of cell-mediated immunity. Infect. Immun. 44:753-755. Koehler, J.E., S. Birkelund, and R.S. Stephens. 1992. Overexpression and surface localization of the Chlamydia trachomatis major outer membrane protein in Escherichia coli. Mol. Microbiol. 6:1087-1094. King, C.H., E.B. Shotts, R.E. Wooley, and K.G. Porter. 1988. Survival of coliforms and bacterial pathogens within prtozoa during chlorination. Appl. Environ. Microbiol. 54:3023-3033. 331 Kiley, M.P., R. Lathe, and J.P. Lecoq. 1983. New versatile cloning vectors based on bacteriophage M13. Gene 26:91-99. Kim, M.J., J.E. Rogers, M.C Hurley, and N.C. Engleberg. 1994. Phosphatase-negative mutants of Legionella pneumophila and their behavior in mammalian cell infection. Microb. Pathog. 17:51-62. Kleckner, N. , J. Bender, and S. Gottesman. 1991. Uses of transposons with emphasis on TnlO. In: Methods in enzymology. ed. J.H. Miller. Vol. 204, pp. 139-180. Academic Press Inc., NY. Kroczek, R.A., and E. Siebert. 1990. Optimization of northern analysis by vacuum-blotting, RNA transfer, and ultraviolet fixation. Anal. Biochem. 184:90-95. Kuchta, J.M., S.J. States, A.M. McNamara, R.M. Wadowsky, and R.B. Yee. 1983. Susceptibility of Legionella pneumophila to chlorine in tap water. Appl. Environ. Microbiol. 46:1134-1139. Kuhn, M., S. Kathariou, and W. Goebel. 1988. Hemolysin supports survival but not entry of the intracellular bacterium Listeria monocytogenes. Infect. Immun. 56:79-82. Lacey, B.W. 1960. Antigenic modulation of Bordetella pertussis. J. Hyg. 58:57-93. Lee, J.V , and A.A. West. 1991. Survival and growth of Legionella species in the environment. J. Appl. Bacteriol. Symp. Supp. 70:121S-129S. Leong, J.M., S. Nunes-Duby, C.F. Lesser, P. Youdeerian, M.M. Susskind, and A. Landy. 1985. The $80 and P22 attachment sites: primary structure and intergration with Eschrichia coli intergration host factor. J. Biol. Chem. 260:4468-4477. Lochner, J.E., R.L. Friedman, R.H. Bigley and B.H. Iglewski. 1983. Effect of oxygen-dependent antimicrobial systems on Legionella pneumophila. Infec. Immun. 39:487-489. I 332 Lorenzo, V. De, M. Herrero, U. Jakkubzik, and K.N. Timmis. 1990. Mini-Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insersion of cloned DNA in gram-negative eubacteria. J. Bacteriol. 172:6568-6572. Lowrie, D.B., P.D. Jackett, and N.A. Ratcliffe. 1975. Mycobacterium microti may protect itself from intracellular destruction by releasing cyclic AMP into phagosomes. Nature 254:600-602. Lowry, P.W., J. Rosalind, W. Grindley, N.J. Troup, and L. Tompkins. 1991. A cluster of legionella sternal wound infections due to postoperative topical exposure to contaminated tap water. N. Engl. J. Med. 324:109-113. Lucero, M.E., and C.C. Kuo. 1985. Neutralization of Chlamydia trachomatis cell culture infections by serovar specific monoclonal antibodies. Infect. Immun. 50:590-597. Ludwig, B., A. Schmid, R. Marre, and J. Hacker. 1991. Cloning, genetic analysis, and nucleotide sequence of a determinant coding for a 19-kilodalton peptideglycan - associated protein (Ppl) of Legionella pneumophila. Infect. Immun. 59:2515-2521. Lundemose, A.G., S. Birkelund, S. Fey, L.P. Mose, and G. Christiensen. 1991. Chlamydia trachomatis contains a protein similar to the Legionella pneumophila mip gene product. Mol. Microbiol. 5:109-115. Lundemose, A.G., D.A. Rouch, S. Birkelund, G. Christiansen, and J.H. Pearce. 1992. Clamydia trachomatis Mip-like Protein. Mol. Microbiol. 6:2539-2548. Marra, A. and H.A. Shuman. 1989. Isolation of a Legionella pneumophila restriction mutant with increased ability to act as a recipient in heterospecific matings. J. Bacteriol. 171:2238-2240. Marra, A., S.J. Blander, M.A. Horwitz, and H.A. Shuman. 1992. Identification of a Legionella pneumophila locus required for intracellular multiplication in human I 333 macrophages. Proc. Natl. Acad. Sci. USA. 89:9607-9611. Marra, A. and H.A. Shuman. 1992. Genetics of Legionella pneumophila virulence. Annu. Rev. Genet. 26:51-69. Mnoil, C, and J. Beckwith. 1985. TnPhoA: a transcription probe for protein export signals. Proc. Natl. Acad. Sci. USA 82:8129-8133. Marrie, T.J., H. Durant, and L. Yates. 1989. Community aquired pnuemonia requiring hospitalization: 5-year prospective study. Rev. Infec. Dis. 11:586-598. Mastro, T.D., B.S. Fields, R.F. Breiman, J. Campbell, D. Plikaytis, and J.S. Spika. 1991. Nosocomial Legionnnaires' disease and use of medication nebulizers. J. Infec. Dis. 163:667-671. McCleary, W.R., J.B. Stock, and A.J. Ninfa. 1993. Is acetyl phosphate a global signal in Escherichia coli? J. Bacteriol. 175:2793-2798. McDade, J.E., C.C. Shepard, D.W. Fraser, T.R. Tsai, M.A. Redus, W.R. Dowdle, and the Laboratory Investigation Team. 1977. Legionnaires' Disease: Isolation of a bacterium and demonstration of its role in other respiratory disease. N. Engl. J. Med. 297:1197-1203. McDade, J.E., D.J. Brenner, and F.M. Bozeman. 1979. Legionnaires' disease bacterium isolated in 1947. Ann. Intern. Med. 90:659-661. McDade, J.E., and C.C. Shepard. 199. Virulent to avirulent conversion of Legionnaires' disease bacterium (Legionella pneumophila)- Its effect on isolation techniques. J. Infect. Dis. 139:707-711. Mekalanos, J.J. 1992. Environmental signals controlling expression of virulence determinants in bacteria. J.Bacteriol. 174:1-7. Mel tor., A.R., and A.A. Weiss. 1989. Environmental regulation of expression of virulence determinants in 334 Bordetella pertussis. J. Bacteriol. 171:6206-6212. Miller, J.F., C.R. Roy, and S. Falkow. 1989. Analysis of Bordetella pertussis virulence gene regulation by use of transcriptional fusions in Escherichia coli. J. Bacteriol. 171:6345-6348. Miller, J.H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Miller, V.L., and J.J. Mekalanos. 1984. Synthesis of cholera toxin is positively regulated at transcriptional level by toxR. Proc. Natl. Acad. Sci. USA 81:3471-3475. Miller, V.L., R.K. Taylor, and J.J. Mekalanos. 1987. Cholera toxin trascriptional activator ToxR is a transmembrane DNA binding protein. Cell. 48:271-279. Miller, V.L., V.J. DiRita, and J.J. Mekalanos. 1989. Identification of toxS, a regulatory gene whose product enhances ToxR-mediated activation of the cholera toxin promoter. J. Bacteriol. 171:1288-1293. Mintz, C.L. and H.A. Shuman. 1987. Transposition of bacteriophage Mu in the legionnaires disease bacterium. Proc. Natl. Acad. Sci. USA 84:4645-4649. Mintz, C.S., J. Chen, and H.A. Shuman. 1988. Isolation and characterization of auxotrophic mutants of Legionella pneumophila that fail to multiply in human monocytes. Infect. Immun. 56:1449-1455. Moffat, J.F., W.J. Black, and L.S. Tompkins. 1994. Further molecular characterization of the cloned Legionella pneumophila zinc metalloprotease. Infect. Immun. 62: 751- 753. Moffat, J.F., W.J. Black, and L.S. Tompkins. 1994. Further molecular characterization of the cloned Legionella pneumophila zinc metalloprotease. Infect. Immun. 62:751-753. Moulder, J.W. 1985. Comparative biology of intracellular parasitism. Microbiol. Rev. 49:298-337. 335 Muller, H.E. 1980. Proteolytic action of Legionella pneumophila on human serum proteins. Infect. Immun. 27:51- 53. Mustafa, A.S., K.E.A. Lundin, and F. Oftung. 1993. Human T cells recognize mycobacterial heat shock proteins in the context of multiple HLA-DR molecules: studies with healthy subjects vaccinated with Mycobacterium bovis BCG and Mycobacterium leprae. Infect. Immun. 61:5294-5301. Nelson, M.B., M.A. Apicella, T.F. Murray, H. Vankeulen, L.D. Spotila, and D. Rekosh. 1988. Cloning and sequencing of Haemophilus influenzae outer membrane protein P6. Infect. Immun. 56:128-134. Newhall, W.J.V., and R.B, Jones. 1983. Disulfide-linked oligomers of the major outer membrane protein of chlamydiae. J. Bacteriol. 154:998-1001. Nguyen, M.H., J.E. Stout, andV.L. Yu. 1991. Legionellosis. Infec. Dis. Clin. North America. 5:561-584. Niedeveld, C.J., F.M. Pet, and P.L. Meenhorst. 1986. Effect of rubbers and their constituents on proliferation of Legionella pnuemophila in naturally conl\ninated hot water. Lancet, ii:180-184. Nikaido, H., and M. Vaara. 1985. Molecular basis of bacterial outer membrane permeability. Microbiol. Rev. 49:1- 32. Nikaido, H. 1992. Porins and specific channels of bacterial outer membranes. Mol. Microbiol. 6:435-442. Nikaido, H. 1993. Transport across the bacterial outer membrane. J. Bioenerg. Biomem. 25:581-589. Nikaido, H. 1994. Porins and specific diffusion channels in bacterial outer membranes. J. Biol. Chem. 269:3905-3908. Orme, I.M., E.S. Miller, A.D. Roberts, S.K. Furney, J.P. Griffin, K.M. Dobos, D. Chi, B. Rivoire, and P.J. r 336 Brennan. 1992. T lymphocytes mediating protection and cellular cytolysis during the course of Mycobacterium tuberculosis infection. J. Immunol. 148:189-196. Orme, I.M., P. Anderson, and W.H. Boom. 1993. T cell response to Mycobacterium tuberculosis. J. Infec. Dis. 167:1481-1497. Parnes, J.R. 1989. Molecular biology and function of CD4 and CD8. Adv. Immunol. 44:265-311. Pasculle, A.W., J.C. Feeley, R.J. Gibson, L.G. Cordes, R.L. Myerowitz, CM. Patton, G.W. Gorman, CL. Carmack, J.W. Ezzell, and J.N. Dowling. 1980. Pittsburgh pneumonia agent: direct isolation from human lung tissue. J. Infect. Dis. 141:727-732. Payne N.R., and M.A. Horwitz. 1987. Phagocytosis of Legionella pneumophila is mediated by human monocyte complement receptors. J. Exp. Med. 166:1377-1389. Payne, N.R., C.G. Bellinger-Kawahara, and M.A. Horwitz. 1987. Phagocytosis of Mycobacterium tuberculosis by human monocytes is mediated by receptors for the third component of complement. Clin. Res. 35:617A. Pearsall, N.N., and R.S. Weiser. 1970. The Macrophage. Lea & Febiger, Philadephia. Peeling, R.W., I.W. Maclean, and R.C Brunham. 1984, Invitro neutralization of Chlamydia trachomatis with monoclonal antibody to epitope on the major outer membrane protein. Infect. Immun. 46:484-488. Pine, L., P.S. Hoffman, G.B. Malcolm, R.F. Benson, and M.G. Keen. 1984. Determination of catalase, peroxidase, and superoxide dismutase within the genus Legionella. J. Clin. Microbiol. 20:421-429. Pine, L., P.S. Hoffman, G.B. Malcom, R.F. Benson, and M.J. Franzus. 1986. Role of keto acids and reduced-oxygen- scavenging enzymes in the growth of Legionella species. J. Bacteriol. 23:33-42. 337 Playfair, H.L., J. Taverne, C.A.W. Bate, and J.B. De Souza. 1990. The malaria vaccine:anti-parasite or anti- disease. Immunol. Today. 11:25-27. Pope, CD., L. Dhand, andN.P. Cianciotto. 1994. Random mutagenesis of Legionella pneumophila with mini-TnlO. FEMS Microbiol. Lett. 124:107-112. Portnoy, D., P.S. Jacks, and D. Hinrichs. 1988. Role of hemolysin for the intracellular growth of L. monocytogenes. J. Exp. Med. 167:1459-1471. Quinn, F.D., CA. Butler, and P.S. Hoffman. 1987. Characterization and the cloning of the disulfide-cross- linked major outer membrane protein of Legionella pneumophila H116. J. Cel. Biochem. 115:116. Quinr. F.D., and L.S. Tompkins. 1989. Analysis of a cloned sequence of Legionella pneumophila encoding a 3 8kDa metalloprotease possessing haemolytic and cytotoxic activities. Mol. Microbiol. 3: 797-805. Ramani, N., L. Huang, and M. Freundlich. 1992. In Vitro interaction of intergration host factor with the OmpF promoter-regulatory region of Escherichia coli. Mol. Gen. Genet. 231:248-255. Rashed, K.A., A, Mason, J. Gibson, J. Francis, A. Fairfax, and P. Daggett. 1986. Legionnaires' disease in Staford: Management of an epidemic. Lancet, i:197-199. Rechnitzer, C, M. Tveda, and G. Doring. 1989. A rapid method for purification of homogenous Legionella pneumophila cytotoxic protease using fast protein liquid chromotography. FEMS Microbiol. 59:3 9-44. Reeves, M.W., L.P. Pine, J.B. Neiland, and A. Balows. 1983. Absence of siderophore activity in Legionella species grown in iron deficient media. J. Bacteriol. 154:324-329. Rodgers, F.G., P.W. Greaves, A.D. Macrae, and M.J. Lewis. 1980. Electron microscope evidence of flagella and pili on Legionella pneumophila. J. Clin. Pathol. 33:1184- 1 338 1188. Rodgers, F.G., and M.R. Davey. 1982. Ultrastructure of the cell envelope layers and surface details of Legionella pneumophila. J. Gen. Microbiol. 128:1547-1557. Rothman, J.E. 1989. Polypeptide chain binding proteins: catalysts of protein folding and related processes in cells. Cell 59:591-601. Rouviere-Yaniv, J., M. Yaniv, and J.E. Germond. 1979. E. coli DNA binding protein HU forms nucleosome-like structure with circular double-stranded DNA. Cell. 17:265- 274. Rowbotham, T.J. 1980. Priliminary repor on the pathogenicity of Legionella pneumophila for fresh water and soli amoebae. J. Clin. Pathol. 33:1179-1183. Rowbotham, T.J. 1986. Current views on the relationships between amoeba, legionellae, and man. Israel J. Med. Sci. 22:678-689. Sadosky, A.B., L.A. Wiater, and H.A. Shuman, 1993. Identification of Legionella pneumophila genes required for growth within and killing of human macrophages. Infect. Immun. 61:5361-5373. Sadosky, A.B., J.W. Wilson, H.M. Steinman, and H.A. Shuman. 1994. The iron superoxide dismutase of Legionella pneumophila is essential for viability. J. Bacteriol. 176:3790-3799. Saha, A.K., J.N. Dowling, K.L. LaMarco, S. Das, A.T. Remaley, N. Olomu, M.T. Pope, and R.H. Glew. 1985. Properties of an acid phosphatase from Legionella micdadei which blocks superoxide anion production by human neutrophils. Arch. Biochem. Biophys. 243:150-160. Saha, A.K., J.N. Dowling, N.K. Mukhopadhyay, and R.H. Glew. 1988. Demonstration of two protein kinases in extracts of Legionella micdadei. J. Gen. Micobiol. 134:1275- 1281. 339 Sambrook, J., E.F. Fritch, and T. Maniatis. 1989 Molecular cloning: a laboratory manual. Cold Spring Harbor laboratory, Cold Spring Harbor, NY. Sampson, J.S., B.B. Plikaytis, and H.W. Wilkinson. 1986. Immunologic response of patients with legionellosis against major protein-contaning antigens of Legionella pneumophila serogroup 1 as shown by immunoblot analysis. J. Clin. Microbiol. 23:92-99. Sanger, F., S. Nicklen, and A.R. Coulsen. 1977. DNA sequenceing with achain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. Schmid, M.B. 1990. More than just "histone-like" proteins. Cell. 63:451-453. Schneider, D., and C Parker. 1982. Effect of pyridines on phenotypic properties of Bodetella purtussis. Infect. Immun. 38:548-553. Shinnick, T.M. 1991. Heat shock proteins as antigens of bacterial and parasitic pathogens. Curr. Top. Microbial. Immunol. 167:145-160. Simons, R.W., F. Houman, and N. Kleckner. 1987. Improved single and multicopy lac-based cloning vactors for protein and operon fusions. Gene. 53:85-96. Silhavey, T.J., M.L. Berman, and L.W. Enquist. 1984. Experiments with gene fusions. Cold Spring Habor Laboratory, Cold Spring Habor, NY. Skerrett, S.J., R. Schmidt, and T.R. Martin. 1989. Impaired clearance of aerosolized L. pneumophila in corticosterois treated rats: a model of legionnaires' disease in the immunocompromised host. J. Infect. Dis. 160:261-268. Slauch, J.M., and T.J. Silhavy. 1989. Genetic analysis of the switch that controls porin genes expression in Escherichia coli K-12. J. Mol. Biol. 210:281-292. I 340 Slauch, J.M., and T.J. Silhavi. 1991. cis-acting ompF mutations that result in OmpR-dependent constitutive expression. J. Bacteriol. 173:4039-4048. Southern, E.M. 1975. Detection of specific sequences among DNA fragments seperated by gel electrophoresis. J. Mol. Biol. 98:503-517. States, S.J., L.F. Conley, M. Ceraso, T.E. Stephenson, R.S. Wolford, R.M. Wadowsky, A.M. Mcnamara, and R.B. Yee. 1985. Effect of metals on Legionella pneumophila growth in drinking water plumbing systems. Appl. Environ. Microbiol. 50:1149-1154. Steel, T.W., J. Lanser, and N. Sangster. 1990. Isolation of Legionella longbeacheae serogroup 1 from potting mixes. Appl. Environ. Microbiol. 56:49-53. Steinert, M., H. Engelhard, M. Flugel, E. Wintermeyer, and J. Hacker. 1995. The Lly protein protects Legionella pneumophila from light but does not directly influence its intracellular survival in Hartmanella vermiformis. Appl. Environ. Microbiol. 61:2428-2430. Steinman, H.M. 1992. Construction of an Escherichia coli K-12 strain deleted for manganese and iron superoxide dismutase genes and its use in cloning the iron superoxide dismutase gene of Legionella pneumophila. Mol. Gen. Genet. 232:427-430. Stephens, R.S., M.R. Tarn, C.C. Kuo, and R.C. Nowinski. 1982. Monoclonal antibodies to Chlamydia trachomatis: antibody specificities and antigen characterization. J. Immunol. 128:1083-1089. Stibitz, S., and M.S. Yang. 1991. Subcellular location and immunological detection of proteins encoded by the vir locus of Bordetella purtussis. J. Bacteriol. 173:4288-4296. Stock, J.B., A.J. Ninfa, and A.M. Stock. 1989. Protein phosphorylation and regulation of adaptive responses in bacteria. Microb. Rev. 53:450-490. 341 Stout, J.E., V.L. Yu, and M.G. Best. 1985. Ecology of Legionella pneumophila within water distribution systems. Appl. Environ. Microbiol. 49:221-228. Straley, S.C., and P.A. Harmon. 1984. Yersinia pestis grows within phagolysosomes in mouse peritoneal macrophages. Infect. Immun. 45:655-659. Summersgill, J.T., L.A. Powell, B.L. Buster, R.D. Miller, and J.A. Ramirez. 1992. Killing of Legionella pneumophila by nitric oxide in yinterferon-activated macrophages. J. Leuk. Biol. 52:625-629. Swanson, A.F., and C.C. Kuo. 1991. Evidence that the major outer membrane protein of Chlamydia trachomatis is glycosylated. Infect. Immun. 59:2120-2125. Swanson, A.F., and C.C. Kuo. 1994. Binding of the glycan of the major outer membrane protein of Clamydia trachomatis to HeLa cells. Infect. Immun. 62:24-28. Szeto, L., and H.A. Shuman. 1990. The Legionella pneumophila major secretory portein, a protease, is not required for intracellular growth or cell killing. Infect. Immun. 58: 2585-2592. Tabor, S. 1990. Expression using the T7 RNA polymerase/promoter system. In F.A. Ansubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl (ed), Current protocols in molecular biology, p. 16.2.1-16.2.11. Green publishing and Wiley Interscience, NY. Tartakovskii, I.S., I.K. Zubashev, and V.F. Koniukhov. 1990. The protective properties of different Legionella antigens in experimental Legionella infection. Zh. Mikrobiol. Epidemiol. Immunobiol. 5:81-85. Thrope, T.C., and R.D. Miller. 1981. In Vitro production of an extracellular protease by Legionella pneumophila. Infect. Immun. 34:299-302. Tison, D.L., D.H. Pope, W.B. Cherry, and C.B. Fliermans. 1980. Growth of Legionella pneumophila in r i i 342 association with blue-green algae (Cyanobacteria). Appl. Environ. Microbiol. 39:256-259. Tison, D.L., J.A. Baross, and R.J. Seidler. 1983. Legionella in aquatic habitats in the mount Saint Helens blast zone. Current Microbiol. 9:345-348. Tobin, J. O'H., J. Beare, M.S. Dunmill, S. Fisher-Hoch, M. French, R.G. Mitchell, P.J. Morris, and M.F. Muers. 1980. Legionnaires' disease in a transplant unit, isolation of the causative agent from shower baths. Lancet. ii:118. Tropschug, M., E. Wachter, S. Mayer, E.R. Schonbrunner, and F.X. Schmid. 1990. Isolation and sequence of an FK506-binding protein from Neurospora crassa which catalyses protein folding. Nature. 346:674-677. Tyndal, R.L., and E.L. Domingue. 1982. Cocultivation of Legionella pneumophila and free-living amoeba. Appl. Environ. Microbiol. 44:954-959. Vickers, R.M., A. Brown, and G.M. Garrity. 1981. Dye containing buffered charcol yeast extract medium for the diffrenciation of members of the family Legionellaceae. J. Clin. Microbiol. 13:380-382. Villeneuve, A., L. Brossay, G. Paradis, and J. Hebert. 1994. Determination of neutralizing epitopes in variable domains I and IV of the major outer membrane protein from Chlamydia trachomatis serovar K. Microbiol. 140:2481-2487. Wadowsky, R.M., and R.B. Yee. 1983. Satellite growth of Legionella pneumophila with an environmental isolate of Flavobacterium breve. Appl. Environ. Microbiol. 46:1447- 1449. Wadowsky, R.M., L.J. Butler, M.K. Cook, S.M. Verma, M.A. Paul, B.S. Fiels, G. Kelti, J.L. Sykorka, and R.B. Yee. 1988. Growth supporting activity for Legionella pneumophila in tap water cultures and implication of hartmanellid amoeba as growth factors. Appl. Environ. Microbiol. 49:1197-1205. Weeratna R., D.A. Stamler, P.H. Edelsteine, M. Ripley, I 343 T. Marrie, D. Hoskin, and P.S. Hoffman. 1994. Human and guinea pig immune responses to Legionella pneumophila protein antigens OmpS and Hsp60. Infect. Immun. 62:3454- 3462. Weiss, M.S., U. Abele, J. Weckesser, W. Welte, E. Schiltz, and G.E. Schulz. 1991. Molecular architecture and electrostatic properties of a bacterial porin. Science. 254:1627-1629. Weiss, M.S., and G.E. Schulz. 1992. Structure of porin refined at 1.8 A resolution. J. Mol. Biol. 227:493-509. Wiater, L.A., A.B. Sadosky, and H.A. Shuman. 1994. Mutagenesis of Legionella pneumophila using Tn903dIIlacZ: identification of a growth-phase-regulated pigmentation gene. Mol. Microbiol. 11:641-653. Widen, R.H., CA. Newton, T.W. Klein, and H. Friedman. 1993. Antibody-mediated enhancement of Legionella pneumophila-induced interleukin 1 activity. Infect, immun. 61:4127-4132. Wilkinson, H.W., D.D. Cruece, and C.V. Broom. 1981. Validation of Legionella pneumophila indirect immunofluorescence assay with epidemic sera. J. Clin. Microbiol. 13:139-146. Williams, A., A. Baskerville, A.B. Dowsett, and J.W. Conlan. 1987. Immunocytochemical demostration of the association between Legionella pneumophila, its tissue- destructive protease, and pulmonary lesions in experimental legioannaires' disease. J. Pathol. 153:257-264. Winn, W.C.Jr., and R.L. Myerowitz. 1981. The pathology of the Legionella pneumonias. A review of 74 cases and the literature. Hum. Pathol. 12:401-422. Winn, W.C.Jr. 1983. Cellular inflammatory defences in Legionella pneumonia. Clin. Immunol. News. 4:83-87. Winn, W.C.Jr. 1988. Legionnaires disease: Historical perspective. Clin. Microbiol. Rev. 1:60-81. n I 344 Wintermeyer, E., U. Rdest, B. Ludwig, A. Debes, and J. Hacker. 1991. Characterization of legiolysin (lly), responsible for haemolytic activity, colour production, and fluorescence of Legionella pneumophila. Mol. Microbiol. 5: 1135-1143. Wintermeyer, E., M. Flugel, M. Ott, M. Steinert, U. Rdest, K-H. Mann, and J. Hacker. 1994. Sequence determination and mutational analysis of the lly locus of Legionella pneumophila. Infect. Immun. 62: 1109-1117. Woo, A.H., V.L. Yu, and A. Goetz. 1986. Potential in- hospital mode of transmission for Legionella pneumophila: Demonstration experiments for dissemination by showers, humidifiers, and rinsing of ventilation bag apparatus. Amer. J. Med. 80:567-573. Yamamoto, Y., T.W. Klein, H. Shinomiya, M. Nakano, and H.Friedman. 1992. Infection of macrophages with Legionella pneumophila induces phosporylation of a 76-kilodalton protein. Infec. Immun. 60:3452-3455. Young, D., R. Lathigra, R. Hendrix, D. Sweetser, and R.A. Young. 1988. Stress proteins are immune targets in leprosy and tuberculosis. Natl. Acad. Sci. USA 85:4267-4270. Zhang, Y.X., S. Stewart, T. Joseph, H.R. Taylor, and H.D. Caldwell. 1987. Protective monoclonal antibodies recognize epitopes located on the major outer membrane protein of Chlamydia trachomatis. J. Immunol. 138:575-581.