UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional and molecular assessment of the role of toxic shock syndrome toxin-1, staphylococcal enterotoxin… De Boer, Monica Lynn 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1996-090701.pdf [ 11.86MB ]
Metadata
JSON: 831-1.0099042.json
JSON-LD: 831-1.0099042-ld.json
RDF/XML (Pretty): 831-1.0099042-rdf.xml
RDF/JSON: 831-1.0099042-rdf.json
Turtle: 831-1.0099042-turtle.txt
N-Triples: 831-1.0099042-rdf-ntriples.txt
Original Record: 831-1.0099042-source.json
Full Text
831-1.0099042-fulltext.txt
Citation
831-1.0099042.ris

Full Text

Functional and Molecular Assessment of the Role of Toxic Shock Syndrome Toxin-1, Staphylococcal Enterotoxin A , and Staphylococcal Enterotoxin B in the Pathogenesis of Toxic Shock Syndrome by MONICA L Y N N De BOER B . S c , The University of Waterloo, 1990 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Microbiology) We accept this thesis as conforming to thej&quired standard  THE UNIVERSITY OF BRITISH COLUMBIA February 1996 c Monica Lynn De Boer, 1996  In  presenting  degree at the  this  thesis in  partial  University of  fulfilment  of  the  requirements  British Columbia, I agree that the  for  an advanced  Library shall make it  freely available for reference and study. I further agree that permission for extensive copying  of  department  this thesis for or  by  his  or  scholarly purposes may be granted her  representatives.  It  is  by the  understood  that  head of copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department  of  rV^faWvOo-  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  ABSTRACT Staphylococcus aureus produces exoproteins including toxic shock syndrome toxin-1 (TSST-1), staphylococcal enterotoxin A (SEA) and staphylococcal enterotoxin B (SEB) which have been implicated in causing an acute multisystem disease called toxic shock syndrome (TSS). These toxins belong to the family of proteins known as superantigens.  Superantigens  cause profound immunological disturbances which are believed to underly the mechanisms causing TSS upon infection with toxigenic strains of S. aureus. Production of TSST-1, SEA and SEB are examples of variable genetic traits in that they are produced by some S. aureus isolates but not others. Certain toxin combinations are more prevalent than others. For example, SEA and TSST-1 are frequently co-produced among menstrual TSS isolates, whereas TSST-1 and SEB appear to rarely, i f ever, be co-produced. The nature of these interactions on both functional and molecular levels was examinined in this thesis. It was hypothesized that co-production of SEA and TSST-1 might yield a more virulent strain than production of TSST-1 alone, providing a putative explanation for why TSST1 / S E A strains are associated with the majority of menstrual TSS cases. To test this +  +  hypothesis, a TSST-1 /SEA" derivative of a TSST-1+/SEA+ clinical strain was constructed +  by the method of plasmid integration. Introduction of the TSST-1 /SEA and TSST+  +  1 /SEA" strains into a tampon-associated vaginal infection model of TSS and a D+  galactosamine sensitized mouse model of lethal shock however did not demonstrate any detectable differences in lethality or morbidity between the strains. In addition, in vitro  ii  analyses revealed that the culture filtrate from the SEA" mutant did not differ in its ability to induce TNFa secretion from rabbit splenocytes or stimulate T cell proliferation at low dilutions when compared to the TSST-1 VSEA" strain. This suggests that production of SEA with TSST-1 in this strain does not increase the virulence of S. aureus. The molecular basis of mutually exclusive production of TSST-1 and SEB in S. aureus was examined. Analysis of TSST-1 isolates indicated that 8 of 8 TSST-1 isolates had +  +  sequences homologous with seb, even though SEB was not produced. PCR analysis with primers that targeted various regions of seb failed to amplify the expected PCR product, suggesting that a full SEB coding sequence is not present. Hybridization of restriction fragments with TSST-1 and SEB genetic element probes suggested that tst was located in close proximity to SEB genetic element sequences on the chromosome. DNA similarities between TSST-1 and S E B isolates were also found upon hybridization with SEB genetic +  +  element probes. These observations could suggest that the SEB genetic element carries a preferred site of TSST-1 genetic element insertion. Upon insertion, the TSST-1 genetic element may interfere with the expression of SEB in as yet some unidentifed manner. Alternatively, the TSST-1 and SEB genetic elements may share a common point of insertion in the S. aureus chromosome linked to DNA sequences homologous to the SEB genetic element. Insertion of one genetic element carrying one of the toxin genes may inhibit the insertion of the other genetic element carrying the alternate toxin gene.  111  T A B L E OF CONTENTS Abstract Table of Contents List of Tables List of Figures List of Abbreviations List of Appendices Acknowledgement  Page* u iv xi xii xvii xviii xix  Chapter 1. Background 1.1. 1.1.1. Historical Perspectives/Epidemiology/Prevalence 1.1.2. Causative Agent of Toxic Shock Syndrome A. Carriage Rate of Staphylococci in the Population 1.1.3. Clinical Spectrum and Diagnosis of Toxic Shock Syndrome 1.1.4. Treatment of Toxic Shock Syndrome 1.1.5. Risk Factors Associated with Menstrual and Nonmenstrual Toxic Shock Syndrome  2 3 4 6 8  1.2. 1.2.1. Evidence implicating toxic shock syndrome toxin-1 in the pathogenesis of toxic shock syndrome 1.2.2. Evidence implicating staphylococcal enterotoxins in the pathogenesis of toxic shock syndrome A. Evidence suggesting that SEA and TSST-1 Co-production may be important in the pathogenesis of menstrual TSS B. Evidence suggesting a putative role for SEB in the pathogenesis of TSS in TSST-1* S. aureus isolates 1.2.3. Evidence implicating toxins other than TSST-1 and the enterotoxins in the pathogenesis of toxic shock syndrome  12 15 16 17 19  iv  1.2.4. Functional consequences of TSST-1, SEA and SEB Production by S. aureus A. TSST-1, SEA and SEB as'Superantigens' B. The MHC Class II molecule is the receptor for TSST-1, SEA and SEB C. Activation of monocytes by staphylococcal superantigens D. Activation of T cells by staphylococcal superantigens E. Evidence implicating a central role for T cell mitogenesis and cytokine secretion in the pathogenesis of superantigenmediated shock 1.2.5. Genetic control of TSST-1, SEA and SEB Production by iSl aureus A. Regulatory loci involved in controlling toxin production B. Genetic analysis of TSST-1 production C. Genetic analysis of SEA production D. Genetic analysis of SEB production 1.3.  Focus of Thesis Proposal  20 22 23 24 26  28 30 33 34 37  Chapter 2. Materials and Methods 2.1.  Bacterial growth media and antibiotics  2.2.  Bacterial strains and plasmids  2.3.  Phenorvpic Analyses of S. aureus 2.3.1. Toxin Production A. Immunoblotting B. Immunoassay  2.4.  40 40  40 40 44  Molecular Biology Techniques 2.4.1. D N A manipulation  46  2.4.2. Isolation of plasmid DNA  46  2.4.3. Transfer of DNA  46  2.4.4. Transduction  47  (  2.5.  2.4.5. Preparation of chromosomal DNA  48  2.4.6. Preparation of DNA probes  48  2.4.7. Southern hybridization analysis  49  2.4.8. Polymerase chain reaction analysis Construction of Isoeenic Mutants  49  2.5.1. S. aureus RN3984 - Wildtype isolate used for isogenic mutant construction  50  2.5.2. Inactivation of tst in S. aureus RN3984 2.5.3. Inactivation of sea in S. aureus RN3984 A. Construction of plasmids B. Introduction of pMLD6873C-l into RN3984 and isolation of SEA" transductants C. Verification of the stability of the SEA" phenotype of the isogenic mutant after in vivo growth 2.6.  51 52 54  Creation of a TSST-1 and SEB producing strain of S. aureus 2.6.1.  Construction of plasmids A. pJWl (^/-containing plasmid) B. pMLD6876-l (^ei-containing plasmid)  2.6.2. Introduction of pJWl and pMLD6876-l into S. aureus 2.7.  51  55 55 56  In Vitro Effects of TSST-1. SEA, and the Culture Filtrates from Isogenic Strains on Human Peripheral Blood Mononuclear Cells and Rabbit Splenocvtes 2.7.1 Purification of TSST-1 and SEA  56  2.7.2. Purification of human peripheral blood mononuclear cells  57  2.7.3  58  Preparation of rabbit splenocytes  2.7.4. Preparation of culturefiltratesfromisogenic strains  58  2.7.5. Immunoassay detection of cytokine production by human PBMC or rabbit splenocytes stimulated with toxin or culture filtrate A. Human TNFa  59  vi  B. Human DL-IB C. Human IL-6 D. Rabbit TNFa 2.7.6. Proliferative response of human PBMC or rabbit splenocytes upon stimulation with toxin or culture filtrate 2.8.  59 59 60 61  Animal Model Studies 2.8.1. Animals  62  2.8.2. Preparation of isogenic mutant inocula for in vivo studies  62  2.8.4. Introduction of isogenic mutants into two animal models of lethal shock A. Tampon-associated vaginal infection model of TSS B. D-Galacatosamine sensitized mouse model  63 64  Chapter 3. Determination of whether TSST-1 plays a pivotal role in the pathogenesis of menstrual TSS by the introduction of isogenic mutants in TSST-1 production in a tampon-associated vaginal infection model 3.1.  Introduction  65  3.2.1. Characterization of the TSST-1" mutant RN7043  66  3.2.2. Virulence of RN3984 and RN7043 in a tampon-associated vaginal infection model of TSS A. Mortality B. Temperataure change C. Weight loss  70 72 72  3.2.3. Effect of Culture Filtrates of Isogenic Strains on Rabbit Splenocytes A. T cell proliferation B. TNFa secretion 3.3.  Discussion  72 75 75  vii  Chapter 4. Effect of the Co-Production of TSST-1 and SEA Superantigens in the Pathogenesis of Menstrual TSS 4.1.  Introduction  4.2.  Results  4.3.  80  4.2.1. In vitro effect of purified TSST-1 and SEA on human PBMC A. T cell proliferation B. Cytokine secretion (TNFa, IL-16, DL-6)  81 84  4.2.2. Characterization of the TSST-17SEA' mutant INT#1  91  4.2.3. Virulence of RN3984 and INT#1 in two animal models of lethal shock A. Tampon-associated vaginal infection model B. D-galacotsamine sensitized mouse model  96 101  4.2.4. In vivo stability of the SEA" phenotype of INT#1  101  4.2.5. In vitro effect of culturefiltratesfromisogenic strains on rabbit splenocytes A. T cell proliferation B. TNFa secretion  102 105  4.2.6. Enhancement of T cell proliferation by TSST-1 and SEA in rabbit splenocytes is concentration dependent  105  Discussion  112  Chapter 5. Demonstration of Whether a Mutually Exclusive Toxin Phenotype in S. aureus is Mediated by Lack of Structural Genes 5.1.  Introduction  11«  viii  5.2.  Results 5.2.1. Detection of tst in TSST-1* S. aureus isolates  118  5.2.2. Detection of seb among TSST-1 S. aureus isolates  123  +  5.2.3. PCR analysis of TSST-1 isolates with oligonucleotide primers specific for seb  126  5.2.4. Immunoblot analysis of putative SEB /TSST-1 S. aureus isolates  129  +  +  +  5.2.4. Examination of SEB isolates with probes derived from sequences flanking tst +  5.3.  Discussion  131 131  Chapter 6. Possible Molecular Mechanisms Dictating Lack of SEB and TSST-1 Co-Production in S. aureus 6.1.  Introduction  6.2.  Results  138  6.2.1. Creation of a SEB*/TSST-1* strain of S. aureus  138  6.2.2. Analysis of TSST-1 isolates with probes derived from sequences flanking the SEB structural gene and located within the SEB genetic element  139  +  6.2.3. Transduction of RN7043 te/::Tc into S. aureus 7690 r  147  6.2.4. Analysis of SEB isolates with the 700 bp EcoBl fragment derivedfromsequences associated within the SEB genetic element  152  6.2.4. Analysis of SEB* isolates with fed-specific gene probes  154  6.2.5. Analysis of SEB* isolates with a probe derivedfromsequences that extend past the junction of the SEB genetic element downstream of seb  157  +  6.3.  Discussion  Chapter 7. Summary of Scientific Contributions and Future Avenues of Research 7  7  L  *« 2  Summary and scientific c o n t r i b u t i o n s  of th^ic  1  6  g  Future Avenues ofB«Par h f  Literature Cited 175  x  LIST O F TABLES TABLE*  Page*  1.  Case Definition for Toxic shock syndrome  7  2.  TSST-1 and SEA Production among S. aureus isolates  3.  S. aureus Reference and Control Strains Used in this Study  4.  S. aureus Clinical Isolates Used in this Study  5.  Plasmids Used in This Study  6.  Detection of TSST-1 levels in the Culture Filtrates of RN3984 and RN7043  7. 8.  9. 10. 11.  Detection of SEA levels in the Culture Filtrates of RN3984 and INT#1  18 41 42 43 68 92  Mortality and Morbidity mediated by RN3984 and INT#1 in a Tampon-Associated Vaginal Infection Model of TSS in the NZW rabbit  98  Effect of Low Concentrations of TSST-1 (100 pg/ml) and SEA (1 or 10 pg/ml) on the Mitogenic Response of Rabbit Splenocytes  110  Effect of High Concentrations of TSST-1 (100 ng/ml) and SEA (1 or 10 ng/ml) on the Mitogenic Response of Rabbit Splenocytes  111  TSST-1 and SEB Production among S. aureus Isolates  137  xi  LIST O F FIGURES Figure #  1.  Plasmid constructs generated to assist in the inactivation of sea in RN3984 by the method of plasmid integration  2.  Immunoblot analysis of culture filtrates from TSST-1" mutants with a polyclonal anti-TSST-1 antibody  3.  Southern hybridization analysis of TSST-1" mutant genomic DNA digested with CM and hybridized with a 297 bp HincU-BamHl tst specific gene probe  4.  Comparison of SEA levels with respect to total protein in the culture filtrates of TSST-1" mutants  5.  Weight loss in rabbits infected with RN3984 and RN7043  6.  Mitogenic response induced by RN3984 and RN7043 culture filtrates on rabbit splenocytes  7.  Increased proliferative response of rabbit splenocytes to RN3984 culture filtrate was consistent with the presence of TSST-1  8.  T N F a secretion induced by RN3984 and RN7043 culture filtrates from rabbit splenocytes  9.  Mitogenic response of human PBMC to TSST-1  10.  Mitogenic response of human PBMC to SEA  11. 12. 13. 14. 15.  16.  17.  18.  19. 20. 21.  22.  23.  Human PBMC mitogenic response induced upon coincubation with TSST-1 and SEA  85  TNFa secretion from human PBMC upon coincubation with TSST-1 and SEA  87  EL-IB secretion from human PBMC upon coincubation with TSST-1 and SEA  89  IL-6 secretion from human PBMC upon coincubation with TSST-1 and SEA  90  Comparison of TSST-1 levels with respect to total protein in the culture filtrates of RN3984 and INT#1  93  Southern hybridization analysis of SEA" mutant genomic DNA with a 624 bp 2JamHI-#//idin sea specific gene probe  94  A possible recombination event between pMLD6873C-l and sea in RN3984 to generate the SEA" mutant INT#1  95  Southern hybridization analysis of SEA' mutant genomic DNA with a 1.4 kb Taql fragment of pRN6441  97  Weight loss in rabbits infected with RN3984 and DMT#1  99  Mitogenic response induced by RN3984 and INT#1 culture filtrates on rabbit splenocytes  103  Increased proliferative response of rabbit splenocytes to RN3984 culture filtrate was consistent with the presence of SEA  104  TNFa secretion induced from rabbit splenocytes by RN3984 and INT#1 culture filtrates  106  Mitogenic response of a pooled rabbit splenocyte preparation to TSST-1  107  xiii  24. 25. 26.  Mitogenic response of a pooled rabbit splenocyte samples to SEA  108  Restriction map of pRN6550 and location of fragments used as probes  120  Southern hybridization analysis of TSST-1" and TSST-1 S. aureus isolates digested with Clal and probed with a 297 bp HincUBamHI tst specific gene probe  121  Southern hybridization analysis of TSST-1 isolates digested with HindlU and probed with a 297 bp Hincll-BamHl tst specific gene probe  122  +  27.  28. 29.  30.  31.  32.  33.  34.  +  Restriction map of pSK155 and the location of the 150 bp Taql seb specific gene probe  124  Southern hybridization analysis of TSST-1 isolates with a 150 bp Taq seb specific gene probe  125  +  Location of the 150 Taql seb specific gene probe used for Southern hybridization analysis and the oligonucleotide primers used for PCR analysis  127  Agarose gel electrophoresis of PCR products after amplification using SEB1 and SEB2 primers  128  Agarose gel electrophoresis of PCR products after amplification using MLD-1 and MLD-2 primers  130  Southern hybridization analysis of SEB S. aureus isolates probed with a 1.0 kb BamSl fragment derived from pRN6550  132  Southern hybridization analysis of SEB'*' S. aureus isolates probed with a 890 bp BamHl-Styl fragment derived from pRN6550  133  +  xiv  35.  36. 37.  38.  39.  40.  41.  42.  43.  44.  45.  Immunoblot analysis of culture filtrates of RN4220 transformed with pJWl and pMLD6876-l  140  Restriction map of pSKISS and the location of fragments used as probes  142  Southern hybridization analysis of TSST-1* isolates with a 700 bp EcoRl SEB genetic element probe  143  Southern hybridization analysis of TSST-1 isolates digested with C M and hybridized with a 1.1. kb HindUl-EcoRI SEB genetic element probe  144  Southern hybridization analysis of TSST-1* isolates digested with HindXH and hybridized with a 1.1. kb Hindlll-EcoRl fragment of pSK155  146  Southern hybridization analysis of Tc* transductants of S. aureus 7690 with a 297 bp HincIL-BamHI tst specific gene probe  149  +  Southern hybridization analysis of T<f transductants of S. aureus 7690 with a 700 bp £coRI fragment of pSK155  150  Southern hybridization analysis of TC* transductants of S. aureus 7690 with a 1.1 kb Hindm-EcdSl fragment of pSK155  151  Southern hybridization analysis of SEB* isolates with a 700 bp EcoRI SEB genetic element probe  153  Southern hybridization analysis of SEB* isolates with a 150 bp Taql seb specific gene probe  155  Illustration of the predicted chromosomal region surrounding seb in S. aureus isolates  156  xv  Southern hybridization analysis of the S E B isolate 0507 with a 1.8 kb Kpnl-Clal seb gene probe +  Southern hybridization analysis of SEB S. aureus isolates with a 1.4 kb XbalHin&TQ. SEB genetic element junction probe +  Mechanisms that may mediate the mutual exclusivity of TSST-1 and SEB co-production in S. aureus.  y  LIST OF ABBREVIATIONS APC agr BHI BSA Cm Em ET 5-EFN i.p. EL LB LD LFA MTSS NMTSS PBS PBS-T PBS-TW PBMC PCR MLEE NZW SDS SDS-Page TCR TH TNF TNFa Tc TSA TSB TSS TSST-1 SEA SEB SEC sar sae  antigen presenting cell accessory gene regulator brain heart infusion bovine serum albumin chloramphenicol erythromycin electrophoretic type gamma-interferon intraperitoneal interleukin Luria-Bertani lethal dose lymphocyte function antigen menstrual toxic shock syndrome nonmenstrual toxic shock syndrome phosphate buffered saline PBS + .05% Tween 20 PBS + . 1 % Tween 20 peripheral blood mononuclear cells polymerase chain reaction multilocus enzyme electrophoresis new Zealand white sodium dodecyl sulfate sodium dodecyl sulfate-polyacrylamide gel electrophoresis T cell receptor T helper tumor necrosis factor tumor necrosis factor alpha tetracycline tryptic soy agar tryptic soy broth toxic shock syndrome toxic shock syndrome toxin-1 staphylococcal enterotoxin A staphylococcal enterotoxin B staphylococcal enterotoxin C staphylococcal accessory regulator S. aureus exoprotein expression  xvii  LIST OF APPENDICES  Page# Appendix # 1.  Ph.D. Publications To Date  197  xviii  ACKNOWLEDGEMENT Many thanks to my supervisor A.W. Chow for his encouragement, helpful advice and knowledgeable contributions over the past 5 years. I also gratefully thank Winnie Kum who aided me considerably over the course of this research, especially in the animal model studies and subsequent in vitro assays. I would also like to acknowledge Swee Han Goh, who was instrumental in setting up most of the molecular biology techniques in our laboratory; Keven Laupland for optimization of the IL-6 assay; the nursing staff at the Cell Separator Unit at Vancouver General Hospital for supplying the plateletpheresis packs; Saleem Khan for providing plasmid pSK155, and especially Kris Gillespie at the Jack Bell Research Centre Animal Laboratory for her expert technical assistance in the animal model studies. Special thanks also to Richard Novick who provided me with the excellent opportunity to learn more about the genetic systems in Staphylococci by allowing me to visit his laboratory. He also provided me with all of the S. aureus strains and plasmids that were needed to carry out most of the molecular biology work.  xix  Chapter 1. B A C K G R O U N D  1.1.  Clinical. Epidemiologic and Microbiologic Features of Toxic Shock Syndrome  1.1.1. Historical Perspectives/Epidemiology/Prevalence Although illnesses clinically resembling the manifestations observed in toxic shock syndrome (TSS) have been described since 1927 (Chesney et al., 1984), the first documented case of TSS was portrayed by Dr. James Todd in 1978 while working at the Children's Hospital in Denver, Colorado (Todd et al., 1978). Dr. Todd named the syndrome and first described its association with the organism Staphylococcus aureus. Shortly thereafter, many studies aimed at determining what risk factors may play a part in the etiology of the disease were carried out. As a result, in 1980, a positive association was found between TSS and tampon use among women during menstruation (Davis et ah, 1980; Shands et al., 1980). In particular, it was noted that women using Rely tampons (distributed by Proctor and Gamble R  Co.) were at a greater relative risk than women using other tampon brands (Schlech et al., 1982). At this time (September 1980) Rely was voluntarily removed from the market by the R  company. TSS was listed as a nationally reportable disease by the United States Center for Disease Control (CDC) in 1982 (Chesney et al., 1984). Using a passive-surveillance system, the number of reported cases of menstrual TSS (MTSS) was found to peak in 1980, followed by a significant decline thereafter (Broome, 1989). Whether the decrease was due to an actual lower incidence, or as a result of removal of Rely brand tampons from the market in R  conjunction with increased public awareness leading to changes in the number of women  1  using tampons, or their menstrually-related habits, is not known. Although a positive association between TSS and tampon use during menstruation was made early on in the identification of this syndrome, cases of TSS in men, children, and nonmenstruating women have also been described (Silver, 1980; Reingold et al., 1982; Bartlett et al., 1982; Chesney et al., 1984). Nonmenstrual TSS (NMTSS) is often associated with the presence of S. aureus in focal wound or soft tissue infections. The temporal trends from 1979-1986 are difficult to assess due to the low incidence rate during this time period (Broome, 1989). However in recent years, NMTSS is becoming more prevalent than MTSS (Chow, 1993). The estimated prevalence of TSS is 0.22 to 1.23 cases per 100,000 (Chow, 1995). Recurrences are quite frequent (15-22%) (Chow, 1995). Today, TSS still remains a potentially life-threatening disease with a mortality rate of 3 to 7% (Chow, 1995). 1.1.2.  Causative Agent of Toxic Shock Syndrome  Staphylococcal TSS is caused by various toxins elaborated by the organism Staphylococcus aureus. The toxins that are strongly implicated in causing the disease include toxic shock syndrome toxin-1 (TSST-1), staphylococcal enterotoxin A (SEA), staphylococcal enterotoxin B (SEB) and staphylococcal enterotoxin C (SEC). S. aureus isolates associated with TSS are phenotypically and genotypically distinct from other S. aureus strains, as reviewed by See and Chow (1989). The most important differentiating factor of TSS-associated strains is the production of TSST-1. TSST-1 was found to be produced by over 90% of staphylococci isolated from TSS patients (Crass and Bergdoll, 1986a). In addition, Todd et al. (1984) showed that TSS-associated strains of S.  2  aureus were also more likely to produce staphylococcal enterotoxins than non-TSS strains. Altemeier et al. (1982) found that the majority of TSS-associated S. aureus isolates belonged to phage group 1, and were lysed by phages 29 and 52. Other investigations have confirmed these findings (Kreiswirth et al, 1984; Marples and Wieneke, 1993). In addition, TSSassociated S. aureus strains were less likely to carry plasmids (Todd et al., 1984), exhibited resistance to arsenate, cadmium and pencillin (Barbour, 1981; Todd et al., 1984; Kreiswirth et al., 1984), were less likely to produce pigment (Todd et al., 1984), exhibited bacteriocin susceptibility (Todd et al., 1984), were more proteolytic for the substrates hemoglobin (Barbour, 1981) and casein (Todd et al., 1984; Todd et al., 1985), and have reduced hemolytic activity (Barbour, 1981; Schlivert et al., 1982; Todd et al., 1984). Vaginal TSS isolates were also less likely to produce lipase and nuclease (Schlievert et al., 1982). TSSassociated S. aureus strains have also been found to be more lethal in rabbit models of TSS (Scott et al., 1983; Arko et al, 1984; Melish et al, 1989). Collectively these observations suggest that S. aureus strains associated with TSS are phenotypically distinct from other S. aureus strains. A.  Carriage Rate of Staphylococci in the Population  Although in general staphylococci are found on the skin, skin glands and mucous membranes of warm-blooded animals, certain staphylococcal species colonize specific anatomical sites (Howard and Kloos, 1987). S. aureus is generally found in the anterior nares. Jacobson et al. (1984) observed a S. aureus nasal carriage rate of 31% in 770 patients admitted to a hospital in Utah. The carrier rate for TSS-associated protein (or TSST-1) positive strains in this study was 6.6%. In a separate study, Martin et al (1982)  3  found a 26.9% nasal carriage rate and a 10.3% vaginal carriage rate in 145 young women. Studies by Guinan et al. (1982) and Chow et al. (1983) found vaginal carriage rates of 9.2 and 6.8% in 808 and 495 women respectively, indicating that S. aureus is not typically part of the vaginal microflora. In the latter study, 2.6% of the women were colonized with TSST-1 producing strains of S. aureus. A close association was found between nasal and vaginal carriage in the study by Martin et al. (1982), in that 60% of the women with vaginal staphylococci also had nasal carriage. This rate of nasal carriage was significantly higher in comparison to nasal carriers whose vaginal cultures were negative for S. aureus. This study provides support for the suggestion that MTSS may be mediated by the deposition of S. aureus in the vagina through the tampon, which may have become contaminated by the woman's own hands (Mortimer, 1982). Consistent with this is the observation made by Linnemann et al. (1982) that S. aureus isolates obtained from nasal and genital cultures of the same individual were of the same phage type, suggesting a clonal relationship. In a similiar manner, nasal carriage of S. aureus may be an important factor dictating predisposition to NMTSS through infection of surgical wounds by autoinoculation (Mortimer, 1982). These studies provide a basis for estimating the percentage of the population that is potentially at risk for developing TSS. 1.1.3.  Clinical Spectrum and Diagnosis of Toxic Shock Syndrome  The pathological manifestations of TSS can be primarily divided into four distinctive features, as summarized by Chesney (1989). First, TSS can be readily characterized by the rapidity of its onset in an otherwise healthy individual. Upon initial presentation, symptoms of fever, chills, malaise, headache, myalgias, muscle tenderness, vomiting and diarrhea  4  commonly exist. The second unique feature of TSS is the rapid onset of hypotension resulting from a loss of vascular tone and the massive leakage of fluid from the intravascular to the interstitial space. In fact TSS has been often referred to as a 'capillary leak syndrome'. The hypotension observed in TSS patients leads in part to the pathological effect that is observed in almost every organ system. The involvement of almost every organ system is the third distinctive aspect observed in TSS patients. Finally, TSS is associated with various dermatologic sequalae. Upon presentation, TSS patients may exhibit a scarlatinform rash, that 10-21 days after onset of the disease may result in full thickness desquamation, in particular of the palms of the hands and soles of the feet. Nonpitting edema and petechiae may be observed in conjunction with inflammation of mucous membranes, particularly of the oropharynx, conjunctiva and vagina (Chow, 1983). In addition, a 'strawberry tongue' has been described in over 25% of one patient group (Chow, 1983). Recently Kain et al. (1993) attempted to determine by multivariate discriminant analysis if the clinical manifestations associated with NMTSS could be differentiated from those observed in MTSS. Their studies revealed that NMTSS was indeed clinically distinct from MTSS in several ways. Patients with NMTSS had an earlier onset of fever and rash with respect to the other clinical features observed in TSS, had less frequent musculoskeletal involvement, more severe anemia, and more frequent central nervous system complications than patients with MTSS. Thus clinically, distinctions between MTSS and NMTSS can be made. A strict case definition for TSS as outlined by the CDC in the United States has been  5  established (Table 1). Due to the variations in the clinical manifestations that may occur among patients, a diagnosis of probable TSS can be made if one of the six criteria outlined in Table 1 is absent. Since a specific diagnostic test for TSS is lacking, a diagnosis must be made based upon the exclusion of other clinical entities that present similiarly. Attempts are made to rule out the presence of Streptococcus pyogenes, which is the causative agent of scarlet fever, and more recently, has been associated with mediating Streptococcal toxic shock syndrome (Strep TSS)(Cone et al., 1987). In addition, attempts are made to exclude Staphylococcal scalded skin syndrome, Kawasaki's disease, Rocky mountain spotted fever, and the possibilty of drug reactions. The isolation of a TSST-l-producing S. aureus from an infected site and the observation that a patient's sera do not have antibody to the toxin, in conjunction with the hematologic and biochemical abnormalities associated with TSS, provides strong support for a diagnosis of TSS. 1.1.4.  Treatment of Toxic Shock Syndrome  In periods of acute illness, management of TSS requires immediate rapid fluid replacement to counteract the massive leakage of intravascular fluid into the interstitial space, and to maintain adequate tissue perfusion (Chow, 1995). Therapy is aimed at eliminating S. aureus strains from the infected site by. removal of the infected foreign body (for example a tampon or intravenous catheter) followed by antibiotic treatment. Most S. aureus strains exhibit resistance to penicillin and ampicillin, so ^-lactamase resistant anti-staphylococcal agents such as cloxacillin may be used (Chow, 1995). In penicillin-allergic patients, vancomycin or clindamycin may be preferred alternatives (Chow, 1995). Since TSS is toxin-mediated,  6  Table 1 Case Definition for Toxic Shock Syndrome*  1.  Fever (temperature ^ 38.9°C)  2.  Rash (diffuse macular erythredema)  3.  Desquamation, 1-2 weeks after onset of illness, particularly of palms and soles  4.  Hypotension (systolic blood pressure < 90 mm Hg for adults or < 5th percentile by age for children < 16 years of age, or orthostatic syncope)  5.  Involvement of three or more of the following organ systems: A. Gastrointestinal (vomiting or diarrhea at onset of illness) B. Muscular (severe myalgia or creatine phosphokinase level > 2 X ULN") C. Mucous membrane (vaginal, oropharyngeal, or conjunctival hyperemia) D. Renal (BUN ' or Cr > 2 X U L N or ^ 5 white blood cells per high power field - in the absence of a urinary tract infection) E. Hepatic (total bilirubin, SCOT or S G F F > 2 X ULN) F. Hematologic (platelets <, 100,000/mm ) G. Central nervous system (disorientation or alterations in consciousness without focal neurologic signs when fever and hypotension are absent) 1  8  1  3  6.  Negative results of the following tests, if obtained: A. Blood, throat, or cerebrospinal fluid cultures B. Serologic tests for Rocky Mountain spotted fever, leptospirosis, or measles  Twice upper limit of normal for laboratory Blood urea nitrogen level 'Creatine level Serum glutamic oxaloacetic transaminase level Serum glutamic pyruvic transaminase level  b  d 6  •(Chesney et al., 1984)  7  irrigation of the infected site may be warranted to remove any pre-formed toxin. In addition, treatment with pooled intravenous immune globulin (IVIG) preparations, which contain high levels of anti-TSST-1 antibody, have been successful in eliciting protection in animal models of TSS (Melish et al., 1987), and may be warranted in critically ill TSS patients. 1.1.5. Risk Factors Associated with Menstrual and Nonmenstrual Toxic Shock Syndrome Different risk factors associated with menstrual TSS and nonmenstrual. TSS have been noted. In concordance with the observed frequency of TSST-1-producing S. aureus strains from TSS patients, women colonized with TSST-1 strains are at a higher risk for +  developing MTSS than their uncolonized counterparts. Age also appears to be a factor, in \  that women less than 30 years of age appear to be at a higher risk for developing MTSS than women greater than 30 years of age (Davis et al., 1980). This may be accounted for in part by the fact that in a healthy population, the presence of antibodies against TSST-1 correlates closely with age. The majority of individuals develop TSST-1 antibody by the age of 20 (Vergeront et al, 1983). This is consistent with the finding that an increased incidence of MTSS is found in the age group 15-19 years (Broome, 1989). No age-related differences have been found with respect to rate of vaginal S. aureus carriage (Davis et al., 1980), although one study did show a trend toward decreased genital colonization in relation to increasing age (Linnemann et al., 1982). Whether age-related or not, women who lack antibody to TSST-1 have a much higher risk of developing MTSS than those with antibody. This is based on studies which have shown that sera from patients with TSS are significantly more likely to lack 8  antibody to TSST-1 in comparison to healthy controls (Bergdoll et al., 1981; Bergdoll et al., 1982; Notermans et al, 1983; Bonventre et al., 1984). This lack of antibody to TSST-1 is even observed in patients upon convalescence (Bonventre et al., 1984; Stolz et al., 1985), and may explain why recurrences are frequent. The importance of the presence of TSST-1 antibody has been shown in a rabbit model of TSS, in which TSST-1 specific monoclonal antibodies were found to confer protection to rabbits after infection with a nonenterotoxigenic strain of S. aureus carrying a plasmid encoding TSST-1 (Best et al., 1988). A racial predisposition for MTSS has also been suggested, in that the majority of MTSS cases occur in white females (Gaventa et al., 1989; Broome, 1989). This racial distinction does not appear to be due to differences in colonization rates, since black women have been found to be genitally colonized with S. aureus, including TSST-1 S. aureus, as frequently, +  or more frequently, than white women (Linnemann et al., 1982). The fact that less MTSS is noted among this group may result from a failure to note the erythematous rash in women with dark skin, or may be due to some undefined host characteristic (Linnemann et al., 1982). Women who have vaginal carriage of Escherichia coli or other facultative lactosefermenting gram-negative bacilli in addition to S. aureus may also have a greater risk of developing MTSS (Chow and Bartlett, 1982). This association may be consistent with observations that endotoxin can synergize with staphylococcal exoproteins to cause disease in vivo (Sugiyama et al., 1964; de Azavedo and Arbuthnott, 1984). Lastly, tampon usage during menstruation is the best documented risk factor for the development of MTSS. As previously mentioned, the use of one brand of tampon, the Rely*  9  brand, has been more frequently associated with MTSS cases than other brands. The association of one particular brand with disease may be related to the chemical composition of the tampon, which has been shown to play an important factor dictating relative risk potential since this aids in determining both the absorbency and M g  2 +  binding ability  (Berkley et al., 1987; Kass and Parsonnet, 1987; Kass, 1989). The higher the absorbency of the tampon, the higher is the relative risk for developing the disease (Berkley et al., 1987; Reingold et al., 1989). Before 1977, tampons were made of rayon, or a blend of rayon and cotton. After 1977, tampons have been made with more absorbent synthetic materials such as polyacrylate fibres and polyester foam (Shands et al., 1980), and this may have led to an increased incidence of the disease. The ability of certain tampon fibres to bind more Mg " " 2 1  than others is also believed to be important since M g  2 +  concentration has been shown to  regulate TSST-1 production (Mills et al., 1985). Additional factors linking tampon usage with the development of TSS in susceptible individuals have been proposed. The presence of blood on a tampon may provide an important nutrient source that allows increased growth of the organism and/or increased toxin production. In support of this, Lee et al. (1987) provided data to suggest that the presence of blood in the culture medium leads to an increased production of TSST-1. The tampon may also act to provide a larger surface area for the organism to colonize. Tampons may also increase the risk of developing MTSS through their ability to cause microulcerations in the vagina (Friedrich and Siegesmund, 1980), which may subsequently lead to an increased absorption of toxin. Also, aerobic conditions are known to favour TSST-1 production more than anaerobic conditions (Todd et al., 1987), and since tampon insertion into the vagina increases the p02 of the  10  microenvironment (Wagner et al, 1984), more toxin may be produced in vivo and disease may ensue. Tampon usage during menstruation has also been found to cause shifts in the vaginal microflora when compared to the vaginal microflora of the same women using napkins (Chow and Bartlett, 1989). What role this alteration of microflora may play in TSS, if any, is presently unknown. Despite all of the above mentioned methods of how tampon usage may significantly increase the risk of developing MTSS, one of the primary roles of the tampon may be to mediate transport of the organism into the vagina. The tampon may act as an intermediary in the transfer of the organism from the hands or perineum of the woman into the vagina. Evidence that possible contamination of the tampon at the site of manufacture is not to blame is based on a study in which 0 of 264 unused tampons tested harboured S. aureus, despite the fact that tampons were not sterilized in the U.S.A. at this time (Shands et al., 1980). Consistent with the postulate that transmission of the organism from the hand to the vagina via the tampon may be important is suggested by the observation that use of a non-applicator type of tampon increases the risk for vaginal S. aureus carriage (Guinan et al., 1982). Studies addressing risk factors for NMTSS have also been carried out. NMTSS can result from S. aureus infection at any body site, but most commonly in surgical or cutaneous wounds. Unlike that of MTSS, NMTSS can occur in men, women and children of any age who lack protective antibody. Different racial groups appear to be equally susceptible to NMTSS (Reingold et al, 1982). In comparison to MTSS, NMTSS cases occur more significantly in association with a period of hospitalization (Kain et al, 1993), and after a history of previous antimicrobial therapy (Kreiswirth et al, 1986; Kain et al, 1993). Use of  11  barrier contraceptives has also been shown to increase the risk of NMTSS (Schwartz et al., 1989). Infection with S. aureus strains that produce toxins other than TSST-1 are also associated more often with NMTSS cases (Garbe et al., 1985; Schlievert, 1986). 1.2.  Toxin Involvement in the Pathogenesis of Toxic Shock Syndrome  Toxic shock syndrome is a toxin mediated disease. This is based on the observation that despite the frequent isolation of S. aureus from TSS patients and the dysfunction that is observed in multiple organ systems, bacteremia is rarely noted (Chow, 1983). In rabbit models of TSS, animals that succumb to lethal infection by toxigenic strains of S. aureus also exhibit "negative" blood cultures (Rasheed et al., 1985; Melish et al., 1989). The observation that the majority of S. aureus strains isolated from TSS patients are toxigenic (Crass and Bergdoll, 1986a) further strengthens this premise. In addition, convalescent sera from patients with TSS more frequently seroconvert to various staphylococcal exoproteins than patients with non-TSS-associated S. aureus infections (Whiting et al., 1989). Clinically, symptoms such as the abrupt onset of vomiting and diarrhea and the erythema of skin and mucousal surfaces observed in TSS resemble that of a toxin-induced pathology (Chow, 1983). 1.2.1.  Evidence implicating Toxic Shock Syndrome Toxin-1 (TSST-1) in the Pathogenesis of Toxic Shock Syndrome  In 1981, Bergdoll et al. (1981) reported the identification of a marker protein for TSS that was produced in over 90% of TSS-associated S. aureus isolates. After determining that the protein could induce vomiting in baboons, it was called Staphylococcal enterotoxin F (SEF). This name was subsequently changed to Toxic Shock Toxin (TST) when the emetic  12  action of the protein could not be reproduced (Reiser et al., 1983). At the same time, Schlievert et al. (1981) reported the purification and characterization of a toxin produced by a strain of S. aureus that was a potent pyrogen, had nonspecific lymphocyte mitogenic activity, and could enhance host susceptibility to the effects of endotoxin. This protein was referred to as Staphylococcal Pyrogenic Exotoxin C (SPEC). When SEF and SPEC were found to be identical (Bonventre et al., 1983; Bergdoll and Schlievert, 1984), the name was changed to Toxic Shock Syndrome Toxin-1, or TSST-1 (Bergdoll and Schlievert, 1984). Strong evidence exists to implicate TSST-1 as a causal agent in TSS. First, a large majority of strains isolated from TSS patients produce TSST-1. Second, the sera of some individuals with TSS upon convalescence were shown to seroconvert to TSST-1 (Stolz et al., 1985; Whiting et al., 1989). Third, one group of investigators found that S. aureus strains from TSS patients produced significantly more TSST-1 in vitro than toxigenic control strains (Rosten et al., 1987). In addition to the above indirect evidence, a role for TSST-1 in pathogenesis has been more directly assessed with the use of animal models. Rabbits are the species of choice for animal models since they can reproduce the majority of the symptoms observed in TSS upon administration of purified TSST-1 or TSS-associated S. aureus strains (de Azavedo, 1989). Monoclonal antibodies against TSST-1 have protected rabbits from TSST-1 administered by constant infusion (Bonventre et al., 1988), and from challenge by a nonenterotoxigenic strain of S. aureus containing a plasmid encoding TSST-1 introduced into subcutaneously implanted polyethylene chambers (Best et al., 1988). Genetically engineered isogenic strains of S. aureus differing in the ability to produce TSST-1 have also been utilized in rabbit studies. Using the subcutaneous infusion model, Rasheed et al. (1985)  13  were among the first investigators to show the importance of TSST-1 in mediating a TSS-like illness in vivo using genetically constructed S. aureus strains. The construct was made by transducing the TSST-1 marker from a clinical TSST-1 isolate into a TSST-1" strain. +  Although the TSST-1 transductant caused more illness and lethality in rabbits in comparison +  to the TSST-1" control strain, of concern is what other DNA sequences besides the TSST-1* marker may have been transduced into the construct. In addition, although the subcutaneous infusion model allows a controlled, localized depot of infection, it does not closely resemble the host situation observed in menstrual TSS. Shortly thereafter, de Azavedo et al. (1985) constructed a pair of isogenic mutants in TSST-1 production by introducing a plasmid-borne TSST-1 structural gene (tst) into a naturally occurring TSST-1" strain. The isogenic strains were introduced into an uterine chamber that maintained the organisms in a localized region while allowing bacterial exoproteins to be released into the circulation. Using this model, the strain that carried thetar-plasmidinduced a TSS-like illness in rabbits whereas the control strain that carried the plasmid with a deletion in tst had no effect. Criticisms for this experimental design include the fact that the uterine model, although having a genital focus, does not resemble menstrual TSS in which organisms become colonized in the vagina. Also in the experimental design, the rabbits had to be injected daily with erythromycin to maintain the m-carrying plasmid. This requirement may have important implications with regard to the observed clinical outcome, since antibiotic administration can affect the sensitivity of rabbits to TSST-1 (Schlievert, 1983). In a subsequent study, Sloane et al. (1991) created TSST-1 isogenic mutants by the method of allele replacement and introduced the strains into the uterine chamber model. Again it was shown that the TSST-1 strain was more virulent +  14  than the TSST-1" isogenic mutant. The method of allele replacement mutagenesis by these investigators represents an ideal method for the creation of isogenic mutants. If this study had made use of a clinical isolate that more closely resembled a TSS strain in phenotypic characteristics in a more clinically relevant animal model, the results would have definitively addressed the role of TSST-1 in menstrual TSS. A vaginal tampon model described by Melish et al. (1989) was the first relevant animal model reported that reproduced the human situation in menstrual TSS. Although TSST-1  +  staphylococci could lead to TSS in this model, the role of TSST-1 in menstrual TSS was not demonstrated specifically since isogenic strains were not used. To date, no study has been performed with isogenic TSST-1 mutants in a clinically relevant animal model of TSS. 1.2.2. Evidence implicating Staphylococcal Enterotoxins in the Pathogenesis of Toxic Shock Syndrome In addition to TSST-1, other staphylococcal exoproteins, in particular the enterotoxins, may play a role in TSS either on their own, or in conjunction with TSST-1. In support, 60% of TSST-1 strains have been found to co-produce one of the enterotoxins (Crass and +  Bergdoll, 1986a). Staphylococcal enterotoxin production has also been found in TSS isolates that were TSST-1", indirectly implicating them as primary causal agents (Bergdoll et al., 1982; Garbe et ah, 1985). Moreover, introduction of TSST-17enterotoxin TSS-associated +  S. aureus strains into the rabbit subcutaneous model of toxic shock have caused a TSS-like illness to develop (Garbe et al., 1985), suggesting that the enterotoxins can induce similiar symptoms as TSST-1 in vivo. In TSS patients infected with TSST-1" S. aureus strains who could develop an antibody response, seroconversion to at least one of the enterotoxins was  15  found to occur more frequently than in TSS patients infected with TSST-1 S. aureus +  (Whiting et al, 1989). A.  -  Evidence Suggesting that SEA and TSST-1 Co-Production may be Important in the Pathogenesis of Menstrual Toxic Shock Syndrome  As indicated, the production of SEA by S. aureus may also partially contribute to TSS. Upon analysis of 30 NMTSS strains for toxin production, Garbe et al (1985) identified one isolate that produced only SEA. Parsonnet et al (1986) also identified 1 of 20 NMTSSassociated strains that produced SEA alone. In 1990, McCollister et al (1990) isolated a SEA /TSST-1' strain of S. aureus (D4508) from a patient with NMTSS. Concentrated +  culture filtrates of the organism administered subcutaneously via osmotic pumps to rabbits induced TSS-like symptoms, whereas culture filtrates pretreated with antibodies against SEA failed to induce the same symptoms. Further evidence that SEA can induce TSS-like symptoms in vivo was shown in several studies in which purified SEA introduced into rabbits or monkeys could cause symptoms similiar to that observed in humans with TSS, including death (Bergdoll, 1983; Parsonnet et al, 1986; McCollister et al, 1990). Instances in which SEA appears to be the sole agent responsible for causing TSS however are limited. Very few TSS-associated strains have been identified that produce SEA alone. Furthermore, Crass and Bergdoll (1986a) found that of 380 TSS-associated strains examined for TSST-1 and enterotoxin production, none were found to produce SEA alone. Although SEA production on its own may not be that important clinically, striking evidence exists to suggest that SEA production in conjunction with TSST-1 by a S. aureus strain may be responsible for causing the majority of menstrual TSS cases. As indicated in Table 2, the ( 16  combination of TSST-1 and SEA production occurs more frequently among MTSS isolates than NMTSS isolates (Crass and Bergdoll, 1986a; Chang et al., 1991; Kain et al, 1993). In addition, among MTSS isolates, this toxin combination is observed more frequently than any other (Crass and Bergdoll, 1986a). Analyses by multilocus enzyme electrophoresis (Chang et al., 1991) and phage typing (Marples and Wieneke, 1993) suggest that most MTSS cases are caused by a single clone of S. aureus which produce both TSST-1 and SEA. B.  Evidence Suggesting a Putative Role for SEB in the Pathogenesis of Toxic Shock Syndrome in TSST-1' S. aureus Isolates  A putative role for SEB in the pathogenesis of TSS is indicated from the results of several studies in which phenotypic characterizations of TSS-associated strains have been performed. Schlievert (1986) found that SEB production was significantly associated with TSS or probable TSS in isolates that did not produce TSST-1. Garbe et al. (1985) noted that TSST1" TSS-associated S. aureus strains were more likely to produce SEB than TSST-1 strains. +  Crass and Bergdoll (1986b) found that 8 of 9 TSST-1' NMTSS isolates produced only SEB, implicating it as the major etiologic agent causing disease in these cases. Finally, Lee et al. (1992) reported that SEB production among TSST-1" isolates was significantly more prevalent in TSS-associated infections than non-TSS associated infections. As with SEA, injection of SEB into monkeys can reproduce many of the symptoms that are observed in TSS, including shock (Beisel 1972; Liu et al., 1978; Bergdoll 1983). In addition, Van Miert et al. (1984) found that TSST-1 and SEB could induce similiar clinical hematological and blood biochemical changes when administered intraveneously to goats. Collectively, these results suggest that SEB may play a role in causing TSS.  17  Table 2. TSST-1 and SEA Production among S. aureus isolates  Patient Group (no. of isolates)  TSST-1* (%)  SEA* (%)  TSST-1*/SEA* (%)  MTSS (24)  83  75  71  NMTSS (41)  59  32  24  Non-TSS (146)  25  25  12  Carrier (139)  28  20  13  a  MTSS, Menstrual toxic shock syndrome; NMTSS, Nonmenstrual toxic shock syndrome; Non-TSS, Non-toxic shock syndrome-associated S. aureus infection; Carrier, No infection; TSST-1*, producing toxic shock syndrome; SEA* producing staphylococcal enterotoxin A "Significantly different from NMTSS, Non-TSS or Carrier strains (P < 0.001, Chi square, 2-tailed) Data compiled from Chang et al. (1991).  Of interest is the observation that SEB appears to be only associated with TSS strains that are TSST-1". In fact, TSST-1 and SEB co-production by the same S. aureus (Strain is extremely rare (Garbe et ah, 1985; Crass and Bergdoll, 1986a; Crass and Bergdoll, 1986b; Bohach et ah, 1990) and has only been described in a few instances (Ewan et ah, 1989; Johnson et ah, 1991; Lee et ah, 1992). Genotypic studies also suggest that the presence of both the TSST-1 (tst) and the SEB (seb) genes within the same isolate is rare. Investigations using oligonucleotide probes in conjunction with hybridzation failed to identify the presence of seb in TSST-1 isolates (Bohach et ah, 1989; Neill et ah, 1990). However Johnston et +  ah (1991) found the TSST-1 and SEB structural genes in 11% (3/28) of their clinical isolates using PGR analysis. 1.2.3.  Evidence Implicating Toxins other than TSST-1 and the Enterotoxins in the Pathogenesis of Toxic Shock Syndrome  Although TSST-1 and the staphylococcal enterotoxins are produced in the majority of TSS-related isolates, there still remains some that do not express TSST-1 or any of the enterotoxins. Crass and Bergdoll (1986a) found that 9 of 380 (2.4%) S. aureus strains isolated from patients with both menstrual and nonmenstrual TSS did not produce TSST-1, SEA, SEB, SEC, SED, or SEE. In addition, Garbe et ah (1985) identified 3 of 10 NMTSS TSST-1" isolates that did not produce enterotoxin. These results suggest that other toxin(s) besides TSST-1 and the enterotoxins may be responsible for causing TSS in some cases.  S.  aureus produces more than thirty different extracellular proteins, and a detailed examination of the frequency of production of every protein among TSS-associated S. aureus isolates has not been carried out. In 1982, elaboration of an unknown epidermal toxin among TSS-  19  associated strains was described (Kapral, 1982), but evidence to suggest whether it plays a causative role in TSS is lacking. Recently a new toxin, Staphylococcal enterotoxin H (SEH), produced by a TSST-1' nonmenstrual strain of S. aureus was identified (Ren et al., 1994). This new toxin induced an acute TSS-like illness when injected intravenously into rabbits. The percentage of TSS-associated strains producing this toxin has not yet been investigated. Whether SEH may play a role in inducing TSS in TSST-1", nonenterotoxigenic isolates is currently unknown. 1.2.4. Functional Consequences of TSST-1, SEA and SEB Production by S. aureus A.  TSST-1, SEA and SEB as 'SUPERANTIGENS'  TSST-1, SEA and SEB belong to a family of proteins designated as 'superantigens' (White et al., 1989). Two main groups of superantigens - exogenous and endogenous, have been noted. The exogenous group of superantigens are proteins produced by microbes, and in addition to the staphylococcal enterotoxins and TSST-1, include the staphylococcal exfoliating toxins, streptococcal pyrogenic exotoxins (SPE) A , B and C, the M protein molecule and a soluble product secreted by Mycoplasma arihritidis called M A M . The endogenous superantigens are comprised of a group of products encoded by unlinked genetic loci in mice, the best known being the minor lymphocyte stimulating antigens (Mis). These type of antigens have now been found to be encoded by mammary tumor viruses (Frankel et al, 1991). Superantigen is used to describe these unique proteins since they can stimulate a large number of T cells at a lower concentration than that required by a conventional antigen. Moreover, these antigens stimulate only T cells bearing specific V components of the T cell 6  20  receptor. Each superantigen has its own unique V profile (Schlievert, 1993). TSST-1 B  stimulates human V 2 T cells (Choi et al, 1990), SEB V 3,12,14,15,17,20 T cells (Choi et B  8  al., 1989), and SEA V 6.3,6.9,7.3,7.4 T cells (Zumla 1992). This unique feature of B  superantigens enables them to interact with more T cells than a conventional antigen since the actual region contacted by the superantigen on the V chain of the T cell receptor is relatively B  invariant within each V„ gene family (Drake and Kotzin, 1992). In addition, superantigens differ from conventional antigens in other aspects: Superantigens do not require processing by the antigen presenting cell before binding to the MHC Class U molecule. Instead, superantigens interact with the MHC Class II molecule outside of the conventional peptide binding groove, and thus superantigen presentation by T cells is not MHC-restricted. Superantigens are unique molecules capable of triggering profound disturbances within the immune system. In addition to causing massive stimulation of T cells, superantigens have been found to induce T cell nonresponsiveness, either through functional inactivation (anergy) (Rellahan et al., 1990; O'Hehrir et al., 1990) or clonal deletion (MacDonald et al., 1991; Kawabe and Ochi, 1991). Superantigens also mediate staphylococcal enterotoxindependent cell-mediated cytotoxicity, or SDCC, in which rapid killing of antigen presenting cells, irrelevant of their nominal specificity, is evoked (Kalland et al., 1991). The cytotoxic activities can be mediated by both CD4 and C D 8 T cells (Dohlsten et al, 1990) and T +  +  cells expressing the gamma-delta T cell receptor (Rust et al., 1990). This latter effect also has the potential for functional inactivations of T-cell mediated responses. It may be speculated that the production of these toxins by S. aureus may represent a method by which the organism avoids recognition by T cells (Kalland et al., 1991). Indeed, SDCC might play  21  a role in inhibiting immunoglobulin production by B cells, since Class U B cells would also +  be destroyed in such a cytotoxic response. As proposed by Mourad et al. (1993), downregulation of the humoral response by these superantigens may provide a putative explanation for why TSS patients do not generate TSS specific-antibodies and why recurrences of this disease are frequent. B.  The M H C Class II Molecule is the Receptor for TSST-1, SEA and SEB  Although TSST-1 is non-cytotoxic for a variety of cell lines (Drumm et al., 1989), specific binding to cervical epithelial (Kushnaryov et ah, 1984a; Kushnaryov et al., 1984b), and vascular endothelial (Kushnaryov et al., 1989) cells has been observed. Binding of TSST-1 to epithelial cells leads to internalization via receptor-mediated endocytosis (Kushnaryov et al., 1984b). Once transported to the inside by epithelial cells, TSST-1 may interact with endothelial cells to become translocated from the extravascular space across the vessel wall into the lumen, where interaction with peripheral blood mononuclear cells may occur. Many researchers have provided evidence to suggest that the M H C Class II molecule on accessory cells such as antigen presenting cells (APC) is the receptor for TSST-1, SEA and SEB (Fischer et al., 1989; Herrmann et al., 1989; Mollick et al., 1989; Scholl et al., 1989a; See et al., 1992a). Although all three toxins bind to the same receptor, unique binding epitopes have been observed. Scholl et al. (1989b) showed by cross-competition binding studies that SEB and TSST-1 bind to distinct sites on both HLA-DR and HLA-DQ transfected cell lines. Two distinct binding sites for TSST-1 and SEB on HLA-DR were also shown by Chintagumpala et al. (1991). Consistent with this, See et al. (1990) were unable to inhibit labelled TSST-1 binding to human peripheral blood mononuclear cells with excess  22  unlabelled SEB. In contrast, the same investigators were able to show that excess unlabelled SEA could inhibit labelled TSST-1 binding to human peripheral blood mononuclear cells, suggesting that TSST-1 and SEA may bind to overlapping epitopes. Similiarly SEA was shown by Fraser (1989) to compete with SEB for binding to HLA-DR. It was thus postulated that SEA may bind to HLA-DR at a site that overlaps the binding epitopes for TSST-1 and SEB (See et al, 1992a). Isotypic and allelic polymorphisms of the MHC molecule have been shown to influence the binding of staphylococcal toxins. TSST-1, SEA, and SEB bind preferentially to the MHC isotypes in the order HLA-DR>DQ>DP (Scholl et al, 1989a; Herman et al, 1990; Uchiyama et al, 1990; Imanishi et al, 1992). Preferential binding of these toxins to specific HLA-DR alleles expressed on L-cells has also been demonstrated (Herman et al, 1990; Scholl et al, 1990). In addition to the MHC Class II molecule, a second toxin binding site for superantigens on murine macrophages may be present. In a recent report by Beharka et al (1994), macrophages obtained from Class II deficient transgenic mice were found to bind various staphylococcal exotoxins and induce cellular activation. Other investigations have also indicated that staphylococcal enterotoxins might bind to sites other than the MHC Class II molecule (Dohlsten et al, 1991; Herrmann et al, 1991). These putative non-MHC Class II toxin binding sites still remain to be defined. C.  Activation of Monocytes by Staphylococcal Superantigens  Interaction of staphylococcal superantigens with the MHC Class II molecule on monocytes leads to transcription of genes for monocyte-derived cytokines such as IL-1B and TNFa  23  (Trede et al., 1991). It has been found that direct contact between monocytes and T cells is required for purified TSST-1 to stimulate TNFa and IL-lfl secretion (See et al. 1992c). Furthermore, studies by Gjorloff et al. (1991), and Fisher et al. (1990) showed that SEA induction of IL-1 and TNFa from monocytes required the participation of T cells. The binding of superantigens to MHC Class II molecules on monocytes is a prerequisite for cytokine induction, as evidenced by the fact that antibodies to Class U molecules in the presence of staphylococcal superantigens can diminish TNFa and IL-1B production (Grossman et al., 1990; Matsuyama et al., 1993). Superantigens can induce TNFa release from monocytes independent of T cell mitogenesis since a SEA mutant toxin that was nonmitogenic for T cells was still able to bind to Class II molecules and induce TNFa release (Grossman et al., 1992). In addition Fisher et al. (1990) used an intracytoplasmic staining technique to show that intracellular TNFa production did occur in monocytes in the presence of T cells after stimulation with SEA. D.  Activation of T Cells by Staphylococcal Superantigens  Interaction of the superantigen/MHC Class n complex with specific V components of the B  T cell receptor results in the activation of T cells. Upon activation, T cells proliferate and secrete cytokines such as TNF, IL-2 and IFN-5. The dependency on the MHC Class II molecule for mediating superantigen-induced T cell activation was suggested by the observation that monoclonal antibodies against the HLA-DR molecule inhibited both TSST-1 and SEA induced T cell proliferation (Fleischer and Schrezenmeier, 1988; Fischer et al., 1989; See et al., 1992a). The secretion of a T cell specific cytokine, TNFB, has also been found to be dependent on the MHC Class II molecule, since T cells from DR4-negative  24  donors, stimulated with TSST-1, were found to secrete TNFfl only in the presence of L cells transfected with DR4 genes and not control L cells (Akatsuka et al., 1994). Moreover this response could be abrogated by the addition of an anti-HLA-DR monoclonal antibody. Although T cell activation by superantigens is dependent on Class U M H C , evidence suggests that superantigens can interact with T cells in the absence of the antigen presenting cell to induce intracellular signals through the T cell receptor. Fleischer and Schrezenmeier (1988) showed that staphylococcal enterotoxins incubated with purified resting T cells caused a rise in intracellular C a  2+  concentrations. Lagoo et al. (1994) also showed that superantigens  induced cytokine gene expression but not cytokine secretion in highly purified blood T cells in the absence of Class n  +  cells. It therefore appears that although signals are induced to the  T cell upon interaction with superantigens, additional accessory signals mediated by crosslinking of the T cell receptor, Class II molecule, and adhesion molecules are required for complete T cell activation. Both C D 4 and C D 8 T cells can be activated to proliferate and secrete cytokines upon +  +  superantigen exposure in the presence of APC (Calvano et al., 1984; Fischer et al., 1990). Activation of C D 8 cells by superantigens has been associated with an inhibition of +  immunoglobulin production (Poindexter and Schlievert, 1987). Class I molecules do not appear to be necessary for the CD8 T cell response since Carlsson et al. (1988) found that +  Daudi cells, which lack Class I molecules, are able to support C D 8 T cell activation. +  Likewise, although C D 4 cells are activated, the presence of CD4 on T cells does not appear +  to be absolutely necessary for the Class II mediated effects since transfection of CD4" murine T cell hybridomas with CD4 cDNA did not affect the proliferative response of the majority  25  of the hybridomas to various superantigens (Sekaly et al., 1991). These data suggest that the V segment on a C D 8 cell is also capable of being cross-linked to M H C Class n molecules +  fl  by superantigens. £.  Evidence Implicating a Central Role for T Cell Mitogenesis and Cytokine Secretion in the Pathogenesis of Superantigen-Mediated Shock  Since the staphylococcal toxins believed to be responsible for causing TSS belong to the superantigen family of proteins, it is reasonable to hypothesize that these unique superantigen-related activities might contribute to the pathogenesis of TSS. The initial immunological response induced by these toxins upon interaction with MHC Class II molecules and T cells is the expansion of T cell clones bearing specific V molecules. In B  addition to proliferation, T cells are activated to release cytokines such as TNFa, IL-lfl, BL-2 and IFN-T. The postulate that T cell mitogenesis followed by excessive secretion of cytokines, particularly TNFa, plays a critical role leading to lethal shock upon superantigen stimulation is based on collected evidence from numerous studies. An indirect role for T cells in TSS was suggested by Choi et al. (1990), who found that during TSS, T cell stimulation occurs on a scale that is not observed in responses to conventional antigens. Direct evidence for a primary role of T cells in the toxicity induced by SEB was shown by Marrack et al. (1990). These investigators found that both T cell deficient nude mice, and mice genetically bred to express a low number of SEB reactive V -T cells, experienced less B  weight loss than control mice upon injection of SEB. These investigators concluded that SEB-induced weight loss in mice was mediated by T cells. Experiments confirming these observations were performed by Miethke et al. (1992) using D-Galactosamine (D-gal)  26  sensitized mice. D-galactosamine is a hepatotoxic agent that sensitizes mice to the biologic effects of superantigens as well as endotoxin by unknown mechanisms. D-Gal sensitized SCID mice, deficient in T and B cells, did not succumb to lethal shock upon injection with SEB. Upon reconstitution with T cells, however, the mice died. The same investigators performed the identical experiments with TSST-1, and also demonstrated a requirement for T cells in TSST-1-mediated lethal shock (Meithke et ah, 1993). These experiments provide the most direct evidence to support a primary role for T cells in mediating lethal shock. In a complementary investigation, Bonventre et al. (1993) identified a critical residue in the TSST-1 molecule that was required for mediating the mitogenic action on T cells. Substitution of the histidine residue at position 135 to an alanine by site-directed mutagenesis yielded a TSST-1 mutant (H135) that retained MHC Class II binding ability, but lost the ability to induce T cell activation (as measured by mitogenesis and cytokine assays) (Bonventre et al., 1993; Cullen et ah, 1995). The mutant toxin failed to induce death in a rabbit model of TSS (Bonventre et al'., 1993) or a murine model of lethal shock (Bonventre et ah, 1995) whereas the wild-type toxin was lethal. TNF has been implicated as the central mediator in the pathogenesis of TSS. TNF introduced into experimental animals can induce most of the symptoms that are observed in TSS patients, including fever (Dinarello et ah, 1986), hypotension, and shock (Ikejima et ah, 1988; Dinarello et ah, 1989). Many studies have shown that anti-TNF monoclonal antibodies can confer protection against the toxic effects of both S. aureus and its purified toxins. For example, antibody against TNF was found to protect rabbits against a lethal infusion of purified TSST-1 (Parsonnet et ah, 1988); an anti-TNF antibody was found to  27  /  prevent death in baboons infused with live S. aureus (Hinshaw et al., 1992); an anti-TNF monoclonal antibody was found to provide protection to mice in a dose-dependent manner against the lethal effects of TSST-1 (Miethke et al, 1993); and anti-TNF antibody was found to protect mice against the lethal effects of SEB (Miethke et al., 1992). Evidence that T cells might be the source of the cytokine(s) mediating toxicity comes from the observation that cyclosporin A (an agent that suppresses lymphokine secretion by T cells) treatment of DGal sensitized mice conferred protection against the lethal effects of SEB (Miethke et al., 1992) and TSST-1 (Miethke et al, 1993). Whether the T cell requirement for toxicity is due to the direct release of TNF from T cells, or is due to another cytokine released from T cells that enhance TNF production from monocytes and/or T cells is not clear. Although evidence exists to suggest that TNF is the primary mediator leading to TSS, other cytokines, such as IL-1, may also be important. IL-1 and TNF have been found to act synergistically to induce a shock-like state in rabbits (Ikejima et al., 1988; Okusawa et al., 1988; Dinarello et al., 1989), and each can induce the production of the other. In addition, TNFa and IL-1B can induce the in vivo production of IL-6 (Shalaby et al., 1989). IL-6 may in fact be an important mediator of TNFa and IL-1B functions in vivo since injection of both cytokines in mice correlated with a greater increase in serum IL-6 and mortality compared to animals injected with either cytokine alone (Shalaby et al., 1989). 1.2.5. A.  Genetic Control of TSST-1, SEA and SEB Production in S. aureus Regulatory Loci Involved in Controlling Toxin Production  Most staphylococcal exoproteins including TSST-1 and SEB, are produced maximally during the post-exponential phase of growth in vitro. Other exoproteins are produced during  28  the exponential phase of growth, including coagulase, Protein A and SEA (Czop and Bergdoll, 1974; Borst and Betley, 1993). During the transition from exponential to postexponential growth, regulatory systems alter the expression of accessory traits. One regulatory locus in S. aureus that appears to be involved in gene activation during this transition is the 'accessory gene regulator', or agr (Recsei et al., 1986) [previously referred to as exp (Morfeldt et al., 1988)]. This locus regulates a number of S. aureus exoprotein' genes, in that it can be both stimulatory and inhibitory for various gene products. For example, while mutations in agr led to the undetectable production of proteins such as ahemolysin (Recsei et al., 1986), TSST-1 (Recsei et al., 1986) and SEB (Gaskill and Khan, 1988), others such as coagulase and Protein A were increased (Recsei et al., 1986). Due to the differential effects on various proteins, it seems probable that agr may act through a number of other intermediate regulatory steps to mediate its effect on gene expression. In this regard, three other related regulatory systems in S. aureus have been identified. Smeltzer et al. (1993) identified a locus called xpr that was also found to be responsible for regulating the production of a number of extracellular staphylococcal proteins. Transposon insertion mutagenesis into xpr led to a decrease in the expression of extracellular proteins that were similiar to the proteins found to be decreased in agr mutants. It was proposed that xpr and agr may interact to regulate genes of pathogenic importance. A second locus called sar, for staphylococcal accessory regulator, was also identified. Sar appears to be involved in regulating the expression of extracellular and cell-wall associated proteins (Cheung et al., 1992). This locus appeared distinct from agr, since sar mutants produced more a-hemolysin and protease than the wild-type, in direct contrast to the agr  29  mutants. The sar locus may therefore act as a counter-regulatory system to that of agr (Cheung et al, 1992). Evidence for a joint role for both loci in the regulation of exoproteins was suggested by Cheung and Projan (1994), who found that optimal transcription of RNATH (a transcript of agr that is believed to be the agr-specific regulator of exoprotein synthesis) was dependent on an intact sar locus. Finally, a third locus designated sae for S. aureus exoprotein expression, has been identified (Giraudo et al., 1994). Mutations in this locus result in diminished or null levels of a- and B-hemolysins, DNAase, coagulase and Protein A. Production of lipase, staphylokinase, SEA and proteases however were not affected. As a result of the novel phenotype of these mutants, this locus appears to be distinct from that of agr, xpr or sar. TSST-1 is under control of the agr locus. Agr exerts its effect on TSST-1 production at the transcriptional level, since agr mutants have less TSST-1 mRNA in comparison to wildtype cells (Rescei et al., 1986). SEB is also under control of the agr, and regulation appears to be at the transcriptional level (Gaskill and Khan, 1988). In addition to agr, the xpr locus may also play a role in regulating SEB production, since xpr mutants were found to produce undetectable levels of SEB (Smeltzer et al, 1993). In contrast to TSST-1 and SEB, SEA is not under control of the agr (Tremaine et al, 1993). B.  Genetic Analysis of TSST-1 Production  TSST-1 is a 22 kD exoprotein produced by some strains of S. aureus. Initial purification and biochemical analyses of the toxin were carried out by Bergdoll et al (1981), Schlievert et al. (1981), and Blomster-Hautamaa et al. (1986b). The TSST-1 structural gene was cloned into E. Coli on a 10.6 kb chromosomal fragment in the plasmid pRN6100 from an  30  alpha-hemolysin negative derivative of the S. aureus strain 3-14 (Kreiswirth et al., 1983). The nucleotide and partial amino acid sequence of TSST-1 was determined by BlomsterHautamaa et al. (1986a). The TSST-1 structural gene (tst) consists of 708 nucleotides. The toxin has a 40 amino acid signal peptide that upon cleavage yields a 194 amino acid mature toxin molecule. Variability in TSST-1 production observed among S. aureus strains is determined by the presence or absence of tst, since the concordance between TSST-1 production and detection of the chromosomal gene with oligonucleotide probes is >99% (Bonventre et al., 1989). The TSST-1 determinant is not plasmid linked (Kreiswirth et al., 1984). Initial studies found that S. aureus strains isolated from patients with TSS were more frequently lysogenized by temperate bacteriophage than non-TSS-associated S. aureus strains. This observation led to the suggestion that the TSST-1 determinant may be carried by a bacteriophage (Schutzer et al., 1983). This theory was refuted by Kreiswirth et al. (1983), when transfer of the TSST1 determinant by lysogenic conversion could not be demonstrated. DNA from TSST-1" " 1  isolates hybridized to probes derived from sequences flanking the TSST-1 gene, whereas DNA from TSST-1' isolates did not. This suggested that the TSST-1 gene was located within a larger segment of DNA present only among TSST-1 isolates. Subsequently, +  Kreiswirth et al. (1989) reported that the TSST-1 gene is located on a unique genetic element of ~ 4-7 kb in size. Digesting whole cell DNA with Clal (a restriction enzyme that does not cleave within the TSST-1 genetic element) followed by hybridization with a TSST-1 specific gene probe revealed Clal fragments of various sizes. This indicated that the TSST-1 genetic element may be mobile. Consistent with this was the observation that protoplast fusion and  31  transformation mapping had placed the TSST-1 gene in at least two places - very close to the tryptophan (trp) operon in some strains, and very close to the tyrosine B (tyrB) locus in another strain (Chu et al., 1988). The trp operon appeared to be the preferred location for the TSST-1 genetic element, since the vast majority of TSST-1 S. aureus strains were +  tryptophan auxotrophs (Chu et al., 1985; Kreiswirth et al., 1989). The observation that the majority of trp' isolates (51 out of 58) shared common phenotypic characteristics and a common pattern by Southern hybridization with a tst probe suggest that these trp' isolates are clonal in nature (Kreiswirth et al., 1989). Multilocus enzyme electrophoresis (MLEE) analyses indicated that a single clone of S. aureus (ET41) was responsible for the majority of female urogenital TSS cases (Musser et al., 1990). M L E E analysis also suggests that since the TSST-1 gene occurred in a number of various clonal lineages representing diverse species of S. aureus, the gene is evolutionarily old, and hence, has not been recently evolved or acquired by S. aureus (Musser et al., 1990). To date, the TSST-1 genetic element is referred to as Heterologous Insertion 555 (Chu et al., 1988). This term is used to indicate the presence of an extra segment of chromosomal DNA in an organism associated with a specific trait (Kreiswirth et al., 1984). The observation that the TSST-1 determinant was not able to be transferred via transduction, in conjunction with 4-7 kb being the estimated size of the TSST-1 genetic element (a size that is smaller than the genome of any staphylococcal phage) provides evidence to support that the TSST-1 determinant is not mediated by phage. In 1989, the original investigators who studied the molecular biology of the TSST-1 determinant proposed that the TSST-1 genetic element is a staphylococcal transposon (Kreiswirth et al., 1989). To date however, no  32  experiments to demonstrate that the TSST-1 genetic element can undergo successive transposition to nonhomologous targets have been reported. The exact nature of the TSST-1 genetic element thus remains unclear. C.  Genetic Analysis of SEA Production  Staphylococcal enterotoxin A is the toxin that is most frequently associated with staphylococcal food poisoning outbreaks caused by the ingestion of preformed toxin in contaminated food sources (Minor and Marth, 1971). In vitro, S. aureus produces very small amounts of SEA, usually not exceeding 5 - 1 0 /xg/ml of medium (Iandolo and Dyer, 1981). In 1978, the SEA structural gene (sea) was proposed to reside on the chromosome since SEA production was not associated with detectable plasmid DNA in a number of isolates (Shafer and Iandolo, 1978b). Chromosomal mapping experiments revealed a preferred location for sea, since 24 of 29 strains had sea located on the chromosome between thepur-110 and i/v-129 markers (Pattee and Glatz, 1980; Mallonee et al., 1982). Subsequently, sea was cloned on a 2.5 kb HindTK fragment and expressed in E. coli (Betley et al., 1984). Further mapping studies by Betley et al. (1984) verified that although sea was located at the pur-ilv site in some strains, other chromosomal locations could be found. The SEA determinant was thus proposed to be mobile, and analysis by Southern blotting found it to be associated with a DNA segment of 8-12 kb in size. Shortly thereafter, Betley and Mekalanos (1985) reported that sea was carried by a family of related phages that differed by restriction fragment length polymorphisms (RFLPs) surrounding the gene. The variability in the sequences of the phages led to differences in the upstream promoter regions of sea, which played a role in dictating the amount of SEA produced by each strain (Borst and  33  Betley, 1994). The phages that carry a functional SEA gene may be either viable or defective (Betley and Mekalanos, 1985). In addition, related phages that do not carry sea have been found (Betley et al., 1984; Betley and Mekalanos, 1985). There was also a report of a phage that carried a SEA-like gene, called sezA (Soltis et al., 1990). Although sezA +  +  had a high degree of nucleotide identity with sea, there was no translation into an enterotoxin-like protein due to the lack of a translation initiation codon. Recently, sea has been found associated with triple converting phages (Coleman et al., 1989). These novel serotype F staphylococcal phages mediated the simultaneous triple lysogenic conversion for B-lysin, staphylokinase and enterotoxin A . The structural genes for staphylokinase and sea were found closely linked in the phage genome. Insertional inactivation of the B-lysin determinant occurs upon lysogenization of the triple converting phage, yielding a B-lysin", staphylokinase , enterotoxin A strain. +  +  In 1988, the nucleotide sequence of sea was reported (Betley and Mekalanos, 1988). The gene is 771 bp and encodes a SEA precursor of 257 amino acids, that upon processing yields the mature form of SEA with a molecular mass of 27,100 kD. D.  Genetic Analysis of SEB Production  Similiar to TSST-1 and SEA, SEB is another example of a variable genetic trait in S. aureus. SEB is produced more abundantly in vitro than SEA, usually exceeding 100 /*g/ml r  of medium (Iandolo, 1990). There is wide variation with respect to the amount of SEB produced by various S. aureus strains. The deduced nucleotide sequences of seb from a high SEB producer and from a low SEB producer were similiar, negating the possiblity that nucleotide sequence differences in the gene or promoter region were responsible for the 34  observed variability (Compagnone-Post et al., 1991). One possibility is that variations in SEB levels between strains arise from variations in cellular proteins involved in transcriptional activation. In this regard, the cloned SEB gene from a high and low producer, when introduced on a plasmid into the same strain, were found to produce comparable amounts of both SEB and SEB mRNA (Compagnone-Post et al., 1991). Thus unknown host factors may play a role in regulating SEB production. One such host factor may be the delta-lysin RNA. High SEB producers were shown to have in general more delta-lysin mRNA than low SEB producers (Compagnone-Post et al., 1991). Initial studies attempting to analyse the SEB determinant in S. aureus focussed on the methicillin resistant strain DU4916. Studies with DU4916 led to the suggestion that SEB production was plasmid mediated in this strain (Dornbusch and Hollander, 1973). Subsequently Shafer and Iandolo (1979) found that seb was chromosomal in two methicillin resistant plasmid-negative strains. It was then concluded that seb could be located either on a plasmid or the chromosome. Further studies by Dyer and Iandolo (1981) led to the suggestion that the plasmid believed to be associated with SEB production (called pSN2) may not contain seb, but may be involved in the regulation of SEB synthesis. Shortly thereafter, Khan and Novick (1982) refuted a regulatory role for pSN2 in SEB production. Their results were based on experiments in which cured derivatives of DU4916 were still found to produce SEB. DNA-DNA hybridization experiments verified that pSN2 had not integrated into the chromosome of the derivatives. Next, the nucleotide sequence of pSN2 was determined, and found to encode a polypeptide of ~ 20 kD. This was different in size than the 28 kD protein that corresponds to SEB. Through their studies, Khan and Novick (1982)  35  thus tentatively labelled the SEB gene in DU4916 as chromosomal. The most recent report to examine the genetics of SEB production appeared in 1988. Johns and Khan (1988) provided evidence to suggest that seb is associated with a genetic element of at least 26.8 kb in size. SEB isolates were shown to have identical sized +  restriction fragments with homology to a probe putatively identified as encompassing the right hand junction of the SEB genetic element and common chromosomal DNA, whereas SEB" isolates had heterogenous sized fragments to the probe. This prompted the investigators to conclude that the 26.8 kb region surrounding seb (termed the SEB genetic element) is found in one chromosomal location in these strains. As observed for the genetics of SEA production, some SEB" strains, although lacking seb, had sequences that were homologous to the SEB genetic element. Whether this represented an incomplete or related element was not established. Johns and Khan (1988) theorized that the SEB genetic element could be a bacteriophage. The possibility that it was a common transposon was ruled out since staphylococcal transposons as large as 26.8 kb have not been described. The size of the SEB genetic element is similiar to that of other phages, including the phage that carries sea. Furthermore, the location of seb at approximately 1.5 kb upstream of one end of the element is consistent with the location of other toxin genes carried by other phages (Betley and Mekalanos, 1985; Johnson et al., 1986). Attempts to induce putative phage carrying the SEB structural gene however have been unsuccessful. This does not exclude the possibility that SEB may be phage-mediated, since defective phage that are unable to produce infectious particles upon induction may be present. Alternatively Johns and Khan (1988) hypothesized that seb could be part of a large plasmid  36  that had integrated into the chromosome. They noted that Altboum et al. (1985) had described a 56.2 kb plasmid, called pZAlO, that frequently integrated and excised from the chromosome with concomitant rearrangements in DNA sequence. Upon transfer of pZAlO into SEB" strains, several transformants became SEB . Further support for plasmid location +  of seb came from the observation that the G-C content of seb is lower than the majority of S. aureus chromosomal genes, but similiar to that of other S. aureus plasmid genes (Jones and Khan, 1986). To date, no further work has been reported in this regard, and thus the exact nature of the SEB genetic element and its chromosomal location still remain elusive. 1.3.  Focus of Thesis Proposal  Staphylococcus aureus is a complex pathogen that produces many extracellular proteins including TSST-1, SEA and SEB, all of which have been implicated in causing TSS. TSST1 is believed to be the primary toxin involved in causing the disease, since much evidence is present to support an etiologic role. Unfortunately however, key experiments addressing its biological role in menstrual TSS have not yet been performed. These experiments consist of using isogenic mutants in TSST-1 production in a relevant animal model. The primary objective of this research was to determine if a role for TSST-1 in menstrual TSS could be established by performing the above experiment. Upon establishment of a role for TSST-1 in menstrual TSS, interaction of TSST-1 with other staphylococcal exoproteins, specifically SEA and SEB, on both a functional and molecular level was warranted. The interaction of TSST-1 with these specific toxins are believed to have important clinical significance due to two very important observations that have been made among TSS-associated S. aureus isolates. First is the observation that MTSS isolates of S. aureus co-produce TSST-1 and  37  SEA more frequently than NMTSS, non-TSS or carrier isolates (Chang et al, 1991). Second is the observation that production of TSST-1 and SEB by S. aureus is mutually exclusive or that co-production of these toxins is extremely rare (Garbe et al., 1985; Crass and Bergdoll, 1986a; Crass and Bergdoll, 1986b; Ewan et al, 1989; Bohach et al, 1990; Johnson et al, 1991; Lee et al, 1992). Co-production of TSST-1 and SEA by S. aureus is postulated to have important functional consequences in vivo due to the designation of both of these toxins as superantigens. The coexpression of two such proteins by a S. aureus strain might lead to a more virulent strain than production of a single 'superantigenic' toxin. If true, this might provide a putative explanation for why a TSST-1 /SEA clone of S. aureus is frequently associated with +  +  MTSS. What role SEA may play in conjunction with TSST-1 specifically in MTSS however is unclear. In addition to TSST-1, SEB has also been suggested to play a role in TSS. Therefore, the phenotypic observation that TSST-1 and SEB production among S. aureus strains appears mutually exclusive has important clinical implications, and a novel regulatory system involved in mediating the control of TSST-1 and SEB expression may exist. Although it is generally accepted that this toxin combination rarely exists at the phenotypic level, controversy still exists at the genotypic level over whether tst and seb can be found within the same isolate. In light of this controversy, an examination of S. aureus isolates for the presence of the TSST-1 and SEB genes and determination of the molecular mechanism(s) involved in regulating this toxin phenotype, was warranted.  38  In view of the above, the following Questions provided the focus for the thesis proposal: 1.  Is there conclusive evidence that TSST-1 plays a causative role in the pathogenesis of Menstrual TSS?  2.  Does co-production of TSST-1 and SEA by a S. aureus strain mediate increased virulence?  3.  Is the "mutually exclusive" production of TSST-1 and SEB due to the absence of structural genes in the chromosome?  4.  What are some possible mechanisms that may explain the "mutual exclusivity" of TSST-1 and SEB co-production?  39  Chapter 2. Materials and Methods  2.1.  Bacterial growth media and antibiotics  Escherichia coli strains were propogated on Luria-Bertani (LB) media (Sambrook et al., 1989) containing antibiotics where appropriate (ampicillin 50 /ig/ml; ampicillin 4- methicillin 80 and 20 /tg/ml respectively; erythromycin [Em] 100 uglraX). Staphylococcus aureus strains were propogated on Tryptic Soy Agar (TSA), Tryptic Soy Broth (TSB) or Brain Heart Infusion media (BHI) (Difco, Detroit, MI) containing antibiotics where appropriate (Em 10 jig/ml; chloramphenicol [Cm] 10 j*g/ml). 2.2.  Bacterial strains and plasmids  The Staphylococcus aureus control or reference strains and clinical isolates used in this study are listed in Tables 3 and 4 respectively. Plasmids used in this study are listed in Table 5. 2.3. 2.3.1.  Phenotypic Analyses of S. aureus Toxin Production A.  Immunoblotting  S. aureus isolates were grown overnight in either BHI or TSB containing antibiotics where appropriate. The post-exponential phase cultures were centrifuged; (1,200 x g; 10 min) and the supernatant was filter-sterilized. Proteins in the culture supernatants were concentrated by either lyophilization or ultrafiltration through a Centricon-10 or Microcon-3 concentrator (10,000 and 3,000 molecular weight cut-offs respectively; Amicon, Beverly, MA). 40  I  ON  ON 00  o\  Q M  5  3  12  2  I  00 ON —<  o 00 »- OS %> —<  |^  •e r •1*5  3  * «> 3 s  U  00  O  3  <  <  "o  t/J  m  3 c  O  .8  1 C  -8  + <  '6 'S  oo  s o  co +  Pi  I3 <U  s 3  Z, T  S o  O  "2 00  ^» 00  S3 CO  o Z GO .2 2 H  3  •a »  6 0 S  ON  o\  U  cu  u D C '« tl  ^  ON 00  si  u o  < t/: 7  CO  v. oo co  Z to  o U  co co  CO  s •< E O  0]  Z co  rr +  O m Z co  co  co  *7 +  s  f2  S <-> Z co  •o 1 + c ee u u & c 2 +  s  a a+ — CD i:  3 u  M  Q. + 3  o  1  s: *-  CO  co  •>»•  00 ON  RN7i  o  RN3  E-  o  RN4  X)  tc c  Csl  «/"> oe  00 IN  ^ _  v> 00  00  00  41  2  Table 4 Staphylococcus aureus Clinical Isolates Used in this Study Cinical Isolates 7051 7054 7128 7140 7142 7145 7163 7294 7333 7335 7370 7371 7372 7385 7402 7405 7407 7423 7429 7431 7513 7599 7616 7660 7690 7692 7694 7695 7785 7835 7836 8254 8270 8292 8297 8315 8385 8395 8426 VI-5  Phenotypic Results TSST-1+ SEB+  + + + + + + + + + + + + + + + + +  Genotypic Results tst+ seb+  +  + +  nd  +  +  + + + +  +  + + +  nd  + + + +  nd + nd  + +. + + + + nd  + + + + + + +  + + + + + + +  + + + +  nd + + + + +  Electrophoretic Type (ET) 5 6 6 21 6 6 6 21 5 22 6 6 6 6 6 6 5 6 6 6 6 6 21 10 49 6 50 51 48 5 50 59 21 50 8 6 62 63 6 nd  Clinical Characteristics Carrier Carrier Carrier Carrier Carrier Carrier Carrier TSSNM TSSNM TSSNM Carrier Carrier Carrier TSSNM Non-TSS Non-TSS Non-TSS Non-TSS Non-TSS Non-TSS TSSNM TSSNM Carrier Carrer Non-TSS Non-TSS Non-TSS Non-TSS TSSM Non-TSS TSSNM Carrier Carrier TSSNM Carrier Carrier Non-TSS Non-TSS Non-TSS TSSNM  Abbreviations: TSST-1+, producing toxic shock syndrome toxin-1; SEB+, producing staphylococcal enterotoxin B; tst*, containing TSST-1 structural gene; seb*, containing SEB structural gene; TSSNM, nonmenstrual toxic shock syndrome, T S S M , menstrual toxic shock syndrome; Non-TSS, non-toxic shock syndrome-associated S. aureus infection; Carrier, no infection; nd, not determined. E T analysis was performed by J. M . Musser in the laboratory of R. K. Selander (Pennsylvania State University)  42  r00  ON  —'  &  g  8  c  Q  w-> 00 —  IS  3  3  (A  c  la  '1 3  "H.  a  oo •ft  3  00 B  i>  >  ll  00  •c o  s  8  l l  O  — o  00  1  2 m  5 a § « 5.8  2  3 8  2 "3.  Q.  c 'Si  oo c  00  T  '5  •is  8 o  82  " 5 c  3 2 = o.  e p  U 13 3 oo &;=  5 s  3 8 •  <  a.  <  a.  <  a.  <  s  a. <  u  a. <  E B.  9[x] < £  •i E  '•5 |8 B  2  3  |5  s  8  8  CO  ft -3  *5  3  6c 1*3  a p Co bj  8  s p  E  IW B  I ."2 E  10.  u  SO  o Q  vo roe vo D -J  1^ 00  Q •—i  a.  c  u 'a.  E co  "a.  K  43  Lyophilized samples were suspended in distilled water. Equal amounts of total protein (BioRad Protein Assay; Bio-Rad Laboratories Ltd., Missisauga, Ontario) were loaded onto a 14% SDS-polyacrylamide gel (SDS-PAGE) prepared by standard procedures (Sambrook et ah, 1989). After electrophoretic separation of the samples, proteins were transferred to a nitrocellulose membrane with the Semi-Dry Electroblotter A (Ancos Dimension Laboratories Inc., Mississauga, Ontario, Canada; 30-45 min, 150 mA, room temperature) using transblot buffer (25 mM TRIS-HC1, 192 mM glycine, 20% methanol; pH 8.3). The membrane was incubated overnight in a 1:100 dilution of polyclonal rabbit anti-serum raised against purified TSST-1 from S. aureus MN8 (Rosten et al., 1989) or 1 /tg/ml of rabbit anti-serum to SEB (Toxin Technology, Sarasota, FLA) in 5% skim milk (Difco) in a volume sufficient to cover the membrane (30-50 ml). Following three 2-min washes with TRIS-buffered saline (1.21 g/1 TRIS-HC1, 9 g/1 NaCl; pH 7.4) containing 1% Tween-20 (TBS-T) filters were incubated with a 1:1,000 dilution of biotinylated goat-anti-rabbit immunoglobulin G (Gibco/BRL) in TRIS-buffered saline containing 0.5% bovine serum albumin (TBS-BSA) for approximately 2 h. After washing with TBS-T, the filter was incubated with a 1:1,000 dilution of streptavidin-horseradish peroxidase conjugate (Gibco/BRL) in TBS-BSA for 30 min. Following a final wash step with TBS-T, colour development was achieved with 4chloronaphthol as described by the manufacturer (Gibco/BRL). B.  Immunoassay  Production of TSST-1 by S. aureus isolates was quantitated by a non-competitive enzymelinked immunosorbent assay (ELISA) procedure similiar to the method previously described (Rosten et ah, 1987) with one modification. Biotinylated-TSST-1 was used as the capture  44  antibody in place of anti-TSST-1 conjugated to alkaline phosphatase. Bidtinylation of TSST1 was accomplished using the Protein Biotinylation System (Gibco/BRL) following the manufacturer's instructions. After addition of biotinylated anti-TSST-1 [diluted 1:2,000 in PBS containing 0.05% Tween-20 (PBS-T)] to the wells (0.1 ml), the ELISA plate was incubated for 90 min at 37°C. This was followed by three 2-min washes with PBS-T. Streptavidin-alkaline phosphatase (Gibco/BRL; diluted 1:2,000 in PBS-T) was added (0.1 ml/well) and plates were incubated for 20 min at 37°C. Detection was performed as described previously (Rosten et al., 1987) using /7-nitophenyl phosphate as the substrate (Sigma Chemical Co., St. Louis, MO). The sensitivity limit of the ELISA was 0.5 ng/ml of TSST-1. SEA production by S. aureus was determined by a non-competitive ELISA procedure. Affinity purified anti-SEA (Toxin Technology) (2 /tg/ml) in 0.05 M bicarbonate-carbonate buffer (pH 9.6), was coated onto microtiter plates (0.1 ml/well) (ImmulonI, Dynatech Laboratories, Chantilly, VA) overnight at room temperature. Unbound antibodies were removed by three 2-min washes with PBS-T. A SEA standard (Toxin Technology), serially diluted from 128 to 0.5 ng/ml in PBS-T or BHI(when appropriate) or culture filtrates pretreated with normal rabbit serum (10% [vol/vol] final concentration) to eliminate protein A (Rosten et al., 1987), were added in 0.1 ml volumes to the wells. After incubation at 37°C for 1.5 h, wells were washed with PBS-T, and 0.1 ml of alkaline phosphatase conjugated anti-SEA diluted 1/250 in PBST (prepared by the method of Nakamura et al., 1986) was added to each well followed by incubation for 1.5 h at 37°C. Finally wells were washed with PBS-T and incubated with 0.1 ml of a 1 mg/ml solution of /?-nitrophenyl  45  phosphate (Sigma) followed by measurement of the O.D. at 410 nm using a Dynatech MR5000 microplate reader (Dynatech Laboratories). The sensitivity limit of the immunoassay was 1 ng/ml of SEA. SEB production by S. aureus isolates was determined by a non-competitive ELISA procedure as described by Lee et al. (1992). The sensitivity limit of the SEB immunoassay was 0.5 ng/ml of SEB. The total amount of protein found in the culture filtrates of strains was quantitated using the Bio-Rad Protein Assay (Bio-Rad Laboratories). 2.4. 2.4.1.  Molecular Biology Techniques DNA Manipulation  DNA modifying enzymes (restriction enzymes, ligase, Klenow, alkaline phosphatase) were purchased from New England Biolabs (Mississauga, Ontario, Canada), Gibco/BRL Life Technologies Inc. (Gaithersburg, MD) or Boehringer Mannheim (Laval, Quebec, Canada) and used according to the manufacturer's instructions. 2.4.2.  Isolation of Plasmid DNA  Plasmid DNA was isolated from Escherichia coli via lysis by alkali using standard procedures (Sambrook et al., 1989). Isolation of plasmid DNA from Staphylococcus aureus was performed by the alkali lysis procedure modified by the use of lysostaphin as described by Novick (1991). 2.4.3.  Transfer of DNA  Plasmids were transformed into competent CaCl -treated cells of E. coli JM109 (Yanisch2  Perron et al., 1985) by standard procedures as described by Sambrook et al. (1989) or into  46  cells of XL-1 Blue (Bullock et al, 1987) following the protocol of Stratagene (1992). Introduction of plasmid DNA into S. aureus RN4220 (restriction-deficient mutant capable of accepting shuttle plasmids propogated in E. coli) (Kreiswirth et al., 1983) was accomplished by electroporation (Bio-Rad Laboratories Ltd.) following the method of Augustin and Gotz (1990). Introduction of plasmid DNA into S. aureus RN3984 was performed via transduction with phage 80a (Novick, 1963). Phage 80a was maintained by propagation on S. aureus RN450. 2.4.4.  Transduction  Phage 80a transducing lysates were prepared on donor strains using the agar layer method described by Swanstrom and Adams (1951). A swab of an overnight culture of donor cells was suspended into 0.2 ml of phage buffer (ImM MgS0 , 4mM CaCl , 50 mM TRIS-HC1 4  2  (pH 7.8), 5.9 g/1 NaCl, 1.0 g/1 gelatin) (Novick, 1991) and inoculated with ~ 10 PFU of 7  phage 80a. Cultures were gently mixed and the phage was allowed to absorb for 5 min at room temperature. Tryptic Soy Broth Top Agar (TSBTA; TSB containing 0.7% agar) (2.5 ml) was added and the mixture poured onto TSA plates containing 5mM CaCl . After 2  overnight incubation, 2.5-5.0 ml of phage buffer was added to each plate and the agar layer scraped into centrifuge tubes. Tubes were shaken vigorously for several minutes and centrifuged at 10,000 X g for 10 min at 4°C. Supernatants were filter^sterilized. Transduction was performed essentially as described by Kasatiya and Baldwin (1967) with the modification of a 20 min incubation at 37°C before addition of 0.02M sodium citrate as described by Schroeder and Pattee (1984). Transductants were selected on BHI agar containing 10 jtg/ml erythromycin and 5 jtg/ml sodium citrate.  47  2.4.5.  Preparation of chromosomal DNA  DNA was prepared from overnight cultures of S. aureus grown at 37°C. The cells were washed by centrifugation in TEE buffer (1 M TRIS-HC1 (pH 7.6) containing 0.1 M EDTA and 10 mM EGTA), resuspended in TEE buffer (450 /xl/ml of culture) containing lysostaphin (10 units/ml; Sigma) and RNaseA (50 /ig/ml; Boehringer Manheim), and incubated at 37°C for 1 h. Cells were lysed by the addition of 30 /xl of 10% SDS and 0.1 mg/ml proteinase K (Boehringer Mannheim), followed by incubation at 65°C for 1 h. Sodium chloride was added to 500 mM (final concentration) followed by 80 fA per ml of volume of a 10% solution of CTAB (hexadecyl trimethylammonium bromide; Sigma) in 0.7 M NaCl. The sample was incubated at 65°C for 10 min and then extracted with phenol-chloroform (Sambrook et al., 1989). The nucleic acids were precipitated with ethanol as previously described (Sambrook et al, 1989). The DNA was dissolved in TE buffer (10 mM TRIS-HC1 (pH 7.6), containing 1 mM EDTA) and stored at 4°C until use. 2.4.6.  Preparation of DNA probes  pRN6550 carrying tst (Kreiswirth et al, 1987) (provided by R. P. Novick and B. N . Kreiswirth, Public Health Research Institute, New York), pSK155 carrying seb (Ranclli et al, 1985) (provided by S. A . Khan, University of Pittsburg), pMJB38 carrying sea (Betley and Mekalanos, 1988) (provided by R. P. Novick) and pRN6441 (Villafane et al, 1987) (provided by R. P. Novick) were digested with the appropriate restriction endonucleases. The DNA fragments to be used as probes were isolated from a low-melting agarose gel (Sigma) with the use of Gelase (Epicentre Technologies, Madison, WI), following the manufacturer's instructions. The DNA fragments were labelled by the random primed  48  incorporation of digoxigenin-labeled dUTP (Boehringer Mannheim) as outlined by the supplier. 2.4.7.  Southern hybridization analysis  Chromosomal DNA from the bacterial isolates was digested with restriction endonucleases Clal, HindHI or Kpn\ under conditions recommended by the supplier. DNA was electrophoresed through a 1 % agarose gel in TRIS-acetate-EDTA buffer (40 mM TRISacetate, 2 mM EDTA). Digoxigenin-labelled molecular size standards (Boehringer Mannheim) were used as the size controls. DNA fragments were transferred to nylon membranes (Boehringer Mannheim) under alkaline conditions (0.4 M NaOH) as described by Sambrook et al. (1989). The hybridization conditions were: 20-50 ng/ml digoxigenin-labelled probe in a solution of 50% formamide, 5X SSC (IX SSC= 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0), 2% blocking reagent (Boehringer Mannheim), 0.1% N-lauroylsarcosine (wt/vol), and 0.02% SDS (wt/vol) at 42°C overnight. Washing (0.1X SSC, 0.1% SDS [wt/vol]) was done at 68°C. Detection by luminescence was accomplished by the addition of a high-affinity alkaline phosphatase-conjugated anti-digoxigenin antibody, followed by the substrate AMPPD (disodium salt) under conditions described by Boehringer Mannheim. Probes were removed from the nylon blots with two 20-min washes at 37°C in 0.2 M NaOH containing 0.1% SDS (wt/vol), followed by two 5-min washes in 2X SSC at room temperature. Removal of the probe was verified by lack of a hybridization signal after the wash step. 2.4.8.  Polymerase chain reaction (PCR) analysis  Sequences for the oligonucleotide primers designated SEB1 and SEB2 targeting sequences  49  located at positions 310-329 and 634-657 respectively of the SEB structural gene were prepared at the Nucleic Acid and Protein Service Unit (University of British Columbia). These primers were designed by Johnson et al. (1991) and give an amplification product of 478 bp in SEB isolates. PCR was done essentially as described by Johnson et al. (1991), +  except that 60 ng of nucleic acid was used as template for the reaction. Sequences for the oligonucleotide primers designated MLD-1 and MLD-2 were chosen with the use of 'Primer Detective' software. The oligonucleotide sequence for MLD-1 was 5'- CTAAACCAGATGAGTTGCACAAATCG -3', and the sequence for MLD-2 was 5'AATTCCCTAACTTAGTGTCC -3'. These primers target SEB structural gene sequences located at positions 17-39 and 162-181 respectively, with an expected PCR product of 165 bp in SEB isolates. +  The PCR was performed in a 50 /xl reaction mixture using 60 ng of  nucleic acid as template and 0.2 /xM of each primer under the following conditions: denaturation for 30 sec at 95°C, annealing of primers for 30 sec at 55°C, primer extension for 1 min at 72°C. This was repeated for 40 cycles. PCR products were separated on a 2.5% agarose gel in TRIS-acetate-EDTA buffer. 2.5. 2.5.1.  Construction of Isogenic Mutants S. aureus RN3984 - Wildtype isolate used for isogenic mutant construction  RN3984 is a TSST-l-producing strain of S. aureus isolated from the vagina of a patient with menstrual TSS (Barbour, 1981). It has been also referred to as the Harrisburg strain (Schlievert et al., 1979). It is beta-hemolytic, tryptophan-negative, coagulase positive, and belongs to phage group 1 (major types 29/52A/79) (Barbour, 1981; Schlievert et al., 1979;  50  Kreiswirth et al., 1989). It produces staphylococcal enterotoxin A but not staphylococcal enterotoxins B, C, D or E as tested by immunoblot analysis described previously in this section with the addition of 1 jtg/ml of rabbit anti-serum to SEA, SEB, SECi. , SED or SEE 3  respectively (Toxin Technology). 2.5.2.  Inactivation of tst in S. aureus RN3984  The TSST-1 structural gene of RN3984 was inactivated by allele replacement by R. P. Novick and B. N . Kreiswirth as previously described (Sloane et al., 1991). In brief, the tst shuttle plasmid pRN6684 [tst::T<?; temperature sensitive origin of replication (f)] was introduced into RN3984 at the permissive temperature for replication. Dilutions of a broth culture were plated onto TSB containing Tc and incubated at 43°C. After integration of the plasmid into tst by a single crossover event, transduction of the co-integrate into wild-type RN3984 followed by Tc screening of the transductants was performed. Transductants were r  scored for Em* to identify recombinants where plasmid excision had occurred and caused allele replacement of tst by tst::T<f. The Tc fragment was ~2.9 kb. One transductant, r  RN7043, was kept for further characterization. 2.5.3.  Inactivation of sea in S. aureus RN3984 A.  Construction of plasmids  A 624 bp ZJamHI-flwdlll fragment (containing sea) was excised from pMJB38 and ligated to Barrim-Hindlll digested pUC19 (2.7 kb) to create pMLD6872 (3.3 kb) containing a unique C/al site within sea. Inactivation of sea in pMLD6872 was accomplished by inserting a 1.4 kb Taql fragment encoding the erythromycin resistance determinant (Em ) of the S. 1  aureus plasmid pRN6442 (Villafane et al., 1987) into the unique Clal site creating  51  pMLD6873 (4.7 kb). All attempts to introduce pMLD6873 as a suicide plasmid into RN3984 by electroporation were unsuccessful. Next EcoKL digested pMLD6873 was ligated to EcoKL digested pRN6441 (-3.8 kb) (Villafane et al, 1987) to create pMLD6873C-l (-8.5 kb) (a shuttle plasmid that can replicate in both S. aureus and E. coli containing a temperature sensitive origin of S. aureus replication). Refer to Figure 1 for an overview of plasmid construction. B.  Introduction of plasmid pMLD6873C-l into RN3984 and isolation of SEA" transductants  pMLD6873C-l was introduced into S. aureus RN4220 by electroporation and transformants selected for resistance to Em. Next, pMLD6873C-l was transferred from RN4220 to RN3984 (TSST-1 /SEA ) by transduction with phage 80a. After incubation at +  +  30-32°C, (the permissive temperature for plasmid replication) transductants were selected by resistance to Em. Several Em transductants of RN3984 were selected and after 8 h growth r  at 32°C in TSB containing Em, the transductants were grown overnight at 43°C (the nonpermissive temperature for plasmid replication). Overnight cultures were plated (0.1 ml) onto solid media (TSA with Em) and grown overnight at 43°C. The passage of cells from solid to liquid media at 43°C was performed for three consecutive cycles. Em colonies were r  analysed for SEA production by immunoassay as previously described in section 2.3.1. At this stage, the genomic DNA of several SEA" transductants were digested with flwdffl and analysed by Southern hybridization analysis with a 624 bp BamHL-Hindlll sea probe (Betley and Mekalanos, 1988). The results indicated that in these transductants, pMLD6873C-l had integrated into the sea chromosomal locus since the expected HindWL fragment that exceeded  52  Figure 1. Plasmid constructs generated to assist in the inactivation of sea in RN3984 by the method of plasmid integration. Em, erytliromcyin; H, Hindm„ H*, Hindm site not definitely known  53  the Hindm fragment observed with RN3984 by 1.4 kb was not observed (the size of the Errf marker used to inactivate sea in pMLD6873). Instead, three Hindm fragments that hybridized to the probe were found in the SEA" transductants. Since pMLD6873C-l contains two flwdm sites (see Figure 1) this would be consistent with recombination of the plasmid into the sea locus via a single crossover event. To increase the likelihood of obtaining a SEA" mutant derived by allele replacement in which only the inactivated sea and not the plasmid backbone had recombined into the chromosome, one SEA /Em transductant was +  r  chosen and the Em' marker again transduced into wild-type RN3984 with phage 80a. This facilitates plasmid excision and allele replacement at a high frequency (McDevitt et al., 1992). Sixty of sixty-four Em transductants screened after this step were SEA". Several r  transductants analysed by Southern hybridization analysis using the sea probe revealed results that were again consistent with integration of pMLD6873C-l into sea. Attempts to remove the plasmid backbone of one transductant by growing without antibiotic selection at the nonpermissive temperature for a number of generations as previously described (Foster et al., 1990) were unsuccessful. One SEA7Em mutant, designated INT#1, was chosen for further r  characterization. C.  Verification of the stability of the SEA" phenotype of INT#1 after in vivo growth  Thirty BALB/c mice were inoculated intraperitoneally (i.p.) with ~ 10*-10 CFU of 7  washed RN3984 (TSST-1 /SEA ) or INT#1 (TSST-1 /SEA) suspended in sterile phosphate+  +  +  buffered saline (PBS) (15 per group). For the following 5 days, bacteria were collected from three mice in each group by injection of 5 ml of sterile PBS i.p., followed by reaspiration.  54  The collected aspirate was centrifuged at 1,200 X g for 10 min and the pellet was resuspended in 0.5 ml of sterile PBS. Samples of the recovered bacteria were plated onto Mannitol salt agar (Difco) and incubated overnight at 37°C. Cells from several isolated colonies on mannitol plates (chosen by the presence of a yellow halo surrounding the colony) were grown overnight in TSB and the culture supernatant was analysed for SEA by immunoassay. 2.6. 2.6.1.  Creation of a TSST-1 and SEB Producing Strain of S. aureus Construction of plasmids A.  p J W l (tst containing plasmid)  Construction of pJWl was performed by Julian Wood (University of Victoria, Co-op Student). In brief, the cloned TSST-1 structural gene resided in the pBR322-based plasmid pRN6550 (Kreiswirth et al, 1987). A 1.6 kb BgU. fragment containing tst was excised from pRN6550 and ligated to Hindi digested pBluescript (Short et al., 1988). Next, tst was excised on a 1.62 kb HindUl-XhoI fragment and ligated to the Smal digested B. subtilis plasmid pHPS9 (Haima et al., 1990) to create pJWl. pJWl is a shuttle vector that can replicate in both S. aureus and E. coli, and contains antibiotic resistant genes for chloramphenicol and erythromycin. B.  pMLD6876-l (seb containing plasmid)  The cloned SEB structural gene resided in the plasmid pSK155 (Ranelli et al., 1985). seb was excised on a 1.8 kb Kpril-Clal fragment and ligated to HincH digested pUC19 to create pMLD6876. A shuttle vector derivative of pMLD6876 was next constructed by ligating pMLD6876 with pRN6441 (Villafane et al, 1987) to create pMLD6876-l.  pMLD6876-l  55  contains a gene encoding resistance to erythromycin. 2.6.2.  Introduction of p J W l and pMLD6876-l into S. aureus  pMLD6876-l was introduced into S. aureus RN4220 by electroporation. Colonies were selected for resistance to Em. SEB production was verified by immunoblot analysis in the Em transformants. Next, pJWl was introduced into RN4220(pMLD6876-l) by r  electroporation. Colonies were selected for resistance to Em and Cm. Production of TSST1 and SEB by Em7Cm transformants was determined by immunoblot analysis. r  2.7.  In Vitro Effects of TSST-1. SEA and the Culture Filtrates of Isogenic Strains on Human Peripheral Blood Mononuclear Cells and Rabbit Splenocytes  2.7.1. Purification of TSST-1 and SEA TSST-1 was purified from culture supernatants of S. aureus MN8 by Winnie Kum (University of British Columbia) by preparative isoelectric focusing and chromatofocusing as previously described (Kum et al., 1993). This procedure results in a pure TSST-1 preparation, as verified by silver staining after SDS-PAGE, by immunoblotting with polyclonal rabbit antiserum raised against the crude culture supernatant of S. aureus MN8, and by autoradiography after iodination and SDS-PAGE (Kum et al., 1993). SEA was purchased from Toxin Technology and further purified by Winnie Kum by chromatofocusing using a pH 6 to 8 gradient polybuffer exchanger (PBE 94; Pharmacia Fine Chemicals). SEA (2 mg) was reconstituted in 5 ml of TRIS-acetate buffer (25 mM, pH 8.3) and applied to a column (K 15.30, 1.5 by 30 cm; Pharmacia) containing 80 ml of PBE 94 equilibrated with TRIS-acetate buffer. Elution of SEA was accomplished with polybuffer 96acetate (pH 6.0; Pharmacia) at a flow rate of 36 ml/h. Fractions were assayed for SEA by  56  immunoassay, and those containing SEA were pooled, dialysed against several changes of deionized water for 72 h at 4°C, and lyophilized. Purity of the SEA preparation was verified by the observation of a single band at ~ 28 kD after separation on SDS-PAGE followed by silver staining. 2.7.2. Purification of Human Peripheral Blood Mononuclear Cells Fresh human peripheral blood mononuclear cells (PBMC) were obtained from healthy volunteers by the method previously described (See et al., 1992c). In brief, cells were obtained by centrifugation of platelet-pheresis buffy coats over Histopaque 1.077 (Sigma). Mononuclear cells at the interface were washed with Hank's balanced salt solution before being separated into T- and non-T cell populations by rosetting with sheep erythrocytes as described previously (Madsen and Johnson, 1979). Non-rosetted cells were separated into monocyte and B lymphocyte fractions by density gradient centrifugation over Percoll (Pharmacia Fine Chemicals, Dormal Quebec, Canada). RPMI 1640 and 10% heat inactivated (56°C, 30 min) fetal calf serum (FCS) (Gibco/BRL) (~ 16 ml) was added to the non-T cell fraction and mixed with Percoll to give a final specific gravity of 1.062 g/ml. One ml of RPMI plus 10% FCS was layered on the top of each suspension and centrifuged at 800 X g for 15 min. The monocyte-containing interface was collected and cells were washed three times with Hank's. This procedure results in a >90% pure monocyte preparation as assessed by nonspecific esterase staining of cytospin preparations (See et al., 1992c). For isolation of purified human T lymphocytes, E-rosetted cells were treated with ammonium chloride to remove sheep erythrocytes, and washed three times with Hank's. At this stage, T cells are >80% C D 2 by flow cytometric analysis as determined previously in our +  57  laboratory by Raymond See. 2.7.3. Preparation of Rabbit Splenocytes Rabbit spleens from female New Zealand White rabbits were removed asceptically, and single cell suspensions were obtained by pressing the spleen through a stainless steel grid into RPMI 1640 medium. Red blood cells were removed from the cell suspension by lysis with 0.85%  NH4CI2  similiar to the method previously described (Mishell et. al., 1980). Cells  were washed in RPMI 1640 and resuspended to the appropriate concentration in RPMI 1640 supplemented with 10% heat inactivated FCS, 2mM L-glutamine, 25 mM Hepes buffer (pH 7.2) and 10 jtg/ml polymyxin B sulfate. 2.7.4. Preparation of Culture Filtrates from Isogenic Strains Culture filtrates from isogenic strains were prepared by growing as indicated for the in vivo model inocula preparation (BHI, post-exponential phase) (see Section 2.8.2) followed by centrifugation at 2,500 X g for 10 min. Culture filtrates were filter-sterilized and stored at -20°C. 2.7.5. Immunoassay Detection of Cytokine Production by Human P B M C or Rabbit Splenocytes Stimulated with Toxin or Culture Filtrate T cells and monocytes (1:1 ratio; 2 X 10 cells/well) in supplemented RPMI 1640 6  (described in section 2.7.3) were cultured in 1.0 ml volumes in conjunction with various concentrations of TSST-1, SEA or TSST-1 plus SEA in 24-well culture plates (BectonDickinson). After 19.5 h at 37°C, 5% C 0 , supernatants were harvested and contaminating 2  cells removed by centrifugation at 800 X g for 5 min. Supernatants were stored at -70°C.  58  A.  Human TNFa  TNFa present in the culture supernatants was assayed by ELISA as previously described (See et al, 1992c). The sensitivity limit of the TNFa ELISA was 250 pg/ml. B.  Human IL-1B  EL-IB in the culture supernatants was assayed by ELISA as previously described (See and Chow, 1992). The sensitivity limit of the IL-1B ELISA was 500 pg/ml. C.  Human IL-6  Levels of IL-6 in culture supernatants was determined by ELISA. Goat anti-human EL-6 (R + D Systems, Minneapolis, M N ; 0.08 ml at 2 itg/ml in 0.05 M bicarbonate-carbonate buffer, pH 9.6) was used to coat flat-bottomed 96-well microliter plates (ImmulonI, Dynatech Laboratories, Inc.) overnight at room temperature. Unbound antibodies were removed by three 2-min washes with PBS containing 0.1% Tween 20 (PBS-TW). A human IL-6 standard (R + D Systems) serially diluted from 2500 to 0.2 pg/ml in PBS containing 3% BSA (PBS-BSA), or test sample, was added in duplicate 0.08 ml volumes to the wells and plates were incubated for 90 min at 37°C. After washing with PBS-TW, 0.08 ml of biotinylated goat anti-human IL-6 (R + D Systems) diluted 1/8,000 in PBS-BSA was added to wells and incubated for 1.5 h. Plates were then washed and wells incubated with 0.08 ml of streptavidin-alkaline phosphatase (Gibco/BRL) diluted 1/8,000 in PBS-BSA for 20 min at 37°C. Wells were washed five times with 0.2 ml of 50 mM TRIS-buffered saline (pH 7.5) and an ELISA amplification system (Gibco/BRL) was used as previously described (See et al., 1992c) to increase the sensitivity of the assay. Optical density at 490 nm was measured in a Titertek Multiscan spectrophotometer (Flow Laboratories, Mississauga, Ontario,  59  Canada). The sensitivity limit of the IL-6 assay was 625 pg/ml. In each case the background values (PBMC in media alone) were subtracted from the observed values. D.  Rabbit TNFa  Rabbit splenocytes (2 X 10 cells/0.5 ml) were incubated with 0.3 ml supplemented RPMI 6  1640 plus 0.2 ml of 10-fold serially diluted culture filtrates of RN3984, RN7043 or INT#1 in 24-well tissue culture plates (Becton Dickinson). Addition of BHI (0.2 ml undiluted + 0.3 ml supplemented RPMI 1640) served as the media control. After incubation at 37°C, 5% C 0 for 24 h, the media was collected and contaminating cells present were removed by 2  centrifugation at 800 X g for 5 min. Supernatants were stored at -70°C. TNFa induced from rabbit splenocytes was quantitated by a rabbit TNFa ELISA. In brief, goat-anti-rabbit TNFa (Pharminogen, San Diego, CA) (4.0 /tg/ml) in 0.05 M bicarbonate-carbonate buffer (pH 9.6), was used to coat microtiter plates (Dynatech Laboratories) overnight at room temperature (0.08 ml/well). Unbound antibodies were removed by three 2-min washes with PBS-TW. Rabbit TNFa conditioned medium (Pharminogen), serially diluted from 5 ng/ml to 0.15 ng/ml in PBS-BSA were added in duplicate 0.08 ml volumes to the wells. Test samples (harvested supernatants of 24 h stimulated rabbit splenocyte preparations) were added in triplicate 0.08 ml volumes. BHI incubated with rabbit splenocytes alone served as the background control. After incubation at 37°C for 1.5 h, wells were washed with PBS-TW and 0.08 ml of biotinylated goat-anti-rabbit TNFa (Pharminogen) (2 /*g/ml) in PBS-BSA was added. Following incubation at 37°C for 1.5 h and washing, plates were incubated with streptavidin-alkaline phosphatase (Gibco/BRL) diluted 1/1,000 in PBS-BSA for 20 min at 37°C. Finally, plates were washed and incubated for 1.5 h at 37°C with p-nitrophenyl  60  phosphate (Sigma) as the substrate. The O.D. at 410 nm was measured using a Dynatech MR5000 microplate reader (Dynatech Laboratories). The sensitiviy limit of the TNFa assay was 150 pg/ml. 2.7.6.  Proliferative Response of Human P B M C or Rabbit Splenocytes Upon Stimulation with Toxin or Culture Filtrate  Mitogenesis of human PBMC was assessed by quantitation of pH]mymidine incorporation into cellular DNA after exposure to TSST-1, SEA or TSST-1 plus SEA. A 1:1 ratio of T cells and monocytes (3 X 10 cells/well) suspended in supplemented RPMI 1640 were s  cultured in 0.2 ml volumes with various concentrations of TSST-1, SEA or TSST-1 plus SEA in 96-well U-bottom tissue culture plates (Falcon Lab ware, Becton-Dickinson) for 3 days at 37°C, 5% C 0 . Cells were labelled with 1 /xCi of pHJthymidine (ICN Flow 2  Laboratories, Irvine CA; 6.7 Ci/mmol) 18 hr before completion of incubation and harvested onto glass fiber filter paper with an automatic harvester (Skatron, Sterling, VA). Radioactivity was detected in a liquid scintillation counter (Beckman LSI800). The proliferative response of rabbit splenocytes to the culture filtrates of isogenic strains was measured as described above. Rabbit splenocytes (3 X 10 cells/0.1 ml) were cultured s  in the presence of 0.05 ml of supplemented RPMI 1640 plus 0.05 ml of 10-fold serially diluted culture filtrates in supplemented RPMI 1640. BHI alone served as the media control. For experiments involving TSST-1 or SEA complementation to RN7043 or INT#1 culture filtrates respectively, 0.05 ml of purified TSST-1 or SEA diluted in supplemented RPMI 1640 was added to the appropriate well to equal the known concentration of TSST-1 or SEA present in the RN3984 culture filtrate sample as quantitated by immunoassay.  61  The proliferative response of rabbit splenocytes to purified TSST-1 and SEA was determined as described above by incubating rabbit splenocytes (3 X 10 cells/0.1 ml) with 5  0.05 ml supplemented RPMI 1640 and various concentrations of TSST-1, SEA, or TSST-1 plus SEA in a total volume of 0.2 ml. For the generation of TSST-1 and SEA induced mitogenic dose response curves, a pooled rabbit splenocyte preparation from 5 rabbits was used in an attempt to minimize the variation in proliferative responses that have been observed among rabbits. 2.8. 2.8.1.  Animal Model Studies Animals  Female New Zealand White (NZW) rabbits were purchased from R and R Rabbitry (Washington, USA). They were housed separately and fed water and rabbit chow ad libitum. Eight week old female BALB/c mice were purchased from Charles River Canada (Quebec). They were housed 5 per cage. 2.8.2.  Preparation of Isogenic Mutant Inocula for In Vivo Studies  RN3984 (TSST-1 /SEA ), RN7043 (TSST-17SEA ) and INT#1 (TSST-1 /SEA) were +  +  +  +  streaked onto BHI and two isolated colonies were inoculated into 20 ml volumes of dialysed BHI broth (Rosten et al., 1989). BHI media and all glassware used for bacterial culture preparation were maintained in pyrogen-free conditions to prevent endotoxin contamination (Rosten et al., 1989). Cultures were incubated with shaking at 150 rpm, 37°C overnight in a controlled environment incubator-shaker (New Brunswick Scientific, Edison, NJ). The postexponential phase S. aureus strains were centrifuged at 550 X g for 10 min at 4°C. After two washes in sterile PBS, cultures were resuspended in BHI. The number of organisms in  62  the staphylococcal suspensions were determined on BHI agar plates. 2.8.3.  Introduction of isogenic mutants into two animal models of lethal shock A.  Tampon-associated vaginal infection model of TSS in the New Zealand White rabbit  NZW rabbits weighing 3.3-4.5 kg were anesthetized with ketamine (30 mg/kg), acepromazine (0.8 mg/kg) and xylaxine (3 mg/kg) and maintained on 4% oxygen and 1.53.5% halothane during the surgical procedure. A suprapubic abdominal incision was made to expose the length of the vagina. A preweighed, sterile tampon (rayon/cotton, Tampax Super Plus, Tambrands Canada Inc., 0.5-0.55 g) placed in an uncuffed endotracheal tube applicator (Tip-to-Tip, 7 mm internal diameter; Mallinckroft, Argyle, NY) was inserted through the urogenital sinus into the vagina and dislodged against the cervix using a sterile wooden dowelling. A loose ligature was placed around the distal end of the vagina with a silk 2-0 sterile nonabsorbable suture to prevent spontaneous extrusion of the tampon. A 2 ml inoculum containing ~ 10 CFU S. aureus in BHI, mixed with 0.1 % of rabbit blood (as 10  simulated menses, obtained from the same rabbit) was injected into the tampon through the vaginal wall with a 25 guage needle (Becton Dickinson, Cockeysville, MD). The abdominal incision was closed with synthetic suture and after administration of 0.02-0.05 mg/kg of buprenephrine, rabbits were allowed to recover for observation. Rabbits were monitored for a period of up to 8 days with respect to mortality, and incidence of lipaemia and diarrhea. Weight and rectal temperature measurements were taken daily. These clinical parameters are indirect indicators of in vivo cytokine secretion, since TNF and EL-1 have been shown to suppress lipoprotein lipase production (Beutler et ah, 1985; Price et al., 1986), induce fever  \  63  (Dinarello et al., 1986), and cause weight loss (Moldawer et al., 1988). B.  D-Galactosamine sensitized mouse model of lethal shock  BALB/c mice were injected i.p. with 20 mg of D(+)-galactosamine-hydrochloride (Sigma) dissolved in sterile saline 2 h prior to bacterial challenge. The mice (30 in total, 15 per group) were inoculated i.p. with ~ 10 CFU of RN3984 or INT#1. In a preUminary 8  experiment with RN3984, this dose corresponded to the i.p. L D  8 0  after 48 h. Infection with  BHI media alone in 2 mice served as the control. Deaths were recorded up to 72 h after inoculation.  64  Chapter 3.  Determination of whether TSST-1 plays a pivotal role in the pathogenesis of MTSS by the introduction of isogenic mutants in TSST-1 production in a Tampon-associated Vaginal Infection Model  3.1.  Introduction  Although S. aureus produces many different extracellular products, the observation that TSST-1 was produced by the majority of TSS isolates caused it to be implicated in TSS early in the investigation of this disease. Nothwithstanding the numerous studies that have been performed that strongly point to TSST-1 as a causative agent in this syndrome, definitive evidence proving a role for TSST-1 in menstrual TSS was lacking. This may have resulted from the lack of an appropriate animal model and/or unavailability of appropriate test strains in earlier studies. These models and constructs are currently available however. A vaginal tampon model developed in the NZW rabbit has been described by Melish et al. (1989). In addition, the construction of various isogenic mutants in TSST-1 production have been reported (Rasheed et al., 1985; de Azavedo et ah, 1985; Sloane et ah, 1991). Before this thesis work began, no one had made use of an appropriate animal model that resembled menstrual TSS in conjunction with isogenic S. aureus mutants in TSST-1 production to verify TSST-1 as an etiologic agent. In light of this, the first objective of this research was to introduce isogenic mutants in TSST-1 production into a vaginal infection model of TSS to 65  determine if the TSST-1 strain caused more mortality or morbidity than the TSST-1" +  construct. In addition, the effects of culture filtrates from isogenic strains on rabbit splenocytes with respect to T cell proliferation and TNFa secretion - two biologic activities which are proposed to play an important role in the pathogenesis of superantigen-mediated shock, were examined. 3.2. 3.2.1.  Results Characterization of the TSST-1" Mutant RN7043  Loss of TSST-1 production by RN7043 was verified by immunoblot analysis (Figure 2) and immunoassay (Table 6) as described in Materials and Methods. As can be seen in Figure 2, a band with a molecular mass of 22 kD (molecular mass of TSST-1) having immunological reactivity to the anti-TSST-1 antibody is missing in the culture filtrate of RN7043 (Lane 1) compared to the culture filtrate of RN3984 (Lane 2). This indicates that no TSST-1 is stably secreted into the medium by RN7043. The size of the Clal fragment containing tst was examined by Southern hybridization analysis of Clal digested genomic DNA of RN7043 with a 297 bp Hincll-BamHL tst specific gene probe (Kreiswirth et al., 1989). As shown in Figure 3, the Clal fragment that hybridized to the tst gene probe in RN7043 was larger than the Clal fragment that hybridized to the probe in the TSST-1 wildtype strain RN3984. The larger Clal fragment observed to +  hybridize with RN7043 DNA as compared to RN3984 DNA was consistent with the expected increase in size (approximately 2.9 kb) if the Tc fragment inserted into tst of RN7043. This r  observation provided support for the supposition that tst in RN7043 had been inactivated by allele replacement.  66  Figure 2. Immunoblot analysis of culture filtrates from TSST-1' mutants separated by SDS-PAGE and blotted with a polyclonal anti-TSST-1 antibody. Lane# 1. RN7043 culture filtrate 2. RN3984 culture filtrate 3. 8 ng of purified TSST-1 4. Bio-Rad Kaleidoscope Pre-Stained Molecular Mass Standards (kD) Arrow indicates the position of the TSST-1 standard.  67  Table 6. Detection of TSST-1 in the Culture Filtrates of RN3984 and RN7043.  Strain  [TSST-1] in Culture Filtrate (ng/ml)  RN3984  107.2 ± 4 . 2  RN7043  nd  TSST-1 levels in culture filtrates were quantitated by immunoassay as described in Materials and Methods. Two single colonies of each strain were inoculated into 20 ml BHI and incubated at 37°C for 10 h. Culture filtrates were obtained by centrifugation and filter-sterilization. Values represent the mean ± S E M of 5 culture filtrates prepared from each strain tested in at least duplicate wells by immunoassay. The sensitivity limit of the immunoasaay was 0.5 ng/ml of TSST-1. nd, not detected  68  1  2  3  Figure 3. Southern hybridization analysis of TSST-1' mutant genomic DNA digested with Clal and hybridized with a 297 bp HincR-BamHLtor-specificgene probe (Kreiswirth et al., 1989) as described in Materials and Methods. Lane# 1. Digoxigenin-labelled DNA Molecular Mass Standards (kb) 2. RN3984 3. RN7043  69  SEA is the only staphylococcal enterotoxin produced by RN3984. Since SEA is in the superantigen family of proteins and could be a causal agent in MTSS, a comparison of the amount of SEA produced by the TSST-1" mutants was performed. For some unexplained reason, in relation to the total amount of protein (pg of SEA/itg of protein), RN7043 was found to produce significantly less SEA than RN3984 (9.7 ± 2.4 vs 2.6 ± 0.3; P = 0.03, Mann-Whitney test, 2-tailed, see Figure 4). This could be important if SEA plays a contributory role in MTSS, since any observed differences between the pathogenicity of the TSST-1" mutants in vivo could be attributed to the higher amount of SEA produced by RN3984. 3.2.2.  Virulence of RN3984 and RN7043 in a tampon-associated vaginal infection model of toxic shock syndrome A.  Mortality  RN3984 was more lethal in the tampon-associated vaginal infection model than RN7043. 4 of 13 (31%) rabbits infected with RN3984 died as compared to 0 of 13 (0%) rabbits infected with RN7043 (P < 0.05, Fisher's test, one-tailed). Three of the four deaths occurred within the first three days of experimental infection. Our definition of mortality included symptoms of 'severe clinical distress' necessitating immediate sacrifice as mandated by the University of British Columbia Animal Care Committee. Conditions which warranted immediate sacrifice for humane reasons in this model was defined as the presence of any two of the following three objective criteria: i) weight loss greater than 20% of that prior to challenge ii) a recumbency position of the rabbit  70  Figure 4.  Comparison of SEA levels with respect to total protein in the culture filtrates of the TSST-1'mutants. Cultures were prepared by inoculating two single colonies into 20 ml of BHI followed by incubation at 37°C for 10 h. Bars represent the mean SEA level (detected by immunoasay) ± SEM per ng of protein in 5 prepared culture filtrate samples. The difference between the values was significant (P = 0.03, Mann-Whitney test; 2-tailed). ;  71  iii) no response (ie. failure to get up) to various stimuli, including sound, pain or touch B.  Temperature change  Rabbits infected with RN3984 exhibited a significantly higher mean average rectal temperature over a 6 day observation period than rabbits infected with the TSST-1" mutant RN7043. The mean highest temperature (°C) ± SEM of the survivors infected with RN3984 was 40.2 ± 0.1 vs 39.7 ± 0.1°C for rabbits infected with RN7043 (P = 0.0003, student's ttest, one-tailed). C.  Weight loss  Although it appeared that rabbits infected with RN3984 lost more weight as compared to pre-infection baseline values (Day 0) than rabbits infected with RN7043 (Figure 5), differences in weight loss between the infected groups over the observation period was not statistically significant. Both groups of rabbits lost weight up until Day 3 following infection, however rabbits infected with RN3984 continually lost weight thereafter, whereas rabbits infected with RN7043 exhibited a gradual increase in weight over the same time period. 3.2.3. Effect of culture filtrates of isogenic strains on rabbit splenocytes A. T cell proliferation A mitogenic effect as measured by [ H] thymidine incorporation into cellular DNA 3  (described in Materials and Methods) was observed when diluted culture filtrate preparations were added to rabbit splenocytes (Figure 6). RN3984 culture filtrate induced a significantly greater mean effect ( ± SEM) in T cell proliferation than the RN7043 culture filtrate at dilutions of 1/1,000 (842,527 ± 178,477 vs 510,372 ± 120,297 cpm; P = 0.01, paired ttest, one-tailed) and 1/10,000 (385,567 ± 96,984 vs 98,529 ± 16,303 cpm; P = 0.01,  72  0 • RN3984 o RN7043 in in <L> o c  %in  -5  -C <a s  ^  i Li  -10  -15 0  1  2  -3  Days from  4  .5  6  7  Infection  Figure 5. Weight loss in rabbits infected with RN3984 and RN7043. Values represent the mean ± SEM in all rabbits (including those that subsequently succumbed to the infection) in which weight values were obtained.  73  1200000H  • RN3984 o RN7043 T Media  1000000  x  800000 o  600000 400000 200000 0 1/10  1/10000  Dilution  Figure 6. Mitogenic response induced by RN3984 and RN7043 culture filtrates on rabbit splenocytes. Values are expressed as the mean ± SEM measured by [ H]thymidine incorporation into cellular DNA in at least 7 rabbit splenocyte samples, tested in a rninimum of triplicate determinations. BHI served as media control. * indicates a significant difference in proliferation compared to RN3984 (P = 0.01, paired t-test, one-tailed) 3  J  74  paired t-test, one-tailed). The decreased level of T cell proliferation caused by RN7043 culture filtrate in comparison to RN3984 culture filtrate could be explained by the lack of TSST-1. If this were true, adding TSST-1 to the RN7043 culture filtrate would be expected to increase the level of T cell proliferation back to the level observed with the RN3984 culture filtrate. To test this, purified TSST-1 was added with a 1/10,000 dilution of RN7043 culture filtrate (at a concentration that equalled the known amount found in the diluted RN3984 culture filtrate as determined by immunoassay) to rabbit splenocytes and the amount of T cell proliferation was measured by [ H]thymidine incorporation into cellular DNA. As 3  observed in Figure 7, purified TSST-1 added to the culture filtrate of RN7043 restored the amount of cell proliferation back to that observed with RN3984. B.  TNFa Secretion  Culture filtrates of RN3984 and RN7043 stimulated rabbit splenocytes to secrete TNFa as measured by a rabbit TNFa specific immunoassay described in Materials and Methods (Figure 8). Incubation of rabbit splenocytes with undiluted RN3984 culturefiltrateinduced significantly more TNFa (mean + SEM) than incubation with undiluted RN7043 culture filtrate (25.74 ± 1.19 vs 21.71 ± 1.47 ng/ml; P = 0.03, paired t-test, one-tailed). Similiarly, a significant increase in TNFa secretion from rabbit splenocytes was observed upon incubation with a 1/10 dilution of RN3984 culture filtrate as compared to a 1/10 dilution of RN7043 culture filtrate (25.57 ± 1.7 vs 22.54 ± 1.76 ng/ml; P = 0.04, paired ttest, one-tailed). 3.3.  Discussion  These studies support the hypothesis that TSST-1 is an important virulence determinant in  75  700000 600000 500000 O  BHI Media  RN3984  RN7043  RN7043 + TSST-1  Figure 7.  Increase in the proliferative response of rabbit splenocytes to RN3984 culture filtrate is consistent with the presence of TSST-1. A 1/10,000 dilution of RN7043 culturefiltratewas added with purified TSST-1 (at a concentration that equalled the known amount found in the diluted RN3984 culture filtrate as determined by immunoassay) to rabbit splenocytes and the amount of T cell proliferation was assessed by incorporation of [ H]thymidine into cellular DNA. Results are the mean ± SEM in S rabbit splenocyte samples each tested in a minimum of triplicate cultures. * indicates a significant difference in proliferation compared to RN3984 (P < 0.05, paired t-test, onetailed); U indicates no significant difference compared to RN3984 (P > 0.05, paired t-test, .one-tailed) 3  v  76  Figure 8. TNFa secretion induced by RN3984 and RN7043 culture filtratesfromrabbit splenocytes. Values represent the mean ± SEM tested in a minimum of duplicate cultures. BHI served as the media control and gave background values of <0.04 ng/ml. Numbers in brackets indicate the number of rabbits tested at each dilution. * indicates a significant difference in proliferation compared to RN3984 (P < 0.05, paired t-test, one-tailed)  77  the pathogenesis of menstrual toxic shock syndrome. The TSST-1 S. aureus strain was able +  to induce more lethality and fever than its TSST-1" isogenic derivative in the tamponassociated vaginal infection model. This is the first study to test a role for TSST-1 in menstrual TSS using isogenic mutants in a relevant animal model. In addition, using mitogenesis and TNFa secretion as indicators of toxicity potential, the in vitro observation that the RN3984 culture filtrate induced an increased amount of T cell proliferation and TNFa secretion from rabbit splenocytes than RN7043, further supports the in vivo data. RN3984 was initially isolated from a patient with menstrual TSS (MTSS) and phenotypically resembles the majority of MTSS isolates since it produces both TSST-1 and SEA (Kain et al., 1993), and requires tryptophan for growth [Chu et al., 1988]. No other enterotoxins besides SEA are produced by RN3984. Similiar to TSST-1, SEA also belongs to the superantigen family of proteins, and as such, can cause profound immunological disturbances. It is puzzling why SEA is produced in lower amounts by RN7043 as compared to RN3984 when grown in vitro. Since growth in a test-tube does not mimic growth in vivo, partly because different environmental signals may exist to regulate the expression of various virulence determinants, it is not known whether RN7043 produces a lower amount of SEA than RN3984 in vivo. It is possible that TSST-1 influences the expression of SEA in some manner, such that inactivation of tst in RN7043 resulted in the reduction of SEA production. Alternatively it is possible that during construction of RN7043, insertion of the Tc cassette r  into tst resulted in polar effects to occur in downstream genes which may have been involved in regulating the expression of several staphylococcal proteins. In terms of superantigen expression, these strains differ in both TSST-1 and SEA  78  production. Whether SEA is an important mediator in menstrual TSS is not known at present. These results suggest that it may not be (at the concentration produced by RN7043) since there was not any mortality observed in rabbits infected with RN7043 (TSST-17SEA ). +  However since the amount of SEA produced by RN7043 was lower than for RN3984, this result does not provide adequate assessment of its potential role. Due to the lower amount of SEA produced by the TSST-1" mutant, determination of whether TSST-1 plays a causative role in MTSS remains inconclusive. To determine if SEA production enhances the virulence of a TSST-1 strain, the pathogenicity of a TSST-1 /SEA and a TSST-1 /SEA" strain were +  +  +  +  next compared upon introduction into two animal models of lethal shock.  79  Chapter  4. Effect of TSST-1 and SEA Superantigen Co-Production in the  Pathogenesis of Menstrual TSS  4.1.  Introduction  In the previous chapter evidence that TSST-1 is an important virulence determinant in the pathogenesis of menstrual toxic shock syndrome remained inconclusive since the TSST-1" mutants differed in the amount of SEA produced. It is believed that production of enterotoxins in conjunction with TSST-1 may exacerbate the illness. Since enterotoxins belong in the superantigen family of proteins, an enterotoxin /TSST-l strain may cause +  +  greater immunological disturbances in an infected host than a strain producing only TSST-1. Co-production of SEA and TSST-1 appears to be clinically important since this toxin combination is produced by menstrual TSS isolates of S. aureus more frequently than nonmenstrual, non-TSS or carrier isolates (Chang et al., 1991). Analyses by multilocus enzyme electrophoresis (MLEE) (Chang et al., 1991) and phage typing (Marples and Wieneke, 1993) suggest that a single clone of S. aureus producing both TSST-1 and SEA is responsible for the majority of menstrual TSS cases. In light of these observations, it was hypothesized that co-production of SEA and TSST-1 by S. aureus may act to increase various superantigen-associated activities which may lead to a more virulent strain than production of TSST-1 alone.  To test this hypothesis, TSST-1, SEA, and TSST-1 plus SEA  at minimally effective concentrations were incubated with human PBMC and the effects on T cell proliferation and cytokine secretion (TNFa, IL-lfi, IL-6) were examined. Since both 80  TSST-1 and SEA are known to bind to the MHC Class U molecule on human PBMC to induce these activities (Grossman et al., 1990; Fleischer and Schrezenmeier, 1988; Fischer et al., 1989; See et al., 1992a) an enhancement in these effects was fully expected upon coincubation at low concentrations. To examine the effects of this toxin combination in vivo, the SEA structural gene (sea) of a TSST-1 /SEA strain of S. aureus was inactivated to +  +  create a TSST-1 /SEA" mutant. The TSST-1 /SEA and TSST-1 /SEA" strains were +  +  +  +  introduced into a rabbit tampon-associated vaginal infection model of TSS and a Dgalactosamine mouse model of lethal shock to determine if the TSST-1 /SEA strain of S. +  +  aureus causes more mortality and morbidity than the TSST-1 /SEA" strain. In addition, the +  in vitro effects (T cell proliferation, TNFa secretion) of culture filtrates from the isogenic pair and purifed TSST-1 and SEA were examined on rabbit splenocytes. 4.2. 4.2.1.  Results In vitro effect of purified TSST-1 and SEA on human peripheral blood mononuclear cells A.  T cell proliferation  TSST-1 and SEA can induce the proliferation of T cells from human PBMC as observed in Figures 9 and 10 (data by Winnie Kum). The dose response curves observed with these cells are similiar to that observed by other investigators in that TSST-1 and SEA were found to mediate T cell proliferation over a broad dose range (Langford et al., 1978; Calvano et al., 1984; Poindexter and Schlievert, 1985; Kum et al., 1993). Direct comparison between T cell populations is difficult due to variation in the amplitude of the proliferative response (Calvano et al., 1984).  81  c o 4-J ra o Q. u O u c  70000 60000 50000 Q_ O  c  40000 30000  TJ  'e >, JZ -M  20000  1—1 ~T~ n i  10000  i  0 -6  -5  -4  -3  - 2 - 1 0  1  Log [TSST-1] ug/ml  Figure 9. Mitogenic response of human PBMC to TSST-1 as measured by [ H]thymidine incorporation into cellular DNA. Each point represents the mean ± SEM from one human PBMC culture tested in triplicate determinations. Background activity has been subtracted. Concentrations of TSST-1 outlined by the boxes indicate the doses used for the co-incubation studies. (Experimental data provided by Winnie Kum). ' , 3  82  120000h  Log  [SEA]  ug/ml  Figure 10. Mitogenic response of human PBMC to SEA as measured by [ H]thymidine incorporation into cellular DNA. Each point represents the mean ± SEM from one human PBMC culture tested in triplicate determinations. Background activity has been subtracted. The concentration of SEA outlined by the box indicates the dose used for the co-incubation studies. (Experimental data provided by Winnie Kum). 3  83  To test if co-incubation of TSST-1 and SEA on human PBMC can result in an enhancement in T cell proliferation as compared to incubation with either toxin alone as expected, concentrations of TSST-1 and SEA that caused a minimal amount of T cell proliferation according to the dose response curves were chosen (Figures 9 and 10, indicated by boxes). Three concentrations of TSST-1 (1, 10 and 100 pg/ml) either alone, or in conjunction with SEA (0.1 pg/ml) were added to human PBMC. Three doses of TSST-1 were used in these studies in an attempt to overcome the anticipated variability of responses between human T cell populations. As observed in Figure 11, a significant increase in T cell proliferation (cpm ± SEM) occurred upon incubation of human PBMC with both TSST-1 (1 pg/ml) and SEA (0.1 pg/ml) (21,224 ± 6613) as compared to either TSST-1 (12,111 ± 3909; P = 0.013, paired t-test, 2-tailed) or SEA (8,046 ± 2310; P = 0.035, paired t-test, 2tailed) alone. In addition, 3 cultures of human PBMC tested with increasing concentrations of TSST-1 (10 and 100 pg/ml) in conjunction with SEA (0.1 pg/ml) also exhibited a significant increase in T cell proliferation as compared to incubation with either toxin alone i (2 of 3 human PBMC cultures using TSST-1 at 10 pg/ml, and 3 of 3 human PBMC cultures using TSST-1 at 100 pg/ml). The increased effect in mitogenesis appeared to be additive rather than synergistic since the mean proliferative response observed upon co-incubation with TSST-1 and SEA did not significantly exceed the sum of the proliferative response observed upon incubation with TSST-1 or SEA alone. B.  Cytokine secretion (TNFa, IL-lfl, EL-6)  The effect of co-incubation of TSST-1 and SEA on human PBMC with respect to cytokine secretion was next investigated. A preliminary experiment with the human PBMC of one  84  \  Figure 11. Human PBMC mitogenic response induced upon coincubation with TSST-1 and SEA as measured by [ H]thymidine incorporation into cellular DNA. The values represent the mean ± SEM in a minimum of quadruplicate determinations performed in 9 human PBMC samples. Background activity has been subtracted. * significantly different from the mitogenic response observed with TSST-1 or SEA alone ( P < 0.05, paired t-test, 2-tailed) 3  85  donor was performed to generate a dose response curve of TSST-1 and SEA induced TNFa, JL-1B, and IL-6 production. Concentrations of TSST-1 and SEA ranging from 0.001 pg/ml to 1 ng/ml were used to stimulate human PBMC for 19.5 h. Although dose response effects were observed for toxin induced TNFa and IL-6 production, this particular donor was a low IL-1B responder to these toxins at the concentrations used. Stimulation with either TSST-1 or SEA at 1 /xg/ml (the highest dose used) resulted in the release of IL-1B into the culture supernatant, but at the low end of the sensitivity limit for the IL-1B immunoassay. Due to the inherent variabilities among human PBMC samples with respect to cytokine induction (de Azavedo et ah, 1988; Endres et al., 1988; Fast et al., 1988; Ikejima et al., 1988; Schindler et al., 1990; Gjorloff et ah, 1991; Muller-Alouf et ah, 1994; Krakauer 1995), the choice of an appropriate concentration of either TSST-1 or SEA that would be suitable for all donors is extremely difficult. Furthermore different concentrations may be needed to stimulate each cytokine. As a result, based on this preliminary experiment, an arbitrary concentration of 10 pg/ml of TSST-1 and either 1 or 10 pg/ml of SEA was chosen. To determine if co-incubation of TSST-1 and SEA with human PBMC enhanced TNFa production as compared to incubation with TSST-1 or SEA alone, five human PBMC samples were stimulated with TSST-1 (10 pg/ml) and SEA (1 pg/ml) either alone, or in combination. As shown in Figure 12, a significant increase in the mean TNFa level (ng/ml ± SEM) present in the culture supernatants harvested from the human PBMC was observed upon addition of both TSST-1 and SEA (1.5 ± 0.21) as compared to either TSST-1 (0.85 ± 0.22; P = 0.034, paired t-test, 2-tailed) or SEA (0.56 ± 0.35; P = 0.017, paired t-test, 2tailed) alone. A significant increase in the mean TNFa level in culture supernatants was also  86  Figure 12. TNFa secretion from human PBMC upon co-incubation with TSST-1 and SEA. Human PBMC were incubated for 19.5 h with TSST-1, SEA or TSST-1 plus SEA and culture supernatants were tested for TNFa by immunoassay. Values represent the mean ± SEM in 5 human PBMC samples each studied in duplicate determinations. Background values (culture media alone) were below the levels of detection of the immunoassay. * significantly different from the TNFa level observed with TSST-1 or SEA alone (P < 0.05, paired t-test, 2tailed)  87  observed when TSST-1 was co-incubated with SEA at 10 pg/ml (2.1 ± 0.3) as compared to incubation with TSST-1 (0.85 + 0.22; P = 0.013, paired t-test, 2-tailed) or SEA (1.4 ± 0.27; P = 0.008, paired t-test, 2-tailed) alone. Similiar to the effect on T cell mitogenesis, incubation of human PBMC with TSST-1 and SEA appeared to cause an increase in TNFa production in an additive rather than synergistic manner, since the concentration of TNFa secreted upon co-incubation with TSST-1 and SEA did not significantly exceed the sum of the response observed with either toxin individually. Unlike the ability of TSST-1 and SEA to enhance TNFa secretion from human PBMC, IL18 or EL-6 levels Were not significantly increased upon co-stimulation with both toxins. The mean EL-IB level (ng/ml + SEM) in the culture supernatants of eight human PBMC samples upon stimulation with TSST-1 (10 pg/ml) and SEA (1.0 pg/ml) (5.34 ± 2 . 4 ) was not significantly greater than incubation with TSST-1 (5.61 ± 2.7; P = 0.84, paired t-test, 2tailed) or SEA (3.71 ± 2.0; P = 0.16, paired t-test, 2-tailed) alone (Figure 13). Increasing the concentration of SEA to 10 pg/ml did not alter the results. Similiarly, the mean EL-6 concentration (ng/ml ± SEM) measured in the culture supernatants of eight human PBMC samples upon addition of both TSST-1 (10 pg/ml) and SEA (1 pg/ml) (1.19 ± 0.43) was not significantly increased when compared to incubation with TSST-1 alone (0.99 ± 0.41; P = 0.5, paired t-test, 2-tailed), although a significant increase was observed in comparison to SEA alone (0.53 ± 0.28; P = 0.02, paired t-test, 2-tailed) (Figure 14). Increasing the concentration of SEA to 10 pg/ml did not change the results.  88  9 8 SEA 1 pg/mi  7 £  6  CD C i—-i  5  1  4  CD. _J  •  TSST-1 10 pg/m TSST-1 + SEA  3 2 1 0  Figure 13. EL-16 secretion from human PBMC upon co-incubation with TSST-1 and SEA. Human PBMC were incubated for 19.5 h with TSST-1, SEA or TSST-1 plus SEA and culture supernatants were tested for BL-16 by immunoassay. Values represent the mean ± SEM in 8 human PBMC samples each studied in duplicate determinations. Values observed with culture media alone were subtracted from the reported values.  ti  89  Figure 14. IL-6 secretion from human PBMC upon co-incubation with TSST-1 and SEA. Human PBMC were incubated for 19.5 h with TSST-1, SEA or TSST-1 plus SEA and culture supernatants were tested for IL-6 by immunoassay. Values represent the mean ± SEM in 8 human PBMC samples each studied in duplicate determinations. Background values (culture media alone) were below the levels of detection of the immunoassay.  90  4.2.2.  Characterization of the TSST-1+/SEA" mutant INT#1  Mutagenesis of sea in RN3984, a TSST-1 /SEA clinical strain isolated from a patient +  +  with menstrual TSS, was accomplished by plasmid integration as described in Materials and Methods. SEA in the culture filtrate of INT#1 was lower than the limit of detection by immunoassay, performed as described in Materials and Methods (Table 7). The amount of TSST-1 detected in the culture filtrate of INT#1 by immunoassay (as described in Materials and Methods) was not significantly different from the amount detected in the culture filtrate of RN3984 (Figure 15). This indicated that inactivation of sea by plasmid integration did not interfere with TSST-1 production. Disruption of the SEA structural gene was verified by performing Southern hybridization analysis on Hindlll digested genomic DNA of INT#1 with a 624 bp BaniHI-HindJn. sea specific gene probe (Betley and Mekalanos, 1988). As can be seen in Figure 16, the sum of j  the INT#1 chromosomal flwdlll fragments that hybridized to the probe (—6.4, 5.1 and <2 <•  kb) was greater than the flwdlll fragment that hybridized in RN3984 (~4.0 kb) and was consistent with the size expected if pMLD6873C-l (~8.4 kb) recombined at the sea chromosomal locus. The presence of three flwdlll fragments that hybridized to the sea probe in INT#1 is proposed to have arisen via a single crossover event between the chromosomal sea of RN3984 and the inactivated sea of pMLD6873C-l near the 5' end of the gene as illustrated in Figure 17. The position of the Hindlll site in pMLD6873C-l indicated by an asterix is not definitively known since the orientation in which EcoRI digested pRN6441 inserted into pMLD6873 to generate pMLD6873C-l was not determined (refer to Materials and Methods). 91  Table 7. Detection of SEA levels in the Culture Filtrates of RN3984 and INT#1. Strain  [SEA] in Culture Filtrate (ng/ml)  RN3984  25.5 ± 3 . 7  INT#1  nd  SEA levels in culture filtrates were quantitated by immunoassay as described in Materials and Methods. Two single colonies of each strain were inoculated into 20 ml BHI and incubated at 37°C for 10 h. Culture filtrates were obtained by centrifugation and filter-sterilization. Values represent the mean ± SEM of 5 culture filtrates prepared from each strain tested in at least duplicate wells by immunoassay. The sensitivity limit of the immunoassay was 1 ng/ml of SEA. nd, not detected  92  1.0  Figure 15. Comparison of TSST-1 levels with respect to total protein in the culture filtrates of RN3984 and INT#1. Cultures were prepared by inoculating two single colonies into 20 ml BHI followed by incubation at 37°C for lOh. Bars represent the mean TSST-1 level (detected by immunoassay) ± SEM per fig of protein in 5 culturefiltratesprepared from RN3984 and 4 culturefiltratesprepared from INT#1.  93  12  3  4  Figure 16. Southern hybridization analysis of SEA" mutant genomic DNA digested with HindlH. and hybridized with a 624 bp 5a7nHI-i7/>?dni sea specific gene probe. Lane # 1. Digoxigenin-labelled DNA molecular size standards (kb) 2. RN450 (sea control isolate) 3. RN3984 4. INT#1  94  sea  624 bp BamHI-Hindlll probe  H  "2.2 kb H-B (624 bp sea • 1.4 kb Taq\)  "3.8 kb E-H (PRN6441)  Figure 17. A possible recombination event between pMLD6873C-l and sea in RN3984 to generate the SEA' mutant INT#1. The indicated point of integration of pMLD6873C-l into sea is consistent with the presence of three Hindlllfragmentsthat hybridized to the 624 bp BamiHL-Hindin. sea probe in Figure 16, one of which was ~6.0 kb (refer to molecular sizes of the constructed plasmids in Materials and Methods). The dotted line represents the chromosomal DNA of RN3984 and the hatched box represent sea in RN3984. The unfilled box represents sea in pMLD6y873C-l inactivated by the 1.4 kb Taqlfragmentof pRN6441 (indicated by solid box) as described in Materials and Methods. H, Hindlll; H*. indicates possible location of Hindlll (see Materials and Methods); B, BamW; E, EcdRl  95  Digestion of INT#1 with Clal and hybridization with the 1.4 kb Taql probe revealed one fragment with homology (Figure 18). The 1.4 kb Taql fragment encoded Em and was used r  to inactivate sea in pMLD6873 (as described in Materials and Methods, section 2.5.3). The faint band observed slightly above the darker band in Figure 18 is believed to have arisen by hybridization with incompletely digested chromosomal DNA. The presence of only one band that hybridized to the probe is consistent with the premise that pMLD6873C-l inserted only once into the chromosome. It should be noted that for some unexplained reason, one of the Clal sites was destroyed in pMLD6873 upon insertion of the 1.4 kb Taql fragment (see Materials and Methods, section 2.5.3). This explains why a Clal fragment larger than 1.4 kb was observed upon digestion of INT#1 with Clal followed by hybridization with the 1.4 kb Taql probe (Figure 18). 4.2.3. Virulence of RN3984 and I N T O in two animal models of lethal shock A.  Tampon-associated vaginal infection model  To determine whether a TSST-1 /SEA strain of S. aureus is more virulent than a TSST+  +  1 /SEA" strain, monitoring of lethality, rectal temperature, weight loss, incidence of lipaemia +  and incidence of diarrhea was performed over eight days in 24 NZW rabbits - 10 infected with either RN3984 or INT#1, and 4 infected with culture media alone. Monitoring of rabbits was performed blinded with respect to infection group. As outlined in Table 8, no difference in lethality was observed between rabbits infected with RN3984 or INT#1. Other clinical parameters such as mean rectal temperature (Table 8), percent weight loss (Figure 20), and incidence of lipaemia (Table 8) were also not significantly different between the groups. Although the trends in weight loss were similiar to those observed for RN3984 and  96  Figure 18. Southern hybridization analysis of SEA" mutant genomic D N A digested with Clal and probed with a 1.4 kb Taql fragment of pRN6441. The presence of only one band that hybridized to the probe indicated that pMLD6873C-l had integrated only once into the chromosome. The faint band with a higher molecular size observed to hybridize to the probe is believed to be the result of incomplete digestion of chromosomal DNA. Lane # 1. Digoxigenin-labelled DNA molecular size standards (kb) 2. RN3984 3. ENT#1  97  u •o o  E  c  CU  ISS  o G  c  -H  o -H  O  ©  o  ©  -H  > 1)  ll  0 CO 08 1 C o a,  oo  1 5  ©*  o  CO  c  8.  o oo  e "J  00  ON  s "3 .S  S  o  ~  M  S 3 T3  e OO  ° CO SH 00  3 (2  S c  m  .a OS  C  o u  2 •3 D 98  0  1  0  '  1  1  1  2  3  .  1  4  5  I  1  6  7  •  8  I  9  Days from infection  Figure 19. Weight loss in rabbits infected with RN3984 (#), INT#1 (&), or Media alone (X). Values represent the mean ± SEM in all available measurements gathered from rabbits on each day.  99  RN7043 in that the INT#1 infected group began to gain more weight in comparison to the RN3984 infected group, which may'be suggestive of less morbidity upon TNT#1 infection, no statistically significant difference was observed between the groups on any observation day. By comparing the weight loss graphs of RN3984 and RN7043 (Figure 5) and RN3984 and INT#1 (Figure 19) it was readily apparent that up until Day 6 there was a greater difference in weight loss between rabbits infected with RN7043 and RN3984 than rabbits infected with INT#1 and RN3984. Detection of circulating rabbit TNFa by immunoassay (described in Materials and Methods) in the sera of the infected rabbits over the 8 day observation period was also attempted. Unfortunately, only 3 of 23 rabbits examined had detectable levels and thus differences between the groups using this as a parameter was not useful. Assaying for systemic TNF has many potential pitfulls, including the presence of specific binding proteins in the sera such as autoantibodies (Jeffes et al., 1989) and soluble TNF receptors (Chouaib et al., 1991), that may interfere with detection. In addition, due to the rapid clearance of TNF in vivo, the timing of blood sampling is a crucial factor (van Deuren, 1994). The collection periods may not have been optimal in this study. As expected, a significant increase was observed in the number of rabbits that exhibited diarrhea upon RN3984 infection in comparison to INT#1 infection (4/8 vs 0/8; P < 0.05, Fisher's test, one-tailed) (Table 8). This indirectly supports SEA production by RN3984 and stability of the SEA" phenotype in INT#1 in vivo, since SEA is a common cause of staphylococcal food poisoning, in which one of the main symptoms is diarrhea.  100  B.  D-galactosamine sensitized mouse model  As another test of the TSST-1 /SEA and TSST-1"7SEA" strains, a second animal model +  +  of lethal shock was used to compare the virulence of RN3984 and INT#1. After 72 h, 12 of 15 mice infected with RN3984 died as compared to 8 of 15 mice infected with INT#1 (P = 0.12; Fisher's test, one-tailed). 2 of 2 mice infected with culture media alone survived. In order to conclude that INT#1 was less virulent than RN3984, a ^50% decrease in mortality was desired. Using a sample size of 15, this experiment yielded a 20% chance of observing a Type II statistical error. 4.2.4.  In vivo stability of the SEA' phenotype of INT#1  Although plasmid integration can be utilized to generate site-specific mutations in S. aureus (Foster et ah, 1990; Proctor 1992), its use has been limited by concerns that this type of mutation may be unstable after prolonged growth in vivo as a result of the presence of direct duplications flanking the insertion (Foster et ah, 1990). In light of this, the stability of the SEA" phenotype of INT#1 was monitored for five days upon inoculation into BALB/c mice. Production of SEA by bacteria isolated from mice was tested by immunoassay as described in Materials and Methods. All of the 97 colonies screened from 15 mice over a five day period were SEA". Stability of the SEA" phenotype of INT#1 after in vivo growth in rabbits was also examined. While performing the experiments in rabbits, bacteria were recovered from tampons between Days 1 and 8 and analysed for SEA production. All of the cells from 26 colonies recovered from 9 rabbits infected with RN3984 produced SEA, in comparison to 0 of 67 cells from colonies recovered from 7 rabbits infected with INT#1. This suggests that the SEA" phenotype of INT#1 is stable in vivo, and that the lack of  101  difference in mortality and morbidity observed upon infection with RN3984 and INT#1 in both animal models is not due to reversion of the SEA" phenotype in INT#1. 4.2.5.  In vitro effect of culture filtrates from isogenic strains on rabbit splenocytes A.  T cell proliferation  To determine if the culture filtrates of the TSST-1 /SEA and TSST-1 /SEA" strains +  +  +  differed in their ability to induce T cell proliferation, diluted culture filtrate preparations of RN3984 and ENT#1 were added to rabbit splenocytes. A mitogenic effect was observed upon incubation of rabbit splenocytes with each of two independently prepared culture filtrate samples from RN3984 and INT#1 (Figure 20). A significantly higher amount of T cell proliferation as assessed by [ H]thymidine incorporation into cellular DNA (described in 3  Materials and Methods) (cpm ± SEM) was observed in splenocytes incubated with RN3984 culture filtrate as compared to INT#1 culture filtrate at a dilution of 1/10,000 (410,831 ± 108,121 vs 202,191 + 76,968; P = 0.038, paired t-test, 2-tailed). The decreased T cell proliferation caused by INT#1 culture filtrate compared to RN3984 culture filtrate could be explained by the lack of SEA. If this were true, adding SEA to the INT#1 culture filtrate would be expected to increase the level of T cell proliferation back to the level observed with RN3984 culture filtrate. To test this, purified SEA was added with a 1/10,000 dilution of INT#1 culture filtrate (at a concentration that equalled the known amount present in RN3984 culture filtrate as determined by immunoassay) to rabbit splenocytes and the amount of T cell proliferation was measured by [ H]thymidine incorporation into cellular DNA. As observed 3  in Figure 21, addition of SEA to the INT#1 culture filtrate restored the amount of T cell proliferation back to wild-type levels.  102  \  1200000-  c o ra o  1000000-  Q. O  800000  u  g  s  600000  XI E  400000  ^  200000 0  1/10  1/100  1/1,000  1/10.000  Dilution  Figure 20. Mitogenic response induced by RN3984 and INT#1 culture filtrates on rabbit splenocytes. Values are expressed as the mean ± SEM measured by [ H]thymidine incorporation into cellular DNA in 7 rabbit splenocyte samples each tested in a minimum of triplicate determinations. BHI served as the media control. * indicates a significant difference in proliferation compared to RN3984 (P = 0.038, paired t-test, 2-tailed) 3  103  c o 4-» CXJ  o  CL L.  o u c c TJ  Cl  O  E >->  BHI Media RN39&4 INT#1  INT#1 + SEA  Figure 21. Increased proliferative response of rabbit splenocytes to RN3984 culturefiltratewas consistent with the presence of SEA. A 1/10,000 dilution of ENT#1 culturefiltratewas added with purifed SEA (at a concentration that equalled the known amount found in the diluted RN3984 culturefiltrateas determined by immunoassay) to rabbit splenocytes and the amount of T cell proliferation was assessed by incorporation of [ H]thymidine into cellular DNA. Results are the mean ± SEM in 5 rabbit splenocyte samples each tested in a minimum of triplicate determinations. * indicates a significant difference in proliferation compared to RN3984 (P < 0.05, paired t-test, one-tailed); ^ indicates no significant difference compared to RN3984 (paired t-test, onetailed) 3  104  B.  TNFa secretion  Culture filtrates of RN3984 and INT#1 stimulated rabbit splenocytes to secrete TNFa (Figure 22). No difference in the amount of TNFa secreted from rabbit splenocytes was observed upon incubation with RN3984 or INT#1 culture filtrate at any dilution examined. Two independently prepared culture filtrate samples of RN3984 and INT#1 were used for these studies. 4.2.6.  Enhancement of T cell proliferation by TSST-1 and SEA in rabbit splenocytes is concentration dependent  The above observations suggest that although there was no difference between the isogenic strains in inducing mortality or morbidity in vivo, RN3984 induced more T cell proliferation than INT#1 in vitro at low concentrations (Figure 20). Since T cells are known to play a central role in mediating toxicity upon superantigen stimulation, their stimulation should have resulted in an in vivo effect (Marrack et al., 1990; Miethke et al., 1992; Miethke et al., 1993). However the data also showed the ability of TSST-1 and SEA to cause an enhancement in T cell proliferation was concentration dependent since a difference in T cell proliferation observed with RN3984 culture filtrate as compared to INT#1 was found only at the highest dilution used (1/10,000). As a further test, the mitogenic response of rabbit splenocytes to TSST-1 and SEA was determined (Figure 23 and 24). Although direct comparisons between rabbit splenocyte preparations is difficult due to the variations that exist in both the amplitude and doses required to induce mitogenesis, the TSST-1 induced mitogenesis curve observed with the pooled rabbit splenoctye sample was similiar to that previously reported by Poindexter and Schlievert (1985), in that TSST-1 induced T cell  105  Figure 22. TNFa secretion induced from rabbit splenocytes by RN3984 and INT#1 culture filtrates. Values represent the mean ± SEM of rabbit splenocyte preparations tested in a minimum of duplicate determinations. BFfl served as the media control and gave background values of <0.04 ng/ml. Numbers in brackets indicate the number of rabbits tested at each dilution. j  106  50000 c o  Log [TSST-1] ug/ml  Figure 23. Mitogenic response of a pooled rabbit splenocyte preparation to purified TSST-1 as measured by [ H]thymidine incorporation into cellular DNA. Each point represents the mean ± SEM from a pooled rabbit splenocyte preparation (S rabbits) studied in a minimum of sue replicate determinations. The solid bar represents the mean ± SEM observed with RPMI culture media alone. Boxes labelled ' A ' and B ' indicate the' high and low concentrations used for determining whether the increase in T cell proliferation observed with RN3984 was concentration dependent. 3  107  100000 c o nl i_ O Q.  S c  80000  60000  40000  E X  20000  0  J  RPMI - 8  -7  -6  1  -5  I  L  -4  -3  - 2 - 1 0  1  Log [SEA] ug/ml  Figure 24. Mitogenic response of a pooled rabbit splenocyte preparation to SEA as measured by [ H]thymidine incorporation into cellular DNA. Each point represents the mean ± SEM from a pooled rabbit splenocyte preparation (S rabbits) studied in a minimum of six replicate determinations. The solid bar represents the mean ± SEM observed with RPMI culture media alone. Boxes labelled A ' and B ' indicate the high and low concentrations used for determining whether the increase in T cell proliferation observed with RN3984 was concentration dependent. 3  108  proliferation over a broad dose range. To determine if there was a concentration dependent effect of RN3984 to induce T cell proliferation, concentrations of TSST-1 and SEA that induced high and low mitogenic responses on rabbit splenocytes were chosen (Boxes ' A ' and ' B ' respectively, Figures 23 and 24). Minimally effective concentrations of TSST-1 (100 pg/ml) and SEA (1 or 10 pg/ml) (Box 'A') were co-incubated in 9 rabbit splenocyte preparations in 13 combinations (refer to Table 9). Next, higher concentrations of TSST-1 (100 ng/ml) and SEA (1 or 10 ng/ml) (Box 'B') were co-incubated in 5 rabbit splenocyte preparations in 9 combinations (refer to Table 10). A significant increase in proliferation was observed in 8 of 13 rabbit splenocyte preparations co-incubated with TSST-1 and SEA at the minimally effective dose (Box 'A') as compared to incubation with either toxin alone (P < 0.05, Mann-Whitney test, 2-tailed; Table 9). In contrast, only 1 of 9 rabbit splenocyte preparations co-incubated with higher concentrations of TSST-1 and SEA (Box 'B') exhibited a significant increase in T cell proliferation when compared to incubation with either toxin alone (P < 0.05, Mann-whitney test, 2-tailed; Table 10). The difference between these ratios was significant, in that low doses of TSST-1 and SEA combined enhanced T cell proliferation more often than did high doses (8/13 vs 1/9; P < 0.05, Fisher's test, 2-tailed) when compared to each toxin independently. These results suggest that the ability of TSST-1 and SEA combined to cause an enhancement in T cell proliferation over either toxin alone is dependent on toxin concentration. In light of these results it may be proposed that although RN3984 culture filtrate induced greater T cell proliferation than INT#1 in vitro (at a 1/10,000 dilution), perhaps this effect was not observed in vivo because TSST-1 and SEA are co-produced at  109  110  1 ©  o r— <s o \o m -H -H —' vo 00 < N  r-_ CO  o** vi"  00  < N  5 CD  00*  VO OS  00  ©*  -H  VO  •H  - 8 5c -H -H  «N*S  \o oo  5  o  -  r-  es —' •H -H «n oo  SI  00  1  §s  •i  00  o  OS  00  o <* VO  <N  -H -H « r> — CM 00 CM^  ~  Ol  VO  ci  Si 3 -H -H  Cvl 00  Ov  "\ **! r*>* 00*  o  Ov  r*T Ov vo«" r— --T r-~ — vo*  <N  00  O  00  ov  co vO CM  f  00  v>  <N O* m  o  o -H  -H  -H ©  V>  VO  00  <  tN O O *  co 00  Ov  °°. OS  vO  00  vO  00  -=  n e e H CO  >  -  co H CO  44 I I e s  w>"5b s B  -a -a S  E*  r i CO  i  co H (- co  CO  •  co H  12  co ,• co H co H  c  H co H ^ co H  co  52l  fa  s!a|E 8  O  c  4)  o  > I . -1 B  111  higher levels than that found in the RN3984 culture filtrate at this specific dilution. It should be noted that under the in vitro conditions used in this study, the undiluted culture filtrate of RN3984 contains -100 ng/ml of TSST-1 and - 30 ng/ml of SEA. At these inputs of toxin, an additive effect in T cell proliferation in comparison to the effect observed with each toxin individually would not be predicted. 4.3. Discussion The importance of TSST-1 in the pathogenesis of menstrual TSS was examined initially in these studies by introducing isogenic mutants in TSST-1 production into a tampon-associated vaginal infection model. The results showed that the TSST-1 strain of S. aureus was more +  lethal in the animal model than its TSST-1" isogenic derivative, suggesting that TSST-1 does play a causative role in MTSS. Confounding these results however was the decreased production of SEA by the TSST-1" mutant. If SEA plays an important role in the pathogenesis of MTSS, then the results surrounding the importance of TSST-1 in MTSS that generated from the use of these isogenic mutants in vivo would not be conclusive. This concern prompted studies to determine whether or not SEA plays a role in conjunction with TSST-1 in causing TSS. It was hypothesized that it might contribute to pathogenesis since TSST-1 and SEA is a clinically relevant toxin combination, and SEA is a superantigen. Coproduction of two superantigens by S. aureus might increase the virulence of the organism through enhancing the induction of T cell mitogenesis and cytokine secretion that are believed to be involved in pathogenesis. Since binding of TSST-1 and SEA to the MHC Class II molecule is a prerequisite for T cell mitogenesis and cytokine induction, it was reasonable to assume that at minimally effective concentrations of both toxins, an enhanced  112  effect in these responses would be observed upon co-incubation. Consistent with this was the observation that co-incubation of TSST-1 and SEA with human PBMC significantly enhanced both T cell proliferation and TNFa secretion compared to the effect observed with either toxin alone. These results were consistent with the hypothesis that production of SEA by TSST-1 strains may increase the virulence of the organism, and further analysis of the +  effects of this toxin combination in vivo was performed. To determine if a TSST-1 /SEA strain was more virulent than a TSST-1 /SEA' strain, +  +  +  introduction of the strains into two animal models of lethal shock was performed. Results in both the tampon-associated vaginal infection model and the D-galactosamine sensitized mouse model indicated that the TSST-1 /SEA strain did not cause more mortality or morbidity +  +  than the TSST-1 "7SEA' strain. Stability of the SEA' phenotype of INT#1 after in vivo growth in both BALB/c mice and NZW rabbits was verified. Although the animal models used in this study were quite different, both the mouse model and the rabbit infection model are relevant for analysing the role of staphylococcal exoproteins in TSS since the mechanisms leading to shock are similiar in the two cases. TSS is believed to be caused by the overstimulation of T cells by superantigens followed by excessive secretion of cytokines such as TNF, which can cause shock in vivo (Okusawa et ah, 1988; Dinarello et ah, 1989). Monoclonal antibodies against TNF are protective in a rabbit model of TSS (Parsonett and Gillis, 1988). T cells have been shown to be important in the rabbit model since TSST-1 administration to rabbits caused a significant increase in PBMC counts as compared to non-toxin treated control rabbits. In addition, the increased PBMC counts correlated with enhanced lethality in the study (Lee et ah, 1991). Lethal  113  shock in D-galactosamine sensitized mice upon superantigen stimulation also occurs via a T cell/TNF-dependent pathway since SCID mice lacking T cells were resistant to the lethal effects of either TSST-1 or SEB (Miethke et al, 1992; 1993). In addition, cyclosporin A treatment (a reagent that suppresses cytokine secretion by T cells) and anti-TNF monoclonal antibodies were protective against superantigen-mediated lethality in this mouse model (Miethke et al., 1992; 1993). In support of the in vivo studies, incubation of rabbit splenocytes with RN3984 or INT#1 culture filtrates in vitro produced no difference in the amount of TNFa secreted. Although RN3984 culture filtrate induced more T cell proliferation from rabbit splenocytes in vitro than INT#1, this effect was concentration dependent. A concentration dependent effect in T cell proliferation would be expected with human cells since TSST-1 and SEA bind to the same receptor, the Class U antigen, on human peripheral blood mononuclear cells (Mollick et al, 1989; Scholl et al., 1989), and superantigen binding to Class II molecules is a prerequisite for the T cell proliferative response (Fleischer and Schlezenmeier, 1988; Fischer et al., 1989; See et al, 1992a). Thus although minimally effective concentrations of TSST-1 and SEA combined significantly increased T cell proliferation in human PBMC compared to incubation with either toxin alone, due to the overlapping receptors these toxins share, at higher concentrations of toxin this additive response would not be anticipated. Whether the concentration dependent effect of T cell mitogenic enhancement from rabbit splenocytes upon TSST-1 and SEA co-incubation may be mediated by the binding of a common receptor is only conjecture at this time. As with T cell mitogenesis, TSST-1 and SEA induced cytokine release from human PBMC  114  is also Class n dependent (Grossman et al, 1990; Matsuyama et al., 1993; Akatsuka et al., 1994), and as such would be expected to be concentration dependent. Therefore, although the combination of TSST-1 and SEA at the specific concentrations used in this study caused a significant increase in TNFa secretion from human PBMC compared to incubation with either toxin alone, it is predicted that this would not occur at higher concentrations of toxin. The ability to show a concentration dependent effect in T N F a secretion from rabbit splenocytes upon stimulation with the culture filtrates of the TSST-1 /SEA and TSST+  +  1 /SEA" strains was restricted by the sensitivity limitations of the rabbit TNFa immunoassay. +  Therefore, since the amount of TSST-1 and SEA produced by S. aureus in vivo is not known, it can not be predicted whether or not an enhancement of T cell mitogenesis and TNF secretion occurs during human disease. The evidence presented here suggests that this effect would not be observed, since the pathogenicity of the strains was not different. The in vitro studies of toxin induced cytokine secretion from human PBMC revealed variability in cytokine responses (Figures 12, 13 and 14). In contrast to TNFa induction, an increase in either IL-1B or IL-6 levels in the culture supernatants of human PBMC stimulated with both TSST-1 and SEA as compared to incubation with either toxin alone could not be demonstrated. It is plausible that TSST-1 and SEA enhance the secretion of one cytokine but hot others, since these cytokines appear to be differentially regulated (See and Chow, 1992; Trede et al., 1994). Cytokine induction may occur via different signal transduction pathways (See et al., 1992b). Whether the different activation pathways induced by these toxins are related to the unique epitopes which TSST-1 and SEA bind on human PBMC (Karp et al., 1990; See et al., 1990; Herman et al, 1991; See et al, 1992a) is not  115  known, but it does introduce another degree of complexity in superantigen induced cytokine expression. These results suggest that when TSST-1 is produced by S. aureus, the concomitant production of SEA is irrelevant in terms of mediating overall enhanced lethality or morbidity. SEA may contribute to the virulence of S. aureus in a manner that was not monitored in these studies. For example, SEA production by TSST-1 strains may provide +  benefit in either colonization or survival. Furthermore, TSST-1 could interact specifically with a determinant associated with SEA in this particular clone to cause enhanced virulence. For example, SEA is carried by a triple converting phage that mediates positive conversion for staphylokinase and SEA expression, and negative conversion for fl-lysin (Coleman et al., 1984). It is possible that it is staphylokinase, a plasminogen activator, and not SEA, that interacts with TSST-1 to cause enhanced virulence, or that inactivating B-lysin gives some advantage to this clone. An association between TSST-1 and SEA production may have arisen entirely by chance. Alternatively, the phage encoding SEA may prefer some aspect in the menstrual TSST-1 host background that favours its insertion, and while no biological +  significance ensues, a correlation between TSST-1 and SEA production was made. In summary, these results provide evidence to suggest that production of TSST-1 by S. aureus appears to be the primary toxin involved in causing menstrual toxic shock syndrome. Although the TSST-1" mutant (RN7043) was found to produce less SEA than the TSST-1  +  strain in Chapter 3, the results in this chapter suggest that SEA does not enhance the virulence of this particular S. aureus strain. The differences observed in vivo and in vitro with RN3984 and RN7043 in Chapter 3 therefore was not likely due to decreased production  116  of SEA by RN7043. It does remain a possibility however that production of SEA with TSST-1 among MTSS isolates aids the organism in other properties not investigated in this study. Exploring other putative benefits SEA may provide to TSST-1 producing S. aureus would be of clinical relevance due to the observed frequency of this toxin combination among MTSS isolates.  117  Chapter 5.  Demonstration of whether the mutually exclusive production of TSST-1 and SEB in S. aureus is mediated by lack of structural genes  5.1.  Introduction  Upon establishing a causal role for TSST-1 in menstrual TSS, subsequent objectives of this research was to examine the interaction of TSST-1 with other toxins, specifically SEA and SEB. TSST-1 and SEA was a toxin combination of interest due to its observed frequency among menstrual TSS isolates. Although this suggested that a S. aureus strain producing both TSST-1 and SEA may be more virulent than a strain producing TSST-1 alone, this did not appear to be the case when TSST-1 /SEA and TSST-1 VSEA" strains were introduced +  +  into two tests of lethal shock in animals. In contrast to the frequent co-production of TSST-1 and SEA, TSST-1 and SEB coproduction appears to be mutually exclusive at the phenotypic level in S. aureus. At the genotypic level, results of various studies that have examined whether TSST-1 and SEB structural genes exist in the same isolate have been inconclusive. Because of this, and in light of my continued interest in the underlying mechanisms that act to favour or exclude certain toxin combinations in organisms, I explored possible mechanisms that may be mediating mutual exclusivity of TSST-1 and SEB in S. aureus. 5.2. 5.2.1.  Results Detection of tst in TSST-1 and TSST-1" S. aureus isolates +  To begin, Southern hybridization analysis was used to determine whether the TSST-1 and SEB toxin determinants could be found within the same S. aureus isolate. Six TSST-1" " 1  118  clinical isolates (7140, 7051, 8270, 7294, 7616, VI-5) chosen from our clinical collection based on their known status of TSST-1 and SEB production as determined by ELISA (Lee et al, 1992) and two TSST-1 reference isolates [RN3984, FRI1169 (Kreiswirth et al, 1989)], +  were confirmed to contain the TSST-1 structural gene by Southern hybridization analysis of either Clal digested (Figure 26) or Hindlll digested (Figure 27) genomic DNA using a 297 bp Hincll-BamH.1ttf-specificgene probe (Kreiswirth et al, 1989) (Refer to Figure 25). One 5. epidermidis isolate and six S. aureus isolates found to be TSST-1" by ELISA (RN450, 0507, 7660, 8527, 8528 and 8530) served as negative controls, and as expected, none hybridized with the tst probe (Figures 26 and 27). Clal is a restriction enzyme that has been reported to not cut within the 4 to 7 kb TSST-1 genetic element [Kreiswirth et al, 1989]. However the presence of fragments less than 4 kb (Figure 26) in 2 TSST-1 isolates (7051, +  7294) could indicate the presence of a Clal restriction site within the TSST-1 genetic element of these isolates, suggesting that there is a restriction fragment length polymorphism (RFLP) in this region. Alternatively, there could have been a deletion in the TSST-1 genetic element. The presence of 3 Clal-tst hybridization patterns in the remaining 6 isolates is consistent with previous observations that tst is found in multiple loci in the S. aureus chromosome as determined by protoplast fusion and transformation mapping (Chu et al, 1988). Hindm digestion also revealed three tst hybridization patterns among the TSST-1 clinical +  isolates (Figure 27). RN7043 is an isogenic derivative of RN3984 derived by the insertion of a tetracycline resistance determinant (with an internal Hindm site) within the tst locus. Expectedly, hybridization with the tst probe revealed a smaller Hindlll-tst fragment in  119  tst PBR322  PBR322  297 bp Hincil-BamH!  1-0 kb BamHI 890 bp BamHI-Sty!  Figure 25.  Restriction map of pRN6550 (pBR322 carrying tst and associated TSST-1 genetic element as depicted by Kreiswirth et al., 1989) and location of fragments used as probes. B, Bamm, H, //well; S, Sty!  120  o  5  05  T- o t m in r»o) f r Z C ' - O 01 cy & 1 cc. co eo_ o n u. s co N s >  n o  2 10 m w  .4.2  !3.  5  -  aijo  n -  m  •  2.0-  Figure 26.  Southern hybridization analysis of TSST-1* (RN450, 8527, 8530, 0507) and TSST-1 (RN3984, FRI1169, 7140, 7051, 8270, 7294, 7616, VI-5) S. aureus isolates digested with Clal and probed with a 297 bp HincUBamHL tst specific gene probe. Numbers at left are digoxigenin-labelled DNA molecular size standards (kb). The arrows indicate the common Clal fragment that hybridized to both the tst probe and the 150 bp Taql seb specific gene probe in Figure 29. Only S. aureus 7051 did not have a common Clal fragment that hybridized to the tst and seb probes. +  121  0)  •H  g  "o  ^  «  "H ^ ro vo  rd  g  <D CO d l ^ O H O H l f l O Q) 7"J  ' £ £j P$ H O VD U) -H  21 5.1-  2.0-  Figure 27. Southern hybridization analysis of TSST-T isolates digested with Hindlll and probed with a 297 bp Hincll-BamHl tst specific gene probe. S. aureus 8528 and 7660 and S. epidermidis served as the negative controls. Numbers at left are the positions of the digoxigenin-labelled DNA molecular size standards (kb).  122  RN7043 as compared to RN3984. Of special note are isolates 0507 (Figure 26) and 8528 (Figure 27), both of which are TSST-17SEB . The Clal- andffwdlH-digestedchromosomal +  DNA of these isolates did not hybridize with the tst probe, confirming that tst and seb do not share DNA homology. 5.2.2.  Detection of seb among TSST-1 and TSST-l" S. aureus isolates +  To examine if TSST-1 isolates do or do not contain the SEB structural gene, Southern +  hybridization analysis was performed on the 8 Clal digested TSST-1 isolates using a 150 bp +  Taql seft-specific gene probe (Johns and Khan, 1988; Figure 28). The Clal digested chromosomal DNA of all 8 TSST-1 isolates hybridized with the seb gene probe (Figure 29). +  The two well-characterized TSST-17SEB- reference isolates RN450, 8530 did not hybridize, while the SEB /TSST-1" reference isolate 0507 did. +  Of special interest is the observation that in all isolates but one (7051), the 150 bp Taql reft-specific gene probe hybridized to a similiar sized Clal fragment that also hybridized to the 297 bp Hincll-BaniHl tst-specific gene probe in Figure 26. This suggested that tst and seb may be located close to each other in the chromosome of these isolates. Also, although all of the TSST-1 isolates appeared to contain the Taql fragment from seb, none produced +  SEB in their culture supernatants as determined by noncompetitive ELISA (Lee et al., 1991). The presence of more than one Clal fragment that hybridized to the seb gene probe in some isolates indicates that seb may be present in multiple copies in these isolates. Alternatively, it could also suggest the presence of frequent RFLPs in the region of seb, such that in some isolates there may be an additional Clal site within seb in the region of the 150 bp Taql fragment.  123  E  H  , P B R 3 2 2  1  I  I  E  K  I  L  C  C  X  E  -  H  seb  www  L IWOAWAWWll I I I  I  PBR322  H 160 bp  Figure 28.  Restriction map of pSK155 (pBR322 carrying seb and SEB genetic element sequences as depicted by Johns and Khan, 1988) and the location of the 150 bp Taql seb specific gene probe. H, Hindlll, E , £coRI; K, Kpril, Q,Clal,X,Xbal •junction of SEB genetic element and common chromosomal DNA as suggested by Johns and Khan (1988)  124  2.0-  Figure 29.  Southern hybridization analysis of TSST-1* (RN3984, FRI1169, 7140, 7051, 8270, 7294, 7616, VJ.-5) and TSST-F (RN450, 8530, 0507) S. aureus isolates digested with Clal and probed with a 150 bp Taql se£-specific gene probe. 0507 is a SEBTTSST-1* strain. Numbers at left are the positions of the digoxigenin-labelled DNA molecular size standards (kb). The arrows indicate the common Clalfragmentthat hybridized to both the 297 bp Hincll-BamHl tst probe (Figure 26) and the 150 bp Taql seb probe in each isolate S. aureus 7051 did not have a common Clal fragment that hybridized to both probes.  125  5.2.3.  P C R analysis of TSST-1* isolates with oligonucleotide primers specific for seb  The observation that all TSST-1 isolates examined contained sequences homologous to +  seb appeared to be at variance with reports of other investigators who rarely found seb among TSST-1 isolates using oligonucleotide probes (Bohach et al., 1989; Neill et al., +  1990; Johnson et al., 1991). It is possible that the discrepancy in results arises from the use of probes located at different positions in seb. The probes used in previous studies were targeted to sequences located futher downstream in the seb nucleotide sequence than the 150 bp Taql fragment located at nucleotide positions 40-195 of the seb structural gene of 798 nucleotides. To resolve this discrepancy and hopefully provide further support for the Southern hybridization results, the seb nucleotide primers SEB1 and SEB2, that targeted sequences downstream from the 150 bp Taql fragment at positions 310-329 and 634-657 respectively of the SEB structural gene (see Figure 30) were used to perform PCR analysis on 7 of the TSST-1 isolates (as described in Materials and Methods). Two reference SEB +  +  isolates (0507, 8528), and 3 SEB clinical isolates (7405, 7431, 8292) served as positive +  controls, while 4 SEB" S. aureus isolates (8527, 8530, 8531, RN450) and 5 non-5, aureus isolates (5. epidermidis, S. haemolyticus, S. pyogenes, B. subtilis and E. coli) were used as negative controls. Even though all of the TSST-1 isolates hybridized with the 150 bp Taql +  sefc-specific gene probe, only 1 of these 7 isolates yielded the expected 478 bp PCR product after amplification with the SEB1 and SEB2 primers (Figure 31). Six of the seven isolates showed no product. Thus the apparent discrepancy between the present results and those of others could be explained in part by the sequence location of the seb oligonucleotide probes used, in conjunction with the possibility that only partial SEB structural gene sequences may  126  5'  3' 798  seb  150 bp Taql  SEB-1  (40-106)  (310-329)  SEB-2 (634-667)  MLD-1 and MLD-2 (17-39)  Figure 30.  (162-181)  Location of the ISO bp Taql seb specific gene probe used for Southern hybridization analysis and the oligonucleotide primers used for PCR analysis with respect to the nucleotide sequence of the SEB structural gene. The numbers in brackets indicate the nucleotides that comprise the probe and primers.  127  TSST-1  +  I STRAINS  + + I a b c d e f g h  i  j k | mnopq  +  r s t  478 b p -  Figure 31.  Agarose gel electrophoresis of PCR products after amplification using oligonucleotide primers (SEB1 and SEB2) specific for seb located at positions 310-329 and 634-657 respectively in the SEB nucleotide sequence (Jones and Khan, 1986). Lanes: a.t.u.z: 100 bp ladder SEB" isolates L C J l s j - . : 0507, 8292, 7405, 8528, 7431 TSST-1 test isolates +  d.e.f.g.h.i.j: RN3984, FRI1169, 7140, 7051, 8270, 7294, 7616 SEB' isolates l.m.n.o.p.q.r.v.w: 8527, 8530, 8531, RN450, S. pyogenes, S. epidermidis, S. haemolyticus, B. subtilis, E. coli y_: Reagent blank  128  be present in the isolates studied. To confirm the results observed with Southern hybridization analysis, primers were designed that targeted sequences in the 150 bp Taql region of seb. Two primers designated MLD-1 and MLD-2, located at nucleotide positions 17-39 and 162-181 of the SEB structural gene (Figure 30) were used to perform PCR analysis on 7 of the 8 TSST-1 test isolates that +  hybridized to the 150 bp Taql probe in Figure 29. Two reference S E B isolates (0507 and +  8528) and 1 S E B clinical isolate (8292) served as positive controls, while 5 SEB isolates +  -  (8527, 8530, 8531, 8587, 7690) served as negative controls. One non-5, aureus isolate (S. pyogenes) also served as a negative control. Even though all TSST-1 isolates hybridized to +  the 150 bp Taql probe, none gave the expected 165 bp PCR product after amplification with the MLD-1 and MLD-2 primers (Figure 32). Although this experiment did not help to corroborate the Southern hybridization results, it indicated that the sequences targeted by either the MLD-1 or MLD-2 primer, or both, were not present in the TSST-1 isolates and +  suggested that sequences surrounding the 150 bp Taql region in these isolates are not completely homologous to seb. 5.2.4.  Immunoblot analysis of putative SEB /TSST-1 S. aureus isolates +  +  Our laboratory had previously reported the existence of 5 putative SEB /TSST-1 isolates +  +  among 344 clinical S. aureus isolates studied phenotypically by ELISA (Lee et ah, 1992). In light of the present findings, it was of interest to reanalyse these putative SEB /TSST-1 +  +  isolates to determine if they did in fact coexpress both toxins. Immunoblot analysis of concentrated culture supernatants of the isolates revealed that in fact none of them produced both toxins; 3 produced TSST-1 only, 1 produced SEB only, and 1 produced neither TSST-1  129  Figure 32.  Agarose gel electrophoresis of PCR products after amplification using oligonucleotide primers (MLD-1 and MLD-2) specific for seb located at positions 17-39 and 162-181 respectively in the SEB nucleotide sequence (Jones and Khan, 1986). "Lanes: a. s: 100 bp ladder SEB* isolates b. c.q: 0507,8528,8292 TSST-1 test isolates +  d.e.f.g.h.i.i.k: RN3984, FRI1169, 7140, 7051, 8270, 7294, 7616, VI-5 SEB' isolates L.m.n.o.p.t: 8527, 8530, 8531, S. pyogenes, 8587, 7690 r: Reagent blank  130  nor SEB. The SEB ELISA (as described in Materials and Methods) was then repeated for these 5 isolates, and the results were consistent with the observations made upon immunoblot analysis. 5.2.5.  Examination of S E B isolates with probes derived from sequences flanking tst +  Since all TSST-1" " S. aureus isolates analysed were SEB", even though all hybridized with 1  the 150 bp Taql seb probe, it was of interest to determine whether S. aureus isolates that did produce SEB also contained the TSST-1 genetic element. Nine S. aureus isolates were chosen that were found by hybridization analysis to be seb , and by ELISA to be SEB +  +  (0507, 7371, 7372, 7407, 7429, 7513, 7694, 7836, 8385). When these isolates were hybridized with the 297 HincZZ-BamflZ tst probe, no homology was evident (not shown). In addition, when these isolates were probed with a 1.0 kb BamHl fragment consisting of sequences upstream of tst and a 890 bp BaniRl-Styl fragment consisting of sequences downstream of tst (refer to Figure 25), no homology was observed (Figure 33 and 34 respectively). These two probes consist of DNA sequences within the region that has been reported to be unique to TSST-l-producing isolates (Kreiswirth et al., 1989). 5.3.  Discussion  The results presented here suggest that TSST-1 isolates do contain sequences that are +  homologous to the SEB structural gene, but none of the isolates produce SEB. In contrast, S. aureus isolates that do produce SEB do not contain tst. These observations are intriguing because it is known that little amino acid sequence homology exists between TSST-1 and SEB (Blomster-Hautamaa et al., 1986; Marrack and Kappler, 1990). Furthermore, there was no cross-hybridization between the 297 bp Hincll-BamEL ^-specific probe and the  131  FSS8.SS8 S C E E S i  Figure 33.  Southern hybridization analysis of Clal digested SEB genomic DNA probed with a 1.0 kb BamHI fragment derivedfrompRN6550 (see Figure 25). RN450, 8527 and 8530 served as the TSST-17SEB* control isolates. RN3984 was the TSST-1 control isolate +  +  132  Figure 34.  Southern hybridization analysis of Clal digested SEB genomic DNA probed with a 890 bp BammStyl fragment derived from pRN6550 (see Figure 25). RN450 and 8527 served as the TSST-17SEB' control isolates. RN3984 was the TSST-1 control isolate. +  +  133  chromosomal DNA of the SEB reference isolates (0507, 8528, Figure 26 and 27). +  With relative confidence, it can be stated that these results do not appear to be due to nonspecificity of the seb probe for the following reasons. First, hybridization reactions were performed under conditions of high stringency so the possibility of false-positive hybridization signals was unlikely. Second, none of the control isolates that did not contain seb hybridized to the 150 Taql probe (RN450, 8530, Figure 29). Third, if the seb probe hybridized nonspecifically to tst or the TSST-1 genetic element, it would be expected that the tst probe or TSST-1 genetic element probes would also show hybridization signals with the SEB isolates. This was not observed in any of the 9 SEB isolates examined. The +  +  possibility that the 150 bp Taql seb probe hybridized with coexisting SEC genetic determinants (known to share amino acid homology with SEB) was unlikely because the seb probe did not hybridize the S E C reference isolate 8530 (Figure 29). +  The observation that 8 of 8 Clal digested TSST-1 S. aureus isolates examined also +  contained SEB structural gene sequences appears at first to be at variance with results reported by other investigators. Using PCR analysis with primers located at positions 310329 and 634-657 in the SEB structural gene, Johnson et al. (1991) found that 3 (11%) of 28 TSST-1 isolates contained both TSST-1 and SEB genes. Neill et al. (1990) did not find any +  evidence of seb in TSST-1 isolates with an oligonucleotide probe located at positions 312+  329 of the seb structural gene sequence. Bohach et al. (1989), using an SEC1 probe derived from sequences internal to the gene and having 68% nucleotide sequence homology with the analogous region of seb, also concluded that TSST-1 isolates did not contain seb. It is +  possible that in our study various deletions within the SEB structural gene have occurred. In  134  an attempt to resolve this discrepancy, PCR analysis was done on the TSST-1+ isolates with the same primers as those used by Johnson et al. (1991), located to the right of the 150 bp Taql seb probe in Figure 30. As expected, most (6 of the 7) isolates did not show the 478 bp PCR amplification product (Figure 31). These results help to explain the incongruencies in various observations between different investigators and suggest that nucleotide sequences distal to the 150 bp Taql sequence are absent in the TSST-1 isolates. The presence of a +  478 bp PCR product in 1 (11%) of 7 TSST-1 isolates in this study correlated with the +  observed frequency (11 %) of TSST-1 isolates positive for seb reported by Johnson et al. +  (1991) using the same olignucleotide probes and PCR technique. Although TSST-1 isolates hybridized to the 150 bp Taql seb probe, this region did not +  appear to be completely homologous with the corresponding region of seb since PCR primers that targeted this region failed to amplify the expected PCR product. Deletions extending to this portion of seb sequences in these isolates is possible. Hybridization may have occurred in these isolates since a probe of 150 bp has a better chance of hybridizing to partial sequences than the shorter PCR primers. Alternatively, these TSST-1 isolates may contain +  a seb-like gene. The only known report that made use of a DNA probe overlapping the 150 bp Taql probe used in this study stated that 0 of 37 TSST-1 isolates hybridized in colony blots with the +  probe (Notermans et al., 1988). The reason for the discordant results between that study and this one is not immediately clear, except that only 41 bp of the 150 bp Taql sequence was present in the seb probe used by these investigators. Even though all TSST-1 isolates tested contained sequences homologous to seb, in not +  135  one instance did these isolates produce SEB in their culture supernatants. Many investigators have reported the observation that TSST-1 and SEB are not co-produced in S. aureus (Crass and Bergdoll, 1986a; Crass and Bergdoll, 1986b; Garbe et al, 1985; Bohach et al, 1990). However, others have provided evidence that some S. aureus strains do produce both SEB and TSST-1 (Ewan et al, 1989; Johnson et al, 1991; Lee et al, 1992). In fact, one of the latter reported examples was a study performed in our laboratory, in which a total of 344 S. aureus isolates were analysed for SEB and TSST-1 production by noncompetitive ELISA (Lee et al, 1992). It was reported that expression of SEB and TSST-1 was usually mutually exclusive, in that only 5 of 344 (1.5%) isolates were found to co-produce both toxins. Upon investigation by immunoblot however, none of these 5 putative SEB /TSST-1 isolates +  +  contained both toxins in their culture supernatants. The fact that determination of toxin profiles via immunological analysis can lead to false positive results is well recognized (Ewald et al, 1990; Johnson et al, 1991). The observations reported here overwhelmingly support the notion that expression of SEB and TSST-1 are mutually exclusive since in the study of Lee et al (1992), it can now be stated that 0 of 344 S. aureus isolates (119 of which were TSST-1 ) co-produce SEB and +  TSST-1 (see Table 11). Due to the important role SEB and TSST-1 play in the pathogenesis of NMTSS and MTSS respectively, further studies that address the mechanisms that underlie the acquisition and expression of these toxin determinants in S. aureus are justified.  136  Table 11. TSST-1 and SEB Production among S. aureus Isolates.  Patient Group (no. of isolates)  TSST-1 (•/.)  SEB* (%)  TSST-lVSEB* ( )  MTSS (22)  82  I  0  NMTSS (41)  68  20  0  Non-TSS (137)  22  11  0  Carrier (144)  26  10  0  +  %  MTSS, menstrual toxic shock syndrome; NMTSS, nonmenstrual toxic shock syndrome; Non-TSS, non-toxic shock syndrome-associated S. aureus infection; Carrier, no infection Data compiled from Chang et al. (1991), Lee et al. (1992) and updated with the results from this study.  137  Chapter 6.  6.1.  Possible Molecular Mechanisms for Mutually Exclusive Toxin Co-Production of TSST-1 and SEB in S. aureus  Introduction  In the previous chapter the observed mutual exclusivity of TSST-1 and SEB co-production in S. aureus was verified. Up until now, the molecular basis for this exclusivity in S. aureus has not been investigated. Hybridization analysis presented in Chaper 5 showed that 8 of 8 TSST-1 /SEB' isolates contained sequences homologous to the SEB structural gene. The +  hybridization results also revealed that tst and seb were localized to a restriction fragment of the same size in the majority of the isolates, suggesting a close proximity in the chromosome. However PCR analysis in Chapter 5 revealed that the sequences homologous to seb in TSST-1 isolates were not consistent with the presence of an entire SEB structural +  gene. As a result of these studies, it appeared that perhaps a novel regulatory system was acting to inhibit the expression of SEB in these TSST-1 isolates. In this chapter, +  exploration of several possibile causes for the observed mutual exclusivity was carried out. Based on the collective data, a molecular mechanism mediating the lack of SEB and TSST-1 co-production in S. aureus is proposed. 6.2. 6.2.1.  Results Creation of a SEB /TSST-1 strain of S. aureus +  +  Since SEB is not produced even though both tst and seb homologous sequences are found within the same isolate, it is possible that these genes may be competing for the same transcriptional activator. As one test of this hypothesis, an attempt was made to construct a 138  SEB /TSST-1 strain of S. aureus. If TSST-1 and SEB compete for a positive regulator, +  +  then TSST-1 and SEB co-expression would not be anticipated to occur in the constructed strain. pMLD6876-l (a plasmid that contained seb) and pJWl (a recombinant B. subtilis plasmid which contained tst) were introduced into the restriction-deficient, nonenterotoxigenic strain of S. aureus RN4220 as described in Materials and Methods. This resulted in the production of both TSST-1 and SEB in 5 of 5 RN4220 transformants tested by immunoblot as described in Materials and Methods (see Figure 35). The location of the band for SEB in Figure 35 at approximately 14 kD as opposed to its molecular mass of 28 kD is presumably due to the presence of "nicked" protein in the commercial source used, as has been previously described (Grossman et. al., 1990). These results suggest that lack of TSST-1 and SEB co-production in S. aureus is not mediated by competition for a common transcriptional activator. 6.2.2.  Analysis of TSST-1 isolates with probes derived from sequences flanking the +  SEB structural gene and located within the SEB genetic element Previously, TSST-1 isolates studied were found to contain seb homologous sequences by +  Southern hybridization analysis with a 150 bp Taql seb-specific gene probe (Figure 29). In addition, the tst and $eZ>-specific gene probes hybridized to the same sized Clal restriction fragment in these isolates, which indicated that the tst and seb genes were located near one another in the chromosome. To seek further evidence for the presence of SEB genetic element DNA in TSST-1 S. aureus isolates, Southern hybridization analysis using two +  additional probes derived from sequences upstream of seb was performed on the same 8 Clal digested TSST-1 genomic DNA examined previously. All SEB isolates are highly +  +  139  A  1  3  2  4  5  6  7 8 9  10  kd 69 30  14  B  1  ~~  2  3  4  ~  5  6  ~  7 F=  ^i =  8  9  10  ; —°l  kd  14  Figure 35. Immunoblot analysis of culture filtrates of CmVEm' transformants of RN4220 run on SDS-PAGE and blotted with a polyclonal anti-TSST-1 antibody (A) or a polyclonal anti-SEB antibody (B). Arrows indicate the postions of TSST-1 and SEB. Lane(s) # 1-5: Cm7Em transformants of RN4220 6: RN4220(pJWl)andRN4220(pMLD6876-l)inAandinB respectively 7: RN4220 8: FRI1169 (TSST-r control) and 0507 (SEB control) in A and in B respectively 9: TSST-1 (in A) and SEB (in B) standards (Toxin Technology) The numbers on the right indicate the positions of the molecular mass standards (kd) as determined by separating the Bio-Rad Kaleidoscope Pre-stained molecular size standards in Lane 10. r  +  140  homologous in a 26.8 kb region around seb (termed the SEB genetic element; Johns and Khan, 1988), and the upstream probes were located in this region.  The membrane that was  used for the experiment described in Figure 26 was washed to remove the probe and then reprobed with the 700 bp EcoKL fragment of pSK155 (Figure 36) that consisted of sequences located approximately 1.1 kb upstream of seb. In a separate experiment, Clal digested TSST-1 genomic DNA was probed with a 1.1 kb Hindlll-EcoRl fragment located further +  upstream of the 700 bp EcoKL probe but still within the SEB genetic element (Figure 36). The specificity of these probes was first evaluated. The 700 bp EcoKL SEB genetic element probe hybridized with a strong signal to Clal digested RN450, a TSST-17SEB" reference isolate cured of all known prophages (Murphy et al., 1981). Faint signals were also observed with two other SEB* reference isolates (8527, 8530, Figure 37). Similiarly, the 1.1 kb HindHl-EcoKL probe hybridized (albeit faintly) with RN450 and 8527 (Figure 38). Hybridization of a SEB genetic element probe to DNA from SEB" cells was not totally unexpected since Johns and Khan (1988) had previously reported that some SEB" isolates may contain an incomplete or related SEB genetic element. The most significant findings were that both flanking SEB genetic element probes hybridized to Clal digested TSST-1 chromosomal DNA, indicating that TSST-1 isolates +  +  also contain sequences that are homologous to the SEB genetic element. More importantly, in 6 of the 8 TSST-1 isolates, the 700 bp EcoKL fragment (Figure 37) and the 1.1 kb +  Hindlll-EcoKL fragment (Figure 38) hybridized to a Clal fragment that was similiar in size to the fragment that hybridized to the 297 bp Hincll-BamHL tst probe shown in Figure 26. Only in isolates 7294 and 7616 was the Clal fragment that hybridized to the SEB genetic  141  (  H  E  K  seb  PBR322  C G X  E  *  H  PBR322  I 1 700 bp  1.1  1.8 kb 14 kb  kb  Figure 36. Restriction map of pSK155 [as depicted by Johns and Khan (1988)] and location of fragments used as probes. H,tfmdTJJ;E, £coRI; K, Kpnl, C, C M ; X, Xbal *, junction of SEB genetic element and common chromosomal DNA as suggested by Johns and Khan (1988)  142  Figure 37.  Southern hybridization analysis of TSST-1* S. aureus isolates with a 700 bp EcoRI SEB genetic element probe. Genomic D N A was digested with Clal and hybridized with a 700 bp EcoBl fragment of pSK155 (see Figure 36). RN450, 8527 and 8530 are TSST-17SEB' control isolates. Numbers at left indicate the positions of the digoxigenin-labelled DNA molecular size standards (kb). The arrows indicate the common sized Clal fragment that hybridized to the 297 bp HincU-BaniHl tst probe in Figure 26. Isolates 7294 and 7616 did not have a common Clal fragment that hybridized to both the tst and the 700 bp EcoRI probe.  143  Figure 38. Southern hybridization analysis of TSST-TS. aureus isolates with a 1 1. kb i//'miin-£coRI SEB genetic element probe. Genomic DNA was digested with Clal and hybridized with a 1.1. kb HindHl-EcoRI fragment of pSK155 (see Figure 36). RN450, 8527 and 8530 are TSST-17SEB" control isolates. Numbers at left indicate the positions of the digoxigeninlabelled DNA molecular size standards (kb). The arrows indicate the common sized Clal fragment that hybridized to the 297 bp HincllBamYH tst probe in Figure 26 and the 700 bp EcoRl probe in Figure 37. Isolates 7294 and 7616 did not have a common Clal fragment that hybridized to these probes.  144  element probes different in size from the fragment that hybridized with the tst probe (Figures 26, 37, and 38), perhaps due to RFLPs in the region upstream of seb in these two isolates. To further examine the observation that tst and the SEB genetic element sequences are located close to each other, Hindlll digested genomic DNA of the TSST-1 isolates were +  hybridized with the 297 bp HinU-BamBIttf-specificprobe (Figure 27) and the 1.1 kb Hindni-EcoKI SEB genetic element probe (Figure 39). The specificities of the SEB genetic element probes were further assessed by including three SEB" isolates (7660, 8531, 8297) and three non-5, aureus isolates (5. epidermidis ATCC 14990, 5. haemofyticus ATCC 29970, B. subtilis ATCC 6633) as negative controls. There was no homology between the 1.1 kb Hindni-EcdRI SEB genetic element probe and the three non-5, aureus isolates. As seen in earlier results in which Clal digested SEB' isolates had weak hybridization signals with the 1.1 kb Hindlll-EcoRl probe (RN450, 8527 and 8530 in Figure 38), the three SEB" isolates 7660, 8531 and 8297 also had a faint hybridization signal to the probe (Figure 39), again perhaps due to the presence of an incomplete or related SEB genetic element. More importantly, as can be observed in Figures 27 and 40, similiar sized Hindm restriction fragments of RN7043, FRI1169, and 7140 hybridized to both the tst and the 1.1 kb MndUIEcoBI SEB genetic element probes. Isolates RN3984, 7051, and 7616 did not show this result likely because of incomplete digestion of chromosomal DNA, since a subsequent experiment did confirm that the tst and the 1.1 kb Hindm-EcoSl probes hybridized to a Hindm fragment of a similiar size in isolates RN3984 and 7051 (not shown). The fact that hybridization data with Clal and Hindlll localized tst and SEB genetic element DNA to similiar-sized restriction fragments provides further evidence that the TSST-1 gene is located  145  M <> I  M U  <d  <u  H M "H G) 0'«i  6  U  rd  ^J* co vo . CO ^ H  © oo Q ) X * ( 0 C A O H O H U > O H O r > •H m N CM  rn «  r«  §  ?  ^  H  °  *°  ^  W  <M N ^ •rl CO  Figure 39. Southern hybridization analysis of TSST-1 isolates digested with M/zdIII and hybridized with a 1.1 kb tfrndlll-fcoRI fragment of pSK155. S. aureus 8528 and 0507 served as the SEB-producing positive control isolates. S. aureus 7660, 8531, 8297 and S. epidermidis, S. haemotyticus, and B. subrilis served as the non-SEB-producing negative control isolates. Arrows indicate the position of the common M/idlll fragment that also hybridized to the 297 bp HincTl-BamKl tst probe in Figure 26. The higher migrating fragments observed in isolates RN3984 and 7616 are the result of incomplete digestion of chromosomal DNA. The DNA of 7051 was degraded. Numbers at left are the positions of the digoxigenin-labelled DNA molecular size standards (kb). +  146  near SEB genetic element sequences in these isolates. 6.2.3.  Transduction of RN7043 [tst::T<f] into S. aureus 7690  One possible mechanism by which TSST-1 and SEB co-production could be mutually excluded in S. aureus is if the mobile TSST-1 genetic element inserts into the SEB genetic element to cause deletion or interruption of SEB structural gene sequences. This would explain the partial seb sequences and the close proximity of the tst and SEB genetic element sequences. In an attempt to determine if this could occur, the tst::T<? cassette of RN7043 (described in Materials and Methods) was transduced into strain 7690 as described in Materials and Methods. 7690 was a putative SEBVTSST-1" clinical isolate of S. aureus previously characterized by immunoassay by Lee et al. (1992). Selection of transductants for resistance to tetracycline assured that the TSST-1 genetic element of RN7043 had been successfully introduced into 7690. During the subsequent screening of T<f transductants of 7690 for loss of SEB production by immunoassay, it was found that the wild-type 7690 did not produce SEB, despite its original designation as a SEB-producer. Lack of SEB production in 7690 was subsequently verified by immunoblotting (as described in Materials and Methods), and absence of seb was verified by performing PCR analysis with the SEB1/SEB2 and the MLD-l/MLD-2 primers (see Figure 32) as described in Materials and Methods. In light of these results, it was no longer possible to examine whether the TSST-1 genetic element, when introduced into a SEB-producing strain, could mediate deletion of seb coding sequences to result in loss of SEB production. This experiment did provide both corroborating and enlightening results however. Successful transduction of the TSST-1 genetic element into 7690 was verified by hybridization of Clal digested genomic DNA of  147  five Tc transductants with the 297 bp Hincll-Bamlil tst probe (Figure 40; Lanes 1-5). Two r  well characterized TSST-1" reference isolates (RN450, 8528) served as negative controls, and as expected, did not hybridize with the tst probe, whereas the TSST-1 control isolate +  FRI1169 did. In addition, lack of tst in 7690 was verified. Hybridization of the Tc isolates with the 700 bp EcoKL probe derived from SEB genetic r  element sequences (see Figure 41) corroborated previous observations that tst was located near SEB genetic element sequences in TSST-1 isolates, since transduction of the tsfc-Tc* +  cassette of RN7043 into 7690 resulted in the transfer of 700 bp EcoKL SEB genetic element sequences on a fragment of the same size as the Clal fragment that hybridized to the tst probe (indicated by the arrow in Figure 41; compare with Figure 40). Interestingly, the genomic DNA of all strains, including wild-type 7690, appeared to contain 700 bp EcoKLrelated sequences, despite the fact that this isolate is SEB". Pre-existing 700 bp EcoRI related sequences were also observed in the nonenterotoxigenic strain RN450, and is consistent with a previous result in this study (Figure 37). Other SEB" isolates that hybridized to the 700 bp EcoKL probe included 8527 and 8530 (Figure 37). As mentioned previously, it is believed that the 700 bp EcoKL sequences in these SEB" isolates may represent an incomplete or related genetic element (Johns and Khan, 1988). Finally, hybridization of the tst-Tcf transductants of 7690 with the 1.1 kb HindUl-EcoKL SEB genetic element probe provided yet further evidence that SEB genetic element sequences are located close to tst since these sequences were also transduced along with the tst: :T<f cassette into 7690. The 1.1 kb Hindlll-EcoKL fragment hybridized to a fragment of the same size as the Clal fragment that hybridized to tst and 700 bp EcoKL sequences in the Tc*  148  O) (0  5 E°N O o)  AC U) (O  u- oo rt.  1 2 3 4 5  Southern hybridization analysis of 7<? transductan.ts of S. aureus 7690 with a 297 bp Hincll-BamHl K/-specific gene. The tst::Tc cassette of RN7043 was transduced into 7690 as described in Materials and Methods. Five Tc? transductants were digested with Clal and hybridized with the tst gene probe (Lane 1-5). FRI1169 served as the TSST-1 control isolate. RN450, 8528 and 7690 were the TSST-1' control isolates. Numbers at left indicate the postions of the digoxigenin-labelled DNA molecular size markers (kb). r  +  149  0>  o <o J2 £ oo o  5 O) 5 ?s » -  CM  1  2  3  4  5  DC U. CO N  23 6.6  *  Figure 41.  Southern hybridization analysis of T f transductants of S. aureus 7690 with a 700 bp EcoKL fragment of pSK155 (see Figure 36). Five Tc transductants were digested with Clal and hybridized with the 700 bp EcoRI probe. The arrow indicates the common-sized Clal fragment that also hybridized with the 297 bp Hincll-BamKL tst gene probe in Figure 40. Numbers at left indicate the positions of the digoxigenin-labelled DNA molecular size standards (kb). 1  150  « O  9  (D  2 ; 5 = Z  00  O  w C U)  en (O  ocu-coh.  I •  12  345  6  Figure 42.  Southern hybridization analysis of To* transductants of S. aureus 7690 with a 1.1 kb Hwdlll-EcoRI fragment of pSK155 (see Figure 36). Five Tc transductants were digested with Clal and hybridized with the 1.1 kb HindlTl-EeoW probe. The Clal fragment that hybridized to this probe appeared to be the same fragment that also hybridized with the 297 bp HincTL-BamHI tst probe (Figure 40) and the 700 bp EcoRI probe (Figure 41). Numbers at left indicate the positions of the digoxigenin-labelled DNA molecular size standards (kb). r  151  transductants (Figure 42). Although wild-type 7690, RN450, 8527 and 8530 contained sequences that were homologous to the 700 bp EcoRI SEB genetic element probe (Figures 37 and 41), these isolates do not contain the 150 Taql fragment of seb. These SEB genetic element-related sequences in these isolates could represent the presence of a related or incomplete SEB genetic element. The fact that extremely weak hybridization signals were evident upon analysis of these strains with the 1.1 kb Mndin-iitoRI probe (Figures 38 and 42) provides further support for the supposition that these SEB genetic element-related sequences do not , represent a complete copy of the SEB genetic element. 6.2.4.  Analysis of S E B isolates with the 700 bp EcoRI fragment derived from +  sequences located within the SEB genetic element Two Clal fragments that hybridized to the 700 bp EcoSI SEB genetic element probe have been observed in 7 of the 8 TSST-1 isolates (Figure 37) and the S E B reference isolates +  +  0507 (Figure 37) and 8528 (Figure 41). Although homologous sequences to the SEB genetic element have been observed in TSST-17SEB" isolates such as RN450, 8527, 8530 and 7690 (Figures 37, 38, 41 and 42), the duplication was not. Since only two SEB isolates were +  examined with the 700 bp EcoRI probe, Southern hybridization analysis was performed on several additional S E B isolates with the 700 bp EcoRI probe to determine if multiple Clal +  fragments that hybridized to this probe are as common in SEB isolates as in TSST-1 +  +  isolates. As observed in Figure 43, 8 of 8 SEB isolates in addition to the reference isolate +  0507 revealed two Clal fragments having homology to the probe. Duplication of SEB genetic element sequences among SEB S. aureus isolates has been observed previously +  152  u M U n  a  o N^cNOoo\t7»nr^Hr>CN  21  5.1  —~m <  2.0  Figure 43.  Southern hybridization analysis of SEB isolates with a 700 bp EcoRI SEB genetic element probe. Genomic DNA was digested with Clal and hybridized with the 700 bp EcoRI fragment of pSK155. RN450 and 8527 are SEB" control isolates. Numbers at left indicate the postions of the digoxigenin-labelled DNA molecular size standards (kb). +  153  (Johns and Khan, 1988). Consistent with previous results (Figures 37 and 41), the TSST-1" /SEB" isolate RN450 and 8527 revealed only one Clal fragment with homology to the 700 bp EcoRI probe. 6.2.5.  Analysis of S E B isolates with st?6-specific probes +  To identify which Clal fragment in the SEB isolates that hybridized to the 700 bp EcoRI +  probe in Figure 43 also contained seb, the Clal digested genomic DNA of the SEB isolates +  were hybridized with the 150 bp Taql seb probe. As observed in Figure 44, 4 of the 9 SEB  +  isolates had seb located on the Clal fragment of — 8.5 kb in size (7372, 7513, 7371, 7429) whereas 5 of the 9 SEB isolates examined had both Clal fragments exhibiting homology to +  the 150 bp Taql probe (0507,7407,7694,8395,7836). This indicated that two copies of seb may be present in these isolates. The observation that the smaller Clal fragment (~3.7 or 4.1 kb) in these isolates had a fainter hybridization signal to the 150 bp Taql probe than the larger Clal fragment ( — 6.1 kb) may indicate that these sequences contain only a part of the 150 bp Taql fragment. Based on the results in Figures 43 and 44, a depiction of the region surrounding seb, in addition to the molecular sizes of the Clal fragments that hybridized to the probes can be proposed as shown in Figure 45. The location of the second Clal fragment in the chromosomes of these isolates with respect to the SEB genetic element is not known. It could be possible that the second Clal fragment is located within the 26.8 kb region associated with seb. In addition, it should be noted that two 150 bp Taql homologous sequences could be present in the SEB isolates 7372, 7513, 7371 and 7429 (Figure-44 and +  Figure 45), but the sequences may be located on the same Clal fragment that contains the  154  u  9  M u  s  o N j t N n o o o o n r ^ H r ^ t N  21  5.1  2.0  _ •  Figure 44.  Southern hybridization analysis of SEB isolates with a 150 bp Taql seb specific gene probe. Genomic DNA was digested with Clal and hybridized with the 150 bp Taql fragment of pSK155. RN450, 8527 and 8530 are SEB' control isolates. Numbers at left indicate the positions of the digoxigenin-labelled DNA molecular size standards (kb). +  155  A 5.5 kb  8  C  C  E  5  E  k b  K  seb  CX  • 4 700 bp  700 bp  1.8 kb 150 bp  AND B  f  3.7 or 4.1 kb  6.1 kb  1  700 bp 160 bp  700 bp  18 kb 150 bp  Figure 45. Illustration of the chromosomal region surrounding seb in S. aureus isolates. ' A ' depicts the region surrounding seb in isolates 7372, 7513, 7371 and 7429 (see Figures 43 and 44). B ' depicts the region surrounding seb in isolates 0507, 7407, 7694, 8395 and 7836 (see Figures 43 and 44). The SEB genetic element (thick line) with the restriction sites are shown as described by Johns and Khan (1988). The probes that were found to hybridize to each restriction fragment are shown below the SEB genetic element, and the sizes of the corresponding C M fragments that hybridized to these probes are indicated above the SEB genetic element. The hatched box represents seb sequences. C, C M ; E, £coRI, K, Kpnl, X, Xbal  156  complete structural gene and thus only one fragment hybridizing to the 150 bp Taql probe would be observed. Analysis of 32 SEB isolates with a 1.8 kb Kpnl-Clal probe (see Figure 36) that contained +  SEB structural gene sequences in addition to flanking sequences gave complete concordance between the size of the Clal fragment that hybridized to the 1.8 kb Kpnl-Clal probe and the 150 bp Taql probe in each of the 8 SEB isolates in Figure 44 as would be expected. +  However, in contrast to the results observed with the 150 bp Taql probe (Figure 44), multiple Clal fragments hybridizing to the 1.8 kb Kpnl-Clal probe were not observed in isolates 0507, 7407, 7694, 8395 and 7836. Results for 0507, as shown in Figure 46, are representative for these isolates. This observation supports the previous contention that the smaller Clal fragment (3.7 or 4.1 kb) in these isolates may contain only partial seb sequences, since if two complete SEB coding sequences were present, the 1.8 kb Kpnl-Clal probe should hybridize to both, yielding two Clal fragments upon Southern blotting. The observation that the 150 bp Taql probe hybridized to two Clal fragments indicates that the smaller Clal fragment in these isolates contained only partial SEB coding sequences. As was the case with TSST-1 isolates, these observations suggest that SEB isolates appear to +  +  contain partial SEB coding sequences. 6.2.6.  Analysis of S E B isolates with a probe derived from sequences that extend past +  the junction of the SEB genetic element downstream of seb The results in section 6.2.5 indicate that TSST-1 and S E B strains have a number of +  +  DNA similarities based upon Southern hybridization patterns with SEB genetic element probes. First, in both types of strains, tst, seb and SEB genetic element sequences appear to  157  /  Figure 46.  Southern hybridization analysis of the SEB isolate 0507 with a 1.8 kb Kpnl-Oal seb gene probe (see Figure 36). 0507 was digested with Clal and hybridized with a 1.8 kb Kpril-Clal fragment derived from pSK155. RN450 is the SEB" control isolate. Numbers at left indicate the postions of the digoxigenin-labelled DNA molecular size standards (kb). +  158  be located close to one another in the chromosome. Second, TSST-1 and SEB isolates +  +  have multiple Clal fragments that hybridize to the 700 bp EcoKL SEB genetic element probe, and both appear to have partial SEB coding sequences. TSST-17SEB" isolates do not contain multiple Clal fragments with homology to the 700 bp EcoKL probe, or 150 bp Taql sequences. One possibility to explain these observations is that the TSST-1 and SEB structural genes are on similiar or closely related genetic elements. Alternatively, the TSST1 genetic element may have a preferred site of insertion in the SEB genetic element. If this were true, DNA similiarities upon hybridization with SEB genetic probes would be expected. One important difference in the characteristics of genetic elements carrying TSST-1 and SEB is their reported mobility. The TSST-1 genetic element is mobile (Chu et al., 1988) whereas the SEB genetic element has been found in only one chromosomal location (Johns and Khan, 1988). To further support either of the two hypotheses, I searched evidence to suggest that the SEB genetic element is mobile. The DNA of 9 SEB isolates analysed in Figure 44 were digested with Kpnl and +  hybridized with the 1.4 kb Xbal-HindlH SEB genetic element junction probe (see Figure 36). Kpnl cuts within the SEB genetic element and to the right past the end of the suggested junction of the element (see Figure 36; Johns and Khan, 1988). If the SEB genetic element is located in more than one chromosomal location, Kpnl fragments of different sizes would be expected to hybridize to the 1.4 kb Xbal-Hin&lll probe. As observed in Figure 47, three Kpnl fragments of different sizes were found to hybridize to the probe in the 9 SEB isolates. The observed heterogeneity in Kpnl fragments indicate that the SEB genetic +  element is present either in a region of frequent RFLPs, or in different chromosomal  159  u  Q M U  id o  Figure 47.  co  Southern hybridization analysis of SEB 5. aureus isolates with a 1.4 kb Xbal-Hindm SEB junction element probe (see Figure 36). Genomic DNA was digested with Kpnl and hybridized with a 1.4 kb Xbal-HindlU fragment derived from pSK155. RN450, 8527, 8530 and RN3984 are SEB" control isolates. The high molecular weight fragment that hybridized with the probe in isolate 7694 is presumed to be due to incomplete digestion of DNA. Numbers at left indicate postions of digoxigenin-labelled DNA molecular size standards (kb). +  160  locations. 6.3. Discussion Lack of TSST-1 and SEB co-production in S. aureus is an interesting phenomenon. Other infrequently observed phenotypic combinations that have been identified in S. aureus include hemolysin and TSST-1 production (Chow et ah, 1983) and staphylokinase and B-lysin production (Winkler et ah, 1965). The molecular mechanisms mediating these have been fully characterized (Coleman et ah, 1986; O'Reilly et ah, 1990). Studies addressing the mechanisms that underly the mutually exclusive production of TSST-1 and SEB however have been limited due to uncertainties surrounding the nature of the genetic elements that carry both toxin determinants. In spite of these uncertainties, a number of observations have been made in this study that allow one to both rule out and extend certain possibilties. First, one putative explanation for the lack of TSST-1 and SEB co-production in S. aureus is that they may be competing for the same transcriptional activator. This is reasonable based on the fact that both TSST-1 and SEB are under control of the agr. The agr is a locus that is known to control the expression of a number of exoprotein genes in S. aureus (Kornblum et ah, 1990). However construction of a strain producing both TSST-1 and SEB argues against this possibility. It has been noted that TSS-associated strains of S. aureus are phenotypically distinct from other S. aureus strains. For example, TSS-associated strains of S. aureus are more proteolytic than non-TSS-associated S. aureus strains (Barbour, 1981; Todd et ah, 1984). This observation has been used to suggest that the TSST-1 genetic element could encode a protease that regulates the SEB product. The ability to create a TSST-1 / S E B strain does +  +  161  not specifically rule out this hypothesis since the m-containing plasmid pJWl used in the construct did not contain the entire TSST-1 genetic element. pJWl contained only approximately 1.6 kb of the -4-7 kb sequences associated with the TSST-1 genetic element. To determine if protease regulation is involved, a plasmid containing the entire TSST-1 genetic element could be introduced into RN4220 containing SEB-producing pMLD6876-l, and screened for TSST-1 and SEB co-production. The plasmid pRN6100, which contains the original cloned TSST-1 gene on a 10.6 kb chromosomal fragment (Kreiswirth et ah, 1983) would be ideal for this purpose. The ability to create a TSST-1 / S E B S. aureus strain +  +  using a plasmid containing the entire TSST-1 genetic element would help to determine if posttranslational modification of SEB can occur through the expression of a protease. Southern hybridization analysis of TSST-1 isolates revealed that tst, seb, and SEB +  genetic element probes hybridized to the same size Clal digested DNA fragments in the majority of isolates. These observations indicate that both TSST-1 and SEB-related genetic element sequences may be located close to one another in the chromosome. In addition, a second restriction enzyme, Hindm, localized tst and the 1.1 kb HindlU-EcoBI fragment of the SEB genetic element to a common-sized fragment. Moreover, transduction of the TSST1 genetic element from the TSST-1 strain RN7043 into a TSST-17SEB' strain of S. aureus +  (7690) resulted in the transfer of SEB genetic element sequences on the same Clal fragment carrying the TSST-1 gene, as evident by a common fragment hybridizing to tst, the 700 bp EcoBI and 1.1 kb Hindlll-EcdKL probes in all transductants. Collectively, these observations strongly support the contention that the TSST-1 structural gene is located in close proximity to SEB genetic element in these isolates. On the other hand, SEB-producing S. aureus  162  isolates did not hybridize to the ttf-probe, or to two probes derived from upstream and downstream sequences that flanked the TSST-1 gene. This observation in conjunction with the fact that the TSST-1 genetic element is mobile, could suggest that the TSST-1 genetic element may be interfering with the expression of SEB through insertion into the SEB genetic element. There are several possibilities to explain how insertion of the TSST-1 genetic element could interfere with SEB expression. The TSST-1 genetic element could insert into the promoter or structural region of SEB, thereby disrupting transcription. Alternatively, once inserted into the SEB genetic element, the TSST-1 genetic determinant could possibly disrupt gene function due to exertion of strong polar effects on SEB. Another possibility is that the TSST-1 determinant could carry a transcriptional terminator that does not allow the transcription of SEB. Although all of these mechanisms are possible and none can be discounted, it may also be possible that the TSST-1 genetic element could cause deletion of SEB coding sequences upon insertion. This possibility appears to be more likely than the previously described mechanisms since PCR results did not show the expected SEB amplified products when primers that targeted various regions of seb were used as probes in the TSST1 isolates, indicating that the entire SEB coding sequence is not present. To determine if +  this mechanism mediates the lack of TSST-1 and SEB co-production in S. aureus, introduction of the TSST-1 genetic element into a SEB clinical strain, perhaps via +  transduction, followed by the observed loss of SEB production and verification of deletion of SEB coding sequences would need to be demonstrated. Although this appears to be a straight-forward approach, especially with the availability of the strain RN7043 (tsf.'.Tcf),  163  two limitations of this approach are noteworthy. First is the observed difficulty of introducing foreign D N A into a clinical S. aureus isolate. The efficiences of this are extremely low, and thereby explains why most genetic manipulations of S. aureus are done in the restriction-deficient mutant RN4220. Second, to perform this experiment, one would ideally require a recA SEB mutant of S. aureus in order to minimize the incorporation of +  the TSST-1 genetic element into the host chromosome by homologous recombination. This would undoubtably enhance the probability of identifying desired transductants that have arisen via independent movement of the TSST-1 genetic element, but unfortunately this type of strain is not currently available. Another possibilty to explain the presence of SEB genetic element sequences in TSST-1  +  isolates is that perhaps the TSST-1 and SEB structural genes are carried by similiar genetic elements. If the two toxin genes were carried by similiar genetic elements, it is likely that these elements would share a preferred site of insertion. Potentially, the insertion of a genetic element carrying either the TSST-1 or SEB structural gene would likely exclude the insertion of the other. Indirect support for a common genetic element is supplied by the observation that TSST-1 and SEB isolates share several DNA similiarities in relation to +  +  hybridization results with SEB genetic element probes. It was found that in contrast to TSST-17SEB" isolates, both TSST-1 and S E B isolates had multiple Clal fragments that +  +  hybridized to the 700 bp £<:oRI SEB genetic element probe. In addition, both TSST-1 and +  SEB isolates appeared to contain partial seb coding sequences. Partial seb coding sequences +  were not found in any of the TSST-17SEB" isolates examined. If the theory that both genes are carried by similiar genetic elements were true however, one would have expected to  164  observe SEB-producing isolates to hybridize to probes derived from the TSST-1 genetic element. This was not observed in these studies (Figures 33 and 34). In addition the results suggest that the SEB genetic element is mobile like the TSST-1 genetic element. Although a preferred location in the chromosome may be possible if both toxins genes are carried by similiar genetic elements, if the elements are mobile, alternative sites of insertion would be expected. .Thus theoretically, TSST-1 /SEB isolates should be observed. The observation +  +  that S E B isolates do not hybridize to TSST-1 genetic element probes, in addition to the +  observation that the SEB and TSST-1 genetic elements are mobile, argue against the hypothesis that mutual exclusivity of TSST-1 and SEB production in S. aureus is mediated by the carriage of both toxin genes on similiar genetic elements that each have the same preferred site of insertion in the chromosome. A duplication of 700 bp EcoSl sequences was also found with a SEC /SEB/TSST-1" +  isolate 8530 (ATCC 19095)(see Figure 37). As with co-production of TSST-1 and SEB, coproduction of SEC1 with SEB has also been found to be rare, although co-production of TSST-1 and SEC1 is relatively frequent (TSST-1 appears to influence the site of insertion of the SEC1 gene) (Bohach et al., 1989). Although the nature of the SEC determinant still remains unknown, it is possible that lack of SEB and SEC co-production among S. aureus isolates is mediated by a mechanism similiar to the one that mediates the mutual exclusivity of TSST-1 and SEB co-production. In summary, a number of putative mechanisms that may underly the mutual exclusivity of TSST-1 and SEB co-production in S. aureus have been explored. Although none of these possibilities can be definitively ruled out, some appear to be more likely than others based on  165  the collective preliminary observations made by genetic analyses. The pattern observed upon Southern hybridization analysis of TSST-1 and SEB strains with the probes used in this +  +  study lead to at least two possible mechanisms that may explain the mutual exclusivity of TSST-1 and SEB co-production in S. aureus as illustrated in Figure 48. The TSST-1 genetic element could have a preferred site of insertion in the SEB genetic element. Upon insertion, the TSST-1 genetic element may interfere with the expression of SEB in as yet some unindentified manner (Figure 48A). Alternatively, the TSST-1 and SEB genetic elements may each have the same preferred site of insertion in the S. aureus chromosome. Insertion of one genetic element carrying one of the toxin genes would inhibit the insertion of the other genetic element carrying the alternate toxin gene. It may be further speculated that the shared point of insertion may be linked to a partial seb coding sequence associated with a DNA sequence that shares homology with the SEB genetic element (Figure 48B). Whether the partial seb coding sequence that has homology to the 150 bp Taql probe represents a seblike gene, a remnant of a seb sequence, or an ancestral sequence that was involved in creating the seb gene found today, is only speculative at present. Future studies in this area are clearly needed in order to determine which, if any, of these possible mechanisms may be mediating the mutual exclusivity of TSST-1 and SEB co-production in S. aureus.  166  167  Chapter 7.  7.1.  Summary of Scientific Contributions and Future Avenues of Research  Summary of Results and Scientific Contributions  The main objective of this research was to study the interaction of the staphylococcal superantigen TSST-1 with SEA, and SEB. In addition to belonging to the superantigen family of proteins, these three toxins are carried by genetic elements, of which only the element carrying sea has been fully characterized. Specific interactions may be occurring with regards to the expression of these toxins in S. aureus since certain of these toxin combinations are more prevalent than others. TSST-J and SEA is a toxin combination frequently found co-produced among menstrual TSS isolates, whereas in contrast, TSST-1 and SEB are rarely (if ever) found co-produced. These two observations provided the framework for the research thesis in that questions pertaining to the functional and molecular aspects of these toxin combinations were formulated and examined.  SUMMARY  1.  Demonstration that TSST-1 plays an important role in the pathogenesis of menstrual TSS  ->  Although TSST-1 has been implicated in the pathogenesis of TSS, a causal role for this toxin in MTSS had not yet been verified. A causative role for TSST-1 in MTSS was 168  definitively established in this study by introducing a pair of isogenic mutants in TSST-1 production into a tampon-associated vaginal infection model of TSS in the NZW rabbit. The TSST-1 strain induced more lethality and fever than the TSST-1' isogenic derivative. The +  in vivo studies were corroborated by the in vitro data which showed that the culture filtrate from the TSST-1 strain induced more T cell proliferation and TNFa secretion from rabbit +  splenocytes than the culture filtrate from the TSST-1" strain. These effects are believed to play an important role in the pathogenesis of superantigen-mediated shock. Although the amount of SEA produced by the TSST-1 isogenic mutants differed, SEA was subsequently shown not to enhance the virulence of the TSST-1 strain. This strengthens the premise that +  TSST-1 plays an important role in the pathogenesis of MTSS and suggests that the observed difference in virulence between RN3984 and RN7043 was not a result of decreased production of SEA by RN7043. 2.  Effect of TSST-1 and SEA Superantigen Co-Production in the Pathogenesis of Menstrual TSS In light of the frequent co-expression of TSST-1 and SEA among MTSS isolates, it was  hypothesized that co-production of two superantigens by a S. aureus strain may increase various superantigen-associated activities which may lead to more virulent strain. Consistent with this hypothesis was the observation that the combination of TSST-1 and SEA significantly enhanced T cell proliferation and TNFa secretion from human PBMC as compared to TSST-1 or SEA alone. To test if this toxin combination may have an effect in vivo, TSST-1 /SEA and TSST-1 /SEA" strains of S. aureus were introduced into two +  +  +  animal models of lethal shock. The TSST-1 /SEA strain did not cause more lethality or +  +  169  morbidity than the TSST-1 /SEA" strain. This suggested that co-production of SEA with +  TSST-1 did not increase the virulence of this strain. This was consistent with the observation that the TSST-1 /SEA culture filtrate did not induce a higher amount of TNFa +  +  secretion from rabbit splenocytes as compared to the TSST-1 /SEA' culture filtrate. Although +  the culture filtrate of the TSST-1 /SEA strain induced more T cell proliferation than the +  +  culture filtrate of the TSST-1 /SEA" strain, the effect was concentration dependent. Since +  the pathogenicity of the strains was not different in two test models, these results suggest that an enhancement in T cell proliferation does not occur in vivo, perhaps since the amount of TSST-1 produced by the strain was already at a high level. This was the first time to my knowledge that a toxin gene of 5". aureus has been inactivated by the method of plasmid integration. The concern that this type of construct may be unstable in vivo was alleviated since the stability of the SEA" phenotype of INT#1 after growth in BALB/c mice and NZW rabbits was verified. This was also the first reported study that used isogenic mutants in animal studies to determine the effect various toxin combinations produced by S. aureus may play in pathogenesis. 3.  Genetic Analyses Focused on the Mutual Exclusivity of TSST-1 and SEB CoProduction in S. aureus  i)  Since TSST-1 and SEB are not co-produced, it was of interest to determine if the TSST-1 and SEB structural genes could be found within the same isolate. Upon Southern hybridization with an intragenic seb probe, 8 of 8 TSST-1 isolates +  hybridized to the probe, even though SEB was not produced. This indicated that TSST-1 isolates do contain .^-homologous sequences. +  170  ii)  Although TSST-1 isolates hybridized to a 150 bp Taql seb probe, a complete copy of +  seb was not present in these isolates since PCR analysis using primers that targeted various regions of the SEB nucleotide sequence failed to amplify the expected PCR products. iii)  The TSST-1 structural gene, seb homologous sequences, and SEB genetic element sequences were all localized to the same sized DNA restriction fragment in the majority of TSST-1 isolates analysed. In addition, transduction of tst from a TSST+  1 strain of S. aureus into a TSST-r strain resulted in the co-transfer of SEB genetic +  element sequences. This suggested that tst was located close to SEB genetic element sequences in these isolates. iv)  SEB producing isolates of S. aureus did not contain the TSST-1 structural gene, or sequences that immediately flanked tst within the TSST-1 genetic element.  v)  TSST-1 and SEB isolates were found to contain multiple Clal fragments that +  +  hybridized to the 700 bp EcoRI SEB genetic element probe. TSST-1 and S E B +  +  isolates also contained partial seb coding sequences. This is in contrast to TSST-1  -  /SEB' isolates which only had one Clal fragment that hybridized to the 700 bp EcoRI probe, and no .seo-homologous sequences. vi)  Results of Southern hybridization analysis suggested that the SEB genetic element may be found in different chromosomal locations.  vii)  In light of the initial observation that although TSST-1 isolates appeared to contain +  reo-homologous sequences, SEB is not co-produced, it was proposed that expression of both toxins may be regulated by the competition for a common transcriptional  171  activator. Demonstration that a TSST-1 /SEB strain of S. aureus could be +  +  genetically constructed argues against this possibility. Based on the collective observations made above, possible mechanisms mediating the lack of TSST-1 and SEB co-production in S. aureus have been proposed. First is the possibility that the TSST-1 genetic element has a preferred site of insertion in the SEB genetic element. Upon insertion, the TSST-1 genetic element may interfere with the expression of SEB in as yet some unidentified manner. Alternatively, the genetic elements carrying either tst or seb may have a common point of insertion in the S. aureus chromosome that may be linked to a partial seb coding sequence. Insertion of one genetic element carrying one of the toxin genes may inhibit the insertion of the other genetic element carrying the alternate toxin gene. This was the first study to address the mechanisms that may mediate the mutual exclusivity of TSST-1 and SEB co-production in S. aureus. As a result of these studies, this research has enabled further characterization of the region surrounding seb in S. aureus (see Figure 45). 7.2.  Future Avenues of Research  The observed frequent production of SEA in conjunction with TSST-1 among MTSS isolates remains enigmatic. Although the results presented in this study appeared to suggest that this toxin combination does not enhance the virulence of this organism, it can still not be ruled out. I believe that it is realistic to assume that the presence of SEA, or as yet some unidentified protein, is responsible for causing this particular clone to be dominant among menstrual-associated TSS infections. Since it has been found that co-production of SEA with TSST-1 did not cause more lethality or morbidity in vivo, the constructed isogenic mutants  172  can next be used to assess whether TSST-1 and SEA co-production has an effect on other parameters important in establishing virulence, such as colonization or growth. Alternatively, the question of whether the expression of staphylokinase or the repression of B-lysin may play a role in increasing the pathogenic potential of this specific clone should be examined. A concentration dependent effect of TSST-1 and SEA-induced T cell proliferation on rabbit splenocytes was demonstrated. One follow-up to this study could be to identify the TSST-1 and SEA binding sites on rabbit splenocytes. If they bind to overlapping epitopes on the same receptor as they do on human PBMC, the usefulness of the rabbit model for dissecting out the role of staphylococcal superantigens in the pathogenesis of TSS would be supported. This would also support the prediction that a concentration effect in T cell proliferation exists upon co-incubation of TSST-1 and SEA with human PBMC. If the concentration of TSST-1 and SEA produced in vivo is a critical factor that may dictate whether or not an additive response in various effects will occur, it would follow that the variability in the concentrations of toxin produced by various TSST-1 /SEA strains may +  +  be an important issue. The production of SEA by a strain of S. aureus that produces a minimally effective concentration of TSST-1 (for example 1 ng/ml), would be more of a benefit to the organism than production of SEA in a strain that produced a higher amount of TSST-1 (for example 100 ng/ml). RN3984 appears to produce a relatively higher amount of TSST-1 in comparison to other TSST-1 strains (Rosten et ah, 1987). It would be +  interesting to determine if the same results in the animal studies would occur if a TSST1VSEA mutant was constructed from a TSST-1 /SEA strain of S. aureus that produced a -  +  +  173  lower amount of TSST-1. Despite many attempts by various investigators, the nature of the genetic elements that carry tst and seb still defy full characterization. Sequence analysis of the genetic elements that carry both tst and seb would help to determine the nature of these elements, and would be an important direction for future research. Although studies that address the role of superantigens in disease are important, it is my hope that the underlying mechanisms that favour the acquisition and expression of certain superantigen combinations are not overlooked. The aspects governing how and why toxins associated with mobile genetic elements are introduced, maintained and excluded in a particular S. aureus strain in my opinion, parallel the importance of studies that address the functional consequences of these toxins. Through these studies I believe I have begun to address some of these important issues, and hopefully enough interest will be generated that such studies will continue in the future.  174  LITERATURE CITED Akatsuka, H . , K . Imanishi, K . Inada, H . Yamashita, M . Yoshida, and T. Uchiyama. 1994. Production of tumor necrosis factors by human T cells stimulated by a superantigen, toxic shock syndrome toxin-1. Clin. Exp. Immunol. 96:422-426. Altboum, Z . , I. Hertman, and S. Sarid. 1985. Penicillinase plasmid-linked genetic determinants for enterotoxins B and C, production in Staphylococcus aureus. Infect. Immun. 47:514-521. Altemeier, W . A . , S. A . Lewis, P. M . Schlievert, M . S. Bergdoll, H . S. Bjornson, J . L . Staneck, and B . A . Crass. 1982. Staphylococcus aureus associated with Toxic Shock Syndrome. Ann. Int. Med. 96:978-982. Arko, R . , J . , J . K . Rasheed, C . Y . Broome, F . W . Chandler, and A . L . Paris. 1984. A rabbit model of Toxic Shock Syndrome: clinicopathological features. J. Infect. 8:205-211. Augustin, J . , and F . Gotz. 1990. Transformation of Staphylococcus epidermidis and other staphylococcal species with plasmid DNA by electroporation. FEMS Microbiol. Lett. 66:203-208. Barbour, A . G . 1981. Vaginal isolates of Staphylococcus aureus associated with Toxic Shock Syndrome. Infect. Immun. 33:442-449. Bartlett, P., A . L . Reingold, D . R . Graham, B . B . Dan, D . S. Selinger, G . W . Tank, and K . A . Wichterman. 1982. Toxic Shock Syndrome associated with surgical wound infections. JAMA. 247:1448-1450. Beharka, A . A . , J . W . Armstrong, J . J . Iandolo, and S. K . Chapes. 1994. Binding and activation of major histocompatibility complex class II-deficient macrophages by staphylococcal exotoxins. Infect. Immun. 62:3907-3915. Beisel, W . R . 1972. Pathophysiology of staphylococcal enterotoxin, type B, (SEB) toxemia after intravenous administration to monkeys. Toxicon. 10:433-440. Bergdoll, M . S., B . A . Crass, R . F . Reiser, R . N . Robbins, and J . P. Davis. 1981. A new staphylococcal enterotoxin, enterotoxin F, associated with toxic-shock-syndrome Staphylococcus aureus isolates. Lancet. 1:1017-1021.  175  Bergdoll, M . S., B . A . Crass, R . F . Reiser, R . N . Robbins, A . C . - M . Lee, P. J . Chesney, J . P. Davis, J . M . Vergeront, and P. J . Wand. 1982. An enterotoxin-like protein in Staphylococcus aureus strains from patients with toxic-shock syndrome. Ann. Intern. Med. 96:969-971. Bergdoll, M . S. 1983. Enterotoxins, p. 559-598. In Staphylococci and staphylococcal infections. C. S. F. Easmon and C. Adlam (ed.). Academic Press Inc. Ltd., London. Bergdoll, M . S., and P. M . Schlievert. 1984. Toxic shock syndrome toxin. Lancet. 2:691. Berkley, S. F . , A . W . Hightower, C . V . Broome, and A . L . Reingold. 1987. The relationship of tampon characteristics to menstrual toxic shock syndrome. JAMA. 258:917920. Best, G. K., D . F . Scott, J . M . Kling, M . R . Thompson, L . E . Adinolfi, and P. F . Bonventre. 1988. Protection of rabbits in an infection model of toxic shock syndrome (TSS) by a TSS Toxin-1-specific monoclonal antibody. Infect. Immun. 56:998-999. Betley, M . J . , S. Lofdahl, B . N . Kreiswirth, M . S. Bergdoll, and R . P. Novick. 1984. Staphylococcal enterotoxin A gene is associated with a variable genetic element. Proc. Natl. Acad. Sci. USA. 81:5179-5183. Betley, M . J . , and J. J . Mekalanos. 1988. Nucleotide sequence of the Type A Staphylococcal enterotoxin gene. J. Bacteriol. 170:34-41. Betley, M . J . , and J. J . Mekalanos. 1985. Staphylococcal enterotoxin A is encoded by phage. Science. 229:S185-S187. Beutler, B , J . Mahoney, N . Le Trang, P. Pekala, and A . Cerami. 1985. Purification of cachectin, a lipoporotein lipase-suppressing hormone secreted by endotoxin-induced RAW 264.7 cells. J. Exp. Med. 161:984-995. Blomster-Hautamaa, D . A . , B . N . Kreiswirth, J . S. Kornblum, R . P. Novick, and P. M . Schlievert. 1986a. The nucleotide and partial amino acid sequence of Toxic Shock Syndrome Toxin-1. J. Biol. Chem. 261:15783-15786. Blomster-Hautamaa, D . A . , B . N . Kreiswirth, R . P. Novick,,and P. M . Schlievert. 1986b. Resolution of highly purified toxic-shock syndrome toxin 1 into two distinct proteins by isoelectric focusing. Biochemistry. 25:554-59. Bohach, G. A . , D . J . Fast, R . D . Nelson, and P. M . Schlievert. 1990. Staphylococcal and Streptococcal pyrogenic toxins involved in Toxic Shock Syndrome and related illness. Crit. Rev. Microbiol. 17:251-272. 176  Bohach, G . A . , B . N . Kreiswirth, R . P. Novick, and P. M . Schlievert. 1989. Analysis of TSS isolates producing staphylococcal enterotoxins B and CI with use of southern hybridization and immunologic assays. Rev. Infect. Dis. ll(Suppl):S75-81. Bonventre, P. F . , H . Heeg, C . Cullen, and C.-J. Lian. 1993. Toxicity of recombinant toxic shock syndrome toxin 1 and mutant toxins produced by Staphylococcus aureus in a rabbit infection model of toxic shock syndrome. Infect. Immun. 61:793-799. Bonventre, P.F., H . Heeg, C . K . Edwards, and C. M . Cullen. 1995. A mutation at histidine residue 135 of toxic shock syndrome toxin yields an immunogenic protein with minimal toxicity. Infect. Immun. 63:509-515. Bonventre, P. F . , C . Linnemann, L . S. Weckbach, J . L . Staneck, C . R . Buncher, E . Vigdorth, H . Ritz, D. Archer, and B. Smith B . 1984. Antibody responses to toxic-shocksyndrome (TSS) toxin by patients with TSS and by healthy staphylococcal carriers. J. Infect. Dis. 150:662-666. Bonventre, P. F . , M . R . Thompson, L . E . Adinolfi, Z . A . Gillis, and J . Parsonnet. 1988. Neutralization of toxic shock syndrome toxin-1 by monoclonal antibodies in vitro and in vivo. Infect. Immun. 56:135-141. Bonventre, P. F . , L . Weckbach, G . Harth, and C. Haidaris. 1989. Distribution and expression of Toxic Shock Syndrome Toxin 1 among Staphylococcus aureus isolates of Toxic Shock Syndrome and Non-Toxic Shock Syndrome Origin. Revs. Infect. Dis. ll(Suppl):S90S95. Bonventre, P. F . , L . Weckbach, J . Staneck, P. M . Schlievert, and M . Thompson. 1983. Production of staphylococcal enterotoxin F and pyrogenic exotoxin C by Staphylococcus aureus isolates from toxic shock syndrome-associated sources. Infect. Immun. 40:1023-1029. Borst, D. W . , and M . J . Betley. 1993. Mutations in the promoter spacer region and early transcribed region increase expression of Staphylococcal enterotoxin A. Infect. Immun. 61:5421-5425. Borst, D. W . , and M . J . Betley. 1994. Phage-associated differences in Staphylococcal enterotoxin A gene (sea) expression correlate with sea allele class. Infect. Immun. 62:113118. Broome, C . V . 1989. Epidemiology of toxic shock syndrome in the United States:Overview. Revs. Infect. Dis. ll(Suppl):S14-S21. Bullock, W . O., J . M . Fernandez, and J . M . Short. 1987. XL-1 Blue: A high efficiency plasmid transforming recA Escherichia coli strain with beta-galactosidase selection. Biotech. 5:376-379. 177  Calvano, S. £ . , F . W . Quimby, A . C . Antonacci, R . F . Reiser, M . S. Bergdoll, and P. Dineen. 1984. Analysis of the mitogenic effects of toxic shock toxin on human peripheral blood mononuclear cells in vitro. Clin. Immunol. Immunopathol. 33:99-110. Carlsson, R . , H . Fischer, and H . O. Sjogren. 1988. Binding of staphylococcal enterotoxin A to accessory cells is a requirement for its ability to activate human T cells. J. Immunol. 140:2484-2488. Chang, A . , J . M . Musser, and A . W . Chow. 1991. A single clone of S. aureus producing both TSST-1 and SEA causes the majority of menstrual toxic shock syndrome. Clin. Res. 39:36A. Chesney, P. J . , M . S. Bergdoll, J . P. Davis, and J . M . Vergeront. 1984. The Disease spectrum, epidemiology, and etiology of Toxic Shock Syndrome. Ann. Rev. Microbiol. 38:315-338. Chesney, P. J . 1989. Clinical aspects and spectrum of illness of Toxic Shock Syndrome. Revs. Infect. Dis. ll(Suppl):Sl-S7. Cheung, A . L . , J . M . Koomey, C . A . Butler, S. J . Projan, and V . A . Flschetti. 1992. Regulation of exoprotein expression in Staphylococcus aureus by a locus (sar) distinct from agr. Proc. Natl. Acad. Sci. USA. 89:6462-6466. /  Cheung, A . L . , and S. J . Projan. 1994. Cloning and sequencing of sar A. of Staphylococcus aureus, a gene required for the expression of agr. J. Bacteriol. 176:4168-4172. Chintagumpala, M . M . , J . A . Mollick, and R . R . Rich. 1991. Staphylococcal toxins bind to different sites on HLA-DR. J. Immunol. 147:3876-3881. Choi, Y . , B . Kotzin, L . Herron, J . Callahan, P. Marrack, and J . Kappler. 1989. Interaction of Staphylococcus aureus toxin "superantigens" with human T cells. Proc. Natl. Acad. Sci. USA. 86:8941-8945. • (. Choi, Y . , J . A . Lafferty, J . R . Clements, J . K . Todd, E . W . Gelfand, J . Kappler, P. Marrack, and B . L . Kotzin. 1990. Selective expansion of T cells expressing VB2 in Toxic Shock Syndrome. J. Exp. Med. 172:981-984. Chouaib, S., D . Branellec, and W . A . Buurman. 1991. More insights into the complex physiology of TNF. Immunol. Today. 12:141-143. Chow, A . ,W., and K . H . Bartlett. 1982. Role of vaginal Enterobacteriaceae in Toxic Shock Syndrome. Clin. Invest. Med. 5:3IB.  178  Chow, A . W . , and K . H . Bartlett. 1989. Sequential assessment of vaginal microflora in healthy women randomly assigned to tampon or napkin use. Revs. Infect. Dis. ll(Suppl):S68-S74. Chow, A . W . , M . J . Gribble, and K . H . Bartlett. 1983. Characterization of the hemolytic activity of Staphylococcus aureus strains associated with toxic shock syndrome. J. Clin. Microbiol. 17:524-528. Chow, A . W . , R . K . L . Percival-Smith, K . H . Bartlett, and A . M . Macfarlane. 1983. Vaginal colonization with toxigenic Staphylococcus aureus in healthy women. Clin. Invest. Med. 6:61. Chow, A . W . 1993. Toxic shock syndrome, p. 1010-1011. In The 5 Minute Clinical Consult. Lea and Febiger, London. Chow, A . W . 1995. Toxic shock syndrome. Med. N . Amer. p. 140-148. (  Chow, A . W . 1983. Toxic shock syndrome. Med. N . Amer. 6:535-542. Chu, M . C , B . N . Kreiswirth, P. A . Pattee, R . P. Novick, M . E . Melish, and J . J . James. 1988. Association of Toxic Shock Toxin-1 determinant with a heterologous insertion at multiple loci in the Staphylococcus aureus chromosome. Infect. Immun. 56:2702-2708. Chu, M . C , M . E . Melish, and J . F . James. 1985. Tryptophan auxotypy associated with Staphylococcus aureus that produce Toxic-Shock-Syndrome Toxin. J. Infect. Dis. 151:11571158. Coleman, D . C , D . J . Sullivan, R . J . Russell, J . P. Arbuthnott, B . F . Carey, and H . M . Pomeroy. 1989. Staphylococcus aureus bacteriophages mediating simultaneous lysogenic conversion of B-lysin, staphylokinase and entertotoxin A: Molecular mechanisms of triple conversion. J. Gen. Microbiol. 135:1679-1697. Compagnone-Post, P., U . Malyankar, and S. A . Khan. 1991. Role of host factors in the regulation of the enterotoxin B gene. J. Bacterid. 173:1827-1830. Cone, L . A . , D . R . Woodard, P. M . Schlievert, and G . S. Tomory. 1987. Clinical and bacteriologic observations of a toxic shock-like syndrome due to Streptococcus pyogenes. N . Engl. J. Med. 317:146-149. Crass, B . A . , and M . S. Bergdoll. 1986b. Involvement of staphylococcal enterotoxins in nonmenstrual toxic shock syndrome. J. Clin. Microbiol. 23:1138-1139. Crass, B . A . , and M . S. Bergdoll. 1986a. Toxin involvement in Toxic Shock Syndrome. J. Infect. Dis. 153:918-926. 179  Cullen, C . M . , L . R . Blanco, P. F . Bonventre, and £ . Choi. 1995. A toxic shock syndrome toxin 1 mutant that defines a functional site critical for T-cell activation. Infect. Immun. 63:2141-2146. Czop, J . K . , and M . S. Bergdoll. 1974. Staphylococcal enterotoxin synthesis during the exponential, transitional, and stationary growth phases. Infect. Immun. 9:229-235. Davis, J . P., P. J . Chesney, P. J . Wand, and M . LaVenture. 1980. Toxic Shock Syndrome. Epidemiologic features, recurrence, risk factors, and prevention. N . Engl. J. Med. 303:1429-1435. de Azavedo, J . C . S. 1989. Animal models for toxic shock syndrome: Overview. Revs. Infect. Dis. ll(Suppl):S205-S209. de Azavedo, J . C . S., A . Drurnm, C . Jupin, M . Parant, J . £ . Alouf, and J . P. Arbuthnott. 1988. Induction of tumor necrosis factor by staphylococcal toxic shock toxin 1. FEMS Microbiol. Immunol. 47:69-74. de Azavedo, J . C . S., and J . P. Arbuthnott. 1984. Toxicity of staphylococcal toxic shock syndrome toxin 1 in rabbits. Infect. Immun. 46:314-317. de Azavedo, J . C . S., T. J . Foster, P. J . Hartigan, J . P. Arbuthnott, M . O'Reilly, B . N . Kreiswirth, and R . P. Novick. 1985. Expression of the cloned toxic shock syndrome toxin 1 gene (tst) in vivo with a rabbit uterine model. Infect. Immun. 50:304-309. Dinarello, C . A . , J . G . Cannon, S. M . Wolff, H . A . Bernheim, B . Beutler, A . Cerami, I. S. Figari, A . Palladino, and J . V . O'Connor. 1986. Tumor necrosis factor (cachectin) is an endogeneous pyrogen and induces production of interleukin 1. J. Exp. Med. 163:14331450. Dinarello, C . A . , S. Okusawa, and J . A . Gelfand. 1989. Interleukin-1 induces a shock-like state in rabbits: Synergism with tumor necrosis factor and the effect of ibuprofen, p. 203215. In Perspectives in Shock Research: Metabolism, Immunology, Mediators, and Models. Alan R. Liss, Inc. Dohlsten, M . , G . Hedlund, S. Segren, P. A . Lando, T. Herrmann, A . P. Kelly, and T. Kalland. 1991. Human major histocompatibility complex class II-negative colon carcinoma cells present staphylococcal superantigens to cytotoxic T lymphocytes: Evidence for a novel enterotoxin receptor. Eur. J. Immunol. 21:1229-1233. Dohlsten, M . , P. A . Lando, G . Hedlund, J . Trowsdale, and T. Kalland. 1990. Targeting of human cytotoxic T lymphocytes to MHC class II-expressing cells by staphylococcal enterotoxins. Immunology. 71:96-100.  180  Dornbusch, K . , and H . O. Hallander. 1973. Transduction of penicillinase production and methicillin resistance-enterotoxin B production in strains of Staphylococcus aureus. J. Gen. Microbiol. 76:1-11. Drake, C. G . , and B . L . Kotzin. 1992. Superantigens: Biology, immunology, and potential role in disease. J. Clin. Immunol. 12:149-162. Drumm, A . , J . C . S. de Azavedo, and J . P. Arbuthnott. 1989. Damaging effect of toxic shock syndrome toxin 1 on chick embryo cells in vitro. Revs. Infect. Dis. ll(Suppl):S275S281. Dyer, D . W . , and J . J . Iandolo. 1981. Plasmid-Chromosomal transition of genes important in Staphylococcal enterotoxin B expression. Infect. Immun. 33:450-458. Endres, S, R. Ghorbani, G . Lonnemann, J . W . M . van der Meer, and C . A . Dinarello. 1988. Measurement of immunoreactive interleukin-lB from human mononuclear cells: Optimization of recovery, intrasubject consistency, and comparison with interleukin-la and tumor necrosis factor. Clin. Immunol. Immunopathol. 49:424-438. Ewan, P., S. Tessier, and E . Ofori. 1989. Toxin testing and phage typing of Staphylococcus aureus strains isolated from suspected cases of TSS in Canada. Rev. Infect. Dis. ll(Suppl):S327. Fast, D . J . P. M . Schlievert, and R. D . Nelson. 1988. Nonpurulent response to toxic shock syndrome toxin 1-producing Staphylococcus aureus. Relationship to toxin-stimulated production of tumor necrosis factor. J. Immunol. 140:949-953. Fischer, H . , M . Dohlsten, U . Andersson, G . Hedlund, P. Ericsson, J . Hansson, and H . O. Sjogren. 1990. Production of TNFa and TNFfi by staphylococcal enterotoxin A activated human T cells. J. Immunol. 144:4663-4669. Fischer H . , M . Dohlsten, M . Lindvall, H.-O. Sjogren, and R. Carlsson. 1989. Binding of staphylococcal enterotoxin A to HLA-DR on B cell lines. J. Immunol. 142:3151-3157. Fleischer, B . , and H . Schrezenmeier. 1988. T cell stimulation by staphylococcal enterotoxins. Clonally variable response and requirement for MHC class II molecules on accessory or target cells. J. Exp. Med. 167:1697-1707. Foster, T. J . , M . O'Reilly, P. Phonimdaeng, J . Cooney, A . H . Patel, and A . J . Bramley. 1990. Genetic studies of virulence factors of Staphylococcus aureus. Properties of coagulase and delta-toxin and the role of a-toxin, B-toxin, and protein A in the pathogenesis of S. aureus infections, p. 403-417. In R. P. Novick (ed.), Molecular Biology of the Staphylococci. V C H Publishers, New York.  181  Frankel, W . N . , C . Rudy, J . M . Coffin, and B . T. Huber. 1991. Linkage of Mis genes to endogeneous mammary tumour viruses of inbred mice. Nature. 349:526-528. Fraser, J . D . 1989. High-affinity binding of staphylococcal enterotoxins A and B to H L A DR. Nature. 339:221-223. Friedrich, £ . G . , and K . A . Siegesmund. 1980. Tampon-associated vaginal ulcerations. Obs. Gynecol. 55:149-156. Garbe, P. L . , R . J . Arko, A . L . Reingold, L . M . Graves, P. S. Hayes, A . W . Hightower, F . W . Chandler, and C . V . Broome. 1985. Staphylococcus aureus isolates from patients with nonmenstrual Toxic Shock Syndrome. JAMA. 253:2538-2542. Gaskill, M . £ . , and S. A . Khan. 1988. Regulation of the Enterotoxin B gene in Staphylococcus aureus. J. Biol. Chem. 263:6276-6280. Gaventa, S., A . L . Reingold, A . W . Hightower, C . V . Broome, B . Schwartz, C . Hoppe, J . Harwell, L . K . Lefkowitz, S. Makintubee, D . R . Cundiff, and S. Sitze, and the Toxic Shock Syndrome Study Group. 1986. Active surveillance for Toxic Shock Syndrome in the United States. Revs. Infect. Dis. ll(Suppl):S28-S34. Giraudo, A . T., C . G . Raspanti, and A . Calzolari. 1994. Characterization of a Tn551mutant of Staphylococcus aureus defective in the production of several exoproteins. Can. J. Microbiol. 40:677-681. Gjorloff, A . , H . Fischer, G . Hedlund, J . Hansson, J . S. Kenney, A . C . Allison, H . - O . Sjogren, and M . Dohlsten. 1991. Induction of interleukin-1 in human monocytes by the superantigen staphylococcal enterotoxin A requires the participation of T cells. Cell. Immunol. 137:61-71. v  Grossman, D . , R . G . Cook, J . T. Sparrow, J . A . Mollick, and R . R . Rich. 1990. Dissociation of the stimulatory activities of staphylococcal enterotoxins for T cells and monocytes. J. Exp. Med. 172:1831-1841. Grossman, D . , J . G . Lamphear, J . A . Mollick, M . J . Betley, and R . R . Rich. 1992. Dual roles for Class U major histocompatibility complex molecules in staphylococcal enterotoxin-induced cytokine production and in vivo toxicity. Infect. Immun. 60:5190-5196. Guinan, M . £ . , B . B . Dan, R . J . Guidotti, A . L . Reingold, G . P. Schmid, £ . J . Bettoli, J . G . Lossick, K . N . Shands, M . A . Kramer, N . T. Hargrett, R . L . Anderson, and C . V . Broome. 1982. Vaginal colonization with Staphylococcus aureus in healthy women. Ann. Int. Med. 96:944-947.  182  Haima, P., D . van Sinderen, H . Schotting, S. Bron, and G . Venema. 1990. Development of a B-galactosidase a-complementation system for molecular cloning in Bacillus subtilis. Gene. 86:63-69. Herman, A . , G . Croteau, R.-P. Sekaly, J . Kappler, and P. Marrack. 1990. HLA-DR alleles differ in their ability to present staphylococcal enterotoxins to T cells. J. Exp. Med. 172:709-717. Herman, A . , N . Labrecque, J . Thibodeau, P. Marrack, J . W . Kappler, and R.-P. Sekaly. 1991. Identification of the staphylococcal enterotoxin A superantigen binding site in the B l domain of the human histocompatibility antigen HLA-DR. Proc. Natl. Acad. Sci. USA. 88:9954-9958. Herrmann, T., R . S. Accolla, and H . R . MacDonald. 1989. Different staphylococcal enterotoxins bind preferentially to distinct major histocompatibility complex class II isotypes. Eur. J. Immunol. 19:2171-2174. Herrmann, T., P. Romero, S. Sartoris, F . Paiola, R . S. Accolla, J . L . Maryanski, and H . R . MacDonald. 1991. Staphylococcal enterotoxin-dependent lysis of MHC Class TJ negative target cells by cytolytic T lymphocytes. J. Immunol. 146:2504-2512. Hinshaw, L . B . , T. E . Emerson, F . B . Taylor, A . C . K . Chang, M . Duerr, G . T. Peer, D . J . Flournoy, G . L . White, S. D . Kosanke, C . K . Murray, R . X u , R . B . Passey, and M . A . Fournel. 1992. Lethal Staphylococcus aureus-induced shock in primates: Prevention of death with anti-TNF antibody. J. Trauma. 33:568-573. Horinouchi, S., and B . Weisblum. 1982. Nucleotide sequence and functional map of pE194, a plasmid that specifies induceble resistance to macrolide, lincosamide, and streptogramin type B antibiotics. J. Bacterid. 150:804-814. Howard, B . J . , and W . E . Kloos. 1987. Staphylococci, p. 231-244. In B. J. Howard, J. Klaas, S. J. Rubin, A . S. Weissfeld, and R. C. Tilton (ed.), Clinical and Pathogenic Microbiology. The C.V. Mosby Company, Washington, D. C. j  Iandolo, J . J . , and D . W . Dyer. 1981. The Staphylococcal enterotoxins - a genetic reveiw. J. Food. Saf. 3:249-264. Iandolo, J . J . 1980. The genetics of staphylococcal toxins and virulence factors, p. 399-425. In The Bacteria. Vol XI. Academic Press Inc., London. flcejima, T., S. Okusawa, J . W . M . van der Meer, and C . A . Dinarello. 1988. Induction by toxic-shock-syndrome toxin-1 of a circulating tumor necrosis factor-like substance in rabbits and of immunoreactive tumor necrosis factor and interleukin-1 from human mononuclear cells. J. Infect. Dis. 158:1017-1025. 183  Imanishi, K . , H . Igarashi, and T. TJchiyama. 1992. Relative abilities of distinct isotypes of human major histocompatibility complex class II molecules to bind streptococcal pyrogenic exotoxin types A and B. Infect. Immun. 60:5025-5029. Jacobson, J . A . , £ . M . Kasworm, and J . Murdock. 1984. Nasal carriage of Toxic Shock Syndrome (TSS) associated Staphylococcus aureus in hopitalized patients. Prog. Abstr. Intersci. Conf. Antimicrob. Agents. Chemother, abstr. 470. Jeffes, £ . W . B . , £ . K . Ininns, K . L . Schmitz, R . S. Yamamoto, C . A . Dett, and G . A . Granger. 1989. The presence of antibodies to lymphotoxin and tumor necrosis factor in normal serum. Arth. Rheum. 32:1148-1152. Johns, M . B . , Jr., and S. A . Khan. 1988. Staphylococcal enterotoxin B gene is associated with a discrete genetic element. J. Bacteriol. 170:4033-4039. Johnson L . P., M . A . Tomai M . A . , and P. M . Schlievert. 1986. Bacteriophage involvement in Group A Streptococcal pyrogenic exotoxin A production. J. Bacteriol. 166:623-627. Johnson, W . M . , S. K . Tyler, E . P. Ewan, F . E . Ashton, D . R . Pollard, and K . R . Rozee. 1991. Detection of genes for enterotoxins, exfoliative toxins, and toxic shock syndrome toxin 1 in Staphylococcus aureus by the polymerase chain reaction. J. Clin. Microbiol. 29:426-430. Jones, C . L . , and S. A . Khan. 1986. Nucleotide sequence of the enterotoxin B gene from Staphylococcus aureus. J. Bacteriol. 166:29-33. Kain, K . C , M . Schulzer, and A . W . Chow. 1993. Clinical spectrum of nonmenstrual Toxic Shock Syndrome (TSS):Comparison with menstrual TSS by Multivariate Discriminant Analyses. Clin. Infect. Dis. 16:100-106. Kalland, T., G . Hedlund, M . Dohlsten, and P. A . Lando. 1991. Staphylococcal enterotoxin-dependent cell-mediated cytotoxicity, In Current Topics in Microbiology and Immunology. 174:81-91. Kapral, F . A . 1982. Epidermal toxin production by Staphylococcus aureus strains from patients with toxic shock syndrome. Ann. Int. Med. 96:972-974. Karp, D . R . , C . L . Teletski, P. Scholl, R . Geha, and E . O. Long. 1990. The a l domain of the HLA-DR molecule is essential for high-affinity binding of the toxic shock syndrome toxin-1. Nature. 346:474-476. Kasatiya, S. S., and J . N . Baldwin. 1967. Nature of the determinant of tetracycline resistance in Staphylococcus aureus. Can. J. Microbiol. 13:1079-1086. 184  Kass, E . H . 1989. Magnesium and the pathogenesis of toxic shock syndrome. Revs. Infect. Dis. ll(Suppl):S167-S175. Kass, E . H . , and J . Parsonnet. 1987. On the pathogenesis of toxic shock syndrome. Revs. Infect. Dis. 9(Supp5):S482-S489. Kawabe, Y . , and A . T. Ochi. 1991. Programmed cell death and extrathymic reduction of VB8 C D 4 T cells in mice tolerant to Staphylococcus aureus enterotoxin B. Nature. 349:245-248. +  +  Khan, S. A . , and R . P. Novick. 1982. Structural analysis of plasmid pSN2 in Staphylococcus aureus: No involvement in Enterotoxin B production. J. Bacteriol. 149:642649. Kornblum, J . , B . N . Kreiswirth, S. J . Projan, J . Ross, R . P. Novick. 1990: Agr: A polycistronic locus regulating exoprotein synthesis in Staphylococcus aureus, p. 373-402. In R. P. Novick (ed.), Molecular Biology of the Staphylococci, New York:VCH. Krakauer, T. 1995. Differential inhibitory effects of interleukin-10, interleukin-4, and dexamethasone on staphylococcal enterotoxin-induced cytokine production and T cell activation. J. Leuk. Biol. 57:474-476. Kreiswirth, B . N . , G . R . Kravitz, P. M . Schlievert, and R . P. Novick. 1986. Nosocomial transmission of a strain of Staphylococcus aureus causing toxic shock syndrome. Ann. Int. Med. 105:704-707. Kreiswirth, B . N . , J . P. Handley, P. M . Schlievert, and R . P. Novick. 1987. Cloning and expression of streptococcal pyrogenic exotoxin A and staphylococcal toxic shock syndrome toxin-1 in Bacillus subtilis. Mol. Gen. Genet. 208:84-87. (. , Kreiswirth, B . N . , S. Lbfdahl, M . J . Betley, M . O'Reilly, P. M . Schlievert, M . S. Bergdoll, and R . P. Novick. 1983. The toxic shock syndrome exotoxin structural gene is not detectably transmitted by a prophage. Nature. 305:709-712. Kreiswirth, B . N . , M . O'Reilly, and R . P. Novick. 1984. Genetic characterization and cloning of the Toxic Shock Exotoxin. Surv. Synth. Pathol. Res. 3:73-82. Kreiswirth, B . N . , S. J . Projan, P. M . Schlievert, and R . P. Novick. 1989. Toxic shock syndrome toxin 1 is encoded by a variable genetic element. Revs. Infect. Dis. ll(Suppl):S83-S89. Kum, W . W . S., K . B . Laupland, R . H . See, and A . W . Chow. 1993. Improved purification and biologic activities of staphylococcal toxic shock syndrome toxin 1. J. Clin. Microbiol. 31:2654-2660. 185  Kushnaryov, V . M . , M . S. Bergdoll, H . S. MacDonald, J . Velinga, and R . Reiser. 1984a. Study of staphylococcal toxic shock syndrome toxin in human epithelial cell culture. J. Infect. Dis. 150:535-545. Kushnaryov, V . M . , H . S. MacDonald, R . Reiser, and M . S. Bergdoll. 1989. Reaction of toxic shock syndrome toxin 1 with endothelium of human umbilical cord vein. Revs. Infect. Dis. ll(Suppl):S282-S288. Kushnaryov, V . M . , H . S. MacDonald, R . Reiser, and M . S. Bergdoll. 1984b. Staphylococcal toxic shock toxin specifically binds to cultured human epithelial cells and is rapidly internalized. Infect. Immun. 45:566-571. Lagoo, A . S., S. Lagoo-Deenadayalan, H . - M . Lorenz, J . Byrne, W . H . Barber, and K . J . Hardy. 1994. IL-2, IL-4, and IFN-5 gene expression versus secretion in superantigenactivated T cells. J. Immunol. 152:1641-1652. Langford, M . P., G . J . Stanton, and H . M . Johnson. 1978. Biological effects of staphylococcal enterotoxin A on human peripheral lymphocytes. Infect. Immun. 22:62-68. Lee, A . C , B . A . Crass, and M . S. Bergdoll. 1987. Investigation by syringe method of effect of tampons on production in vitro of Toxic Shock Syndrome Toxin 1 by Staphylococcus aureus. J. Clin. Microbiol. 25:87-90. Lee, P. K . , J . R . Deringer, B . N . Kreiswirth, R . P. Novick, and P. M . Schlievert. 1991. Fluid replacement protection of rabbits challenged subcutaneously with toxic shock syndrome toxins. Infect. Immun. 59:879-884. Lee, V . T. P., A . H . Chang, and A . W . Chow. 1992. Detection of staphylococcal enterotoxin B among TSS- and non-TSS-associated Staphylococcus aureus isolates. J. Infect. Dis. 166:911-915. Linnemann, C . C , J . L . Staneck, S. Hornstein, T. P. Barden, J . L . Rauh, P. F . Bonventre, C . R . Buncher, and A . Beiting. The epidemology of genital colonization with Staphylocccus aureus. Ann. Int. Med. 96:940-944. L i u , C . T., R . K . DeLauter, M . J . Griffin, and R . T. Faulkner. 1978. Effect of staphylococcal enterotoxin B on cardiorenal functions in rhesus macaques. Am. J. Vet. Res. 39:279-286. MacDonald, H . R., S. Baschieri, and R . K . Lees. 1991. Clonal expansion precedes anergy and death of VB8 peripheral T cells responding to staphylococcal enterotoxin B in vivo. Eur. J. Immunol. 21:1963-1966. +  186  Madsen, M . , H . E . Johnson. 1979. A methodological study of E-rosette formation using AET-treated red blood cells. J. Immunol. Methods. 27:61-74. Mallonee, D . H . , B . A . Glatz, and P. A . Pattee. 1982. Chromosomal mapping of a gene affecting Enterotoxin A production in Staphylococcus aureus. Appl. Env. Microbiol. 43:397402. Marples, R . R . , and A . A . Wieneke. 1993. Enterotoxins and Toxic Shock Syndrome Toxin-1 in non-enteric staphylococcal disease. Epidemiol. Infect. 110:477-488. Marrack, P., and J . Kappler. 1990. The staphylococcal enterotoxins and their relatives. Science. 248:705-711. Marrack, P., M . Blackman, E . Kushnir, and J . Kappler. 1990. The toxicity of Staphylococcal enterotoxin B in mice is mediated by T cells. J. Exp. Med. 171:455-464. Martin, R . , V . Buttram, P. Besch, J . J . Kirkland, and G . P. Petty. 1982. Nasal and vaginal Staphylococcus aureus in young women:Quantitative studies. Ann. Int. Med. 96:951953. Matsuyama, S, Y . Koide, T. O. Yoshida. 1993. H L A class II molecule-mediated signal transduction mechanism responsible for the expression of interleukin-16 and tumor necrosis factor-a genes induced by a staphylococcal superantigen. Eur. J. Immunol. 23:3194-3202. McCollister, B . D . , B . N . Kreiswirth, R . P. Novick, and P. M . Schlievert. 1990. Production of toxic shock syndrome-like illness in rabbits by Staphylococcus aureus D4508: Association with enterotoxin A. Infect. Immun. 58:2067-2070. McDevitt, D . , P. Vaudaux, and T. J . Foster. 1992. Genetic evidence that bound coagulase of Staphylococcus aureus is not clumping factor. Infect. Immun. 60:1514-1523. Melish, M . E . , K . Frogner, S. Hirata, and M . S. Murata. 1987. Prophylaxis and therapy in experimental toxic shock syndrome (TSS). Clin. Res. 35:220A. Melish, M . E . , S. Murata, C . Fukunaga, K . Frogner, and C . McKissick. 1989. Vaginal tampon model for toxic shock syndrome. Revs. Infect. Dis. ll(Suppl):S238-S247. Miethke, T., K . Duschek, C . Wahl, K . Heeg, and H . Wagner. 1993. Pathogenesis of the toxic shock syndrome: T cell mediated lethal shock caused by the superantigen TSST-1. Eur. J. Immunol. 23:1494-1500. Miethke, T., C . Wahl, K . Heeg, B . Echtenacher, P. H . Krammer, and H . Wagner. 1992. T cell-mediated lethal shock triggered in mice by the superantigen Staphylococcal enterotoxin B: Critical role of tumor necrosis factor. J. Exp. Med. 175:91-98. 187  Mills, J . T., J . Parsonnet, Y . - C . Tsai, M . Kendrick, R . K . Hickman, and £ . H . Kass. 1985. Control of production of toxic-shock-syndrome toxin-1 (TSST-1) by magnesium ion. J. Infect. Dis. 151:1158-1161. Minor, T. E . , and E . H . Marth. 1971. Staphylococcus aureus and Staphylococcal food intoxications. A review. II. Enterotoxins and epidemiology. J. Milk Food Technol. 35:21-29. Mishell, B . B . , S. M . Shiigi, C . Henry, E . L . Chan, J . North, R . Gallily, M . Slomick, K . Miller, J . Marbrook, D . Parks, and A . H . Good. 1980. Preparation of mouse splenocytes, p. 23. In B. B. Mishell, and S. M . Shiigi (ed.), Selected methods in cellular immunology. W.H. Freeman and Company, San Francisco, CA. Moldawer, L . L . , C . Anderson, J . Gelin, and K . G . Lundholm. 1988. Regulation of food intake and hepatic protein synthesis by recombinant derived cytokines. Am. J. Physiol. 254:G450-G456. Mollick, J . A . , R . G . Cook, and R . R . Rich. 1989. Class II molecules are specific receptors for staphylococcus entertotoxin A. Science. 244:817-820. Morfeldt, E . , L . Janzon, S. Arvidson, and S. Ldfdahl. 1988. Cloning of a chromosomal locus (exp) which regulates the expression of several exoprotein genes in Staphylococcus aureus. Mol. Gen. Genet. 211:435-440. Mortimer, E . A . 1982. Possible mechanisms for vaginal infection with Staphylococcus aureus. Ann. Int. Med. 96:934-936. Mourad, W . , R . Al-Daccak, T. Chatila, and R . S. Geha. 1993. Staphylococcal superantigens as inducers of signal transduction in MHC class n-positive cells. Sem. Immunol. 5:47-55. Miiller-Alouf, II., J . E . Alouf, D . Gerlach, J . H . Ozegowski, C . Fitting, and J . M . Cavaillon. 1994. Comparative study of cytokine release by human peripheral blood mononuclear cells stimulated with Streptococcus pyogenes superantigenic erythrogenic toxins, heat-killed streptococci, andlipopolysaccharide. Infect. Immun. 62:4915-4921. Murphy, E . , S. Phillips, I. Edelman, and R . P. Novick. 1981. Tn554-isolation and characterization of plasmid insertions. Plasmid. 5:292-305. Musser, J . M . , P. M . Schlievert, A . W . Chow, P. Ewan, B . N . Kreiswirth, V . T. Rosdahl, A . S. Naidu, W . Witte, and R . K . Selander. 1990. A single clone of Staphylococcus aureus causes the majority of cases of toxic shock syndrome. Proc. Natl. Acad. Sci. USA. 87:225-229. 188  Nakamura, R . M . , A . Voller, and D . E . Bidwell. 1986. Enzyme immunoassays: heterogeneous and homogeneous systems, p. 27.1-27.20. In D . M . Weir (ed.), Volume 1: Immunochemistry. Blackwell Scientific Publications. Palo Alto, CA. Neill, R . J . , G . R . Fanning, F . Delahoz, R . Wolff, and P. Gemski. 1990. Oligonucleotide probes for detection and differentiation of Staphylococcus aureus strains containing genes for enterotoxins A , B, and C and toxic shock syndrome toxin 1. J. Clin. Microbiol. 28:15141518. Notermans, S., K . J . Huevelman, and K . Wernars. 1988. Synthetic enterotoxin B DNA probes for detection of enterotoxigenic Staphylococcus aureus strains. Appl. Environ. Microbiol. 54:531-533. Notermans, S., W . J . Van Leeuwen, J . Dufrenne, and P. D . Tips. 1983. Serum antibodies to enterotoxins produced by Staphylococcus aureus with special reference to enterotoxin F and toxic shock syndrome. J. Clin. Microbiol. 18:1055-1060. Novick, R . P. 1963. Analysis by transduction of mutations affecting penicillinase formation in Staphylococcus aureus. J. Gen. Microbiol. 33:121-136. Novick, R . P. 1991. Genetic systems in Staphylococci. In J. H . Miller (ed.), Bacterial genetic systems. Methods in Enzymology. 204:587-636. O'Hehrir, R . E . , and J . R . Lamb. 1990. Induction of specific clonal anergy in human T lymphocytes by Staphylococcus aureus enterotoxins. Proc. Natl. Acad. Sci. USA. 87:88848888. Okusawa, S., J . A . Gelfand, T. Ikcjima, R . J . Connolly, and C . A . Dinarello. 1988. Interleukin 1 induces a shock-like state in rabbits. Synergism with tumor necrosis factor and the effect of cyclooxygenase inhibitor. J. Clin. Invest. 81:1162-1172. O'Reilly, M . , B . N . Kreiswirth, and T. J . Foster. 1990. Cryptic a-toxin gene in toxic shock syndrome and septicaemia strains of Staphylococcus aureus. Mol. Microbiol. 4:19471955. Parsonnet, J . , and Z . A . Gillis. 1988. Antibody directed against tumor necrosis factor protects rabbits from lethality induced by toxic shock syndrome. Program Abstr. 1988 Intersci. Conf. Antimicrob. Agents Chemother. (ICAAC), abstr 625. Parsonnet, J . , Z . A . Gillis, and G . B . Pier. 1986. Induction of interleukin-1 by strains of Staphylococcus aureus from patients with nonmenstrual toxic shock syndrome. J. Infect. Dis. 154:55-63.  189  Pattee, P. A . , and B . A . Glatz. 1980. Identification of a chromosomal determinant of enterotoxin A production in Staphylococcus aureus. Appl. Env. Microbiol. 39:186-193. Poindexter, N . J . , and P. M . Schlievert. 1987. Binding of toxic-shock syndrome toxin-1 to human peripheral blood mononuclear cells. J. Infect. Dis. 156:122-129. Poindexter, N , J . , and P. M . Schlievert. 1985. Toxic-shock-syndrome toxin 1-induced proliferation of lymphocytes: Comparison of the mitogenic response of human, murine, and rabbit lymphocytes. J. Infect. Dis. 151:65-72. Price, S. R., S. B . Mizel, and P. H . Pekala. 1986. Regulation of lipoprotein sipases lipase synthesis 3T3-LI adipocyte metabolism by rIL-1. Biochem. Biophys. Acta. 889:374-381. Proctor, R . A . 1992. The use of selective mutagenesis to study the pathogenesis of grampositive bacterial diseases. J. Lab. Clin. Med. 119:5-10. Ranelli, D . M . , C . L . Jones, M . B . Johns, G . J . Mussey, and S. A . Khan. 1985. Molecular cloning of staphylococcal enterotoxin B gene in Escherichia coli and Staphylococcus aureus. Proc. Natl. Acad. Sci. USA. 82:5850-5854. Rasheed, J . K . , R . J . Arko, J . C . Feeley, F . W . Chandler, C . Thornsberry, R . J . Gibson, M . L . Cohen, C . D . Jeffries, and C . V . Broome. 1985. Acquired ability of Staphylococcus aureus to produce toxic shock-associated protein and resulting illness in a rabbit model. Infect. Immun. 47:598-604. Recsei, P., B . Kreiswirth, M . O'Reilly, P. Schlievert, A . Grass, and R . P. Novick. 1986. Regulation of exoprotein gene expression in Staphylococcus aureus by agr. Mol. Gen. Genet. 202:58-61. Reingold, A . L . , C . V . Broome, S. Gaventa, A . W . Hightower, and the Toxic Shock Syndrome Study Group. 1989. Risk factors for menstrual toxic shock syndrome: Results of a multistate case-control study. Revs. Infect. Dis. ll(Suppl):S35-S42. Reingold, A . L . , N . T. Hargrett, B . B . Dan, K . N . Shands, B . Y . Strickland, and C . V . Broome. 1982. Nonmenstrual toxic shock syndrome. Ann. Int. Med. 96:871-874. Reiser, R . F . , R . N . Robbins, G . P. Khoe, and M . S. Bergdoll. 1983. Purification and some physiochemical properties of toxic-shock toxin. Biochemistry. 22:3907-3912. Rellahan, B . L . , L . A . Jones, A . M . Kruisbeek, A . M . Fry, and L . A . Matis. 1990. In vivo induction of anergy in peripheral VJJ8 T cells by staphylococcal enterotoxin B. J. Exp. Med. 172:1091-1100. +  190  Ren, K . , J . D . Barman, V . Pancholi, A . L . Cheung, J . C . Robbins, V . A . Fischetti, and J . B . Zabriskie. 1994. Characterization and biological properties of a new staphylococcal exotoxin. J. Exp. Med. 180:1675-1683. Rosten, P. M . , K . H . Bartlett, and A . W . Chow. 1987. Detection and quantitation of toxic shock syndrome toxin 1 in vitro and in vivo by noncompetitive enzyme-linked immunosorbent assay. J. Clin. Microbiol. 25:327-332. Rosten, P. M . , K . H . Bartlett, and A . W . Chow. 1989. Purification and purity assessment of TSST-1. Rev. Infect. Dis. ll(Suppl):S110-S116. Rust, C . J . J . , F . Verreck, H . Vietor, and F . Koning. 1990. Specific recognition of staphylococcal enterotoxin A by human T cells bearing receptors with the V39 region. Nature. 346:572-574. Sambrook, J . , E . F . Fritsch, and T. Maniatis. 1989. Molecular cloning, a laboratory manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N . Y . Schlech, W . F . , K . N . Shands, A . L . Reingold, B . B . Dan, G . P. Schmid, N . T. Hargrett, A . Hightower, L . A . Herwaldt, M . A . Neill, J . D . Band, and J . V . Bennett. 1982. Risk factors for development of Toxic Shock Syndrome. JAMA. 248:835-839. Schindler, R, J . Mancilla, S. Endres, R . Ghorbani, S. C . Clark, and C . A . Dinarello. 1990. Correlations and interactions in the production of interleukin-6 (IL-6), JL-1, and tumor necrosis factor (TNF) in human mononuclear cells: IL-6 suppresses TL-1 and TNF. Blood. 75:40-47. Schlievert, P . M . 1983. Alteration of immune function by staphylococcal pyrogenic exotoxin type C: Possible role in toxic-shock syndrome. J. Infect. Dis. 147:391-398. Schlievert, P. M . , M . T. Osterholm, J . A . Kelly, and R . D . Nishimura. 1982. Toxin and enzyme characterization of Staphylococcus aureus isolates from patients with and without Toxic Shock Syndrome. Ann. Int. Med. 96:937-940. Schlievert, P. M . 1993. Role of superantigens in human disease. J. Infect. Dis. 167:9971002. Schlievert, P. M . , D . J . Schoettle, and D . W . Watson. 1979. Purification and physiochemical and biological characterization of a staphylococcal pyrogenic exotoxin. Infect. Immun. 23:609-617. Schlievert, P. M . , K . N . Shands, B . B . Dan, G . P. Schmid, and R . D . Nishimura. 1981. Identification and characterization of an exotoxin from Staphylococcus aureus associated with toxic shock syndrome. J. Infect. Dis. 143:509-516. 191  Schlievert, P. M . 1986. Staphylococcal Enterotoxin B and Toxic Shock Syndrome Toxin-1 are significantly associated with non-menstrual TSS. Lancet. 1:1149-1150. Scholl, P. R . , A . Diez, and R . S. Geha. 1986b. Staphylococcal enterotoxin B and toxic shock syndrome toxin-1 bind to distinct sites on HLA-DR and HLA-DQ molecules. J. Immunol. 143:2583-2588. Scholl, P. R . , A . Diez, R . Karr, R . P. Sekaly, and J . Trowsdale, and R. S. Geha. 1990. Effect of isotypes and allelic polymorphism on the binding of staphylococcal exotoxins to MHC class H molecules. J. Immunol. 144:226-230. Scholl, P. R . , A . Diez, W . Mourad, J . Parsonnet, R . S. Geha, and T. Chatila. 1989a. Toxic shock syndrome toxin 1 binds to major histocompatibility complex class U molecules. Proc. Natl. Acad. Sci. USA. 86:4210-4214. Schroeder, C . J . , and P. A . Pattee. 1984. Transduction analysis of transposon Tn551 insertions in the trp-thy region of the Staphylococcus aureus chromosome. J. Bacterid. 157:533-537. Schutzer, S. E . , V . A . Fischetti, and J . B . Zabriskie. 1983. Toxic Shock Syndrome and lysogeny in Staphyloccocus aureus. Science. 220:316-318. Schwartz, B . , S. Gaventa, C . V . Broome, A . L . Reingold, A . W . Hightower, J . A . Periman, P. H . Wolf, and the Toxic Shock Syndrome Study Group. 1989. Nonmenstrual Toxic Shock Syndrome associated with barrier contraceptives:Report of a case-control study. Revs. Infect. Dis. ll(Suppl):S43-S49. Scott, D . F . , J . M . Kling, J . J . Kirkland, and G . K . Best. 1983. Characterization of Staphylococcus aureus isolates from patients with toxic shock syndrome using polyethylene infection chambers in rabbits. Infect. Immun. 39:383-387. See, R . H . , and A . W . Chow. 1989. Microbiology of Toxic Shock Syndrome: Overview. Revs. Infect. Dis. ll(Suppl):S55-S60. See, R . H . , and A . W . Chow. 1992. Staphylococcal toxic shock syndrome toxin 1-induced tumor necrosis factor alpha and interleukin-IB secretion by human peripheral blood monocytes and T lymphocytes is differentially suppressed by protein kinase inhibitors. Infect. Immun. 60:3456-3459. See, R . H . , G . Krystal, and A . W . Chow. 1990. Binding competition of toxic shock syndrome toxin 1 and other staphylococcal exoproteins for receptors on human peripheral blood mononuclear cells. Infect. Immun. 58:2392-2396.  192  See, R. H . , G . Krystal, and A . W . Chow. 1992a. Receptors for toxic shock syndrome toxin-1 and staphylococcal enterotoxin A on human blood monocytes. Can. J. Microbiol. 38:937-944. See, R. H . , G . Krystal, and A . W . Chow. 1992b. TSST-1 and staphylococcal enterotoxins induce distinct patterns of protein phosporylation in human monocytes. Clin. Res. 40:53A. See, R. H . , W . W . S. Kum, A . H . Chang, S. Goh, and A . W . Chow. 1992c. Induction of tumor necrosis factor and interleukin-1 by purified staphylococcal toxic shock syndrome toxin 1 requires the presence of both monocytes and T lymphocytes. Infect. Immun. 60:2612-2618. Sekaly, R. P., G . Croteau, M . Bowman, P. Scholl, S. Burakoff, and R. S. Geha. 1991. The CD4 molecule is not always required for the T cell response to bacterial enterotoxins. J. Exp. Med. 173:367-371. Shafer, W . M . , and J . J . Iandolo. 1979. Genetics of Staphylococcal enterotoxin B in methicillin resistant isolates of Staphylococcus aureus. Infect. Immun. 25:902-911. Shafer, W . M . , and J . J . Iandolo. 1978b. Staphylococcal enterotoxin A: a chromosomal gene product. Appl. Environ. Microbiol. 36:389-391. Shands, K . N . , G . P. Schmid, B . B . Dan, D . Blum, R. J . Guidotti, N . T. Hargrett, R. L . Anderson, D . L . Hill, C . V . Broome, J . D . Band, and D . W . Fraser. 1980. Toxic Shock Syndrome in Menstruating Women. Association with tampon use and Staphylococcus aureus and clinical features in 52 cases. N . Engl. J. Med. 303:1436-1442. Short, J . M . , J . M . Fernandez, J . A . Sorge, and W . D . Huse. 1988. Lambda ZAP: A bacteriphage lambda expression vector with in vivo excision properties. Nucleic Acid Res. 16:7583. Silver, M . A . , and G . L . Simon. 1980. Toxic Shock Syndrome in a male postoperative patient. J. Trauma. 21:650-651. Sloane, R., J . C . S. de Azavedo, J . P. Arbuthnott, P. J . Hartigan, B . Kreiswirth, R. Novick, and T. J . Foster. 1991. A toxic shock syndrome toxin mutant of Staphylococcus aureus isolated by allelic replacement lacks virulence in a rabbit uterine model. FEMS Microbiol. Lett. 78:239-244. Smeltzer, M . S., M . E . Hart, and J . J . Iandolo. 1993. Phenotypic characterization of xpr, a global regulator of extracellular virulence factors in Staphylococcus aureus. Infect. Immun. 61:919-925.  193  Soltis, M . T., J . J . Mekalanos, and M . J . Betley. 1990. Identification of a bacteriophage containing a silent staphylococcal variant enterotoxin gene (sezA ). Infect. Immun. 58:16141619. +  Stolz, S. J . , J . P. Davis, J . M . Vergeront, B . A . Crass, J . Chesney, P. J . Wand, and M . S. Bergdoll. 1985. Development of serum antibody to toxic shock toxin among individuals with toxic shock syndrome in Wisconsin. J. Infect. Dis. 151:883-889. Stratagene. 1992. Epicurian coli competent cells. Instruction manual. Stratagene, La Jolla, CA. Sugiyama, J . , £ . M . McKissic, Jr., M . S. Bergdoll, and B . Heller. 1964. Enhancement of bacterial endotoxin lethality by staphylococcal enterotoxin. J. Infect. Dis. 114:111-118. Swanstrom, M . , and M . H . Adams. 1951. Agar layer method for production of high titer phage stocks. Proc. Soc. Exp. Biol. 78:372-374. Todd, J . , M . Fishaut, F . Kapral, and T. Welch. 1978. Toxic Shock Syndrome associated with phage-group-1 Staphylococci. Lancet i i : 1116-1118. Todd, J . K . , A . Franco-Buff, D . W . Lawellin, and M . L . Vasil. 1985. Distinctive protease activity of toxic shock syndrome-associated Staphylococcus aureus strains, p. 111113. In Proceedings of the 5th International Symposium of Staphylococci and Staphyloccal Infections. Gustav Fischer Verlag, New York. Todd, J . K . , A . Franco-Buff, D . W . Lawellin, and M . L . Vasil. 1984. Phenotypic distinctiveness of Staphylococcus aureus strains associated with Toxic Shock Syndrome. Infect. Immun. 45:339-344. Todd, J . K . , B . H . Todd, A . Franco-Buff, C . M . Smith, and D . W . Lawellin. 1987. Influence of focal growth conditions on the pathogenesis of toxic shock syndrome. J. Infect. Dis. 155:673-681. Trede, N . S., R. S. Geha, and T. Chatila. 1991. Transcriptional activation of IL-1B and tumor necrosis factor-a genes by MHC Class II ligands. J. Immunol. 146:2310-2315. Trede, N . S., T. Morio, P. R. Scholl, R. S. Geha, and T. Chatila. 1994. Early activation events induced by the staphylococcal superantigen toxic shock syndrome toxin-1 in human peripheral blood monocytes. Clin. Immunol. Immunopathol. 70:137-144. Tremaine, M . T., D . K . Brockman, and M . J . Betley. 1993. Staphylococcal enterotoxin A gene (sea) expression is not affected by the accessory gene regulator (agr). Infect. Immun. 61:356-359. \  194  Uchiyama, T., S. Saito, H . Inoko, X . - J . Yan, K . Imanishi, M . Araake, and H . Igarashi. 1990. Relative activities of distinct isotypes of murine and human major histocompatibility complex class U molecules in binding toxic shock syndrome toxin 1 and determination of CD antigens expressed on T cells generated upon stimulation by the toxin. Infect. Immun. 58:3877-3882. van Deuren, M . 1994. Kinetics of tumour necrosis factor-alpha, soluble tumour necrosis factor receptors, interleukin 1-beta and its receptor antagonist during serious infections. Eur. J. Clin. Microbiol. Infect. Dis. 13(Suppl):12-16. Van Miert, A . S. J . P. A . M . , C . T. M . Van Duin, and A . J . H . Schotman. 1984. Comparative observations of fever and associated clinical hematological and blood biochemical changes after intravenous administration of staphylococcal enterotoxins B and F (toxic shock syndrome toxin-1) in goats. Infect. Immun. 46:354-360. Vergeront, J . M . , S. J . Stolz, B . A . Crass, D . B . Nelson, J . P. Davis, and M . S. Bergdoll. 1983. Prevalence of serum antibody to staphylococcal enterotoxin F among Wisconsin residents: implications for Toxic Shock Syndrome. J. Infect. Dis. 148:692-698. Villafane, R . , D . H . Bechhofer, C . S. Narayanan, D . Dubnau. Replication control genes of plasmid pE194. J. Bacteriol. 169:4822-4829. Wagner, G . , L . Bohr, P. Wagner, and L . Norgaard Petersen. 1984. Tampon-induced changes in vaginal oxygen and carbon dioxide tensions. Am. J. Obstet. Gynecol. 148:147150. White, A . , and J . Smith. 1963. Nasal reservoir as the source of extranasal Staphylococci. Antimicrob. Agents Chemother. 3:679-683. White, J . , A . Herman, A . M . Pullen, R . Kubo, J . W . Kappler, and P. Marrack. 1989. The VB-specific superantigen staphylococcal enterotoxin B: Stimulation of mature T cells and clonal deletion in neonatal mice. Cell. 56:27-35. / Whiting, J . L . , P. M . Rosten, and A . W . Chow. 1989. Determination by western blot (immunoblot) of seroconversions to toxic shock syndrome (TSS) toxin 1 and enterotoxin A , B, or C during infection with TSS- and Non-TSS-Associated Staphylocccus aureus. Infect. Immun. 57:231-234. Winkler, K . C , J . De Waart, and C . Grootsen. 1965. Lysogenic conversion of staphylococci to loss" of B-toxin. J. Gen. Microbiol. 39:321-333.  195  Yanisch-Perron, C , J . Vieira, and J . Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene. 33:103-119. Zumla, A . 1992. Superantigens, T cells, and Microbes. Clin. Infect. Dis. 15:313-320.  196  Appendix 1. Ph.D. Publications To Date  De Boer, M . L . , W . W . W . Kum, and A . W . Chow. July 1995. Staphylococcal toxic shock syndrome toxin-1 and enterotoxin A induce T cell proliferation and TNFa production synergistically in human blood mononuclear cells. Programmes and Abstracts of the 19th International Congress of Chemotherapy, pp. 248C. De Boer, M . L . , and A . W . Chow. 1994. The staphylococcal enterotoxin B gene is amplified as part of a tandem duplication in diverse clones of Staphylococcus aureus. Programmes and Abstracts of the 44th Annual Canadian Society of Microbiologists Meeting, pp. 56. De Boer. M . L . , and A . W . Chow. 1994. Toxic shock syndrome toxin 1-producing Staphylococcus aurues isolates contain the staphylococcal enterotoxin B generic element but do not express staphylococcal enterotoxin B. J. Infect. Dis. 170:818-827.  De Boer M . L . , and A . W . Chow. 1993. Analysis of the staphylococcal enterotoxin B and toxic shock syndrome toxin-1 genetic elements. Abstracts of the 1993 Infectious Diseases Society of America (IDSA) Meeting, pp. 533.  De Boer M . L . , R . P. Novick, and A . W . Chow. 1993. A tampon-associated vaginal infection model of toxic shock syndrome using isogenic S. aureus mutants derived by tst mactivation and allele replacment. Clin. Res. 41(1) :44A.  197  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099042/manifest

Comment

Related Items