Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Comparison between standard in vitro virulence associated assays and human coproantibody siga production.. Fletcher, Kathleen Margaret 1987-07-08

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


831-UBC_1987_A6_7 F58.pdf [ 3.08MB ]
JSON: 831-1.0096892.json
JSON-LD: 831-1.0096892-ld.json
RDF/XML (Pretty): 831-1.0096892-rdf.xml
RDF/JSON: 831-1.0096892-rdf.json
Turtle: 831-1.0096892-turtle.txt
N-Triples: 831-1.0096892-rdf-ntriples.txt
Original Record: 831-1.0096892-source.json
Full Text

Full Text

COMPARISON BETWEEN STANDARD IN VITRO VIRULENCE ASSOCIATED ASSAYS AND HUMAN COPROANTIBODY SIGA PRODUCTION AS PREDICTORS OF YERSINIA ENTEROCOLITICA AND YERSINIA ENTEROCOLITICA-LIKE ORGANISM ASSOCIATED MOUSE VIRULENCE AND HUMAN DISEASE PRESENTATION By KATHLEEN MARGARET FLETCHER (B.Sc, The University of British Columbia, 1985) A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Medical Microbiology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1987 c Kathleen Margaret Fletcher In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of PATHOLOGY The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date OCTOBER 9, 1987  DE-6(3/81) - i i -I. ABSTRACT A semi-quantitative indirect immunofluorescence assay was developed which distinguishes two types of patients from whom yersiniae are recovered: those who produce a strong yersiniae specific coproantibody secretory IgA (SIgA) response and those who do not. This SIgA response appeared to be yersiniae specific as faecal supernatant controls from patients whose stools where shown to yield negative or positive cultures for Salmonella, Campylobacter, or Clostridia were SIgA negative. Organisms isolated from patients with high SIgA titers had a higher incidence of virulence associated characteristics although SIgA response was not associated with most other commonly recognized assays of virulence. A strong association was shown to exist between SIgA titre and mouse virulence, the gold standard of bacterial virulence. Clinical examination of patients culture positive with yersiniae documented a strong association between acute enteric illness and high SIgA titre. This association was not dependant on the cultured yersiniae species. No single jji vitro virulence associated assay was found to be a reliable predictor of animal virulence. The virulence of nine Y.frederikseni i and one Y.kristenseni i, previously thought to be non-pathogenic in man, was also documented. -iii-II. TABLE OF CONTENTS page I. Abstract ii II. Table of Contents iiIII. List of Tables iv IV. List of Figures v V. Introduction A. Yersinia Characteristics 1 B. Virulence Associated Characteristics 4 C. Coproantibody Secretory Immunoglobulin A 8 VI. Methods A. Specimens and Organisms 12 B. Organism Associated Characteristics 1. Autoagglutination 13 2. Calcium Dependance 4 3. Congo Red Absorption 14. Plasmid Profile 15. Mouse Virulence 6 C. Human Coproantibody SIgA Response 1D. Murine Serology 17 E. Diagnostic Test Evaluation 18 VII. Results 20 VIII. Discussion 4 IX. Summary 28 X. Bibliography 9 XI. Appendix A 3XII. Appendix B 40 XIII. Appendix C 1 XIV. Acknowledgments 53 - iv -III. LIST OF TABLES Table § page I. Biochemical Characteristics of Yersinia Species 42 II. Biochemical Features of Yersinia Biogroups 43 III. Summary of Patient and Organism Characteristics 44 IV. Relationship Between In Vitro Yersinia Virulence 46 Associated Assays and Acute Enteric Disease (All Isolates Included) V. Relationship Between In Vitro Yersinia Virulence 47 Associated Assays and Acute Enteric Disease (Excluding Strains Isolated From Patients With Inflammatory Bowel Disease) VI. Relationship Between In Vitro Yersinia Virulence 48 Associated Assays and Mouse Virulence (All Isolates Included) VII. Relationship Between In Vi tro Yersinia Virulence 49 Associated Assays and Mouse Virulence (Excluding Strains Isolated From Patients With Inflammatory Bowel Disease) VIII. Comparison of Positive Virulence Associated 50 Characteristic Expression Between IgA Positive (2+) and Negative (1+ or -) Strains IX. Comparison of Mouse Serological Response to Human Intestinal Response 51 X. Comparison of Days Cold Enrichment to Recovery of IgA Stimulating Organisms 52 - V -IV. LIST OF FIGURES Figure § page 1 Typical Indirect Immunofluorescence SIgA Assay Results 21 - 1 -V. INTRODUCTION The genus Yersinia consists of gram negative fermentative bacilli belonging to the Enterobacteriacae family. Originally members of the genus Pasteurella, these organisms were assigned to a new genus in 1944 based on their oxidase negativity (1), and later, in 1964, they were separated into three species: Y.pestis, Y.pseudotuberculosis. and Y.enterocolitica (2). Y.pestis is the known causative agent of Bubonic Plague in man, Y.pseudotuberculosis is associated with granulomatous lesions of the intestine, and Y.enterocolitica is associated with a variety of enteric disorders including diarrhea, abdominal pain, gastroenteritis, and mesenteric lymphadenitis (3-6). A. Yersiniae Characteristics The first report of Y.enterocolitica-related enteric illness came from Carlsson et a! in 1964 who recovered the organism from the mesenteric lymph nodes of a patient with acute terminal ileitis (7). In 1966 Winblad e_t a_i documented patients culture and sero-positive with V.enterocolitica who had presented with acute diarrhea or appendicitis-like symptoms (8). Although the majority of Y.enterocolitica isolates reported were primarily restricted to Northern Europe, Canadian isolates were first reported in 1967 (9). Since then Y.enterocolitica has been increasingly associated with acute, - 2 -self-limited enteric illness and several post-infection syndromes, such as polyarthritis and erythema nodosum (5,10-13). Y.enterocolitica has also been associated with pseudoappendicitis in children resulting in surgery (14). In addition, severe yersiniae associated septicaemia in children and immunosuppressed patients has been documented (15,16). In 1980, a series of atypical Y.enterocolitica-1ike organisms were subclass ified on the basis of DNA-DNA hybridization homology (17) and unique biochemical properties into three new species: Y.frederikseni i. V.kristenseni i. and Y.intermedia (18-20). Table I outlines the biochemical properties of these yersiniae and those of two additional species, Y.ruckeri and Y.aldovae (21), solely found as aquatic isolates. Y.enterocolitica does not refer to a single organism. Instead, this species may be subdivided by biochemical reactivity into 8 biotypes (Table II) and by 56 1ipopolysaccharide 0 antigens and 19 flagellar H antigens into serotypes (22). When labelling a strain however, only the biotype and 0 antigen type are routinely included. Phage typing systems which further differentiate the serotype/biotypes have also been developed, but are usually reserved for epidemiologic studies (23). Serotype (24) and biotype (25) have both been associated with the organism's pathogenic potential. In 1983, the World Health Organization outlined the syndromes associated with various serotypes (10). Types 0:3 and 0:9 were termed pathogenic as they were associated only with the "classic" gastroenteritis syndrome in man. "Other" serotypes, ie 4, 5, 6, 7, 8, 13, etc, were found only in healthy carriers or cases of - 3 -supra-infection and were therefore considered non-pathogenic. This same report noted that strains belonging to Biotype 1 were not usually associated with human disease. Since then there have been several reports of infection with these "non-pathogenic" strains (26-31), ie Biotype 1 strains have been implicated in septicaemia cases (28), 0:5, 0:5,27, 0:6,30, 0:6,31, and 0:21 in human enteritis cases (32-35), and 0:8 in food bourne outbreaks (36). The varied reports of pathogenic potential may be the result of varied patient ages across the studies, i.e. Caprioli et ai noted a rise in the incidence of "other " serotype with increasing patient age. It is important to note that some "pathogenic" strains have also been recovered from healthy human and animal carriers (10,37,38). Currently the most commonly pathogenic strains are 0:3, 0:5,27, 0:8, and 0:9. Yersinia species are psychotropic and will therefore grow at temperatures lower than competing flora. In 1974 Eiss (45) documented a 40% increase in yersiniae recovery after cold enrichment, however the usefulness of cold enrichment in the recovery of pathogenic organisms has been questioned. Van Noyen (46) reported that cold enrichment did not significantly increase recovery of serotypes 0:3 and 0:9 although increased recovery of other serotypes was found. Y.enterocolitica sensu stricto, and the newly defined strains have been isolated from a variety of environmental and clinical sources including human and animal stools, fish, small mammals, water, milk, raw and cooked pork, tofu, and ice cream (39-44). Y.kristenseni i. Y.frederikseni i, and Y.intermedia have primarily been isolated from the - 4 -environment or skin and wound infections and were not originally associated with enteric illness (18-20). Recently, however, these organisms have been recovered from patients presenting with acute enteric symptoms (6,32). B. Virulence Associated Characteristics Virulence of Y.enterocolitica and related organisms has repeatedly been associated with a variety of in vivo and j_n vitro characteristics. In vivo assays are animal model based, and in vi tro assays range from differential media to plasmid analysis. In vivo virulence has been defined as the ability to cause keratoconjunctivitis in guinea pigs (Sereny Test)(47) , to cause mouse death after intraperitoneal injection (48), and to cause mouse diarrhea or to invade the intestine and deep organs (spleen and liver) of rabbits and mice after oral challenge (49-51). The guinea pig Sereny Test is limited in its usefulness as a yersiniae virulence test as some commonly pathogenic strains (i.e. serotypes 0:3, 0:9, and 0:5,27) fail to evoke a, conjunctivitis response (50). However, when mice are utilized in place of guinea pigs the assay's sensitivity does increase significantly (52). Overall the mouse is currently the animal model of choice as mouse symptoms following inoculation via numerous routes closely mimic those of humans, i.e. intravenous or intragastric inoculation yields mouse intestinal lesions histologically similar to those seen in humans (50,53). Injection of dilute bacterial suspensions intraperitoneally (i.p.) into mice will detect the virulence of numerous serotypes. The - 5 -spectrum of the i.p. assay can be further improved by the addition of iron dextran to the inoculum (54). Oral inoculation mouse models with subsequent mouse diarrhea better parallel the events of human disease, but onset of mouse diarrheal symptoms may be difficult to assess with accuracy. Infiltration of mouse deep organs, liver and spleen, following oral challenge with Yersinia species provides the simplest method of assessing virulence. This model correlates well with animal models, yet has an easier, more definable end point, i.e. <10 orgs/spleen considered avirulent and 50-100 orgs/spleen considered virulent. In vi tro virulence associated characteristics include a requirement of calcium for growth (55), ability to autoagglutinate (56), absorption of a haemin type dye, Congo Red (57), hydrophobicity (58), expression of outer membrane proteins (74), and carriage of a 40-48 Megadalton (Mda) DNA plasmid (59), all of which are expressed in virulent strains at 37°C but not at 25°C. The requirement of calcium for growth by virulent Yersinia species was first reported in 1959 by Higuchi et aj_ (60). They documented virulent Y.pestis strains requiring the cation for growth in liquid media at 37°C. When starved for calcium these virulent strains undergo restriction, a slow down of metabolism, involving a reduction in stable RNA and nucleotide triphosphate formation and decreased adenylate energy stores. Growth on calcium deficient media at 37°C is therefore pin-point in nature. Although no mechanistic role for calcium dependance has been elucidated, this phenomenon may relate to the low i |  | intraphagosomal Ca and Mg ion concentrations encountered during - 6 -ingestion by mammalian cells (55). Since 1959 calcium dependance of yersiniae has been frequently associated with the organism's ability to cause disease in animals (55) and man (61). More recently this characteristic has been assigned to a specific, highly conserved 9kb region on the virulence associated plasmid (62). It is important to note that this characteristic has been detected in environmental non-pathogenic isolates as well as clinical isolates (63,64). Autoagglutination of Yersinia species in tissue culture media at 37°C was reported by Laird and Cavanaugh in 1980 (56). They noted that virulent strains had a greater tendency to clump at 37°C than avirulent strains. In 1985 Prpic et al (58) described autoagglutination by Laird and Cavanaugh's method as a reliable predictor of mouse virulence in organisms carrying an additional virulence associated characteristic, Congo Red absorption (see below). Autoagglutination has been associated with the expression of two outer membrane proteins, V and W antigens, originally defined in Y.pestis and is thought to result from development of a protein fiber matrix linking the cells together (62,65). As with calcium dependance, autoagglutination has been associated with the virulence plasmid although in this case no specific loci has been elucidated (65,66). Y.enterocolitica species do not generally produce exogenous siderophores and therefore rely on the presence of free ferric iron in the environment, or uptake of siderophores released by other organisms in their vicinity, i.e. desferrioxamine released by Streptomyces pilosus (17). Absorption of Congo Red is thought to parallel iron absorption, - 7 -a characteristic with great advantages to organisms attempting to survive in the mammalian host. Absorption was originally associated with virulence in Y.pestis (67,68), but has more recently been documented in Y.enterocolitica species (57). In 1987, the expression of a series of high molecular weight Iron Regulated Proteins (IRPs) on the outer membrane of Yersinia species under iron starvation conditions was documented (69). Such expression could account for iron or congo red uptake although absorption studies on these organisms were not included in their report. In addition, aerobactin producing strains of Y.frederikseni i, Y.kristenseni i. and Y.intermedia have recently been reported (70). Although Y.enterocolitica and V.pseudotuberculosis strains examined in the same study did not produce the hydroxymate siderophore, its role in potential iron absorption should not be ignored. Several studies have documented a family of 40-48 Mda plasmids present in Y.enterocolitica strains that can be associated with virulence (48,52,60,71,72,73,74). The first of this series was documented during a milk related yersiniosis outbreak in 1980 (47). Its association with virulence and tissue invasiveness was established through ethidium bromide curing experiments (47), and at this time it was found that repeated subculture at 37°C also resulted in curing of the plasmid. Since 1980, 50-100% DNA homology has been shown to exist within Y.enterocolitica plasmids and 50% DNA homology between Y.enterocolitica and Y.pestis plasmids, perhaps indicating a common ancestral origin (75). In addition to calcium dependance and - 8 -autoagglutination plasmid carriage has been associated with expression of a unique series of outer membrane proteins (74), cytotoxicity to human epithelial cells (73,75), and inhibition of phagocytosis and chemiluminescence of normal human neutrophils (76). It is important to note that non-pathogenic strains have also been shown to carry plasmids within this size range, and that plasmid-less pathogenic strains have been documented (77). A recent study has suggested the existence of movable genetic elements in Yersinia species that could account for the presence of plasmid characteristics in the absence of a plasmid (78). Each of the _in vi tro characteristics discussed above has some ability to differentiate organisms with potential virulence in man (79), but none considers the human response to the organism. The first objective of this study was to compare the in vi tro virulence associated characteristics of Y.enterocolitica and Y.enterocolitica-1ike organisms recovered from symptomatic patients to animal virulence, the gold standard of bacterial virulence, and to the disease presentation in humans. C. Coproantibody Secretory Immunoglobulin A In the cells of the intestinal lamina propria (Gut Associated Lymphoid Tissue) there is a predominance of Immunoglobulin A (IgA) producing plasma cells, accounting for >80% of the Ig secreting cells (79). Two IgA monomers are linked via disulfide bonds to a 15,000mw J-chain and actively transported out of the lamina propria (79). - 9 -Epithelial cells synthesize a glycoprotein or secretory component which in turn acts as a basal receptor for the IgA dimer. This complex (SIgA) is then transported through the epithelial cell layer to the gut lumen and released. This type of active transport ensures its presence in large quantities in intestinal secretions. Of the immunoglobulins, SIgA is the best suited to deal with antigenic material in the gut lumen as it resists phagocytosis and proteolysis by intestinal enzymes, is quadravalent, and has an affinity for the mucosa (80). SIgA is thought to be the most important resistance mechanism for bacterial elimination as it can both agglutinate the organisms and act as an immune barrier in the intestine by preventing adherence and thereby invasion of the organisms (81). This type of blocking activity is preferred in the gut as it does not activate any further antimicrobial activity, i.e. complement activation, that could lead to degranulation and tissue damage. Several recent studies have documented a rise in serum titre of yersiniae specific antibody early in infection (82-84) primarily accounted for by IgA class antibodies in particular the 11S or dimer form normally found in secretions, however the reliability of serological diagnosis of yersiniae infection remains controversial. Agglutination assays, either by tube or microtitre, have great potential for cross-reactivity as whole organisms are utilized (85). In addition, although high antibody titers can be considered diagnostic, low antibody titers may result from some underlying illness, i.e. co-infection or immunodeficiency, and cannot exclude yersiniosis. -10-In addition, agglutination assays detect primarily IgM and IgG as IgA does not agglutinate. In fact, the presence of IgA may even decrease the apparent titre of IgM and IgG by binding to antigenic sites and blocking IgM and IgG interactions (84). Although ELISA assays provide a more sensitive method for detecting class specific antibody, they still do not overcome the problems associated with IgA blocking (85,86). In addition to methodological problems, no criteria exist for ascribing significance to an increased IgA titre (87), i.e. healthy individuals often exhibit high background antibody levels to yersiniae. Therefore the diagnostic applications of serum IgA levels in a case of suspected yersiniosis are limited, and any serology for this purpose requires at least one sera from both acute and convalescing phases (13,88). Overall, in spite of the difficulties discussed, an increased IgA titre has been implicated in the prediction of post-infection aseptic arthritis, thereby remaining important (85,87). In 1971 Reed and Williams (89) documented the presence of Secretory IgA (SIgA) class shigella agglutinating antibodies in faeces from patients with recent onset diarrhea. In the past 25 years several other investigators have reported coproantibody responses to enteric organisms in rectal mucosa and faeces using indirect immunofluorescence assays (90-92). Although no mechanistic role has been suggested, a good association between coproantibody production and stage of enteric illness was clearly demonstrated. The second, objective of this study was to determine whether or not an immunological marker of human immune response, coproantibody SIgA, correlated first with human disease presentation and secondly with any of the established virulence associated markers including mouse virulence. It was suggested by Sarasombath et aj. in 1987 (93) that systemic antibody responses and intestinal SIgA responses were linked during the course of S.typhi infections. Such a linkage potentially exists during yersiniae infection. The third objective of this study was to compare organisms eliciting varied human intestinal immune responses to serological titers in a mouse model. VI. METHODS - 12 -A. SPECIMENS AND ORGANISMS Faecal specimens were received in the UBC-HSCH Microbiology Laboratory in Cary-Blair Transport Media or sterile containers and routinely processed for detection of enterics as previously described (6). Briefly, faecal material was plated on CIN Yersinia Selective Agar (PML Microbiologicals) for 18-24 hours at 37°C. Remaining specimen was incubated at 4°C in broth culture and subcultured after 1, 3, 7, and 14 days. Colonies with the characteristic bulls-eye appearance were screened on Triple Sugar Iron, Urea, Lysine, Indole, and motility assays (PML Microbiologicals). Strains were further characterized on API 20E Enterobacteriacae System (API Laboratory Products). All Y.enterocolitica strains were then biotyped and serotyped by an independent reference laboratory. Faecal specimens were simultaneously examined for the presence of parasites, but viral cultures were not included. Original faecal specimens were maintained at -4°C until bacteriology was completed. When a Yersinia species was recovered, the faeces was processed for coproantibody SIgA detection. Organisms were maintained at -70°C in Glycerol and Dimethyl sulfoxide (DMSO) until assayed for virulence. Clinical history of all patients was reviewed and is presented in Table 3. - 13 -In cases where the organisms were isolated and frozen prior to the onset of this study, faecal supernatant was not available. These organisms were included in the virulence testing for comparison purposes, but were not tested for SIgA. B. ORGANISM VIRULENCE ASSOCIATED ASSAYS 1. Autoagglutination Interpretation of autoagglutination is commonly based on subjective observation of complete tube clearing. Several authors have expressed difficulty in assessment of autoagglutination with Laird and Cavanaugh's method (56), i.e. strains failing to grow in tissue culture media, high rate of false positives, varying results dependant on inoculum size, and faulty interpretation as a result of partially cleared tubes (94,95). For these reasons the method of Laird and Cavanaugh (56) was modified to include a spectrophotometric analysis of clearing. Organisms were grown overnight in tryptone yeast extract broth (pH 7.0) (Difco) at 25°C, centrifuged, washed in phosphate buffered saline (PBS), and 9 resuspended to McFarlane Standard 3 (10 orgs/ml). The culture was divided, incubated 18 hours at 25°C or 37°C, and the optical density (O.D.) read at 460 nm. Known autoagglutination positive and negative strains (Strains 700L and 700S, Dr.C.Pai, Calgary) were included in ten experimental - 14 -runs and the difference between twice their Standard Error of the Mean (Appendix A) values was calculated to be 0.1 O.D. units. Strains where the O.D. at 25°C exceeded that at 37 C by 0.1 were interpreted as autoagglutination positive. 2. Calcium Dependance Direct plating on Magnesium Oxalate Agar (Appendix B) as described by Higuchi and Smith (96) was used to detect calcium dependant strains. Colonies from organisms requiring calcium for growth (Calcium Dependant) were <1 mm diameter after 36 hours at 37°C and those not requiring calcium (Calcium Independent) were >1 mm diameter. 3. Congo Red Absorption The Congo Red Acidmorpholinepropanesulfonic Acid agar (Appendix B) described by Prpic et aj. (57) was used to assess absorption of Congo Red. Organisms that actively take up the dye form red colonies after 36 hours at 37°C and are scored as positive, those that do not form white or slightly pink colonies and are scored as negative. 4. Plasmid Profile Presence of a 40-48 Mda plasmid was determined with a modified Birnboim and Doly Alkaline Lysis Extraction method (97). Organisms were grown overnight in beef heart infusion broth (Difco) at 25°C and harvested by centrifugation, washed, and 9 resuspended to 10 orgs/ml in PBS. Cells were resuspended in lOOul fresh, chilled Lysis Buffer (50mM Glucose, lOmM EDTA, - 15 -25mM Tris-HCl (pH 8.0), and 4mg/ml Lysozyme) and allowed to stand 5 minutes at room temperature to weaken the cellular walls. 200ul of fresh, chilled Alkaline Buffer (0.2 N NaOH and 1% SDS) was added, mixed twice by inversion, and allowed to stand 5 minutes on ice to lysis cells. 150ul fresh, chilled Acid Buffer (60 ml Potassium Acetate, 11.5 ml Glacial Acetic Acid and 28.5 ml distilled ^0) was added, inverted and gently vortexed for 10 seconds, and allowed to stand 5 minutes on ice to precipitate chromosomal DNA. Samples were centrifuged (12,000 xg for 5 min) and the supernatant added to an equal volume of Phenol/Chloroform. Samples were then vortexed and centrifuged (12,000 xg for 2 min) to extract contaminating proteins. The supernatant was removed and allowed to stand in 2 volumes of 70% ethanol at -20°C for 20 minutes to precipitate plasmid DNA. Samples were recentrifuged, additional ethanol poured off, and the pellet air dried. Pellets were resuspended in 40ul of RNase Buffer (30mM Tris (pH 8.0), 5mM EDTA, 50mM NaCl, and 20 ug/ml RNase) to eliminate contaminating RNA. 15ul of DNA solution and 5ul of Stop Mix (50 ml Glycerol, 7ml EDTA (0.5 M), 5mg Bromphenol Blue, in 100 ml Distilled HgO) were loaded on a Tris Borate Buffered Agarose Gel (8.9mM Tris Base, 8.9mM Boric Acid, 0.2mM EDTA, and 0.6% agarose in distilled HgO). Gels were electrophoresed for 2 hours at 50mV on a Mini-Gel Apparatus (Bethesda Research Laboratories) and post-stained in 5ug/ml - 16 -ethidium bromide for 20 minutes. Standards included T4, T5, T7, and lambda phage DNA (Sigma) and a strain known to carry the 40Mda plasmid (700S). Those strains with plasmids banding in the same region as the control strain were considered plasmid carrying. 5. Mouse Virulence For animal virulence assessment the method of Bakour et aj_ (50) was modified. Adult Swiss mice (Charles River) were dehydrated for 24 hours then allowed to drink ad 1ib from a 0.1% peptone 9 broth containing 10 orgs/ml, from an overnight tryptone soy agar plate (Difco), for 36 hours. Mice were sacrificed by carbon dioxide and their spleens homogenized then plated on C1N yersinia selective agar (PML Microbiologicals) for 24 hours at 37°C. Colonies with the characteristic bulls-eye appearance were further characterized on TSI and Urea slants (PML Microbiologicals). A strain was defined as mouse virulent if the homogenized spleen contained more than 100 organisms. C. HUMAN C0PR0ANTIB0DY SIGA RESPONSE Intestinal SIgA response was assessed with a newly established semi-quantitative indirect immunofluorescence assay. Faecal samples were stored at -4°C until a yersiniae species was recovered. Faecal specimens were then diluted 1:1 with PBS, centrifuged (30 min at 15,000xg) and the supernatant stored at -70°C. 200 ul of serial - 17 -supernatant dilutions (1/10,1/50,1/100) were incubated with 40 ul of 9 o 10 orgs/ml in PBS on covers lips (18 hours at 25 C). Covers lips were air dried, dip washed in distilled water, and incubated with 120 ul of FITC labelled goat anti-human IgA (Sigma) (30 min at 37°C). Slips were dip washed, air dried, inverted and mounted for fluorescence microscopy. Results were defined as follows: no fluorescence - ; single organisms 1+; homogeneous microagglutination 2+ (Figure 1). D. MURINE SEROLOGY Adult Swiss mice were orally inoculated with IgA 2+, 1+, and -strains as described above. Animals were sacrificed 7 days after inoculation and their heart blood immediately withdrawn. Blood was centrifuged and sera stored at -70°C until time of assay. Formalinized antigen was prepared as described by Winblad (98). Organisms were grown 1-2 days on Blood Agar plates (PML Microbiologicals) at room temperature, harvested by centrifugation, and washed in saline before suspending in 10% Buffered Formalin. A loopful of antigenic material was streaked on Blood Agar plates and incubated overnight at 37°C to check sterility. Prior to the microagglutination assay the antigenic material was washed in saline and resuspended to McFarlane Standard 5. 0.5 ml of sera was serially diluted in duplicate round bottom microtitre plates (Becton Dickinson) and 0.5 ml of formalinized antigen was added to each well. Plates were spun at 700 - 18 -rpm for 10 minutes. The inverse of the last dilution showing halo-type agglutination was recorded as the antibody titre. E. DIAGNOSTIC TEST EVALUATION Clinical evaluation of any diagnostic test is based on the test's accuracy, that is, how close is the test measurement to the true value (99). The true value in many cases is unreadable, therefore new tests are compared against established or "gold standard" tests. In this study, two gold standards of organism virulence were utilized. Animal virulence is a widely accepted standard of organism pathogenicity, and association with acute human enteric symptoms further approximates the true human pathogenic potential of the organism. When selecting a test for use it becomes important to know the probability that the test will be positive when applied to an organism that is truly virulent, i.e. its sensitivity, and the probability that the test will be negative if applied to an organism that is truly avirulent, i.e. its specificity. Although these values are useful when selecting a test for use, they are of limited use in evaluating a single test result. In that case it is necessary to know the probability that an organism will be virulent if the test result is positive, that is, the predictive value of the test needs to be calculated. Positive predictive value refers to the ratio of true positives to test positives and negative predictive value refers to the ratio of true negatives to test negatives. Predictive values are influenced not only by the - 19 -sensitivity and specificity of the test, but also by the probability of virulence in the population prior to testing (99). Testing of a population with a high incidence of positives increases the predictive values. Appendix C shows sample calculations of positive predictive value, negative predictive value, sensitivity, and specificity for the Coproantibody SIgA assay. However, since this study was concerned primarily with the probability that an organism is pathogenic in man when the test value is positive, only positive predictive values for remaining assays are reported in the text. - 20 -VII. RESULTS Figure 1 illustrates typical fluorescence microscope fields for a 2+, a 1+, a negative, and a negative in phase contrast. A 2+ result is defined as large clusters of agglutinating organisms fluorescing bright apple green and is indicative of a strong specific IgA class antibody response (Figure IA). A 1+ result is defined as single fluorescent organism and is indicative of a weak specific antibody response (Figure IB). When the cultured yersiniae species stimulates no antibody response, the field is apparently empty (Figure 1C). This absence of fluorescence can not be accounted for by an absence of bacteria however as organisms are detected in phase contrast of the same field (Figure ID). Supernatant only and organism only control slides (photos not included) showed some homogeneous background fluorescence, but no organism-like structures. Faecal supernatants from salmonella (Pt# 64), Campylobacter (Pt# 65, 66, 67), or Clostridia (Pt# 68) associated diarrhea patients and asymptomatic, organism-negative patients (Pt# 69, 70, 71) were examined for cross-reactive agglutination with yersinia. No specific agglutinating cross-reactivity was found. However, non-agglutinating halo-type staining of single organisms was detected when supernatants were tested against control organisms including salmonella, Campylobacter, Clostridia, and Escherichia coli. - 21 -• Figure 1 - Typical Indirect Immunofluorescence SIgA Assay Results. A, ++, Homogenous Microagglutination B, +, Single Organism and Non-Specific Background; C, -, No Specific Fluorescence; D, - in Phase Contrast, Non-clumping Organisms Coating Slide. Table III contains a complete listing of all patients and organism characteristics used in this research and referred to in the text. Table IV documents the ability of the virulence associated assays to differentiate organisms from patients with acute enteric symptoms. Although not absolute, the positive predictive value of SIgA response for acute disease (P.P. - 85.7%) was greater than that of any single in  vi tro (Autoagglutination - 56.0%, Calcium Dependance - 56.5%, Congo Red - 58.5% , Plasmid Carriage - 53.7%) and paralleled that of mouse virulence (P.P. - 73.7%). Of note, exceptions to this association included patients # 6, 8, and 47 who suffered inflammatory bowel disease - 22 -yet elicited a strong SIgA response (see discussion) and patients #9, 10, 24, and 25 who presented with acute diarrhea yet had a low or absent SIgA response. Patient #10 was shown to be Clostridium difficile toxin positive, although the bacterium was not cultured, and patient #9 carried B. hominis. When organisms recovered from patients with inflammatory bowel disease are excluded from data analysis on the basis that these patients may be presenting with diarrheal symptoms as a result of underlying illness, SIgA production remains a better predictor of acute enteric symptoms (P.P. - 75.0%) than any other in vitro assay (Autoagglutination - 52.2%, Calcium Dependance - 52.3%, Congo Red - 55.3% , Plasmid Carriage - 51.4%) and still parallels the gold standard of mouse virulence (P.P. - 70.6%)(Table V). Overall positive predictive values (P.P.) of the in vitro virulence associated assays for mouse virulence were: Autoagglutination - 44.0%; Calcium Dependance - 39.1; Congo Red Absorption - 34.1%; and Plasmid Carriage - 29.3% (Table VI), although these values fluctuated slightly with the yersiniae species. These results agree with those of Prpic e_t al (58), which found autoagglutination to be the most reliable predictor of mouse.virulence in congo red positive organisms. The newly established coproantibody SIgA assay predicted mouse virulence better than any of the established virulence associated assays (P.P. -85.7%)(Table VI) and if organisms recovered from patients with inflammatory bowel disease are again excluded the SIgA assay appears to be an absolute predictor of mouse virulence (Table VII). - 23 -Organisms cultured from SIgA positive (2+) patients have a higher incidence of some virulence associated characteristics. Table VIII documents no significant difference between SIgA 2+ and SIgA 1+/-2 strains in expression of congo red absorption (X =0.00), calcium 2 2 dependance (X =0.14), and plasmid carriage (X =0.15), however, a 2 significant difference was apparent with mouse virulence (X =17.65; p<0.002) and autoagglutination (X2=5.39; p<0.05). Table IX relates the ability of organisms to stimulate a human coproantibody response to their ability to elicit a serological response in another animal, the mouse, similiarly orally challenged. Although no definite conclusions could be drawn from only four organisms, it was noted with interest that organisms coming from SIgA producing patients (2+) elicited high antibody responses in the mouse whereas those coming from low SIgA producers (1+ or -) elicited little or no antibody response. These findings are consistent with those of Sarasombath et al (93) that a correlation between intestinal and systemic antibody responses does exist. Simultaneous human sera was not available from these patients, therefore similar experiments in man were not included in the study. Table X compares the days cold enrichment required for recovery to the type of SIgA stimulation. The number of SIgA stimulators did not increase or decrease substantially from 0 to 14 days and although a slight increase in SIgA non-stimulators was seen in the first week, this increase did not persist into the second week. Overall, 58.3% of the mouse virulent, antibody-indueing yersiniae were recovered only after cold enrichment. VIII. DISCUSSION The virulence test results in this study agree with those of Kay et al (100) that no single virulence associated assay correlates well with yersiniae virulence as assessed by the mouse model. In addition, this study demonstrates a relationship between coproantibody SIgA stimulation, mouse virulence, and acute human enteric illness. As discussed earlier, V.frederikseni i, Y.kristenseni i. and V.intermedia were not originally considered enteric pathogens but have recently been recovered from patients presenting with acute diarrhea. Prpic e_t aj_ (58) found that some V. enterocol it ica-1 ike organisms expressed some of the j_n vi tro virulence associated characteristics, but were still mouse avirulent. Similarly, this study documents the expression of one or more _in vi tro characteristics in Y.frederikseni i. Y.intermedia, and Y.kristenseni i species. It also documents the existence of Y.frederikseni i. Y.kristenseni i, and atypical Y.enterocolitica strains (i.e. 0:41,43, 0:36, etc) which are mouse virulent and immunostimulatory in humans. This is consistent with the previously discussed observation that strains of Y.enterocolitica other than 0:3, 0:5,27, etc, may be pathogenic in man. Although the usefulness of cold enrichment techniques is still in question, this study found it to be important in recovery of some enterically pathogenic yersiniae, i.e. 58.3% of the mouse virulent, - 25 -immunostiraulatory yersiniae were recovered only after cold enrichment. The existence of coproantibody SIgA in faecal specimens following enteric infections was documented as early as 1947 when Harrison & Banvard (101) reported the detection of agglutinating antibody in faecal specimens from patients with both acute and chronic bacillary dysentery. Since then several studies have reported the presence of high coproantibody titers in faeces (102) and intestinal mucosa (90) of patients with ulcerative colitis and in volunteers orally immunized with live S. typhi vaccine (91,92). In the current investigation the presence of high SIgA titers in faecal specimens has been documented and strongly associated with acute enteric illness and carriage of mouse virulent organisms. Of note, when patients with inflammatory bowel disease were excluded from analysis because their underlying disease represented confounding data, the positive predictive value of SIgA for mouse virulence became 100%. In addition, this association was not dependent on the cultured yersiniae species. In 1951 Barksdale et aj. (102) noted that certain strains of coliform bacilli contained antigenic components common to certain shigella and salmonella species. Such cross-reactions could conceivably limit the diagnostic value of a coproantibody assay, therefore appropriate controls, including supernatants from salmonella, Campylobacter, or Clostridia diarrhea patients and enteropathogen negative asymptomatic patients, were included in this study. Since these specimens were negative, a strong agglutinating SIgA response appeared to be yersiniae specific. However, as the number of control supernatants in this study was low, no overall specificity could be - 26 -determined. Two exceptions to the association between SIgA response, acute disease, and mouse virulence in the current study were found in patients with Crohn's Disease and acute colitis, both of whom exhibited a strong SIgA response to an apparently avirulent yersiniae strain. This observation was consistent with Barksdale et ai's case of a chronic ulcerative colitis patient from whom neither shigella nor salmonella were isolated, yet whose coproantibodies agglutinated three salmonella species and four or more shigella species (102). Toivanen et aj_ (103) suggested that persistence of yersiniae within the intestine, i.e. in epithelial or lymphoid tissues, could provide a stimulus for prolonged antibody production in cases of reactive arthritis. Such a mechanism may also be involved in maintenance of antibody production in inflammatory bowel disease. Monteira et al (92) suggested that strains of enteric bacteria which would produce only transient reactions in non-colitic patients may be able to elicit antibody production in mucosa of colitic patients. Further explanations include the possibility that mouse-virulent organisms could lose their temperature-dependent virulence plasmid while in the gut lumen or that strains of yersiniae may exist which are immunostimulatory in humans yet inactive in the mouse model. The second explanation is supported by Hancock et al's (104) finding that certain strains of mice are naturally resistant to some yersiniae strains. For the aforementioned reasons, the coproantibody SIgA assay as outlined here may produce false positives and should not be relied on to predict the animal virulence potential of - 27 -yersiniae strains recovered from patients with known inflammatory bowel disease . Carriage of yersiniae in the presence of another causal agent would account for cases of acute diarrhea where low or absent SIgA responses were detected. The mouse avirulent nature of the cultured yersiniae species in these cases and the observation that some of these serotypes are normally found only in carriers or in suprainfections (10) supports this concept. Two such patients (Pt# 9, 10) carried bacterial or parasitic pathogens which may have accounted for their symptoms. No additional organisms were cultured from the remaining patients (Pt# 24, 25), however, since electron microscopy of faecal samples was not done virally induced enteritis could not be excluded. A number of patients culture positive with a mouse virulent yersiniae carried a concurrent parasite, i.e. B.hominis. D.fragilis, or I.butchli i. These patients may represent concurrent yersiniae and parasitic infection, or infection with one organism predisposing the patient to the other. The actual sequence of infection, however, could not be determined from a single faecal specimen. In these cases the coproantibody SIgA assay may be of use in differentiating transient non-immunostimulatory yersiniae isolates from those eliciting a strong intestinal immune response. - 28 -IX. SUMMARY In summary this thesis has outlined the use of a semi-quantitative indirect immunofluorescence assay which facilitates the performance of a serological assay without the need for repeat specimens. This assay was shown to distinguish two types of patients from whom yersiniae are recovered: those who produce a strong yersiniae specific coproantibody secretory IgA (SIgA) response and those who do not. A positive (2+) SIgA response was a better predictor of acute human enteric illness and mouse virulence than any of the established in vi tro assays of virulence and appeared to be yersiniae specific. In addition, the virulence of nine Y.frederikseni i and one Y.kristenseni i, previously thought to be non-pathogenic in man, was documented. - 29 -X. BIBLIOGRAPHY 1 Van Loghem JJ. The classification of the plague bacillus. Journal of Microbiology and Serology, 10, 15, 1944. 2 Frederiksen U. A study of some Versinia pseudotuberculosis-1ike bacteria ("Bacterium enterocoliticum" and "Pasteurella X"). Proceedings of the XIV Scandinavian Congress of Pathology and Microbiology, Oslo, 103, 1964. 3 Mollaret HH, Thai E. Genus XI Yersinia. In RE Buchanan^ N Gibbons (eds) Bergey's Manual for Determinative Bacteriology 8 ed. The Williams and Wilkins Co., Baltimore, 330, 1974. 4 Hoogkamp-Korstanje JAA. Yersiniosis in childhood: Incidence and diagnosis. In C Simons P Wilkinson (eds) Diagnosis of Infectious Diseases-New Aspects. Schattauer, New York, 69, 1986. 5 Leino R, Granfors K, Havia T, Heinonen R, Lampinen M, Toivanen A. Yersiniosis as a gastrointestinal disease. Scandinavian Journal of Infectious Disease, 19, 63, 1987. 6 Simmonds SD, Noble MA, Freeman HJ. Gastrointestinal features of culture-positive Yersinia enterocolitica infection. Gastroenterology, 92, 112, 1987. 7 Carlsson MG, Ryd H, Sternby NH. A case of human infection with Pasteurella pseudotuberculos is X. Acta Pathologica and Microbiologica Scandinavica, 62, 128, 1964. 8 Winblad S, Nilehn B, Sternby N. Yersinia enterocolitica (Pasteurella X) in human enteric infections. British Medical Journal Dec, 1363, 1966. 9 Toma S. Survey on the incidence of Yers inia enterocolitica in the province of Ontario. Canadian Journal of Public Health Sept/Oct 1973. 10 World Health Organization Group Meeting. Yersiniosis. EURO Reports & Studies, Copenhagen. 1983. 11 Leino R, Kalliomaki L. Yersiniosis as an internal disease. Annals of Internal Medicine. 81, 458, 1974. - 30 -12 Granfors K, Toivanen A. IgA-anti-yersinia antibodies in yersinia triggered reactive arthritis. Annals of the Rheumatic Diseases, 45, 561, 1986. 13 Attwood SEA, Cafferkey MT, West AB, Healy E, Mealy K, Buckley TF, Boyle N, Keane FBV. Yersinia infection and acute abdominal pain. The Lancet, Mar 7, 529, 1987. 14 Saebo A The Yersinia enterocolitica infection in acute abdominal surgery. A clinical study with a 5-year follow-up period. Annals of Surgery, 198, 760, 1983. 15 Bouza E, Dominguez A, Meseguer M, Buzon L, Boixeda D, Revillo MJ, DeRafael L, Martinez-Beltran J. Yersinia enterocolitica septicemia. American Journal of Clinical Pathology, 74, 404, 1980. 16 Foberg U, Fryden A, Kihlstrom E, Persson K, Wei land 0. Yers inia  enterocolitica septicemia: Clinical and microbiological aspects. Scandinavian Journal of Infectious Diseases, 18, 269, 1986. 17 Bottone, EJ. An approach to laboratory identification with reference to deoxyribonucleic acid hybridization groups and biochemical characteristics. In : Bottone EJ, Ed, Yers inia  enterocolitica. CRC Press, Boca Raton, Florida. 17, 1981. 18 Bercouvier H, Ursing J, Brenner DJ, Steigerwalt AG, Fanning GR, Carter GP, Mollaret HH. Yers inia kristenseni i: A new species of Enterobacteriaceae composed of sucrose - negative strains (formerly called atypical Yers inia enterocolitica or Vers inia enterocolitica—1 ike). Current Microbiology, 4, 219, 1980. 19 Brenner DJ, Bercouvier H, Ursing J, Alonso JM, Steigerwalt AG, Fanning GR, Carter GP, Mollaret HH. Yers inia intermedia: A new species of Enterobacteriaceae composed of rhamnose-positive, melibiose-positive, raffinose-positive strains (formerly called Yers inia enterocolitica 18 Yers inia enterocolitica-1ike). Current Microbiology, 4, 207, 1980. 20 Ursing J, Brenner DJ, Bercouvier H, Fanning R, Steigerwalt AG, Brault J, Mollaret HH. Yers inia frederikseni i: A new species of Enterobacteriaceae composed of rhamnose-positive strains (formerly called atypical Yers inia enterocolitica or Yers inia enterocolitica-1ike). Current Microbiology, 4, 213, 1980. - 31 -21 Bercouvier H, Steigerwalt AG, Guiyoule A, Huntley-Carter G, Brenner DJ. Yers inia aldovae (formerly Yers inia enterocolitica-1ike Group X2): A new species of Enterobacteriaceae isolated from aquatic ecosystems. International Journal of Systematic Bacteriology, 34, 166, 1984. 22 Aleskic S, Bockemuhl J, Lange F. Studies on the serology of flagellar antigens of Yersinia enterocolitica and related Yersinia species. Zentroblatt Bakteriologie Hygiene Section A, 261, 299, 1986. 23 Baker PM, Farmer JJ. New bacteriophage typing system for Yers inia  enterocolitica. Yers inia kri stenseni i, Yers inia frederikseni i and Yers inia intermedia: Correlation with serotyping, biotyping and antibiotic susceptibility. Journal of Clinical Microbiology, 15, 491, 1982. 24 Une T. Studies on the pathogenicity of Yersinia enterocolitica. Microbiology and Immunology, 21, 349, 1977. 25 Schiemann DA, Devenish JA. Relationship of HeLa cell infectivity to biochemical, serological and virulence characteristics of Yers inia enterocolitica. Infection and Immunity, 35, 497, 1982. 26 Bottone EJ. Current trends of Yers inia enterocolitica isolates in the New York City area. Journal of Clinical Microbiology, 17, 63, 1983. 27 Karmali MA, Toma S, Schiemann DA, Ein SH. Infection caused by Yersinia enterocolitica serotype 0:21. Journal of Clinical Microbiology, 15, 596, 1982. 28 Seto WH, Lau JTK. Septicaemia due to Yers inia enterocoli tica biotype I in Hong Kong. Journal of Infection, 8, 28, 1984. 29 Marymont JE, Durfee KK, Alexander H, Smith JP. Yers inia  enterocolitica in Kansas. American Journal of Clinical Pathologists June, 753, 1982. 30 Synder JD, Christenson E, Feldman RA. Human Yers inia  enterocolitica infections in Wisconsin. Clinical, laboratory and epidemiological features. American Journal of Medicine, 72, 768, 1982. 31 Bissett MJ. Yersinia enterocolitica isolates from humans in California. Journal of Clinical Microbiology, 4, 137, 1979. 32 Lewis AM, Chattopadhyay B. Faecal carriage rate of Yersinia species. Journal of Hygiene, 97, 281, 1986. - 32 -33 Toma S, LaFleur L. Yers inia enterocolitica infections in Canada 1966 to August 1978. In: Bottone EJ, Ed, Yers inia enterocolitica. CRC Press, Boca Raton, Florida. 9, 1981. 34 Martin T, Kasian GF, Stead S. Family Outbreak of Yersiniosis. Journal of Clinical Microbiology 16, 622, 1982. 35 Ratnam S, Mercer E, Picco B, Parsons S, Butler R. A nosocomial outbreak of diarrheal disease due to Yers inia enterocolitica serotype 0:5, biotype 1. Journal of Infectious Disease, 145, 242, 1982. 36 Shayegani M, Stone WB, DeForge I, Root T, Parsons LM, Maupin P. Yers inia enterocolitica and related species isolated from wildlife in New York State. Applied and Environmental Microbiology Sept, 420, 1986. 37 Toma S, Deidrick VR. Isolation of Yersinia enterocolitica from swine. Journal of Clinical Microbiology, 2, 478, 1975. 38 Adesiyun AA, Agbonlahor DE, Lombin LH, Kwaga JKP. Occurrence of virulence markers in species of Yersinia isolated for animals in Nigeria. Veterinary Microbiology 12, 289, 1986. 39 Bercouvier H, Mollaret HH. Genus XIV Yersinia. In NR Kreig JG Holt (eds) Bergey's Manual of Systemic Bacteriology (Vol 1). The Williams and Wilkins Co., Baltimore, 498, 1984. 40 Schiemann DA. Isolation of toxigenic Yers inia enterocolitica from retail pork products Journal of Food Protection 43, 360, 1980. 41 Tauxe RV, Wauters G, Goossens V, Van Noyen R, Vandepitte J, Martin SM, DeMol P, Thiers G. Yers inia enterocolitica infections and pork: The missing link. The Lancet May 16, 1129, 1987. 42 Delmas CL, Vidon DJ. Isolation of Yers inia enterocolitica and related species from foods in France. Applied and Environmental Microbiology, 50, 767, 1985. 43 Kapperud G. Studies on the pathogenicity of Yers inia  enterocolitica and Y. enterocolitica-1ike bacteria. Acta Pathologica Microbiologica Scandinavica, 88, 287, 1980. 44 Langeland G. Yers inia enterocolitica and Yers inia enterocolitica-1ike bacteria in drinking water and sewage sludge. Acta Pathologica Microbiologica et Immunologica Scandinavia 91, 179, 1983. - 33 -45 Eiss J. Selective culturing of Yers inia enterocolitica at low temperature. Scandinavian Journal of Infectious Disease, 7, 249, 1975. 46 Van Noyen R, Vandepitte J, Uauters G. Nonvalue of cold enrichment of stools for isolation of Yers inia enterocolitica serotypes 3 and 9 from patients. Journal of Clinical Microbiology, 11, 127, 1980. 47 Zink DL, Feeley JC, Wells JG, Vanderzant C, Vickery JC, O'Donovan GA. Plasmid-mediated tissue invasiveness in Yers inia  enterocolitica. Nature, 283, 224, 1980. 48 Gemski P, Lazere JR, Casey T. Plasmid associated with pathogenicity and calcium dependency of Yers inia enterocolitica. Infection and Immunity, 27, 682, 1980. 49 Alonso JM, Vilmer E, Mazigh D, Mollaret HH. Mechanisms of acquired resistance to plaque in mice infected by Yers inia enterocolitica 0:3. Current Microbiology. 4, 117, 1980. 50 Bakour R, Balligand G, Laroche Y, Cornells G, Wauters G. A simple adult-mouse test for tissue invasiveness in Yers inia enterocoli tica strains of low experimental virulence. Journal of Medical Microbiology, 19, 237, 1985. 51 Pai CH, Mors V, Seemayer TA. Experimental Yers inia enterocoli t i ca enteritis in rabbits. Infection and Immunity, 28, 238, 1980. 52 Aulisio CCG, Hill WE, Stanfield JT, Sellers RL. Evaluation of virulence factor testing and characteristics of pathogenicity in Yers inia enterocolitica. Infection and Immunity, 40, 330, 1983. 53 Carter PB. Pathogenicity of Vers inia enterocolitica for mice. Infection and Immunity, 11, 164, 1975. 54 Smith RE, Carey AM, Damare JM, Hetrick FM, Johnston RW, Lee WH. Evalutaion of iron dextran and mucin for enhancement of the virulence of Yers inia enterocoli tica serotype 0:3 in mice. Infection and Immunity, 34, 550, 1981. 55 Brubaker RR. Low-calcium response of virulent Yersiniae. Medical Microbiology and Immunology, 175, 43, 1986. 56 Laird WJ, Cavanaugh DC. Correlation of autoagglutination and virulence of Yersiniae. Journal of Clinical Microbiology, 11, 430, 1980. - 34 -57 Prpic JK, Robins-Browne RM, Davey RB. Differentiation between virulent and avirulent Yers inia enterocolitica isolates by using congo red. Journal of Clinical Microbiology, 18, 486, 1983. 58 Prpic JK, Robins-Browne RM, Davey RB. In vitro assessment of virulence in Yers inia enterocolitica and related species. Journal of Clinical Microbiology, 22, 105, 1985. 59 Gemski P, Lazere JR, Casey T. Plasmid associated with pathogenicity and calcium dependency of Vers inia enterocolitica. Infection and Immunity, 27, 682, 1980. 60 Higuchi K, Kupferberg LL, Smith JL. Studies on the nutrition and physiology of Pasteurella pestis. III. Effects of calcium ions on the growth of virulent and avirulent strains of Pasteurella pestis. Journal of Bacteriology, 77, 317, 1959. 61 Delmas CL, Vidon DJ. Isolation of Yersinia enterocolitica and related species from foods in France. Applied and Environmental Microbiology, 50, 767, 1985. 62 Cornells G, Laroche Y, Balligand G, Sory MP, Wauters. Yers inia  enterocolitica, A primary model for bacterial invasiveness. Reviews of Infectious Diseases, 9, 64, 1987. 63 Noble MA, Barteluk RL, Freeman.HJ, Subramaniam R, Hudson JB. Clinical relevance of virulence assays of Yersinia species. Journal of Clinical Microbiology, 25, 802, 1987. 64 Carter PB, Varga CF, Keet EE. A new strain of Yers inia  enterocolitica pathogenic for rodents. Applied Microbiology, 26, 1016, 1973. 65 Balligand G, LaRoche Y, Cornells G. Genetic analysis of virulence plasmid from a serogroup 9 Yers inia enterocolitica strain: Role of outer membrane protein PI in resistance to human serum and autoagglutination. Infection and Immunity, 48, 782. 1985. 66 Skurnik M. Lack of correlation between the presence of plasmids and fimbriae in Yersinia enterocolitica and Yers inia  pseudotuberculosis. Journal of Applied Bacteriology, 56, 355, 1984. 67 Payne SM, Finkelstein RA. Detection and differentiation of iron-responsive avirulent mutants on congo red agar. Infection and Immunity, 18, 94, 1977. - 35 -68 Surgalla MJ, Bees ley ED. Congo red agar plating medium for detecting pigmentation in Pasteurella pestis. Applied Microbiology, 18, 834, 1969. 69 Carniel E, Mazigh D, Mollaret HH. Expression of iron regulated proteins in Yersinia species and their relation to virulence. Infection and Immunity, 55, 277, 1987. 70 Stuart SJ, Prpic K, Robins-Browne RM. Production of Aerobactin by some species of the genus Yersinia. Journal of Bacteriology, 166, 1131, 1986. 71 Portnoy DA, Moseley SL, Falkow S. Characterization of plasmids and plasmid-associated determinants of Yers inia enterocoli t ica pathogenesis. Infection and Immunity, 31, 775, 1981. 72 Ben-Gurion R, Shafferman A. Essential virulence determinants of different yersinia species are carried on a common plasmid. Plasmids, 5, 183, 1981. 73 Portnoy DA, Moseley SL, Falkow S. Characterization of plasmids and plasmid-associated determinants of Yers inia enterocolitica pathogenesis. Infection and Immunity, 31, 775, 1981. 74 Martinez RJ. Plasmid-mediated and temperature-regulated surface properties of Yers inia enterocolitica. Infection and Immunity, 41, 921, 1983. 75 Heesmann J, Keller C, Morawa R, Schmidt N, Siemans HJ, Laufs R. Plasmids of human strains of Yers inia enterocolitica: Molecular relatedness and possible importance for pathogenesis. Journal of Infectious Disease, 147, 107, 1983. 76 Lian CJ, Pai CK. Inhibition of human neutrophil chemiluminescence by plasmid mediated outer membrane proteins of Yers inia  enterocolitica. Infection and Immunity, 49, 145, 1985. 77 Hill WE, Payne WL, Aulisio CCG. Detection and enumeration of virulent Yers inia enterocolitica in food by DNA colony hybridization. Applied and Environmental Microbiology, 46, 636, 1983. 78 Ginzburg AL. Movable genetic elements of pathogenic Yersinia bacteria. Genetika, 23, 581, 1987. 79 Newby TJ. Protective Immune Responses in the Intestinal Tract. In TJ Newby and CR Stokes (ed) Immune Responses of the Gut. CRC Press. 143, 1984. - 36 -80 Mayrhofer G. Physiology of the Intestinal Immune System. In TJ Newby and CR Stokes (ed) Immune Responses of the Gut. CRC Press. 1, 1984. 81 Stokes CR. Induction and Control of Intestinal Immune Responses. In TJ Newby and CR Stokes (ed) Imm Responses of the Gut. CRC Press. 97, 1984. 82 Granfors K, Toivanen A. IgA-Anti-Yersinia Antibodies in Yersinia Triggered Reactive Arthritis. Annals of the Rheumatic Diseases. 45, 561, 1986. 83 Gronblad EA, Makela 0. salivary and serum antibodies in patients with Yersinia enterocolitica infection. Microbial Pathogenesis, 1. 565, 1986. 84 Larsen JH, Hartzen SH, Parm M. The determination of specific IgA-antibodies to Yers inia enterocolitica and their role in enteric infections and their complications. Acta Pathologica Microbiologica et Immunologica Scandinavica Section B, 93, 337, 1985. 85 Heesmann J, Eggers CH, Schroder J, Laufs R. Serological diagnosis of Yersiniosis by the immunoblot technique using plasmid-encoded antigens of Yersinia enterocolitica. In C Simons P Wilkinson (eds) Diagnosis of Infectious Diseases-New Aspects. Schattauer, New York, 69, 1986. 86 Bottone EJ, Sheehan DJ. Yers inia enterocolitica: Guidelines for serologic diagnosis of human infections. Reviews of Infectious Diseases, 5, 898, 1983. 87 Granfors K. Measurement of immunoglobulin M (IgM), IgG, and IgA antibodies against Yers inia enterocolitica by enzyme-linked immunosorbent assay: Persistence of serum antibodies during disease. Journal of Clinical Microbiology, 9, 336, 1979. 88 Winblad S. Immune responses to Yersinia and Pasteure^a. In NR Rose H Friedman (eds) Manual of Clinical Immunology 2 ed, American Society for Microbiology, Washington, 474, 1980. 89 Reed WP, Williams RC. Intestinal Immunoglobulins in Shigellosis. Gastroenterology, 61, 35, 1971. 90 Monteiro E, Fossey J, Shiner M. Antibacterial Antibodies in Rectal and Colonic Mucosa in Ulcerative Colitis. The Lancet, Feb 6, 249, 1971. - 37 -91 Cancellieri V, Fara GM. Demonstration of specific IgA in human feces after Immunization with Live Ty21a Salmonella typhi Vaccine. The Journal of Infectious Diseases 151, 482, 1985. 92 Corridori S, Negretti F. Immunomicrobiological Monitoring of Orovaccinal Therapy in Acute Intestinal Infections. International Symposium on Enteric Infections in Man and Animals; Standardization of Immunological Procedures, Dublin, Irelend, 1982, Developemental Biology Standards 53, 285, 1983. 93 Sarasombath S, Banchuin N, Sukosol T, Rungpitarangsi, Manasatit. Systemic and intestinal immunities after natural typhoid infection. Journal of Clinical Microbiology, 25, 1088, 1987. 94 Mehlman IJ, Aulisio C, Romero A, Sanders AC. Recovery of potentially enteropathogenic Yersinia species from food and clinical specimens. In TA Roberts, G Hobbs, JHB Christian, & N Skovgaard (eds) Psychrotropic Microorganisms in Spoilage and Pathogenicity. Academic Press, London, 1982. 95 Swaminathan B, Harmon MC, Mehlman IJ. A review, Yersinia enterocolitica. Journal of Applied Bacteriology, 52, 151, 1982. 96 Higuchi K, Smith JL. A differential plating medium for the estimation of the mutation rate to avirulence. Journal of Bacteriology, 81, 605, 1961. 97 Maniatis T, Fritsch EF, Sambrook J (ed.) Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory. 1982. 98 Uinblad S, Nilehn B, Sternby N. Yersinia enterocolitica (Pasteurella X) in Human Enteric Infections. British Medical Journal, December, 1363, 1966. 99 Mulley AG. Screening for the healthy patients. Medical Clinics of North America, 71, 4, 625, 1987. 100 Kay BA, Uachsmuth K, Gemski P, Fee ley JC, Quan TJ, Brenner DJ. Virulence and phenotypic characterization of Yers inia  enterocolitica isolated from humans in the United States. Journal of Clinical Microbiology. 15, 1161, 1983. 101 Harrison PE, Banvard J. Coproanitbody Excretion During Enteric Infections. Science, Aug 29, 188, 1947. 102 Barksdale UL, Ghoda A, Okabe K. Coproagglutinins in Ulcerative Colitis. Journal of Infectious Disease. 89, 47, 1951. - 38 -103 Toivanen A, Granfors K, Lahesman-Rantala R, Leino R, Stahlberg T, Vuento R. Pathogenensis of yersinia-triggered reactive arthritis: Immunological, microbiological and clinical aspects. Immunological Reviews, 86, 47, 1985. 104 Hancock GE, Schaedler RU, MacDonald TT. Yersinia enterocolitica infection in resistant and susceptible strains of mice. Infection and Immunity, 53, 26, 1986. 105 Edwards WH. Edwards^nd Ewing's Identification of Enterobacteriacae. 4 ed Elsevier, New York. 461, 1986. 39 APPENDIX A EVALUATION OF AUTOAGGLUTINATION STANDARDS Value Autoagglutination + Number of Runs 10 10 Mean Value 0.0425 0.1268 Standard Error of the Mean 0.0156 0.0224 Upper Boundary 0.0737 Lower Boundary - 0.1729 Twice the Standard Error 0.0992 =0.10 of the Mean APPENDIX B MAGNESIUM OXALATE AGAR 880 ml Distilled Water 40 gm Blood Agar Base Bo i1, Coo1, and Add: 40 ml MgCl2 - 6H20 80 ml Sodium Oxalate Autoclave 20 min at 121°C. CRAMP AGAR 15 gm Noble Agar 2 gm Galactose 2 gm Casamino Acids 1000 ml Distilled Water 10 ml Congo Red (5,000 ug/ pH 7.0 Autoclave 25 min at 121°C - 41 -APPENDIX C  DIAGNOSTIC TEST EVALUATION 2x2 DATA TABLE SAMPLE CALCULATIONS Gold Standard Value /a+b /d+c Gold Standard Value a = # True Positives b = # False Positives c = # False Negatives d = # True Negatives + Diagnostic + a b Test Value - c d Positive Predictive Value = a Negative Predictive Value = d Sensitivity = a/a+c Specificity = d/d+b EXAMPLE: 2x2 DATA TABLE FOR SIGA ASSAY Diagnostic +12 2 Test Value - 0 14 Positive Predictive Value = 12/12+2 = 85.7% Negative Predictive Value = 14/14+0 = 100% Sensitivity = 12/12+0 = 100% Specificity = 14/14+2 = 87.5% TABLE I BIOCHEMIGAL CHAMCTERISTTCS OF YT^RSINIA SPECIES ^21'105^ Characteristic Y.pestis Y.pseudo- Y.entero- Y. inter- Y.freder- Y.kris Y. Y. tuiberculosis colitica media iksenii tensenii ruckeri aldovae Motility - + + + + + V + Lysine - - - - - - + -Decarboxylase Ornithine - - + + + + + + Decarboxylase Urease - + + + + + - + Citrate - - - + V - + + Voges-Proskauer - - + + + - V + Indole - - V + + V - -Rhamnose - + - + + - - + Sucrose - - + + + - - -Cellobiose - - + + + + - -Melibiose V + - + - - - -Sorbose - - + + + + - -Sorbitol — — + + + + — + TABLE II BICCHEMICAL FEATURES OF YERSINIA BIOGROUPS^64^ Criaracteristic Biogroup 1 "American" Strains 2 3 1 2 SA^ 4 5 Lipase Activity (Hydrolyzes Tween) + + - - - - '-Esculin Hydrolysis - — — "/+ — — Indole Production + + + — — — — Acid Production From Xylose + + + + + — v Voges-Proskauer Test + + + + — + + Ornithine [Decarboxylase, Acid Production from Trehalose, and Nitrate Reduction + + + + + + — Pymzinaraidase Activity + + 1 - Positive for acid production from sorbose and inositol 2 - Negative for acid production from sorbose and inositol - 44 -TABLE III ~ SUMMARY OF PATIENT AND ORGANISM CHARACTERISTICS Patient Species Biotype Synptoms IgA Mouse Calcium Congo Red Autoagglu Plasmid Other # Serotype Response Virulence Dependance Absorption -tination Profile Orqanism 1 Y.enterocolitica 1,0:6,31 Acute Diarrhea ++ + + D.fragilis 2 Y.enterocolitica 1,0:6,36 Abdcminal Pain ++ + - — + + -3 Y.enterocolitica 1,0:7,13 Acute Diarrhea ++ + - - - - -4 Y.enterocolitica 4,0:3 Acute Diarrhea ++ + + + + + -5 Y.enterocolitica 4,0:3 Acute Diarrhea ++ + + + + — 6 Y.enterocolitica l,nt Acute Colitis ++ + — — — - — 7 Y.enterocolitica l,nt Acute Diarrhea ++ + • — — — - — 8 Y.enterocolitica 1,0:5 Crohn's Disease ++ - — — - p — — 9 Y.enterocolitica 1,0:4,32 Acute Diarrhea + - + +  • + B.harunis 10 Y.enterocolitica 1,0:6,30 Acute Colitis + - - — _ i — C.diff.Toxin 11 Y.enterocolitica 1,0:7,8 Fever NYD + - — — — — — 12 Y.enterocolitica 1,0:7,13 Crohn's Disease + - + — — + — 13 Y. enterocolitica 1,0:36 Weight Loss + - — + + — B.hcminis 14 Y. enterocolitica 1,0:41,43 Abdominal Pain + - — — + — — 15 Y. enterocolitica l,nt Rectal Bleeding + - — + — — D. fragilis 16 Y. enterocolitica 1,0:5 Abdominal Pain - — — — • — — 17 Y. errterocolitica 1,0:7,8 Abdominal Pain - — + — ; — Salmonella 18 Y.enterocolitica 1,0:5,27 Septicaemia ND + + + — + — 19 Y. frederiksenii ND Acute Diarrhea ++ + + + + + — 20 Y.f rederiksenii ND Acute Diarrhea ++ + - + - + — 21 Y.f rederiksenii ND Hyporat^remia ++ + + + + + — 22 Y.f rederiksenii ND Acute Diarrhea -H- + — — — — 23 Y.frederiksenii ND Acute Diarrhea ++ + + + + + — 24 Y.frederiksenii ND Acute Diarrhea + - + + - + — 25 Y.f rederiksenii ND Acute Diarrhea - + — — + — 26 Y. frederiksenii ND Abdominal Pain - + — - + Trophozoites 27 Y.frederiksenii ND Chronic Diarrhea - - — - — — 28 Y.frederiksenii ND Crohn's Disease - - + — — — 29 Y. frederiksenii ND Reactive Arthritis ND + + + + + — 30 Y.frederiksenii ND Acute Diarrhea ND + - + — — B.hcaninis 31 Y.frederiksenii ND Acute Diarrhea ND + - + + - — 32 Y.frederiksenii ND AIDS N + + + - + I.butchlii 33 Y.frederiksenii ND Acute Colitis ND + - + - + — 34 Y.f rederiksenii ND Acute Diarrhea ND - - + + + B.haninis 35 Y.frederiksenii ND Acute Diarrhea ND - - + — — S.ireuchin 36 Y.frederiksenii ND Abdominal Pain ND - - — + + — - 45 -TABLE III - SUMMARY OF PATIENT AND ORGANISM CHARACTERISTICS - CONT'D Patient # Species Biotype Serotype Symptoms IgA Response Mouse Calcium Congo Red Virulence Dependance Absorption Autoagglu Plasmid -tination Profile Other Organism 37 Y.frederiksenii ND Acute Diarrhea ND + _ _ 38 Y.frederiksenii ND Abdorrtinal Pain ND - + + + + — 39 Y.frederiksenii ND Chronic Diarrhea ND - - + — + — 40 Y.frederiksenii ND Chronic Diarrhea ND - - + + + — 41 Y.frederiksenii ND Abdominal Pain ND • - + + + + B.hcminis 42 Y.frederiksenii ND Acute Diarrhea ND - - + + — Isospora 43 Y.frederiksenii ND Abdominal Pain ND - + + + + — 44 Y.frederiksenii ND Acute Diarrhea ND - - + — — Giardia 45 Y.frederiksenii ND Acute Diarrhea ND - - — —; + — 46 Y.frederiksenii ND Acute Diarrhea ND - + — + + — 47 Y. kristensenii ND Acute Colitis ++ - + — + + — 48 Y.kristensenii ND Acute Diarrhea ND + + + + + B.hcminis 49 Y. kristensenii ND Abdominal Pain ND - + — — i + — 50 Y. kristensenii ND Acute Diarrhea ND - + + + + — 51 Y.kristensenii ND Rectal Bleeding ND - - + — + C. jejuni 52 Y.kristensenii ND Acute Diarrhea ND - - — — • + — 53 Y.kristensneii ND AMcminal Pain ND - - + —: — — 54 Y.kristensenii ND Abdominal Pain ND - + — + + — 55 Y.kristensenii ND Acute Appendicitis ND - - + — + S.typimuriui 56 Y.intermedia ND Nausea ND - - + — + — 57 Y.intermedia ND Acute Diarrhea ND - - + — + — 58 Y.intermedia ND Acute Colitis ND - - + + + — 59 Y.irrtermedia ND Acute Diarrhea ND - + + — + — 60 Y.intermedia ND Abdominal Pain ND - - + - + D. fragilis 61 Y.intermedia ND Acute Diarrhea ND - - — — + — 62 Y.intermedia ND Acute Diarrhea ND - - + — + — 63 Y. intermedia ND Abdominal Pain ND - - + — + — 64 S. enter ididis ND Acute Diarrhea -* ND ND ND ND ND — 65 C. jejuni ND Acute Diarrhea -* ND ND ND ND ND — 66 C. jejuni ND Acute Diarrhea -** ND ND ND ND ND — 67 C. jejuni ND Acute Diarrhea -4c* ND ND ND ND ND — 68 C.difficile ND Acute Diarrhea -** ND ND ND ND ND — 69 E.coli 0157:H7 Acute Diarrhea -** ND ND ND ND ND — 70 Negative ND Asymptomatic -* ND ND ND ND ND — 71 Negative ND Asymptomatic -* ND ND ND ND ND — 72 Negative ND Acute Colitis -* ND ND ND ND ND — * - Tested against yersinia isolates from patient # 1. ** - Tested against yersiniae isolates from patients # 2, 4, 11, 18. - 46 -TABLE IY RELATIONSHIP BETWEEN IN VITRO YERSTNTA  VIRULENCE ASSOCIATED ASSAYS AND ACUTE ENTERIC DISEASE (All Isolates Included) (# True Positives/ Test Positives) Species Autoagglu Congo Red Calcium Plasmid SIgA Mouse -tination Absorption Dependance Carriage Response Virulence Y.enterocolitica Y.frederikseni i Y.kri stensenii Y.intermedia Total 3/6 7/14 3/4 1/1 4/8 13/21 3/5 4/7 4/5 5/12 3/5 1/1 4/6 9/19 5/8 5/8 7/8 4/5 1/1 ND 6/8 7/10 1/1 ND 14/25 24/41 13/23 23/41 12/14 14/19 Value {%) (56.0) (58.5) (56.5) (53.7) (85.7) (73.7) - 47 -TABLE V RELATIONSHIP BETWEEN IN VITRO YERSINIA  VIRULENCE ASSOCIATED ASSAYS AND ACUTE ENTERIC DISEASE (Excluding Strains Isolated From Patients With Inflammatory Bowel Disease) (# True Positives/ Test Positives) Species Autoagglu Congo Red Calcium Plasmid SIgA Mouse -tination Absorption Dependance Carriage Response Virulence Y.enterocolitica 3/6 4/8 3/4 3/5 5/6 5/7 Y.frederikseni i 7/14 11/19 5/12 8/19 4/5 6/9 Y.kristenseni i 2/3 3/5 2/4 4/7 0/1 1/1 Y.intermedia ND 3/6 VI 4/7 ND ND Total 12/23 21/38 11/21 19/37 9/12 12/17 Positive Predictive Value (96) (52.2) (55.3) (52.3) (51.4) (75.0) (70.6) - 48 -TABLE VI RELATIONSHIP BETWEEN IN VITRO YERSINIA  VIRULENCE ASSOCIATED ASSAYS AND MOUSE VIRULENCE (All Isolates Included) (# True Positives/ # Test Positives) Species Autoagglu -tination Congo Red Absorption Calcium Dependance Plasmid Carriage SIgA Response Y.enterocolitica 4/6 4/8 3/5 4/6 7/8 Y.frederiksenii 6/14 9/21 5/12 7/19 5/5 Y.kristenseni i 1/4 1/5 1/5 1/8 0/1 Y.intermedia 0/1 0/7 0/1 0/8 ND Total 11/25 14/41 9/23 12/41 12/14 Positive Predictive Value (%) (44.0) (34.1) (39.1) (29.3) (85.7) - 49 -TABLE VII RELATIONSHIP BETWEEN IN VITRO YERSINIA  VIRULENCE ASSOCIATED ASSAYS AND MOUSE VIRULENCE (Excluding Strains Isolated From Patients With Inflammatory Bowel Disease) (# True Positives/ # Test Positives) Species Autoagglu Congo Red Calcium Plasmid SIgA -tination Absorption Dependance Carriage Response Y.enterocolitica 4/6 4/8 3/4 4/5 6/6 Y.frederikseni i 6/14 8/19 5/12 6/18 5/5 Y.kristenseni i 1/3 1/5 1/4 1/7 ND Y.intermedia ND 0/6 0/1 0/7 ND Total 11/23 13/38 9/21 11/37 11/11 Positive Predictive Value (%) (47.8) (34.2) (42.9) (29.7) (100) - 50 -TABLE VIII COMPARISON OF POSITIVE VIRULENCE ASSOCIATED CHARACTERISTIC EXPRESSION  BETWEEN IGA POSITIVE (2+) AND NEGATIVE (1+ OR -1 STRATNS (# of Isolates With Positive Reaction)(%) IgA # Org- Autoagglu Congo Red Calcium Plasmid Mouse Response anisms -tination Absorption Dependance Carriage Virulence 2+ 14 9(64.3) 6(42.9) 7(50 0) 7(50.0) 12(85.7) 1+/0 14 2(14.3) 5(35.7) 7(50.0) 5(35.7) 0(0) - 51 -TABLE IX COMPARISON OF MOUSE SEROLOGICAL RESPONSE  TO HUMAN INTESTINAL RESPONSE SIgA Species Serotype Murine Serum Response Antibody Titre 2+ Y.enterocolitica 0:6,36 1/1024 2+ Y.enterocolitica 0:7,13 1/2048 1+ Y.enterocolitica 0:5 1/1 0 Y.frederiksenii ND 0 - 52 TABLE X COMPARISON OF DAYS COLD ENRICHMENT  TO RECOVERY OF SIGA STIMULATING ORGANISMS # Days IgA Stimulators IgA Non-Stimulators Recovered Recovered 0 5 2 1-7  7 8-14 4 5 - 53 -ACKNOWLEDGEMENTS This work was supported by a grant from the British Columbia Health Care Research Foundation (#5-52403). For his valuable advice and assistance in many aspects of this thesis I would like to thank Dr.M.A. Noble. For their technical assistance in yersiniae isolation, serotyping, and fluorescence photography I also thank R.L. Barteluk, S. Toma, and A.W. Vogl respectively. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items