Jul 1, 1987 - organisms, including autoagglutination, calcium dependence, plasmid carriage, and absorption of Congo red, mouse virulence, and clinical ...
Vol. 26, No. 2
JOURNAL OP CLINICAL MICROBIOLOGY, Feb. 1988, p. 287-292 0095-1137/88/020287-06$02.00/0 Copyright C 1988, American Society for Microbiology
Human Coproantibody Secretory Immunoglobulin A Response to Yersinia Species K. M. FLETCHER, C. M. MORRIS, AND M: A. NOBLE* Department of Pathology, Division of Medical Microbiology, University of British Columbia, Vancouver, British Columbia V6T 2BS, Canada Received 1 July 1987/Accepted 5 October 1987
A semiquàntitative indirect immunofluorescence assay to detect coproantibody secretory IgA (SIgA) was established to investigate the human intestinal immune response to Yersinia species. This assay was based on microagglutination of SIgA in fecal specimens with the patient's homologous organism. Two populations of patients were defined, those who produced an agglutinating (2+) SIgA response and those who did not. A comparison between SIgA production and standard in vitro virulence-related characteristics of infecting organisms, including autoagglutination, calcium dependence, plasmid carriage, and absorption of Congo red, mouse virulence, and clinical presentation, was performed. A positive (2+) SIgA result was associated with acuté enteric illness (positive predictive value, 78.6%) and mouse virulence (positive predictive value, 85.7%). When patients With active inflammatory bowel disease were excluded, the positive predictive value of SIgA for mouse virulence and acute enteric disease became 100%. In addition to strains of Yersinia enterocolitica 4,0:3, strains generally characterized as nonpathogenic, including Yersiniafrederiksenii, were found to be associated with acute disease, mouse virulence, and stimulation of SIgA. The indirect immunofluorescence assay for detection of SIgA response appears to be a useful indicator of pathogenic strains of yersiniae recovered from énteric specimens.
Yersinia enterocolitica is a causative agent of a variety of enteric disorders, including acute diarrhea, gastroenteritis, and pseudoappendicitis. Yersiniosis is generally self-limited, but Yersinia-related chronic diarrhea and postreactive syndromes (i.e., arthritis and erythema nodosum) have also been documented (8, 17, 18>. Other Yersinria species, Yersiniafrederiksenii, Yersinia kristensenii, and Yersinia.iintermedia, have been considered enterically avirulent in humans (3, 5, 30). Yersiniae have been shown to have toxigenic properties and cellular adhesion factors that are associated with plasmid carriage (16, 23). Virulence of Yersinia species in humans has been associated with a number of in vivo and in vitro bacterial characteristics, including deep organ infiltration in orally inoculated mice (1), dependence on calcium for growth (11), ability to absorb a hemin-type dye, Congo red (24), and carriage of a 40- to 48-megadalton plasmid (31). Each characteristic has some reliability in identifying organisms with potential virulencé in humans (25), but none considers the human response to the organism. Several recent studies have documented a rise in serum titér of Yersinia-specific antibody early in infection (8, 9, 15), primarily accounted for by immunoglobulin A (IgA) class antibodies, in particular the 11S or dimer form normally found in secretions (8). Since methods for detecting IgA in serum are not entirely reliable (15) and no criteria exist for ascribing significance to an increased IgA titer (4), the diagnostic applications of serum IgA levels are limited. In addition, serological analysis, either by agglutination or by enzyme-linked immunosorbent assay, requires a second postillness specimen that is not generally readily available. In 1971, Reed and Williams (26) documented the presence of secretory IgA (SIgA) class Shigella-agglutinating antibodies in feces from patients with recent-onset diarrhea. Although no mechanistic role was suggested, a good correla*
tion between coproantibody production and stage of enteric illness was clearly demonstrated. A number of investigators have since reported coproantibody responses to enteric organisms in rectal mucosa and feces by using indirect immunofluorescence assays (6, 7, 21). This study observes that a semiquantitative indirect immunofluorescence assay can determine human intestinal immune response to Yersinia species and compares this response with both in vivo and in vitro virulence-associated characteristics and human disease presentation.
MATERIALS AND METHODS Specimens and organisms. Fecal specimens were received in the microbiology laboratory of the university hospital în Cary-Blair transport medium or sterile containers and proëlessed as described previously (22). Fecal samples were maintained at -4°C until bacteriology, including cold entichment, was completed. When a yersinia strain was recovered, the clinical history of the patient was reviewed and the feéés wete processed for coproantibody SIgA determination. Crganismns were maintained at -70°C in glycerol and dimethyl sulfoxide until assayed for virulence. Virulence-associated in vitro assays. Before organisnis were tested for virulence-associated characteristics, they were retrieved and transferred three times at 25°C on blQod agar plates (PML Microbiologicals). The virulence assays performed included calcium dependence (12), Congo red
absorption (24), plasmid profiling (19), autoagglutination (14), and mouse virulence (1). The first three assays were performed as described previously (22). Standards for plasmid profiling included T4, T5, T7, and lambda phage DNA (Sigma) and a strain known to carry the 40-megadalton plasmid (701S; C. Pai, Calgary, Alberta, Canada). Those strains with plasmids banding in the same region as the control strain were considered to be plasmid carrying. Autoagglutination. For autoagglutination, the method of Laird and Cavanaugh (14) was modified to take into account
Corresponding author. 287
288
J. CLIN. MICROBIOL.
FLETCHER ET AL.
bacteria which appeared to have significantly autoagglutinated at 25°C but had not completely settled on the bottom of the tube. Organisms were harvested from an overnight tryptone-yeast extract broth culture, washed, and suspended in phosphate-buffered saline (PBS). The culture was divided and incubated for 18 h at 25 or 37°C. Both tubes were observed for growth, but were not agitated to prevent resuspension of bacteria into the medium. The optical density (00) of the unagitated tubes was read in the middle of the tube at 460 nm (Beckman 25). Known autoagglutinationpositive and -negative organisms were included in 10 experimental runs, and the difference between twice the standard error of the mean values was calculated to be 0.1 OD units. Strains were interpreted as autoagglutination positive if the OD at 25C exceeded that at 37°C by 0.1 unit. Mouse virulence. The assay for mouse virulence was described by Bakour et al. (1). Swiss mice (Charles River Breeding Laboratories) were dehydrated for 24 h and allowed to drink ad libitum from a 0.1% peptone broth containing 109 organisms per ml for 36 h. Mice were sacrificed, and the spleens were homogenized and plated on CIN Yersinia selective agar (PML Microbiologicals) (24 h at 37°C). Colonies with the characteristic Yersinia appearance (27) were characterized further on triple sugar iron and urea slants (PML Microbiologicals). A strain was defined as mouse virulent if more than 5 x 102 organisms were recovered from the homogenized splenic tissue. Human coproantibody SIgA response. A semiquantitative
indirect immunofluorescence antibody assay was established to detect intestinal SIgA response to yersinia. Fecal samples were stored at -4°C until a Yersinia species was recovered. Fecal specimens were diluted 1:1 with PBS and centrifuged for 30 min at 15,000 x g (Eppendorf centrifuge), and the supernatant was stored at -70°C. A 200-,ul amount of supernatant dilutions (1/10, 1/50, 1/100) was incubated with 40 ,uI of 109 homologous organisms per ml in PBS on cover slips (18 h at 25°C). Cover slips were air dried, dip washed in distilled water, and incubated with 120 ,uI of fluoresceinisothiocyanate-labeled 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; 1+, single organisms; 2+, homogeneous microagglutination (Fig. 1). Statistical analysis. Positive predictive value (P.P.) was calculated as the number of true positives, assessed by mouse virulence, divided by the number of test positives and multiplied by 100. RESULTS Figure 1 demonstrates typical fluorescence microscope fields for a 2+, a 1+, a negative, and a negative in-phase response. 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 (Fig. 1A). A 1+ result is defined as single fluorescent organisms
t E.
go g FIG. 1. Typical indirect immunofluorescence SIgA assay results. (A) Homogenous microagglutination (2+); (B) single organism (1+) and nonspecific background; (C) no specific fluorescence (-); (D) in-phase contrast, nonclumping organisms coating slide (-).
YERSINIA COPROANTIBODY SIgA
VOL. 26, 1988
and is indicative of a weak antibody response (Fig. 1B). When the cultured Yersinia species stimulated no antibody response, the field was apparently empty (Fig. 1C). This absence of fluorescence cannot be accounted for by an absence of bacteria, as organisms were detected in phase contrast of the same field (Fig. 1D). Supernatant-only and organism-only control slides showed some homogeneous background fluorescence, but no organismlike structures. Fecal supernatants from diarrheic patients infected with Salmonella species (patients 29 to 31), Campylobacter species (patients 32 to 37), Clostridium species (patient 38), Shigella species (patient 39), or Escherichia coli (patient 40), and from asymptomatic, pathogen-negative patients (patients 41 to 43) were examined for cross-reactive agglutina-
289
tion with yersiniae. No specific agglutinating cross-reactivity found; however, some nonagglutinating halo-type staining of single organisms similar to the 1+ reaction described above was detected when supernatants were tested against these control organisms. Therefore, for the purpose of interpretation, an agglutinating reaction was considered essential for positivity in the SIgA assay. Table 1 documents a positive correlation between agglutinating (2+) SIgA response and acute disease (P.P., 78.6%). Exceptions include patients 1, 6, and 14, who suffered inflammatory bowel disease yet elicited a strong SIgA response (see Discussion), and patients 15, 16, 20, 21, 22, 23, and 28, who presented with acute diarrhea yet had a low or absent SIgA response. Patient 15 was shown to be positive
was
TABLE 1. Summary of patient and organism characteristics Patient
Biotype,
no.
Species
serotypeb
Symptoms
1 2 3 4 5 6 7 8 9 10
Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. frederiksenii Y. frederiksenii Y. frederiksenii Y. frederiksenii Y. frederiksenii Y. kristensenii Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. enterocolitica Y. frederiksenfi Y. frederiksenii Y. frederiksenii Y. enterocolitica Y. enterocolitica Y. frederiksenii Y. frederiksenii S. enteritidis S. infantis S. typhimurium C. jejuni C. jejuni C. jejuni C. jejuni C. jejuni C. jejuni C. difficile
1, 0:5 1, 0:6,30 1, 0:6,31 1, 0:7,13 1, nt 1, nt 4, 0:3 4, 0:3 ND ND ND ND ND ND
Crohn's disease Abdominal pain Acute diarrhea Acute diarrhea Acute diarrhea Acute colitis Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute colitis Acute diarrhea Acute colitisd Fever NYD Crohn's disease Abdominal pain Weight lossd Rectal bleeding' Acute diarrhea Acute diarrhea Crohn's disease Abdominal pain Abdominal pain Abdominal pain Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Acute diarrhea Asymptomatic Asymptomatic Acute colitis
il
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
1,0:4,32 1,0:6,30 1,0:7,8 1,0:7,13 1, 0:41,43 1, 0:36 1, nt ND ND ND
S. sonnei E. coli
1,0:5 1, 0:7,8 ND ND ND ND ND ND ND ND ND ND ND ND ND 0157:H7
Negative Negative Negative
ND ND ND
a Only the SIgA assay was done for patients 29 to 43. nt, Not typeable; ND, not determined. C Patient also culture positive for Clostridium difficile toxin. d Patients also culture positive for Blastocystis hominis. e Patient also culture positive for Dientamoeba fragilis. f Tested against Yersinia isolate from patient 3. 9 Tested against Yersinia isolate from patient 7. h Tested against Yersinia isolates from patients 2, 7, and 17 and
SIgA
Mouse
Calcium
response
virulence
dependence
2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+ 2+
-
+ +
+
+ +
1+
1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+
-
_
Congo red absorption
-g
-g
_g _g
-f -f
2+f
b
an
0:5,27 Yersinia strain.
Cold enrichment
(days)
-
-
0
+
+
2
-
+
+
-
14
_ _ _
-
-
4
-
-
7
-
-
O
+ + +
+ + + +
+ + + +
+ + + +
O
+
_
O
1 14
+
-
+
O
+
+
+
+
14
+
-
O
+ +
-
+
+
-
+ +
16 1
_ _
_ _
-
-
7
-
-
7
-
+
14 14
_
_
+
_ + +
_
_
_h _h
Plasmid profile
_
+
-
+
+
-
6
+
-
-
-
-
+
-
+ +
14 14 7
+
-
-
1
-
-
-
14
-
-
+
O O
-
1
-
+
-f
tination
_ _ _
+
-f
Autoagglu-
_
_ -
290
J. CLIN. MICROBIOL.
FLETCHER ET AL. TABLE 2. Relationship between in vitro assays and mouse virulence No. of true-positives/no. testing positive Species
Y. enterocolitica Y. frederiksenùi Y. kristensenii P.P. (%)
Autoagglutination
Congo red
Calcium
Plasmid
absorption
dependence
carriage
4/6 4/4 0/1 72.7
3/6 4/6 0/0 58.3
2/5 3/6
3/5 4/7
7/8
0/1 41.7
0/1 53.8
0/1 85.7
for Clostridium difficile toxin, although the bacterium was not cultured; patients 16 and 20 carried Blastocystis hominis, and patient 21 carried Dientamoeba fragilis. P.P. values for mouse virulence were as follows: autoagglutination, 72.7%; calcium dependence, 41.7; Congo red absorption, 58.3%; and plasmid carriage, 53.8% (Table 2). These results are consistent with those of Prpic et al. (25), who also noted that autoagglutination was the most reliable of the in vitro virulence-associated characteristics as a predictor of mouse virulence. Agglutinating (2+) reactivity in the coproantibody SIgA assay predicted mouse virulence better than any of the established virulence-related assays (P.P., 85.7%). When organisms recovered from patients with inflammatory bowel disease were excluded, the P.P. of SIgA for mouse virulence became 100%. Organisms cultured from SIgA-positive (2+) patients had a higher incidence of virulence-related characteristics, although SIgA response itself was not associated with any single virulence-related characteristic. DISCUSSION This study confirms the previous findings of Kay et al. (13) that no single virulence-related characteristic correlates well with Yersinia virulence as assessed by the mouse model and also demonstrates that strains of Yersinia associated with human coproantibody SIgA stimulation are likely to be mouse virulent. In addition, the study demonstrates that strains of Y. frederiksenii can be mouse virulent and immunostimulatory in the human gastrointestinal tract. Y. frederiksenii, Y. kristensenii, and Y. intermedia were originally associated only with skin and wound infections or found as environmental isolates (3, 5, 30); however, they have been recovered from a number of patients presenting with acute diarrhea (22, 27). Prpic et al. (25) found that some Y. enterocolitica-like organisms expressed some of the in vitro virulence-associated characteristics but were still avirulent in mice. This study also documents the expression of one or more in vitro virulence-associated characteristics in Y. frederiksenii and Y. kristensenii species. In addition, the existence of Y. frederiksenii and atypical Y. enterocolitica strains which are mouse virulent and immunostimulatory in humans was observed. This is consistent with our clinical observation that strains of Y. enterocolitica other than 0:3, 0:5,27, etc., may be pathogenic in humans. In addition, 58.3% of the mouse-virulent, antibody-inducing yersiniae were recovered only after cold enrichment (four recovered after 7 days, three recovered after 14 days), further supporting the importance of cold enrichment in the recovery of some enterically pathogenic yersiniae (22). SIgA is the best suited of the immunoglobulin to deal with antigenic material in the gut lumen, as it resists proteolysis by intestinal enzymes, has an affinity for the mucosa, and is the predominant immunoglobulin produced in the gut-associated lymphoid tissue in humans (19). As it can both agglutinate organisms and prevent their adherence, SIgA is
SIgA response
5/5
thought to be the most important resistance mechanism for bacterial elimination from the intestine (28). The existence of coproantibody SIgA in fecal specimens following enteric infections was documented as early as 1947, when Harrison and Banvard (11) reported the detection of agglutinating antibody in fecal specimens from patients with both acute and chronic bacillary dysentery. Since then several studies have reported the presence of high coproantibody titers in the feces (2) and intestinal mucosa (21) of patients with ulcerative colitis and in volunteers orally immunized with live Salmonella typhi vaccine (6, 7). In the current investigation, the presence of high SIgA titers in fecal specimens has been clearly associated with acute enteric illness and carriage of mouse-virulent organisms. Of note, when patients with inflammatory bowel disease were excluded because their underlying illness created confounding data, the P.P. of SIgA for mouse virulence and acute enteric disease became 100%. In addition, this association was not found to be dependent on which species of Yersinia was isolated. In 1951, Barksdale et al. (2) noted that certain strains of coliform bacilli contain antigenic components common to certain Shigella and Salmonella species. Such cross-reactions could conceivably limit the diagnostic value of a coproantibody assay, and therefore appropriate controls were included in this study. Since specimens from patients who are culture negative or positive for an enteropathogen, i.e., Salmonella species, Campylobacterjejuni, Clostridium difficile, Shigella sonnei, or E. coli, were negative, an agglutinating (2+) SIgA response appeared to be Yersinia specific. However, as the number of controls was low, no overall specificity could be 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 Yersinia strain. This was consistent with the findings of Barksdale et al. (2), who reported a patient with chronic ulcerative colitis from whom neither Shigella nor Salmonella organisms were isolated, yet whose coproantibodies agglutinated three Salmonella species and four or more Shigella species. Toivanen et al. (29) suggested that the 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. Monteiro et al. (21) suggested that strains of enteric bacteria which would produce only transient reactions in noncolitic patients may be able to elicit antibody production in mucosa of colitic patients. Further explanations include the possibility that a mouse-virulent organism could lose its 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 latter
YERSINIA COPROANTIBODY SIgA
VOL. 26, 1988
is supported by the documentation by Hancock et al. of mice naturally resistant to some yersiniae (10). For the aforementioned reasons, the coproantibody SIgA assay as outlined here should not be relied on to predict the animal virulence potential of Yersinia strains recovered from patients with known inflammatory bowel disease. Acute diarrhea, during which low (1+) or absent SIgA responses were seen, could represent carriage of yersiniae in the presence of another causal agent. This is supported by the non-mouse-virulent nature of the cultured Yersinia species. Four such patients carried bacterial or parasitic pathogens which could account for their symptoms. No additional organisms were cultured from the remaining patients; however, since electron microscopy of fecal samples was not done, virus-induced enteritis could not be ruled out. A number of patients who were culture positive for a mouse-virulent Yersinia strain carried a parasite, i.e., Blastocystis hominis or Dientamoeba fragilis. These patients may represent concurrent Yersinia and parasite infection, or infection with one organism predisposing the patient to the other; however, the sequence of infection could not be determined from a single fecal specimen. In these cases, the coproantibody SIgA assay may be of use in differentiating transient nonimmunostimulatory Yersinia isolates from those eliciting a strong intestinal immune response. Since the majority of SIgA-positive patients presented only with acute short-lived enteric illness, sequential stool specimens were not always available. However, in two cases that were followed, patients 3 and 6, the reaction changed from 2+ to 1 + as the stool changed from fluid to formed and the symptoms disappeared. This appears to support the relationship between SIgA and acuteness of disease in humans. In summary, a reproducible semiquantitative indirect immunofluorescence assay has been established which facilitates the performance of a serological assay for enteric yersiniosis with a single fecal specimen. A positive SIgA response predicted acute enteric illness and mouse virulence better than any of the established in vitro virulence-associated assays. ACKNOWLEDGMENTS This work was supported by a grant from the British Columbia Health Care Research Foundation (5-52403). For their valuable assistance in Yersinia isolation, serotyping, and fluorescence photography, we thank R. L. Barteluk, S. Toma, and A. W. Vogl, respectively.
6.
7.
8.
9. 10. 11.
12.
13.
14. 15.
16. 17.
18. 19. 20. 21. 22.
1.
2. 3.
4. 5.
LITERATURE CITED Bakour, R., G. Balligand, Y. Laroche, G. Cornelis, and G. Wauters. 1985. A simple adult-mouse test for tissue invasiveness in Yersinia enterocolitica strains of low experimental virulence. J. Med. Microbiol. 19:237-246. Barksdale, W. L., A. Ghoda, and K. Okabe. 1951. Coproagglutinins in ulcerative colitis. J. Infect. Dis. 89:47-51. Bercovier, H., J. Ursing, D. J. Brenner, A. G. Steingerwalt, G. R. Fanning, G. P. Carter, and H. H. Mollaret. 1980. Yersinia kristensenii: a new species of Enterobacteriaceae composed of sucrose-negative strains (formerly called atypical Yersinia enterocolitica or Yersinia enterocolitica-like). Curr. Microbiol. 4:219-224. Bottone, E. J., and D. J. Sheehan. 1983. Yersinia enterocolitica: guidelines for serologic diagnosis of human infections. Rev. Infect. Dis. 5:898-901. Brenner, D. J., H. Bercovier, J. Ursing, J. M. Alonso, A. G. Steingerwalt, G. R. Fanning, G. P. Carter, and H. H. Mollaret. 1980. Yersinia intermedia: a new species of Enterobacteriaceae
23.
24.
25. 26. 27. 28.
291
composed of rhamnose-positive, melibiose-positive, raffinosepositive strains. Curr. Microbiol. 4:207-212. Cancellieri, V., and G. M. Fara. 1985. Demonstration of specific IgA in human faeces after immunization with live Ty2la Salmonella typhi vaccine. J. Infect. Dis. 151:482-484. Corridori, S., and F. Negrett. 1983. Immunomicrobiological monitoring of orovaccinal therapy in acute intestinal infections. International symposium on enteric infections in man and animals: standardization of immunological procedures. Dublin, Ireland. Dev. Biol. Stand. 53:285-297. Granfors, K., and A. Toivanen. 1986. IgA-anti-yersinia antibodies in yersinia triggered reactive arthritis. Ann. Rheum. Dis. 45: 561-565. Gronblad, E. A., and O. Makela. 1986. Salivary and serum antibodies in patients with Yersinia enterocolitica infection. Microb. Pathol. 1:565-571. Hancock, G. E., R. W. Schaedler, and T. T. MacDonald. 1986. Yersinia enterocolitica infection in resistant and susceptible strains of mice. Infect. Immun. 53:26-31. Harrison, P. E., and J. Banvard. 1947. Coproantibody excretion during enteric infections. Science 29:188-190. Higuchi, K., and J. L. Smith. 1961. A differential plating medium for the estimation of the mutation rate to avirulence. J. Bacteriol. 81:605-608. Kay, B. A., K. Wachsmuth, P. Gemski, J. C. Feeley, T. J. Quan, and D. J. Brenner. 1983. Virulence and phenotypic characterization of Yersinia enterocolitica isolated from humans in the United States. J. Clin. Microbiol. 15:1161-1163. Laird, W. J., and D. C. Cavanaugh. 1980. Correlation of autoagglutination and virulence of yersiniae. J. Clin. Microbiol. 11:430-432. Larsen, J. H., S. H. Hartzen, and M. Parm. 1985. The determination of specific IgA-antibodies to Yersinia enterocolitica and their role in enteric infections and their complications. Acta Pathol. Microbiol. Immunol. Scand. Sect. B 93:337-339. Lee, W. H., P. P. McGrath, P. H. Carter, and E. L. Eide. 1977. The ability of some Yersinia enterocolitica strains to invade HeLa cells. Can. J. Microbiol. 23:1714-1722. Leino, R., and L. Kalliomaki. 1974. Yersiniosis as an internal disease. Ann. Intern. Med. 81:458-461. Leino, R., K. Granfors, T. Havia, R. Heinonen, M. Lampinen, and A. Toivanen. 1987. Yersiniosis as a gastrointestinal disease. Scand. J. Infect. Dis. 19:63-68. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual, p. 368-369. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Mayrhofer, G. 1984. Physiology of the intestinal immune system, p. 1-96. In T. J. Newby and C. R. Stokes (ed.), Local immune responses of the gut. CRC Press, Boca Raton, Fla. Monteiro, E., J. Fossey, and M. Shiner. 1971. Antibacterial antibodies in rectal and colonic mucosa in ulcerative colitis. Lancet i:249-251. Noble, M. A., R. L. Barteluk, H. J. Freeman, R. Subramaniam, and J. B. Hudson. 1987. Clinical significance ofvirulence-related assay of Yersinia species. J. Clin. Microbiol. 25:802-807. Okamoto, K., H. Ichikawa, Y. Kawamoto, A. Miyama, and S. Yoshi. 1980. Heat-stable enterotoxin produced by Yersinia enterocolitica isolated from patients. Microbiol. Immunol. 24:407-408. Prpic, J. K., R. M. Robins-Browne, and R. B. Davey. 1983. Differentiation between virulent and avirulent Yersinia enterocolitica isolates by using Congo red agar. J. Clin. Microbiol. 18: 486-490. Prpic, J. K., R. M. Robins-Browne, and R. B. Davey. 1985. In vitro assessment of virulence in Yersinia enterocolitica and related species. J. Clin. Microbiol. 22:105-110. Reed, W. P., and R. C. Williams. 1971. Intestinal immunoglobulins in shigellosis. Gastroenterology 61:35-45. Simmonds, S. D., M. A. Noble, and J. H. Freeman. 1987. Culture-positive Yersinia enterocolitica infections: a two year experience with 47 patients. Gastroenterology 92:112-117. Stokes, C. R. 1984. Induction and control of intestinal immune responses, p. 97-142. In T. J. Newby and C. R. Stokes (ed.),
292
FLETCHER ET AL.
Local immune responses of the gut. CRC Press, Boca Raton, Fla. 29. Toivanen, A., K. Granfors, R. Lahesman-Rantaia, R. Leino, T. Stahlberg, and R. Vuento. 1985. Pathogenesis of yersinia-triggered reactive arthritis: immunological microbiological and clinical aspects. Immunol. Rev. 86:47-70. 30. Ursing, J., D. J. Brenner, H. Bercouvier, R. Fanning, A. G. Steigerwalt, J. Brault, and H. H. Moliaret. 1980. Yersinia frede-
J. CLIN. MICROBIOL.
riksenii: a new species of Enterobacteriaceae composed of rhamnose-positive strains (formerly called atypical Yersinia enterocolitica or Yersinia enterocolitica-like). Curr. Microbiol. 4:213-217. 31. Zink, D. L., J. C. Feeley, J. G. Wells, C. Vanderzant, J. C. Vickery, and G. A. O'Donovan. 1980. Plasmid-mediated tissue invasiveness in Yersinia enterocolitica. Nature (London) 283: 224-225.
