Vol. 29, No. 9JOURNAL OF CLINICAL MICROBIOLOGY, Sept. 1991, p. 2002-20060095-1137/91/092002-05$02.00/0Copyright © 1991, American Society for Microbiology

Outbreak of Keratoconjunctivitis due to Salmonella weltevredenin a Guinea Pig Colony

M. JOHN ALBERT,* M. ANSARUZZAMAN, SHAH M. FARUQUE, KHALEDA HAIDER, FIRDAUSI QADRI,M. MOYENUL ISLAM, A. K. M. G. KIBRIYA, AND SAUL TZIPORI

International Centre for Diarrhoeal Disease Research, Dhaka, Bangladesh

Received 21 February 1991/Accepted 20 June 1991

The purpose of this report is to demonstrate that the ability to produce keratoconjunctivitis (KC) is a

property found in Salmonella weltevreden. This observation is contrary to previous reports that Salmonella spp.

do not produce KC. An outbreak of KC due to S. weltevreden occurred in a guinea pig colony, and the animalscarried the organism in the intestinal tract. The same Salmonella serotype that caused an epidemic of diarrheain humans and a routine laboratory isolate also possessed the ability to induce KC. Unlike Shigella spp. (theprototype organisms positive for KC), S. weltevreden induced KC and bound Congo red dye even when grownat 30°C. It invaded HeLa cells in culture but did not hybridize with a DNA probe for invasiveness of Shigellaspp. and enteroinvasive Escherichia coli even though it harbored plasmids. It was susceptible to all theantibiotics tested, was hydrophobic, and showed mannose-sensitive hemagglutination. It did not haveenterotoxic or cytotoxic activities.

On inoculation into the conjunctival sac of guinea pigs,members of the genus Shigella and the closely relatedenteroinvasive Escherichia coli produce keratoconjunctivitis(KC) (5, 13, 36). The ability to produce KC is correlated withthe ability of these organisms to produce invasive diarrhea(8, 18). The KC test was introduced by Sereny and is knownas the Sereny test (36). Although other enteric pathogenssuch as strains of Salmonella, Campylobacter, Aeromonas,and Yersinia also produce invasive diarrhea, reports to datesuggest that all except Yersinia spp. are uniformly negativein the Sereny test (8, 10, 11, 19, 28, 40). Yersinia enteroco-litica, however, produces only conjunctivitis and not KC(35). We report here an outbreak of naturally occurring KCdue to Salmonella weltevreden in the guinea pig colony ofthe animal house of the International Centre for DiarrhoealDisease Research, Dhaka, Bangladesh.

MATERIALS AND METHODSReference bacterial strains. One S. weltevreden strain

(MY8863) implicated in an outbreak of diarrhea in a Bangla-deshi village (unpublished data); one S. weltevreden strain(D21411) and one Shigella flexneri 2a strain (611), bothobtained from the routine clinical laboratory; and the non-pathogenic E. coli K-12 were included for comparison.Guinea pig colony. The guinea pig colony has approxi-

mately 150 animals housed in a separate room of the animalhouse. Between March and June 1990, 13 animals developedspontaneous KC.

Bacteriological investigation. The affected eyes of the ani-mals were swabbed and cultured on blood agar and MacCon-key agar. Gut contents of two animals that developed KCand three consecutive stool samples of all five animal han-dlers were cultured for enteric bacterial pathogens by stan-dard methods (42). Salmonella organisms were serotyped byO and H antisera (Pasteur Institute, Paris, France) bystandard methods (7).Sereny test. The organisms were grown on Trypticase soy

agar (BBL) overnight either at 37 or at 30°C. The growth was

* Corresponding author.

suspended in normal saline to 1010 organisms per ml, and0.02 ml of the suspension was dropped into the left eye ofeach guinea pig. Each organism was tested in two guineapigs, and the animals were observed for up to 7 days for thedevelopment of KC.

Histology. After development of KC, all animals weresacrificed. The eyeball and eyelid from one animal inocu-lated with S. weltevreden and those from another inoculatedwith S. flexneri 2a were removed and placed in neutralbuffered formalin. Hematoxylin- and eosin-stained sectionswere viewed under a light microscope for histopathologicchanges. Negative controls included the uninoculated righteyes from these animals.HeLa cell invasion assay. Invasion assay was performed as

described previously, with some modifications (39). Insteadof HEp-2 cells, HeLa cells were used. Bacterial cells weregrown to mid-exponential phase in Trypticase soy broth(BBL) at 37°C. Approximately 1 x 1010 to 2 x 1010 CFU wasadded to a HeLa cell monolayer (1 x 105 cells in a 6-ml vial[Kimble, Toledo, Ohio] containing Eagle's minimum essen-tial medium), centrifuged at 850 x g for 10 min, and thenincubated at 37°C for 3 h in a 5% C02-95% air atmosphere.The plates were then washed 10 times with phosphate-buffered saline (PBS) and incubated for an additional 2 h inminimum essential medium containing 100 ,ug of gentamicinper ml to kill extracellular bacteria. After this, the monolayerwas washed three times with PBS, and internalized bacteriawere released by lysis of the monolayer with a solutioncontaining 0.25% trypsin and 0.5% sodium desoxycholate indistilled water and quantified by plate count. The positiveand negative controls included were S. flexneri 2a and E. coliK-12, respectively. Each strain was tested in duplicatethrice, and the values were averaged.

Plasmid analysis. Bacterial plasmids were extracted by themethod of Birnboim and Doly (2), separated by agarose gelelectrophoresis, and stained by ethidium bromide as de-scribed previously (27).

Plasmid curing. Bacteria were treated with novobiocin (26)and sodium dodecyl sulfate (14), and colonies were exam-ined for plasmids as described above.DNA hybridization. The DNA probe used was constructed

2002

on February 23, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

KC DUE TO S. WELTEVREDEN 2003

from the invasive plasmid of S. flexneri 5 (M9OT) andconsisted of a 17-kb EcoRI digestion fragment of pRM17(37). The appropriate restriction fragment was purified asdescribed by Moseley et al. (29). The probe was labeled bynick translation (22) with [ac-32P]dCTP (Amersham Interna-tional plc., Aylesbury, Buckinghamshire, United Kingdom)and a nick translation kit (BRL). Colony blots were pre-pared, processed, and hybridized under stringent conditionsas described by Echeverria et al. (6).

Enterotoxin and cytotoxin production. The organisms weregrown in Trypticase soy broth at 37°C for 20 h, and thesupernatants were tested for heat-labile enterotoxin in Y-1adrenal tumor cells (33), heat-stable enterotoxin in sucklingmouse (4), and cytotoxin in HeLa cells (9).Congo red binding. Congo red binding of the organisms

was determined by streaking the organisms on Congo redagar and observing for development of pigmented coloniesas described previously (31).

Salt aggregation test. Fresh colonies of strains obtainedafter growth on Trypticase soy agar for 18 h at 37°C wereused to determine cell surface hydrophobicity by usingdifferent concentrations of ammonium sulfate (31).

Hemagglutination test. Bacteria grown in Trypticase soybroth were used for agglutination of a 2% suspension ofguinea pig erythrocytes (30). Inhibition of hemagglutinationwas tested by using 1% solutions of mannose, glucose,galactose, and N-acetyl neuraminic acid.

Antibiogram. The antibiotic susceptibilities of bacteriawere tested by the disk diffusion method (1).

RESULTS

From the eyes of all 13 animals which developed sponta-neous KC, a Salmonella sp. was isolated. Serotyping of theisolates revealed that all of them possessed an 0 antigen 3,10; a phase 1 H antigen, r; and a phase 2 H antigen, Z6. Thisantigenic composition is identical with that of S. weltevre-den, which belongs to serologic group E.The KC test was carried out with two guinea pig isolates,

GP-1 and GP-2, the human epidemic diarrheal isolateMY8863, the clinical laboratory isolate D21411, and S.flexneri 2a after the bacteria were grown at 37°C. TheShigella isolate produced conjunctivitis at 48 h and keratitisat 72 h. All the Salmonella isolates produced conjunctivitisat 48 h and keratitis at between 72 and 96 h. The KCproduced by strain GP-2 is shown in Fig. 1. When the KCtest was repeated with GP-2 and S. flexneri 2a grown at 30°C,GP-2 still produced KC, but S. flexneri 2a did not.

Histology of the eyes of the animals sacrificed 72 h afterinoculation with Salmonella spp. and Shigella sp. (afterdevelopment of KC) showed a similar picture. The eyelidsand palpebral conjunctivas showed severe acute inflamma-tion with polymorphonuclear cell infiltration, edema, andhyperemia. A similar picture was seen in the corneas.Shigella sp. produced a much more severe inflammatoryreaction in the cornea than did Salmonella spp. The histol-ogy of S. weltevreden KC is shown in Fig. 2.

All four Salmonella strains and the Shigella strain invadedHeLa cells. The survival of salmonellae in the gentamicin-HeLa cell assay ranged from 2 x 103 to 4 x 103 organismsper ml, whereas that of Shigella sp. was 4 x 104 organismsper ml. The survival rate of E. coli K-12 was 0.The guinea pig strains GP-1 and GP-2 possessed two

plasmids with molecular masses of 60 and 2.6 MDa, whereasthe isolates MY8863 and D21411 possessed only the largeplasmid. None of the strains hybridized with the invasive

FIG. 1. Development of KC in a guinea pig eye 72 h afterinoculation with guinea pig strain S. weltevreden GP-2 (A) comparedwith the eye of a normal uninoculated guinea pig (B). The whiteshadow in panel B is due to the reflection of light.

plasmid probe. Attempts to cure the plasmids were notsuccessful. None of the strains produced enterotoxins orcytotoxin. All of them were susceptible to ampicillin, tetra-cycline, chloramphenicol, streptomycin, gentamicin, tri-methoprim-sulfamethoxazole, nalidixic acid, and furaxone.All strains formed pigmented colonies on Congo red agar at30 and 37°C. The surface hydrophobicity of the strains asdetermined by the salt aggregation test was found to be 1.5M. All strains gave a strong and rapid agglutination of guineapig erythrocytes which was inhibited by mannose only.

Culture of the gut contents of two animals that developedKC was positive for S. weltevreden, and none of the animalhandlers carried the organism.

DISCUSSION

Although both Salmonella spp. and Shigella spp. areinvasive enteric pathogens, there are differences in thepathogenesis of diarrhea caused by these organisms. Shigel-lae invade and multiply within colonic epithelial cells, caus-ing ulceration of the mucosa. The lesion rarely extendsbeyond submucosa, and bacteremia is relatively uncommon(17). However, nontyphoid salmonellae invade both terminalileum and colon, and the organisms are transported by

VOL. 29, 1991

on February 23, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

2004 ALBERT ET AL. J. CLIN. MICROBIOL.

A ~ x ;s, .x.

'C'-f - ,0I

r0s* ew* ,*^t S ftF ..' ',$r

eqe1t

'a a '-,Wr,. _ww, n,t~~~~~ -C ,., 'a ,/,f

-. - ..sAp~ -

t?,-0-

'p. f

I

_.

*fI

414i0sSs. ;*

.:+ :~~~~~~~~~wIteA -

BICtU.'t %.

.- .* t -"

t ,'p.- -r ait - .0'.r

r _ * ,.'-- .'put- - r ...

at~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~f; - , . _ .. *r _07 4 _S..a..........s to', _ a -a

ftw*

:M *t tjE o~~~~~_V-=_ts:*

FIG. 2. Histology of the eye of a guinea pig (Fig. 1A) showing severe KC (A) compared with that of the eye of a normal guinea pig (B)(Fig. 1B) (hematoxylin and eosin staining). Magnification, x66.

macrophages across the mucosa and deeper into the tissueand mesenteric lymph nodes. Bacteremia is more common insalmonellosis than in shigellosis (3). In shigellae, althoughthe ability to cause KC is encoded on both chromosomal andlarge invasive plasmid genes, invasiveness of HeLa cells isdirectly correlated with the latter (18). Loss of the plasmidresults in loss of the ability to both produce KC and invadeHeLa cells (12). Similarly, in Y. enterocolitica, another

invasive pathogen, the ability to cause conjunctivitis de-pends upon both chromosomal and plasmid genes. How-ever, the ability to invade tissue culture cells alone isdependent upon chromosomal genes (21, 32). In many sal-monellae, for full pathogenicity the presence of a crypticplasmid is required, although for invasiveness of tissueculture cells, as in Y. enterocolitica, chromosomal genesalone are sufficient (12). All the S. weltevreden strains had a

..1.

"

m.,I.ICC

a-a -

#:-

on February 23, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

KC DUE TO S. WELTEVREDEN 2005

common 60-MDa plasmid, and we wanted to determine therole of this plasmid in the induction of KC. However,attempts to cure the strains of this plasmid were not suc-cessful. Salmonella spp., for example, S. typhimurium, donot cause KC yet invade HeLa cells in culture (10). Thus,the abilities to produce KC and to invade HeLa cells areindependent characteristics. S. weltevreden possessed theabilities to produce KC as well as to invade HeLa cells.Congo red binding is correlated with the invasiveness of

shigellae and enteroinvasive E. coli (24, 31). Moreover, inthese organisms, the abilities to bind Congo red and toinduce KC are temperature dependent; the organisms losethese properties when cultured at less than 37°C (25, 31).However, S. weltevreden bound Congo red as well asinduced KC when cultured both at 37 and at 30°C, suggestingthat in this organism, these abilities are not strictly temper-ature dependent. A DNA probe constructed from the inva-sive 140-MDa plasmid of Shigella sp. failed to hybridize withS. weltevreden, suggesting that the invasive genes in theseorganisms may not be related. Multidrug resistance in shigel-lae is a serious problem in Bangladesh (38), but S. weltevre-den was susceptible to all the antibiotics tested. S. weltevre-den, like other salmonellae (41), showed mannose-sensitivehemagglutination, suggesting the presence of type I fimbriae.They also agglutinated at a 1.5 M salt concentration, sug-gesting that they are relatively hydrophobic (20, 31). Theseattributes strengthen the virulence potential of S. weltevre-den. There have been reports of salmonellae producingheat-labile enterotoxin (34), heat-stable enterotoxin (15), orcytotoxin (16), but S. weltevreden was negative for all theseproperties.

Investigation of the source of infection suggested that theanimals were the carriers of the organism and fecal contam-ination of the eye resulted in KC. A strain of S. weltevredenthat caused an epidemic of diarrhea in a Bangladeshi villagein 1983 (unpublished data) and a routine clinical isolate alsopossessed the ability to induce KC. Our unpublished datasuggest that intestinal infection with S. weltevreden is notuncommon in Bangladesh. This organism is also known to befrequently isolated in Hawaii (23). The potential of thisorganism to cause KC should be borne in mind.Thus, contrary to the observation that salmonellae are

incapable of causing KC, we have demonstrated that at leastS. weltevreden is capable of inducing this reaction. It will beinteresting to study the molecular basis of this virulenceproperty.

ACKNOWLEDGMENTS

This research was supported by the International Centre forDiarrhoeal Disease Research, Bangladesh (ICDDR,B). ICDDR,B issupported by countries and international agencies which share itsconcern for the health problems of developing countries. Currentdonors include the aid agencies of the governments of Australia,Bangladesh, Belgium, Canada, Denmark, France, Japan, The Neth-erlands, Norway, Sweden, Switzerland, the United Kingdom, andthe United States; international organizations, including the UnitedNations Capital Development Fund, the United Nations Develop-ment Programme, the United Nations Children's Fund, and theWorld Health Organization; and private foundations, including theFord Foundation and the Sasakawa Foundation.We thank Priyatosh Sukul for typing the manuscript.

REFERENCES1. Bauer, A. W., W. M. M. Kirby, J. C. Sherris, and M. Turck.

1966. Antibiotic susceptibility by a standardized single diskmethod. Am. J. Clin. Pathol. 45:493-496.

2. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction

procedure for screening recombinant plasmid DNA. NucleicAcids Res. 1:1513-1523.

3. Butler, T. 1990. Salmonellosis, p. 757-772. In K. S. Warren andA. F. Mahmood (ed.), Tropical and geographical medicine, 2nded. McGraw-Hill Book Co., New York.

4. Dean, A. G., Y. C. Ching, R. G. Williams, and L. B. Harden.1972. Test for Escherichia coli enterotoxin using infant mice:application in a study of diarrhea in children in Honolulu. J.Infect. Dis. 125:407-411.

5. DuPont, H. L., S. B. Formal, R. B. Hornick, M. J. Snyder, J. P.Libonati, D. G. Sheahan, E. H. LaBrec, and J. P. Kalas. 1971.Pathogenesis of Escherichia coli diarrhea. N. Engl. J. Med.285:1-9.

6. Echeverria, P., D. N. Taylor, J. Seriwatana, J. E. Brown, and V.Lexomboon. 1989. Examination of colonies and stool blots fordetection of enteropathogens by DNA hybridization with eightDNA probes. J. Clin. Microbiol. 27:331-334.

7. Ewing, W. H. 1986. Edwards and Ewing's identification ofEnterobacteriaceae, 4th ed. Elsevier Publishing Co., Inc., NewYork.

8. Formal, S. B., T. L. Hale, and P. J. Sansonetti. 1983. Invasiveenteric pathogens. Rev. Infect. Dis. 5:S702-S707.

9. Gentry, M. K., and J. M. Dalrymple. 1980. Quantitative micro-titer cytotoxicity assay for Shigella toxin. J. Clin. Microbiol.12:361-366.

10. Gianella, R. A., 0. Washington, P. Gemski, and S. B. Formal.1973. Invasion of HeLa cells by Salmonella typhimurium: amodel for the study of invasiveness of Salmonella. J. Infect.Dis. 128:69-75.

11. Guerrant, R. L., R. G. Lahita, W. C. Winn, and R. B. Roberts.1978. Campylobacteriosis in man: pathogenic mechanisms andreview of 91 blood stream infections. Am. J. Med. 65:584-592.

12. Hackett, J., I. Kotlarski, V. Mathan, F. Franki, and D. Rowley.1986. The colonization of Peyer's patches by a strain of Salmo-nella typhimurium cured of cryptic plasmid. J. Infect. Dis.153:1119-1125.

13. Harris, J. R., I. K. Wachsmuth, B. R. Davis, and M. L. Cohen.1982. High-molecular-weight plasmid correlates with Esche-richia coli enteroinvasiveness. Infect. Immun. 37:1295-1298.

14. Inuzuka, N., S. Nakamura, M. Inuzuka, and M. Tomoeda. 1969.Specific action of sodium dodecyl sulfate on the sex factor ofEscherichia coli K-12 Hfr strains. J. Bacteriol. 100:827-839.

15. Jiwa, S. F. H. 1981. Probing for enterotoxigenicity among thesalmonellae: an evaluation of biological assay. J. Clin. Micro-biol. 14:463-472.

16. Ketyi, I., S. Pacsa, L. Emody, A. Vertenyi, B. Kocsis, and B.Kuch. 1979. Shigella dysenteriae 1-like cytotoxic enterotoxinsproduced by Salmonella strains. Acta Microbiol. Acad. Sci.Hung. 26:217-223.

17. Keusch, G. T., and M. Bennish. 1989. Shigella, p. 265-282. InM. J. G. Farthing and G. T. Keusch (ed.), Enteric infections:mechanisms, manifestations and management. Chapman andHall Medical, London.

18. Kopecko, D. J., M. Venkatesan, and J. M. Buysse. 1989. Basicmechanisms and genetic control of bacterial invasion, p. 42-63.In M. J. G. Farthing and G. T. Keusch (ed.), Enteric infections:mechanisms, manifestations and management. Chapman andHall Medical, London.

19. Ljungh, A., and T. Kronevi. 1982. Aeromonas hydrophila tox-ins-intestinal fluid accumulation and mucosal injury in animalmodels. Toxicon 20:397-402.

20. Ljungh, A., and T. Wadstrom. 1982. Salt aggregation test formeasuring cell surface hydrophobicity of urinary Escherichiacoli. Eur. J. Clin. Microbiol. 1:388-393.

21. Maki, M., T. Vesikari, I. Rantala, C. Sundquist, and P.Gronross. 1983. Pathogenicity of 42-44 MDal plasmid positiveand negative Yersinia pseudotuberculosis and Yersinia entero-colitica 0:8 and 0:9 studied in the guinea pig eye model (Serenytest). Acta Pathol. Microbiol. Immunol. Scand. Sect. B 91:241-244.

22. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual, p. 109-112. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

VOL. 29, 1991

on February 23, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

2006 ALBERT ET AL.

23. Martin, S., N. Bean, and R. Tauxe. 1986. An atlas of Salmonellain the United States-serotype specific surveillance 1968-1986.CDC Public Health Service, Centers for Disease Control, At-lanta.

24. Maurelli, A. T., B. Blackmon, and R. Curtiss III. 1984. Loss ofpigmentation in Shigella flexneri 2a is correlated with loss ofvirulence and virulence-associated plasmid. Infect. Immun.43:397-401.

25. Maurelli, A. T., and P. J. Sansonetti. 1988. Identification of achromosomal gene controlling temperature-regulated expres-sion of Shigella virulence. Proc. Natl. Acad. Sci. USA 85:2820-2824.

26. McHugh, G. L., and M. N. Swartz. 1977. Elimination of plas-mids from several bacterial species by novobiocin. Antimicrob.Agents Chemother. 12:423-426.

27. Meyers, J. A., D. Sanchez, L. P. Elwell, and S. Falkow. 1976.Simple agarose gel electrophoretic method for the identificationand characterization of plasmid deoxyribonucleic acid. J. Bac-teriol. 127:1529-1537.

28. Morgan, D. R., P. C. Johnson, H. L. DuPont, T. K. Satterwhite,and L. V. Wood. 1985. Lack of correlation between knownvirulence properties ofAeromonas hydrophila and enteropatho-genicity for humans. Infect. Immun. 50:62-65.

29. Moseley, S. L., P. Echeverria, J. Seriwatana, C. Tirapat, W.Chaicumpa, T. Sakuldaipeara, and S. Falkow. 1982. Identifica-tion of enterotoxigenic Escherichia coli by colony hybridizationusing three enterotoxin gene probes. J. Infect. Dis. 145:863-869.

30. Qadri, F., S. Haq, and I. Ciznar. 1989. Hemagglutinatingproperties of Shigella dysenteriae type 1 and other Shigellaspecies. Infect. Immun. 57:2909-2911.

31. Qadri, F., S. A. Hossain, I. di1nar, K. Haider, A. Ljungh, T.Wadstrom, and D. A. Sack. 1988. Congo red binding and saltaggregation as indicators of virulence in Shigella species. J.Clin. Microbiol. 26:1343-1348.

32. Robins-Browne, R. M., and J. K. Prpic. 1985. Effects of ironand desferrioxamine on infections with Yersinia enterocolitica.

Infect. Immun. 47:774-779.33. Sack, D. A., and R. B. Sack. 1975. Test for enterotoxigenic

Escherichia coli using Y-1, adrenal cells in miniculture. Infect.Immun. 11:334-336.

34. Sandefur, P. D., and J. W. Peterson. 1977. Neutralization ofSalmonella toxin-induced elongation of Chinese hamster ovarycells by cholera antitoxin. Infect. Immun. 15:988-992.

35. Schiemann, D. A., and J. A. Devenish. 1980. Virulence ofYersinia enterocolitica determined by lethality in Mongoliangerbils and by the Sereny test. Infect. Immun. 29:500-506.

36. Sereny, B. 1955. Experimental Shigella keratoconjunctivitis: apreliminary report. Acta Microbiol. Acad. Sci. Hung. 2:293-296.

37. Sethabutr, O., S. Hanchalay, P. Echeverria, D. N. Taylor, and U.Leksomboon. 1987. A non-radioactive DNA probe to identifyShigella and enteroinvasive Escherichia coli in stools of chil-dren with diarrhoea. Lancet ii:1095-1097.

38. Shahid, N. S., M. M. Rahaman, K. Haider, H. Banu, and N.Rahman. 1985. Changing pattern of resistant Shiga bacillus(Shigella dysenteriae type 1) and Shigella flexneri in Bang-ladesh. J. Infect. Dis. 152:1114-1119.

39. Small, P. L. C., R. R. Isberg, and S. Falkow. 1987. Comparisonof the ability of enteroinvasive Escherichia coli, Salmonellatyphimurium, Yersinia pseudotuberculosis, and Yersinia entero-colitica to enter and replicate within HEp-2 cells. Infect. Im-mun. 55:1674-1679.

40. Tenner, C., P. Racz, and B. Redey. 1971. Significance of theepithelial phase in experimental Salmonella conjunctivitis. ActaMicrobiol. Acad. Sci. Hung. 18:167-177.

41. Wadstrom, T. 1988. Adherence traits and mechanisms of micro-bial adhesion in the gut. Bailliere's Clin. Trop. Med. Commun.Dis. 3:417-433.

42. World Health Organization. 1987. Programme for control ofdiarrhoeal diseases (CDD/83.3 Rev. 1), p. 9-20. In Manual forlaboratory investigations of acute enteric infections. WorldHealth Organization, Geneva.

J. CLIN. MICROBIOL.

on February 23, 2020 by guest

http://jcm.asm

.org/D

ownloaded from