DNA Fingerprinting and Serotyping of Campylobacter jejuni Isolates ...

JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 1996, p. 892–896 0095-1137/96/$04.0010 Copyright q 1996, American Society for Microbiology

Vol. 34, No. 4

DNA Fingerprinting and Serotyping of Campylobacter jejuni Isolates from Epidemic Outbreaks ¨ GREN,1 KJETIL MELBY,2 LENA LIND,1* EVA SJO

AND

BERTIL KAIJSER1

Department of Clinical Bacteriology, University of Go ¨teborg, Go ¨teborg, Sweden,1 and Department of Microbiology, Ullevaal University Hospital, Oslo, Norway2 Received 31 July 1995/Returned for modification 26 October 1995/Accepted 16 January 1996

The aim of the present investigation was to compare DNA fingerprinting and serotyping (heat-stabile and heat-labile antigens) of isolates from epidemic outbreaks as well as of solitary isolates. Campylobacter jejuni isolates from two epidemic outbreaks in Sweden, one milkborne (35 isolates) and one waterborne (17 isolates), and one waterborne outbreak in Norway (11 isolates), as well as 30 solitary isolates from Swedish patients with gastroenteritis, were analyzed. A total of 93 isolates were analyzed. In the waterborne outbreak in Norway, only one serotype with one DNA pattern was found. In the milkborne outbreak in Sweden, two serotypes (HS2:HL4 and HSNT:HL4) with two different DNA patterns were found. The isolates from the waterborne outbreak in Sweden were different serotypes. For two isolates of the same serotype, different DNA patterns were seen. This was also recorded for isolates from solitary cases. It was concluded that serotyping is a useful tool in most epidemiological situations but sometimes lacks sufficient discriminatory power. DNA fingerprinting can add valuable epidemiological information to that supplied by serotyping and can in some situations provide sufficient epidemiological information when used alone. small lake which is used for drinking water in that part of Norway. The contamination possibly came from feces of wild birds in the surrounding area. The second epidemic outbreak was in Sweden, and the organisms originated from a small river, where the river water via backflow contaminated the drinking water system. Organisms of the third small outbreak originated from infected cows. The owner family drank unpasteurized milk. In epidemic outbreaks caused by bacteria, there is an urgent need for a rapid, reliable, and simple typing procedure in order to detect and eliminate the source of infection. For Campylobacter spp. as well as for many other enteropathogenic microorganisms, a number of methods, such as serotyping, ribotyping, auxotyping, and plasmid typing, have been established (17, 18). A commonly used serotyping system is based on surface antigens (8). The aim of this investigation was to compare the usefulness of the recently described DNA fingerprinting typing procedure with heat-stable (HS)- and heat-labile (HL)-antigen serotyping techniques using isolates from singly infected persons as well as isolates from epidemic outbreaks (13).

Campylobacter jejuni is a commensal organism normally found in the wild as well as in domestic animals (6, 23). Furthermore, it is found at a fairly high frequency in river water and ponds, during both warm and cold seasons (2, 24). Most persons infected with C. jejuni have solitary infections originating from contaminated food. Secondary cases are uncommon (1, 3, 8). Various species of Campylobacter have been known for 15 years to have an important role in human gastrointestinal infections. Originally, two species were described, C. jejuni and C. coli. Thereafter, more species with various degrees of clinical significance were recognized, some pathogenic and some nonpathogenic. In the mid-1980s, Campylobacter pylori was recognized as a specific genus and was named Helicobacter. Many Campylobacter species cause diarrhea. Helicobacter spp. contribute to the etiology of gastritis and ulcers. C. jejuni is one of the most common bacterial etiologies of waterborne outbreaks of diarrhea (12, 14, 25). One explanation is that C. jejuni may survive under different conditions in a nonculturable form (7, 15, 21). Several waterborne epidemics have previously been described. Three well-known and well-documented epidemics occurred in Vermont in 1982 (25), in Grums, Sweden, in 1986 (14), and in northern Norway in 1988 (12). The outbreaks often are small, affecting a limited number of individuals. Those mentioned above, however, involved hundreds of people. In addition to water, unpasteurized milk has been recognized as a source of epidemic outbreaks of diarrhea (22). Approximately 5% of cattle (23) have Campylobacter spp. as a component of the gut flora. We have recently studied three epidemic outbreaks. One waterborne outbreak occurred in northern Norway, originating from contaminated drinking water. It was concluded that the source of the organism was the polluted surface water of a

MATERIALS AND METHODS Bacterial isolates. Isolates of C. jejuni from three epidemic outbreaks were investigated. From the milkborne outbreak, C. jejuni isolates from the feces of 30 cows the milk of 1 cow, and the stools of four farmers with diarrhea were investigated. From a waterborne outbreak in Sweden, C. jejuni isolates from 13 patients with diarrhea and four water samples were tested. From a waterborne outbreak in Norway, C. jejuni isolates from 11 patients with diarrhea were investigated. Thirty isolates from patients (18 with serotype HS2:HL4, 5 with serotype HS18:HL2, and 7 with serotype HS1:HL2) without any epidemiological connection to an outbreak were also tested. All isolates were kept freeze-dried until tested. Preparation of chromosomal DNA from plate cultures. Freeze-dried isolates of Campylobacter were mixed with 1 ml of sterile broth and plated on Columbia blood agar containing 5% horse blood. The plates were incubated overnight at 428C under microaerophilic conditions (5% O2, 10% CO2, and 85% N2). The Campylobacter cells were scraped off and resuspended in 1.5 ml of 0.9% NaCl in an Eppendorf tube and spun down at 6,000 3 g for 5 min. The pellet was resuspended in 540 ml of 0.01 M Tris-EDTA (TE) buffer (pH 8.0) containing 5 mg of lysozyme (Sigma) per ml. A 60-ml volume of 10% sodium dodecyl sulfate

* Corresponding author. 892

DNA FINGERPRINTS AND SEROTYPES OF C. JEJUNI ISOLATES

VOL. 34, 1996

893

TABLE 1. Testing of bacterial strains with different restriction endonucleases Enzyme

No. of strains tested

ApaI NotI HindIII BamHI Bsu 36 I

3 3 18 18 18

HaeIII PstI

93 55

Result

Not digested Not digested Partly digested Partly digested No distinct pattern when DNA fragments were .10 kb Distinct pattern No distinct pattern when DNA fragments were .10 kb

(SDS) was added, and the solution was incubated at 378C for 15 min. One hundred fifty microliters of a solution containing 25 mg of pronase (Boehringer Mannheim GmbH) per ml was added, and the combination was mixed and reincubated overnight at 378C and then centrifuged at 10,000 3 g for 15 min. Thereafter, 700 ml of the supernatant was transferred to another Eppendorf tube (4). For DNA extraction, 700 ml of phenol was added and the mixture was vigorously shaken and then centrifuged at 10,000 3 g for 5 min. Approximately 75% of the viscous supernatant above the white interface was transferred to a fresh Eppendorf tube, an equal volume of chloroform-isoamyl alcohol was added, and the mixture was carefully combined and centrifuged at 10,000 3 g for 2 min. The supernatant was once more mixed with an equal volume of chloroform-isoamylic alcohol and centrifuged as described above, and the supernatant was again transferred to a fresh Eppendorf tube. The supernatant was made 1 M by adding 5 M NaCl, and the DNA was precipitated with 2 volumes of 96% ethanol for 30 min at 2208C and centrifuged at 10,000 3 g for 10 min. The pellet, which consisted of DNA, was suspended in 100 ml of TE buffer (pH 7.6) and treated with 0.5 ml of a solution containing 10 mg of RNase (Boehringer Mannheim GmbH) per ml for 45 min at 378C. A 50-ml volume of 7.5 M NH4 acetate was added, and the DNA was precipitated with 375 ml of 96% ethanol for 30 min at 2208C and centrifuged at 10,000 3 g for 10 min. The pellet was resuspended in 100 ml of TE buffer (pH 7.6). The final DNA concentration was measured spectrophotometrically (Gene Quant II; Pharmacia Biotech) at 260 nm (5). Restriction endonuclease digestion of DNA. Different restriction endonucleases were examined to find out which one gave the most distinct DNA pattern.

TABLE 2. Campylobacter isolates, serotypes, and HaeIII restriction endonuclease digest DNA patterns from epidemic outbreaks and from solitary patients with diarrhea Origin and source of specimen

Waterborne outbreak, Norway Patients Waterborne outbreak, Sweden Patients

Water Milkborne outbreak, Sweden Patients Cow milk Cow feces Solitary patients with diarrhea

No. of isolates

11 3 7 1 1 1 2 1 1 4 1 29 1 18 7

Total a

NT, not typeable.

93

Serotypea

FIG. 1. DNA pattern I of the 11 isolates (lanes 1 to 11) from the Norwegian waterborne outbreak.

DNAs from three isolates of C. jejuni were digested with either ApaI or NotI; DNAs from 18 isolates were digested with either HindIII, BamHI, or Bsu 36 I; DNAs from 93 isolates were digested with HaeIII; and DNAs from 55 isolates were digested with PstI (see Table 1) (4, 5, 9, 10, 16). All enzymes were obtained from Scandinavian Diagnostic Service. Approximately 5 mg of the purified DNA was used for each restriction endonuclease digest. Thirty units of each of the enzymes and a specific buffer for each enzyme made according to the supplier’s specifications were used. The volume of the digest was adjusted to 30 ml with distilled water. The digestion was stopped with 0.5 ml of EDTA, and 10 ml of SB buffer (25% sucrose, 5 mM sodium acetate, 0.05% bromophenol blue, 0.1% SDS) was added (16). Electrophoresis and photography. The digested bacterial DNA was loaded on a horizontal slab gel (0.7%) covered with electrophoresis buffer (40 mM Tris acetate [pH 7.9], 2 mM sodium EDTA). The gel was run overnight at 40 V. After electrophoresis, the gel was stained for 45 min with ethidium bromide (0.5 mg/ml) and photographed with UV illumination and Polaroid film (Polaroid 667). Serotyping. The 93 C. jejuni isolates included were serotyped for HS (19, 20) and HL antigens (11). A total of 65 different absorbed antisera were used for HS typing, and 125 absorbed antisera were used for HL typing. The absorption was performed as described by Penner et al. (19) and Lior et al. (11), respectively.

DNA pattern(s)

HS7:HL5a

I

HSNT:HL55 HS23:HL7 HS1,44:HL10,13 HS23:HLNT HS1,44:HL10,13 HS23:HL7 HSNT:HL51 HS22,23:HL13

II III IV V VI II VII VIII

HS2:HL4 HS2:HL4 HS2:HL4 HSNT:HL4 HS2:HL4 5HS18:HL2 HS1:HL2

IX IX IX X XI–XVIII XIX–XXII XXIII–XXVII 27

FIG. 2. DNA fingerprinting profiles II to VIII of the isolates from the Swedish waterborne outbreak. Lanes 1, 2, and 4, DNA pattern II; lanes 3, 8 to 12, and 14 to 16, pattern III; lane 5, pattern IV; lane 6, pattern V; lane 7, pattern VI; lane 13, pattern VII; and lane 17, pattern VIII.

894

LIND ET AL.

J. CLIN. MICROBIOL.

FIG. 3. DNA patterns IX and X of the isolates from the milkborne outbreak in Sweden. Pattern IX: lanes 1 to 4, isolates from patients; lane 5, isolate from cow milk; lanes 6 to 9 and 11 to 18, isolates from cow feces. Pattern X: lane 10, cow feces strain with serotype HSNT:HL4.

RESULTS Neither ApaI nor NotI digested the DNA at all. With either HindIII or BamHI, the DNA was only partly digested. Digestion with either Bsu 361 or PstI gave no distinct DNA pattern

FIG. 5. DNA fingerprinting profiles XIX to XXII of the isolates with serotype HS18:HL2 from solitary patients. Lane 1, pattern XIX; lane 2, pattern XX; lane 3, pattern XXI; and lanes 4 and 5, pattern XXII.

when the DNA fragments were .10 kb. HaeIII digestion gave the best-resolved pattern even when the DNA fragments were .10 kb, so DNAs from all isolates were digested with HaeIII (Table 1). Eleven isolates from the Norwegian waterborne outbreak were all identified as serotype HS7:HL5. These isolates also had the same DNA pattern (Table 2 and Fig. 1). The results for isolates from the waterborne outbreak in Sweden were more complex. Six serotypes and seven DNA patterns were identified among the 17 C. jejuni isolates (Table 2 and Fig. 2). For two serotypes, HS23:HL7 and HS1,44: HL10,13, two different DNA patterns were found. For the 35 isolates from the milkborne outbreak in Sweden, two serotypes, HS2:HL4 and HSNT:HL4, were identified among the specimens from farmers, cow milk, and cow feces. Thirty-four isolates had the same DNA pattern. For the isolate from the feces of a single cow, with the serotype HSNT:HL4, a different DNA pattern was also found (Table 2 and Fig. 4). In the group of 30 isolates with no epidemiological connection, we found eight different DNA patterns among the 18 isolates of serotype HS2:HL4, four DNA patterns among five isolates of serotype HS18:HL2, and five DNA patterns among the seven isolates of serotype HS1:HL2 (Table 2 and Fig. 3 through 6). Among the 93 isolates that were tested, a total of 27 DNA patterns were identified. DISCUSSION

FIG. 4. DNA fingerprinting profiles XI to XVIII of the isolates with serotype HS2:HL4 from solitary patients. Lane 1, DNA pattern XI; lane 2, pattern XII; lanes 3 and 4, pattern XIII; lanes 5 to 7 and 10 to 16, pattern XIV; lane, 8 pattern XV; lane 9, pattern XVI; lane 17, pattern XVII; and lane 18, pattern XVIII.

In order to find a wide spectrum of epidemiologically different isolates, we selected C. jejuni isolates from solitary infected individuals as well as isolates from epidemic outbreaks.

VOL. 34, 1996

DNA FINGERPRINTS AND SEROTYPES OF C. JEJUNI ISOLATES

895

Enteropathogens other than C. jejuni were also recovered from fecal samples from the infected patients. The origin of this outbreak was identified as wastewater contamination of the community drinking water pipeline system. Several different strains might therefore have caused the outbreak. For the solitary isolates from patients with no epidemiological connection, several DNA patterns could also be identified within each serotype (Table 2; Fig. 4 through 6). Thus, for a number of our isolated serotypes, several DNA patterns were found. However, from a reversed perspective, for each DNA pattern never more than one serotype was found. We conclude that serotyping is a useful tool in most epidemiological situations and easy to perform. In some of the isolates with one serotype pattern, several different DNA patterns were found. Thus, DNA fingerprinting adds to the epidemiological information supplied by serotyping. In laboratories in which DNA fingerprinting but not Campylobacter serotyping is available, the DNA technique may, in the same instance, supply sufficient information for a limited epidemiological investigation. ACKNOWLEDGMENTS

FIG. 6. DNA fingerprinting profiles XXIII to XXVII of the isolates with serotype HS1:HL2 from solitary patients. Lane 1, pattern XXIII; lanes 2, 5, and 6, pattern XXIV; lane 3, pattern XXV; lane 4, pattern XXVI; and lane 7, pattern XXVII.

The waterborne outbreaks chosen were one in Sweden, in which several isolates of different serotypes were identified as causing the infection (personal communication), and one in Norway, in which all isolates belonged to one serotype (12) (Table 2). Serotyping based on HL antigens and HS, lipopolysaccharide antigens is a commonly used system for characterization of C. jejuni (11, 19). Serological typing, in most epidemiological situations, provides sufficient discriminative power. DNA fingerprinting is based upon the fact that a restriction enzyme makes sequence-specific cuts in DNA and thus can separate different C. jejuni isolates with different sequences of nucleic acids in the DNA. As DNA structures differ in different bacteria, several restriction endonuclease enzymes need to be tested. We analyzed seven endonucleases. Of these, only one, HaeIII, gave a distinct enough pattern for analysis. HaeIII was therefore used throughout the investigations. In the Norwegian waterborne outbreak as well as in the milkborne outbreak in Sweden, only one serotype with one DNA pattern was isolated (12, 13) (Table 2; Fig. 1 and 2). During investigation of the latter outbreak, a Campylobacter isolate with a serotype (HSNT:HL4) very similar to that of an isolate causing diarrhea (HS2:HL4) was found in feces of one cow. With the DNA fingerprinting method, we could show that the two isolates had two distinct DNA patterns, verifying that serotype HSNT:HL4 was not involved in the outbreak. This exemplifies the usefulness of extended typing procedures. In the Swedish waterborne outbreak, several serotypes were found. For two serotypes, HS23:HL7 and HS1,44:HL10,13, two different DNA patterns were found (Table 2 and Fig. 2).

This investigation was supported by grants from the Swedish Medical Research Council (8673) and the Swedish Council for Forestry and Agricultural Research (472/91). The skillful typing of the manuscript by Ann Barkenfelt was greatly appreciated. The technical laboratory analyses were competently performed by Elisabeth Pettersson. REFERENCES 1. Albert, M. J., A. Leach, V. Asche, J. Hennessy, and J. L. Penner. 1992. Serotype distribution of Campylobacter jejuni and Campylobacter coli isolated from hospitalized patients with diarrhea. J. Clin. Microbiol. 30:207–210. 2. Bolton, F. J., D. Coates, D. N. Hutchinson, and A. F. Godfree. 1987. A study of thermophilic campylobacters in a river system. J. Appl. Bacteriol. 62: 167–176. 3. Bradbury, W. C., A. D. Pearson, M. A. Marko, R. V. Congi, and J. L. Penner. 1984. Investigation of a Campylobacter jejuni outbreak by serotyping and chromosomal restriction endonuclease analysis. J. Clin. Microbiol. 19:342– 346. 4. Collins, D. M., and D. E. Ross. 1984. Restriction endonuclease analysis of Campylobacter strains with particular reference to Campylobacter fetus ss. fetus. J. Med. Microbiol. 18:117–124. 5. Dawson, C., R. J. Owen, and A. Beck. 1987. A method for rapid extraction of Providencia chromosomal DNA and restriction enzyme digestion in agarose pellets. Lett. Appl. Microbiol. 5:41–45. 6. Garcia, M. M., H. Lior, R. B. Stewart, G. M. Ruckerbauer, J. R. Trudel, and A. Skijarevski. 1985. Isolation, characterization and serotyping of Campylobacter jejuni and Campylobacter coli from slaughter cattle. Appl. Environ. Microbiol. 49:667–672. 7. Jones, D. M., E. M. Sutcliffe, and A. Curry. 1991. Recovery of viable but non-culturable Campylobacter jejuni. J. Gen. Microbiol. 137:2477–2482. 8. Kaijser, B., and E. Sjo ¨gren. 1985. Campylobacter strains in Sweden. Serotyping and correlation to clinical symptoms. Acta Pathol. Microbiol. Immunol. Scand. Sect. B 93:315–322. 9. Kakoyiannis, C. K., P. J. Winter, and R. B. Marshall. 1984. Identification of Campylobacter coli isolates from animals and humans by bacterial restriction endonuclease DNA analysis. Appl. Environ. Microbiol. 48:545–549. 10. Kakoyiannis, C. K., P. J. Winter, and R. B. Marshall. 1988. The relationship between intestinal Campylobacter species isolated from animals and humans as determined by BRENDA. Epidemiol. Infect. 100:379–387. 11. Lior, H., D. L. Woodward, J. A. Edgar, L. J. Laroche, and P. Gill. 1982. Serotyping of Campylobacter jejuni by slide agglutination based on heatlabile antigenic factors. J. Clin. Microbiol. 15:761–768. 12. Melby, K., O. P. Dahl, L. Crisp, and J. L. Penner. 1990. Clinical and serological manifestations in patients during a waterborne epidemic due to Campylobacter jejuni. J. Infect. 21:309–316. 13. Melby, K., L. A. Holmen, J. G. Svendby, T. Eggebø, and B. M. Anderssen. 1991. Epidemiologic outbreak of Campylobacter infection. Tidskr. Nor. Laegeforen. 111:1530–1534. 14. Mentzing, L. O. 1981. Waterborne outbreak of Campylobacter enteritis in central Sweden. Lancet ii:352–354.

896

LIND ET AL.

15. Moran, A. P., and M. E. Upton. 1987. Factors affecting production of coccoid forms by Campylobacter jejuni on solid media during incubation. J. Appl. Bacteriol. 62:527–537. 16. Owen, R. J., A. Beck, and P. Borman. 1985. Restriction endonuclease digest patterns of chromosomal DNA from nitrate-negative Campylobacter jejunilike organisms. Eur. J. Epidemiol. 1:281–287. 17. Owen, R. J., M. Costas, and C. Dawson. 1989. Application of different chromosomal DNA restriction digest fingerprints to specific and subspecific identification of Campylobacter isolates. J. Clin. Microbiol. 27:2338–2343. 18. Patton, C. M., I. K. Wachsmuth, G. M. Evins, J. A. Kiehlbauch, B. D. Plikaytis, N. Troup, L. Tompkins, and H. Lior. 1991. Evaluation of 10 methods to distinguish epidemic-associated Campylobacter strains. J. Clin. Microbiol. 29:680–688. 19. Penner, J. L., J. N. Hennessy, and R. V. Congi. 1983. Serotyping of Campylobacter jejuni and Campylobacter coli on the basis of thermostable antigens. Eur. J. Clin. Microbiol. 2:378–383. 20. Penner, J. L., N. Hennessy, S. D. Mills, and W. C. Bradbury. 1983. Appli-

J. CLIN. MICROBIOL.

21. 22. 23. 24. 25.

cation of serotyping and chromosomal restriction endonuclease digest analysis in investigating a laboratory-acquired case of Campylobacter jejuni enteritis. J. Clin. Microbiol. 18:1427–1428. Rollins, D. M., and R. R. Colwell. 1986. Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Appl. Environ. Microbiol. 52:531–538. Sharp, J. C. M., G. M. Paterson, and N. J. Barrett. 1985. Pasteurization and the control of milkborne infection in Britain. Br. Med. J. 291:463–464. Svedhem, Å., and B. Kaijser. 1981. Isolation of Campylobacter jejuni from domestic animals and pets: probable origin of human infection. J. Infect. 3: 37–40. Taylor, D. N., K. T. McDermott, J. R. Little, J. G. Wells, and M. J. Blaser. 1983. Campylobacter enteritis from untreated water in the Rocky Mountains. Ann. Intern. Med. 99:38–40. Vogt, R. L., H. E. Sours, T. Barrett, R. A. Feldman, R. J. Dickinson, and L. Witherell. 1982. Campylobacter enteritis associated with contaminated water. Ann. Intern. Med. 96:292–296.