Hippurate Hydrolysis Test Research Articles

Campylobacter jejuniHS:23 and Guillain-Barré Syndrome, Bangladesh

To the Editor: Guillain-Barre syndrome (GBS) is an acute peripheral neuropathy triggered by a preceding infectious illness. Gastroenteritis caused by Campylobacter jejuni is the most frequently reported antecedent event (1). In Japan, South Africa, China, and Mexico, Campylobacter strains with certain Penner heat-stable (HS) serotypes, including HS:19 and HS:41, are overrepresented among isolates from GBS case-patients, compared with isolates from enteritis case-patients (2,3). Several studies indicate that C. jejuni HS:19 and HS:41 have a clonal population structure and suggest that these serotypes might have unique virulence properties that are intricately linked to development of GBS (4). However, data from the United Kingdom and the Netherlands suggest that such virulence properties may not be restricted to specific HS serotypes because many other serotypes can be cultured from patients with GBS (5). We report a non-HS:19 and non-HS:41 C. jejuni serotype and sequence type (ST)–3219 that are overrepresented among isolates from GBS patients in Bangladesh. We conducted a prospective case-control study of the serotype and genotype of C. jejuni associated with GBS in Bangladesh. Case-patients were 97 persons with GBS admitted to Dhaka Medical College Hospital, Bangabandhu Sheikh Mujib Medical University, and Dhaka Central Hospital during July 2006–June 2007. All fulfilled the diagnostic criteria for GBS of the National Institute of Neurological Disorders and Stroke of the US National Institutes of Health (Bethesda, MD, USA) (6). The control group comprised 97 patients with other neurologic diseases, matched with case-patients by sex, age, and date of admission to the hospital. A second control group comprised 97 healthy family members of case-patients. Up to 3 stool samples were cultured from each case-patient and control. Campylobacter strains were presumptively identified with Gram stain, oxidase, and hippurate hydrolysis tests and confirmed with a C. jejuni species-specific PCR. Serotyping was performed at the National Laboratory for Enteric Pathogens, Canadian Science Centre for Human and Animal Health, Winnipeg, Manitoba, Canada. All strains were serotyped according to the HS serotyping schemes of Penner et al. (7). To determine the class of lipooligosaccharides (LOS) locus in each of the C. jejuni strains, genomic DNA was isolated by using the DNeasy tissue kit (QIAGEN, Venlo, the Netherlands). PCR analysis was performed with primer sets specific for classes A, B, C, D, and E (8). We isolated C. jejuni from fecal samples of 10 case-patients. Campylobacter strains were not isolated from the control groups (p<0.001). Serotyping of the 10 GBS-related strains showed 4 different HS serotypes. C. jejuni HS:23 was found in 5 (50%) strains; HS:19, in 2 (20%); HS:55 and HS:21, in 1 strain each. One strain was untypeable according to the HS typing scheme. In a collection of clinical C. jejuni isolated during the same period from patients with enteritis, HS:23 was encountered in 9 (28%) of 32 patients. Serotypes previously associated with GBS were HS:1, HS:2, HS:4, HS:4/50, HS:5, HS:10, HS:13/65, HS:16, HS:19, HS:23, HS:35, HS:37, HS:41, HS:44, and HS:64 (5,9). Nine (90%) of the C. jejuni isolates from the case-patients had the class A or class B LOS, which are highly associated with the presence of ganglioside-mimicking structures in LOS (10). Godschalk et al. found that 14 (82%) of 17 GBS-associated isolates possessed a class A/B/C locus (8). Parker et al. (10) found that all GBS-related strains and 64% of the other clinical and environmental isolates belonged to LOS class A/B/C loci. The expression of ganglioside-mimicking structures in Campylobacter, LOS is considered essential for the induction of autoantibodies that lead to GBS. Godschalk et al. (8) demonstrated that specific genes involved in C. jejuni LOS biosynthesis are crucial for the induction of antiganglioside antibodies that lead to GBS. We performed multilocus sequence typing to examine the overall genomic variation among 10 GBS-related C. jejuni strains. We identified 6 different STs among the GBS-related C. jejuni strains (Table). However, ST-3219 has a new combination of alleles and was identified in 4 strains. Concordantly, the analysis demonstrated that C. jejuni isolates with serotype HS:23 were all ST-3219. Of particular interest, ST-985 (BD-67) shared 5 alleles (aspA, uncA, glnA, glyA, pgm) with ST-3219 (Table). Table Serotyping and multilocus sequence typing analysis of Campylobacter jejuni strains associated with Guillain-Barre syndrome, Bangladesh* Our findings of a C. jejuni HS:23 serotype and ST-3219 that is highly prevalent among GBS-related C. jejuni strains from Bangladesh are consistent with previous observations that specific LOS types and serotypes are overrepresented among GBS-related C. jejuni strains. These observations support the hypothesis that, although a great variety of C. jejuni serotypes can be isolated from GBS patients in some geographic areas, specific clonal serotypes and multilocus types are prevalent in GBS patients in other places. The association of GBS with C. jejuni LOS class A/B/C is the only consistent finding when universal collections of GBS-associated strains are considered.

Isolation and identification of Campylobacter spp. and Campylobacter coli from poultry carcasses by conventional culture method and multiplex PCR in Mashhad, Iran

Summary The genus Campylobacter is of great importance to public health because it includes several species that may cause diarrhoea. Poultry and poultry products are known as important sources of human campylobacteriosis. In this study, during the autumn months of 2005, a total of 100 samples from poultry carcasses, representing 20 broiler flocks were obtained by rinse test, after the chilling stage of processing. The samples were enriched in Preston broth, followed by streaking on selective media. Then, the suspected colonies were isolated on sheep blood agar and tested for morphology, motility and Gram-staining. Biochemical tests and hippurate hydrolysis activity were also performed. Concurrently, a multiplex PCR assay (m-PCR) with two sets of primers was employed for identification of Campylobacter genus and Campylobacter coli. The m-PCR assay was applied on bacterial cultures harvested from selective media plates. By conventional culture method, including hippurate hydrolysis test from suspected colonies, 76% of samples were positive for Campylobacter spp. and 2% for C. coli. In m-PCR assay 28% of the harvested cultures, were positive for Campylobacter genus but C. coli were not detected in any of the samples. According to this preliminary study, it seems that the contamination rate of poultry carcasses with other species of Campylobacter genus is higher than contamination with C. coli.

Correct identification and discrimination between Campylobacter jejuni and C. coli by a standardized hippurate test and species-specific polymerase chain reaction

Hippurate hydrolysis test results of 240 Campylobacter strains were compared with those of two multiplex polymerase chain reaction (PCR) assays. Of the 152 strains identified in Finnish clinical microbiology routine laboratories as C. coli (hippurate-negative), 11% were C. jejuni (hippurate-positive) by standardized hippurate test and 39% by PCR in the reference laboratory. Two of the 81 hippurate-positive strains were identified as C. coli. Standardizing the hippurate test by determining minimum and maximum turbidity limits (McFarland 6 and McFarland 10, OD(450) values 0.8 and 1.4, respectively) for the bacterial cell suspension eliminated the false-positive results, but 32% of the 145 hippurate-negative strains were still identified as C. jejuni by PCR. The species identification of Campylobacter isolates in Finland could be improved by using a standardized hippurate hydrolysis test to identify hippurate-positive C. jejuni and testing hippurate-negative strains by molecular methods. This would also improve the epidemiological data on this important zoonotic pathogen.

Discrimination between &lt;i&gt;Campylobacter jejuni&lt;/i&gt; and &lt;i&gt;C. coli&lt;/i&gt; Using a Macrolide and Quinolone Line Probe Assay (MQ-LiPA)

Two hundred and forty nine Campylobacter isolates were discriminated between C. jejuni and C. coli by the hippurate hydrolysis test, three PCR methods using C. jejuni or C. coli specific primer sets, and the macrolide and quinolone line probe assay (MQ-LiPA). The results among the five methods were in agreement for all strains except for six strains, which were specified as C. coli by the hippurate hydrolysis test but as C. jejuni by the other methods. The data of 16S rDNA sequences in two out of these six strains, confirmed the strains were C. jejuni. These results suggest that MQ-LiPA is an effective method and can replace the hippurate hydrolysis test as other PCR methods.

Application of biochemical and polymerase chain reaction assays for identification of Campylobacter isolates from non-human primates

A multiplex polymerase chain reaction (PCR) assay was performed on 167 thermophilic campylobacters isolated from non-human primates. Samples were first identified by phenotypic methods resulting in 64 Campylobacter jejuni and 103 C. coli strains. Four strains identified biochemically as C. coli, were then determined to be C. jejuni by PCR. Comparison of methodologies showed that the main discrepancies were attributed to the hippurate hydrolysis test and sensitivity to cephalothin and nalidixic acid. Analysis of data showed that the application of phenotypic methods should be supplemented by a molecular method to offer a more reliable Campylobacter identification.

Species identification of erythromycin-resistant Campylobacter isolates and optimization of a duplex PCR for rapid detection

This study describes the approach used to verify the species identity of 23 erythromycin-resistant Campylobacter isolates whose identity was initially determined based mainly on the results of the rapid hippurate hydrolysis test or the results of the API-Campy identification system. Species identification of the isolates investigated was confirmed by repeating hippurate hydrolysis using a modification of the rapid hydrolysis test, in addition to performing three genetic-based assays. The original identification was verified in 69.6% of the isolates. The remaining isolates showed discrepancies in identity as determined by results of the identification assays performed. A duplex PCR assay, targeting the hipO and aspA genes, indicated the existence of mixed cultures of C. jejuni and C. coli in the frozen stocks of two of these isolates.

Species level identification of thermotolerant campylobacters

The aim of this study was to compare the phenotypic and genotypic based methods for species identification of thermotolerant campylobacters of human and food origin from the Czech Republic. Phenotypic methods are time-consuming and sometimes lead to intermediate results, therefore replacement by more specific and rapid methods are needed. Out of a total of 911 campylobacter strains tested, 800 human isolates were received from the clinical bacteriology laboratories from 5 regions and 111 foodstuff isolates (raw chicken and pork meat from retail market) originated from the routine examination in our laboratory. Based on the PCR method 85.1% of these strains were identified as C. jejuni, 12.5% as C. coli and 2.3% as mixed cultures of C. jejuni and C. coli. When species determination of campylobacters was based on conventional methods (hippurate hydrolysis test), 28.5% of the isolates were not identified correctly. The mixed cultures of campylobacters have not been detected without further subculturing of strains, which takes several days and enormously extends the identification process. The use of the PCR method showed to be a useful tool for species identification of Campylobacter spp.

Species identification by genotyping and determination of antibiotic resistance in Campylobacter jejuni and Campylobacter coli from humans and chickens in Sweden

Campylobacter is today the most common cause of human bacterial enteritis in Sweden, as well as in most other industrialized countries. Common sources of infection are undercooked chicken meat, unpasteurized milk and contaminated drinking water. One aim with our present study was to identify the species Campylobacter jejuni and Campylobacter coli strains from humans and chickens using a polymerase chain reaction/restriction enzyme analysis (PCR/REA) method, as well as traditional hippurate hydrolysis test. Another aim was to investigate the antibiotic resistance pattern of the human domestic C. jejuni/ C. coli isolates from infected patients and isolates from healthy Swedish chicken, as well as isolates from humans infected abroad. If discrimination between C. jejuni and C. coli was based on testing for hippurate hydrolysis, 95% of the human domestic strains and 88% of the chicken strains were identified as C. jejuni. Based on genotyping by PCR/REA, 100% of the human domestic strains and 98% of the chicken strains were attributed to C. jejuni. The E-test and disc diffusion methods were used for phenotypic antibiotic resistance studies. The two methods gave similar results. Most Swedish C. jejuni/ C. coli isolates both from humans and chickens were sensitive to doxycycline and erythromycin, which are antibiotics used to treat human infection. Only 7% of the human domestic strains and 2% of the chicken strains were resistant to the quinolones tested. As a comparison, more than 94% of strains isolated from travelers to Asia and southern Europe showed antibiotic resistance to one or more drugs.

Molecular-based identification and typing of Campylobacter jejuni and C. coli.

Thermophilic Campylobacter spp., mainly Campylobacter jejuni and to a lesser extent C. coli are recognized as the most common bacteriological causes of gastroenteritis in humans. As enteric infection with Campylobacter organisms cannot be distinguished from that caused by other enteric pathogens, a definitive diagnosis can only be made by isolating or detecting the organism from the feces. The epidemiology of Campylobacter enteritis has been complicated by the ubiquitous nature of the organism (commonly found as a commensal in the intestines of domestic animals, in milk, and in water). Furthermore, identification is carried out only to genus level by most clinical laboratories. Because of the biochemical similarity known to exist between C. jejuni and C. coli, the hippurate hydrolysis test is often used as the only phenotypic test capable of differentiating the two species. This test, however, has some acknowledged technical limitations and is dependent on inoculum size; results can be difficult to interpret accurately. Furthermore, almost all C. jejuni isolates possess the hippuricase gene, fewer C. jejuni isolates express the hippuricase gene. For this reason, certain polymerase chain reaction (PCR)-based species identification methods, for both C. jejuni and C. coli, and for the other thermophilic species, provide more reliable identification; they also help to highlight mixed species cultures, should they occur. However, even with these methods, false negatives or nonspecifically amplified product(s) can occur in a minority of isolates tested owing to genomic anomalies. Thus a second molecular identification method may be required in these circumstances. Gonzalez et al. developed a species-specific PCR assay for the identification of C. jejuni and C. coli based on the ceuE gene, which is involved in siderophore transport. Using this method two primer sets are employed in separate PCR amplification reactions. Another method, developed by Eyers et al., performs PCR amplification of 23S rRNA gene fragments, based on regions specific for C. jejuni, C. coli, C. lari, and C. upsaliensis. In addition, Hani and Chan developed a PCR assay that detected and amplified the hippuricase gene. This molecular approach may offer a more reliable means of identifying C. jejuni strains compared with the phenotypic hippurate hydrolysis test alone.

Comparison of the BAX® System with a multiplex PCR method for simultaneous detection and identification of Campylobacter jejuni and Campylobacter coli in environmental samples

The Campylobacter detection is performed by conventional culture methods and the identification of Campylobacter jejuni and Campylobacter coli is principally based on the hippurate hydrolysis test. The two major drawbacks of this biochemical test for species identification include the inconsistency of the results and the presence of atypical strains, which can lead to the misidentification of an isolate. As an alternative, multiplex polymerase chain reaction (mPCR) protocols for the simultaneous detection and identification of different Campylobacter species have been developed. This study examined the performances of an experimental BAX® System assay for the C. jejuni and C. coli identification in comparison to a multiplex PCR protocol recently published. The samples tested were represented by 106 environmental swabs collected on Teflon strips and tables, stainless steel saws, hooks and trays, ceramic floors and walls, as well as equipment surfaces, located in a swine ( N=50) and a poultry ( N=56) slaughterhouse. The highest Campylobacter detection rate was obtained after 48 h of enrichment by using both the PCR procedures. After 24 h, the BAX® System provides a more rapid and accurate Campylobacter detection and identification assay than the multiplex PCR. Except for two samples, all the broths where Campylobacter cells were detected after 24 or 48 h of enrichment, with at least one of the PCR protocols, gave Campylobacter colonies using the culture method.

Identification of thermophilic Campylobacter spp. by phenotypic and molecular methods

The differences between phenotyping and genotyping (polymerase chain reaction- restriction fragment length polymorphism) of Campylobacter jejuni, Campylobacter coli, Campylobacter lari and Campylobacter upsaliensis were assessed. A total of 51, 63 and 88 strains from dogs, pigs and humans, respectively, were examined. The strains were first typed by biochemical methods, then by PCR-RFLP using AluI and Tsp509I. None of the strains were typed as Camp. lari by the PCR-RFLP. The biggest differences were found in the identification of Camp. jejuni and Camp. coli. The main discrepancies were caused with the hippurate hydrolysis test and sensitivity to cephalothin and nalidixic acid. Strains which were identified biochemically as Camp. coli and by digestion with AluI as Camp. jejuni (eight strains) were tested for the presence of the hippuricase gene. The PCR typing results showed the presence of the hippuricase gene as unique to Camp. jejuni. A reliable identification of Campylobacter spp. should be supplemented with a molecular method.

Development of a m-PCR assay for simultaneous identification of Campylobacter jejuni and C. coli.

Multiplex PCR assay (m-PCR) with three sets of primers was developed for simultaneous identification of Campylobacter jejuni and C. coli. Poultry faecal samples were enriched in Preston broth for 24 h and streaking on selective media was performed before and after enrichment. m-PCR was applied on bacterial cultures harvested from media plates. The data showed a selective effect of Preston broth which favoured the growth of C. coli. Identification of the species by the hippurate hydrolysis test and by the m-PCR was performed on 294 isolates of Campylobacter. The efficiency of the identification by the biochemical test is only 34% in comparison to 100% efficiency with the PCR. The use of our m-PCR in combination with the culture method allowed reliable detection and identification of C. jejuni and C. coli within 3-4 d.

Serotyping of Campylobacter jejuni based on heat-stable antigens: relevance, molecular basis and implications in pathogenesis.

Serotyping of Campylobacter jejuni based on heat-stable antigens: relevance, molecular basis and implications in pathogenesis.

Identification of Serpulina species associated with porcine colitis by biochemical analysis and PCR.

A PCR system for the detection and identification of group IV spirochetes (Serpulina pilosicoli) was designed to complement biochemical tests, e.g., the hippurate hydrolysis and beta-glucosidase tests, and to verify the accuracy of a previously proposed biochemical classification system. The PCR assay was based on amplification of a segment of the 16S rRNA gene. Both primers were constructed to selectively amplify the 16S rRNA gene of Serpulina pilosicoli. All analyzed Serpulina strains exhibiting the capacity to hydrolyze hippurate and lacking beta-glucosidase activity, including the type strain for spirochetal diarrhea, P43, were amplified with the PCR system. All other tested strains, including type and field strains of different phenotypes of Serpulina species, as well as Salmonella species, Campylobacter species, and Escherichia coli strains, were negative in the assay. Among the tested strains were 18 Scottish field isolates originating from the mucosae of pigs with colitis. A simple classification scheme, suitable for routine classification of porcine intestinal spirochetes, is also presented. The scheme is based on hemolysis, indole production, and the hippurate hydrolysis test.

Prevalence of Serpulina species in relation to diarrhea and feed medication in pig-rearing herds in Sweden

Abstract Objective To determine prevalence of various pheno- and genotypes of Serpulina sp in young pigs in relation to diarrhea and feed medication in Swedish pig-rearing herds. Design Isolation of spirochetes. Phenotypical and genotypical classification. Sample Population Young pigs (n = 358) in 19 pig-rearing herds. Procedure Serpulina isolates were classified according to a biochemical scheme based on hemolysis, indole production, hippurate hydrolysis, and α-galactosidase, α-glucosidase, and β-glucosidase activities. The 16S rRNA sequences for 10 of the field strains and 2 type strains of Serpulina spp were aligned and compared. Minimum inhibitory concentrations of olaquindox for 9 of the strains were determined. Results Weakly β-hemolytic intestinal spirochetes (WBHIS) were isolated from 17 of the herds and 65% of the samples. More than 1 phenotype of WBHIS was found in 12 of the 19 herds. S hyodysenteriae was not isolated in any of the herds. Hippurate-positive WBHIS were isolated in 6 of 7 herds affected by diarrhea, but in only 1 of 8 herds without diarrhea. Hippurate-positive strains were closely related to the pathogenic strain P43 if judged from sequence comparisons. Strains with the same biochemical profile isolated within a herd had identical sequences, but when isolated from different herds, sequence differences were observed. The prevalence of WBHIS was reduced in herds medicated with olaquindox. Investigated field strains had minimum inhibitory concentration values ≤ 1 μg/ml for olaquindox. Conclusion The presence of WBHIS, with the ability to hydrolyze hippurate, was related to diarrhea in pig herds. Clinical Relevance Potentially pathogenic WBHIS can be distinguished from nonpathogenic strains by the hippurate hydrolysis test. (Am J Vet Res 1996;57:807–811)

Bacteremia caused by nonhemolytic group B streptococci

Case Report A 75-yr-old woman with a history of type 2 noninsulin-dependent diabetes and hepatic cirrhosis with previous acute episodes of encephalopathy was admitted to the hospital because of abdominal swelling, tenderness of several day’s duration, and nausea On physical examination, the patient was afebrile and had tender hepatomegaly, ascites, and jaundice with normal central nervous system findings. The white blood cell count was 8,100/mm3 (85% neutrophils, 12% lymphocytes, and 3% monocytes). Prythrocyte sedimentation rate was 102 mm at the first hour. Other hematologic studies, metabolic profile, and liver function tests showed increased serum amylase (2,832 U/L), hyperglycemia, and hyperbilirubinemia An ultrasonogmphic study was done and a clinical diagnosis of acute pancreatitis was confiied. Medical therapy was begun, and the disease was self-limited within 3 d after treatment was instituted On the fifth day after admission, the patient underwent gynecological exploration because of vaginal discomfort, and a sample of the purulent vaginal discharge was sent to microbiology laboratory for culture. On the sixth day, she suddenly developed fever (39”C), and three sets of blood cultures were obtained. Empiric treatment with cefotaxime (2 g tid i.v.) was begun, yet the patient remained febrile. The vaginal sample grew a pure culture of a gram-positive, nonhemolytic coccus subsequently identified as Sneptoc0ccus agalactiue (3). The same microorganism also grew in all three blood culture sets after overnight incubation (BacTAlert system, Organon Teknika). A Gram stain of the blood culture showed gram-positive cocci resembling streptococci. The organism grew on sheep blood agar at 37°C with 5% COZ as small, white, slightly convex, nonhemolytic colonies, and hemolysis was not detectable on lifting a colony from the agar. The isolate was catalase-negative, PYRand bile esculin-negative, bacitracin-resistant, and did not grow in the presence of 6.5% sodium chloride. In addition, positive reactions for voges-Proskauer and hippurate hydrolysis tests were obtained. Group B streptococcal antigen was demonstrated with the PhadeBact test system (Karo Bio Diagnostics, Sweden), and a biochemical profile consistent with group B streptococci was obtained with the API 20 Strep test (bioM&ieux, France). Serotyping of the isolate by agglutination with antisera to GBS serotypes I through V (Dako A/S, Denmark) produced agglutination with the type II antisera. Repeated subcultures of the isolate onto blood agar under aerobic and anaerobic conditions failed to produce beta hemolysis. No pigment production was seen when the isolate was cultured in Granada medium, a selective and differential pigmentenhancing culture medium (4).

Identification of Campylobacter jejuni and C. coli by gel electrophoresis of the outer membrane proteins.

Analysis of the electrophoretic profiles of the outer membrane proteins could be used to differentiate Campylobacter jejuni (16 strains) from Campylobacter coli (10 strains). This observation was confirmed by the study of DNA homology obtained by a quantitative filter hybridization method. The hippurate hydrolysis test gave a poor correlation with the results of differentiation obtained by DNA homology studies and outer membrane protein profile.

Serotyping of and hippurate hydrolysis by Campylobacter jejuni isolates from human patients, poultry and pigs in the Netherlands.

Three hundred strains of Campylobacter jejuni, isolated from human patients, poultry and pigs were serotyped according to the Penner-Lauwers system and furthermore tested for hippurate hydrolysis. Serotyping showed a close relationship between human and chicken strains, whereas there was little relationship between human and pig strains. A similar conclusion was drawn from the results of the hippurate hydrolysis test: 86% of human and 94% of poultry strains were positive, whereas 91% of pig strains were negative. These data confirm the conclusions from epidemiological investigations, according to which poultry is a major source of human campylobacteriosis, while pigs are rarely so.

Base composition and sequence homology of deoxyribonucleic acid of thermotolerantCampylobacter from human and animal sources

A total of 31Campylobacter strains were studied. Of these, 24 were thermotolerant field isolates and 7 were reference strains. Mol% G+C was estimated by thermal denaturation and DNA-DNA-relatedness by the hydroxyapatite method. The strains identified asC. jejuni orC. coli and differentiated by the hippurate hydrolysis test formed separate, homogenous DNA homology groups. The two species were genetically more related to each other than to otherCampylobacter species. Results also show that the group of hippurate-negative thermotolerant campylobacters resistant to nalidixic acid is genetically heterogenous.

Serotyping of Campylobacter jejuni and Campylobacter coli on the basis of thermostable antigens.

Hippurate hydrolysis tests performed on the serotype reference strains of the serotyping scheme based on thermostable antigens under development for Campylobacter jejuni showed that 42 strains were Campylobacter jejuni and 17 were Campylobacter coli. Moreover, only four (0.2%) of 2025 hippurate positive Campylobacter jejuni isolates reacted in Campylobacter coli antisera and 12 (4.3%) of the 282 Campylobacter coli reacted in Campylobacter jejuni antisera. Evidently each species has its own array of antigenic specificities. Separate schemes for serotyping Campylobacter jejuni and Campylobacter coli are advocated.