DNA restriction digest and ribosomal RNA gene patterns of Campylobacter jejuni: a comparison with bio-, sero-, and bacteriophage-types of United Kingdom outbreak strains

Knowledge gaps in control of Campylobacter for prevention of campylobacteriosis

Campylobacteriosis is an important, worldwide public health problem with numerous socio-economic impacts. Since 2015, approximately 230,000 cases have been reported annually in Europe. In the United States, Australia and New Zealand, campylobacteriosis is the most commonly reported disease. Poultry and poultry products are considered important sources of human infections. Poultry meat can become contaminated with Campylobacter during slaughter if live chickens are intestinal carriers. Campylobacter spp. can be transferred from animals to humans through consumption and handling of contaminated food products, with fresh chicken meat being the most commonly implicated food type. Regarding food-borne disease, the most important Campylobacter species are Campylobacter jejuni and Campylobacter coli. In humans, clinical signs of campylobacteriosis include diarrhoea, abdominal pain, fever, headache, nausea and vomiting. Most cases of campylobacteriosis are sporadic and self-limiting, but there are post-infection complications, for example, Guillain-Barrés syndrome. This review summarizes an analysis undertaken by the DISCONTOOLS group of experts on campylobacteriosis. Gaps were identified in: (i) knowledge of true number of infected humans; (ii) mechanisms of pathogenicity to induce infection in humans; (iii) training to prevent transfer of Campylobacter from raw to ready-to-eat food; (iv) development of effective vaccines; (v) understanding transmission routes to broiler flocks; (vi) knowledge of bacteriocins, bacteriophages and antimicrobial peptides as preventive therapies; (vii) ration formulation as an effective preventive measure at a farm level; (viii) development of kits for rapid detection and quantification of Campylobacter in animals and food products; and (ix) development of more effective antimicrobials for treatment of humans infected with Campylobacter. Some of these gaps are relevant worldwide, whereas others are more related to problems encountered with Campylobacter in industrialized countries.

Combined use of ribotyping, PFGE typing and IS431 typing in the discrimination of nosocomial strains of methicillin-resistant Staphylococcus aureus

We have previously reported the phenotypic characterization of methicillin-resistant Staphylococcus aureus (MRSA) clinical strains isolated in Malaya University Hospital in the period 1987 to 1989 using antibiogram, coagulase typing, plasmid profiles, and phage typing. Here, we report the analysis of the same strains with three genotyping methods; ribotyping, pulsed-field gel electrophoresis (PFGE) typing, and IS431 typing (a restriction enzyme fragment length polymorphism analysis using an IS431 probe). Ribotyping could discriminate 46 clinical MRSA strains into 5 ribotypes, PFGE typing into 22 types, and IS431 typing into 15 types. Since the differences of the three genotyping patterns from strain to strain were quite independent from one another, the combined use of the three genotyping methods could discriminate 46 strains into 39 genotypes. Thus, the powerful discriminatory ability of the combination was demonstrated.

Distribution of serotypes of Campylobacter jejuni and C. coli from Danish patients, poultry, cattle and swine

The number of human cases of enteritis caused by Campylobacter jejuni and C. coli is increasing in Denmark and other European countries. No systemic typing has earlier been performed on Campylobacter isolates of Danish origin. The primary purpose of this study was to provide a serotype distribution of Campylobacter isolates from Danish patients and the major food production animals. In addition, the occurrence of intestinal carriers of thermophilic campylobacters among these food production animals was examined. In a nationwide survey, the individual isolation rate was 36% for broiler chickens, 47% for cattle and 46% for swine when sampled at the slaughterhouse. C. jejuni accounted for 83-91% of the thermophilic Campylobacter spp. in broiler chickens and cattle, whereas 95% of the isolates from swine was C. coli. In human patients with Campylobacter enteritis, 94% of the isolates were C. jejuni and 6% were C. coli. Heat-stable serotyping (the 'Penner scheme') was performed on a total of 398 isolates from the four sources: human patients (n = 145), broiler chickens (n = 94), swine (n = 111) and cattle (n = 48). Among human isolates, serotype O:1,44, O:2 and the O:4-complex accounted for 62% of the C. jejuni isolates. These serotypes were also common in samples from broilers and cattle. In swine, C. coli O:30 and O:46 were most common. The serotype distribution of human clinical isolates showed large overlap with the serotype distribution of campylobacters in cattle and chickens, and on this basis both could be major sources of human campylobacteriosis.

DNA fingerprinting of Campylobacter fetus using cloned constructs of ribosomal RNA and surface array protein genes

DNA fragments coding for the ribosomal RNA and the surface array proteins of Campylobacter fetus have been cloned from a genomic library constructed in Escherichia coli. They were used in the molecular characterization of C. fetus (subsp. fetus; subsp. venerealis) strains by restriction fragment length polymorphism (RFLP) method. Ribotyping results showed that all strains of the two subspecies can be classified under one ribogroup implying very close relatedness. The sapA gene DNA marker, however, discriminated all the strains regardless of the subspecies when chromosomal DNA was restricted with HindIII, HaeIII, XbaI or EcoRV. These results illustrate that the sapA probe is potentially useful in fingerprinting C. fetus strains and in determining the relationships of strains for epidemiological purposes.

A milk-borne campylobacter outbreak following an educational farm visit

After a nursery school trip to a dairy farm, 20 (53%) of 38 children and 3 (23%) of 13 adult helpers developed gastrointestinal infection. Campylobacter jejuni was isolated from 15 primary cases and from 3 of 9 secondary household cases. A cohort study of the school party found illness to be associated with drinking raw milk (relative risk 5.4, 95% confidence interval 1.4-20.4, P = 0.001). There was a significant dose response relationship between amount of raw milk consumed and risk of illness (chi 2-test for linear trend 12.1, P = 0.0005) but not with incubation period, severity of symptoms or duration of illness. All 18 human campylobacter isolates were C. jejuni resistotype 02 and either biotype I (number 16) or biotype II (number 2). Campylobacter was also isolated from samples of dairy cattle and bird faeces obtained at the farm but these were of different resisto/biotypes. Educational farm visits have become increasingly popular in recent years and this outbreak illustrates the hazard of exposure to raw milk in this setting.

Comparison of polymerase chain reaction and pulsed-field gel electrophoresis for the epidemiological typing of Campylobacter jejuni

Seventeen sporadic Campylobacter jejuni enteritis cases occurred in Taichung City, Taiwan between July 1995 and September 1995. Pulsed-field gel electrophoresis (PFGE) and enterobacterial repetitive intergenic consensus (ERIC-1) primed polymerase chain reaction (PCR) techniques were compared for the epidemiological typing of the 17 C. jejuni isolates. Fourteen distinct PFGE fingerprint patterns were observed. Fifteen distinct PCR fingerprint patterns were demonstrated. Two clusters of isolates (isolates 5 and 6; isolates 10, 11 respectively) were found to be genetically indistinguishable by both methods. In conclusion, we consider that PFGE is a highly reproducible method for determining the relatedness among the C. jejuni isolates in this study, although their limited numbers of restriction fragments may reduce the discriminatory power. Although less reproducible than PFGE typing, ERIC-1 primed PCR can be used as a simple and rapid tool to discriminate different strains of C. jejuni.

Resistotyping of campylobacters: fulfilling a need

A 9-month trial of a simple typing scheme for ¿thermophilic' enteric campylobacter isolates at a large Public Health Laboratory is described. Resistotyping was performed with six agents in a method modified by Bolton and colleagues from an earlier scheme, and biotyping was performed by a modified Lior scheme involving three tests. Reproducibility was excellent in both schemes, with test variation < 2%. Five household clusters and one larger presumptive milk-borne outbreak were identified in this scheme, and confirmed in pyrolysis mass spectrometry. The 328 isolates from new patients, excluding duplication from these clusters, were divided into 35 resistotypes with the largest group comprising 22% of isolates. In combined bio- and resistotyping, 86 types were found, with the largest group comprising 9.5% of isolates. The results are contrasted with salmonella sero- and phage-typing, where, on the same basis, the 176 isolates in the same period were divided into 40 groups, with the largest comprising 45% of isolates. Resistotyping, with or without additional biotyping, proved to be a convenient, simple, rapid, highly discriminatory, reproducible and inexpensive method well suited to use in local laboratories. It is a strong candidate for first-line national and local surveillance of campylobacter infections, fulfilling a need for monitoring of this important cause of enteric disease.

Comprehensive ribotyping scheme for heat-stable serotypes of Campylobacter jejuni

Strains from diverse sources belonging to all 47 heat-stable Penner serotypes of Campylobacter jejuni were examined for polymorphism around the 16S rRNA genes. Penner serotype reference strains and a group of nonserotypeable isolates were included in the study. Complete typeability was obtained; 30 distinct PstI and 42 HaeIII polymorphisms were found. Three bands were detected in almost all strains with these enzymes, confirming that three copies of the 16S rRNA gene are typical for C.jejuni. By combination of the two enzyme polymorphisms, 77 16S ribotypes were defined among the 261 strains analyzed. With two exceptions, no specific association was observed between these ribotypes and heat-stable serotypes. Nine serotypes were homogeneous with respect to the 16S ribotype. Most nonserotypeable strains belonged to ribotypes defined elsewhere in the study. The 16S ribotypes of C.jejuni described here were not found in strains of Campylobacter coli, and vice versa.

Ribotypes and AP-PCR fingerprints of thermophilic campylobacters from marine recreational waters

Thirty-two strains of thermophilic campylobacters isolated from marine recreational water and seven reference strains were biotyped and analysed by chromosomal DNA HaeIII ribopatterns and AP-PCR profiles based on a random 10-mer primer (5'-CAA TCG CCG T-3'). The majority of seawater isolates (90%) were Campylobacter coli, and three strains were Camp. jejuni. Southern blot hybridization analysis showed differences between the strains, and in a numerical analysis three main clusters were formed at the 45% similarity level, that corresponded to Camp. jejuni subsp. jejuni, Camp. coli, and a combination of Camp. coli and Camp. jejuni subsp. doylei. AP-PCR profiles also differentiated between the species but were less discriminatory than ribotyping because six strains (17%) could not be typed by this method. Numerical analysis gave four main clusters at the 45% similarity level, corresponding to Camp. jejuni subsp. jejuni, Camp. coli (two clusters) and Camp. lari. The study shows that strains within each species are diverse genomically. Both molecular methods were highly discriminatory, although some strains with identical ribotypes could be distinguished by AP-PCR, and they are valuable new alternatives to traditional typing in epidemiological studies of environmental campylobacters.

Comparison of PFGE, ribotyping and phage-typing in the epidemiological analysis of Campylobacter jejuni serotype HS2 infections

In this study we have evaluated the ability of three typing methods, pulsed field gel electrophoresis (PFGE), phage-typing and ribotyping, to discriminate not only between strains of differing serotypes but also between strains within a single serotype, heat stable serotype 2 (HS2). Forty-five isolates derived from cases of campylobacter enteritis occurring in the Cardiff area were examined. These included 18, mostly HS2, strains associated with an outbreak. The typing results for these and a further 39 epidemiologically unrelated strains of serotype HS2 were compared. This is the first report documenting the use of PFGE in an epidemiological investigation of Campylobacter jejuni in the UK. The results presented suggest that this technique is the most discriminatory of the three subtyping methods examined.

Random amplified polymorphic DNA fingerprinting of Campylobacter jejuni and C. coli isolated from human faeces, seawater and poultry products

The polymerase chain reaction (PCR) technique was used to obtain randomly amplified polymorphic DNA (RAPD) profiles from 64 type and serotype reference strains and 114 isolates of Campylobacter jejuni and C. coli from food, seawater and human faeces. Genetic diversity was detected among the strains as a total of 118 different RAPD profiles were obtained, each one containing from 4 to 11 bands between 0.30 and 1.50 kb. The discriminatory power of a random 10-mer primer (sequence 5'-CAATCGCCGT-3') was assessed. In general, no profiles were common to strains of the same Penner serogroup, but occasional strains from different Penner serotypes shared identical band profiles. RAPD analysis also differentiated between the species, and after numerical analysis, five main clusters were defined at the 40% similarity level, corresponding to C. jejuni, C. coli and C. lari with some exceptions. RAPD profiling of Campylobacter is highly discriminatory and is a valuable new alternative to traditional typing in epidemiological studies.

Molecular subtyping scheme for serotypes HS1 and HS4 of Campylobacter jejuni

We describe a molecular subtyping scheme for two principal O (heat-stable [HS]) serotypes of Campylobacter jejuni, HS1 and the HS4 complex. A 16S rRNA gene-specific probe confirmed that almost all the C. jejuni strains had three copies of this gene, and strains could be assigned with complete typeability to 1 of 16 combined (Pst1 and HaeIII) 16S ribotypes. Macrorestriction profiles (mrps) consisting of up to 10 SmaI fragments from approximately 40 to approximately 480 kbp were resolved by pulsed-field gel electrophoresis (PFGE). There were 11 mrps among the HS1 strains and 9 mrps among HS4 strains which corresponded to valid types--they occurred in multiple isolates, hosts, places, and times. There were 14 additional single-strain mrp fingerprints in HS1 and 20 in HS4. PFGE exhibited complete typeability when formaldehyde fixation of cells was employed, and PFGE was generally more differential than ribotyping. The data presented elucidate a high-resolution genotypic subtyping scheme for these common subspecific phenotypes of C. jejuni, which is both coherent and efficient for epidemiological purposes.

Flagellin gene polymorphism analysis of Campylobacter jejuni infecting man and other hosts and comparison with biotyping and somatic antigen serotyping

Flagellin gene sequence polymorphisms were used to discriminate amongst 77 strains of Campylobacter jejuni from sporadic and outbreak-associated human enteric infections, and from chickens, sheep and calves. The results were assessed in relation to Lior biotyping and serotyping (Penner somatic antigens). Eight DNA PCR-RFLP patterns (genotypes) were identified by analysis of HinfI fragment length polymorphisms in flagellin gene (flaA) polymerase chain reaction (PCR) products. One genotype (F-1) was a feature of 55% of strains. Strains within the genotypes were heterogeneous with respect to somatic antigens with 12 serogroups represented amongst the C. jejuni isolates of flaA type F-1. Serogroups Pen 1, 2 and 23 were the commonest (45%) amongst the 20 different serogroups represented. Several unique clusters of isolates with diverse biotypes were defined, and one cluster (F-7/Pen 23) contained epidemiologically implicated outbreak strains as well as sheep and calf isolates. We conclude that HinfI flaA typing is reproducible and offers high typability, and its combination with serogrouping provides a novel approach to characterizing isolates of C. jejuni with improved discrimination.

Molecular subtyping by genome and plasmid analysis of Campylobacter jejuni serogroups O1 and O2 (Penner) from sporadic and outbreak cases of human diarrhoea

Ribosomal RNA gene patterns, randomly amplified polymorphic genomic DNA (RAPD) profiles and plasmid profiles were used to discriminate between 28 strains of Campylobacter jejuni serogroups O1 and O2 (Penner). Most isolates were biotype I (Lior). The strains were representative isolates from a UK school outbreak of enteritis (7 cases) and from 21 sporadic human cases of enteritis in 4 countries. The molecular techniques discriminated to various degrees between strains in each of the serogroups. The outbreak strains were homogeneous in most molecular features but a variety of types was detected amongst the isolates from the sporadic cases. Five groups of two or more strains with identical ribopatterns were identified and within each, strains from different patients were homogenous with respect to serogroup. RAPD profile typing based on numerical analysis generally matched ribotyping. Plasmid profiling overall gave least discrimination but was useful in separating some strains similar in other features. We concluded that optimal discrimination of C. jejuni could best be achieved using a combination of phenotypic and genotypic properties. Hae III ribotyping was the single most discriminatory and reproducible technique investigated. Several strains of C. jejuni from sporadic infections had similar molecular profiles which have potential for general typing purposes.

Molecular typing of thermotolerant species of Campylobacter with ribosomal RNA gene patterns

Ribosomal RNA gene restriction patterns (ribopatterns) of 72 strains representing Campylobacter jejuni (subspecies jejuni and doylei), C. coli, C. upsaliensis and C. lari including the urealytic (UPTC) biovar were determined using four common restriction endonucleases (HaeIII, HindIII, PstI and PvuII). The relative effectiveness of these enzymes for general molecular typing of the thermotolerant campylobacters was assessed. Ribotypes (HaeIII) were defined on the basis of computer-assisted numerical analysis. For C. jejuni, C. coli and C. lari, the HaeIII ribopatterns provided a high level of typability and discrimination, with patterns that were reproducible and easy to code for numerical analysis. There was evidence of diversity within three of the HaeIII types, and PstI ribopatterns proved the most reliable for detecting such differences. C. upsaliensis also could be ribotyped with HaeIII, but HindIII, PvuII and PstI were less satisfactory for this species. For such campylobacters, the HindIII ribopatterns generally were complex and difficult to compare, and the PvuII profiles provided the least discrimination. We conclude that the choice of restriction endonuclease is of critical importance when applying ribotyping to different species of Campylobacter. HaeIII ribopatterns were the most effective means of typing strains of different thermotolerant species of campylobacters and, when combined with PstI ribopatterns, offered a highly discriminatory basis for molecular typing.

Flagellin gene typing of Campylobacter jejuni by restriction fragment length polymorphism analysis

We developed and studied a molecular typing approach for Campylobacter spp. with restriction fragment length polymorphism (RFLP) analysis of the flagellin gene flaA in C. jejuni. Using polymerase chain reaction, we amplified the flaA gene from strains comprising different HL:O serotypes by using a primer set directed at the conserved 5' and 3' flaA gene sequence to generate a 1.7-kb amplicon. The amplicon was further digested with the restriction enzyme DdeI, and the fragments generated were analyzed by agarose gel electrophoresis. In 43 non-outbreak strains of six common HL serotypes (HL 1, 2, 4, 5, 9, and 36) in the United States, 18 RFLP patterns were observed. In U.S. outbreak strains previously studied by 10 other typing methods, flaA typing correlated with the HL serotype within each outbreak, and six additional flaA types were identified. Our results suggest that RFLP analysis of the flaA gene from Campylobacter spp. has sufficient discrimination to be useful as a practical typing method for clinical and epidemiologic investigations.

Ribosomal RNA gene restriction patterns of Helicobacter pylori: analysis and appraisal of Hae III digests as a molecular typing system

Ribosomal RNA gene restriction patterns (ribopatterns) of 162 strains of Helicobacter pylori from 93 patients were studied to assess their suitability for use as the basis of a molecular typing system. Computer-assisted numerical analysis of Hae III ribopatterns of 122 strains from 9 countries in 4 continents showed that almost every strain had a distinct and unique ribopattern and only strains from the same individual were genomically matched in all bands or with minor (1-2 band) differences. Hae III ribopatterns offered high typability and reproducibility but were too discriminatory for large-scale epidemiological typing purposes because no rational basis for grouping strains was evident. In contrast, composite band profiles based on 21 band loci within the Hae III ribopatterns, which were used to compare strain sets selected on toxigenicity and geographical origin, were more conserved. Minor differences between some strain sets in several band loci were detected but generally the composite profiles were a reproducible feature of H. pylori. We conclude that Hae III ribopatterns provide an excellent fingerprint for small-scale studies of genomic variation in defined populations, such as sequential patient isolates, but are too specific for general typing of H. pylori.

Ribosomal RNA gene patterns of Helicobacter pylori from surgical patients with healed and recurrent peptic ulcers

Fifty-two strains of Helicobacter pylori were examined by DNA restriction endonuclease digestion, ribosomal (r)RNA gene probe hybridization and biotyping. Most (49) strains originated from gastric (antral) biopsies of patients before or after elective surgery for duodenal ulcers. Chromosomal DNA Hind III ribopatterns showed 9 strain clusters of which the largest contained 12 strains each with 3 common bands (1.50, 3.45, and 4.26 kb) but which were heterogeneous with respect to biotype and total digest pattern. Isolates from post-operative patients with either healed or recurrent ulcers showed ribopattern heterogeneity and exhibited a similar distribution of H. pylori ribopattern types; no single type predominated in any patient group or was more highly associated with recurrent ulcers than with healed ulcers. Multiple isolates from two surgical patients had only minor genomic variations in each set whereas isolates from two brothers had different ribopatterns. We conclude that Hind III ribopatterns in conjunction with total digest patterns might provide the basis for future epidemiological typing studies.

Chemiluminescent universal probe for bacterial ribotyping

We describe a novel procedure for direct covalent coupling of horseradish peroxidase to rRNA and ribotyping by using enhanced chemiluminescence. Compared with their 32P-end-labeled counterparts, chemiluminescent rRNA probes are stable and easy to synthesize and provide results as good as or superior to those obtained with isotopic labeling. Direct chemiluminescent labeling of Escherichia coli rRNA produces a sensitive, universal probe suitable for clinical laboratory use in the investigation of nosocomial outbreaks.

Characterization of Neisseria meningitidis isolated by ribosomal RNA gene restriction patterns and restriction endonuclease digestion of chromosomal DNA

The use of ribosomal RNA (rRNA) gene restriction patterns to study the molecular epidemiology of Neisseria meningitidis was investigated. Ninety-four isolates of Neisseria meningitidis were characterized by their rRNA gene restriction patterns with 16 + 23 S rRNA from Escherichia coli as a probe. Thirteen rRNA gene restriction patterns were recognized; each of these patterns represented between 1 and 30 isolates. Isolated with the outbreak-associated phenotype B15P1.16 (sulphonamide resistant) all gave a single rRNA gene restriction pattern but this pattern also contained isolates with other phenotypes. Further discrimination between isolates was achieved by comparison of banding patterns resulting from restriction endonuclease digestion of chromosomal DNA with Bgl II. This gave a banding pattern consisting of about ten bands which was simple to interpret. Using this technique 94 isolates were classified in 54 patterns containing between 1 and 14 isolates. Restriction endonuclease analysis with Bgl II characterized outbreak-associated isolates with the phenotype B15P1.16 and enabled strains not typable by conventional methods to be identified as probable outbreak-associated isolates. The techniques should prove useful for epidemiological studies.