Robert C. Clarke

Amplification of an invA gene sequence of Salmonella typhimurium by polymerase chain reaction as a specific method of detection of Salmonella

Amplification of nucleotide sequences within the invA gene of Salmonella typhimurium was evaluated as a means of detecting Salmonella. A collection of 630 strains of Salmonella comprising over 100 serovars, including the 20 most prevalent serovars isolated from animals and humans in Canada, was examined. Controls consisted of 142 non-Salmonella strains comprising 21 genera of bacteria. Cultures were screened by inoculating a single colony of bacteria directly into a polymerase chain reaction (PCR) mixture which contained a pair of primers specific for the invA gene. The specific PCR product was a 284 bp DNA fragment which was visualized in 2% agarose gels. With the exception of two S. litchfield and two S. senftenberg strains, all Salmonella strains were detected. In contrast, none of the non-Salmonella strains yielded the specific amplification product. Non-specific amplification of a few non-Salmonella strains resulted in a product that was distinctly different in size from the specific 284 bp product. Specificity of amplification was further confirmed by demonstration of hybridization of a 32P-labelled invA gene fragment only to the specific 284 bp product. The detection of 99.4% of Salmonella strains tested and the failure to specifically amplify DNA from non-Salmonella strains confirm that the invA gene contains sequences unique to Salmonella and demonstrate that this gene is a suitable PCR target, with potential diagnostic applications.

Growing concerns and recent outbreaks involving non-O157:H7 serotypes of verotoxigenic Escherichia coli

Verocytotoxin-producing E. coli (VTEC) of serotype O157:H7 have been shown to be important agents of foodborne disease in humans worldwide. While the majority of research effort has been targeted on this serotype it is becoming more evident that other serotypes of VTEC can also be associated with human disease. An increasing number of these non-O157:H7 VTEC have been isolated from humans suffering from HUS and diarrhea. Recently a number of foodborne outbreaks in the USA, Australia, and other countries have been attributed to non-O157:H7 VTEC serotypes. Surveys of animal populations in a variety of countries have shown that the cattle reservoir contains more than 100 serotypes of VTEC, many of which are similar to those isolated from humans. The diversity and complexity of the VTEC family requires that laboratories and public health surveillance systems have the ability to detect and monitor all serotypes of VTEC.

Prevalence of the eaeA gene in verotoxigenic Escherichia coli strains from dairy cattle in Southwest Ontario

This study determined the prevalence of the eaeA gene and its relationship to serotype and type of verotoxin produced in a collection of 432 verotoxigenic Escherichia coli (VTEC) obtained from the faeces of healthy cows and calves in a systematic random survey involving 80 dairy farms in Southwest Ontario. A PCR amplification procedure involving primer pairs which target the conserved central region of the O157:H7 eaeA gene showed that 151 (35.2%) strains were positive for the eaeA gene. All isolates (9-21 for each O group) of O groups 5, 26, 69, 84, 103, 111, 145 and 157 were positive, whereas all isolates (7-34 for each O group) of O groups 113, 132, and 153 and serotype O156:NM (38 isolates) were negative for eaeA. Seventy-three percent of 130 isolates of eaeA-positive serotypes produced VT1 only compared with 20% of 253 isolates of eaeA-negative serotypes. We conclude that there is a strong association between certain O groups and the eaeA gene, that serotypes of eaeA-positive and eaeA-negative VTEC implicated in human and cattle disease are present at high frequency in the faeces of healthy cattle, that VT1 is more frequently associated with eaeA-positive than with eaeA-negative serogroups, and that the eaeA gene is more frequently found in VTEC from calves compared with VTEC from adult cattle.

Polymerase chain reaction for detection of verocytotoxigenic Escherichia coli isolated from animal and food sources

Animals and their by-products have been implicated as important sources of verocytotoxigenic Escherichia coli (VTEC) associated with disease in humans. VTEC comprise a wide range of serotypes and produce a variety of closely related verocytotoxins (VT). A pair of oligonucleotide primers, targeting conserved sequences found in VT1, VT2 and VTE genes, was used to develop a polymerase chain reaction (PCR) procedure to detect all types of VTEC. Supernatants of boiled broth cultures of VTEC (223 strains) isolated from ground beef, ground pork, raw milk, bovine faeces and porcine faeces; non-VTEC E. coli (72 strains); and other enteric and food bacteria (76 strains) were tested by PCR. The verocytotoxigenicity of these strains was verified by the Vero cell assay. All 223 VTEC isolates, comprising over 50 different serotypes, were detected by the PCR procedure. Shigella dysenteriae type 1 was the only other bacterium that was positive in this assay. As little as 1 pg of VTEC DNA and as few as 17 cfu of VTEC could be detected with this method. The results indicate that these primers detect VTEC over a wide range of serotypes. This method may be applicable as a screening procedure for the detection of VTEC in samples of foods and faeces.

Prevalence of verocytotoxigenic Escherichia coli in ground beef, pork, and chicken in southwestern Ontario

Samples of ground beef (225), pork (235) and chicken (200) were randomly selected from meat processing plants in the southwestern Ontario area. Supernatants of broth cultures of the samples were tested for verocytotoxins using a Vero cell assay. Neutralization of cytotoxic activity using antisera specific for three types of verocytotoxin (Verotoxin 1, Verotoxin 2 and Shiga-like toxin II) was performed on positive samples. Isolation of verocytotoxigenic Escherichia coli (VTEC) was attempted from positive samples. VTEC were confirmed as E. coli biochemically, tested for drug resistance, and serotyped. Based on neutralization studies, the prevalence of VTEC in beef and pork was at least 36.4% and 10.6%, respectively. This is much higher than has been reported from a survey of retail meats in which a method designed to detect only E. coli O 157.H7 was used. Isolations of VTEC were made from 10.4% of the beef samples and 3.8% of the pork samples. No VTEC were recovered from the chicken samples. The majority of VTEC isolates were susceptible to commonly used antimicrobial agents. A number of the serotypes of the VTEC isolates recovered have been associated with human disease; however, no VTEC of serotype O 157.H7 were isolated.

Distribution and characteristics of verocytotoxigenic Escherichia coli isolated from Ontario dairy cattle.

Faecal swabs obtained from a random sample of 1131 cows and 659 calves on 100 southern Ontario dairy farms were examined for verocytotoxigenic Escherichia coli (VTEC) using a Vero cell assay. Five isolates from each positive culture were tested similarly. Positive colonies were examined with DNA probes for Shiga-like toxin I (SLT-I) and SLT-II sequences. Probe-negative colonies were tested for neutralization of verocytotoxicity using anti-SLT-I and anti-SLT-IIv antisera. Colonies showing no neutralization response were examined in a polymerase chain reaction procedure. Colonies positive by any test were confirmed to be E. coli biochemically, serotyped, biotyped and tested for antimicrobial resistance. Faecal culture supernatants which were positive in the Vero cell assay, but culture negative, were examined using the neutralization assay. Animals were classified positive by faecal culture supernatant or by positive VTEC isolate. The prevalence rates of VTEC infection in cows and calves were estimated to be 9.5 and 24.7%, respectively. The proportion of animals infected on each farm ranged from 0 to 60% for cows and 0 to 100% for calves. Of 206 VTEC isolates identified, few were of serotypes which have been isolated from humans and none were E. coli O 157.H7.

The prevalence of Salmonella enteritidis and other Salmonella spp. among Canadian registered commercial layer flocks

A survey was conducted to estimate the prevalence of Salmonella enteritidis and other salmonellas among Canadian commercial egg producing flocks. Environmental (faecal and eggbelt) samples from 152 of 295 (52.9%) randomly selected flocks were contaminated with salmonellas. Thirty-five different salmonella serovars were isolated. Eggbelt samples were more often contaminated with salmonellas than faecal samples (25.7 v. 10.1%). The most prevalent serovars were S. heidelberg, S. infantis, S. hadar, and S. schwarzengrund; they were isolated from samples of 59/295 (20%), 18/295 (6.1%), 17/295 (5.8%), and 15/295 (5.1%) flocks, respectively. Feed samples of 21/295 (7.2%) flocks were contaminated with salmonellas. Salmonella enteritidis was isolated from the environmental samples of 8/295 (2.7%) flocks. Salmonella enteritidis phage type (PT) 8 was isolated from 5 flocks, PT 13a from 2 flocks, and PT 13 from 1 flock.

Isolation of verocytotoxin-producing Escherichia coli from milk filters in south-western Ontario.

A total of 1012 milk filters collected from 498 dairy farms in south-western Ontario during three study periods (December 1985-March 1986) were tested for the presence of verocytotoxin-producing Escherichia coli (VTEC). VTEC were detected and isolated using a Vero cell assay. Supernatants from 20 of the milk filter cultures had verocytotoxic activity and 7 VTEC strains were isolated from these positive samples. The prevalence of VTEC in the samples in each of the three study periods were 0.44, 0.65 and 0.99% respectively. All seven VTEC strains isolated were sensitive to commonly used antimicrobial agents. The serotypes of these strains were O 26.H11, O43.H2, O 153.H25, O ?. H8, O? .H19, O ?. non-motile, and Orough.H19. Two of these serotypes (O 153.H25 and O 26.H11) have previously been associated with disease in humans.

Risk factors for bovine infection with verocytotoxigenic Escherichia coli in Ontario, Canada

Risk factors for prevalent infection with verocytotoxigenic Escherichia coli (VTEC) were studied in a random sample of 1131 cows and 659 calves < 3 months of age on 100 dairy farms in southern Ontario. Faecal-culture supernatants from each animal were screened for verocytotoxicity using a Vero cell assay. Five E. coli isolates from positive samples were tested for verocytotoxin production using a test sequence consisting of (1) hybridization with DNA probes for Shiga-like toxin I (SLT-I) and SLT-II sequences, (2) neutralization of verocytotoxicity by a mixture of anti-SLT-I and anti-SLT-IIv antisera, and (3) amplification of SLT genes by a polymerase chain reaction procedure. Neutralization assays were performed on positive faecal-culture supernatants from which no positive isolates were obtained. Farm managers were interviewed using a standardized questionnaire to obtain information on farm and individual animal-level characteristics and management practices. Calves > 2 weeks of age were at significantly greater risk of infection than those under 2 weeks (OR = 2.0). Farm-level calf infection was negatively associated with herd size, the use of nipple bottles for feeding calves, the use of traditional tie-stall housing as opposed to loose housing or other methods, and the maintenance of a closed herd. Farm-level cow infection was negatively associated with herd size, the use of loose housing, the maintenance of a relatively open herd, and the use of milking parlours or pipelines as opposed to bucket-type milking machines. These variables provide potential markers for the VTEC infection status of dairy herds in the target population, but the causal mechanisms underlying these observed associations are unknown.

Estimation of the under-reporting rate for the surveillance of Escherichia coli O157:H7 cases in Ontario, Canada.

Two models estimating the proportion of Escherichia coli O157:H7 cases not reported in the Ontario notifiable diseases surveillance system are described. The first model is a linear series of adjustments in which the total number of reported cases is corrected by successive underreporting coefficients. The structure of the second model is based on a relative difference in the proportion of E. coli O157:H7 cases which are hospitalized between the surveillance database and the underlying population. Based on this analysis, the rate of under-reporting of symptomatic cases of E. coli O157:H7 infection in Ontario ranges from 78 to 88% corresponding to a ratio of 1 reported case for approximately 4-8 symptomatic cases missed by the surveillance system. This study highlights the need to increase awareness among public health workers of the potential biases that may exist in the interpretation of routine surveillance data.