Antimicrobial resistance and virulence characteristics in 3 collections of staphylococci from bovine milk samples.

Mastitis is a prevalent disease in dairy cattle, and staphylococci are among the most common causative pathogens. Staphylococci can express resistance to a range of antimicrobials, of which methicillin resistance is of particular public health concern. Additionally, Staphylococcus aureus carries a variety of virulence factors, although less is understood about the virulence of non-aureus staphylococci (NAS). The aim of our study was to identify and characterize 3 collections of staphylococcal isolates from bovine milk samples regarding antimicrobial resistance, with emphasis on methicillin resistance, and their carriage of virulence genes typically displayed by Staph. aureus. A total of 272 staphylococcal isolates collected in Norway and Belgium in 2016 were included, distributed as follows: group 1, Norway, 100 isolates; group 2, Flanders, Belgium, 64 isolates; group 3, Wallonia, Belgium, 108 isolates. Species identification was performed by use of MALDI-TOF mass spectrometry. Phenotypic resistance was determined via disk diffusion, and PCR was used for detection of methicillin resistance genes, mecA and mecC, and virulence genes. Antimicrobial resistance was common in Staphylococcus epidermidis and Staphylococcus haemolyticus from all different groups, with resistance to trimethoprim-sulfonamide frequently occurring in Staph. epidermidis and Staph. haemolyticus as well as in Staph. aureus. Resistance to penicillin was most frequently observed in group 1. Ten Belgian isolates (1 from group 2, 9 from group 3) carried the methicillin resistance determinant mecA: 5 Staph. aureus from 2 different farms and 5 NAS from 3 different farms. Almost all Staph. aureus isolates were positive for at least 3 of the screened virulence genes, whereas, in total, only 8 NAS isolates harbored any of the same genes. Our study contributes to the continuous need for knowledge regarding staphylococci from food-producing animals as a basis for better understanding of occurrence of resistance and virulence traits in these bacteria.

Genetic characterization of Shigatoxigenic and enteropathogenic Escherichia coli O80:H2 from diarrhoeic and septicaemic calves and relatedness to human Shigatoxigenic E. coli O80:H2.

AIM: The purpose of this work was to identify and genetically characterize enterohemorrhagic Escherichia coli (EHEC) and enteropathogenic E. coli (EPEC) O80:H2 from diarrhoeic and septicaemic calves in Belgium and to comparing them with human EHEC after whole genome sequencing. METHODS AND RESULTS: Ten EHEC and 21 EPEC O80 identified by PCR between 2009 and 2018 from faeces, intestinal content and a kidney of diarrhoeic or septicaemic calves were genome sequenced and compared to 19 human EHEC identified between 2008 and 2019. They all belonged to the O80:H2 serotype and ST301, harboured the eaeξ gene, and 23 of the 29 EHEC contained the stx2d gene. Phylogenetically, they were distributed in two major sub-lineages: one comprised a majority of bovine EPEC whereas the second one comprised a majority of stx2d bovine and human EHEC. CONCLUSIONS: Not only EPEC but also EHEC O80:H2 are present in diarrhoeic and septicaemic calves in Belgium and are genetically related to human EHEC. SIGNIFICANCE AND IMPACT OF THE STUDY: These findings support the need to assess cattle as potential source of contamination of humans by EHEC O80:H2 and to understand the evolution of bovine and human EHEC and EPEC O80:H2.

Comparison of quantitative PCR and MALDI-TOF mass spectrometry assays for identification of bacteria in milk samples from cows with subclinical mastitis.

AIMS: The objective of this study was to compare qualitatively and quantitatively the results of identification of the bacteria present in milk samples from cows with subclinical mastitis using multiplex qPCR assay and matrix-assisted laser desorption-ionization time of flight mass spectrometry (MALDI-TOF MS® ) after bacteriological growth. METHODS AND RESULTS: A total of 182 samples were aseptically collected from 119 cows with high somatic cell counts (>2·105 SCC per ml) on 11 farms in Belgium in 2014. The mutiplex qPCR assay was carried out on 350 µl of milk with the PathoProof® Complete-16kit. Ten microlitre of milk was streaked on Columbia blood agar and three selective agar plates. Growing colonies were identified by MALDI-TOF MS. Of the 182 samples, 90 gave positive results with either or both tests for one or two bacterial species/genera. Total qualitative agreement of the bacteria identified was observed in 41 mono- or bi-bacterial samples (46%) and partial agreement in 19 bi-bacterial samples at both or either tests (21%). The results of both tests on those mono- and bi-bacterial samples were not significantly different (McNemar test; P = 0·395) with a fair agreement (Cohen's kappa test; k = 0·375; P = 0·055). Moreover, quantitative correlation between the qPCR intensity and the numbers of growing colonies was observed in half of the 60 samples with qualitative matching results. CONCLUSIONS: Both methods give identical qualitative and quantitative results with approximately a half and a quarter of the mono- and bi-bacterial samples respectively. Several reasons can explain the differences. The multiplex qPCR assay only targets the most important mammary gland pathogens and can detect DNA of bacteria both alive and dead. Conversely, bacteria only grow when alive and the MALDI-TOF MS databases do not include all bovine milk-associated bacterial species yet. SIGNIFICANCE AND IMPACT OF THE STUDY: This study further highlights the limitations and complementarity of the genetic and phenotypic tests for the identification of bacteria present in milk samples.

Low prevalence of the 'gang of seven' and absence of the O80:H2 serotypes among Shigatoxigenic and enteropathogenic Escherichia coli (STEC and EPEC) in intestinal contents of healthy cattle at two slaughterhouses in Belgium in 2014.

AIMS: The purpose of this survey was to estimate the respective prevalence of the 'gang of seven' and 'non-gang of seven' serotypes of Shigatoxigenic and enteropathogenic Escherichia coli and to identify the O80:H2 serotype in 245 intestinal contents collected at two slaughterhouses in Belgium in 2014. METHODS AND RESULTS: After overnight enrichment growth, the 69 intestinal contents testing positive with PCR targeting the eae, stx1 and stx2 genes were inoculated onto four agar media. Of the 2542 colonies picked up, 677 from 59 samples were PCR confirmed. The most frequent virulotypes were eae+ in 47 (80%) samples, stx2+ in 20 (34%) samples and eae+ stx1+ in 16 (27%) samples. PCR-positive colonies belonged to different virulotypes in 36 samples. No colony was O80-positive, whereas two eae+ colonies from two samples were O26:H11, 50 eae+ stx1+ and eae+ from eight samples were O103:H2 and two eae+ stx1+ stx2+ colonies from one sample were O157:H7. CONCLUSIONS: The 'non-gang of seven' serotypes are more frequent than the 'gang of seven' serotypes and the O80:H2 serotype was not detected among Shigatoxigenic and enteropathogenic Escherichia coli in the intestines of cattle at these two slaughterhouses. SIGNIFICANCE AND IMPACT OF THE STUDY: Although the identification protocols of Shigatoxigenic Escherichia coli focus on the 'gang of seven' serotypes, several other serotypes can be present with possible importance in public health. Innovative selective identification procedures should be designed.

Identification of Shiga toxin-producing (STEC) and enteropathogenic (EPEC) Escherichia coli in diarrhoeic calves and comparative genomics of O5 bovine and human STEC.

Escherichia coli producing Shiga toxins (Stx) and the attaching-effacing (AE) lesion (AE-STEC) are responsible for (bloody) diarrhoea in humans and calves while the enteropathogenic E. coli (EPEC) producing the AE lesion only cause non-bloody diarrhoea in all mammals. The purpose of this study was (i) to identify the pathotypes of enterohaemolysin-producing E. coli isolated between 2009 and 2013 on EHLY agar from less than 2 month-old diarrhoeic calves with a triplex PCR targeting the stx1, stx2, eae virulence genes; (ii) to serotype the positive isolates with PCR targeting the genes coding for ten most frequent and pathogenic human and calf STEC O serogroups; and (iii) to compare the MLSTypes and virulotypes of calf and human O5 AE-STEC after Whole Genome Sequencing using two server databases (www.genomicepidemiology.org). Of 233 isolates, 206 were triplex PCR-positive: 119 AE-STEC (58%), 78 EPEC (38%) and 9 STEC (4%); and the stx1+eae+ AE-STEC (49.5%) were the most frequent. Of them, 120 isolates (84% of AE-STEC, 23% of EPEC, 22% of STEC) tested positive with one O serogroup PCR: 57 for O26 (47.5%), 36 for O111 (30%), 10 for O103 (8%) and 8 for O5 (7%) serogroups. The analysis of the draft sequences of 15 O5 AE-STEC could not identify any difference correlated to the host. As a conclusion, (i) the AE-STEC associated with diarrhoea in young calves still belong to the same serogroups as previously (O5, O26, O111) but the O103 serogroup may be emerging, (ii) the O5 AE-STEC from calves and humans are genetically similar.

Genotypic and phenotypic characterization of methicillin-resistant Staphylococcus aureus (MRSA) isolated from milk of bovine mastitis.

The aim of this study was to evaluate the presence of methicillin-resistant Staphylococcus aureus (MRSA) among a (S. aureus) collection (n = 430) isolated from milk of cows suffering from mastitis in Belgium and to compare their genotypic as well as phenotypic characteristics. Pulsed field gel electrophoresis (PFGE) and PCR-based typing techniques (MLST, spa, SCCmec, and agr typing) have been applied and supplemented by capsule serotyping, biofilm production quantification and antimicrobial susceptibility testing. Nineteen MRSA were isolated. Seven distinct ApaI PFGE patterns were observed. All isolates, except one, were identified as ST398 strains. Three spa types (t011, t567 and t108) and two SCCmec types (IV and V) were identified. All isolates belonged to agr type I and capsule type 5 and were Panton-Valentine leukocidin (PVL) negative. All isolates produced biofilm in TSBglc , whereas the majority did not in milk serum. Twelve resistance patterns were observed, with almost two-thirds of the isolates being resistant to at least six antibiotics, including penicillin and tetracycline. Our study confirms that the emerging ST398 LA-MRSA clone has attained Belgian cattle. With regard to genotypic and phenotypic typing, the 19 MRSA isolated in this study form a homogenous group and do not differ much from one another, neither from what has been previously described.

Typing of O26 enterohaemorrhagic and enteropathogenic Escherichia coli isolated from humans and cattle with IS621 multiplex PCR-based fingerprinting.

AIMS: This study evaluated a typing method of O26:H11 enterohaemorrhagic and enteropathogenic Escherichia coli (EHEC and EPEC) based on the variation in genomic location and copy numbers of IS621. METHODS AND RESULTS: Two multiplex PCRs, targeting either the left (5') or right (3') IS/chromosome junction of 12 IS621 insertion sites and one PCR specific of another truncated copy, were developed. Thirty-eight amplification profiles were observed amongst a collection of 69 human and bovine O26:H11 EHEC and EPEC. Seventy-one per cent of the 45 EHEC and EPEC with identical IS621 fingerprints within groups of two, three or four isolates had >85% pulsed field gel electrophoresis (PFGE) profile similarity, including four groups of epidemiologically related EHEC or EPEC, while most of the groups had <85% similarity between each others. Epidemiologically related EHEC from each of three independent outbreaks in Japan and Belgium also exhibited identical IS621 fingerprints and PFGE profiles. CONCLUSIONS: The IS621 fingerprinting and the PFGE are complementary typing assays of EHEC and EPEC; though, the former is less discriminatory. SIGNIFICANCE AND IMPACT OF THE STUDY: The IS621 printing method represents a rapid (24 h) first-line surveillance and typing assay, to compare and trace back O26:H11 EHEC and EPEC during surveys in farms, multiple human cases and outbreaks.

Cattle enterotoxaemia and Clostridium perfringens: description, diagnosis and prophylaxis.

Cattle enterotoxaemia is one of numerous pathologies caused by Clostridium perfringens. These anaerobic Gram-positive bacteria are naturally present in the intestinal flora of mammals, but their uncontrolled multiplication under certain conditions results in the overproduction of toxins in the intestinal tract. Major clinical signs are induced by the systemic spread of these toxins in the blood and tissues. Enterotoxaemia may be acute or peracute, and sudden death is often reported in rapidly growing, apparently healthy cattle. Enterotoxaemia can be prevented only with better understanding of its risk factors and pathogenesis. This paper provides an up-to-date overview of knowledge concerning the aetiology of the syndrome, its epidemiological context, pathogenesis, clinical signs and lesions, the diagnostic procedures and prophylactic tools, with specific attention to field aspects that are directly relevant to practitioners and clinical researchers.

Purification of the recombinant beta2 toxin (CPB2) from an enterotoxaemic bovine Clostridium perfringens strain and production of a specific immune serum.

Overgrowth of Clostridium perfringens clones with production of one or more of its toxin(s) results in diverse digestive and systemic pathologies in human and animals, such as cattle enterotoxaemia. The so-called beta2 toxin (CPB2) is the most recently described major toxin produced by C. perfringens. In this study, the cpb2 ORF (cpb2FM) from a cattle C. perfringens-associated enterotoxaemia was cloned and sequenced. The cpb2FM and its deduced nucleotide sequence clearly corresponded to the cpb2 allele considered as "consensus" and not to "atypical" allele, despite its "non-porcine" origin. Expression assays of the recombinant toxin CPB2FM were performed in Escherichia coli and Bacillus subtilis with the expression vector pBLTS72, and by genomic integration by double recombination in B. subtilis. Highest level of production was obtained with the expression vector in B. subtilis 168 strain. The recombinant CPB2FM protein was purified and a specific rabbit polyclonal antiserum was produced. Polyclonal antibodies could detect CPB2 production in supernatants of C. perfringens from enterotoxaemic cattle.

The expression of Clostridium perfringens consensus beta2 toxin is associated with bovine enterotoxaemia syndrome.

Clostridium perfringens has been implicated in a broad array of enteric infections including the fatal haemorrhagic enteritis/enterotoxaemia syndrome in cattle. The beta2 toxin (CPB2), encoded by cpb2, is suspected to be implicated in this syndrome. However, among C. perfringens isolates from cattle suspected of clostridial disease, an atypical allele was recently found to predominate at the cpb2 locus and atypical corresponding CPB2 proteins were shown to be poorly expressed, thus arguing against a biologically significant role of the beta2 toxin in clostridial diseases in cattle. This study compared genotype and phenotype of the beta2 toxin between C. perfringens isolates from a group of healthy calves (n=14, 87 isolates) and from a group of enterotoxaemic calves (n=8, 41 isolates). PCR results revealed the exclusive presence of the typical "consensus"cpb2 in the enterotoxaemic group. Western blot analysis demonstrated that the typical variant of CPB2 was often expressed in isolates from enterotoxaemic calves (43.9%) and infrequently in isolates from healthy cattle (6.9%). These data suggest that the typical variant of the CPB2 toxin may play a role in the pathogenesis of cattle enterotoxaemia.

2F3 monoclonal antibody recognizes the O26 O-antigen moiety of the lipopolysaccharide of enterohemorrhagic Escherichia coli strain 4276.

Enterohemorrhagic Escherichia coli (EHEC) and enteropathogenic E. coli (EPEC) organisms are groups of pathogenic strains whose infections are characterized by a typical lesion of enterocyte attachment and effacement. They are involved in enteric diseases both in humans and in animals, and EHEC strains can be responsible for hemolytic uremic syndrome in humans. Previously, it was shown that the 2F3 monoclonal antibody (MAb) is specific for the O26 EHEC and EPEC strains (P. Kerr, H. Ball, B. China, J. Mainil, D. Finlay, D. Pollock, I. Wilson, and D. Mackie, Clin. Diagn. Lab. Immunol. 6:610-614, 1999). As these groups of bacteria play an important role in pathology, the aim of this paper was to characterize the antigen recognized by the 2F3 MAb and its genetic determinant. A genomic locus containing the entire O-antigen gene cluster and half of the colanic acid gene cluster from an O26 EHEC strain was shown to be sufficient for the production of the antigen recognized by the 2F3 MAb in an E. coli DH5alpha strain. By transposon mutagenesis performed on the recombinant plasmid, all 2F3 enzyme-linked immunosorbent assay-negative mutants had their transposons inserted into the O-antigen gene cluster. The O-antigen gene cluster was also cloned from an O26 EHEC strain into the E. coli DH5alpha strain, which then produced a positive result with the 2F3 MAb. Further analysis of the type of lipopolysaccharides (smooth or rough) produced by the clones and mutants and of the O antigen of the 2F3-positive clones confirmed that the epitope recognized by the 2F3 MAb is located on the O antigen in the O26 EHEC and EPEC strains and that its genetic determinant is located inside the O-antigen gene cluster.

Adherence to various host cell lines of Mycoplasma bovis strains differing in pathogenic and cultural features.

Mycoplasma bovis is known to be responsible for pneumonia and arthritis in calves, as well as mastitis in dairy cows. Despite clear evidence of its pathogenic potential, little is known about mechanisms of cytadherence and the molecular factors involved. The purpose of this work was to compare adherence rates of M. bovis field strains to different host cell lines and study the effects of cloning and sub-culturing M. bovis strains on their adherence properties. Eighteen metabolically labeled M. bovis strains isolated from different pathological backgrounds were examined in adherence trials using four different host cell lines, i.e. embryonic bovine lung (EBL), embryonic bovine trachea (EBTr), Madin Darby bovine kidney (MDBK) and rabbit kidney (RK) cells. Although large interstrain variations in adherence rates (3.4-19.1%) were measured they could not be correlated to the pathological background (pneumonia, arthritis or mastitis). Adherence rates to the fibroblast cell line (EBTr) were significantly lower than those to the three epithelial cell lines (EBL, MDBK and RK). The only non-pathogenic strain (221/89) exhibited lower adherence rates than three isolates from clinical mastitis. Interestingly, adherence rates were significantly reduced after in vitro passaging. In contrast, no effect of single cloning of strains on adherence was observed. There was no general correlation between expression of variable surface proteins (Vsps) as monitored by immunoblotting and adherence rates, although alterations in Vsp expression profiles were seen as a consequence of passaging. As there is probably a large number of adhesins, variable and non-variable, on the surface of M. bovis cells the issue is very complex, and the most active components have yet to be identified.

A role for the Clostridium perfringens beta2 toxin in bovine enterotoxaemia?

Non-enterotoxigenic type A Clostridium perfringens are associated with bovine enterotoxaemia, but the alpha toxin is not regarded as responsible for the production of typical lesions of necrotic and haemorrhagic enteritis. The purpose of this study was to investigate the putative role of the more recently described beta2 toxin. Seven hundred and fourteen non-enterotoxigenic type A C. perfringens isolated from 133 calves with lesions of enterotoxaemia and high clostridial cell counts (study population) and 386 isolated from a control population of 87 calves were tested by a colony hybridisation assay for the beta2 toxin. Two hundred and eighteen (31%) C. perfringens isolated from 83 calves (62%) of the study population and 113 (29%) C. perfringens isolated from 51 calves (59%) of the control population tested positive with the beta2 probe. Pure and mixed cultures of four C. perfringens (one alpha+beta2+, one alpha+enterotoxin+ and two alpha+) were tested in the ligated loop assay in one calf. Macroscopic haemorrhages of the intestinal wall, necrosis and haemorrhages of the intestinal content, and microscopic lesions of necrosis and polymorphonuclear and mononuclear cell infiltration of the intestinal villi were more pronounced in loops inoculated with the alpha and beta2-toxigenic C. perfringens isolate. These results suggest in vivo synergistic role of the alpha and beta2 toxins in the production of necrotic and haemorrhagic lesions of the small intestine in cases of bovine enterotoxaemia. However, isolation of beta2-toxigenic C. perfringens does not confirm the clinical diagnosis of bovine enterotoxaemia and a clostridial cell counts must still be performed.

Replicon typing of F18 fimbriae encoding plasmids of enterotoxigenic and verotoxigenic Escherichia coli strains from porcine postweaning diarrhoea and oedema disease.

The presence of fimbrial adhesin F18 is frequently found in enterotoxigenic Escherichia coli (ETEC) and verotoxigenic E. coli (VTEC) strains responsible for diarrhoea and oedema disease of weaned pigs. The F18 adhesin occurs in two antigenic variants: F18ab is characteristic of VTEC while F18ac is more typical for ETEC. F18 encoding plasmids of 17 phenotypically characterized porcine E. coli isolates (10 ETEC, 6 VTEC and 1 ETEC/VTEC) were tested with a DNA probe for F18 fimbrial adhesin and with replicon probes for the RepFIa, RepFIb and for the RepFIc family of basic replicons. In all the cases, the F18 probe hybridized to only one plasmid band of size higher than 42MDa. All F18 plasmids were determined to be unireplicon plasmids belonging to the RepFIc replicon family of the F incompatibility complex. There was no difference between F18ac plasmids of ETEC and F18ab plasmids of VTEC strains in terms of replicon type or subtype. However, the size of F18ab plasmids of the VTEC strains varied between 42 and 98MDa, in contrast to F18ac plasmids of ETEC strains (constantly approximately 98MDa).

Genotypic and phenotypic characterization of potential virulence of intestinal avian Escherichia coli strains isolated in Algeria.

In order to characterize potential pathogenic Escherichia coli strains isolated from diarrheic hens and chickens originating from intensive battery rearing in North Algeria, the presence of a large range of virulence factors and markers was studied in 50 strains by DNA-DNA hybridization on colonies and phenotypic tests. The sequences we focused on were those coding for adhesins F5, F41, F17, Pap, Afa, and Sfa; intimin Eae; and toxins STa, STb, LT1, Stx1, Stx2, CNF1, and CNF2. The phenotypes explored were the colicins, aerobactin, hemolysins, and hemagglutinin production and serum resistance. The genotypic and phenotypic tests enabled us to categorize the isolates into two distinct groups: those with a potential to invade the host (27 strains were serum resistant and/or produced aerobactin), among which three strains were also potentially diarrheagenic, one strain was LT1 + F17+ Afa+ Pap+ (enterotoxigenic E. coli) and the two others were Stx1 (verotoxigenic E. coli). Twenty-three strains were colicinogenic, including 19 strains producing colicin V. This latter factor was also detected in isolates negative for the other virulence factors. On the basis of the type of erythrocytes agglutinated, we established 14 mannose-resistant hemagglutination patterns among the 37 strains tested, including 22 serum-resistant and/or aerobactin producing strains and 15 strains negative for these two characters. None of the strains produced alpha hemolysin, whereas two strains produced beta hemolysin and enterohemolysin, respectively. Congo red fixation was observed in 25 strains. No relationship could be detected between Congo red fixation and the presence of other virulence markers, such as serum resistance and aerobactin production. This study shows that among isolates originating from the feces of diarrheic chickens, the proportion of potentially diarrheagenic E. coli strains is low.

Bacterial intestinal flora associated with enterotoxaemia in Belgian Blue calves.

The enterotoxaemia syndrome in Belgian Blue calves is characterised by a high case fatality rate, sudden death, lesions of haemorrhagic enteritis of the small intestine and, quite often an absence of other clinical signs but its cause has not been yet identified. As a first step in this identification, the aerobic and anaerobic intestinal flora of a population of 78 calves, originating from farms located in southern Belgium and that died in circumstances defined as "calf enterotoxaemia" (study population) and of 64 calves that died in other circumstances (control population) were studied qualitatively and quantitatively. The colonies were identified after subcultures with appropriate API sugar sets. Anaerobically Clostridium perfringens was isolated in higher numbers (mean values of 10(7)-10(7.5) colony forming units (CFU) versus 10(4)-10(5) CFU per ml of intestinal content) and from more animals (79 versus 19%) in the study population than in the control population, although individual results from both populations could overlap. Other clostridial species, i.e. mainly urease-negative C. sordellii and C. bifermentans, were isolated in high numbers (>10(6) CFU per ml of intestinal content) from a few animals in the study population only. All but one of the 705 C. perfringens isolates from both populations belonged to the A toxin type and none of the urease-negative C. sordellii was toxigenic. Gram-negative anaerobes were not isolated in high numbers from any of the samples. Aerobically beta-haemolytic E. coli were significantly more frequent among the study population, but were isolated from only 25% of the animals. Salmonella Typhimurium was isolated from only two animals in the study population. Less than 1% of the E. coli isolated were verotoxigenic and one-third were necrotoxigenic. At this stage only non-enterotoxigenic type A C. perfringens are thus statistically associated with the enterotoxaemia syndrome in Belgian Blue calves and fulfil the first of the Koch's postulates.

Characterization of intestinal cnf1+ Escherichia coli from weaned pigs.

Escherichia coli isolated from 204 cases of porcine postweaning diarrhoea were tested by PCR for the genes of cytotoxic necrotic factors (CNF) and of cytolethal dystending toxin (CDT). Selected strains were also examined by PCR for the presence of papC-, sfa-, f17-, f18-, and afa-specific sequences encoding P, S, F17, F18 fimbriae and afimbrial adhesins. A 5.9% (12/204) of the strains had cnf1 gene, and two of them had cdt gene as well. Further six cdt+ strains were detected which were cnf-negative. Most of the cnf1+ strains belonged to serogroups O2, O6, O8, O54 characteristic of necrotoxic E. coli (NTEC) of humans. All the cnf1+ strains possessed the genes for P or S fimbriae or both, but were negative for F4, F17, or F18 or afimbrial adhesins. Results suggest that these enteric isolates may have entero- and/or uropathogenic significance in weaned pigs, and may have zoonotic potential for humans.

The afa-related gene cluster in necrotoxigenic and other Escherichia coli from animals belongs to the afa-8 variant.

Six hundred and nine necrotoxigenic Escherichia coli type 1 and 2 (NTEC1 and NTEC2) and non-NTEC isolated in Western and Southern Europe, North Africa and Canada from diseased calves, pigs, humans, poultry, and 55 isolated from asymptomatic calves were studied for the identification of afa-related sequences to the recently described afa-7 and afa-8 gene cluster variants from two bovine Escherichia coli (Lalioui et al., 1999). Colony hybridization and PCR assays for the afaD-7, afaE-7, afaD-8 and afaE-8 identified the afa-related sequences to the afa-8 gene cluster in most (67/79; 85%) of the E. coli positive with the Afa-f family probe and in 14 additional strains negative with the Afa-f probe. No E. coli was positive for the afa-7 gene cluster. The existence of afa-8 positive strains was thus confirmed among bovine E. coli and for the first time among porcine, poultry and human E. coli. Sequencing of the afaE-8 amplicon of nine strains from the different host species showed a high degree of conservation (>95% at the DNA level; >92% at the amino-acid level). The afa-8 gene cluster was more frequent in E. coli from diseased calves (18%) than from piglets (12%), humans (6%) and poultry (5%). Bovine NTEC2 (26%) were more frequently positive than NTEC 1 (20%) and non-NTEC (11%). E. coli isolated from asymptomatic calves were rarely positive: one NTEC2 (3%) and no non-NTEC. The afa-8 gene cluster was located on the Vir plasmid in 11/23 NTEC2, but no plasmid localization was detected in NTEC1 or non-NTEC.

Identification of the F17 fimbrial subunit- and adhesin-encoding (f17A and f17G) gene variants in necrotoxigenic Escherichia coli from cattle, pigs and humans.

Putative colonization factors of the F17 family of fimbrial adhesins have been identified in necrotoxigenic Escherichia coli Type 1 and Type 2 (NTEC1 and NTEC2) from calves, pigs, and humans. The f17A and f17G gene variants, coding respectively for the major subunit and for the adhesin of the F17 fimbriae, were typed in 70 E. coli carrying f17-related sequences (15 NTEC1, 51 NTEC2, and four non-NTEC) by colony hybridisation with gene probes derived from the different f17A gene variants (a, b, c, and d) and by PCRs specific for each f17A and f17G (I and II) gene variants. Typing of f17A genes was not possible by colony hybridisation, as most 70 E. coli were positive with more than one gene probe. On the other hand, the PCRs allowed the typing of the f17A gene in 37 E. coli and of the f17G gene in all 70 E. coli. The f17Ab gene variant was detected in 13 NTEC2; the f17Ac, in all 15 NTEC1, six NTEC2 and two non-NTEC; and the f17Ad, in one non-NTEC. Seven additional NTEC2 were positive with the PCRs for two variants: f17Ab and f17Ac in three of them; f17Ac and f17Ad in four of them. Either these seven NTEC2 harbour two variants or the variant present can be detected by two PCRs. The remaining 25 NTEC2 and one non-NTEC tested negative with the PCRs for the four f17A gene variants, suggesting the existence of other variant(s). In contrast, all 70 E. coli were positive with the PCR for the f17GII gene variant and none with the PCR for the f17GI gene variant. The f17-related sequences were present on the CNF2/Vir plasmids in 27 out of the 46 NTEC2 from which plasmid DNA could be extracted: all but one of those positive for the f17Ab gene variant and various proportions of those positive for other variants. In contrast, no plasmid carried f17-related sequences in NTEC1 and non-NTEC.