Identification of virulence attributes of gastrointestinal Escherichia coli isolates of veterinary significance

The pathogenic strains of Escherichia coli recovered from the intestinal tract of animals fall into categories called enterotoxigenic, enteropathogenic, enterohemorrhagic and necrotoxigenic. The other two categories, enteroinvasive and enteroaggregative, have not been reported in animals. The pathogenicity of these strains is determined by the presence of certain genes that encode adhesins and toxins, are generally organized in large blocks in chromosomes, large plasmids or phages, and are often transmitted horizontally between strains. In this review, we summarize current knowledge of the virulence attributes that determine the pathogenic potential of E. coli strains and the methods available to assess the virulence of the strains. We also discuss the clinical symptoms, the gross and histological lesions, and the molecular diagnostic methods our laboratories have implemented for detecting pathogenic strains of E. coli that are isolated from the gastrointestinal tract of animals.     

Screening petting zoo animals for the presence of potentially pathogenic Escherichia coli.

Several outbreaks of Escherichia coli O157 have been reported in petting zoos, resulting in hospitalization of many children. At present, no standard procedure has been adopted to monitor the presence of enterohemorrhagic E. coli (EHEC) or Shiga-toxin-producing E. coli (STEC) in petting zoo animals. Direct detection of these strains from rectal swabs of animals in petting zoos was developed and obviated the need to culture the organisms. DNA extracted from bacteria in the swabs was tested for the presence of wecA gene specific for E. coli by polymerase chain reaction (PCR). The wecA positive samples were further tested for Shiga-toxin genes stxl and stx2, and the intimin eae by multiplex PCR and for the presence of O157 and H7. Swabs (n=104) from 15 animal species in a petting zoo were tested; 7 goats and 3 cows were found to carry STEC. The method is rapid and convenient for monitoring potentially pathogenic E. coli in petting zoo animals.     

Hemolytic-uremic syndrome in a postpartum mare concurrent with encephalopathy in the neonatal foal.

A postpartum mare and foal were presented for evaluation of fever and lethargy in the mare. The mare was diagnosed with endometritis and initially responded well to treatment. On the second day of hospitalization, the mare developed renal insufficiency characterized by oliguria, azotemia, hemolysis, and thrombocytopenia. Concurrently, the foal developed rapidly progressive central nervous system signs culminating in refractory seizures. Both animals failed to respond to treatment and were euthanized. Thrombotic microangiopathy involving glomeruli was evident on microscopic examination of the mare's kidneys. Microscopic evidence of brain edema was the principal postmortem finding in the foal. No specific etiology was confirmed in either case. Notably, Escherichia coli 0103:H2 was isolated from the mare's uterus and the gastrointestinal tracts of both animals. To the authors' knowledge, this is the first report in which an organism implicated as a cause of hemolytic-uremic syndrome was isolated from an animal with clinical signs and postmortem findings consistent with the disease.     

Detection of O antigens in Escherichia coli

Lipopolysaccharide on the surface of Escherichia coli constitutes the O antigens which are important virulence factors that are targets of both the innate and adaptive immune systems and play a major role in host-pathogen interactions. O antigens are responsible for antigenic specificity of the strain and determine the O serogroup. The designation of O serogroups is important for classifying E. coli strains, for epidemiological studies, in tracing the source of outbreaks of gastrointestinal or other illness, and for linking the source to the infection. For conventional serogroup identification, serotyping by agglutination reactions against antisera developed for each of the O serogroups has been used. In the last decade, many O-antigen gene clusters that encode for the enzymes responsible for the synthesis of the variable oligosaccharide region on the surface of the bacteria have been sequenced and characterized. Unique gene sequences within the O-antigen gene clusters have been targeted for identification and detection of many O groups using the polymerase chain reaction and microarrays. This review summarizes current knowledge on the DNA sequences of the O-antigen gene clusters, genetic-based methods for O-group determination and detection of pathogenic E. coli based on O-antigen and virulence gene detection, and provides perspectives on future developments in the field.     

Occurrence of pathogenicity island I(APEC-O1) genes among Escherichia coli implicated in avian colibacillosis

Colibacillosis caused by avian pathogenic Escherichia coli (APEC) is a leading cause of economic loss to the poultry industry worldwide. The ability of APEC to cause disease is determined by certain virulence markers, some of which are located on pathogenicity islands (PAls). We recently described one such PAI in an APEC O1:K1 strain (APEC-O1). This PAI, termed PAI I(APEC-O1), carries the genes of the pap operon, a region similar to the tia invasion determinant of enterotoxigenic E coli; ireA, a gene that encodes an iron-responsive element; and a novel 1.5-kb region, ORF 54. Here, the occurrence of six selected loci of PAI I(APEC-O1) (papA, papC, papG, ireA, tia, and ORF 54) among APEC and fecal E. coli strains from apparently healthy chickens (avian commensal E. coli) was determined using polymerase chain reaction (PCR) techniques. None of the commensal E. coli was positive for all six traits, whereas 7.2% of the APEC isolates were positive for all the traits. Although there was no significant difference in the occurrence of ORF 54 among APEC and commensal E. coli, tia, ireA, papC, and papG genes were predominantly present in APEC rather than in avian commensal E. coli. papA was detected in only 6.3% of APEC, perhaps because of the presence of allelic variants of the gene. Additionally, the presence of all six traits was tested with PCR in APEC isolates collected in the 1980s, and these results were compared with those obtained with the APEC isolated in the 1990s. There was no significant difference in the occurrence of tia, ireA, papC, papG, and ORF 54 between APEC isolates collected during the different decades. However, papA was more frequently present in APEC from the 1980s than it was in APEC from the 1990s. Phylogenetic group of an isolate did not correlate with pathogenicity or the presence of PAI traits, except that more APEC of the low-pathogenicity group belonged to the phylogenetic group B1. However, PAI traits occurred more frequently in isolates belonging to the intermediate- and high-pathogenicity groups than in isolates of low pathogenicity.     

Bronchopneumonia associated with extraintestinal pathogenic Escherichia coli in a horse

Extraintestinal pathogenic Escherichia coli (ExPEC) strains carrying distinct virulence attributes are known to cause diseases in humans and animals and infect organs other than the gastrointestinal tract. A fatal case of bronchopneumonia in a 12-year-old female Quarterhorse was investigated. Following postmortem examination, E. coli, Enterococcus sp., and Klebsiella pneumonia were isolated from the lungs, which contained multifocal intra-alveolar accumulations of neutrophils and macrophages with edema, hemorrhage, and fibrin. The strain of E. coli belonged to O2H21 and carried virulence genes cnf1, sfa, foc, fimA, and papG allele I that are known to be associated with ExPEC strains. The strain was resistant to several antimicrobials including clindamycin, erythromycin, oxacillin, penicillin, and rifampin. This is the first report, to the authors' knowledge, in which ExPEC O2H21 has been associated with fatal bronchopneumonia in a horse.     

Isolation of necrotoxigenic Escherichia coli from a dog with hemorrhagic pneumonia

A 7-month-old sexually intact male Cocker Spaniel was admitted to the North Carolina State University Veterinary Teaching Hospital for evaluation of lethargy, panting, and excessive salivation that had become progressively severe during a 5-hour period. Despite intensive medical care, the dog died within the first 24 hours of hospitalization, and death was attributed to acute, severe, necrotizing pneumonia. Lung tissue collected at necropsy by use of swabs was cultured and yielded an isolate of Escherichia coli; because of the rapid progression of illness in an otherwise healthy dog, the isolate underwent virulence typing and was determined to be a necrotoxigenic E. coli. Necrotoxigenic E. coli produce a toxin called cytotoxic necrotizing factor and are known to be involved in extraintestinal infections, including urinary tract infection, in humans and animals. Virulence typing of E. coli isolates from dogs with peracute pneumonia is recommended to further characterize the epidemiologic characteristics and public health importance of necrotoxigenic E. coli.     

Pathogenic and fecal Escherichia coil strains from turkeys in a commercial operation

The biochemical phenotypes and antimicrobial susceptibility patterns of 105 clinical Escherichia coli isolates from flocks with colibacillosis in a turkey operation were compared with 1104 fecal E. coli isolates from 20 flocks in that operation. Clinical isolates and 194 fecal isolates with biochemical phenotypes or minimum inhibitory concentrations for gentamicin and sulfamethoxazole similar to clinical isolates were tested for somatic antigens and the potential virulence genes hylE, iss, tsh, and K1. The predominant biochemical phenotype of clinical isolates contained 21 isolates including 14 isolates belonging to serogroup 078 with barely detectable beta-D-glucuronidase activity. Thirty-five fecal isolates had biochemical phenotypes matching common phenotypes of clinical isolates. Sixty-six (63%) clinical isolates exhibited intermediate susceptibility or resistance to gentamicin and sulfamethoxazole compared with 265 (24%) fecal isolates (P < 0.001). Seventy-seven clinical isolates reacted with O-antisera, of which 51 (66%) belonged to the following serogroups: O1, O2, O8, O25, O78, O114, and O119. In comparison, 8 of 35 (23%) fecal isolates subtyped on the basis of biochemical phenotype belonged to these serogroups and four of 167 (2%) fecal isolates subtyped on the basis of their antimicrobial resistance patterns belonged to these serogroups. Iss, K1, and tsh genes were detected more often among clinical isolates than these fecal isolates (P < 0.05). In summary, a small subgroup of E. coli strains caused most colibacillosis infections in this operation. These strains existed at low concentration in normal fecal flora of healthy turkeys in intensively raised flocks. The data suggest that colibacillosis in turkey operations may be due to endogenous infections caused by specialized pathogens.     

Extraintestinal pathogenic Escherichia coli-induced acute necrotizing pneumonia in cats

Extraintestinal pathogenic Escherichia coli (ExPEC) are pathogens involved in several disease conditions, ranging from urinary tract infection to meningitis in humans and animals. They comprise epidemiologically and phylogenetically distinct strains, affecting most species and involving any organ or anatomical site. Here, we report fatal cases of necrotizing pneumonia in cats. Over a 1-week period, 13 cats from an animal shelter in Stamford, Connecticut were presented for necropsy. All had a clinical history of acute respiratory disease. The gross and microscopic findings for all the cats were consistent. Escherichia coli was uniformly isolated from the lungs of all the tested cats. All the isolates were haemolytic, genetically related as determined by enterobacterial repetitive intergenic consensus PCR, and harboured genes encoding for cytotoxic necrotizing factor-1 and fimbriae and adhesions that are characteristic of ExPEC, implying a point source clonal outbreak. As cats are common household pets, this report raises concerns regarding zoonotic potential (in either direction) for these ExPEC strains.