First isolation in Argentina of a highly virulent Shiga toxin-producing Escherichia coli O 145 : NM from a domestic cat

Introduction: Hemolytic uremic syndrome (HUS) is distributed worldwide. In Argentina, more than 450 cases of HUS, mostly sporadic, are reported annually. The main serotype isolated is O157:H7, and among non-O157 STEC, O145:NM is the most frequent strain. We studied the relationship of companion animals living in contact with a child with sporadic HUS, as carriers of Shiga toxin-producing Escherichia coli (STEC) strains. Methodology: Duplicate rectal swab samples were taken weekly from the household cat and dog at the home of a patient with HUS. Samples were plated on MacConkey and sorbitol MacConkey-CT agar. Confluent growth from each plate was screened for the presence of stx1, stx2 and rfbO157 gene by PCR assays. Up to 300 individual colonies taken from positive plates at screening were retested by PCR. Results: The strain from the cat belonged to the highly virulent serotype O145:NM. Although this strain differed antigenically from the strain isolated from a child with HUS living in the same house, both carried the stx2, eae and ehxA virulence genes. The strain isolated from the dog belonged to the serotype O178:H19. Conclusions: An asymptomatic household cat may harbour the high virulent STEC strain, such as O145:NM, the second most frequently STEC serotype associated with HUS in Argentina. Companion animals are probably exposed to the same sources as the humans. More studies are needed to establish dogs and cats as sources of infection in the epidemiological cycle of infections caused by STEC strains, and to develop effective control strategies for this pathogen.


Introduction
Hemolytic uremic syndrome (HUS) is widely distributed around the world.In Argentina, more than 450 cases of HUS are reported annually, and in the last five years the notification rate has increased by 1.7% [1].HUS is the main cause of acute renal failure and is responsible for 20% of the renal transplants among children in this country [2].Humans frequently become infected with Shiga toxin-producing Escherichia coli (STEC) by the ingestion of contaminated food or water, person-toperson transmission, or direct contact with animals that represent a threat for STEC infection (e.g., farm animals, ruminants) [3,4,5].Food has been identified worldwide as the source of infection in approximately half of the cases of HUS [6].Despite that ruminants in general and cattle in particular have been implicated as the main STEC reservoirs [7], other animals can also maintain the strains in the environment [8,9].A carrier rate of E. coli O157:H7 in companion animals was recently described by Kataoka et al. in Japan [10].The aim of this study was to evaluate the relationship of household pets, a cat and a dog, living in contact with a child with HUS, as carriers of Shiga toxin-producing Escherichia coli (STEC) strains and their potential role in the epidemiological cycle of STEC infection.

Ethical considerations
The pets were handled under the supervision of licensed veterinarians according to standards set by the Institutional Ethics Committee, Faculty of Veterinary Science, University of Buenos Aires, Argentina.

Patient
The patient was an 11-year-old boy with diagnosis of HUS, living with his parents and a 14year-old brother in an apartment of Buenos Aires City, Argentina.Ingestion of raw/undercooked meat, raw milk, and contact with farm animals were excluded.Virulence characteristics of the O157:H7 strain (stx 2 , eae, ehxA) isolated from the patient, by methods described elsewhere [11], were reported by the service Fisipopatogenia, Instituto Nacional de Enfermedades Infecciosas, ANLIS "Dr C.G. Malbrán "Buenos Aires, Argentina (2007).One dog and one cat were the companion animals of the family.

Companion animals
Both animals lived in close contact with the owners, usually climbing on the tables and beds.They were fed only with commercial food, and contact with other animals was excluded.

Sample collection
Duplicate rectal swab samples were taken weekly from the household pets at the home of a patient with HUS.The first samples were collected during the same week of HUS diagnosis (on day 4 after the HUS confirmation).Then samples were weekly collected (days 11, 18, 25, 32 and 39).Sampling was continued until two successive negative results were found for both animals.

Enrichment and plating procedures
Samples, collected in Stuart's medium, were processed the day of collection by inoculating culture tubes with tryptic soy broth (TSB) and tryptic soy broth plus tellurite and cefixime (TSB-CT).After 6 hours of incubation at 37ºC and 42ºC respectively in aerobiosis, cultures from TSB were streaked onto MacConkey agar (MAC), and those from TSB-CT onto sorbitol MacConkey-CT agar (SMAC-CT) plates and incubated in aerobiosis at 37ºC for 18 hours.

Results and discussion
The first sample taken from the cat during the week of detection of the HUS case (day 4) was positive for the stx 2 gene, but the sample collected from the dog was negative (Table 2).On days 18 and 25, samples from the dog were positive for the stx 2 gene.On days 11, 32 and 39, samples from both animals were negative in the screening and therefore samplings were stopped.Both animals were asymptomatic during the time of sampling.The STEC strains isolated from the cat on day 4 belonged to the serotype O145:NM.The genetic profile of the isolates corresponded to stx 2 + eae + ehxA + .This strain belonged to the genotype stx 2 as determined by RFLP-PCR, and exhibited the enterohemolytic phenotype.
We were able to isolate the STEC 0178:H19 strain from the dog on day 18.The strain carried the stx 2(vh-a) gene; however, it was neither enterohemolytic nor carrying eae and saa genes.Despite the positive result on screening, no STEC isolates were recovered from the sample collected from the dog on day 25.All STEC isolates were susceptible to all antibiotics tested.
In Argentina, approximately 60% of HUS cases are due to O157:H7, and among the non-O157 STEC serotypes, O145:NM is the most frequent strain [24].Although a high virulent O145:NM STEC serotype had been isolated from the household cat in the week of the detection of the HUS case, an O157:H7 strain was isolated from the child presenting with clinical signs of HUS.
Since screening for STEC in human samples was stopped when an STEC strain was recovered, due to the limitation of the methodology [25], one cannot exclude the possibility of isolating more than one STEC serotype from the patient with HUS and from the cat as well.In this patient case, we did not screen specifically for the O145:NM serotype.Although immunomagnetic separation-based detection of O157 is routinely used in human samples, IMS-based detection of serotypes other than O157 were not available in 2007.Moreover, IMS to O157 strains was not performed in the pets' samples.

Conclusion
A close relationship between children and their pets was seen in the city of Buenos Aires, Argentina [26].Although it is not possible to determine the source of infection in this home, and any of the strains from the companion animals were directly associated with the sporadic case of HUS, the isolation of STEC strains in both animals from the same home is striking, especially given the local prevalence previously determined for these species which was as low as 1.1% (5/450) in dogs, and 2.6% (4/149) in cats [13].Virulent STEC strains may be harboured by asymptomatic household dogs and cats, which are probably exposed to the same sources as the humans.More studies are needed to establish companion animals as a source of infection or accidental carriers of STEC in the epidemiological cycle of infections caused by STEC strains, and to develop effective control strategies for this pathogen.The isolation of O145:NM from a cat has already been documented in Germany [9].This strain was also isolated from cattle in Argentina [27], but this is the first time that this serotype with an enterohemorrhagic E. coli virulence repertoire has been recovered from a domestic cat in our country.The potential role of dogs and cats in the epidemiological cycle of STEC human infection is still poorly investigated.Due to the close cohabitation between companion animals and their owners, the transmission of sporadic cases of HUS is probable.

Table 1 .
Primers, DNA sequences, target genes, conditions, product sizes and references of PCR used in the study

Table 2 .
Sampling time, screening and profile of STEC strains isolated from household pets related with a sporadic case of HUS