Cats as a Risk for Transmission of Antimicrobial Drug-resistant Salmonella

Cats can shed antimicrobial drug−resistant Salmonella serotypes in the environment.

S almonella infections are still a leading cause of human foodborne infections in the world (1,2). These infections primarily originate from eating contaminated food, especially chicken eggs and egg products, and also meat products from pigs and chickens (3,4). Considering the high frequency of food contamination and the emergence of multidrug−resistant Salmonella strains, control of Salmonella in food-producing animals has become a worldwide challenge. Other environmental sources can lead to accidental human infections with Salmonella as well. The role of pet animals as a source of Salmonella has not been fully investigated, but severe human infections originating from reptiles, especially pet turtles, have been reported (5).
Cats and dogs are the most widely kept pet animals, yet the incidence of Salmonella in these animals is largely unknown, and the risk that these animals pose for transmission of Salmonella to humans is unclear. In particular, cats that can freely roam outside, and are therefore able to scavenge or hunt food of unknown quality, are potential candidates for Salmonella carriage. Most reports concerning Salmonella and cats are case studies of clinical salmonellosis, which resulted in septicemia and death (6,7). Subclinical infections and carrier animals, however, are much more important with respect to transmission to humans. In this study, rectal swabs from cats of different origin (house cats, group-housed cats, diseased cats) were cultured for Salmonella. The serotype and phage type of the Salmonella isolates were determined, and the isolates were characterized with respect to their antimicrobial drug-resistance pattern and interaction with human intestinal epithelial cells.

Collection of Fecal Samples
A total of 278 rectal swab samples from house cats of different age, sex, and breed were taken between July and November 2003. All house cats came from different owners. The animals came from all over the Dutch-speaking part of Belgium, i.e., north of Brussels. Rectal swab specimens were also taken from 58 cats that were submitted for autopsy to the Faculty of Veterinary Medicine, Ghent University. The latter died or were euthanized because of incurable disease. All cats came from different owners, except three cats that had feline immunodeficiency virus (FIV), which came from one owner. Finally, rectal samples of 35 kittens (all <4 months of age) were taken at a facility where the animals were group-housed, waiting to be adopted. These animals came from 16 different owners.

Bacteriologic Analysis
Bacteriologic analysis was performed by enrichment of the rectal swabs. The samples were first pre-enriched in buffered peptone water (BPW) (Oxoid, Basingstoke, Hampshire, UK) overnight at 37°C, after which 1 mL of this suspension was added to 9 mL of tetrathionate brilliant green broth (Oxoid) (enrichment). After incubation overnight at 37°C, a drop of this suspension was spread on brilliant green agar (BGA) (Oxoid). Both the serotype and phage type of positive isolates were determined.

Polymerase Chain Reaction (PCR)
For PCR, a loop of bacterial culture was resuspended in 50 µL of water, and DNA was released from bacterial cells by boiling for 20 min. After the mixture was spun for 1 min in a microfuge at 14,000 x g, 2 µL of the supernatant was taken as a template DNA for PCR. PCR was carried out in 20-µL volumes by using PCR Master Mix from Qiagen (Hilden, Germany), according to the manufacturer's instructions. All the resistant strains were tested for the presence of the genes typical for particular resistance. The genes determined and primers used are listed in Table 1. Cycling consisted of 50-s incubations at 92°C, 55°C, and 72ºC, which were repeated 25 times. After PCR, amplification products were detected by electrophoresis in 2% agarose gel, stained with ethidium bromide, and visualized under UV light. Antimicrobial drug−sensitive strain S. Typhimurium F98 was used as a negative control in all the amplifications. S. Typhimurium strains 8420, 6237, 3520, 2200, and 5833 were used as positive controls.
All Salmonella strains were tested for the presence of the SopB gene. The primers were GATAGGAAA-GATTGAGCACCTCTG and TACAGAGCTTCTAT-CACTCAGCTTC, and the PCR cycle consisted of 30 cycles of (30 s 95°C, 1 min 58°C, 1 min 72°C).

Pulsed-Field Gel Electrophoresis (PFGE)
The bacteria were grown while being shaken overnight at 37°C in Luria-Bertani broth (LB). The XbaI PFGE patterns were determined for all 21 S. Typhimurium strains by using previously described PFGE methods (16,17) with some slight modifications. The patterns were grouped in a dendrogram with GelCompar II software (Applied Maths, St.-Martens-Latem, Belgium) by using the Dice coefficient and the unweighted pair group method with an arithmetic averages clustering algorithm.

Invasion of the Human intestinal Epithelial Cell Line T84
The capacity of all cat Salmonella isolates and the human S. Typhimurium isolates 8420, 6237, 3520, 2200, and 5833 to invade human intestinal epithelial cells was determined. Cells of the human colon carcinoma cell line T84 were seeded in 96-well cell culture plates (Greiner, Frickenhausen, Germany) at a density of 5.10 5 cells/mL culture medium (DMEM + 10% fetal calf serum + 2% Lglutamine, without antimicrobial drugs) and grown for 24 h. Bacteria were grown for 20 h in LB-medium, after which the suspension was diluted 1:50 in fresh LB-medium. After 4 h of incubation at 37°C, suspensions were centrifuged and resuspended in DMEM with 10% fetal calf serum (FCS). The number of colony-forming units (CFU)/mL was determined by plating 10-fold dilutions on BGA. The suspensions were stored overnight at 4°C. The next day, 10 6 CFU in 200 µL were added to the T84 cell cultures, which were then centrifuged for 10 min at 1,500 rpm to make close contact between the bacteria and the colon cells. The plates were incubated for 1 h at 37°C and 5% CO 2 . The cells were then rinsed three times with Hanks' Balanced Salt Solution (HBSS, Life Technologies, Paisley, Scotland). Cell culture medium with gentamicin (50 µg/mL) was added, and plates were incubated for 1 h at 37°C and 5% CO 2 . Hereafter, the cells were rinsed three times with PBS and analyzed with 1% Triton X-100 (Sigma, St. Louis, MO) in distilled water. From this lysate, 10-fold dilution series were made. From each dilution, 6 x 20 µL was added to BGA, to determine the number of CFU Salmonella per mL. The assays were performed in triplicate. The percentage of intracellular bacteria, relative to the number of Salmonella bacteria, initially incubated with the cells, was calculated. The previously mentioned human isolates of S. Typhimurium were used for comparison between the cat isolates and human isolates. Statistical analysis was performed by analysis of variance methods using the SPSS 11.0 software.

Characterization of Salmonella Isolates from Cats
Of 278 healthy house cats, 1 Salmonella strain was isolated, an S. Enteritidis phage type 21 strain, sensitive to all tested antimicrobial drugs. Five strains were isolated from cats that died from or were euthanized because of incurable disease. Feline AIDS (caused by feline immunodeficiency virus [FIV]) was diagnosed in three cats, one died due to feline panleukopenia parvovirus infection, and one was poisoned. Three isolates were identified as being ampicillin-resistant S. Typhimurium phage type 193, harboring the bla TEM gene. They had the same pulsed-field gel electrophoresis (XbaI) pattern, indicating that the isolates were of clonal origin (Figure 1). The three cats came from the same owner. One isolate was an antimicrobial drug-sensitive Salmonella Bovismorbificans strain. One isolate was Salmonella 4:i:-, which was resistant to ampicillin, chloramphenicol, sulfonamides, tetracycline and sulfamethoxazole-trimethoprim (ACSuTSxt), harboring the bla TEM , cat, sul2, tet(A), and dfrA1 antimicrobial drug-resistance genes. Eighteen strains were isolated from the group-housed cats. All of these were S. Typhimurium phage type 120/ad. Fourteen of these strains showed acquired resistance to ampicillin, chloramphenicol and tetracycline and harbored the bla TEM , cat, and tet(A) antimicrobial drug−resistance genes, while four isolates were resistant to chloramphenicol only and only harbored the cat gene (Table 2). Pulsed-field gel electrophoresis showed that the isolates from the group-housed cats were of the same XbaI PFGE type, and that three subtypes within this type were present, indicating a clonal origin ( Figure  1). One subtype contained the 14 strains that were resistant to the three mentioned antimicrobial drugs. All Salmonella strains harbored the SopB gene.

Invasion of the Human Intestinal Epithelial Cell Line T84
All isolates invaded T84 cells, with the cat isolates of S. Typhimurium PT193 (strains 1147, 1145, and 55, which belong to the same clone) and the human isolate S. Typhimurium strain 2200, the most invasive, yielding a percentage of invasion of 8% to 10%. The multidrug− resistant cat isolate Salmonella 4:i:-(strain 11) was the least invasive strain, having an invasion percentage of about 0.5%. Invasion percentages of the different isolates are shown in Figure 2. Of the strains of the same PFGE type, only one was shown in Figure 2, since no significant differences were detected between the invasion percentages of these strains. Statistically significant differences are shown in the figure.

Discussion
This study concluded that, although cats can transmit Salmonella strains, healthy house cats are generally safe. Earlier reports regarding isolation of pathogens from healthy cats showed low percentages (mostly around 1%) of Salmonella-positive rectal swabs (18,19). In our study, 1 of 278 healthy cats was found to be positive.
Immunodeficiency and nonhygienic housing can be predisposing factors for cats to shed Salmonella in the feces, resulting in contamination of the environment. Rectal swabs from 18 of 35 group-housed kittens were Salmonella-positive in our study. The fact that the 35 kittens were derived from more than 10 different owners before being group-housed and that one PFGE type (three subtypes) of S. Typhimurium 120/ad was isolated, indicates spread of the Salmonella strain between the cats or a common source. The age of these animals may also play a role, since all animals in this group were <4 months. Young animals are more susceptible to Salmonella infection. Also immunodeficiency can result in Salmonella excretion. One outbreak of fatal salmonellosis in cats has been reported after mild immunosuppression induced by live panleukopenia virus vaccination (7). In our study, animals infected with FIV and one animal that had panleukopenia shed Salmonella. Three animals that were infected with FIV were derived from the same owner, which indicates that the animals were infected with Salmonella from the same source or that one animal contaminated the others.
In our study, serotypes Typhimurium, Enteritidis, Bovismorbificans, and 4:i:-were isolated from cats. The isolated serotypes indicate that the cats were infected from the same sources compared with other animals and man. Indeed, serotypes Typhimurium and Enteritidis are the most widespread serotypes and the serotype Bovismorbificans is not uncommon in other animals, including humans (2,20).
Generally, invasion in the human intestinal epithelial cell line T84 was comparable between the cat isolates and isolates from humans. Invasion in intestinal epithelial cells is the primary step in the pathogenesis of Salmonella that causes gastrointestinal problems (21). This finding implies that the cat isolates are potentially pathogenic for humans. Moreover, all cat isolates harbored the SopB gene, which is involved in blocking the closure of chloride channels in gut epithelium and thus in inducing diarrhea. As in most other animal species, the cat isolates of the serotype Typhimurium harbored antimicrobial drug-resistant genes, raising concerns about spreading antimicrobial drugresistant strains to humans.
Since the 1990s, concerns have arisen about the emergence and spread of multidrug−resistant Typhimurium strains, especially the multidrug−resistant ACSSuT type, which is resistant to ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline (2). In our study, some S. Typhimurium isolates from cats were resistant to a single drug such as ampicillin or chloramphenicol, while most isolates from the group-housed cats (same clone) were resistant to ampicillin, chloramphenicol, and tetracycline. Resistance genes were found to be bla TEM (ampicillin), cat (chloramphenicol), and tet(A) (tetracycline). The genes in the class 1 integron of the multidrug−resistant genomic island in ACSSuT type S. Typhimurium, required for the resistances to the above three mentioned antimicrobial drugs, are bla PSE1 , floR, and tet(G) (22). This illustrates that these isolates did not acquire their resistance genes from horizontal transfer from pentadrug−resistant ACSSuT type strains. The isolate Salmonella 4:i:-was resistant to ampicillin, chloramphenicol, sulfonamides, tetracycline, and sulfamethoxazole/trimethoprim (ACSuTSxt-type), encoded by bla TEM (ampicillin), cat (chloramphenicol), sul2 (sulfonamides), tet(A) (tetracycline), and dfrA1 (trimethoprim). Also the resistance shown by this example had no relationship to the typical S. Typhimurium DT104 multidrug− resistant genomic island.
In conclusion, healthy house cats are generally safe with regard to excretion of Salmonella in the environment. Cats that are sick or are receiving medication resulting in immune deficiencies can potentially pose a threat to public health. Young children, the elderly, and immunocompromised persons are at risk because of their high sensitivity for the infection. All persons should follow good hygiene practices when keeping cats as pets. . The y-axis shows the percentages of intracellular bacteria 2 hours postinfection, relative to the initial number of bacteria, incubated with the cells. The x-axis shows isolate numbers. All isolates derived from the group-housed cats had the same invasion percentage as strains 198 and 355 (data not shown). Isolates 55 and 1145 had the same invasion percentage as strain 1147 (data not shown). Data not sharing superscript numbers indicate statistically significant differences (p < 0.05).