Virulence of Escherichia coli B2 Isolates from Meat and Animals in a Murine Model of Ascending Urinary Tract Infection (UTI): Evidence that UTI Is a Zoonosis

Abstract

In vivo evidence of a connection between urinary tract infections (UTI) and foods is lacking. The virulence of 13 Escherichia coli B2 isolates from healthy animals and fresh meat was investigated in a murine model of ascending UTI. All isolates produced positive bladder cultures (10² to 10⁷ CFU), and nine isolates produced positive kidney cultures (10² to 10⁵ CFU).

Urinary tract infections (UTI) are one of the most common human bacterial infections. They are responsible for severe morbidity and loss of productivity, and the overall costs associated with UTI amount to $1 billion in the United States alone (12). In approximately 80% of cases, UTI is caused by Escherichia coli (12). These E. coli isolates belong to the pathogenic group of extraintestinal E. coli (ExPEC), stemming mainly from the host's own fecal flora, and in the majority of cases belong to phylogroup B2 (14, 15). It has been suggested that one exterior reservoir of uropathogenic E. coli (UPEC) in the human intestine could be retail food or animals related to retail food (10). Many studies have investigated this hypothesis by using molecular techniques, including typing, virulence gene detection, and genome sequencing (8-10). Mostly, researchers have focused on comparing E. coli from retail food and UTI patients or avian-pathogenic E. coli (APEC) with ExPEC in vitro, and their molecular characterizations have indicated a connection between UTI and food (7, 8). Despite many in vitro studies, the zoonotic risk of animal and meat isolates remains unclear due to a lack of in vivo studies. In this study, we set out to investigate the virulence of E. coli B2 isolates from healthy pigs, broiler chickens, pork, and broiler chicken meat in a murine model of ascending UTI, which is representative of UTI in humans (2, 4, 11).

A total of 13 E. coli strains from healthy Danish broiler chickens (fecal, n = 4), Danish broiler chicken meat (n = 2), imported broiler chicken meat (n = 2), healthy Danish pigs (fecal, n = 2), Danish pork (n = 2), and imported pork (n = 1), previously published (6), were used for this study. The isolates belonged to phylogroup B2 and carried two or more of the virulence genes iutA, kpsM II, papA, papC, focG, sfaS, and hlyA, as described previously (5). All isolates used were found positive for type 1 fimbriae by use of a modification of a phenotypic test described earlier (3). The clinical UTI E. coli isolate C175-94 was used as a positive control and an E. coli O rough:H− isolate as a negative control in the murine experiments (4). The virulence of the strains was tested in a murine model of ascending UTI (Danish Ministry of Justice animal ethics committee approval no. 2004/561-835) described by Hvidberg et al. (4). In short, mouse bladders were emptied by gently pressing the abdomen, and 50 microliters (5 × 10⁶ CFU) of each bacterial suspension was slowly inoculated transurethrally in 4 to 6 outbred female albino CFW1 mice (26 to 30 g; Harlan Netherlands, Horst, Netherlands) by use of plastic catheters. The mice, which were housed 4 to 6 to a cage, were given free access to chow and 5% glucose-containing water. Seventy-two hours after inoculation, urine was collected from each mouse. The mice were then euthanized by cervical dislocation, and bladder and kidneys were removed and stored in Eppendorf tubes. The urine samples were processed the same day by spotting (20 μl) of a series of 10-fold dilutions (10⁰ to 10⁻⁶) in duplicate on bromthymol blue agar plates (SSI Diagnostika, Hillerød, Denmark). The bladder and kidneys were stored in 0.9% saline solution and kept at −80°C until processing. They were then incubated at room temperature for 1 h and subsequently homogenized using a TissueLyser (Qiagen, Ballerup, Denmark). Plates for bacterial counting were processed as described above. The detection limit was 25 CFU/sample.

Figure 1 shows the bacterial counts in urine, bladder, and kidneys of mice 72 h after inoculation with 13 B2 isolates from animals and meat and control strains. The negative-control strain failed to produce median bacterial counts above the detection limit (Fig. 1o). A positive culture for a given strain was defined as median bacterial counts above the limit of detection. The positive control and all 13 isolates produced positive bladder cultures, ranging from 10² to 10⁷ CFU per bladder (Fig. 1a to n); 11 of these isolates (Fig. 1a and d to m) also produced positive urine cultures (10³ to 10⁷ CFU per ml urine). Further, nine isolates produced positive kidney cultures (10² to 10⁵ CFU per 2 kidneys) (Fig. 1a, d, e, g, and i to m).

FIG. 1.

Bacterial counts in urine, bladder, and kidneys of mice killed 72 h after inoculation with the different E. coli B2 strains from healthy production animals or meat. Each point represents the results from one mouse. The number of urine samples was smaller than the number of bladder and kidney samples in one case due to unsuccessful urine sample collection. The solid horizontal line represents the median bacterial count. The dotted line indicates the detection limit (25 CFU/sample). Danish b.c. meat, Danish broiler chicken meat; Imp. b.c. meat, imported broiler chicken meat; Pos. control, positive control; Neg. control, negative control.

In recent years, the possible role of food-borne E. coli in causing UTI has been debated increasingly. Similarities among animal, meat, and UTI E. coli pheno- and genotypes have been illustrated (1, 7-10, 13, 16). However, the question of whether E. coli from animals and from fresh meat available at retail shop counters can cause UTI in humans and thus constitute a real zoonotic risk has remained unanswered. So far, emphasis has been placed on APEC. In support of the argument for similarity between APEC and ExPEC, researchers have shown that human ExPEC isolates (including UTI isolate CFT073) are virulent in chicken infection models (1, 9, 16) and that transfer of an APEC plasmid into a commensal avian E. coli strain significantly enhances the virulence of the commensal strain in murine kidneys but not in urine or bladder (13). Czirók et al. showed lethality of APEC after intraperitoneal inoculation in mice (1). However, the relevance of this murine model for human UTI is questionable. Recently, a study by Zhao et al. showed similar tendencies of expression of virulence and antibiotic resistance genes for APEC and a UPEC isolate in a murine model of UTI (16). However, in this model, the mice were reinoculated every 2 days to maintain infection levels, which brings the true virulence of the APEC isolate in the UTI model into question. No studies have included fecal isolates from healthy production animals or meat. Solid proof of animals and food being reservoirs of E. coli capable of causing human UTI has therefore been lacking. Here we show for the first time, to our knowledge, virulence of fresh-meat isolates and animal fecal isolates in a murine model of ascending UTI. Interestingly, all meat and animal isolates caused UTI, and most isolates also produced positive kidney cultures, testifying to the pathogenic nature of E. coli from otherwise healthy animals and meat. The murine model is representative of human UTI, and our results therefore provide important circumstantial evidence that UTI is a zoonosis (8-10). More studies are needed to confirm this, e.g., studies determining a direct link between E. coli-contaminated food products and a subsequent E. coli UTI in humans after intake of the meat products. Further, it is important to investigate more potential reservoirs of UPEC (e.g., other food sources and animals but also environmental samples).

Limitations of the study include the limited number of strains, phylogroups, and sources investigated. Strengths include use of the in vivo model of ascending UTI, which is representative of human UTI.

In conclusion, we demonstrated that all 13 tested E. coli B2 isolates from healthy animals and fresh meat caused positive bladder cultures and that 9 out of 13 caused renal infection in the murine model of ascending UTI. This is the first report of solid in vivo studies providing important circumstantial evidence that UTI is a zoonosis.