Antimicrobial resistance of Escherichia coli in cats and their drinking water: drug resistance profiles and antimicrobial-resistant genes

Abstract

Background

Antimicrobial resistance (AMR) is a global health concern that is exacerbated by the transmission of bacteria and genetic material between humans, animals and the environment. This study investigated AMR of Escherichia coli (E. coli) isolated from cats’ feces and their drinking water. The study compared the AMR of fecal and environmental E. coli isolates from pet cats.

Results

A total of 104 samples (52 cat feces and 52 cat drinking water samples) was cultured for E. coli. The study compared the AMR of fecal and environmental E. coli isolates from pet cats. An analysis of carbapenemase and extended-spectrum β-lactamase (ESBL)-producing E. coli genes (bla_TEM, bla_SHV and bla_CTX-M) and phylogroups of E. coli was also performed. E. coli was identified from all fecal (100%) and almost half of drinking water (44.2%) samples. All E. coli isolate was susceptible to amikacin or imipenem. Clindamycin showed the highest resistance rate. β-lactam was the most found with co-resistance profiles, comprising β-lactams with aminoglycosides, quinolones, sulfonamides, macrolides or carbapenems. Very strong positive correlations of bactericidal agents were found among quinolones (r > 0.8, p < 0.01). Within the group of bacteriostatic agents, moderate correlation was observed between azithromycin and sulfa-trimethoprim (r = 0.5253, p < 0.01). Carbapenemase gene was not detected in this study. Extended-spectrum β-lactamase-producing E. coli genes (bla_TEM, bla_SHV and bla_CTX-M) were identified in E. coli isolates, with bla_TEM being the most predominant. Furthermore, phylogroup B2 was the dominant segregation among the E. coli, particularly in fecal isolates.

Conclusions

This study identified AMRin E. coli isolated from cats’ feces and their drinking water.  The results revealed that the phylogroup B2 was predominant, with bla_TEM being the most widespread ESBL gene.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-024-04435-8.

Background

Escherichia coli (E. coli) can be classified into eight phylogroups (A, B1, B2, C, D, E, F and clade I) based on the pathogenicity islands present in its genome [1]. Phylogroups A, B1, B2 and D are found most often [1, 2]. Non-pathogenic E. coli strains are predominantly found in groups A or B1, while other groups are responsible for pathogenic strains [2]. Extended-spectrum β-lactamase (ESBL)-producing E. coli strains are commonly reported in healthy and diseased animals [3–8]. The ESBL genes have been found in both commensal and clinical E. coli isolates obtained from cats and their owners, indicating the bidirectional spread of plasmids containing gene-encoding ESBL-producing E. coli through the environment, humans and cats [8–11]. The bla_TEM, bla_SHV and bla_CTX−M genes are the most frequently detected genes in ESBL-producing E. coli [12]. They can hydrolyze and break down the lactam antimicrobial, leading to antimicrobial resistance (AMR) [12].

AMR is a worldwide health issue in humans and animals [13–16]. The extensive use of antimicrobial drugs, as well as low public awareness and knowledge, has contributed to the spread of AMR bacteria [13–16]. A high frequency of AMR has been found in E. coli isolates from both companion animals and humans [8, 17, 18]. The exchange of bacteria and genetic material among humans, animals and the environment significantly contribute to the spread of AMR [14, 19]. Close contact between cats and their owners creates favorable conditions for the transmission of zoonotic pathogens, either directly through physical contact or indirectly through a contaminated environment [14, 19]. Cat drinking water often is contaminated with coliform, especially in households with longhaired cats [20]. However, there is limited knowledge about the AMR of E. coli strains in cat feces and their drinking water. Monitoring AMR in these two sources allows for the assessment of differences in the prevalence of AMR between cats and their environment and for the identification of potential sources of resistance transmission. Understanding patterns of AMR in E. coli could provide insights into beneficial strategies to promote the appropriate and responsible use of antimicrobials among veterinarians, public health officials and researchers.

To understand the role of AMR E. coli strains in cats and their surroundings, the objectives of this study were to determine the prevalence and association of the AMR of E. coli between cat feces and cat drinking water, analyze AMR patterns and evaluate the resistance of E. coli strains to bactericidal or bacteriostatic agents. The E. coli phylogroups and the presence of carbapenemase and ESBL-producing E. coli genes were also investigated.

Methods

Animals and sample collection

A total of 52 pet cats were enrolled in this study. The cats were randomly selected during routine visits to the Kasetsart University Veterinary Teaching Hospital, Bangkok, Thailand. The cats’ median age (range) was 4.0 (0.7–15.9) years. There were 17 females and 35 males. The cats’ breeds were Domestic Shorthair (n = 38), Persian (n = 8), British Shorthair (n = 3) and Scottish Fold (n = 3). Ethical approval for the study was obtained from the Kasetsart University Institutional Animal Care and Use Committee (approval number #ACKU62-VET-030) and by the Ethical Review Board of the Office of the National Research Council of Thailand.

For each cat, a fresh fecal sample was obtained per rectal swab by a veterinarian at the Kasetsart University Veterinary Teaching Hospital. The swab was immediately stored in a sterile tube and submitted to the microbiology laboratory at the Kasetsart University Veterinary Teaching Hospital within 15 min. In addition, a water sample of approximately 50 ml was collected from each cat’s drinking water bowl at home by the cat’s owner using a sterile 50 ml syringe (Nipro disposable syringe, Nipro Corporation Limited, Sri Ayutthaya, Thailand) and capped with a sterile syringe cap (3M 5-200 Curos disinfecting cap, 3M Health Care, MN, USA). The water sample was immediately placed on ice and transported to the microbiology laboratory for further analysis.

Bacterial isolation and identification

Fecal samples were processed to isolate E. coli strains using a conventional bacterial culture. Each fecal swab sample was streaked onto MacConkey agar and incubated at 37 °C for 18–24 h. Lactose fermentation-positive colonies were selected and subsequently confirmed through triple sugar iron agar, indole, methyl red, Voges-Proskauer and citrate (IMViC) tests. Among the isolates that displayed lactose fermentation with gas production within 48 h, those testing negative for hydrogen sulfide and exhibiting IMViC patterns of positive-positive-negative-negative were identified as presumptive E. coli. Presumptive E. coli colonies underwent further biochemical tests, including indole and citrate utilization tests, to confirm their identity as E. coli.

For the water samples, one ml was spread on MacConkey agar and incubated at 37 °C for 24 h. Colonies resembling E. coli were further confirmed using the same biochemical tests as described for fecal E. coli.

Antimicrobial susceptibility test

Antimicrobial susceptibility testing was conducted using the disc diffusion method. Fourteen antimicrobial agents were used: amikacin (30 µg), gentamycin (10 µg), imipenem (10 µg), meropenem (10 µg), ceftriaxone (30 µg), cephalexin (30 µg), amoxicillin (10 µg), amoxicillin + clavulanic acid (20 µg), ciprofloxacin (5 µg), enrofloxacin (5 µg), norfloxacin (10 µg), clindamycin (2 µg), azithromycin (15 µg) and sulfamethoxazole + trimethoprim (25 µg) (Mastdiscs^® AST, Mast Group Ltd., Merseyside, England). The AMR breakpoints were interpreted following the Clinical and Laboratory Standards Institute (CLSI) Performance Standards for Antimicrobial Susceptibility Testing (21). E. coli ATCC 25922 was used as the quality control strain.

Pure cultures of E. coli strains isolated from fecal and water samples were prepared and standardized to match the turbidity of a 0.5 McFarland standard. Each standardized suspension was then spread onto Mueller-Hinton agar plates using a sterile swab. Antimicrobial discs were placed on the agar surface, and the plates were incubated at 37 °C for 16–20 h. The diameter of the inhibition zones was measured and interpreted according to the CLSI guidelines [21] to determine the susceptibility or resistance of each E. coli strain to the tested antimicrobial agents.

Detection of β-lactamase genes and Clermont phylotyping

The bacterial DNA from E. coli isolate was extracted using a commercial kit (GF-1 Bacterial DNA extraction kit, Vivantis Technologies Sdn Bhd, Selangor Darul Ehsan, Malaysia). The presence of carbapenemase genes (bla_IMP, bla_KPC, bla_NDM and bla_{OXA−48−like}) and ESBL genes (bla_TEM, bla_SHV and bla_CTX−M) were identified using multiplex polymerase chain (PCR) reactions, as described [22, 23]. The PCR amplification conditions were set as follows: initial activation at 95 °C for 3 min; 30 cycles of 95 °C for 5 s, 56 °C for 30 s, 72 °C for 45 s; and a final extension at 72 °C for 5 min.

The Clermont phylotyping method was applied to the E. coli isolates using multiplex PCR assays [24, 25]. The PCR amplification conditions were set as follows: initial activation at 94 °C for 4 min; 30 cycles of 94 °C for 5 s and 57 °C for 20 s (group E), 59 °C for 20 s (group A, B, C, D and F), or 63 °C for 20 s (group B2); and a final extension at 72 °C for 5 min. The PCR products were analyzed by 30-minute gel electrophoresis on 2% agarose gels in 0.5X TBE buffer [26]. The gel, stained with ethidium bromide, was evaluated under ultraviolet light (SynGene, Cambridge, UK). The PCR product sizes were compared to a molecular-sized standard (GeneRuler 100 bp Plus DNA ladder; Thermo Fisher Scientific, Waltham, Ma, USA). All primer sequences used in this study were shown in supplemental Table 1.

Statistical analysis

Data obtained from the antimicrobial susceptibility tests were analyzed using commercially available statistical software packages (GraphPad Prism version 6.0, GraphPad Software, Inc., La Jolla, California, USA; STATA version 12, StataCorp, College Station, Texas, USA). The prevalence of AMR in fecal and environmental E. coli strains was calculated as a percentage. The association of AMR between feces and drinking water was determined using Fisher’s exact test. Additionally, the correlation of sensitivity patterns among bactericidal or bacteriostatic agents against E. coli strains was assessed using a pairwise correlation coefficient. The significance level was set at p < 0.05. Negligible, weak, moderate, strong and very strong relationships corresponded to a pairwise correlation coefficient of 0.00–0.09, 0.10–0.39, 0.40–0.69, 0.70–0.89 and 0.90–1.00, respectively [27].

Results

E. coli was isolated from all 52 (100%) cat feces samples and 23 (44.2%) cat drinking water samples. The antimicrobial sensitivity analysis of fecal E. coli strains isolated from pet cats revealed diverse resistance patterns to different antimicrobial classes. The prevalence of AMR is shown in Table 1. The isolates showed no resistance to Amikacin, while minor resistance to gentamycin was shown in both fecal (7.7%) and water (4.4%) samples. A very low number of isolates were resistant to carbapenem (imipenem and meropenem). Only a single meropenem-resistant E. coli isolate was found in cat feces. Resistance to penicillin, such as amoxicillin and amoxicillin with clavulanic acid, was exhibited in both types of samples. In fecal samples, there were 19 (38.5%) amoxicillin-resistant and 3 (5.8%) amoxicillin + clavulanic acid-resistant isolates. In drinking water, there were 13 (56.5%) amoxicillin-resistant and 7 (30.4%) amoxicillin + clavulanic acid-resistant isolates. Resistance to clindamycin was consistently present in E. coli isolates obtained from both cat feces (76.9%) and drinking water (73.9%). Azithromycin showed varying resistance in E. coli isolates from cat feces (7.7%) and drinking water (4.4%). An association of AMR between cat feces and cat drinking water was identified with amoxicillin + clavulanic acid (p = 0.0038), whereas other drugs did not show an association between cat feces and cat drinking water (p ≥ 0.05; Table 1).

Table 1.

Prevalence of antimicrobial resistance patterns in E. Coli isolated from feces and drinking water of 52 cats

Group of antimicrobials	Antimicrobial name	E. coli from feces		E. coli from cat’s drinking water		p-value
		No. of isolates	No. of isolates with resistant E. coli (%)	No. of isolates	No. of isolates with resistant E. coli (%)
Bactericidal agents
Aminoglycosides	Amikacin	52	0	23	0	-
	Gentamycin	52	4 (7.7)	23	1 (4.4)	0.5924
Carbapenems	Imipenem	52	0	23	0	-
	Meropenem	52	1 (1.9)	23	0	0.5032
Cephalosporins	Ceftriaxone	52	4 (7.7)	23	1 (4.4)	0.5924
	Cephalexin	52	9 (17.3)	23	7 (30.4)	0.2007
Penicillin	Amoxicillin	52	19 (38.5)	23	13 (56.5)	0.1463
	Amoxicillin + clavulanic acid	52	3 (5.8)	23	7 (30.4)	0.0038
Quinolones	Ciprofloxacin	52	9 (17.3)	23	3 (13.0)	0.6423
	Enrofloxacin	52	9 (17.3)	23	2 (8.7)	0.331
	Norfloxacin	52	9 (17.3)	23	2 (8.7)	0.331
Bacteriostatic agents
Lincomycins	Clindamycin	52	40 (76.9)	23	17 (73.9)	0.7784
Macrolides	Azithromycin	52	4 (7.7)	23	1 (4.4)	0.5924
Sulfonamides	sulfamethoxazole + trimethoprim	52	7 (13.5)	23	1 (4.4)	0.2384

Isolates from cat feces exhibited significant susceptibility to antimicrobial agents (46.2%), whereas isolates from cat drinking water displayed lower susceptibility (17.4%; Table 2). A predominant resistance pattern to β-lactam was observed in isolates from both cat feces (26.9%) and drinking water (56.5%). Moreover, isolates from cat feces displayed resistance patterns against combinations of β-lactam with one, two and three other antimicrobial groups, whereas isolates from cat drinking water were mainly resistant to combinations of β-lactam with one and two other antimicrobial groups.

Table 2.

Antimicrobial resistance patterns of E. Coli from cat feces and cat drinking water

Class of antimicrobial agents	Resistance patterns	Cat feces		Cat drinking water
		No. of isolates (%)	95% CI	No. of isolates (%)	95% CI
0	-	24 (46.2)	32.2–60.5	4 (17.4)	5.0-38.8
1	β-lactams	14 (26.9)	15.6–41.0	13 (56.5)	34.5–76.8
2	β-lactams + quinolone β-lactams + sulfonamide β-lactams + aminoglycoside β-lactams + macrolide β-lactams + carbapenem	3 (5.8) 1 (1.9) 1 (1.9) 1 (1.9) 1 (1.9)	1.2–16.0 0.1–10.3 0.1–10.3 0.1–10.3 0.1–10.3	2 (8.7) 1 (4.3) - 1 (4.3) -	1.1–28.1 0.1–22.0 - 0.1–22.0 -
3	β-lactams + quinolone + aminoglycoside β-lactams + sulfonamides + macrolide	1 (1.9) 1 (1.9)	0.1–10.3 0.1–10.3	1 (4.3) 1 (4.3)	0.1–22.0 0.1–22.0
4	β-lactams + quinolone + sulfonamide + aminoglycoside β-lactams + quinolone + sulfonamide + macrolide	2 (3.9) 1 (1.9)	0.5–13.2 0.1–10.3	- -
5	β-lactams quinolone + sulfonamide + aminoglycoside + macrolide	2 (3.9)	0.5–13.2	-

The pairwise correlation analysis of bactericidal antimicrobial agents revealed a very strong positive correlation among quinolones, including ciprofloxacin, enrofloxacin and norfloxacin (r > 0.8, p < 0.01; Table 3). A moderate correlation was found within the cephalosporin group (r = 0.5586, p < 0.05). Additionally, a moderate correlation was found between aminoglycoside (gentamycin) and quinolones (ciprofloxacin, enrofloxacin and norfloxacin) as well as between amoxicillin + clavulanic acid and cephalexin (r = 0.4–0.6, p < 0.05). For bacteriostatic antimicrobials, the pairwise correlation analysis of antimicrobial sensitivity revealed a moderate correlation between azithromycin and sulfa-trimethoprim (r = 0.5253, p < 0.01; Table 4).

Table 3.

Pairwise correlation of antimicrobial resistance among bactericidal agents

Antimicrobial agents	Gentamycin	Meropenem	Ceftriaxone	Cephalexin	Amoxicillin	Amoxicillin +  clavulanic acid	Ciprofloxacin	Enrofloxacin	Norfloxacin
Gentamycin		-0.0429	0.3624**	0.2358*	0.2418*	0.0488	0.4549**	0.4731**	0.4813**
Meropenem	-0.0135		-0.0395	-0.0719	0.1024	-0.0545	-0.0557	-0.0551	-0.0532
Ceftriaxone	0.3624**	-0.0395		0.5586*	0.2991**	0.2696*	0.3603**	0.3756**	0.3829**
Cephalexin	0.2358*	-0.0719	0.5586**		0.2687*	0.5262**	0.2844*	0.2232	0.2349*
Amoxicillin	0.2418*	0.1024	0.2991**	0.2687*		0.2480*	0.3523	0.3459**	0.3316**
Amoxicillin + clavulanic acid	0.0488	-0.0545	0.2696*	0.5262**	0.2480*		0.2382*	0.1577	0.1664
Ciprofloxacin	0.4549**	-0.0557	0.3603**	0.2844*	0.3523	0.2382*		0.9442**	0.9558**
Enrofloxacin	0.4731**	-0.0551	0.3756**	0.2232	0.3459**	0.1577	0.9442**		0.9886**
Norfloxacin	0.4813**	-0.0532	0.3829**	0.2349*	0.3316**	0.1664	0.9558**	0.9886**

*p < 0.05, **p < 0.01

Table 4.

Pairwise correlation of antimicrobial resistance among bacteriostatic agents

Antimicrobial agents	Azithromycin	Sulfamethoxazole + trimethoprim
Clindamycin	0.0444	0.0476
Azithromycin		0.5253**
Sulfamethoxazole + trimethoprim	0.5253**

**p < 0.01

The presence of the carbapenemase gene was not identified in all E. coli isolates. As shown in Table 5, an analysis of ESBL-producing E. coli isolated in this study revealed that bla_TEM was detected most often (88.7%), whereas bla_TEM + bla_CTX−M, bla_TEM + bla_SHV and bla_TEM + bla_SHV+ bla_CTX−M were present in 8.5%, 1.4% and 1.4% of isolates, respectively. Only 7 isolates (9.9%; 6 from fecal samples and 1 from drinking water) harbored bla_CTX−M, which has the potential for cephalosporin degradation (Table 5). Phylogroup typing showed that phylogroup B2 (42.3%) was predominant, particularly among isolates from feces (32.4%). The second most predominant phylogroup was phylogroup B1 (15.5%), followed by unknown phylogroup (14.1%), F (9.8%), A or C (7.0%), E or clade I (7.0%) and D or E (4.2%). The phylogroup B2 isolates mainly carried bla_TEM. One isolate from the fecal samples (1.4%) of phylogroup D or E carried triple coexisting bla genes (bla_TEM, bla_SHV and bla_CTX−M). The distribution of bla genes in each individual phylogroup is also summarized in Table 5.

Table 5.

Distribution of extended-spectrum β-lactamase genes (bla_TEM, bla_SHV and bla_CTX−M) detected in E. Coli from cat feces (n = 54) and cat drinking water (n = 17)

Clermont phylogroup	Isolation source (n; %)	blaTEM (n; %)	blaTEM+, blaSHV (n; %)	blaTEM, blaCTX−M (n; %)	blaTEM, blaSHV, blaCTX−M (n; %)
A or C	Feces (n = 4; 5.6%)	3 (75.0%)	N/A	1 (25.0%)	N/A
	Water (n = 1; 1.4%)	1 (100.0%)	N/A	N/A	N/A
B1	Feces (n = 7; 9.9%)	7 (100.0%)	N/A	N/A	N/A
	Water (n = 4; 5.6%)	4 (100.0%)	N/A	N/A	N/A
B2	Feces (n = 23; 32.4%)	22 (95.7%)	N/A	1 (4.3%)	N/A
	Water (n = 7; 9.9%)	5 (71.4%)	1 (14.3%)	1 (14.3%)	N/A
D or E	Feces (n = 3; 4.2%)	2 (66.7%)	N/A	N/A	1 (33.3%)
	Water (n = 0; 0%)	N/A	N/A	N/A	N/A
E or Clade I	Feces (n = 3; 4.2%)	3 (100.0%)	N/A	N/A	N/A
	Water (n = 2; 2.8%)	2 (100.0%)	N/A	N/A	N/A
F	Feces (n = 5; 7.0%)	4 (80.0%)	N/A	1 (20.0%)	N/A
	Water (n = 2; 2.8%)	2 (100.0%)	N/A	N/A	N/A
Unknown	Feces (n = 9; 12.7%)	7 (77.8%)	N/A	2 (22.2%)	N/A
	Water (n = 1; 1.4%)	1 (100.0%)	N/A	N/A	N/A
Total	71 (100%)	63 (88.7%)	1 (1.4%)	6 (8.5%)	1 (1.4%)

N/A = Not available

Discussion

This study investigated the prevalence of antimicrobial resistance among cat feces and environmental (cat drinking water) E. coli isolates. Isolates had the highest susceptible to aminoglycosides and carbapenems and the highest resistance to lincomycin. Only amoxicillin + clavulanic acid had an association of resistance between cat feces and drinking water. Most of the E. coli isolates from both feces and drinking water showed resistant patterns to ß-lactam. The bla_TEM, bla_SHV and bla_CTX−M genes were detected in both feces and drinking water. Phylogroup B2 was the most prevalent phylogroup. These findings give insights into the AMR of E. coli and offer valuable guidance for selecting appropriate antimicrobial agents for cats with bacterial infections.

Both fecal and environmental E. coli isolates showed higher resistance to clindamycin. The increased resistance to cephalosporin and penicillin observed among isolates can be attributed to the widespread preference for penicillin use in humans [28], as well as cross-resistance within E. coli strains carrying the β-lactamase gene, which can degrade the β-lactam ring of these antimicrobial groups. In general, fecal E. coli displayed greater resistance than environmental strains, except to cephalexin, amoxicillin and amoxicillin + clavulanic acid. An association of amoxicillin + clavulanic acid resistance between feces and drinking water was also identified. These findings raise concerns about AMR in humans and cats residing in the same household.

Co-occurring resistance to multiple antimicrobial agents, including β-lactams with aminoglycosides, quinolones, sulfonamides, macrolides or carbapenems, was observed in both fecal and water E. coli isolates. Some isolates showed complex resistance patterns across multiple antimicrobial classes, highlighting the potential challenges in treating infections caused by these strains. Furthermore, fecal E. coli demonstrated a higher percentage of resistance to one to two antimicrobial groups, whereas environmental E. coli showed a higher percentage of resistance to three to five antimicrobial groups. This finding suggests that the levels of AMR in fecal and environmental E. coli may differ, emphasizing the importance of appropriate management strategies to control antimicrobial resistance in cats and their surroundings.

The pairwise correlation analysis revealed various correlations. Very strong positive correlations indicate shared mechanisms or similarity in modes of action or bacterial resistance development. Conversely, weaker correlations indicate different resistance patterns among antimicrobial groups. The presence of moderate to very strong correlations among antimicrobial drugs within the same category suggests a resistance pattern of E. coli within these specific antimicrobial categories. Therefore, avoiding the selection of antimicrobials from the same categories is essential to prevent the development of bacterial resistance. Interestingly, meropenem showed no significant correlations with any other antimicrobial agents, indicating differences in their sensitivity profiles. These findings carry significant implications for selecting appropriate treatments, optimizing therapeutic outcomes, and addressing further resistance. Additionally, they underscore the importance of conducting culture and sensitivity profiles before initiating antimicrobial therapy.

The presence and increase of E. coli strains carrying antimicrobial-resistant genes within humans, animals and the environments are a major global concern. The presence of ESBL-producing E. coli in cats and their environment may be linked to the spread of drug-resistant genes. Several studies have reported the dissemination of β-lactamase genes in cats [8, 10, 11]. It is known that AMR E. coli can also be present in healthy animals [6, 7], and the presence of ESBL genes in healthy animals can be transferred horizontally to pathogenic strains and may potentially spread resistance to humans through fecal-oral contamination [9, 29, 30]. β-lactamase genes and Clermont phylotyping were also investigated in this study. The analysis of drug-resistant strains revealed distinct patterns between fecal and environmental E. coli isolates. The detection of β-lactamase genes (bla_TEM, bla_SHV and bla_CTX−M) revealed a presence of antimicrobial resistance in these E. coli isolates. The bla_TEM gene was the predominant β-lactamase gene identified in E. coli isolates. Importantly, more bla_CTX−M isolates were obtained from the feces of cats, suggesting a high possibility of transmission from cats to the environment and humans. However, carbapenemase genes were not identified in the present study, consistent with low resistance to meropenem in only one E. coli isolate. These results may be attributed to the limited use of carbapenem as a restricted antimicrobial [31]. Therefore, carbapenems should be reserved as a last resort rather than being used as empirical antimicrobial agents.

E. coli isolates found in cats visiting a veterinary hospital displayed increased resistance to most of antimicrobial tested. This emphasizes the widespread presence of AMR and ESBL-producing E. coli in these cat populations. This poses a potential threat to both public and animal health. The microbes and AMR genes from cats can spread to humans through direct contact during petting and indirectly through fomites such as cat litter and water bowls (Fig. 1). Various factors, including antimicrobial usage, disease, stress, aging, poor dietary habits and lifestyle, influence differences in gut microbiota [32]. Inflammatory bowel disease and difficult-to-treat bacterial infections, attributed to AMR, have been linked with dysbiosis [32]. The U.S. Centers for Disease Control and Prevention (CDC) has identified slivered onions contaminated with E. coli O157:H7 as the cause of the outbreak at the fast food restaurant [33]. Since companion animals, especially domestic cats, can serve as potential reservoirs for multidrug-resistant enteropathogenic E. coli strains [34]. One health preventive measures should be implemented to improve food safety and security. It is advisable for cat owners to wash their hand after handling, cleaning up litters or feeding cats. Given that E. coli has been identified in cats with gingivitis [35], immunocompromised patients should avoid receiving pet licks on their face or having direct contact with sick cats.

Fig. 1.

The horizontal transmission of microbes and antimicrobial resistance genes from cats to humans occurs through direct contact with cats and indirect contact with fomites. The gut microbiota transfers antimicrobial resistance genes, colonizing both human and cat bodies. Occasionally, transmission of enteropathogenic E. coli by the fecal-oral route can induce acute diarrhea. The expansion number of E. coli are also associated with gut dysbiosis leading to intestinal inflammation and the progression of various intestinal disorders

Phylogroups A and B1, predominantly composed of commensal E. coli strains, are the most commonly detected phylogroups in human sources, while phylogroup B2 is prevalent among herbivorous and omnivorous mammals [36]. Additionally, phylogroup E is associated with cattle [37]. Notably, nearly half of the E. coli strains, particularly those isolated from fecal samples, were categorized into phylogroup B2, indicating potential pathogenicity from cats. Human and pet interactions can transmit both their pathogens and AMR genes [7, 38]. To counteract these issues, it is crucial to implement stewardship programs for controlling antimicrobial usage and encouraging responsible prescribing in health care settings [39]. Enforcing stringent infection control measures, incorporating proper hygiene practices and using effective sanitation protocols are also essential to control the spread of AMR bacteria within health care facilities and communities [39]. Furthermore, investing in research and development of new antimicrobial is vital to staying ahead of evolving resistance mechanisms, particularly the discovery of novel drugs that can effectively combat ESBL-producing bacteria. 

One limitation of this cross-sectional study was the inability to identify the clinical history or pre-existing diseases of enrolled cats. Consequently, the specific conditions of a cat’s health or previous exposure to antimicrobial agents could not be linked to the presence of AMR. Additionally, the minimum inhibitory concentration testing was not performed, which limits the ability to quantify resistance level. Although this study successfully demonstrated the presence of AMR of E. coli in cats, additional studies should investigate the specific health conditions of cats that lead to the presence of AMR. It is noteworthy that E. coli isolates from cats and their drinking water may not necessarily reflect the common clinical scenario involving E. coli as an enteropathogen in cats. Moreover, exploring the resistance patterns of E. coli across humans, pets and the environment is crucial from a One Health perspective. This study focused on the resistance patterns between pets and their environment, but did not investigate human E. coli strains.   Thus, more research is needed to understand the relationship of AMR between humans, pets and the environment.

Conclusions

The present study has provided valuable insights into the complex dynamics of AMR in cats and their environment, focusing on E. coli strains isolated from cat fecal samples and drinking water. Isolates showed the lowest resistance to carbapenems and the highest resistance to lincomycin (clindamycin) among the antimicrobial agents. The resistance to multiple antimicrobial agents and the predominance of phylogroup B2 underscore the challenges in managing AMR in the feline population. In addition, bla_TEM was the frequently identified β-lactamase gene in E. coli isolates from cat feces and their drinking water, suggesting the potential for β-lactam resistance and highlighting the complexity of AMR in cats.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

AMR

Antimicrobial resistance

ESBL

Extended-spectrum β-lactamase

E. coli

Escherichia coli

PCR

Polymerase chain reaction

Author contributions

PS: conceptualization, clinical case handling, data curation, writing—original draft preparation, formal analysis. SS, AK: clinical case handling, complementary examinations, reviewed manuscript. SA, PA, WS: comparative analysis of antimicrobial resistance, reviewed manuscript. NT: conceptualization, supervision, data curation, writing—original draft preparation, reviewed manuscript. All authors have read and approved the final manuscript.

Funding

This research is financial supported from Faculty of Veterinary Medicine, Kasetsart University: FFK (VET.KU2023-KPVRF01).

Data availability

The data used and/or analyzed in the present study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This animal use protocol was submitted and reviewed by the Kasetsart University Institutional Animal Care and Use Committee (ID#ACKU62-VET-030) and found to be in accordance with the guidelines of animal care and use under the Ethical Review Board of the National Research Council of Thailand for the conduct of scientific research. The committee approved and permitted the animal care and use to be conducted as stated in the animal use protocol for this research study. The study was carried out in compliance with the ARRIVE guidelines.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.