Isolation of Human Lineage, Fluoroquinolone-Resistant and Extended-β-Lactamase-Producing Escherichia coli Isolates from Companion Animals in Japan

An increase in human and veterinary fluoroquinolone-resistant Escherichia coli is a global concern. In this study, we isolated fluoroquinolone-resistant E. coli isolates from companion animals and characterized them using molecular epidemiological analysis, multiplex polymerase chain reaction to detect E. coli ST131 and CTX-M type extended-spectrum β-lactamases (ESBL), and multi-locus sequence typing analysis. Using plain-CHROMagar ECC, 101 E. coli isolates were isolated from 34 rectal swabs of dogs and cats. The prevalence of resistance to fluoroquinolone and cefotaxime was 27.7% and 24.8%, respectively. The prevalence of fluoroquinolone-resistant isolates (89.3%) was higher when CHROMagar ECC with CHROMagar ESBL supplement was used for E. coli isolation. The prevalence of cefotaxime resistance was also higher (76.1%) when 1 mg/L of ciprofloxacin-containing CHROMagar ECC was used for isolation. The cefotaxime-resistant isolates possessed CTX-M type β-lactamase genes (CTX-M-14, CTX-M-15, or CTX-M-27). Seventy-five percent of fluoroquinolone-resistant isolates were sequence types ST131, ST10, ST1193, ST38, or ST648, which are associated with extensive spread in human clinical settings. In addition, we isolated three common fluoroquinolone-resistant E. coli lineages (ST131 clade C1-M-27, C1-nM27 and ST2380) from dogs and their respective owners. These observations suggest that companion animals can harbor fluoroquinolone-resistant and/or ESBL-producing E. coli, in their rectums, and that transmission of these isolates to their owners can occur.


Introduction
For many people, companion animals such as dogs and cats are a part of their daily lives. These animals may be given antimicrobial agents to treat bacterial infections, such as urinary tract infections, pneumonia, and blood stream infections [1][2][3].
Antimicrobial resistance is of a major concern worldwide. In Japan, fluoroquinolones and third-generation cephalosporins are commonly and frequently used for the treatment of bacterial infections in humans and their companion animals [4,5]. Consequently, fluoroquinolone-resistant and/or extended-spectrum β-lactamase (ESBL)-producing Enterobacterales have been isolated from both human and companion animals [6,7]. Previously, we reported that fluoroquinolone-resistant and/or ESBL-producing E. coli colonized the guts of companion animals [7][8][9]. This suggests that the transmission of companion animalderived resistant E. coli isolates to humans could occur by direct contact between owners and their animals. E. coli is one of the principal pathogens for urinary tract infection in humans [10]. Thus, transmission of companion animal-derived antimicrobial-resistant E. coli might pose a potential risk to human health [11]. The international high-risk E. coli clone, sequence type (ST) 131, is the fluoroquinolone-resistant and/or ESBL-producing E. coli most frequently isolated from human infections [12,13]. Recent studies also reported isolation of E. coli ST131 from companion animals [14][15][16]. However, it is unclear whether direct transmission of E. coli ST131 occurs between humans and companion animals.
To evaluate potential hazards associated with companion animal-derived antimicrobialresistant bacteria to human health, investigation of the prevalence of antimicrobial resistance in companion animal bacterial isolates is required, particularly in relation to commonly and frequently used antimicrobial agents such as fluoroquinolones and thirdgeneration cephalosporins. In addition, molecular epidemiological analysis of antimicrobialresistant bacteria isolated from companion animals and their owners will shed light on the role of these animals in transmission of such isolates to humans, vice versa.
In this study, E. coli were isolated from the rectums of dogs and cats and the prevalence of fluoroquinolone-resistant and/or ESBL-producing E. coli was investigated to understand their possessions, especially focused on E. coli ST131, in rectum of companion animals. In some cases, E. coli isolates from companion animals and their owners was compared to estimate the possibility of the domestic cross-transmission.
For E. coli isolates obtained from the plain-CHROMagar ECC, the minimum inhibitory concentration (MIC) of ciprofloxacin ranged from <0.25 to >32 mg/L (the MIC 50 and MIC 90 were <0.25 and >32 mg/L, respectively). Cefotaxime MIC ranged from <0.25 to >128 mg/L (MIC 50 and MIC 90 were 0.5 and >128 mg/L, respectively). The prevalence of both of ciprofloxacin and cefotaxime resistance was 20.1% (21/101 isolates). For E. coli isolates obtained from the ciprofloxacin-added CHROMagar ECC, all isolates exhibited ciprofloxacin resistance, the ciprofloxacin MIC ranged from 8 to >32 mg/L (MIC 50 and MIC 90 were 32 and >32 mg/L, respectively). Cefotaxime MIC ranged from <0.25 to >128 mg/L (both the MIC 50 and MIC 90 were >128 mg/L). For E. coli isolates obtained from CHROMagar ECC with ESBL supplement, the ciprofloxacin MIC ranged from 0.5 to >32 mg/L (MIC 50 and MIC 90 were 32 and >32 mg/L, respectively). All isolates exhibited cefotaxime resistance, and the cefotaxime MIC ranged from 16 to >128 mg/L (both of MIC 50 and MIC 90 was >128 mg/L).

Isolation Rate of Ciprofloxacin-or Cefotaxime-Resistant E. coli Isolates from Companion Animals
The isolation rate of ciprofloxacin-and cefotaxime-resistant organisms among 34 companion animals using plain-CHROMagar ECC was 32.4% and 29.4%, respectively. Although the rates were mostly higher in ciprofloxacin-added CHROMagar ECC and CHRO-Magar ECC with ESBL supplement (Table 2), there were no significant differences between plain-CHROMagar ECC and ciprofloxacin-added CHROMagar ECC or CHROMagar ECC with ESBL supplement (p > 0.05).

Detection of CTX-M Type β-Lactamase
Multiplex polymerase chain reaction (PCR) detected a CTX-M type β-lactamase gene in 28 cefotaxime-resistant E. coli isolates obtained from CHROMagar ECC with ESBL supplement. Six isolates possessed a CTX-M group 1 β-lactamase gene, whereas 21 possessed a CTX-M group 9 β-lactamase gene. One strain possessed a β-lactamase gene that did not belong to CTX-M group 1, group 2, or group 9. All CTX-M group 1 β-lactamase genes were identified as CTX-M-15 by DNA Sanger sequencing. CTX-M group 9 genes were shown to be one of two types: CTX-M-14 (13 isolates) or CTX-M-27 (eight isolates).

MLST Analysis of Ciprofloxacin-Resistant E. coli Isolates from Companion Animals and Their Owners
Rectal swab samples were obtained from the owners of three dogs (#1, #2, and #3; Table 3). Ciprofloxacin-resistant E. coli isolates were obtained from dog #1 and its owner under all three culture conditions, and all were ST131 C1-M27 isolates. Ciprofloxacinresistant E. coli isolates were isolated from dog #2, with both plain-and ciprofloxacinadded CHROMagar ECC, and from its owner with ciprofloxacin-added CHROMagar ECC. All isolates from dog #2 and its owner were ST131 C1-nM27. Ciprofloxacin-resistant E. coli isolates were isolated from dog #3 with both ciprofloxacin-added CHROMagar ECC and CHROMagar ECC with ESBL supplement, and both were ST2380 isolates. An ST2380 isolate was also recovered from one of the two owners using ciprofloxacin-added CHROMagar ECC.

Discussion
In this study, we isolated fluoroquinolone-resistant or ESBL-producing E. coli isolates from the rectums of dogs and cats in Japan using three different selective media to evaluate the potential prevalence of antimicrobial-resistant E. coli in companion animals.
The prevalence of ciprofloxacin-and cefotaxime-resistant E. coli isolated using plain-CHROMagar was 27.7% and 24.8%, respectively. These prevalence values were lower than the prevalence of similar antibiotic-resistant E. coli isolates from diseased dogs (38.8% for ciprofloxacin-resistant and 26.5% for cefotaxime-resistant) and cats (37.5% for ciprofloxacinresistant and 26.6% for cefotaxime-resistant) reported in a national surveillance study carried out in Japan, Japanese Veterinary Antimicrobial Resistance Monitoring System (JVARM) [17]. The current study was based on swabs taken from animals attending companion animal clinics. As such, the data presented may reflect a selection bias because the samples came from dogs and cats without overt bacterial infections and which are less likely to have been previously exposed to antimicrobial treatments. On the other words, it indicates that companion animals harbour the ciprofloxacin-and cefotaxime-resistant E. coli in certain manner in their rectums. The prevalence of both ciprofloxacin and cefotaxime resistance was 20.1% in plain-CHROMagar, and the prevalence of ciprofloxacin-or cefotaximeresistant E. coli isolates was significantly increased by the presence of ciprofloxacin and ESBL supplement in the selection medium (Table 1). This observation is very important because if the ciprofloxacin-resistant E. coli isolates were selected, these mostly exhibit resistant to another frequent-used antimicrobial agents, cephalosporins (ESBL-producing). The co-resistance of fluoroquinolone-resistance and the third-generation cephalosporins should be taken into account in clinical settings in order to optimize antimicrobial treatment efficacy. Most of the cefotaxime-resistant E. coli isolates possessed either CTX-M-14, CTX-M-15, or CTX-M-27 type β-lactamases. These CTX-M types are widespread in human clinical settings in Japan, and CTX-M-possessing isolates frequently show co-resistance to fluoroquinolones [18,19] as well as companion animals [7,20]. This suggests a common feature between humans and companion animals. Detection of CTX-M-14, CTX-M-15, or CTX-M-27 type β-lactamases genes has also been reported among dogs and cats from other Asian regions, America, Europe, and Oceania (only CTX-M-15 among the three types was found in Africa) [6]. Thus, spread of the CTX-M type β-lactamases should be international concerns not only in human [21] but also in companion animals.
In this study, the isolation rate of ciprofloxacin-resistant and cefotaxime-resistant E. coli isolates from companion animals was 32.4% and 29.4%, respectively. Although there were no significances, slightly higher isolation rates of resistant isolates were observed when using ciprofloxacin-added or ESBL supplement agar. In the latter case, ciprofloxacin-resistant or ESBL-producing E. coli isolates should be predominant in the presence of additional antimicrobial selection, as shown by the increase of ciprofloxacin-and cefotaxime-resistant E. coli isolates in the selective medium (Table 1). Consistent with this, a previous study demonstrated that administration of first-generation cephalosporins (cefazolin and cephalexin) to dogs selected ESBL-producing E. coli isolates in feces [22].
According to MLST analysis, we identified eight fluoroquinolone-resistant E. coli lineages (ST131, ST10, ST1193, ST2380, ST38, ST648, ST5150, and ST5163). The ST131, ST10, ST1193, ST38, and ST648 lineages constituted 75.0% (12/16) of the total fluoroquinoloneresistant isolates, and all have been isolated previously from human specimens [23,24]. This result suggests that fluoroquinolone-resistant E. coli are present in companion animals and, since these animals are in close contact with humans, they are a potential hazard to human health. The most isolated ST in this study was ST131 (31.3%; 5/16 isolates). ST131 is recognized as an international high-risk clone in human clinical settings [15]. Previous studies in Japan also reported isolation of ST131 from ESBL-producing E. coli isolates in dogs and cats, and that ST131 was the predominant lineage [16,20]. Studies on fluoroquinolone-resistant E. coli isolates in human clinical settings reported that ST131 lineages accounted for more than 50% of these isolates [18,25]. Collectively, these studies suggest that ST131 is the predominant lineage in humans, but that it may be maintained in both humans and companion animals.
The recently emerged ST1193 lineage is also a pandemic fluoroquinolone-resistant clone [24]. In our previous study, neither ST131 nor ST1193 lineages were isolated from companion animals in the year of isolation, 2005 [26]. Moreover, our recent study revealed that, although ST1193 was not isolated from the human clinical setting in Japan (Sapporo) in 2008, it was subsequently isolated in 2020 [19]. These observations suggest that the pattern of emergence of fluoroquinolone-resistant E. coli isolates matches the E. coli clones emerging in humans. Consistent with this, all of the paired samples between companion animals and their owners shared the same fluoroquinolone-resistant E. coli lineages, including ST131 (Table 3). Additionally notable was that the same E. coli lineage, ST2380, was isolated from dog #3 and one of its two owners. This result also supports the possibility of domestic direct transmission of fluoroquinolone-resistant E. coli between humans and companion animals. In the limitations, the current study was conducted by small numbers of canine samples and paired (dogs and their owners) samples from limited animal clinics. Thus, the expanded studies will be required in future to further verify the probability. Although recent studies favor the possibility of cross transmission of fluoroquinolone-resistant E. coli between human and dogs [27][28][29], evidence of the isolation of identical lineages between dogs and their owners have fewer reported. Only two cases of the isolation of identical lineages of ESBL-producing E. coli (ST131 and ST38) from both of human and the household dogs were reported [30]. Thus, current observations worth to strengthen the domestic transmission in dogs and their owners.

Bacterial Isolation
All 34 rectal samples obtained by rectal swab were collected from companion animals (28 dogs and 6 cats) that attended three animal clinics in Sapporo, Japan. These swabs were spread on CHROMagar ECC (Kanto Chemical, Tokyo, Japan). One to three blue colonies of suspected E. coli were picked, sub-cultured on CHROMagar ECC, and stored as 20% glycerol stocks at −80 • C until use for subsequent experiments.
For the selection of fluoroquinolone-resistant and ESBL-producing E. coli isolates, swabs were inoculated into 1 mL of Trypticase Soy Broth (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) and incubated overnight at 37 • C. The resulting cultures were spread on ciprofloxacin (Bayer, Osaka, Japan) (1 mg/L)-added CHROMagar ECC and CHROMagar ECC supplemented with CHROMagar ESBL supplement (Kanto Chemical, Tokyo, Japan). Three colonies were picked up from each agar plate of all selective media. Identification of E. coli was performed using MALDI Biotyper (Bruker, Billerica, MA, USA). After receiving consent from the animal owners, E. coli from their feces was isolated as described above.

Antimicrobial Susceptibility
Susceptibility to ciprofloxacin and cefotaxime were determined by the broth microdilution method according to Clinical and Laboratory Standards Institute (CLSI) guidelines [31]. Isolates were inoculated into 1 mL of Trypticase Soy Broth and cultured overnight at 37 • C. After the cultivation, the suspensions were diluted and 2 µL of the dilutions (final inoculation was 5 × 10 5 cfu/mL) were added into each well of 96 well plate containing ciprofloxacin or cefotaxime in 100 µL of cation-adjusted Mueller-Hinton broth (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). MICs were determined after incubation for 20 h at 37 • C. Isolates that exceeded breakpoints (ciprofloxacin was 2 mg/L, and cefotaxime was 4 mg/L) were identified as resistant. E. coli ATCC 25922 was used as the reference strain.

Genetic Analysis
ST131, their clades (A, B, and C) and C subclades (C1-M27, C1-nM27, and C2) were identified by multiplex PCR, as described by Matsumura et al. [32]. Other STs of E. coli isolates were identified by MLST analysis as previously described by using primer sets of seven housekeeping genes (adk, fumC, icd, purA, gyrB, recA, and mdh) [33]. β-Lactamase gene CTX-M-1, 2, and 9 groups were detected by the multiplex PCR described by Dallenne et al. [34], and their variants were identified by Sanger DNA sequencing as described in a previous study [35]. All PCRs were performed by using Quick Taq HS Dye Mix (TOYOBO, Osaka, Japan) and MiniAmp Thermal Cycler (Thermo Fisher Scientific, Waltham, MA, USA) according to the same reaction conditions described in above references.

Statistical Analysis
Fisher's exact test was used for the statistical analysis. A p-value of <0.05 was considered statistically significant.

Ethical Approvemnet
This study was approved by the Ethical committee of Sapporo Medical University (No. 1-2-23).

Conclusions
This study demonstrates that companion animals possess fluoroquinolone-resistant and ESBL-producing E. coli isolates in their rectums, the majority of which comprise the same fluoroquinolone-resistant lineages (ST131, ST10, ST1193, ST38, and ST648) detected previously in human clinical samples. In addition, we found that most fluoroquinoloneresistant E. coli isolates exhibited co-resistance to cefotaxime due to the presence of CTX-M type β-lactamase genes (CTX-M-14, CTX-M-15, and CTX-M-27). The increased prevalence of fluoroquinolone-resistant and/or ESBL-producing E. coli isolates when ciprofloxacin or ESBL supplement was included in the selective medium, suggests the need for prudence in the use and administration of antimicrobials in veterinary settings to prevent the selection of these resistant E. coli. Furthermore, it is also important for dog and cat owners to implement infection control measures in their homes to prevent domestic transmission of fluoroquinolone-resistant and/or ESBL-producing E. coli between humans and companion animals.