The emergence and spread of Enterobacteriaceae with reduced susceptibility to quinolone drugs is a growing concern. While the introduction of quinolone drugs initially suggested a delay in resistance development due to their synthetic nature, this expectation has not been fully realized. Clinically significant resistance can arise from rare simultaneous mutations in multiple target genes. This study aimed to understand how fluoroquinolone use influences resistance selection and emergence. A previous study demonstrated that a single dose of enrofloxacin in calves increased fecal *E. coli* resistant to ciprofloxacin. The high relevance of quinolone resistance to human medicine, given its critical role in treating serious infections, necessitates further research to identify specific resistance mechanisms selected by fluoroquinolones in cattle. This is crucial for generating evidence-based data to evaluate the impacts of fluoroquinolone use in cattle and to propose targeted interventions if needed. Early research suggested plasmid-mediated quinolone resistance (PMQR) in *E. coli* was possible, even if not naturally occurring. Studies using multidrug resistance (MDR) systems demonstrated the in vitro existence of quinolone-resistant bacteria and transferable quinolone resistance genes in the environment. The first PMQR gene, *qnr*, was identified on a multi-resistance plasmid in *Klebsiella pneumoniae*. Five families of *qnr* genes have been described, along with other PMQR genes such as efflux pumps (*oqxAB*, *qepA*) and a variant of an aminoglycoside acetyltransferase, *aac(6')-Ib-cr*. These PMQR genes have been associated with similar mobile genetic elements as extended-spectrum β-lactamases (ESBLs). The use of quinolone drugs as a selective pressure for quinolone resistance in clinical settings has been reported, including the potential for co-selection of multidrug-resistant isolates.

Several studies have provided insights into the mechanisms and prevalence of quinolone resistance. Gomez-Gomez et al. (1997) showed the possibility of plasmid-mediated quinolone resistance (PMQR) in *E. coli*. Martinez et al. (1998) demonstrated how quinolone-resistant bacteria and transferable quinolone resistance genes could exist in the environment. The discovery of the *qnr* gene on a multi-resistance plasmid highlighted the role of plasmid-mediated transfer in spreading resistance. Subsequent research identified several PMQR gene families and their association with mobile genetic elements similar to those of ESBLs. Studies have also shown the selection pressure exerted by quinolone use on the emergence of quinolone resistance and co-selection of multidrug resistance. The use of fluoroquinolones in veterinary medicine, particularly enrofloxacin, raises concerns about the potential contribution to the selection and spread of resistance, impacting human health due to the critical role of these drugs in human medicine and the potential for transmission of resistant bacteria between animals and humans.

This study involved 84 calves randomly allocated to three groups: enrofloxacin (ENR, n=22), tulathromycin (TUL, n=24), and control (CTL, n=21). Calves were selected based on high risk for bovine respiratory disease (BRD). Fecal samples were collected longitudinally at 0, 7, 28, and 56 days. *E. coli* was cultured, and DNA was extracted from isolates. PCR was used to screen for cephalosporin, quinolone, and tetracycline resistance genes. QRDR screening was performed using Sanger sequencing. A total of 264 *E. coli* isolates were selected for genotypic characterization (phenotypically resistant to ceftriaxone or to both ceftriaxone and ciprofloxacin). Antimicrobial susceptibility testing was done using Kirby-Bauer disk diffusion with a modified NARMS panel of 12 antimicrobials. Two multiplex PCR protocols were used: one for PMQR genes (*qnrA, qnrD, qnrB, qnrS, oqxAB, Aac(6')Ib-cr, qepA, and qnrC*) and another for β-lactamase (*bla-TEM, bla-CTX-M, bla-OXA*) and tetracycline genes (*tetA* and *tetB*). Data were analyzed using descriptive and chi-square analyses in JMP 15, mixed logistic regression (PROC GLIMMIX, SAS), and Fisher exact tests (PROC FREQ, SAS). Sensitivity and specificity were calculated to assess the accuracy of genotype predictions of resistant antimicrobial phenotypes. A heatmap visualized the distribution of resistance genes and QRDR mutations in isolates with phenotypic resistance to ciprofloxacin.

The study detected *aac(6')Ib-cr*, *bla-CTX-M*, *bla-TEM*, *tetA*, and *tetB* resistance genes. The most common mutation detected was *gyrA* 594 (T→C). Five QRDR mutations showed significantly higher risk in ciprofloxacin-resistant isolates: *parC* 239 (G→T), *gyrA* 248 (C→T), *gyrA* 259 (G→A), *gyrA* 570 (C→T), and *parE* 1372 (T→G). *E. coli* isolates with *aac(6')Ib-cr* also had mutations in *parC* 239, *gyrA* 248, and *gyrA* 259. The ENR group showed significantly higher *aac(6')Ib-cr* at day 7 (P=0.048) and *tetA* at day 28 (P=0.03) compared to the CTL group. Mixed logistic regression revealed a significant association between the *gyrA* 248 (C→T) mutation and the ENR group at days 7 (OR: 11.59; P=0.03) and 28 (OR: 9; P=0.055) compared to the CTL group. The most common genotypic resistance profile was *bla*-TEM, *tetA*, *tetB*, and the most common phenotypic profile was resistance to multiple drug classes (amoxicillin/clavulanic acid, amoxicillin, enrofloxacin, cefoxitin, ceftriaxone, chloramphenicol, ciprofloxacin, nalidixic acid, sulfisoxazole, tetracycline, and trimethoprim sulfamethoxazole). The QRDR mutation profile with the highest sensitivity and specificity for ciprofloxacin resistance included simultaneous mutations in *parC* 239, *gyrA* 248, *gyrA* 259, *gyrA* 570, and *parE* 1372.

The study found that enrofloxacin treatment resulted in a significantly higher prevalence of the *aac(6')Ib-cr* gene at day 7 and the *gyrA* 248 mutation at days 7 and 28 compared to the control group. The *aac(6')Ib-cr* gene, a variant of an aminoglycoside acetyltransferase, confers reduced ciprofloxacin susceptibility. The limited temporal effect of *aac(6')Ib-cr*, with its absence at day 56, suggests a limited longitudinal selective advantage. This could be due to the limited competitive advantage conferred by enrofloxacin or increased microbial competition. The observed co-occurrence of *aac(6')Ib-cr* with specific QRDR mutations suggests clonal expansion of resistant strains. The high sensitivity and specificity of a particular QRDR mutation profile for ciprofloxacin resistance support the hypothesis of multiple stepwise genomic adaptations in response to selective pressures. The prevalence of ESBL genes (*bla-CTX-M*, *bla-TEM*) and the observed correlations between multidrug resistance genotypes and phenotypes highlight the co-selection of resistance to multiple antimicrobial classes, reducing the effectiveness of critical antimicrobials. The presence of PMQR genes on plasmids co-existing with ESBL genes underscores the risk of co-selection and transfer of resistance to multiple drugs.

Enrofloxacin treatment significantly increased the detection of the *gyrA* 248 mutation and the *aac(6')Ib-cr* gene in preweaned calves. Correlations between multidrug resistance genotypes and phenotypes indicate the need for further research to identify factors contributing to the selection of multidrug-resistant isolates. This is particularly relevant for isolates with simultaneous resistance to fluoroquinolones and cephalosporins, emphasizing the importance of antimicrobial stewardship strategies to mitigate the spread of resistance.

The study used breakpoints primarily based on CLSI criteria for human isolates, potentially affecting the interpretation of results for cattle isolates. The relatively small sample size may limit the generalizability of the findings. Future studies could focus on a larger sample size with species-specific breakpoints and a more comprehensive analysis of the fecal microbiota.