Evaluation and Characterization of Quinolone-Resistant Escherichia coli in Wastewater Treatment Plant Effluents

Abstract

The increasing global incidence of quinolone antimicrobial resistance poses a considerable public health concern. The aquatic environment, particularly wastewater treatment plants (WWTPs), serves as a major reservoir for antibiotic-resistant bacteria (ARB) and antibiotic resistance genes (ARGs), leading to the dissemination of antibiotic resistance. This study aimed to assess the prevalence and factors contributing to quinolone antibiotic resistance in Escherichia coli isolates obtained from effluents of 33 WWTPs. A total of 1082 E. coli isolates were analyzed, 32.6% and 17.1% of which showed resistance to nalidixic acid and ciprofloxacin, respectively. Phenotypic and genotypic analyses of antibiotic resistance demonstrated that quinolone resistance primarily originated from chromosomal mutations in the gyrA, parC, and parE genes, known as quinolone resistance-determining regions (QRDRs). The amino acid substitution at codon 83 in gyrA was closely associated with nalidixic acid resistance, whereas substitutions at codon 87 in gyrA and codon 80 in parC were significantly associated with ciprofloxacin resistance. The plasmid-mediated quinolone resistance (PMQR) genes qnrS and qnrB were identified in 41 isolates (11.5%) and 15 isolates (4.2%), respectively. Thus, we confirmed that the quinolone resistance in E. coli in WWTPs primarily occurs through QRDR mutations rather than through the acquisition of PMQR genes. Phylogenetic analysis revealed that most quinolone-resistant isolates belonged to the B1, A, B2, and D phylogenetic groups. Notably, the B2 group, which is responsible for extraintestinal infections, exhibited the highest rate of quinolone resistance. These findings provide novel insights into the presence and mechanisms of quinolone resistance in E. coli isolates from WWTPs, emphasizing the need for further research and understanding of quinolone resistance in the environment.

1. Introduction

Quinolones are antibacterial drugs widely used to treat various bacterial infections in humans. Owing to the extensive use of these medications, the population of quinolone-resistant bacterial strains has continuously increased since the 1990s [1]. The clinical use of the first quinolone, nalidixic acid, dates back to 1962. In the mid-1980s, ciprofloxacin, a fluoroquinolone with broad antibacterial properties, became available for clinical use [2]. Ciprofloxacin is one of the most widely used fluoroquinolones globally and is the second generation of this class of antibiotics. Similar to the case in other antibiotics, the increasing prevalence of quinolone resistance poses a threat to the efficacy of this class of drugs.

Quinolone resistance mechanisms can be classified into two main categories: mutations and acquisition of resistance genes [3]. Bacteria become resistant to quinolones when they acquire mutations in quinolone resistance-determining regions (QRDRs), including DNA gyrase and DNA topoisomerase IV (gyrA, gyrB, parC, and parE). These mutations affect the ability of the drug to bind to target enzymes by altering their binding affinity [2,4]. In particular, Escherichia coli strains with high levels of fluoroquinolone resistance are typically linked to multiple mutations in the QRDRs of topoisomerase enzymes [5]. Although mutations in bacterial gyrase and topoisomerase IV genes were previously believed to cause fluoroquinolone resistance, evidence shows that plasmid-associated resistance factors are involved and may contribute to the spread of antibiotic resistance in the environment [6]. Plasmid-mediated quinolone resistance (PMQR) has been observed in clinical and environmental bacteria worldwide. These mechanisms provide limited resistance to fluoroquinolones and increase the likelihood of selecting other chromosome-encoded resistance mechanisms for this class of antimicrobial agents [7]. To date, researchers have identified six qnr gene families, which are referred to as qnrA, qnrB, qnrC, qnrD, qnrS, and qnrVC [7].

The aquatic environment contains a large and varied number of antibiotic-resistant bacteria (ARB) and antibiotic resistance genes (ARGs), making it a hub for numerous clinical bacteria and pollutants [8]. One of the major sources of ARB and ARGs in the environment is effluent from wastewater treatment plants (WWTPs), which are the main anthropogenic sources. WWTPs, which are crucial reservoirs of various mobile antibiotic resistance elements, play a pivotal role in the recombination and dissemination of ARGs in the environment [9,10]. Active monitoring of ARB and ARGs in WWTPs has been conducted to confirm the occurrence and transmission patterns of antibiotic resistance. With the development of molecular biological techniques, recent monitoring studies have used culture-independent methods, such as quantitative PCR (qPCR), high-throughput qPCR, and digital PCR [11,12]. However, these methods cannot directly compare the phenotypic and genotypic characteristics of individual bacterial strains. Thus, culture-based methods are necessary to compare these characteristics.

The objective of this study was to investigate the quinolone resistance profiles of E. coli in WWTPs and elucidate the underlying genetic mechanisms by using a culture-based method. During 2020–2021, quinolone-resistant E. coli was monitored in the effluent of 33 WWTPs, and the residual antibiotic concentrations were measured. Furthermore, the correlation between quinolone-resistant isolates was analyzed to gain a better understanding of the spread of high-risk clones.

2. Materials and Methods

2.1. Isolation and Identification of E. coli

During 2020–2021, effluent samples were collected from 33 wastewater plants in Korea. Samples were collected once in 2020, whereas samples were collected during the dry and rainy seasons in 2021. Typically, in South Korea, following the activated sludge process, wastewater treatment plants employ methods including chlorine disinfection, ultraviolet (UV) disinfection, and ozone disinfection to achieve compliance with the effluent water quality standard for the total coliform group, set at 3000 cfu/mL. To obtain a pure isolate of E. coli, 100 μL of wastewater effluents was streaked onto the CHROMagar™ orientation agar (CHROMagar, Paris, France) by using a disposable bacterial loop. After 24 h of incubation at 35 °C, pink colonies were selected and transferred on NA-MUG agar (Difco laboratories, Sparks, MD, USA) to confirm the species. After 24 h of incubation, fluorescent colonies were selected under ultraviolet conditions. The isolated strains were determined to be E. coli using both 16S rRNA sequencing and matrix-assisted laser desorption ionization-time of flight mass spectrometry.

2.2. Phenotyping Antibiotic Resistance Testing

The phenotypic profiles of the E. coli isolates were determined using the microdilution method. Broth microdilutions are typically used to test antimicrobial susceptibility. Bacterial suspensions adjusted to 0.5 McFarland scale were inoculated onto a 96-well microtitration plate (KRNV5F, Daejeon, Republic of Korea) that was commercialized in the livestock sector with 16 antimicrobial agents. After incubation for 1 day at 35 °C, the minimum inhibitory concentration (MIC) was checked with the naked eye or an automated reader, which facilitates reading microdilution tests and recording results with a high ability to discern growth in the wells. For the determination of MIC endpoint, the breakpoints of 16 antimicrobial agents in the panel were set in accordance with the guidelines of the Clinical Laboratory Standard Institute [13] and European Committee on Antimicrobial Susceptibility Testing [14]. The KRNV5F panel has a MIC test range of 2–128 μg/mL for nalidixic acid and 0.12–16 μg/mL for ciprofloxacin. The MIC breakpoints of nalidixic acid and ciprofloxacin were 32 and 1 μg/mL, respectively. The distribution curve was obtained by plotting the number of isolates against the quinolone concentration.

2.3. Molecular Analysis

Genomic bacterial DNA was extracted using the DNeasy Plant Mini Kit (Qiagen Inc., Valencia, CA, USA). In the case of PMQR, qnrA, qnrB, qnrC, qnrD, qnrS, and qnrVC were analyzed using multiplex PCR updated by Kraychete et al. [7], whereas aac(6′)-Ib-cr and qepA were analyzed using single PCR [15]. In the case of QRDR, gyrA, gyrB, parC, and parE were analyzed and sequenced to investigate chromosomal mutations in these regions [16,17,18,19]. Information regarding each primer is presented in Table S1. The PCR amplicons were visualized using 1.5% agarose gel electrophoresis to confirm the band size. The products were purified using a QIAquick PCR Purification Kit (Qiangen Inc., Hilden, Germany) and sequenced by Macrogen Sequencing Service (Macrogen, Seoul, Republic of Korea).

2.4. Phylo-Typing of Escherichia coli

All E. coli isolates (A, B1, B2, C, D, E, and F) were phylotyped using quadruplex PCR with four phylogenetic markers (chuA, yjaA, TspE4.C2, and arpA) following the method developed by Clermont et al. [20]. A second PCR was performed targeting the arpA and trpA regions to identify groups E and C, respectively. Groups B2 and F were subdivided into new group, G, by performing the triplex PCR of trpA, cfaB, and ybgD using the Clermont method [21]. The primers used in this study are detailed in Table S2.

2.5. Determination of Antimicrobial Residue

The 16 antimicrobial residues from the wastewater samples were determined as follows. As pretreatment, 500 mL of each sample was filtered with a 0.2 μm PVDF filter, and 900 µL was transferred into amber auto-sampler vials and combined with 100 µL of 1% acetic acid, 40 mg/mL ethylene-diaminetetraacetic acid disodium salt dihydrate (Na₂EDTA), and 10 µL of 10 ng/mL isotopically labeled standards. Then, 200 μL of the pretreated sample was analyzed using high-performance liquid chromatography coupled with tandem mass spectrometry. The limits of quantification of ciprofloxacin and nalidixic acid were 0.006 and 0.011 μg/L, respectively.

2.6. Statistical Analysis

The correlation between quinolone resistance determinants was predicted using Fisher’s exact test or chi-square test. Chi-square test was used to determine any significant differences between the proportions of quinolone-resistant isolates and phylogenetic groups of E. coli. When the requirements for chi-square testing were not met, Fisher’s exact test was performed. A two-sided p-value of <0.05 was considered to indicate statistical significance.

3. Results

3.1. Antibiotic Susceptibility Profiles

In total, 1082 E. coli isolates were acquired from wastewater samples using a selective method, and 358 quinolone-resistant E. coli isolates (33.1%) were confirmed via antibiotic resistance tests. A total of 353 isolates were resistant to nalidixic acid (32.6%), and 185 isolates were resistant to ciprofloxacin (17.1%). Among the 353 nalidixic acid-resistant E. coli isolates, 180 had MIC > 128 (51.1%), 85 had MIC = 128 (24%), 63 had MIC = 64 (17.8%), and 25 had MIC = 32 (7.1%). Among the 185 ciprofloxacin-resistant isolates of E. coli, 65 had MIC > 16 (34.9%), 47 had MIC = 16 (25.8%), 27 had MIC = 8 (14.5%), 8 had MIC = 4 (4.3%), 10 had MIC = 2 (5.4%), and 28 had MIC = 1 (15.1%). The distribution of E. coli isolates based on the MIC results is shown in Figure 1.

3.2. Phenotypic and Genetic Characterization

Quinolone-resistant E. coli isolates were investigated for PMQR genes and QRDR mutations. Of the total isolated E. coli, 358 isolates had phenotypic characteristics of quinolone resistance, of which 261 isolates had QRDR mutations or carried PMQR genes and 97 isolates had none of the quinolone resistance determinants investigated in this study.

In the PMQR analysis, 358 E. coli isolates were used to identify plasmid-mediated quinolone resistance genes. Specifically, qnrA, qnrB, qnrC, qnrD, qnrS, qnrVC, aac(6′)-Ib-cr, and qepA were investigated using PCR. PMQR analysis results revealed 57 isolates harboring at least one quinolone resistance gene (

Table 1. Status of PMQR gene possession in quinolone-resistant E. coli isolates (n = 358).

Quinolone-Resistant Gene	qnrB	qnrS	acc(6′)-Ib-cr	At Least One Resistant Gene
Number of isolates with resistant genes	15 (4.2%)	41 (11.5%)	9 (2.5%)	57 (15.9%)

). Of the isolates that showed quinolone resistance, only 15.9% had resistance genes carried by plasmids. Among the qnr family genes, only qnrS and qnrB were detected in this study. qnrS was found in 41 isolates (11.5%), and qnrB was found in 15 isolates (4.2%). Another PMQR gene, acc(6′)-Ib-cr, was detected in nine isolates (2.5%). Although we only examined isolates during 2020–2021, the results indicated that PMQR genes were less widely dispersed in E. coli in WWTPs.

In the QRDR analysis, as a result of investigation on chromosomal mutations in gyrA, gyrB, parC, and parE, 262 (73%) of the 358 quinolone-resistant isolates were shown to have mutations in gyrA, parC, and parE. No mutations were observed in gyrB. Quinolone resistance is caused by mutations in gyrA (codons 83 and 87), parC (codons 80 and 84), and parE (codons 416 and 458). The mutation and frequency results are shown in

Table 2. Status of QRDR mutations in quinolone-resistant E. coli isolates (n = 358).

Gene	Position	Mutation	Number of Isolates (Percent)
gyrA	83	Ser83Leu	249 (69.4%)
		Ser83Ala	6 (1.7%)
	87	Asp87Asn	94 (26.2%)
		Asp87Tyr	22 (6.1%)
		Asp87Gly	1 (0.3%)
parC	80	Ser80Ile	124 (34.5%)
	84	Glu84Gly	3 (0.8%)
		Glu84Lys	2 (0.6%)
		Glu84Val	9 (2.5%)
ParE	416	Leu416Phe	30 (8.4%)
	458	Ser458Ala	37 (10.3%)

.

The predominant amino acid mutations in the QRDR of gyrA were serine to leucine at codon 83 (69.4%) and aspartic acid to asparagine at codon 87 (26.2%). parC showed a serine-to-isoleucine mutation at position 80 (34.5%). parE had a leucine-to-phenylalanine mutation at position 416 (8.4%) and a serine-to-alanine mutation at position 458 (10.3%). Almost all isolates with high levels of ciprofloxacin resistance (≥32 mg/L) carried double or triple mutations in gyrA, parC, and parE. Amino acid mutation in the QRDR of gyrA, parC, and parE in 358 E. coli isolates, along with the corresponding MICs of nalidixic acid and ciprofloxacin were presented in Table S3.

We analyzed the association between QRDR mutations or PMQR gene retention and the type of quinolone antibiotic used in 358 isolates exhibiting nalidixic acid and/or ciprofloxacin resistance. A dendrogram showing the relationship among these isolates, PMQR genes, and QRDR mutations indicated that nalidixic acid resistance was closely related to the amino acid substitution at codon 83 in gyrA, whereas ciprofloxacin resistance was significantly associated with the amino acid substitution at codon 87 in gyrA and codon 80 in parC (Figure 2). Additionally, we confirmed through the chi-square test that the amino acid mutations at codon 87 in gyrA and codon 80 in parC were significantly related to ciprofloxacin resistance.

3.3. Determination of Antibiotic Residue

Residual antibiotics were detected in the wastewater effluent samples collected from 33 WWTPs. Nalidixic acid was not detected, but ciprofloxacin was detected in all WWTPs. During the dry season, the concentration of ciprofloxacin ranged from 0.015 µg/L to 0.29 µg/L, while during the rainy season, it ranged from 0.01 µg/L to 0.18 µg/L.

3.4. Phylogenetic Group Analysis

Most of the 358 quinolone-resistant isolates were categorized into four phylogenetic groups, with group B1 having the highest proportion (27%), followed by groups A (23%), B2 (22%), and D (17%). When analyzing all 1082 E. coli isolates, group B1 had the highest number, followed by groups A, D, and B2, in descending order. As the total number of isolated strains was high, groups B1 and A had more quinolone-resistant strains than the other groups. However, considering the number of isolated strains in each group, group B2 had the highest quinolone resistance rate, followed by group D (

Table 3. Quinolone resistance rates and ratio of QRDR and PMQR by phylogroup of E. coli.

	A	B1	B2	C	D	E	F	G	CladeI	Unknown
Total isolates (n = 1082)	294	388	129	19	145	38	20	23	6	20
Quinolone-resistant isolates (n = 358)	81	96	77	7	62	5	12	8	3	7
Quinolone-resistant rate (%)	27.6	24.7	59.7	36.8	42.8	13.2	60.0	34.8	50	35
QRDR rate (%)	61 (75.3)	61 (63.5)	70 (90.9)	6 (85.7)	45 (72.6)	2 (40.0)	7 (58.3)	5 (62.5)	2 (66.7)	3 (42.9)
PMQR rate (%)	14 (17.3)	20 (20.8)	2 (2.6)	2 (28.6)	9 (14.5)	2 (40.0)	2 (16.7)	1 (12.5)	0 (0)	5 (35.0)

, Figure 3a). Additionally, the results of the Pearson chi-square test confirmed a significant difference in quinolone antibiotic resistance rates among the phylogenetic groups (p < 0.05).

Analysis of the resistance acquisition (PMQR, QRDR) for the four major phylogenetic groups confirmed significant differences between the groups (p < 0.05). Although isolates harboring PMQR genes were associated with groups A and B1, isolates with QRDR mutations were significantly associated with group B2 (Figure 3b,c).

Groups A, B1, and D showed similar proportions of isolates harboring PMQR genes and/or QRDR mutations, but more than 90% isolates from group B2 acquired quinolone resistance due to QRDR mutations (Figure 3b). In particular, qnrS, qnrB, and acc(6′)-Ib-cr were widely distributed in groups A and B1 (Figure 3c). In the QRDR analysis, mutations in gyrA (codons 83 and 87) and parC (codon 80) were distributed evenly among the phylogenetic groups (A, B1, B2, and D), while mutations in parE (codon 458) mainly occurred in group B1 and mutations in parE (codon 416) occurred only in group B2 (Figure 3d).

4. Discussion

In this study, we used a culture-based method to assess the prevalence of and factors contributing to quinolone antibiotic resistance. Among the E. coli isolates obtained from WWTPs, the resistance rates for nalidixic acid and ciprofloxacin were 32.6% (353/1082) and 17.1% (185/1082), respectively. Recently, the World Health Organization (WHO) has issued cautious guidelines for the use of fluoroquinolones in human and veterinary medicine because of the strong link between their use and increased resistance [22]. Nalidixic acid is not used in human or animal settings, whereas ciprofloxacin is predominantly used in human medicine. In Korea, the defined daily doses per 1000 inhabitants per day (DID) for ciprofloxacin used in humans ranged from 1.8 to 1.9, and the observed ciprofloxacin resistance rate in E. coli during 2020–2021 was between 41% and 41.3% [23]. On the other hand, in the field of livestock, although sales data for nalidixic acid and ciprofloxacin have not been reported since 2019, the antibiotic resistance rates of E. coli isolated from fecal samples of livestock such as cattle, pigs, chickens, and ducks were 30.9% for ciprofloxacin and 38.7% for nalidixic acid. Also, the resistance rates of E. coli derived from slaughtered livestock were higher, with 41.2% for ciprofloxacin and 49.6% for nalidixic acid, surpassing the rates observed in E. coli from fecal samples in 2021 [24]. In particular, chickens exhibited the highest quinolone resistance rate, which is comparable to findings reported in European countries [22]. The antimicrobial resistance rates of E. coli to nalidixic acid and ciprofloxacin were markedly elevated in both human and animal domains. Annual surveillance is consistently performed to analyze and report monitoring outcomes. However, the quinolone resistance rates are relatively moderate in the environmental sector compared with other sectors. Thus, continuous monitoring is necessary in the environmental sector to enhance our understanding of the resistance escalation and transmission dynamics in the environment.

Antibiotic resistance tests were conducted first to compare the phenotypic and genotypic characteristics of antibiotic resistance. Considering the absence of specifically tailored MIC breakpoints for environmental settings, we evaluated antibiotic resistance in accordance with criteria applied in clinical settings. However, applying a clinical definition of resistance to the environment based on the probability of treatment failure in human patients may be inappropriate. For instance, in this study, the MIC breakpoint for ciprofloxacin was set to 1, following the clinical definition of quinolone resistance. However, when examining the distribution of quinolone resistance in E. coli isolated from WWTPs based on antimicrobial concentrations, the resistance tended to decrease until the MIC reached 4 and then increased thereafter. Therefore, considering the resistance patterns, we adjusted the MIC breakpoint for environmental E. coli isolates separately. Martìnez et al. [25] proposed ecological breakpoints that appear suitable for analyzing environmental microorganism resistance, suggesting the need for further research in environmental settings to gain a better understanding of resistance in the environment.

Resistance to quinolone antibiotics can develop through mutations in the target sites of antibiotics and acquisition of ARGs. In the present study, quinolone resistance in WWTPs primarily resulted from mutations in the QRDR of the chromosome despite the presence of isolates carrying PMQR genes. A recent study on quinolone-resistant E. coli from Norwegian turkey meat has also yielded comparable results [26]. Genetic mutations in gyrA and parC were identified as significant contributors to resistance, with limited dissemination of PMQR in E. coli. Mutations in gyrA, parC, and parE were the main mechanisms underlying quinolone resistance in WWTP-derived E. coli. Most quinolone-resistant isolates harbored one or two mutations in gyrA and one mutation in either parC or parE. Previous studies have highlighted amino acid changes at positions 83 and 87 in gyrA as common and significant mutations associated with quinolone resistance [27]. Our study corroborated these findings, revealing a strong correlation between ciprofloxacin resistance and mutations at position 87 in gyrA, contrary to the previous understanding of mutations at position 83 [28].

In the context of PMQR, analysis of qnr family genes revealed the presence of qnrS and qnrB, as well as a few instances of aac(6′)-Ib-cr, which is known for its ability to acetylate fluoroquinolones, such as ciprofloxacin and norfloxacin [3]. In previous studies, qnrB was the most frequently detected gene within the qnr family [29]. However, in the context of wastewater effluents, a higher frequency of qnrS than of qnrB was found in the present study. In an investigation on PMQR retention in quinolone-resistant E. coli isolated from wastewater in the Czech Republic, qnrS was identified as the most common gene, followed by qnrB [22]. Kaplan et al. [6] observed that qnrS is the predominant plasmid-mediated resistance determinant in raw sewage and sludge. Recent survey results in Nepal also validated that qnrS is among the abundant ARGs in WWTPs [30]. These findings suggest that qnrS is the dominant PMQR gene in E. coli populations found in wastewater. Similar to qnr family genes, qepA and aac(6′)-Ib-cr, which are genes associated with antibiotic resistance, were previously believed to be prevalent in aquatic environments [31]. However, the current study found no evidence of qepA, and only partial confirmation was obtained for aac(6′)-Ib-cr. These results indicate that E. coli is not the primary vector for the dissemination of qepA and aac(6′)-Ib-cr in WWTPs and that PMQR genes are not widely distributed in WWTPs. In certain isolates, quinolone antibiotic resistance has been identified even in the absence of PMQR genes and the absence of mutations in the QRDR genes. This phenomenon might be elucidated by mutations in regulatory genes that govern the expression of natural efflux pumps localized to bacterial membranes—a quinolone resistance mechanism [3] that remained unexplored in the present study. However, this study only investigated isolates obtained in 2020–2021. Thus, continuous monitoring of WWTPs is necessary because favorable conditions may vary over time.

Li et al. [32] performed a qPCR analysis of quinolone resistance genes in wastewater using a culture-independent method and demonstrated a high prevalence of qnrD, qnrS, qepA, oqxB, and oqxA in nearly all wastewater samples. However, in the present study, the frequency of these genes was very low in quinolone-resistant E. coli strains from wastewater. Röderova et al. [22] and Hooper et al. [3] reported that the presence of PMQR genes has a limited effect on the overall level of quinolone resistance. However, PMQR genes can facilitate the selection of high-level resistance in the presence of quinolones. Therefore, further monitoring and surveillance are needed to explore whether other microorganisms are involved in the dissemination of resistance determinants.

Recent observations have identified the aquatic environment as the primary source of PMQR genes that are transmitted to bacterial isolates through mobile genetic elements [2]. However, this study revealed that the transmission pattern via plasmids in E. coli was not significant in wastewater environments. Instead, a greater resistance effect was observed owing to gene mutations caused by residual antibiotics. Li et al. [32] indicated a strong correlation between the PMQR and quinolone residues in soil and wastewater samples. Thus, we also anticipated a significant correlation between QRDR and residual antibiotics. Recent studies have demonstrated that repeated exposure to high concentrations of ciprofloxacin can lead to mutational changes in quinolone target genes, even outside the established canonical QRDR [33]. In the present study, ciprofloxacin was detected with nearly 100% prevalence in the effluents of 33 WWTPs. Moreover, ciprofloxacin detection was not limited to Korea but was observed worldwide, surpassing a detection frequency of 90% [34]. Consequently, there is a strong likelihood of ongoing genetic variation caused by the residue.

Clermont et al. [20,21] proposed a classification system for E. coli strains, categorizing them into eight phylogenetic groups (A, B1, B2, C, D, E, F, and G) with an additional cryptic Escherichia clade I. The analysis revealed that groups B2 and D demonstrated higher quinolone resistance than the other isolates. Groups B2 and D are typically associated with extraintestinal infections, whereas groups A and B1 are common among commensal isolates [35,36]. B2 is associated with high virulence and urinary tract infections [37]. In a recent study of quinolone-resistant E. coli isolated from food-producing animals and animal-derived food in the Philippines, phylogroups A, B1, and D were predominant. However, WWTPs had a higher prevalence of group B2, which exhibited significantly increased quinolone resistance.

A significant majority (over 70%) of E. coli isolates obtained from human specimens belong to groups B2 and D, whereas over 70% of E. coli isolates recovered from animals belong to groups A and B1. Ciprofloxacin is predominantly used in humans, and the prevalence of ciprofloxacin resistance is higher in E. coli isolates obtained from humans than in WWTPs. Thus, the higher quinolone resistance of groups B2 and D compared with the major E. coli groups A and B1 can be explained by this factor. Belotindos et al. [29] demonstrated that E. coli isolates carrying mutations in QRDRs and PMQR determinants are not predominantly affiliated with group B2, whereas isolates with amino acid substitutions in QRDRs are significantly associated with group B1. However, the present study revealed that more than 90% of the isolates belonging to group B2 harbored mutations in QRDRs (Table 3), and the majority of qnr genes were detected in groups A and B1. This observation provides evidence that commensal isolates belonging to groups A and B1, commonly obtained from animals, exhibit a propensity to acquire quinolone resistance through plasmid exchange. By contrast, group B2, predominantly detected in patients, is believed to exhibit a higher tendency for genetic mutations, possibly resulting from sustained antibiotic exposure. However, as sequence type analysis provides a more accurate approach to elucidating the relationship between the acquisition of quinolone resistance in E. coli within WWTPs and its origin, additional comprehensive research on this subject will be necessary.

The presence of E. coli was significantly elevated in WWTP effluents. However, establishing it as the primary factor contributing to the dissemination of quinolone resistance determinants is challenging. The primary driver of quinolone resistance in these environments is primarily associated with genetic mutations rather than the transfer of antibiotic resistance factors, such as plasmids. Nonetheless, research on this matter should be subjected to comprehensive evaluation through continuous monitoring.

5. Conclusions

This study investigated the prevalence of and factors contributing to quinolone resistance in E. coli isolates from WWTP effluents. The primary mechanism of quinolone resistance in WWTPs involved QRDR mutations, despite the presence of isolates carrying PMQR genes. Specific amino acid substitutions in gyrA and parC are associated with nalidixic acid and ciprofloxacin resistance, respectively. Analysis of PMQR genes revealed the presence of qnrS, qnrB, and aac(6′)-Ib-cr. The quinolone-resistant isolates were classified into phylogenetic groups B1, A, B2, and D, with group B2 showing the highest resistance. Isolates with PMQR genes were associated with groups A and B1, whereas those with QRDR mutations were significantly associated with group B2. These findings suggest that group B2, typically isolated from patients, may develop antimicrobial resistance through genetic mutations due to exposure to ciprofloxacin. Furthermore, although WWTPs are recognized as major reservoirs for horizontal gene transfer, E. coli is not a significant bacterium involved in the spread of quinolone resistance genes. Therefore, the acquisition of resistance in individual species within WWTPs warrants further investigations to gain a comprehensive understanding of the dissemination of resistance in the environment.