Prevalence of extended-spectrum beta-lactamase-producing Escherichia coli strains in dairy farm wastewater in Chiang Mai

We investigated the prevalence of extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli strains in dairy farm wastewater in Chiang Mai, Thailand. We analyzed wastewater samples collected from 150 dairy farms and found that 88.7% of the farms (n = 133) were positive for ESBL-producing E. coli . Multiplex polymerase chain reaction (PCR) amplification was performed to characterize the presence of bla CTX-M , bla TEM , and bla SHV in ESBL-producing isolates. blaCTX-M was found in all isolates (n = 133), followed by blaTEM (80/133, 60.2%), whereas bla SHV was not detected in any isolate. blaCTX-M and blaTEM were present in 60.2% (80/133) of the isolates, and 39.8% (53/133) isolates carried bla CTX-M alone. Subgroup analysis showed that CTX-M -1 was the most prevalent subgroup among the isolates (129/133, 97.0%), followed by CTX-M -8 (2/133, 1.5%) and CTX-M -9 (2/133, 1.5%). The distribution of the phylogenetic groups was as follows: group A (100/133, 75.2%), followed by B1 (14/133, 10.5%), D (6/133, 4.5%), F (6/133, 4.5%), B2 (4/133, 3.0%), and E (3/133, 2.3%). Based on enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) and dendrogram analysis, 24 isolates were classified into clades I (n = 21), II (n =1), and III (n =2). Minor genetic differences were found in all clade I isolates. Our data suggest that the circulating of ESBL-producing E. coli carried at least one bla gene strain distributed in dairy farm wastewater in Chiang Mai. Genetic analysis was performed to characterize the 133 isolates for bla CTX-M , bla SHV , and bla TEM using multiplex PCR with specific primers. bla CTX-M was found in every isolate (133/133, 100%), followed by bla TEM in 60.2% of (80/133). bla was not detected in bla CTX-M co-existing TEM were by isolates, of regression


INTRODUCTION
Extended-spectrum beta-lactamase (ESBL)-producing bacteria from the Enterobacteriaceae family have been frequently reported (Coque et al., 2008;Dhillon and Clark, 2012;Rawat and Nair, 2010;Anuchatkitcharoen et al., 2020). ESBLs are enzymes that can inactivate the third-generation cephalosporins cefotaxime and ceftazidime (M'Zali et al., 2000). The presence of ESBL-producing Escherichia coli in food-producing animals has created a public health problem (Madec et al., 2017;Pitout and Laupland, 2008). Several investigations have shown that food-producing animals serve as a reservoir for ESBL-producing E. coli (Madec et al., 2017;Michael et al., 2017;Umpiérrez et al., 2017). Improper antibiotic use in animal husbandry has been associated with the spread of ESBLs (Landers et al., 2012;Marshall and Levy, 2011;Mellata, 2013;Natthida Sooksai et al., 2019). In food-producing animal farms, antibiotics are used to treat and prevent bacterial diseases (Landers et al., 2012;Michael et al., 2017) and as a result affect the normal flora of the animals, including E. coli. Survival adaptations by these bacteria involve acquiring antibiotic-inactivating enzymes encoded by antimicrobial resistance genes (Courvalin, 1994;West et al., 2011). The genes encoding ESBL acquired by E. coli in food-producing animals could then spread to humans via the food supply chain, contaminated water, and healthy fecal carriers. These bacteria are normal flora in the animal intestinal tract (Mellata, 2013;Michael et al., 2017;Watson et al., 2012).
The phylogenetic E.coli group was classified using quadruplex PCR to detect the presence or absence of genes, including chuA, yjaA, TspE4.C2, and asp. Therefore, seven phylogenetic groups were established: A, B1, B2, C, D, E, and F (Clermont et al., 2013). Phylogenetic group A was the most frequently isolated group in the mammalian intestinal tract, feces, and wastewater (Coura et al., 2015). Phylogenetic groups B2 and D were frequently isolated from extraintestinal E. coli, causing diseases such as mastitis and urinary tract infection (Horcajada et al., 2005;Le Gall et al., 2007;Liu et al., 2014).
ERIC-PCR generates DNA fingerprints using a pair of primers (Wilson and Sharp, 2006) and is an accurate, rapid, and reliable diagnostic approach (Meacham et al., 2003). The ERIC-PCR results and the dendrogram analysis of E.coli could be investigated for determination of genetic diversity and genetic source (Casarez et al., 2007;da Silveira et al., 2002;Leung et al., 2004).
In Chiang Mai, Thailand, ESBL-producing E. coli have been isolated from chicken and pig manure (Nuangmek et al., 2018;Rodroo et al., 2021). However, this is the first report of such isolates in dairy farm wastewater. ESBL-producing E. coli in wastewater can likely spread and contaminate inland water and natural environments.

351
This study aimed to investigate the prevalence, bla gene, phylogenetic group, and genetic diversity of ESBL-producing E. coli in wastewater collected from dairy farms in Chiang Mai, Thailand.

Sample collection
Samples were collected from 150 dairy farms located in three districts in Chiang Mai, Thailand: Mae On, Mae Wang, and Chai Prakan (n = 50 per districts). The wastewater samples were collected in sterile 50-mL sterilized bottles and stored on ice until use. The sample collection periods were as follows: Mae On in December 2019 and Mae Wang and Chai Prakan in January 2020. All farms were small holder farms with a dairy herd size of 10-50. The farm wastewater management system comprised wastewater collection in a manhole.

Bacterial isolation
E. coli strains were isolated by the direct plating method (Schauss et al., 2015). In brief, 10 mL of wastewater was inoculated in 90 mL of Luria-Bertani (LB) broth for 24 h at 37°C. The cultures were streaked on LB agar plates containing 2 mg/L cefotaxime and incubated for 24 h at 37°C. The resultant colonies were streaked on fresh LB agar plates and evaluated using the indole, methyl red, Voges-Proskauer, and citrate tests. After the biochemical tests, PCR was performed to detect a specific E.coli gene, uspA, using a specific pair of primers showed in Table 1.

Screening for ESBL-producing E. coli
ESBL-producing E. coli were screened using a standard double-disc synergy test according to Clinical Laboratory Standard Institute recommendations (CLSI, 2012). Briefly, bacterial suspensions were normalized to McFarland 0.5 turbidity standard in normal saline. Mueller-Hinton agar plates were inoculated with the bacterial suspensions and overlaid with disks containing ceftazidime (30 μg), ceftazidime (30 μg) + clavulanic acid (10 μg), cefotaxime (30 μg), and cefotaxime (30 μg) + clavulanic acid (10 μg). ESBL production was indicted by a growth inhibition zone of ≥5 mm around the combination drug disks versus the single-drug disks. CLSI recommends performing phenotypic confirmation of potential ESBL-producing E. coli isolates by testing both cefotaxime and ceftazidime, alone and in combination with clavulanic acid.

Genomic DNA extraction
In this study, we performed genomic DNA extraction using bacterial culture boiling in media as previously described (Madico et al., 1995). ESBL-producing E. coli were cultured on LB agar at 37°C for 24 h, and then a single colony was cultured in LB broth at 37°C for 24 h. A 200-μL aliquot of bacterial culture was added to 800 μL of distilled water, boiled at 95° for 10 min, and clarified by centrifugation at 12,000 rpm for 5 min. Then, the supernatant was collected and stored at −20°C.

Detection of blaTEM, blaSHV, blaCTX-M, and CTX-M subgroups
Multiplex polymerase chain reaction (PCR) was performed to detect the presence of bla TEM , bla SHV , and bla CTX-M in 133 ESBL-producing E. coli isolates. All CTX-M ESBL-producing isolates were then classified using multiplex PCR with subgroup-specific primers for CTM-X-1, CTM-X-2, CTM-X-8, CTM-X-9, and CTM-X-25. The primers used in this study are shown in Tables 1 and 2. PCR consisted of an initial enzyme activation step at 94°C for 5 min; 35 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 30 s; and final extension at 72°C for 7 min. The multiplex PCR reaction mixture contained 10 μL of 2× HS Quick Taq Dye Mix (including Taq polymerase; Toyobo, Osaka, Japan), 0.2 μL of each primer, 8.2 μL of DNase-free water, and 1 μL of bacterial DNA.

Phylogenetic grouping
The distribution of phylogenetic groups of ESBL-producing E. coli isolates was examined using a previously described protocol (Clermont et al., 2013). The E. coli strains could be classified into the following phylogenetic groups: A, B1, B2, C, D, E, and F. To examine the phylogenetic group of ESBL-producing E. coli isolates, we used a quadruplex PCR method based on the presence or absence of the chuA, yjaA, TspE4.C2 and arpA genes. ESBL-producing E. coli isolates, identified as phylogenetic group E or C isolates by quadruplex PCR, were further subjected to phylogenetic grouping by specific primers for further investigation of phylogenetic groups E and C. A list of the primers is provided in Table 3. PCR consisted of an initial enzyme activation step at 94 °C for 5 min; 35 cycles of denaturation at 94 °C for 30 s, annealing at 52 °C for 30 s, and extension at 72 °C for 3 min; and final extension at 72 °C for 7 min.  (Woodford et al., 2005) Primer

Enterobacterial repetitive intergenic consensus (ERIC)-PCR
ERIC-PCR was performed to assess the genetic diversity of eight ESBL-producing E. coli isolates per district. The ERIC primers used were 5′-ATGTAAGCTCCTGGGGATTCAC-3′ and 5′-AAGTAAGTGACTGGG-GTGAGCG-3′ (Wilson and Sharp, 2006). The cycling conditions included an initial enzyme activation step at 94°C for 5 min; 35 cycles of denaturation at 94°C for 30 s, annealing at 50°C for 30 s, and extension at 72°C for 7 min; and final extension at 72°C for 7 min.
The amplified products were separated using 1% agarose gel electrophoresis with RedSafe TM (Scientifix, NSW, Australia), and the green bands were visualized under ultraviolet light. The banding pattern was used as the principal clade determinant by PyElph (Pavel and Vasile, 2012).

Biosafety protocols
The biosafety protocols (number CMUIBC A-076202) were approved by the Institutional Biosafety Committee of the Faculty of Veterinary Medicine, Chiang Mai University.

Statistical analysis
Poisson regression was performed to evaluate the statistical significance of bla CTM-X and bla TEM co-expression and bla CTX-M expression alone.

Prevalence of ESBL-producing E. coli in dairy farm wastewater
Wastewater samples from a total of 150 dairy farms were screened. The prevalence of the ESBL-producing E. coli isolates in each district were as follows: 30.8% (n = 41) were from Mae On, 35.3% (n = 47) were from Mae Wang, and 33.8% (n = 45) were from Chai Prakan. The results showed that 88.7% of farms (n = 133) were positive for ESBL-producing E. coli.

bla and CTX-M subgroup distribution
Genetic analysis was performed to characterize the 133 isolates for bla CTX-M , bla SHV , and bla TEM using multiplex PCR with specific primers. bla CTX-M was found in every isolate (133/133, 100%), followed by bla TEM in 60.2% of isolates (80/133). bla SHV was not detected in any isolate. bla CTX-M co-existing with bla TEM were carried by 60.1% (80/133) of the ESBL-producing E. coli isolates, and 39.8% (53/133) of the isolates carried bla CTX-M alone. Poison regression analysis indicated that the likelihood of bla CTX-M co-expessing with bla TEM was highly significant (Z = 26.81).

ERIC-PCR
To determine the genetic diversity of ESBL-producing E. coli, we performed ERIC-PCR fingerprinting using the ERIC-1 and ERIC-2 primers. The selection criteria to investigate the genetic diversity of ESBL-producing E. coli strains included the phylogenetic group and the district. The first criteria selected was phylogenetic group A as this strain was predominant in both this and other studies. As the second criteria, we randomly selected each of the eight ESBL-producing E. coli strains per district from Mae On, Mae Wang, and Chai Prakan. Dendrogram analysis of the 24 isolates showed three putative clades: clade I, clade II, and clade III (Figure 1). Twenty-one isolates belonged to clade I (Mae On, 8; Mae Wang, 6; Chai Prakan, 7). ESBL-producing E. coli strains in clade I showed a minor genetic differences. A single isolate from Mae Wang belonged to clade II, and the remaining isolates belonged to clade III one each from Mae Wang and Chai Prakan.

DISCUSSION
We investigated the prevalence of ESBL-producing E. coli isolated from wastewater samples collected from dairy cattle farms in Chiang Mai as these sites may be the origin of contaminants that leach into natural water sources. The waste management of dairy farms usually involves the collection of wastewater in manholes. The land surrounding the farm is arable and has irrigation systems. Irrigation systems use natural water as reservoirs to store water. The reservoirs are used for human consumption and cultivation (Ngammuangtueng et al., 2019) This irrigation system might be the link between the dairy farms and the natural water sources.
ESBL-producing E. coli were identified in 88.7% of wastewater samples in three districts: Mae On, 30.8%; Mae Wang, 35.3%; and Chai Prakan, 33.8%. The prevalence of ESBL-producing E. coli in this study may be explained by the improper use of antibiotics. Although this study did not investigate the use of antibiotics in dairy farms, a survey was previously conducted in Mae On showing that livestock farmers lack understanding of the negative effects of antibiotic use. (Sooksai et al., 2019;Suriyasathaporn et al., 2012). There is also a causal relationship between antibiotic use and the development of antimicrobial resistance. ESBL-producing E. coli are resistant to many commonly used antibiotics (Hardy, 2002;Landers et al., 2012). Multidrug-resistant E. coli has been reported in Chiang Mai and in the central region of Thailand (Boonyasiri et al., 2014). ESBL-producing E. coli strains have been isolated in the cattle intestinal tract and the farm environment, including pen floors, collecting yards, and slurry ponds (Boonyasiri et al., 2014). Our data suggest that farm wastewater, where cow feces and water are collected, is a reservoir of ESBL-producing E. coli.
There are many ESBLs, such as CTX-M, SHV, TEM, OXA, PER, VEB, BES, GES, SFO, TLA, and IBC (Rawat and Nair, 2010). CTX-M, SHV, and TEM are the most prevalent in ESBL-producing E. coli worldwide (Bonnedahl et al., 2009;Cantón et al., 2012;Coque et al., 2008;Sasaki et al., 2010;Watson et al., 2012). In this study, bla CTX-M was found in 100% (n = 133) of samples, 60.2% of which (n = 80) also carried bla TEM . These findings are consistent with prior reports of bla CTX-M as the dominant bla gene in dairy cattle in many parts of the world (Cormier et al., 2016;Rehman et al., 2017;Tamang et al., 2013). Similar to a study of ESBL-producing E. coli in Thailand, bla CTX-M has been reported as the dominant bla gene in clinical isolates, human feces, pig feces, farm waste, and canal water (Boonyasiri et al., 2014;Bubpamala et al., 2018;Niumsup et al., 2018;Nuangmek et al., 2018;Runcharoen et al., 2017;Sasaki et al., 2010). A study conducted in Lamphun, which neighbors Chiang Mai, showed a predominance of bla CTX-M and bla TEM co-expression in human feces and pig feces (Seenama et al., 2019). The co-expression of bla CTX-M with bla TEM was significant (z = 26.81), as determined using Poison regression. Therefore, our data suggested that at least two antibiotic resistance genes in ESBL-producing E. coli provide best-fit adaptation.
Based on genetic differences, CTX-M can be classified into five groups: CTM-X-1, CTM -X-2, CTM-X-8, CTM-X-9, and CTM-X-25, and each group is composed of a number of variants (Lewis et al., 2007;Livermore and Hawkey, 2005;Sasaki et al., 2010;Tamang et al., 2013;Watson et al., 2012). CTX-M subgroups can be characterized using multiplex PCR with specific primers or nucleotide sequences (Tamang et al., 2013;Woodford et al., 2005). In this study, CTX-M-1 was found to be the most prevalent type of ESBL in all the study areas, with CTX-M-2 and CTX-M-25 in far fewer locations. We were unable to investigate the most common CTX-M-1 variant, CTX-M-15, because of limited resources and facilities. Further studies will be needed to explore the enzyme kinetics of bla CTX-M espectailly CTM-X-1 with bla TEM in vitro and in vivo.
A quadruplex PCR amplification of the genes chuA, yjaA, TspE4. C2, and asp was performed for phylogenetic classification (Clermont et al., 2013). E. coli phylogenetic groups differ in their phenotypic and ecological characteristics (Duriez et al., 2001). Phylogenetic group A are the most well adapted to different environments and are the most frequently isolated group in the mammalian intestinal tract, feces, and wastewater (Coura et al., 2015). In contrast, phylogenetic groups B2 and D are frequently isolated from extraintestinal E. coli, which cause diseases such as mastitis and urinary tract infection (Horcajada et al., 2005;Le Gall et al., 2007;Liu et al., 2014). In this study, ESBL-producing E. coli isolates predominantly belonged to phylogenetic group A (75.2%), followed by B1 (10.5%), D (4.5%), F (4.5%), B2 (3%), and E (2%). The predominance of groups A and B1 is consistent with reports of these groups in household wastewater investigations conducted worldwide (Figueira et al., 2011). The variability in E. coli antibacterial resistance and virulence genes among phylogenetic groups is considered a significant concern. ESBL-producing E. coli strains from cow mastitis has been associated with phylogenetic group A (Dogan et al., 2012;Goldstone et al., 2016;Tomazi et al., 2018) and E. coli phylogenetic group B1 isolates carrying eae were isolated from diarrheic calves (Coura et al., 2017). Importantly, the most prevalent ESBL-producing E. coli isolate in this study belonged to phylogenetic group A or B1, and these groups can be disseminated to natural water sources and thus reach human communities that consume the contaminated water.
In this study, ERIC-PCR and dendrogram analysis were used to classify 24 group A ESBLs-producing E. coli isolates into clades I, II, and III. Slight genetic differences were found in clade I isolates (Mae On,8;Mae Wang,6;and Chai Prakan,7). These findings suggest that phylogenetic group A ESBL-producing E. coli were distributed in dairy farms in Chiang Mai. Bacteria constantly undergo adaptation and exchange genetic materials with other bacteria via plasmids, integration, and bacteriophages (Médigue et al., 1991). These genetic modifications provide survival advantages and increase population fitness, leading to the emergence of clones (Brisson-Noël et al., 1988;Price et al., 2008).

CONCLUSION
We studied the prevalence of ESBL-producing E. coli isolates in dairy cattle farm wastewater in Chiang Mai. The results showed a high prevalence of ESBL-producing E. coli isolates. We also studied the distribution of antimicrobial resistance genes, phylogenetic groups, and genetic diversities of the isolated strains. Further studies should assess whether ESBL-producing E. coli strains isolated from dairy cattle are a source of inland water contamination or human infection.