Lentic and effluent water of Delhi-NCR: a reservoir of multidrug-resistant bacteria harbouring blaCTX-M, blaTEM and blaSHV type ESBL genes

Antimicrobial resistance is not restricted to clinics but also spreading fast in the aquatic environment. This study focused on the prevalence and diversity of extended-spectrum β-lactamase (ESBL) genes among bacteria from lentic and effluent water in Delhi-NCR, India. Phenotypic screening of 436 morphologically distinct bacterial isolates collected from diverse sites revealed that 106 (∼24%) isolates were ESBL positive. Antibiotic profiling showed that 42, 60, 78 and 59% ESBL producing isolates collected from Ghazipur slaughterhouse, Lodhi garden pond, Hauz Khas lake and Jasola wastewater treatment plant, respectively, were multidrug-resistant (MDR). The multiple antibiotic resistance (MAR) index varied from 0.20 to 0.32 among selected locations. The prevalence of ESBL gene variants blaSHV, blaTEM and blaCTX-M were found to be 17.64, 35.29 and 64%, respectively. Furthermore, the analysis of obtained gene sequences showed three variants of blaCTX-M (15, 152 and 205) and two variants of blaTEM (TEM-1 and TEM-116) among ESBL producers. The co-existence of 2–3 gene variants was recorded among 48% ESBL positive isolates. New reports from this study include the blaCTX-M gene in Acinetobacter lwoffii, Enterobacter ludwigii, Exiguobacterium mexicanum and Aeromonas caviae. Furthermore, the identification of blaTEM and blaSHV in an environmental isolate of A. caviae is a new report from India.


INTRODUCTION
Extensive use of antibiotics in developed as well as in developing countries had led to a rapid increase in their concentration in the aquatic environment, as a significant fraction of antibiotics is unutilized and released into wastewater or effluent from treated wastewater (Zhou et al. 2011;Sultan et al. 2020). Due to the frequent use of third-generation cephalosporins for the treatment of infections caused by members of Enterobacteriaceae, the upshot has been recorded in resistance to these antibiotics. Extended-spectrum β-lactamases (ESBLs) are recognized to provide resistance to most of the β-lactam antibiotics along with the other type of antibiotics like aminoglycosides, monobactam, carbapenem, ansamycin, tetracycline and polymyxins (Taco et al. 2014;Gogry et al. 2019;Siddiqui et al. 2020). The presence of various classes of antimicrobials in the wastewater contributes as an important factor towards the emergence of resistance and their dissemination (Aminov & Mackie 2007). There are numerous reports of antibiotic-resistant bacteria (ARB) in environmental settings such as wastewater treatment plants (WWTPs), rivers, hospital effluents, lakes and ponds (Falodun & Ikusika 2019;Bhattacharyya et al. 2020;Smyth et al. 2020). The emergence and distribution of antibiotic resistance genes (ARGs) among pathogenic organisms can impair human and animal fitness. In the environment, ARG transmission occurs through the horizontal gene transfer (HGT) mechanism (Zhang et al. 2011). Reuse of wastewater for different domestic as well as agricultural purposes favours the uptake of ARGs by plants, humans and animals (Fatta-Kassinos et al. 2011). Wastewater and treated effluents have contributed towards an increase in antibiotic-resistant microorganisms in the environment and thereby enhanced the ARG load into the water bodies (Karthikeyan & Meyer 2006). The WWTPs are measured as significant reservoirs for ARGs since activated sludge facilitates the richness of microbial community and also supports the HGT using mobile genetic elements (MGEs) of ARGs (Zhang et al. 2011;Moura et al. 2012).
To prevent the probable health risks originating from ARB in the aquatic environment, the evaluation of such infectious and harmful bacteria is of high importance. Therefore, the present study was carried out to determine the prevalence and diversity of ESBL genes among bacterial isolates from lentic and effluent water in Delhi-NCR, India.

Inspection of sites and sample collection
In this study, four different sites were selected to investigate the occurrence and distribution of ARGs among bacterial isolates from lentic and effluent water in Delhi-NCR, one of the largest metropolitan cities of India with a population size of approximately 24 million. Sampling site Hauz Khas lake is a 125 acres 14th-century water body in New Delhi, now shrunk in size to 15 acres and has a maximum depth of 1-2.5 m. The lake is a site of recreation for people, and its water is used for irrigation, animal use in a nearby park (Deer Park). The second site, the Lodhi garden pond, present in Lodhi estatea tourist site in Delhi, is a duck-fish integrated aquaculture pond. The pond is land-locked and does not receive any sewage water. The third site, Ghazipur slaughterhouse, has a capacity of 1,000 head of cattle to be slaughtered per day. A bio-methanation plant converts digested and undigested stomach contents of ruminants into gas and slurry. The blood released during the slaughtering process is treated before being released into the river Yamuna. The samples were collected from effluent water of the wastewater treatment plant. The fourth site, Jasola wastewater treatment plant, is the biggest treatment plant in South Delhi having the potential to impact the environment to a large extent. Out of four sites, two sites were effluent treated water, viz. Ghazipur slaughterhouse, Ghazipur U.P. India (28°62 0 N, 77°32 0 E) and Jasola wastewater treatment plant, Jasola Vihar, New Delhi (28°5 4 0 N, 77°28 0 E), and the other two sites were lentic water bodies, viz. Hauz Khas lake, Hauz Khas, New Delhi, Delhi (28°55 0 N, 77°19 0 E) and Lodhi garden pond, Lodhi Estate, New Delhi, Delhi (28°59 0 N, 77°21 0 E). Water samples were aseptically collected in a 100 mL sterile glass bottle, stored in an icebox (∼4°C) and transported within 4 h from the sampling sites to the Microbiology Research Laboratory. Before transportation, temperature and pH were recorded for each sampling site. The samples were collected during the year 2015-2016.

Isolation and identification of ESBL positive bacteria
For isolation, samples were serially diluted up to 1 Â 10 À4 , after dilution of the samples 100 μL of each dilution were spread on Luria Agar (LA) plates and kept for 16-18 h incubation at 37°C . After incubation, colonies with different morphology (based on size, appearance, colour, texture, etc.) were picked and streaked on LA plates for further detection of ESBL producing isolates. Morphologically distinct isolates from all four sites were screened by a preliminary test. It was performed using third-generation antibiotics cephalosporins and monobactam by the disk diffusion method. Suspected ESBL producers in the preliminary test were further confirmed by the phenotypic disc confirmatory test (PDCT) which was performed according to Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI 2016) for ESBL production. Escherichia coli ATCC 25922 and Klebsiella pneumoniae ATCC 700603 were used as ESBL negative and positive control, respectively. 2012). The multiple antibiotic resistance (MAR) index of isolates against different classes of antibiotics was calculated. MAR index is calculated by formula a/b, where 'a' signifies the number of antibiotics to which the isolate was resistant and 'b' represents the number of antibiotics to which the isolate was subjected. MAR index was inferred for each location and specific isolate as detailed by Krumperman (1983). Minimum inhibitory concentration (MIC) of eight antibiotics, i.e., ceftazidime (CAZ), ciprofloxacin (CIP), ampicillin (AMP), rifampicin (RIF), cefotaxime (CTX), chloramphenicol (C), colistin (CL) and trimethoprim (TR), against ESBL positive isolates was determined by the broth microdilution method as per CLSI guidelines. Luria Broth (LB) was used for the culture of bacterium and as a reference. 10.24 mg/mL stock solution of antibiotics were prepared and inoculated into the first well of each column onto the polystyrene plate. The progressive serial broth dilution was performed to obtain the required concentrations of 1, 024,512,256,128,64,32,16,8,4 and 2 μg/mL in 96 well microtitre plate, and medium was inoculated with 10 μL of 1,000 times diluted 0.1 OD600 culture and incubated overnight at 37°C. The optical density (OD) was measured at 600 nm using a microplate reader (Thermo Scientific MultiscanGo).

Isolation of genomic and plasmid DNA
For isolation of genomic and plasmid DNA, a pure bacterial colony was suspended into sterile LB media and incubated overnight to obtain mid to late log phase of the bacterial culture. The cells were pelleted by centrifugation. Further, the genomic DNA from ESBL þ samples was extracted by Phenol:Chloroform:Isoamyl alcohol (PCI) procedure and plasmid DNA isolated using alkaline lysis methods. Extracted DNA was visualized over the ultraviolet (UV) transilluminator (GeNei-India). These genomic and plasmid DNA were used as a template in a polymerase chain reaction (PCR) for amplification of various ESBL genes. For amplification of genes, 50 μL master mix were prepared that comprises of 39.5 μL PCR grade water, 5 μL of 10Â buffer (with 25 mM MgCl 2 ), 1 μL of deoxynucleotide triphosphate mixture (10 mM dNTP), 1 μL of forward primer, 1 μL of reverse primer, 0.5 μL Taq polymerase (5 U/μL, GeNei-India) and 2 μL of extracted DNA template. PCR conditions were kept as follows: Initial denaturation for 5 min at 95°C, followed by 30 cycles of cyclic denaturation for 1 min, annealing for 45 s at respective annealing temperature (Tm) of different genes, an extension for 1 min at 72°C and a final extension for 10 min at 72°C. After that amplified PCR products were loaded in well (1% agarose gel). After running of gel, DNA bands were visualized over UV transilluminator (GeNei-India). Primers used in PCR amplification for various genes are mentioned in Table 1.
This study CTX-25R

Molecular characterization of CTX-M type ESBL genes
Phenotypically ESBL positive isolates were screened for the presence of different ESBL genes using specific primers (Table 1). PCR products of various blaCTX-M, blaTEM and blaSHV genes were purified and sequenced with corresponding primers. Sequencing was carried out by Agrigenome, India. The obtained gene sequence data were examined using FinchTV and Bio-Edit. Nucleotide sequences were subject to Basic Local Alignment Search Tool (BLAST) to identify different gene variants.
Bacterial identification using 16S rRNA approach All the CTX-M positive samples were characterized by 16S rRNA sequencing. To perform it, 16S rRNA-specific primers were used for PCR amplification of 16S rRNA gene and the obtained products were sequenced. The obtained gene sequence data were analyzed, and homology was searched using BLAST at NCBI.

Conjugation transfer experiment
Conjugation was performed to learn about the capacity of ESBL positive bacteria to move their resistant factors (blaCTX-M, blaTEM and blaSHV) to recipient bacteria. Four isolates, i.e., JST71, HK106, SH52 and LG1, from different sampling sites harbouring ESBL genes and having MDR phenotype were selected as a donor, and the E. coli J 53 (Sodium Azide resistant) was selected as recipient. The primary culture of donors was grown in cefotaxime (CTX 2 μg/mL) supplemented LB media and the recipient grown in azide (100 μg/mL) supplemented LB medium. Secondary culture of both donor and recipient were mixed in equal volume in LB and kept at 37°C for 24 h (HiMedia, India). After that, serial dilutions were prepared in LB medium and spread on LA plates containing NaN 3 (100 μg/mL) and CTX (2 μg/mL). Plasmid DNA was isolated from transconjugant and used as a template for PCR amplification to detect the presence of ESBL genes.

Screening of ESBL producers
A total of 436 non-duplicate bacterial isolates were isolated from Ghazipur slaughterhouse (103), Lodhi garden pond (98), Hauz Khas lake (140) and Jasola wastewater treatment plant (95). Based on the disc diffusion method, the preliminary and PDCT depicts that 106 isolates were ESBL positive. Thus, the prevalence of ESBLs producing bacteria in Ghazipur slaughterhouse, Lodhi garden pond, Hauz Khas lake and Jasola wastewater treatment plant was found to be 32, 23, 20 and 23%, respectively (Table 2).

Conjugation experiment
The conjugation assay revealed that donor bacterial isolates including LG1, HK 106, JST 71 and SH52 from different sites were able to transfer their resistance genes to the recipient bacteria ( Figure 2). The blaCTX-M genes were successfully amplified in all four tarns-conjugants and blaTEM from HK106 and SH52. None of the four transconjugant exhibited amplification of the blaSHV gene.   (Zurfluh et al. 2013;Chen et al. 2016). There are few reports of a higher percentage of ESBL producers with MDR bacteria from freshwater bodies (Taco et al. 2014;Bajaj et al. 2015). The occurrence of the high percentage of ESBL producing bacteria worldwide is challenging and worrisome. In our study, the majority of ESBL positive isolates exhibited resistance towards cefotaxime, ceftazidime, ceftriaxone, aztreonam, ampicillin and cefazolin similar to report from China urban river sediment (Lu et al. 2010). Furthermore, resistance towards colistin was found at a higher scale than earlier reports of clinical and environmental samples (Ansari et al. 2015). Thereby, a heightened resistance towards colistin which is the last drug available in the current pipeline to treat infections caused by MDR bacteria is alarming. Among ESBLs producing isolates 42, 60, 78 and 59% from the Ghazipur slaughterhouse, Lodhi garden pond, Hauz Khas lake and Jasola wastewater treatment plant, respectively, were found to be MDR. From the site Hauz Khas lake, the resistance pattern was found similar to those reported from surface water and wastewater by Taco et al. (2014). The MAR index value for each site was .0.2 which suggests a deleterious situation and a high pollution load in a particular sampling area indicating that the antibiotic exposure of a specific area could be higher (Krumperman 1983). High MIC value (2-. 1,024 μg/mL) for ceftazidime, cefotaxime, ampicillin, colistin and trimethoprim showed elevated resistance to β-lactam and folate pathway inhibitors among ESBL producers. Our findings are in line with previous reports from the freshwater of Delhi-NCR (Bajaj et al. 2015;Azam et al. 2016). The blaCTX-M gene was detected among 64% ESBL positive isolates which corroborates previous reports from India and Australia (Reinthaler et al. 2010;Bajaj et al. 2015). In this study, we observed that blaCTX-M-15 was the most prevalent type which is similar to other reports (Azam et al. 2016;Maravićet al. 2016). The present study also justifies that the blaCTX-M-15 is the frequently dispersed ESBL among Enterobacteriaceae (Bevan et al. 2017). The co-occurrence of ESBL variants (blaCTX-M, blaTEM and blaSHV) identified in this study is in line with findings from the river Yamuna in Delhi-NCR . blaTEM-116 was the most prevalent type among blaTEM similar to earlier reports of environmental isolates of Enterobacteriaceae (Maravićet al. 2016). blaSHV was found to be the least prevalent type of ESBL. The blaSHV gene is mainly dominant in clinical isolates (Maravićet al. 2016); however, it has been reported in freshwater from Switzerland (Zurfluh et al. 2013). The test isolates predominantly include members of Enterobacteriaceae (70.5%) followed by Moraxellaceae (20.5%), Bacillaceae (4.4%) and others (4.5%, i.e., Aeromonadaceae, Brucellaceae and Pseudomonadaceae). Our data demonstrate a high percentage of E. coli similar to other samples of environmental origin followed by Acinetobacter spp., which is in line with other reports (Lu et al. 2010;Taco et al. 2014;Maravićet al. 2016). New reports from this study include the blaCTX-M gene in A. lwoffii, E. ludwigii, E. mexicanum and A. caviae. Furthermore, the identification of blaTEM and blaSHV in an environmental isolate of A. caviae is a new report from India.

Accession numbers of the analyzed nucleotide sequence
Results of our conjugation experiment revealed the successful transfer of ESBL genes from donor to recipient. The results of conjugation assay are in line with previous reports where movement of TEM, CTX-M and SHV genes were possible through HGT like the phenomenon of conjugation (Siddiqui et al. 2020). Journal of Water and Health Vol 19 No 4,601