Mass bathing events in River Kshipra, Central India- influence on the water quality and the antibiotic susceptibility pattern of commensal E.coli

Background Antibiotic resistance is one of the major global health emergencies. One potential source of dissemination of resistant bacteria is mass gatherings, e.g. mass bathing events. We evaluated the physicochemical parameters of water quality and the antibiotic resistance pattern in commensal Escherichia coli from river-water and river-sediment in pre-, during- and post-mass bathing events in river Kshipra, Central India. Method/Design Water and sediment samples were collected from three selected points during eight mass bathing events during 2014–2016. Water quality parameters (physical, chemical and microbiological) were analyzed using standard methods. In river water and sediment samples, antibiotic susceptibility patterns of isolated E. coli to 17 antibiotics were tested. Results pH, turbidity and dissolved oxygen were significantly lower and total dissolved solid, free carbon dioxide were higher during mass bathing, whilst TSS, BOD and COD were lowest in pre-bathing and highest in post-bathing period. E.coli with multi drug resistance (MDR) or extended spectrum beta-lactamase (ESBL) production were between 9–44% and 6–24%, respectively in river-water as well as river-sediment. Total coliform count/ml and E. coli count were higher during-and post-bathing in river water than in pre-bathing period. Thus, the percentage of resistance was significantly higher during and post-bathing period (p<.05) than in pre-bathing. Colony forming unit (CFU)/ml in river-sediment was much higher than in river-water. Percentage of resistance was significantly higher in river-water (p<.05) than in river-sediment. Conclusions Antibiotic resistance in E.coli isolated from the Kshipra River showed significant variation during mass bathing events. Guidelines and regulatory standards are needed to control environmental dissemination of resistant bacteria.

Introduction Clean, pure and safe water exists briefly in nature, as it is immediately polluted by prevailing environmental factors and anthropogenic activities. Industrialization and urbanization has polluted rivers mostly via agricultural runoffs and industrial effluents containing many used or unused antibiotics [1]. Antibiotics alter the ecology of the environment and generate antibiotic resistance [2]. It has been shown that the accumulation of antibiotics in the environment facilitate bacterial adaptation response to develop antibiotic resistant genes [3,4].
The organic pollution of river-water and at river banks is an increasing problem of worldwide concern [5][6][7]. Escherichia coli (E. coli), usually a commensal coliform of humans and animals, enters into the river mainly from livestock operations, their waste products and human septate [8]. These bacteria may accumulate in the river-sediment and might disseminate to distant sites with the antibiotic resistant genes acquired through horizontal transfer of genetic material. The presence of antibiotic resistant bacteria in freshwater sources has been documented throughout the world [4,9,10].
In India, river-water is considered sacred and mass bathing in some rivers is an age-old ritual. River Kshipra, one of the sacred rivers of India, is a source of domestic water supply for Ujjain City and around. Due to the religious importance of Kshipra River, mass gatherings including mass bathing of large or small scales occur throughout the year at its bank. Mass gathering is a pre-planned assemblage of more than 1000 persons at a particular place for a certain period [11]. During these mass gathering events, pilgrims from all over the country visit Ujjain for bathing in river. People take a 'holy-dip' as a part of religious ritual in this river in the form of mass bathing. They not only take 'holy-dip' in the river-water for bathe but also drink a handful of 'holy-water'. There is a high possibility of ingestion of antibiotic resistant bacteria when they bathe in the river and/or drink river water. Thus, the pilgrims are exposed to antibiotic resistant genes and disseminate it to the wider world.
Occurrence of antibiotic resistant bacteria has been reported from different rivers of India [12][13][14]. It is important to develop knowledge about the role and dynamicity of presence of antibiotic resistance in river-water during mass gatherings to design approaches to contain antibiotic resistance in rivers. In general, there is a paucity of information on the presence and concentration of antibiotic resistance in rivers associated with mass bathing in India. The influence of mass gathering and associated mass bathing on the antibiotic resistance pattern in bacteria in the water and sediment of river Kshipra, Central India, is not studied. Thus, our aim wasto compare and correlate the antibiotic susceptibility pattern and total burden of commensal E.coli from river-water and river-sediment in pre-, during-and post-mass bathing events in river Kshipra. We further aimed to evaluate the physicochemical parameters of water quality during these time events.

Study setting and sample collection
The present study is a part of an ongoing project, which has been described in detail previously [15]. In brief, the study samples were collected from River Kshipra, in its 93 km flow through Ujjain district, before, during and after mass-bathing events. Approximately 50,000-200,000 pilgrims take 'holy-dip' during any mass bathing event. Samples of river-water and river-sediment were collected from pre-identified and described sites (Fig 1) at pre-, during-, and postbathing time during eight mass bathing events that occurred between August -2014 to July 2016 (the sites were selected as they have the highest number of pilgrim visits for mass bathing).
The pre-bathing samples were collected one day prior of the bathing event, post-bathing samples were collected 6-7 hours after bathing event while during-bathing samples were collected at the peak hours of the bathing event, generally between 9 to 11AM. Each time, riverwater and -sediment were collected in sterile screw-capped bottles and conical falcon tube respectively as described in detail [15]. Approximately 2Kg of sediment and 3.5L river-water were collected for analysis of water quality parameters, colony counts and antibacterial susceptibility as described elsewhere [15]. All the collected samples were stored in icebox at 4-6˚C and transported to central research laboratory of R D Gardi Medical College, Ujjain within one hour.
Ethical permission for the study was obtained from the Institutional Ethics Committee of R D Gardi Medical College, Ujjain, MP, India (No 2013/07/17-311). No special permission was required for water and sediment samples as it was taken only from public land.

Physico-chemical and biological parameters
The physical and chemical parameters (ambient and water temperature, pH, conductivity, total dissolved solids (TDS) free carbon dioxide (free CO 2 ), carbonate alkalinity and dissolved oxygen (DO)) of collected river-water and river-sediment samples were analyzed in field [15]. After transport to the laboratory, samples were analyzed for turbidity, hardness, chloride, alkalinity, nitrate nitrogen available phosphorous, total suspended solids (TSS), biochemical oxygen demand (BOD), chemical oxygen demand (COD) and total phosphorous [15].

Microbiological and molecular methods
The samples were processed as described previously [15] for (a) Bacterial enumeration-Tenfold serial dilutions (1:100, 1:1000 as per turbidity of sample) of river-water were done with 0.9% normal saline. The river-sediment sample was first added to 100ml 0.9% normal saline and then ten-fold serial dilutions were performed. All diluted samples were processed following standard membrane filtration technique using 47mm diameter and pore size 0.45μm membrane filters. After filtration, the membranes were directly inoculated on selective and differential media HiCrome Coliform Agar for 24 hours at 37˚C for cultivation, isolation and identification of E. coli (blue-violet colony) and non E. coli isolates. The estimation of total coliform and total E. coli count in colony forming unit (CFU)/100ml was performed, (b) Identification of six E. coli isolates per river-water and -sediment sample was done followed by isolation and confirmation of these isolates by PCR and (c)The confirmed E. coli isolates were tested for antibiotic susceptibility to nine commonly used classes of antibiotics (cephalosporins; quinolones; nalidixc acid; tetracyclins; penicillins; carbapenems; aminoglycosides; sulpha drugs; polymyxins) by Kirby Bauer disc diffusion test on Muller Hinton (MH) Agar with the bacterial suspension of 0.5 McFarland turbidity. The zone diameter of bacterial growth inhibition was interpreted as per Clinical and Laboratory Standard Institute guidelines [16]. The results were interpreted as-number of resistant or susceptible isolates per antibiotic type per sample (out of six isolates), production of beta-lactamase (only where beta-lactamase production is indicated as a possible mechanism explaining observed resistance) by testing the combined disc diffusion method(isolates resistant to either ceftazidime or ceftriaxone (third generation cephalosporin), the presence of co-resistance (phenotypic resistance to two antibiotics of same or different group per isolate), multiple drug resistance (phenotypic resistance to more than two antibiotics of same or different group per isolate), and multi-drug resistance (MDR) (co-resistance involving three or more antibiotics of three different groups) in each isolate. E. coli reference strain ATCC 25922 was used for quality control. Intermediate resistant isolates were categorized as resistant.
The total bacterial DNA from E. coli isolates was extracted using the alkaline lysis method. The genetic confirmation of E. coli was done through PCR with genus-specific oligonucleotide primers [17]. The confirmed E.coli were tested for the presence of various genes (mcr-1, chuA, yjaA and TspE4C2) for detecting colistin resistance and phylogenetic grouping as mentioned in detail previously [10,18].

Data management and analysis
A pilot study was conducted prior to the main study to train research assistants for field and laboratory work and validate the sampling and analysis methods. All results were recorded in a single dataset using bar codes for unique identification and track of samples at the field and within the laboratories. All data were double-checked for quality assurance, cleaned and entered in to IBM SPSS Statistics 23.0 (SPSS Inc., Chicago, IL, USA). Results were noted for the variation in coliform load (in terms of CFU per 100ml) and antibiotic susceptibility in river-water and -sediment were analyzed and compared between different time-points during events and sites. A paired t test was applied for comparing the means of the water quality parameters between pre-, during-, and post-bathing sessions. A z-test for difference of proportion was applied for comparing the antibiotic susceptibility pattern to selected antibiotics in the pre-, during, and post-and the first, second, and third bathing sessions. A significant association determined by p-values < 0.05.

Study samples
Samples were collected from eight important mass bathing events during the study period as shown in Table 1. Total 144 river-water and 72 river-sediment samples were collected and totally 807/864 and 353/486 E. coli were finally isolated, tested and analyzed from river-water and river-sediment respectively. Six river-water (two pre-and four during-bathing) samples showed no growth of E.coli.

Physico-chemical properties of water
Various water quality parameters showed significant difference in during-and/or post-bathing samples as compared to pre-bathing samples ( Table 2). Ambient temperature, pH, turbidity and dissolved oxygen were significantly lower and total dissolved solid, free carbon dioxide were higher during-bathing whilst TSS, BOD and COD were lowest in pre-bathing and highest in post-bathing period. Similarly, total and organic phosphorus were highest in post-bathing samples.

CFUs of coliform and E. coli
The E. coli CFU/ml for the three sites during all events in river-water was between 0.09x10 3 -51x10 3 (mean 5.2x10 3 ) and in river-sediment 2.2x10 3 -9590x10 3 (mean 474x10 3 ) whiles the  mean total coliform CFU/ml for river-water and-sediment was 94x10 3 and 48.7x10 5 respectively. CFU during-and post-bathing time-point was higher as compared to pre-bathing in river-water and in river-sediment (Table 3). E. coli CFU/ml was higher in river-water duringbathing which took place in monsoon and summer, while CFU was extremely high (about 200 to 300 times) in river-sediment for events in summer season.

Antibiotic susceptibility to various antibiotic groups
The antibiotic resistance in E. coli from river-water and river-sediment for different antibiotics is shown in Fig 2. The resistance pattern to various antibiotics was similar in E.coli sampled from river-water and river-sediment though the percentage of resistant bacteria was significantly higher in river-water (p<.05) than in river-sediment. However, considering the total number of coliform in CFU/ml in river-sediment, the total amount of resistance was significantly higher (p = .04) in river-sediment than in river-water. Resistance to polymyxin (colistin) group was not detected in any of the samples. Resistance to penicillin (27-40%) and cephalosporin (14-31%) was highest while it was lowest for tigecycline (0.3-4%) and gentamycin (1-3%). The MDR combinations having cephalosporin+quinolones+ penicllin and cephalosporin+quinolones + sulfonamides groups of drugs were more common than the cephalosporin+quinolone+aminoglycosides or carbapenem combinations. Most of the isolates from all the sources showed resistance simultaneously to ceftazidime, cefotaxime, cefapime, ampicillin, tetracycline, and co-trimoxazole.

Resistance of E. coli isolates between time-points of an event
The mean percentage of resistant E.coli was high in river-water for most commonly used antibiotics such as ampicillin (p = .04), ceftazidime (p = 0.00), cefotaxime (p = 0.02), cefepime, nalidixic acid, ciprofloxacin, and nitrofurantoin during-and post-bathing while the percentage resistance to rarely used antibiotics in Indian clinical settings such as gentamicin, amikacin, tetracycline, sulfamethizole, imipenem (p = 0.00) and meropenem (p = 0.03)was more in preevent period in both river-water and river-sediment (Table 4). E. coli isolates from both river-water and -sediment belonged to phylogenetic groups A (69%), B1 (14%), B2 (3%), and D (14%).

Discussion
The physio-chemical and microbiological quality of river Kshipra deteriorated during-and after mass bathing events. Very high total coliform count/ml and E. coli CFU/ml was seen during-and post-bathing in river-water as compared to pre-bathing period. E. coli CFU/ml in river-sediment was much higher than in river-water. The antibiotic resistance was high with many MDR (9-44%) and ESBL producing (6-24%) E. coli in river-water as well as in riversediment. High total dissolved solid and free carbon dioxide were found during-bathing while high TSS, total and organic phosphorus, BOD and COD were detected in post-bathing period in river Kshipra. Thus, water quality parameters of river water deteriorate and do not remain fit to be used for drinking or household purposes [19][20][21][22]. The changes could be attributed to the use of soap, offering milk, flowers, oil etc. in river-water, dirt, and sweat of bathers', resuspension of sand and clay particles due to the discharge of wastes. Further, the highest worsening of water quality was also found to be correlated with the number of pilgrims that took bath during an event and most populated sites on the river bank [21].
Presence of indicator bacteria in water suggests poor water quality because of faecal contamination. Studies have reported that apart of release of domestic waste, leaching from septic tanks and farm wastes and mass bathing causes significant increase in faecal coliform load [19,23,24]. We found high number of E.coli during-and post-bathing in river-water and in riversediment (Table 3). We interestingly, found that CFU/ml of total coliforms in river-sediment is either same or more in pre-bathing period than during-and post-bathing which is in contrast to the finding for river-water. The increase in bacteria count in river-water during-and post-bathing thus could be attributed to mixing of river-sediment bacteria with the upper layer due to constant and heavy movement of water by movement of number of people during mass bathing. E.coli concentration of river-sediment was found to be re-suspended contributing to pollution of river-water [23]. Matson et al (1978) found mean faecal coliform counts in sediment to be 2,500 times greater than in overlying water [24]. It is therefore, suggested that microbial studies of bottom sediment be considered a part of water quality evaluations.
We also reported high number of E.coli in river-water and in river-sediment in events that took place in monsoon and summer. Bacterial count in river-water and -sediment varies with seasons being highest in summer [25,26]. Gonzales et al also reported seasonal variation in coliform bacteria displaying an increased prevalence during the rainy season [27]. The levels of indicator bacteria is a sign of level of contamination and thus of unhealthy water with high potential of health risk [28].
Antibiotic resistant and MDR coliforms were between 20-38% in isolates from river-water and river-sediment which raise concern regarding risk of acquiring multiple antibiotic resistant genes from non-clinical settings via horizontal transfer to a wide range of bacterial species [3,4,9,10,29,30]. Considering the number of E.coli (about 200 to 300 times higher) in riversediment during summer season, the possibility of transfer of resistance genes may be alarming. Thus, river-sediment is appropriately considered as a natural reservoir of resistance genes [31]. Bacteria with highest level of resistance in river-sediment are the main reservoir of antibiotics and antibiotic resistance in environment [31,29]. Our results, also showed a high percentage of commensal E.coli.
We noticed that commensal coliform are co-resistant to quinolones and cephalosporin which are mostly used in clinical settings, suggesting that environmental and clinical pathogens may have a common origin. The frequent use of quinolones group of antibiotics for treatment of enteric infections caused by gram-negative bacteria, could explain the high (25-40%), resistance of detected MDR E.coli against quinolones. We have previously shown that the frequency of resistance, co-resistance and resistance genes are high and similar in coliforms from humans and their environment [26,32]. Also, as previously seen the ESBL-producing E. coli isolates from river-water and -sediment belonged to phylogenetic groups A and B1 to which commensal E. coli are classified [26,32,33]. Thus, the emergence of MDR in E. coli and the other commensal coliform may also pose a risk for the emergence of new multi resistant pathogen.
The study has some limitations such as colistin resistance should have been performed and confirmed by MIC method and not by disk diffusion method. We, however, simultaneously performed mcr-1 gene detection by PCR. The disk diffusion method for AST is considered as a robust procedure though in some (5%) cases double and blurred zones were detected which were classified after repeat inoculation and confirmation by PCR. In addition, the zones of inhibition were checked by random blind assessment by another technologist. We, thus believe, that our findings of antibiotic resistance in river environment in India within a typical socio-behavioral context of religious mass bathing implies many important issues. Firstly, river acts as a medium for mixing pathogen and non-pathogenic bacteria from many sources like human, animal and environment; secondly, pathogenic bacteria may exchange antibiotic resistance genes with commensal; lastly, river-water acts as a natural reservoir and source of resistance genes for emerging pathogens. As during mass bathing in river Kshipra people from various parts of India, not only take a holy dip in the river but also use it for drinking, though high prevalence of coliform and the antibiotic resistance is an ideal setting for the acquisition and spread of antibiotic resistance genes and bacteria to distant places.