Multiplex PCR based genotypic characterization of pathogenic vancomycin resistant Enterococcus faecalis recovered from an Indian river along a city landscape

Background Enterococci are normal commensals of human gut, but vancomycin-resistant enterococci (VRE) are a severe threat to human health. Antimicrobial-resistant enterococci have been reported previously from Indian surface waters. However, the presence of antimicrobial resistance and virulence markers in Enterococcus faecalis, the most dominant enterococci is yet to be investigated. Objectives The goal of this study was to analyse concentration of enterococci and distribution of antimicrobial resistance and virulence markers in E. faecalis isolates from river waters along an important north Indian city landscape. Methods We enumerated enterococci in river water samples (n = 60) collected from five sites across the Lucknow city landscape using the most probable number and membrane-filtration methods. The antimicrobial sensitivity profile of E. faecalis isolate was generated with the Kirby-Bauer antimicrobial disc diffusion assay. The multiplex PCR was used for genotypic characterization of vancomycin-resistance and virulence in E. faecalis isolates. Results Enterococci density (p < 0.0001) increased from up-to-down-stream sites. Multiplex PCR based genotypic characterization has shown a significant distribution of virulence-markers gelE, ace or efaA in the E. faecalis isolates (p < 0.05). The range of antimicrobial-resistance varied from 5 to 12 in the landscape with the frequency of vancomycin-resistant E. faecalis (VRE) ranging from 22 to 100 %. Conclusion The occurrence of pathogenic VRE in river Gomti surface water is an important health concern. The observed high background pool of resistance and virulence in E. faecalis in river waters has the potential to disseminate more alarming antimicrobial resistance in the environment and poses serious health risk in developing countries like India as VRE infections could lead to increased cost of healthcare. Electronic supplementary material The online version of this article (doi:10.1186/s40064-016-2870-5) contains supplementary material, which is available to authorized users.


Background
Enterococci is considered to be a suitable 'indicator' of fecal contamination in aquatic environment as it survives longer compared to other fecal streptococci and coliforms (Cabelli et al. 1982;USEPA 2003). Enterococcus faecalis, most prevalent enterococci, is an enteric bacterium of mammalian intestinal tract. It is notorious as an opportunistic pathogen causing various infections (Thurlow et al. 2010;Parsek and Singh 2003). Majority of enterococcal infections (~80 %) are caused by E. faecalis due to its ability to acquire virulence traits and resistance to multiple antimicrobials (Gordon et al. 1992;Huycke et al. 1998;Ghoshal et al. 2006). The genome sequence of a clinical E. faecalis isolate V583 has revealed that more than a quarter of its genome consists of mobile or foreign DNA, including genes for vancomycin resistance and virulence markers (Paulsen et al. 2003). Vancomycin, once considered as the most powerful antimicrobial of "last resort, " is no longer efficacious in managing antimicrobial-resistant enterococci (Arias and Murray 2012). Further, the incidence of E. faecalis exhibiting antimicrobial-resistance viz. vancomycin resistance and virulence markers such as gelatinase (gelE), collagen binding protein (ace), endocarditis-associated antigen (efaA) and enterococcal surface protein (esp), have been explored mostly in clinical settings worldwide (Coque et al. 2002;Lowe et al. 1995;Eaton and Gasson 2001;Ghoshal et al. 2006;Mathur et al. 2003). However, very little is known about the distribution of antimicrobial-resistance and virulence-markers in E. faecalis isolates recovered from river waters (Novais et al. 2005;Moore et al. 2008;Lata et al. 2009b). River waters in developing countries are vulnerable to contamination due to discharge of untreated or partially treated municipal and industrial wastes, domestic and wild animal fecal waste and recreational activities on the banks (Ahmed et al. 2005;Ram et al. 2007). Such uninterrupted contamination of rivers lead to a bigger problem of nutrient enrichment or eutrophication of river waters. The eutrophication further leads to growth of aquatic macrophytes assisting the proliferation and prolonged survival of pathogens increasing public health risk (Whitman et al. 2003;Lata et al. 2009b).
River Gomti, a tributary of the river Ganga meets the major water requirement of more than 3.5 million populations of Lucknow city (Fig. 1). This river receives a discharge of 450 mLd untreated waste through different point sources due to the rapid industrialization and urbanization in the region (Singh et al. 2004). In India, river waters are used for drinking, bathing and carrying out various household chores. This demands determination of microbiological quality of river waters for remediation to protect public health. Reported prevalence of diarrheogenic and enterotoxigenic E. coli in the river waters of river Ganga and its tributary river Gomti has emphasized the necessity of information on recreational water quality 'indicator' , E. faecalis in such riverine systems (Ram et al. 2007(Ram et al. , 2008Singh et al. 2010). This study investigates the prevalence and dissemination of vancomycin-resistant and pathogenic E. faecalis along the up-to-downstream landscape of Lucknow city (Fig. 1). The genotypic characterization is done by multiplex PCR assay panels comprising simultaneous amplification of four genes: sodA, vanB, gelE and either ace or efaA. In addition, the prevalence of esp virulence marker and multiple-antimicrobial-resistance was also examined amongst E. faecalis isolates from the river Gomti water.

Concentration of enterococci
Median concentrations of enterococci increased gradually and significantly (p < 0.0001) in the river Gomti waters from up-to-down-stream sites in the landscape. The most downstream site was found to be 123 and > 4500-fold more polluted than the most upstream site, using MPN test and membrane filtration method, respectively (Table 1; Additional file 1: Table S1). Concentrations of enterococci at downstream sites were about 10, 23 and 69-fold higher at site 2, 3 and 4, respectively when compared to the most upstream site, using MPN test (Additional file 1: Table S1). At the same time, median concentration of enterococci using membrane filtration method was 45, 296, 512-fold higher at site 2, 3 and 4, respectively (Additional file 1: Table S1).

Multiplex PCR for vancomycin resistant pathogenic E. faecalis
The multiplex PCR based genotypic characterization was carried out to determine the prevalence of vanB resistance and gelE, ace or efaA virulence markers in E. faecalis isolates from river Gomti (Tables 2 and 3; Fig. 2). The genotypic characterization has shown a significant distribution of virulence-markers along the landscape (p < 0.05). The median count investigated throughout the landscape was 3 (Table 4). According to genotypic characterization the frequency of vanB resistant E. faecalis is 27 % in the landscape. However, 100 % E. faecalis isolates recovered from river Gomti exhibit prevalence of gelE (gelatinase) and esp virulence markers. The esp gene was characterized as described earlier (Lata et al. 2009a). The ace gene was observed in ca. 58 % E. faecalis and the frequency of efaA virulence marker was ca. 97 % in the landscape. Further the frequency of various combinations of multiple-virulence-markers has been observed in the landscape (Table 3).

Gelatinase exhibiting E. faecalis phenotypes
The phenotypic assay for determination of gelatinase activity shows high prevalence of gelatinase positive E. faecalis isolates in river Gomti throughout the landscape. Site 2 exhibits highest prevalence (80 %) of such isolates followed by the most downstream site 5 (67 %) and then most upstream site 1 (60 %) and 57 % for both site 3 and 4 (Table 3).

Antimicrobial-resistance in E. faecalis isolates
A high frequency of multiple-antimicrobial-resistance was recorded among E. faecalis tested. The range for the number of antimicrobials against which resistance was observed in each E. faecalis isolates increased significantly (p < 0.05) towards downstream in the landscape (Tables 5, 6). The median of antimicrobial-resistance per isolate was nine at Gau-Ghat, the most upstream site and increased to a value of eleven at site 3 and subsequently declined to nine at site 4 and 5. In general, E. faecalis isolates were resistant to at least five antimicrobials throughout the landscape. However, the range of antimicrobial-resistance varied from 5 to 12 in the landscape (

Discussion
This study was carried out to investigate the landscape dependent density and distribution of enterococci, prevalence of antimicrobial-resistance and virulence-markers among E. faecalis isolates recovered from river Gomti at Lucknow city. A multiplex PCR based genotypic characterization of vancomycin resistant pathogenic E. faecalis was carried out. The river Gomti exhibits high enterococci density at selected sites. The MPN index/100 mL values of enterococci recorded at all the sampling stations exceeded the standards set by regulatory authorities for river water reservoirs used for drinking and recreational purposes (US EPA 2003). The quantitative enumeration of enterococci by membrane filtration method has also shown a gradual increase in the concentration  Table 1 Quantitative enumeration (log transformed) of enterococci collected from sites (n = 5) located on river Gomti in up-to-down-stream fashion ND not detected *** Statistically significant at alpha < 0.05 a Lower 95 % CI-upper 95 % CI is adopted from of enterococci towards the downstream sites in the landscape. These sites are highly polluted because of the recurring fecal pollution from nearby residential areas and cremation practices at local crematorium. Population pressure alongwith poor sanitation facilities might be responsible for the increased bacterial contamination in river waters of developing countries (Ram et al. 2007(Ram et al. , 2008Lata et al. 2009b). In the present study, an increasing trend of enterococci concentration was observed in the range of 1.5-1.6 × 10 2 CFU/100 mL, 6.8-6.9 × 10 3 CFU/100 mL, 4.4-4.6 × 10 4 CFU/100 mL, 7.7-7.8 × 10 4 CFU/100 mL and 6.8-6.9 × 10 5 CFU/100 mL at sites 1, 2, 3, 4 and 5 respectively. The most downstream site in the landscape, Gandhi setu receives contamination from city through two major wastewater point sources, including Kukrail nala and city's largest crematorium located just upstream. Internationally, the single-sample advisory limit of enterococci for fresh water is 61 CFU/100 mL, and the 5-day geometric mean should not exceed 33 CFU/100 mL, while Indian standards have not defined the limit for enterococci in terms of CFU/100 mL (CPCB 2002;USEPA 2003). A panel of multiplex PCR assay was developed in the current study for detection of vancomycin-resistant pathogenic E. faecalis in environmental waters (Table 2; Fig. 2). The frequency of E. faecalis isolates exhibiting multiplevirulence-markers and vanB genotype is 27 % in the landscape. However, 100 % E. faecalis isolates recovered from river Gomti exhibit prevalence of gelE and esp virulence markers. Gelatinase, a virulence marker widely reported in enterococci, hydrolyzes a broad range of bioactive substances and has been implicated in various infection models through biofilm formation (Parsek and Singh 2003;Nallapareddy et al. 2006). A recent study has shown the prevalence of gelatinase phenotype of enterococci with biofilm formation potential in agricultural environment and implicated it as a reservoir of clinically relevant strains (Macovei et al. 2009). The most frequent multiplevirulence-markers were gelE + esp + efaA + and gelE + esp + efa A + ace + . Further the frequency of various combinations of multiple-antimicrobial-resistance and multiple-virulencemarkers has been observed in the landscape. Among all, 39 % isolates were characterized with gelE + esp + efaA + virulence markers and vancomycin resistance phenotype whereas 45 % were having methicillin resistance, VRE phenotype and gelE + esp + efaA + ace + multiple-virulencemarkers. Additionally 21 % of gelE + esp + efaA + ace + isolates were observed to exhibit resistance to aminoglycoside-βlactam-vancomycin-teicoplanin (G-Lactam-Va-Te); see Fig. 3 for details. The high frequency of VRE in the landscape, its association with other widely disseminated antimicrobials and virulence markers may lead to evolution of potential pathogenic VRE (Cetinkaya et al. 2000;Arias and Murray 2012). Additionally, 29 % isolates exhibit resistance to a macrolide, erythromycin, and rifampicin in association with tetracycline. A study from Denmark has reported carriage of coupled resistance of copper and antimicrobials (macrolide and glycopeptide) in pigborne enterococci isolates contemplating persistence of the background pool of antimicrobial-resistance (BPAR) in that geographic region (Hasman and Aarestrup 2005). A set of other studies have reported the phenomenon of sustained BPAR in poultry and local population and its environmental carriage by plasmid maintenance systems (Sletvold et al. 2007b;Garcia-Migura et al. 2008). Moreover, previous study has implicated the possible transfer of  (2001) linked virulence-traits and antimicrobial-resistance viz. vancomycin resistance (Heaton et al. 1996). The persistence of VRE in the environment even in the absence of antimicrobial selection pressure has been attributed to multiple types of plasmid maintenance systems that might play a pivotal role in persistence and dissemination of The highest levels of multiple-antimicrobial-resistant E. faecalis were isolated from site 3 and 5 followed by sites 2, 4 and 1. Prevalence of such a high background pool of multiple-antimicrobial-resistance may be due to the lack of dilution of the contamination that is added to the river via various point and non-point sources. The river Gomti exhibits sluggish flow except for the monsoon period in the landscape. Although the specific sources of contaminants were not definitively determined, it is likely that the contaminants from nearby slums, cattle herds, poultries and the hospitals contributed to this contamination. Twenty-five city drains in the Lucknow area drain into the river Gomti (CPCB 2005). Heavy silting of the river reduces its carrying capacity. However, agricultural practices, poultry and dairy farming along site 1, the most upstream site might be a contributory factor for high prevalence of multiple-antimicrobial-resistance in the E. faecalis isolates recovered from this site. Although there is no data available from India in this context, the feed  The VRE, both VanA and VanB phenotypes were ubiquitously present throughout the landscape. The increased frequency (100 %) of VRE at site 2 may be due to addition of contamination from the hospital located just upstream. The river Gomti identified as one of the most polluted rivers in India, is the lifeline of Lucknow city (population size ~3.5 million) as it meets about half of the total domestic water demand (about 550 mLd) of the city (Singh et al. 2004). In India, large population depends on untreated river water for carrying out daily needs (Hamner et al. 2006;Ram et al. 2007). The prevalence of vancomycin-resistant pathogenic E. faecalis isolates in river Gomti is a threat to public health as superbugs like VRE are among the most critical human health challenges of this century (Pruden et al. 2006). This study reports vancomycin-resistant pathogenic E. faecalis in river Gomti waters along an up to downstream city landscape. The emerging VRE and high background pool of multiple-antimicrobial-resistant and multi-virulent E. faecalis can contribute to the dissemination of    resistance and virulence-markers in the diverse enterococci and other bacteria. Therefore, the presence of antimicrobial-resistant pathogenic E. faecalis in river waters of developing nations like India demands immediate environmental intelligence and surveillance for risk assessment and prevention strategies for protection of public health.

Conclusion
The occurrence of pathogenic VRE in river Gomti surface water is an important health concern. The observed high background pool of antimicrobial resistance and virulence in E. faecalis in river waters has the potential to disseminate more alarming antimicrobial resistance in the environment and poses serious health risk in developing countries like India as VRE gut inhabitation could lead to increased economic burden for healthcare.

Study site and Sample collection
The study was conducted along 10 km stretch of river Gomti in Lucknow city (Latitude: 25.55′N, Longitude: 80.58′E and Altitude: 123 m). River Gomti, flowing through the northern part of the country, is a major tributary of the Ganga River in India. Originating from a natural reservoir in the swampy and densely forested area near Madho-Tanda (altitude 200 m; latitude 28°34′N and E longitude 80°07′E) in the foothills of Himalayas, the river traverses a distance of about 730 km in the Indo-Gangetic alluvial region before its confluence with river Ganga. The river flowing through the Lucknow city (population: ~3.5 million), the State capital of Uttar Pradesh, meets about half of the total domestic water demand (about 550 mLd) of the town. The average flow of the river at Lucknow varies between 500 mLd in summers and 55,000 mLd during the monsoons. The river contains water throughout the year and exhibits sluggish flow except for the monsoon period. The Gomti River, identified as one of the most polluted tributary of river Ganga, receives 450 mLd of untreated domestic waste water in Lucknow city (Singh et al. 2004). In this study, five sites located in up-to-downstream on the river, namely Gaughat (site 1: located in upstream of river), Daliganj Bridge (site 2: bathing ghat and holy spot), Shaheed Smarak (site 3: bathing ghat and picnic spot), Lakshaman mela ground (site 4: recreational and picnic spots), Gandhi setu (site 5, most downstream location in the landscape) were selected based on various human activities on the banks of river Gomti. A cross-sectional approach was used to collect river water samples. Samples were collected in triplicate (n = 15) from five locations situated in up-todown-stream fashion (Fig. 1). In brief, three transects were established randomly at each site and water samples (1 L) were collected 30 cm below the water surface from left, mid and right banks of the river along each transect. The samples of river water were stored in sterile glass bottles, labeled and transported on ice to the laboratory for analysis. Sample processing and analysis were conducted within 6 h after sample collection.

Isolation and enumeration of Enterococci
Quantitative enumeration of fecal streptococci or enterococci at selected sites was performed using the standard multiple tube fermentation technique (APHA 1998).
In brief, tubes containing 20 mL Azide Dextrose Broth (ADB) were prepared. A series of tubes of ADB was inoculated with appropriate graduated quantities of sample (10, 1 and 0.1 mL). The inoculated tubes were incubated at 37 ± 0.5 °C and examined for turbidity after 24-48 h and calculated MPN index/100 mL surface water depending upon the number of positive tubes. Enterococci were recovered from each sample using standard membrane filtration method (APHA 1998). Briefly, tenfold dilutions of each water sample was prepared (10 0 , 10 −1 , 10 −2 and 10 −3 ), and 10 mL of each dilution were filtered through 0.45-µm, 47-mm mixed cellulose ester filters (Millipore, Bedford, MA, USA) The filters were then placed on mE agar plates incubated at 41 ± 0.5 °C for 48 h for detection and enumeration of enterococci (Hi-Media, India). After 48 h, membrane filters from mE agar plates were placed on Bile Esculin Azide (BEA) agar plates and incubated at 41 ± 0.5 °C for 20 min. Colonies characteristic of enterococci, ranging from pink to dark red on mE agar and producing a brown to black precipitate on BEA agar, were considered presumptive enterococci (APHA 1998). Negative control filters and agar plates were included in each membrane filtration analysis. All colonies were counted, and concentration of enterococci per 100 mL water was determined from dilution plates containing 30-300 CFU. Presumptive enterococci recovered (n = 60) from each sample were identified by biochemical tests, including catalase test and PYR test. The growth of isolates was determined in 6.5 % NaCl, pH 9.6, and at 10°s and 45 °C, respectively. All confirmed enterococci isolates were archived in tryptic soy broth with 15 % (v/v) glycerol at -80 °C for further analyses.

Multiplex PCR assay
All enterococci isolates confirmed by biochemical tests were subjected to genotypic characterization by multiplex Polymerase Chain Reaction (PCR) technique. The multiplex PCR assay format was designed comprising four genes; either ace or efaA gene with sodA (specific for detection of E. faecalis), vanB and gelE gene. All the primers used in this assay have been reported earlier in singleplex PCR format for detection of respective genes in clinical settings (Table 2; supplementary information). Genomic DNA was extracted from overnight grown cells of enterococci isolates (n = 60) using Genelute bacterial genomic DNA kit (Sigma, USA). PCR assay was performed in reaction volume of 50 µl. In brief, The mastermix consisted of 1× hot start Taq buffer, 3.0 mM MgCl 2 , 0.4 mM deoxynucleotide triphosphate mix, 1.5 U of Hotstart Taq polymerase (MBI Fermentas), 300 pg of template DNA, sterile molecular grade water, forward and reverse primer (Metabion, Germany) concentration optimized for this study has been described in (

Gelatinase phenotype determination
Gelatinase production was detected by spotting midlog phase grown pure culture of individual E. faecalis isolates (n = 33) on freshly prepared tryptic soy agar plates (Himedia, India) containing 1.5 % of skimmed milk. Plates were incubated overnight at 37 °C and then cooled to ambient temperature for 2 h. The appearance of a transparent halo around the colonies was considered to be a positive indication of gelatinase production (Gilmore et al. 2002).

VanB phenotyping
Antimicrobial Vancomycin (Van) was used for this study (Sigma-aldrich, USA). The E. faecalis isolates from each sampling site were subjected to tube macrodilution method of antimicrobial susceptibility testing (Van: 4-1024 µg/mL) with slight modification. The tube macrodilution method CLSI (2005) for antimicrobial susceptibility testing (Van: 4-1024 µg/mL) was adopted with slight modification. In brief, well characterized E. faecalis isolates were grown in tubes containing 5 mL Brain Heart Infusion broth and incubated in rotary shaker (INNOVA 4230,New Brunswick,USA) at 150 rpm and 37 + 1 °C for 4-6 h to yield a suspension of 10 6 cells/mL. The inoculum of 10 6 cells was added to sterile Muller-Hinton broth tubes containing Van: 4-1024 µg/mL in twofold dilution series. The tubes were incubated immediately at 35 + 1 °C for 24 h. The tubes showing consistent growth were recorded as VanB positive. E. faecalis ATCC 51299 served as VanB positive and sterile water as negative control.

Statistical analyses
The concentrations of enterococci obtained using MPN analysis test and membrane filtration method from upto-downstream river water samples was compared, Chi square test for trend was applied for the purpose. The prevalence and distribution of antimicrobial-resistance and virulence-markers among isolates from up-to-downstream landscape was assessed using wilcoxon rank-sum tests. All statistical tests were performed using GraphPad Prism version 6.0 for Windows (GraphPad Software, San Diego, CA, USA, www.graphpad.com).

Authors' contributions
PL carried out the phenotypic antimicrobial susceptibility profiling, determination of gelatinase activity, multiplex PCR based characterization for virulence markers and vancomycin resistance in E. faecalis isolates, analysed the data, wrote the manuscript and made illustration for study site. SR contributed to sample collection and microbiological analysis. RS conceived the study, analysed the data and wrote the manuscript. All authors contributed to the general content and structure of the final manuscript.