Genetic diversity of Escherichia coli isolates from surface water and groundwater in a rural environment

The water quality is strongly influenced by urbanization, livestock and agricultural activity in agroecosystems. Increasing public concern over access to clean water has led to improve water source protection and management activities. Rural environments present unique challenges because they serve as water sources for a variety of activities (e.g. drinking water for humans and livestock, irrigation and recreation). Furthermore, both point and non-point sources of microbial pollution may be found (e.g. livestock operations, household septic systems and wildlife) (Chin 2006; Sinclaira et al. 2009).

Identifying sources of fecal pollution in waters is necessary to reduce the potential for human contact with enteric pathogens. Bacterial indicator organisms are commonly used to assess the presence of enteric pathogens in the water. The most common indicator microorganisms are fecal coliforms (FC) and its subgroup Escherichia coli. These microorganisms share a common feature: they all can inhabit the intestines of warm-blooded animals, including wildlife, livestock and humans, and therefore they can be excreted in the feces (Ibekwe et al. 2011; Carlos et al. 2012). However, E. coli presence in contaminated water will not provide information as to the actual source of contamination. This information is important, as fecal contamination resulting from human sources will establish a high public health risk because of the possible presence of pathogenic organisms. Additionally, if the fecal source is known, suitable management actions can be implemented to prevent further contamination and to mitigate the health risks (Harwood et al. 2000).

While E. coli has diverse genotypic and phenotypic characteristics, some are shared among strains exposed to similar environments due to selection pressure. If some of the characteristics among E. coli strains can be grouped by origin of isolation, then it is possible to use these genotypes as a tool to determine the source of unknown bacteria. The level of selective pressure exerted in environment may be a useful criterion for identifying the fecal contamination sources in water (Unno et al. 2010; Ibekwe et al. 2011).

Methods to determine sources of fecal pollution have usually included phenotypic and genetic characterization of fecal indicator bacteria isolate from surface waters. The study of the genetic diversity of E. coli isolates from water may inform on the diverse origins of contaminant sources. DNA fingerprinting techniques, such as pulsed-field gel electrophoresis (PFGE), ribotyping and repetitive element palindromic polymerase chain reaction (Rep-PCR), are reported to be an effective way to track the source of microbial pollution without many of the complications of antibiotic resistance analysis. (McLellan 2004; Mohapatra et al. 2007; Bonnet et al. 2009; Cuevas López et al. 2009; Lyautey et al. 2010; Unno et al. 2010). The Rep-PCR method uses naturally occurring conserved intergenic palindromic sequences for amplification by PCR. The PCR amplification uses primers, including the BOX A1R primer, complimentary to the palindromic sequences to produce unique DNA banding patterns (fingerprints) that are strain-specific. The Rep-PCR provides high taxonomic resolution and may act as a rapid detector of diversity and evolution of the microbial genomes being studied (Versalovic et al. 1994; Dombek et al. 2000).

Many investigators have evaluated the use of Rep-PCR method to determine sources of environmental fecal bacteria in waters. Therefore, Dombek et al. (2000) reported the use of Rep-PCR DNA fingerprint to differentiate E. coli strains isolated from human and animal sources. Somarelli et al. (2007) reported molecular fingerprinting (BOX-PCR) of fecal microorganism isolated from the Conesus lake. Ma et al. (2011) used BOX-PCR to differentiate sources of environmental E. coli and Carlos et al. (2012) to discriminate E. coli strains from animal and waste sources. The investigators showed that this technique has greater discriminatory power. Studies on the genetic characterization and diversity of E.coli isolated from groundwater in order to assess the fecal contamination source are not fully developed in general. To our knowledge, our investigation group has made the first studies in Argentina using BOX-PCR DNA fingerprinting to evaluate genetic diversity of E.coli isolated groundwater in order to assess fecal contamination source. Gambero et al. (2016) have reported genetic and antibiotic studies of E. coli isolated from a sedimentary aquifer in the rural environment in Cordoba, Argentina.

In this framework, the aim of this study was to define water aptitude for human consumption considering general bacteriological characteristics and to elucidate the diversity of E. coli water populations, identifying the main sources of fecal contamination and exploring linkages with the hydrogeological environment and land uses.

The surface water and groundwater salinity values are between 508–2,050 μS/cm and 654–1,078 μS/cm, respectively (Figure 2). The water geochemical features are variable, from calcium–sodium bicarbonate to sodium sulphate water types, which depend on the geochemical processes described for this aquifer in Giuliano Albo & Blarasin (2014).

Environmental water samples (n = 41) were collected in the wet season (spring). Thirty-six samples were collected from groundwater and five from surface water. To ensure the representativeness of groundwater samples, each well was emptied three times. The water samples (500 mL) were taken for bacteriological indicator analysis according to Argentine Food Code (AFC 2012). All samples were collected aseptically in sterile bottles, stored at 4°C and analyzed within 24 h of collection in the laboratory of the National University of Rio Cuarto (Argentina).

The determination of Heterotrophic Plate Counts (HPC) was carried in nutrient agar, incubated at 35°C during 24 h. The total coliforms (TC) and FC were determined through the multiple-tube fermentation (MTF) technique. Probability tables (McCrady tables) were used to determine the Most Probable Number (MPN) and to estimate the number of coliform organisms per 100 ml of water. TC were incubated in MacConkey broth at 35°C during 24–48 h and FC in Brilla broth at 44.5°C during 24 h. The presence of E. coli was determined in 100 ml of sample in MacConkey broth incubated at 35°C during 24–48 h. Then, an aliquot was spread onto EMB agar plates and incubated at 35°C for 24 h. Isolates were confirmed as E. coli by using a series of biochemical tests, including indole, Voges-Proskauer, methyl red tests and the inability to grow on citrate agar (IMViC). The presence of Pseudomonas aeruginosa was determined on a volume of 100 ml of sample in Crystal Violet broth incubated at 35°C during 24–48 h. The isolation was carried out in Cetrimide agar plates and colonies were confirmed by the following biochemical tests: oxidase, growth at 42°C, and pigments production in Agar P and F. The methodology was carried out according to Standard Methods for the Examination of Water and Wastewater (APHA 2005).

Genomic DNA from individual pure cultures of E. coli isolates was extracted by the use of Real Genomic DNA Extraction kit (Real Biotech Corporation, RBC), according to the manufacturer indications. The extracted DNA was used as the template DNA for Rep-PCR fingerprinting. The Rep-PCR assay mixture was prepared as described by Rademaker & Bruijn (1996) modified. The final reaction mix (25 μl) consisted of 1X PCR buffer (Promega, Biodynamics), 1.5 mM MgCl₂, 0.2 mM of each deoxynucleoside triphosphate (Promega, Biodynamics), 10 pmol/μl of the BOX A1R primer (5′-CTACGGCAAGGCGACGCTGACG-3′) (BOX-PCR), 0.04 U/μl of Taq polymerase (Promega) and 3 μl of dilution template DNA.

Amplification was performed in a MJ Research thermocycler as follows: after an initial denaturation at 94°C for 5 min, 35 cycles of denaturation (94°C, 1 min), annealing (40°C, 2 min) and extension (72°C, 2 min) were performed, followed by a final extension (72°C, 10 min). A negative (distilled water) control was included in the PCR experiments. The PCR products were separated by horizontal gel electrophoresis on a 1.5% agarose gel. Gels were stained in 0.5 μg/ml ethidium bromide solution and the gel images were taken in a MiniBisPRO in TIF Format. The Rep-PCR fingerprints of the isolates were normalized with a 1 kb DNA ladder.

The Rep-PCR fingerprints were subjected to cluster analysis using the SPSS, v.11.5. software. This analysis was also used to evaluate the similarity of the strains. In this program, the obtained results were converted to a two dimensional rectangular matrix data of binary codes for all E. coli strains, [+] for presence and [–] for absence of the PCR product band. The relationship between the fingerprints was estimated by the proportion of shared bands using the Dice coefficient and dendrograms were generated using the UPGMA method (Unweighted Pair Group Method with Arithmetic-Averages), with the NTSYS v. 2.1 software. (Dombek et al. 2000; Baldy-Chudzik & Stosik .2005; Somarelli et al. 2007).

The results of bacteriological analysis of groundwater samples are shown in Table 1. It was seen that 53 per cent of samples were unfit for human consumption taking into account all the general bacteriological indicators. FC and E. coli were detected as evidence of fecal contamination. Thus, seven wells were positive for FC (range, 4–430 MPN/100 ml) and eight wells were positive for E. coli. B17, B31 and B43 wells were greatest affected by the bacteriological contamination. The results of bacteriology analysis of surface water samples are shown in Table 2. It was detected that three samples were positive for FC (range, 75–2,100 MPN/100 ml) and all the samples were positive for E. coli.

From the 41 water samples extracted, 17 strains were confirmed as E. coli by biochemical tests, from which eight were isolated from groundwater and nine from surface water. The land use activities neighboring the sites where the E. coli strains were isolated are listed in Table 3.

BOX primers were used to generate PCR fingerprints for E. coli isolates obtained from surface water and groundwater to determine their diversity and the population structure. Strains with BOX-PCR fingerprint patterns with ≥85% similarity were considered clonal populations (Malathum et al. 1998; McLellan 2004). The optimized technique of BOX-PCR method in this study produced high visualization of DNA fragments and was replicable.

The total E. coli (n = 17) isolates produced high quality BOX-PCR fingerprinting and generated DNA fragments from 250 to 2,500 bp. Products of 750, 400, 300 and 250-bp have commonly occurred. The 300-bp band was the only band common to all the isolates. The greatest diversity of DNA fragments were observed within the range of 1,000–2,500-bp.

Bacteriological groundwater quality measurements were compared with the AFC water quality standards. The results indicated that 50% of the samples were unfit for human consumption. In addition, most of the sampling sites were affected by fecal contamination, because of the high levels of FC and the presence of E. coli. These values were related with land uses. Thus, it was found that most samples taken from wells located in the proximity of livestock had higher counts of total and FC and presence of E. coli. The hydrogeological setting plays an important role in groundwater contamination, specially the water table depth and USZ lithology, parameters that control the aquifer vulnerability to contamination. Thus, the USZ exerts a protection on the aquifer through physical (advection, dispersion and dilution), chemical (adsorption, oxidation, hydrolysis, redox reaction, dissolution) and biological (degradation or reduction) processes which produce the disappearance or attenuation of contaminants, influencing their arrival to the aquifer (Díaz Delgado et al. 2005). However, as was observed in this study, many contaminants persist through infiltration and reach the aquifer, in spite of the fact that the found USZ lithology was represented by fine sands and silts. In addition, the most contaminated wells showed shallow depths of the water table (2.58 to 13.2 m). This aspect is also greatly influenced by the contaminant load (duration, concentration, and so forth). In this case, the bacteriological analysis carried out in samples taken near livestock punctual sources, showed high concentration of total and FC and E. coli.

In the present study, the analysis of BOX-PCR fingerprinting revealed a low level of genetic diversity among E. coli water isolates, with two clusters, highlighting that cluster 1 has 100% of genetic similarity, gathering more than 60% of total strains. The whole dendrogram clearly indicates that this hydrological basin contains dominant strains with multiple isolates exhibiting identical (100%) or highly similar (greater than 85%) DNA fingerprints.

These results may be related to the arrival to surface water and groundwater of E. coli coming from the same contamination source. The contamination from a single fecal source would likely result in less diversity between isolates. Thus, different authors indicate that in water environments impacted by only one contamination source, the populations of E. coli were significantly less diverse than impacted by different sources (urban, industrial, sewage and agricultural activity) (Anjos Borges et al. 2003; McLellan 2004; Orsi et al. 2007; Lyautey et al. 2010). Thus, in Barranquita-Knutzen basin, livestock activity would probably be the main source of fecal contamination in surface and groundwater environments.

The multivariate statistical analysis of DNA fingerprinting allowed differentiating two clusters. Cluster 1 corresponds to all strains isolated from Knutzen sub-basin and cluster 2 corresponds to Barranquita sub-basin. This proves that hydrological factors influence the distribution of E. coli in the studied environment. Interestingly, these results have implications to manage the fecal contamination and the associated waterborne disease risks, in rural environment.

In the present study, the bacteriological analysis showed that more than 50% of water samples were unfit for human consumption.

The genetic diversity DNA fingerprinting analysis by BOX-PCR indicated low genotypic diversity of E. coli isolates from surface water and groundwater. The low genetic diversity suggested the presence of a dominant source of fecal contamination in the Barranquita-Knutzen basin.

The relationship between low genotypic diversity of E. coli and land use in the basin would prove that water contamination comes from livestock.

The genetic diversity of E. coli isolated from surface water was less than that identified in groundwater. This is influenced by the water body characteristics, considering that the stream samples are hydraulic linked whereas in the aquifer the samples are less hydraulic connected taking into account the different groundwater flow line directions. Moreover, the different local lithologies and geochemical water types in the aquifer may generate diverse microcosms in bacteria population.

The analysis of genetic diversity by BOX-PCR demonstrates that the hydrological dynamics have a great influence on microbial distribution. Accordingly, the water divide between the two sub-basins controls the two big different strains groups, showing that hydrological features can exert selective pressure on bacteria DNA.

This work was supported by poner PID 35/08 y Secyt UNRC.