Diversity of Mercury Resistant Escherichia coli Strains Isolated from Aquatic Systems in Rio de Janeiro, Brazil

1 Departamento de Ciências Biológicas, Escola Nacional de Saúde Pública Sergio Arouca (Ensp), Fundação Oswaldo Cruz (FIOCRUZ), Rua Leopoldo Bulhões 1480 Manguinhos, 21041-210 Rio de Janeiro, RJ, Brazil 2 Departamento de Microbiologia Médica, Instituto de Microbiologia Paulo de Góes, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil 3 Departamento de Microbiologia Geral, Instituto de Microbiologia Paulo de Góes, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil


Introduction
Chemical contamination of aquatic systems consists of a relevant pollution pattern causing drastic impacts on human, animal, and ecosystem health [1].Among the various chemical contaminants, mercury plays an important role and once released in aquatic systems, mercury can resist to natural degradation processes and persist for a long time in these environments without losing its toxicity [2].
The concern about environmental contamination by this metal is due to its high toxicity, especially to the nervous system, and its bioaccumulation and biomagnification, providing persistence and wide distribution in global aquatic environment.Even regions with no mercury discharging may be affected [2][3][4][5][6][7].
Mercury toxicity to humans and other organisms is related to the chemical form to which the organisms were exposed, the route and time of exposure, dose, nutritional status, individual susceptibility, and genetic predisposition [3,4,6,8].Symptoms and contamination sources are rather different in exposure to elemental mercury, inorganic or organic mercury compounds [3,4].Human contamination by this metal may occur by different pathways such as vapors inhalation, contaminated food and/or water consumption, and to a lesser extent through skin contact [3,6].Mercury exposure triggers a series of effects including neurological, renal, cardiovascular, respiratory, gastrintestinal, hepatic, genotoxic, immunological, dermal, reproductive, and neoplasic disorders.Exposure during pregnancy may lead to malformations, mental retard, cerebral palsy, seizures, and death [3,4,6,8].Mercury resistance is one of the most studied toxic metals resistance mechanisms [7].It has been reported that some bacteria and fungi isolated from different sources have developed resistance mechanisms that enable them to survive even in environments highly contaminated by mercury [9].There are several described bacterial mechanisms that confer protection to harmful concentrations of mercury [10].Among them, we highlight the mercury enzymatic detoxification, promoted by the mercuric reductase protein (MerA), which catalyze the reduction of Hg(II) to volatile Hg(0) [11,12].Considering the genetic of MerA expression, the Hg resistance (mer) operon presents a fundamental role in regulation, Hg binding, and organomercury degradation.It consists of essential genes as merR (responsible for the regulation of the operon), merT/merP (transport of mercury into the bacterial cell), merA (reduction of ionic mercury), and accessory genes such as merB, merC, merD, merE, merF, and merG, that encode proteins that add other skills to microorganisms [13,14].MerR protein can act both as a repressor and activator of transcription.In the absence of Hg 2+ , MerR acts as repressor by binding to the mer operon operator region and preventing the transcription of merTPCAD.In presence of Hg 2+ , it binds to one of two MerR binding sites forming a complex that acts as an activator of mer operon transcription [15].Mermediated approaches have had broad applications in the bioremediation of mercury-contaminated environments and industrial waste streams [8,11,12,16,17].
Mercury resistance in bacteria has been observed in both Gram-positive (S. aureus, Bacillus sp.) and Gram-negative bacteria (E.coli, P. aeruginosa, Serratia marcescens, and Thiobacillus ferrooxidans) [12,16].Mercury resistance is encoded on genetic elements such as plasmids and transposons, which contributes to horizontal dissemination among different bacteria and widespread occurrence in different bacterial groups and environments [12].
In Gram-negative bacteria, including E. coli, the mer operon has already been described [18].However, epidemiological and genetic studies related to mercury resistance are scarce.Therefore, the investigation of the mercury resistance features has been crucial to improve bioremediation processes in contaminated environments in order to minimize human exposure and consequent adverse health effects.
In the present study, E. coli isolates from aquatic systems, in the city of Rio de Janeiro, Brazil, were characterized by phenotypic and genotypic traits related to mercury resistance.Bacteriological tests were carried out in order to determine mercury susceptibility, and molecular approaches based on amplification assays were used to investigate the presence and diversity of mercury resistance gene (merA).

Materials and Methods
2.1.Water Sampling.Samples were selected and grouped according to potential contamination sources in the city of Rio de Janeiro, Brazil.We studied five aquatic environments: residential, industrial, agricultural, and hospital wastewaters and recreational waters (Table 1).

Sample Collection.
Collection procedure consisted of membrane filtration method with some modifications [19].An aliquot of 60 mL of water was aspirated from the upper layer of the water column to a depth of approximately 30 centimeters with a syringe holder adapted to sterile filtration.The aspirate was filtered on a 0.22 m cellulose acetate membrane (Milllipore) and transported under refrigeration for immediate laboratory processing.

Mercury Resistance Phenotype.
In order to classify E. coli as resistant or sensitive to mercury, each strain was tested on nutrient agar (NA, Difco) supplemented with 5 M of Hg 2+ .Evidence of bacterial growth after a period of 24-48 h at 37 ∘ C allowed to classify E. coli strains as Hg resistant (Hg R ).E. coli ATCC 35218 (Hg resistant) and E. coli ATCC 23724 (Hg susceptible) were used as control strains.When no growth was observed, a strain was considered as sensitive.These tests were done in duplicate.

Minimal Inhibitory Concentration (MIC)
. MIC determination was performed following the methodology described by Andrews [21] with some modifications.Overnight cultures of the isolates in nutrient broth (NB, Difco) containing 1 M Hg were adjusted in saline NaCl 0.9% (w/v) in order to contain 1.5 × 10 9 bacterial cells/mL (McFarland 0.5).An aliquot of 50 L was inoculated in nutrient agar plates containing 10 to 40 M Hg.After 24-48 h at 37 ∘ C, the MIC value was determined by observing bacterial growth on agar plates in the presence of the lowest Hg concentration.MIC tests were performed with those E. coli strains exhibiting mercury resistance phenotype ≥5 M Hg.All experiments were performed in duplicate.
2.6.merA Detection.All strains were screened for the presence of merA sequence by PCR amplifications as described by Ní Chadhain and colleagues [22], with some modifications.Each reaction was carried out in 25 L PCR mixture containing 3 L of bacterial DNA obtained through thermal extraction of 18-24 h bacterial growth in tryptic soy broth (TSB, Difco), 2.5 L of 10X buffer (Invitrogen), 2 mM MgCl 2 (Invitrogen), 0.2 mM dNTP (Invitrogen), 30 M of each primer, and 1 U of Platinum Taq DNA polymerase (Invitrogen).The pair of primers used was A1s-n.F (5  -TCCGCAAGTNGCVACBGTNGG-3  ) and A5-n.R (5  -ACCATCGTCAGRTARGGRAAVA-3  ).PCR reaction was conducted in Mastercycler Personal thermocycler (Eppendorf) under the following amplification conditions: initial denaturing step at 94 ∘ C for 5 min, followed by 45 cycles at 94 ∘ C for 10 sec, 68 ∘ C for 40 sec, and 72 ∘ C for 1 min with a final extension at 72 ∘ C for 7 min.Approximately 10 L of the resulting amplification products was added to 2 L of running buffer (gel loading buffer, Invitrogen) and separated by electrophoresis on agarose gel at 1.3% concentration (w/v) prepared in Tris-Borate-EDTA 0.5X (5X-0.89M Tris-HCl (LGC Biotech) 0.89 M boric acid (Merck) and 0.024 M EDTA (LGC Biotech) (pH 8.4)) at a constant voltage of 70 V.Electrophoresis gel was stained with 0.5 g/mL ethidium bromide solution (Invitrogen) over a period of 15 min and washed in distilled water for about 30 min.Gel was visually inspected by using an ultraviolet light transilluminator (UVITec, Cambridge, UK) and photographed in digital image capture system (silver UVIPro, Cambridge, UK).To estimate the size of the fragments a 100 bp DNA ladder standard (Invitrogen) was used.E. coli strains ATCC 35218 (Hg resistant) and ATCC 23724 (Hg sensitive) were used as controls.

Random Amplification of Polymorphic DNA (RAPD-PCR)
. RAPD-PCR analysis was performed according to the methodology described by Pacheco and colleagues [23].Each reaction was carried out in a 30 L PCR mixture containing 2 L of bacterial DNA, 3 L of 10X buffer (Invitrogen), 250 M each dNTP (Invitrogen), 3 mM MgCl 2 (Invitrogen), 1 U of Taq DNA polymerase (Invitrogen), and 30 M of each primer.Primers used were 1247 (5  -AAGAGCCCGT-3  ), 1254 (5  -CCGCAGCCAA-3  ), 1290 (5  -GTGGATGCGA-3  ), and A04 (5  -AATCGGGCTG-3  ).The reaction was conducted in a Mastercycler Personal thermocycler (Eppendorf) under the following amplification conditions: an initial denaturing step at 94 ∘ C for 1 min, followed by 4 cycles at 94 ∘ C for 4 min, 37 ∘ C for 4 min, and 72 ∘ C for 4 min, 30 cycles at 94 ∘ C for 1 min, 37 ∘ C for 1 min, and 72 ∘ C for 2 min with a final extension at 72 ∘ C for 10 min.Reaction products were analyzed by electrophoresis in 1.5% agarose gels and stained with ethidium bromide.RAPD profiles were inspected visually and defined according to the presence or absence and intensity of polymorphic bands.A 1 kb DNA ladder was used as a molecular weight marker (GIBCO, BRL, Gaithersburg, MD, USA).Semiautomated analysis used the UVI Soft Image Acquisition and Analysis Software, program UVIPro bandmap version 11.9 (UVItec, Cambridge, UK).Cluster analysis was done by using the unweighted pair group method with arithmetic averages (UPGMA) of the Image Analysis System.The percentages of similarity were estimated by the Dice coefficient.The reproducibility of the RAPD amplifications was assessed using the selected primers with different DNA samples isolated independently from the same strain and amplified at different times.

Denaturing Gradient Gel Electrophoresis (DGGE).
DGGE analysis was performed according to the methodology described by Muyzer and colleagues with some modifications [24].AxyPrep DNA Gel Extraction kit(Axygen Biosciences) was used for purificating the PCR-merA DNA fragment (285 bp).For PCR-DGGE reaction a final volume of 25 L in amplification reactions containing 3 L of purified DNA, 2.5 L 10X buffer (Invitrogen), 2 mM MgCl 2 (Invitrogen), 0.2 mM dNTPs (Invitrogen), 30 M of each primer, 1% formamide, and 1 U of Platinum Taq DNA polymerase (Invitrogen) was used.The pair of primers for amplification was A1sn.F (5  -TCCGCAAGTNGCVACBGTNGG-3  ) and A5-n.R (5  -ACCATCGTCAGRTARGGRAAVA-3  ).The reaction was conducted in Mastercycler Personal thermocycler (Eppendorf) and programmed for an initial denaturation of 94 ∘ C for 5 min followed by 45 cycles of 94 ∘ C for 10 sec, 68 ∘ C for 40 sec, and 72 ∘ C for 1 min, with a final extension of 72 ∘ C for 7 min.Approximately 25 L of amplified PCR product was added to 15 L of DNA electrophoresis dye (0.005 g  Bromophenol blue, 0.005 g xylene cyanol, 7 mL glycerol P.A., and 3 mL deionized water) and ran on a polyacrylamide gel (8% w/v of acrylamide/bisacrylamide ratio 37.5 : 1) with a linear denaturant gradient ranging from 55% to 80% (where 100% is a solution of 7 M urea and 40% formamide v/v).Electrophoresis was performed in equipment using the Dcode Universal Mutation System (BIO-Rad) and conducted at constant voltage of 100 V at 60 ∘ C for 6 h in 0.5X Tris-acetate (10 mM Tris-acetate, 5 mM Sodium Acetate, 25 mM EDTA, and pH 7.4).After electrophoresis the gel was stained with Sybr Green (Molecular Probes, OR, USA) for 30 minutes and visualized under UV transilluminator.The reproducibility of the assay was tested by loading three PCR products for each sample on DGGE gels.

Mercury Resistance Phenotype and Minimal Inhibitory Concentration (MIC).
A total of 164 strains were classified as mercury resistant (Hg R ) and represented 92.1% of the E. coli isolates (164/178).All Hg R exhibited the Hg MIC value of 10 M.

Mercury Resistance Phenotype and Minimal Inhibitory Concentration (MIC).
Many studies have been conducted in order to determine the mercury resistance in environmental bacteria by testing the minimum inhibitory concentration [25][26][27][28].There is not a standard protocol for determining the MIC of heavy metals.Liquid and/or solid media with different chemical compositions have been commonly used for these assays, as well as variation of metals concentrations.Methodology itself may offer some obstacles such as precipitation and volatilization of the solution and complexes between the metal and culture medium components.These variations, if not minimized before its application, may directly influence the result obtained [26].So, it is very difficult to compare the obtained results with previous studies because of the great diversity of MIC values and the procedures adopted, especially considering the broad spectrum of mercury resistant bacteria that require specific conditions for growing and laboratory processing.The ubiquity of bacterial mercury resistance has been observed in environments worldwide and is supposed to be the result of external interference by humans and other animals through environmental contamination for several years [5,12,26].There were no reports about mercury contamination in the sampling sites; however, Hg resistance was widely detected.Bacterial resistance to mercury present in the environment is considered as one of many examples of genetic and physiological adaptation of microbial communities exposed to contaminants.Several factors have been found to contribute to this phenotype in the rural and urban areas including the use of mercury-based fungicide in the paper industry, agriculture, and hospital disinfectants.These factors may encourage selective activities and result in mercury resistance in open environment [29].Additionally, toxic metal resistance genes are commonly found in environmental bacteria, and these genes may confer coresistance or crossresistance to antimicrobial drugs codified on the same genetic element [26,30].So, selection of microbial communities exposed to toxic levels of the metal or submitted to the coselection mechanisms has led to high rates of circulation of these resistant bacteria in aquatic systems [31,32].

PCR Amplification of merA
Gene.The genetic system evolved as mer operon is the only well-known bacterial mercury resistance system with high yield transformation of its toxic target into volatile nontoxic forms [27,31,[33][34][35], particularly in Gram-negative bacteria [14,22].The mer locus is found to be widely distributed among bacterial lineages, and mer-like sequences have been described.Several biochemical mechanisms are identified, and the complexity among the ecological niche of mercury-resistant microbes is still not fully described [10,35].merA plays a key role on mercury resistance of bacterial community exposed to mercury contamination, but the combinatorial action of genetic determinants seems to confer a broad spectrum mercury detoxification system [10,35].So, the involvement of additional genetic determinants not investigated here, acting as effectors or regulators genes, must be considered for the expression of mercury resistance phenotype among merA negative E. coli strains.merA gene was detected in E. coli isolated from residential wastewaters (11/75), industrial wastewaters (1/9), and hospital wastewaters (2/29).The higher frequency of merA+ E. coli strains obtained from residential wastewaters compared to industrial and hospital wastewaters may be related to several factors such as the representative sampling of each area investigated and involvement of additional genetic determinants as well as related to the intrinsic characteristics of the rural and urban locations.

Random Amplification of Polymorphic DNA (RAPD-PCR)
. RAPD-PCR is a recognized powerful tool showing high discriminatory potential, reproducibility, sensibility, and specificity under well-standardized protocols.Random amplification of polymorphic DNA (RAPD) has been successfully used as a molecular typing system for studies on diversity of E. coli population [23,36].
RAPD typing revealed levels of polymorphism that are consistent with previously reported observations for E. coli and has been attributed to the high plasticity of this bacterial species.Several molecular approaches mainly based on genetic techniques have been successfully applied in order to assess the clonal nature and variability within species [23,36].The occurrence of distinct patterns of E. coli phylogenetic distribution provides evidence of both vertical and horizontal transmission [37][38][39].The mechanisms of genetic diversification contribute to E. coli evolution and creation of new variants, as this bacterial species is often subjected to DNA rearrangements, excisions, transfers, and acquisitions [37,40].There are several highly adapted clones that have acquired specific virulence elements which confer an increased ability to adapt to new niches.Such plasticity may confer ability to acclimate environmental bacteria to new niches allowing these microorganisms to become members of microbial communities in a variety of environments, even facing conditions very different from their primary habitat [36,38,39,41].RAPD-PCR approach was used to investigate the overall chromosomal relatedness among merA+ strains and revealed a high genetic diversity population suggesting that mercury resistance is widely dispersed in E. coli.The observed genotypic diversity led us to suppose that, in Rio de Janeiro, merA+ E. coli isolates consist of nonrelated epidemiological strains and may represent distinct evolutionary lineages.Despite the genetic variability, clustering analysis revealed that the degree of diversity was to a lesser extent among E. coli strains obtained from the same aquatic environment evidencing the circulation of closely related strains.

Denaturing Gradient Gel Electrophoresis (DGGE).
DGGE fingerprinting is a technique widely used in microbial ecology studies and has been focused on studies of genetic diversity and bacterial communities from several environments [17,42].Variability within merA gene has been described, and diverse MerA protein homologs have been identified in both archaeal as well as bacterial genomes but not in eukaryal genomes [17].The increased complexity of mer operons can be attributed to the gradual addition of functions involved in the regulation of the operon by Hg, Hg transport, and organomercury resistance [17].The diversity of merA gene in Gram-negative and Gram-positive bacteria has been accessed by several approaches including those using restriction fragment assays [27,31,33,34].In all these studies, a high genetic variability was detected in merA determinant carried by bacterial species from different environments.However, RFLP technique is limited since it relies on specific target, requiring prior knowledge of the sequences to be analyzed.In the present study, DGGE was used to investigate the merA+ variability among E. coli mercury resistant.
DGGE typing revealed diversity within the 285 bp merA fragment corroborating previous findings that described the occurrence of genetic exchanges in mercury resistance gene as a result of addition, rearrangements, excisions, and horizontal transfer.
Ní Chadhain and colleagues [22] developed a protocol using degenerated primers and detected high diversity within merA sequence from evolutionary distinct Gram-bacteria.In our study, this methodology allowed the detection of variability in the 285 bp merA fragment among 14 E. coli strains (Figure 3).These results are in agreement with previous findings regarding the widespread occurrence and diversity of mercury resistance markers among distinct microbial populations from several environments, including soils and sediments, aquatic systems, animals, and clinical isolates [13,14,27,28,31,32].The high plasticity found in the bacterial genome contributes to the diversity and dissemination of genetic markers favoring its circulation in geographically dispersed environments, even between distinct evolutionary lineages.
E. coli isolates sharing similar RAPD profiles were found to exhibit the same merA DGGE pattern suggesting the circulation of conserved or partially conserved merA sequence among closely related strains.The molecular approaches used as fingerprint tools were found to be accurate and useful methods in distinguishing between closely related bacteria.The obtained results are relevant to our understanding on the characteristics of mercury resistant E. coli circulating in natural environments in aquatic systems in Rio de Janeiro.Our findings substantially expand our knowledge about mer evolution and biodiversity of these microorganisms, and contribute to studies on bioremediation process and environmental management of Hg contamination.

Conclusions
The present study detected a wide dissemination of E. coli isolates resistant to mercury in distinct aquatic systems in the city of Rio de Janeiro possibly due to selective activities with varying patterns of exposure to Hg. Genetic analysis of merA+ strains revealed high degree of diversity among the bacterial population indicating that mercury resistance is widely dispersed in E. coli.These findings suggest that, in the city of Rio de Janeiro, merA+ E. coli may constitute bacterial communities epidemiologically independent and may represent distinct evolutionary lineages.The variability detected within the 285 bp merA fragment possibly reflects the occurrence of specific genetic events.E. coli strains sharing RAPD profile and DGGE band pattern reinforce the hypotheses of circulation of conserved merA sequence among closely related strains.In the light of the pathogenicity attributed to E. coli population, more accurate analyses are required for applications in bioremediation processes.

Figure 2 :
Figure 2: Dendrogram generated by the Dice coefficient and clustering by unweighted pair group method with arithmetic mean and respective RAPD profiles of merA+ Escherichia coli isolates using 1254 primer.

Table 1 :
Origin of Escherichia coli strains included in this study.
). Electrophoresis technique on denaturing gradient gel enabled the detection of variability within the 285 bp gene fragment associated with mercury resistance (merA).Supporting the results obtained from RAPD-PCR, RM 7, RM 8, and RM

Table 2 :
Genetic and phenotypic traits of E. coli strains carrying the 285 bp merA fragment according to aquatic systems and sampling sites.