Characterization of the microbial community in a lotic environment to assess the effect of pollution on nitrifying and potentially pathogenic bacteria

This study aimed to investigate microbes involved in the nitrogen cycle and potentially pathogenic bacteria from urban and rural sites of the São Pedro stream. Water samples were collected from two sites. A seasonal survey of bacterial abundance was conducted. The dissolved nutrient content was analysed. PCR and FISH analysis were performed to identify and quantify microbes involved in the nitrogen cycle and potentially pathogenic bacteria. The seasonal survey revealed that the bacterial abundance was similar along the year on the rural area but varied on the urban site. Higher concentration of dissolved nutrients in the urban area indicated a eutrophic system. Considering the nitrifying microbes, the genus Nitrobacter was found, especially in the urban area, and may act as the principal bacteria in converting nitrite into nitrate at this site. The molecular markers napA, amoA, and nfrA were more accumulated at the urban site, justifying the higher content of nutrients metabolised by these enzymes. Finally, high intensity of amplicons from Enterococcus, Streptococcus, Bacteroides/Prevotella/Porphyromonas, Salmonella, S. aureus, P. aeruginosa and the diarrheagenic lineages of E. coli were observed at the urban site. These results indicate a change in the structure of the microbial community imposed by anthrophic actions. The incidence of pathogenic bacteria in aquatic environments is of particular importance to public health, emphasising the need for sewage treatment to minimise the environmental impacts associated with urbanisation.


Introduction
Rivers and streams are important reserves of water available for human consumption, animal life, agriculture, and industry (Kenzaka et al., 2001).Therefore, there is need to conserve this resource.However, in the modern world, extensive and growing urbanisation is a threat to the ecosystem of streams, through factors associated with the discharge of sewage.This phenomenon has been called "urban stream syndrome" (Paul and Meyer, 2001).Some paper explores mechanisms driving the syndrome, and identifies appropriate goals and methods for ecological restoration of urban streams (Komínková, 2012).
Water systems must meet certain criteria to be considered healthy.In the past, these criteria included only physicochemical parameters (Murray et al., 2001).Indeed, several studies have shown that urbanisation increases the concentration of some nutrients, such as nitrogen, ammonia, nitrate, and phosphorus, decrease the concentration of oxygen and is responsible for pH changes in rivers and streams (Hoare, 1984;Meybeck, 1998;Wernick et al., 1998;Martinelli et al., 2010;Cumar and Nagaraja, 2011;Padmalal et al., 2012).However, the analysis of the chemical and physical characteristics of an ecosystem becomes limited when the objective is the understanding of its complexity as the biological components should also be taken into account.In this regard, many studies have demonstrated that knowledge of the structure and dynamics of the microbial community in rivers and streams is essential, mainly due to their role in several biogeochemical cycles (Brümmer et al., 2000;Araya et al., 2003;Kostanjšek et al., 2005).This is especially important for environments modified by anthropogenic action (Böckelmanna et al., 2000;Kenzaka et al., 2001;Tiquia, 2010).
Since it has been shown that urbanisation is related to the incresase in nitrogen, ammonium and nitrate levels (Hoare, 1984;Meybeck, 1998;Wernick et al., 1998;Martinelli et al., 2010;Cumar and Nagaraja, 2011;Padmalal et al., 2012), one can hypothesise that it could influence the density and diversity of nitrifying and denitrifying microbes in impacted ecosystems.Considering the global nitrogen cycle, nitrification and denitrification are important steps.Microorganisms are essentially the group that mediates these processes; bacteria are the major players, followed by fungi and archaea.The nitrifying bacteria include a number of genera.Nitrosomonas and Nitrosospira are generally known as ammonia oxidising bacteria (AOB), while Nitrobacter and Nitrospira are nitrite oxidising bacteria (NOB) (Head et al., 1993;Teske et al., 1994;Herbert, 1999).
In addition, urbanisation can be a source of pathogenic bacteria through the discharge of human sewage into water environments (Girones et al., 2010).Through the monitoring of microbes in water, it is possible to identify potential pathogens.Species such as Pseudomonas aeruginosa, Escherichia coli (including diarrheagenic E. coli), Aeromonas hydrophila, Staphylococcus aureus, Salmonella sp., Enterococcus sp., Streptococcus sp., and Bacteroides sp. have been found in urban systems, and the vast majority of these are associated with the fecal material of humans and other animals (Savichtcheva et al., 2007;Gonzalez et al., 2010;Willems et al., 2011).Therefore, it is acknowledged that the presence of these groups in water environments constitutes a potential health hazard.
In spite of the great impact that urbanisation imposes on the microbial community in freshwater, the determination of microbial composition has been a great challenge to microbial ecologists.Conventional methods based on microbiological culture cannot provide a representative composition of the microbial community (Böckelmanna et al., 2000).Alternatively, molecular biology techniques can be used.Although several studies have used molecular tools to study the microbial community in polluted freshwater, to the best of our knowledge, few of those targeted groupspecific bacteria (He and Jiang, 2005;Muniesa et al., 2006;Savichtcheva et al., 2007;Gonzalez et al., 2010).Because urbanisation has been related to the increase of nitrogen compounds in the aquatic systems and also a source of pathogenic bacteria through the discharge of sewage, we decided to target the microbes related to those specific aspects posed by the urbanisation phenomenon.Thus, the aim of this study was to investigate microbes from the nitrogen cycle and potentially pathogenic bacteria at urban and rural sites along a stream that receive discharge of domestic sewage and does not have appropriated wastewater management.We hypothesized that there would be a change on the structure of microbial community between the urban and rural region and our finds contribute to the understanding of the anthropogenic impacts on aquatic ecosystems.

Study area
The São Pedro stream is located in the city of Juiz de Fora, Brazil, and is responsible for supplying water to approximately 10% of the population of this city.A part of the stream that passes through the city is polluted, especially due to the discharge of domestic sewage (Latuf, 2004).A previous study in this area that analysed chemical and biological parameters indicated that the São Pedro stream can be divided in two distinct sites and that this difference may be caused by anthropic actions through the urbanisation process (G.Alfenas et al. manuscript in preparation).Site 1 (661799E/7591070N) is an urban area with homes nearby.At this point, the water has an unpleasant odour and very dark colour.The site 2 (668307E/7591772N) is located in a rural area in a farming region with clean and clear water.

Seasonal survey
A previous sampling was conducted between May, 2005 andApril, 2006, when 1 L of water was collected from the subsurface of the urban and rural areas of São Pedro stream.20% (w/v) paraformaldehyde in phosphate buffered saline (PBS) was added to an aliquot of the collected sample to a final concentration of 2%.Then, the cells were concentrated from the water sample (1.0 mL) on polycarbonate filters (25 mm in diameter, pore size 0.2 mm, Whatman).Filters were labelled with 100 µL of DAPI at a final concentration of 2 mg/mL for 3 minutes at room temperature.Bacterial cells on the filter sections were observed using a BX60 microscope (Olympus, Japan).The microbes were analysed in 10 random fields from each sample.From all the fields analysed, the mean abundance and standard deviations were calculated.

Point sampling
Approximately 1 L and 10 L of water samples from the subsurface at the urban and rural sites, respectively, of the São Pedro Stream, were collected in April 2010.The water samples were kept separately in a 15 L bottle previously rinsed three times with a sample from each site.

Analysis of physicochemical parameters
Water temperature and pH were measured in situ with Handheld meter pH 330i (WTW, Germany).An aliquot of water from each site was used to analyse the dissolved nutrients.The concentration of nitrite, nitrate, ammonium nitrogen, total organic nitrogen, and total phosphorus were measured following the methodology described by Wetzel and Likens (1991).The total nitrogen content was calculated as a sum of the concentration of nitrite, nitrate, ammonium nitrogen, and total organic nitrogen.

Detection of nitrifying bacteria
In order to identify nitrifying microbes and molecular markers of the different steps of the nitrogen cycle, PCR analyses were performed.The collected water samples were sonicated on ice using a Vibra Cell VCX130PB (Sonics & Materials, U.S.A) three times for 60 seconds, at an amplitude of 90%.The samples were filtered using a 3M filter in order to eliminate insects and small leaves, followed by GF/F filter to eliminate zooplankton and phytoplankton.The filtered water was centrifuged at 8000 rpm for 15 minutes in 500 mL bottles.
DNA was extracted by chemical digestion with phenolchloroform, according to Smith and Callihan (1992).Detailed, the pellet was resuspended in 500 µL of lysis solution (20% sucrose, 10 mM EDTA, 40 mM Tris-HCl and 1.3 mg/µL lysozyme) plus RNase (Promega Corporation, Madison, WI, U.S.A).The solution was incubated for 20 minutes at room temperature, followed by 10 minutes at 37 °C in a water bath.SDS was added to a final concentration of 0.2% and incubated 30 minutes at room temperature.The resulting lysate was extracted once with equal volumes of phenol:chloroform:isoamylalcohol (25:24:1; Sigma, St Louis, MO) and twice with chloroform:isoamylalcohol (24:1).Finally, nucleic acids were precipitated with an equal volume of cold isopropanol at -20 °C for 30 minutes.The pellet was washed with cold 70% ethanol, air dried and resuspended in 1 µL of RNase and DNase-free water.DNA integrity was checked by agarose gel electrophoresis and quantified spectrophotometrically in a NanoDrop ND 1000 instrument (Thermo Scientific, DE, USA).The DNA was stored at -20 °C until use.
Table 1 shows the primer sequences used to amplify specific DNA fragments from microbes related to the nitrogen cycle (Nitrosomonadaceae, Nitrosospira, Nitrospira, Nitrobacter, amoA -ammonia monooxygenase, napAnitrate reductase A, nfrA -nitrate ammonification).PCR amplifications were carried out as previously described.All the primers used in this study were fully characterised in previous studies, in which they were used as primers or fluorescent in situ hybridisation (FISH) probes (Amann et al., 1990;Mobarry et al., 1996;Hovanec et al., 1998;Watanabe et al., 2001;Mohan et al., 2004;Geets et al., 2007;Smith et al., 2007).The PCR products were separated by electrophoresis on a 1% agarose gel and visualised by staining with 0.5 mg of ethidium bromide per mL of gel.
The PCR assays were performed at least in duplicate.The data were analysed through a comparison of the patterns of band obtained on the urban and rural areas.The more intense the band, the more abundant is the amplified product since the initial concentration of DNA template were the same (20 ng) of all PCR reactions.To verify if the quantity and quality of the DNA extracted from both sites were comparable, PCR analysis was performed using primers for the domain Bacteria (positive control).The negative control was the PCR reaction containing all components needed to perform this analysis, except for DNA.
After the identification of microbes involved in the nitrogen cycle, FISH analysis was carried out in order to quantify some of the microbes in the environment.For better characterise our studied areas, we also investigated the density of the domains Bacteria and Archaea through FISH analysis.As the first step, 20% (w/v) paraformaldehyde in phosphate buffered saline (PBS) was added to an aliquot of the collected sample to a final concentration of 2%.Then, the cells were concentrated from the water sample (1.0 mL) on polycarbonate filters (25 mm in diameter, pore size 0.2 mm, Whatman).The filters were placed on glass slides and covered with 40 µL of hybridisation solution containing the probes at a final concentration of 2.5 ng/ µL (0.9 M NaCl; 20 mM Tris-HCl pH 7.2; 5 mM EDTA; 0.01% SDS; a variable concentration of formamide).The probe sequences and hybridisation conditions are given in Table 2.The filters were incubated in a hybridisation chamber at 42 °C overnight.Then, they were transferred to a 96-well plate with 1 mL of pre-warmed washing solution (20 mM Tris-HCl pH 7.2; 10 mM EDTA; 0.01% SDS; a variable concentration of NaCl) and incubated at 48 °C for 15 minutes.Filters were labelled with 100 µL of DAPI at a final concentration of 2 mg/mL for 3 minutes at room temperature.Subsequently, they were gently washed three times with 80% ethanol, dried on paper and mounted on glass slides using a glycerol:PBS (7:3) solution.Bacterial cells on the filter sections were observed using a BX60 microscope (Olympus, Japan).The microbes were analysed in 10 random fields for each probe from each sample.For each microscopic field, two categories were determined: total DAPI-stained cells and cells stained with the specific probe.From all the fields analysed, the mean abundance and standard deviations

Detection of potentially pathogenic bacteria
In order to identify bacteria of human health interest in the water of the studied environments, PCR analysis were performed using primers for potentially pathogenic bacteria: Streptococcus, Enterococcus, Bacteroides/Prevotella/ Porphyromonas, Salmonella, A. hydrophila, S. aureus, P. aeruginosa and diarrheagenic lineages of E. coli (see Table 1).The DNA used was the same as that used for the nitrifying bacteria analyses, and PCR amplifications were carried out as previously described by (Amann et al., 1990;Trebesius et al., 2000;Watanabe et al., 2001;Fukushima et al., 2003;Qin et al., 2003;Aranda et al., 2004;Rinttilä et al., 2004;Chotár et al., 2006).The negative control was the PCR reaction containing all the components needed to perform this analysis, except DNA.As the positive control, PCR amplifications were carried out on DNA extracted from a reference culture of the bacterial group targeted by the primer.For the lineage ETEC of E. coli, PCR amplification specific for the gene elt was carried out and its amplicon was purified using the QIAquick Gel Extraction Kit (Qiagen, Inc.).The DNA fragment obtained was sequenced and blasted against the nucleotide database of the National Center for Biotechnology Information (NCBI) to confirm its specificity.

Statistical analysis
The data were tested for normality and homogeneity of variance.The single criterion variance analysis (ANOVA -one way) and an a posteriori Tukey's test were used for normal data (seasonal survey) and the Mann-Whitney test was used for non-normal data (FISH analysis) using the program SigmaPlot 11.0.There was an attempt to transform the non-normal data to a normal distribution but to no avail.In both cases, values of P < 0.05 were considered significant.

Seasonal survey
The annual mean of bacterial abundance were 17.4×10 6 mL -1 (±6.97) on the urban area and 4.85×10 6 mL -1 (±1.83) on the rural.The bacterial abundance obtained in all months analysed was significantly higher in the urban site compared to the rural.Considering the rural area, the bacterial abundance was similar throughout the year, except for the months October and May.However, the urban area presented a seasonal variation.April was the month that had bacterial abundance closer to the mean annual abundance (17.97×10 6 mL -1 ).Furthermore, the bacterial abundance obtained in April was statistically similar to the abundance detected in most of the months analysed, thus justifying the choice of the month April for the posterior analysis.

Physicochemical parameters
The water temperature was 22 °C in both studied areas.The water pH at the urban site was 6.5 and at the rural site it was 6.8.The concentration of dissolved nutrients is shown in Table 3.The concentration of nitrite, nitrate, ammonium nitrogen, total organic nitrogen, and total phosphorus were higher in the urban area.The total nitrogen obtained in the urban area was 4.107 mg/L; this concentration was 0.482 mg/L in the rural area.

Nitrifying bacteria
Through PCR analysis shown in Figure 1, the groups Nitrosomonadaceae and Nitrosospira that function as AOB and Nitrospira (one of the NOB groups) were detected in both areas of study.Bacteria of the genus Nitrobacter, which convert nitrite into nitrate, were present at urban area and in the rural area were less than the detection limit of our assay.The genes amoA, napA, and nfrA, important molecular markers of the nitrogen cycle, were present in both areas.However, the higher intensity of the bands from the urban area compared to the rural one suggests that the amplicons for amoA, napA and nfrA were more abundant in this site (see Figure 1).Through FISH analysis, it was observed that the groups Nitrosomonadaceae, Nitrospiraceae, and Nitrobacter had significantly higher abundance in the urban area (see Figure 2).Considering the domains Bacteria and Archaea, the abundance of Bacteria was significantly higher in the urban area (9.48×10 6 mL -1 ) compared to the rural area (0.80×10 6 mL -1 ).For the domain Archaea, the abundance was 2.76×10 6 mL -1 in the urban area and 0.35×10 6 mL -1 in the rural area.

Potentially pathogenic bactéria
The Figure 3 and 4 show the results of PCR using primers specific for genera and species of already known human pathogens.The genera Enterococcus sp. was present in the urban area and was not detected in the rural area.The amplicons for the genera Salmonella sp., Streptococcus sp., and Bacteroides/Prevotella/Porphyromonas had a higher intensity, suggesting that they were more accumulated in the urban area.The same results were observed for the species P. aeruginosa and S. aureus.In contrast, the amplicons for A. hydrophila were present at the same intensity at both sites (as shown in Figure 3).
For E. coli, the amplicons were more abundant in the urban area.Enteropathogenic E. coli (EPEC) was identified according to the genotype eae+/bfp+ in both studied areas.Enterohemorrhagic E. coli (EHEC) was identified according to the genotype stx1+/stx2+ only in the urban area and stx1+/stx2-in the rural area.Enteroinvasive E. coli (EIEC) was classified as ipaH+ in the two areas and the strain ETEC (enterotoxigenic E. coli) was identified according to the genotype elt+/est-in the urban area and elt-/est-in the rural area.It is important to mention that the amplicons obtained for all the diarrheagenic E. coli were more intense at the urban site when compared to the rural site (see Figure 4).

Discussion
Increasing urbanisation has serious impacts on freshwater ecosystems.It is known that the pollution and discharge of sewage alters the microbial community of freshwater environments (Paul and Meyer, 2001;Walsh et al., 2005;Girones et al., 2010).However, there is still a need to establish biological standards that can evaluate the water quality.The study presented here attempted to identify the specific differences in the microbial community at two sites along the São Pedro stream that is influenced by urbanisation.The abundance of bacteria (1×10 7 ) obtained in the urban area of study was in agreement with Kenzaka et al. (2001) and within the range reported for other polluted rivers that were considered eutrophic systems by Yamaguchi et al. (1997).The concentration of dissolved nutrients obtained in this study in the urban area also corroborated with classifying this area as an eutrophic system since this classification is within the range of nutrients found at this site.This high abundance can be justified by the influx of organic material in this environment, resulting from domestic sewage.The high abundance of Archaea in the urban area can be explained since these microbes can easily survive in environments with a higher content of nutrients and could be active in the recycling of nitrogen (Dridi et al., 2011).
PCR analysis showed that the group Nitrobacter, was less than the detection limit of our assay in the rural area.This result corroborates with FISH analysis, through which we observed a higher abundance of Nitrobacter   in the urban area compared to the rural site (as shown in Figure 1 and 2).The high density of the Nitrobacter group is in agreement with the data of dissolved nutrients; these results showed,(at the urban site) a low level of nitrite which was below the detection level, and a high quantity of nitrate (see Table 3).For the group Nitrospiraceae, which includes the genus Nitrospira, no difference was observed between the two habitats by PCR analysis (see Figure 1).These results suggest that, in these habitats, the genus responsible for converting nitrite into nitrate is Nitrobacter.The high amount of nitrate compared to nitrite in both studied areas, especially the urban one, could explain the presence of the napA gene in these environments since the enzyme codified by this gene needs it substrate (nitrate) to perform its function.The majority of the Nitrosomonadaceae group was observed in the urban area, according to FISH analysis (as shown in Figure 2).This result is in agreement with and explains the extremely high amount of ammonium nitrogen found in this area, since Nitrosomonadaceae, including the genus Nitrosomonas, is one of the groups responsible for oxidising ammonia to nitrite.The presence of the amoA and nrfA genes could also be explained by the high amount of ammonium nitrogen in the urban area since the enzyme ammonium monooxygenase needs its substrate (ammonium nitrogen) to perform the nitrification process; the gene nfrA is responsible for supplying this nutrient to the environment.It is important to mention that the analysis of the nutrient content and the presence of the group or enzyme that metabolises it must be considered within the timescale of nutrient metabolism.
In addition to the nitrifying bacteria, our study also investigated bacteria with important clinical interests that are potential human pathogens, since a high number of these microbes are found in urban sewage and may be considered environmental contaminants.Most of those pathogens can be removed from the water by a sewage treatment but, when no treatment is performed, these pathogens can be discharged into effluents and can be a threat to public health (Girones et al., 2010).In Brazil, non-treated sewage is a major cause of water pollution.An important genus of this family Enterobacteriaceae is the genus Salmonella.It is composed of two species, Salmonella enterica and Salmonella bongori, with many subgroups and serovars, all of which are capable of causing human illness.Contamination with Salmonella could occur through drinking contaminated water, swimming in contaminated water, or eating food washed with contaminated water (Boyd et al., 1996;Hsu et al., 2011).The presence of Salmonella sp. in more abundance in the urban area of this study (Figure 3) is in agreement with the findings of Gonzalez et al. (2010), indicating a high prevalence of Salmonella sp. in the Brazilian Lagoon that is an urban ecosystem undergoing accelerated degradation especially due to pollution.Savichtcheva et al. (2007) also found that Salmonella sp. was the most frequently detected enteropathogen in samples of rivers with different levels of pollution, as well as in samples of sludge and wastewater treatment.
Finally, we considered the most studied organisms in the Enterobacteriaceae family, E. coli.Several virulence properties are well-characterised among E. coli populations, differentiating them from commensal and pathogenic strains.Considering the pathogenic E. coli, these virulence determinants are genetically encoded either by chromosomal, plasmid, or bacteriophage DNA and are represented by selected genes such as eae (attaching and effacing lesions), bfpA (localized adherence), ipaH (enteroinvasive mechanism), the genes encoding heat-labile toxin (elt), and heat-stable toxin (est), and the genes encoding the Shiga toxins, stx1 and stx2 (Aranda et al., 2004).Several studies have aimed at investigating the incidence of diarrheagenic E. coli in different water environments, and the majority of them have demonstrated that these lineages and the related virulence genes are present in urban and polluted environments.Our data (see Figure 4) confirm these findings.The presence of EHEC and ETEC in sludge and rivers with different degrees of pollution was detected by Savichtcheva et al. (2007).The stx genes were found in different streams with high urbanisation and also in preserved forest (García-Aljaro et al., 2005;Higgins et al., 2005;Muniesa et al., 2006).Our results show that EHEC was genotyped as stx1 + /stx2 + at the urban site; however the stx2 gene was less than the detection limit in the rural area of the study, which is relevant and worrisome since the stx2 gene is believed to be commonly associated with more severe illnesses (Oliveira et al., 2008).An interesting observation is that the same pathotypes were also found in regional studies with urban pigeons and isolates from human diarrheic feces, indicating the circulation of the same genes in different samples from the city of Juiz de Fora (Silva et al., 2009;Garcia et al., 2011).
The genera Enterococcus, Streptococcus, and Staphylococcus are low-GC Gram-positive bacteria belonging to the phylum Firmicutes.All these genera have species that are classified as human commensals, and as such are part of the normal human microbial community.However, studies have shown that under adverse conditions (uncolonised areas or imbalanced homeostasis) these species can behave as opportunistic pathogens and became virulent and resistant to multiple antibiotics (Willems et al., 2011).Our results (see Figure 3) indicate the presence of Enterococcus sp., Streptococcus sp., and S. aureus in a water environment.He and Jiang (2005) also detected, using molecular techniques, the presence of Enteroccocus in coastal waters and sewage.S. aureus is not associated with the fecal material of humans, but the presence of this species in water suggests that this environment is a potential source of community-acquired S. aureus infections (Goodwin and Pobuda, 2009).
Other groups that are part of the normal human microbiota are the anaerobic bacteria Bacteroides sp., Prevotella sp., and Porphyromonas sp.(Wexler, 2007).As well as Grampositive cocci, they can develop virulence and became important human pathogens.Different studies have used the microbes of the genus Bacteroides as fecal pollution indicators (Bernhard and Field, 2000;Boehm et al., 2003).Our results show the presence of these microbes in a water environment especially in the area subjected to urbanisation (see Figure 3).Bacteroides spp.are important indicators, and are more sensitive than coliforms, since they can be found in greater quantities in humans.This bacteria is an obligate anaerobe and the high level of Bacteroides in water indicates recent contamination (Fiksdal et al., 1985).
The genera Aeromonas is involved in several diseases in humans and other animals.A hydrophila was found in both study areas (see Figure 3).This result was expected because these microbes are primarily aquatic organisms and, as reported by Scoaris et al. (2008), A. hydrophila can be found in treated and non-treated water.P. aeruginosa, a free-living bacteria ubiquitous in the environment, is another potential human pathogen.This species can be found either in oligotrophic environments or in high nutrient environments (Mena and Gerba, 2009), which is in agreement with our data.Our results show that P. aeruginosa was present in the rural area with poor nutrients, but it was more abundant in the urban area considered a eutrophic system.Garcia-Armisen et al. (2011) have recently shown the incidence of P. aeruginosa resistant to multiple antibiotics in rivers contaminated with sewage.
In conclusion, our study presents a comparison of the community composition of a polluted and preserved stream.The results show a major incidence of nitrifying and potentially pathogens at the polluted site, which is influenced by urbanisation, especially by the discharge of sewage.These findings, specially the high incidence of bacteria from gastrointestinal tract of human and others animals indicate a change in the structure of the microbial community imposed by human occupation.The survival and persistence of pathogenic bacteria in natural environments is of particular importance to public health.This study confirms the need for sewage treatment and policies to minimise the environmental impacts associated with urbanisation.Our results show the scenario of a local stream but reflect the situation of many rivers and streams in Brazil.This is alarming, since these resources are responsible for supplying water for part of the population.Our results suggest a range of genes and microbes that can be used as marker of water deteriorated by human action through urbanisation.In the future, these markers can be included in the biological parameters evaluated for urban streams and can be used to better classify the water quality.

Table 1 .
Sequence of primers used for PCR analysis in this study.
were calculated.All counts were corrected by subtracting the counts obtained with the negative control probe.The experiments were performed in duplicate.

Table 2 .
Sequence of Oligonucleotide probes used for Fluorescence in situhybridisation in this study.
* For this probe a competitor of NIT3 was used; †Formamide concentration in hybridisation buffer; ‡ Sodium chloride concentration in washing buffer.

Table 3 .
Concentration of Nutrients (mg/L) dissolved in the studied areas.