Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Microbial Quality and Phylogenetic Diversity of Fresh Rainwater and Tropical Freshwater Reservoir

  • Rajni Kaushik,

    Affiliations Singapore-Delft Water Alliance, National University of Singapore, Singapore, Singapore, Department of Civil and Environmental Engineering, National University of Singapore, Singapore, Singapore

  • Rajasekhar Balasubramanian ,

    Affiliations Singapore-Delft Water Alliance, National University of Singapore, Singapore, Singapore, Department of Civil and Environmental Engineering, National University of Singapore, Singapore, Singapore

  • Hugh Dunstan

    Affiliation School of Environmental and Life Sciences, The University of Newcastle, Callahan, NSW, Australia

Microbial Quality and Phylogenetic Diversity of Fresh Rainwater and Tropical Freshwater Reservoir

  • Rajni Kaushik, 
  • Rajasekhar Balasubramanian, 
  • Hugh Dunstan


The impact of rainwater on the microbial quality of a tropical freshwater reservoir through atmospheric wet deposition of microorganisms was studied for the first time. Reservoir water samples were collected at four different sampling points and rainwater samples were collected in the immediate vicinity of the reservoir sites for a period of four months (January to April, 2012) during the Northeast monsoon period. Microbial quality of all fresh rainwater and reservoir water samples was assessed based on the counts for the microbial indicators: Escherichia coli (E. coli), total coliforms, and Enterococci along with total heterotrophic plate counts (HPC). The taxonomic richness and phylogenetic relationship of the freshwater reservoir with those of the fresh rainwater were also assessed using 16 S rRNA gene clone library construction. The levels of E. coli were found to be in the range of 0 CFU/100 mL – 75 CFU/100 mL for the rainwater, and were 10–94 CFU/100 mL for the reservoir water. The sampling sites that were influenced by highway traffic emissions showed the maximum counts for all the bacterial indicators assessed. There was no significant increase in the bacterial abundances observed in the reservoir water immediately following rainfall. However, the composite fresh rainwater and reservoir water samples exhibited broad phylogenetic diversity, including sequences representing Betaproteobacteria, Alphaproteobacteria, Gammaproteobacteria, Actinobacteria, Lentisphaerae and Bacteriodetes. Members of the Betaproteobacteria group were the most dominant in both fresh rainwater and reservoir water, followed by Alphaproteobacteria, Sphingobacteria, Actinobacteria and Gammaproteobacteria.


Airborne microorgansims can be transferred to aquatic systems through atmospheric fallout of coarse particles (dry deposition) [1] and rainfall (wet deposition) [2][3], leading to changes in the microbial composition of receiving water bodies [4][5]. The presence of bacterial pathogens in airborne particulate matter (PM) is of particular concern from the public health perspective as these aerosolized bacteria can form new cells in PM [6] and be metabolically active with the potential to biogeochemically mediate atmospheric chemistry [7][9]. The transfer of PM containing viable bacterial pathogens from the atmosphere to water bodies can affect not only ecotoxicology, but also human health through various exposure pathways.

Pathogenic episodes in lakes and reservoirs are often associated with rain events and riverine inflows [4], [10]. Excessive rainfall has been reported to be a significant contributor to historical waterborne disease outbreaks due to mobilization and transport of bacterial pathogens [10][12]. Stormwater runoff is a major cause of deterioration of surface water quality in urban areas. When rainfall occurs on paved surfaces, large volumes of water are swiftly carried to drains and discharged into receiving surface waters. Thus, the transport of microbial pollution from rain runoff to lakes and reservoirs is a major concern for management of natural waters worldwide [13]. Changes in the abundance of heterotrophic and coliform bacteria resident in stored water bodies have recently been reported in relation to incoming bacterial loads following rain events [5]. Hence, it is very important to understand the impact of rainfall carrying the live microbial aerosols on the quality of freshwater reservoirs in terms of changes in the abundance of pathogens, the microbial community composition and diversity.

Water treatment organizations and environmental protection agencies frequently use the presence of bacterial indicator organisms (e.g. fecal coliforms and enterococci) and their abundance in surface waters as a surrogate for the risk of contamination by actual pathogenic microorganisms [14][16]. Most of the studies reported in the literature characterized some bacteria residing in bulk water at various points in the drinking water supply system using cultivation-based approaches (14,17–18]. In general, heterotrophic plate counts (HPC) are used to assess the overall bacterial quality of drinking water, or natural waters [19]. Furthermore, most of the bacterial cells in natural communities are present in a viable but nonculturable (VBNC) state and therefore are non-culturable by current cultivation methods [20][21].

The real composition and dynamics of bacterial communities in freshwater resources remain largely unknown. In particular, limited studies have been conducted to determine the levels of bacterial pathogens in fresh rainwater prior to its collection and storage in order to assess its possible impact on the quality of roof-harvested rainwater and also that of aquatic systems upon deposition [1], [9]. The microbial quality of roof harvested rainwater is not only influenced by that of fresh rainwater, but also by the types of rooftop surfaces, the level of water tanks cleanliness, and the presence of insects' and birds' feces. However, no systematic study has been reported in the literature so far on the phylogenetic diversity of fresh rainwater describing the taxonomic richness of readily culturable organisms present in this water medium although such studies have been conducted for roof-harvested rainwater [22]. Therefore, there is a strong need for quantifying the levels of bacterial pathogens and phylogenetic diversity in both fresh rainwater and reservoir water following their simultaneous collection at the same sampling site of the reservoir.

Molecular approaches-based methods are good tools to elucidate composition of microbial communities residing in various aquatic environments [23][24]. These approaches exploit the use of rRNA as taxonomic marker for microorganisms [25][26]. The 16 S rRNA gene is tailor-made for microbial identification due to its universal presence in bacteria, extreme species sequence conservation and evolution-induced interspecies variability [27]. With the recent development and application of large-scale high throughput pyrosequencing-based method [28], community-wide spatial and temporal information on microbial community functional structure and potential activity can be rapidly obtained. Although the pyrosequencing-based approach is able to identify new sequences, it suffers from very high sensitivity to random sampling errors, dominant populations and contaminated non-target DNA [29]. The 16 S rRNA gene clone library method is a relatively old method that has been widely used in bacterial community analyses in various environments [30][32]. This method has been successfully applied to understand the role of freshwater bacterioplankton in global biogeochemical processes in the aquatic ecosystems [33][34]. However, the 16 S rRNA application to fresh rainwater has not been reported to date. The study of phylogenetic diversity in fresh rainwater and reservoir water could provide a basic understanding of the complex communities of environmental bacteria present in fresh rainwater and their impact on the composition of microbial communities in reservoir water intended for human consumption. The objectives of this work were to (1) study the impact of microbial loading of fresh rainwater on the water quality of a tropical reservoir in Singapore (characterized by heavy rainfall) upon deposition using total bacteria and traditional bacterial indicators and (2) examine the kind of bacterial diversity that fresh rainwater and reservoir water harbor using traditional 16 s rRNA cloning and sequencing method.

Materials and Methods


Singapore's climate is characterized by uniform temperature and pressure, high humidity, and abundant rainfall. Singapore receives an annual rainfall of about 2400 mm. There are no distinct wet, or dry seasons as rainfall occurs every month of the year. The two main seasons, based on the prevailing dominant winds, are the Northeast monsoon season (from late November to March), and the Southwest monsoon season (from late May to September). April to early May and October to early November are generally the transitional months separating the monsoons. December is usually the wettest month with an average rainfall of 280 mm.

Both rainwater samples and reservoir samples were collected at a tropical reservoir (coordinates-1°22′03″N 103°48′07″E) as part of the Singapore-Delft Water Alliance project. There was no need for any special permission to carry out the water sampling, and the activities did not involve endangered or protected species. Reservoir water samples were collected from four different sites at the reservoir using a grab sampling method along with rainwater samples. Both types of water samples were collected concurrently at the same four sites (No. 1, 2, 3 and 4). The reservoir has a capacity of 27.8 million of water over 304 hectares of water surface. Site 1 was located at the centre of the reservoir with no land use. Site 2 was located at the corner of the reservoir with influences from a local drainage system. Site 3 was situated near a major highway and site 4 was near a golf course, having anthropogenic influence from this land use type (Fig.1).

Figure 1. Outline of the four sampling sites at the tropical reservoir of Singapore used in this study.

Both rainwater and reservoir water samples were collected from the sampling locations at regular weekly intervals and within 24 h of rainfall events between January 2012 and April 2012. A total of 33 rain samples on an event-to-event basis were collected (n = 6, 5, 12, and 10 for January, February, March and April, respectively). On each sampling occasion two replicate grab samples (1 L) were collected 1 m from the shore at a depth of ∼0.5 m in sterile containers. 1 Liter of fresh rainwater samples was collected in sterilized glass bottles (amber colored glass bottles with a sterilized funnel attached to them) placed at all the four sites used for reservoir sampling. Both fresh rainwater and reservoir water samples (immediately after rainfall) collected from all the four sites were transferred to 2 L sterilized bottles for microbiological and chemical analyses. These water samples were transported to the laboratory in a chilled-cold box and processed within 6 h of collection. The samples were first analyzed for physico-chemical analyses; pH, conductivity, turbidity and conductivity were measured for all the samples as provided in Text S1,Table S1.

Microbiological Analyses

Concentrations of total coliform bacteria and E. coli were determined using the m-ColiBlue24 membrane filtration system (Millipore, Cat #M00PMCB24, Bedford, Massachusetts). For all the collected water samples, 100 mL of the sample was filtered onto cellulose ester membranes using vacuum filtration and the membrane was then incubated for 24 h in sterile petri dishes containing absorbent pads soaked with 2 mL of m-ColiBlue24 broth at 37°C. This was performed in duplicate for all the samples. The colonies in blue color were indicative of E. coli, while total coliforms were enumerated by counting the colonies in red color. The average CFU/100 mL values obtained for rain events in each month for E. coli and total coliforms were estimated.

HPC and Enterococci counts were also determined based on average counts obtained for all the samples collected at the four sites in this study. Briefly, for HPC enumeration, one mL of each water sample was serially diluted and the dilutions were aseptically plated in duplicates onto a plate count agar (Sigma-Aldrich, USA) and incubated at 37°C for a maximum duration of 48 h. The average colony counts were expressed as CFU/mL.

For Enterococci, the enumeration was performed as per the USEPA Method [35]. In brief, 100 mL of the water sample was filtered onto cellulose esters membranes using vacuum filtration in duplicate and the membranes were then placed on top of the membrane-Enterococcus Indoxyl-β-D-Glucoside Agar (mEI Agar, BD, NJ, USA) incubated for 24 h at 41±0.5°C. Colonies with a blue halo, regardless of color, were enumerated as Enterococci. The colony counts were expressed as average CFU/100 mL.

DNA Extraction

One liter of each fresh rainwater and reservoir water samples (duplicates for each sampling site and date) was first filtered through a pre-combusted glass fiber filter (90-mm diameter, GF/F, Whatman) followed by a 0.22 µm hydrophilic polycarbonate filter (Millipore Corporation, Bedford, MA). The filters were cut into small pieces and suspended in 1.0 mL of phosphate buffered saline and DNA was extracted using Ultraclean Microbial DNA Isolation kits (MO BIO laboratories, Carlsbad, CA) according to the manufacturer's instructions. The extracted DNA solution was stored at −80°C.

Construction of 16 S rRNA gene libraries

Clone libraries were constructed from composite DNA samples of fresh rainwater and composite DNA samples of reservoir water obtained at the four sites. Briefly, the 16 S rRNA gene was amplified from pooled environmental DNA samples using primers 27f (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492r (5′-TACCTTGTTACGACTT-3′) [36]. The PCR products were purified using UltraClean PCR Clean-up DNA purification kits (MO BIO Laboratories) and made up to 30 µL with water. Amplified fragments were cloned into TOPO-TA plasmids using the TOPO-TA cloning kit (Invitrogen, Carlsbad, CA) and transferred into Escherichia coli DH5a cells (TakaraBio, Otsu, Japan) to construct 16 S rRNA gene libraries.

Cloned plasmid inserts were amplified directly from cells as described [37] using vector primers. The 16 S rRNA gene portion of the cloned DNA was initially sequenced using the ABI Prism BigDye terminator v3.1 and cycle sequencing kit (PE Applied Biosystems). We sequenced 150 clones for each bacterial gene library.

Phylogenetic analyses

Cloned gene sequences were vector-trimmed and aligned using the NAST (Nearest Alignment Space Termination) algorithm for creating multiple sequence alignments [38]. NAST aligned sequences were chimera checked with chimera slayer. All bioinformatics processes were performed in the MOTHUR environment [39]. Classification of sequences was done using Bayesian classifier method implemented in MOTHUR against Greengenes database [40]. The division level groupings were determined by taxonomic assignment performed by the Ribosomal Database Project 10.0 Classifier tool [41]. Pairwise distances of the aligned clone sequences were calculated and Operational Taxonomic units (OTUs) were grouped using the average neighbor method with a cut-off of 0.03 sequence similarity. Coverage of clone libraries was calculated according to the method recommended by Good (1953) [42]. A Phylogenetic tree was constructed using the neighbor- joining method of ARB with Jukes-Cantor correction model and with 1000 bootstrap replications [43].

Community Statistical Analyses

Microbial quality data were subjected to the student t-test for determining statistical significance. For microbial diversity, analyses of beta diversity with replicate data from rainwater and its respective reservoir water were used for comparison of the two microbial communities. All profiles were inter-compared in a pair-wise fashion to determine a dissimilarity score and stored in a distance matrix. The Unifrac distance metric, as described in Lozupone et al. 2006, was utilized for the phylogenetic distance between OTUs to determine the dissimilarity between the two communities [44]. Weighted Unifrac was used considering the OTU abundance along with the Adonis test of statistical significance (p<0.05), which is a non-parametric multivariate analysis of variance (MANOVA) with the Adonis function and utilizes the sample-to-sample distance matrix directly for finding significant differences among these communities [45].

Results and Discussion

Microbiological quality of fresh rainwater and reservoir water

Microbial quality is usually assessed by measuring ‘fecal indicator bacteria’ (also referred to as fecal indicator organisms, or FIOs) that are generally opportunistic pathogens, present in large numbers in fecal materials. Their presence in water samples is used to indicate the presence of fecal pollution and the possibility that fecal associated pathogens may also be present. The most commonly examined FIOs are thermotolerant coliforms (also termed fecal coliforms), Escherichia coli and intestinal Enterococci (also termed fecal streptococci). E. coli are considered the best indicators of fecal contamination in water. The presence of thermotolerant coliforms/E. coli in water is unacceptable from the public health perspective as it indicates that a major health risk exists. However, it should be noted that certain members of the coliform group live outside of the gastrointestinal tract in the environment and may create a false indication of fecal contamination [46]. Also, several strains of Escherichia coli which are of environmental origin can be thermotolerant, thus giving false positive results [47]. It was therefore necessary to evaluate a range of relevant indicator organisms and total bacteria.

Figs. 2, 3 and 4 show these microbial indicators for both fresh rainwater and reservoir water over the four months of sampling. Levels of thermotolerant coliforms/E. coli are expressed as colony forming units (CFU) per 100 mL (CFU/100 mL). The levels of E. coli were found to be in the range of 0 CFU/100 mL–75 CFU/100 mL for fresh rainwater and in the range of 10 CFU/100 mL–94 CFU/100 mL in reservoir water (Fig. 2). According to the WHO guidelines for treated drinking water (2006), the levels of E. coli should be less than 1 CFU/100 mL [48]. Thus, fresh rainwater is non potable for direct consumption and needs to be treated for potable purposes.

Figure 2. E. coli counts from both rainwater samples and reservoir samples collected from four sites from January to April, 2012.

Figure 3. Total coliforms counts from both rainwater samples and reservoir samples collected at four sites from January to April, 2012.

Figure 4. Enterococcus counts from both rainwater samples and reservoir samples collected at four sites from January to April, 2012.

Total coliforms were previously considered indicators of fecal contamination. The NHMRC and AWRC “Guidelines for Drinking Water Quality in Australia” (2004) do not consider total coliforms as useful indicators of fecal contamination in the absence of E. coli, and have not proposed any guideline value for total coliforms [49]. As the case with HPC, total coliforms can also be used as an indicator for the effectiveness of any treatment program. However, it should be noted that while fecal coliforms are enteric organisms and therefore virtually exclusively of fecal origin, the total coliform count may comprise environmental organisms that are not necessarily common to the digestive tracts of vertebrate animals [50]. The total coliforms were found in the range of 10 CFU/100 mL–220 CFU/100 mL in reservoir water as compared to 10 CFU/100 mL–139 CFU/100 mL in fresh rainwater (Fig. 3). The highest counts were obtained from site #3 in both rainwater and reservoir water, which is near a major highway with residential buildings in the vicinity. This anthropogenic influence could be due to the abundance of airborne microbial pathogens in urban areas with high population density [1], [9]; Turkum et al. (2008) reported that most of the species found in rainwater were derived from aerosol and gas-phase components [51].

Enterococci are a specific group of bacteria that are found in high numbers in both human and animal faeces, and are therefore a valuable indicator for determining the extent of fecal contamination of a water source [16], [52][53]. Enterococci were found to be in the range of 0 CFU/100 mL–11 CFU/100 mL in fresh rainwater and 0 CFU/100 mL–35 CFU/100 mL in the reservoir water (Fig. 4). Overall, the four microbial indicators were found to be the highest at site #3 in both types of water samples. Thus, the results of our microbiological indicator analyses suggested that both the fresh rainwater and the reservoir water samples are not suitable for human consumption without any treatment similar to the findings on the roof harvested rainwater [22], [52], [54].

Total HPC bacteria of the reservoir water samples were two orders of magnitude higher than those of the rainwater. The baseline levels for the reservoir water ranged from 330 CFU/mL to 7.9×104 CFU/mL, as compared to 280 CFU/mL to 7.2×102 CFU/mL in fresh rainwater (Fig. 5). The reason for the reservoir water having two fold higher magnitudes of total HPC bacteria and slightly higher number of total coliforms could be the additional contribution from sediments apart from rainfall and riverine inflows. HPC bacteria are carried by runoff into the reservoir from highland agricultural areas during intensive rainfall events. Thus, rain events may lead to an inflow of high nutrient concentrations as well as high loads of microbes [55][56]. Based on the microbial indicator numbers, there was no statistically significant difference between the rainwater and reservoir water quality with respect to E. coli and Enterococci. However, the microbial quality of rainwater and reservoir was found to be significantly different (p≤0.05) for total heterophic counts and total coliforms according to the t-test. It was also observed that the concentrations of all the microbial indicators were higher in the month of March as compared to those in other months. This can be attributed to a higher frequency of rainfall in March as compared to the relatively dry period in the months of January and February.

Figure 5. Total heterotrophic plate count bacteria from both rainwater samples and reservoir samples collected at four sites from January to April, 2012.

It should be noted that E. coli and Enterococci are indicative of recent fecal contamination as compared to heterophic counts and total coliforms. The influence of E. coli and Enterococci of atmospheric origin on the reservoir water quality appears to be relatively less as compared to that of total bacteria HPC and total coliforms, this could due to their lower abundance in the atmosphere and/or the influence of physical processes such as advection, dispersion, diffusion resulting in low residence time [22]. Other processes such as competitive exclusion and nutrient change may also play role to regulate the survival of these incoming bacteria [5].

Microbial composition and Phylogenetic analyses of fresh rainwater and reservoir water

Microbial communities are fundamental to the functioning of aquatic ecosystems. With the availability of metagenomic technologies and approaches [57], it is important to know the changes in total microbial diversity of both rainwater and reservoir water in addition to the changes in microbial indicator numbers. Gaining insights into the total microbial community would provide a better understanding of its role in mediating aquatic ecological processes and biogeochemical cycling [57][58]. However, the roles of most microorganisms in natural systems are unclear as most of them cannot be cultivated for investigation by current conventional culturing methods [59]. The introduction of culture-independent molecular methods has shed light on the determination of community compositions, and laid the foundation for gaining a deep understanding of aquatic microbial ecology. Rainwater has received considerable attention as a potential alternative source of potable and non-potable water in regions where there is water scarcity [60]. Rainwater collection is an ancient practice and currently still being practiced, especially in areas with no running water. However, the use of 16 S rRNA sequencing techniques has not been applied to fresh rainwater investigations. Thus, there is a lack of information on its microbial communities and their role in aquatic microbial ecology.

In this study, 10 classes of bacteria were detected in fresh rainwater and four classes of bacteria in reservoir water. In fresh rainwater, sequences were affiliated with Betaproteobacteria, Alphaproteobacteria, Sphingobacteria, Actinobacteria, Gammaproteobacteria, Lentisphaerae, CH21, Phycisphaerae, Chlorobia and Spirochaetes. In contrast, the reservoir water library detected sequences affiliated with only Betaproteobacteria, Alphaproteobacteria, Sphingobacteria and Gammaproteobacteria. Betaproteobacteria was found to be the dominant class in both the libraries (Fig. 6). The fresh rainwater had higher diversity and taxonomic richness at the class level than those of reservoir water as the numbers of OTUs were found to be 63 and 29 for fresh rainwater and reservoir water, respectively. However, the proportion of Alphaproteobacteria in the bacterial community was found to be higher in the reservoir water than in the fresh rainwater. Alphaproteobacteria, at least at the class level, are resistant to predation: their relative higher abundance in reservoir water may be due to the capacity to degrade recalcitrant organic compounds such as humic substances accompanied by a tendency of members of the Alphaproteobacteria to form filaments, aggregates, or Caulobacter-like stalked cells that sometimes can make up a majority of the Alphaproteobacteria population in freshwater lakes [61].

Figure 6. Bacteria taxon richness on (A) class level (B) genus level and (C) species level observed in the fresh rainwater and reservoir water.

The taxonomic richness observed in both types of water samples and their relative distribution appears to be similar to that observed in many freshwater and marine systems, as determined by both cultivation and molecular techniques [22], [62][63]; these findings are consistent with the results from previous studies where the dominance of Betaproteobacteria has been observed consistently in freshwater systems, particularly among free-living groups, while α and γ sub-classes appear to dominate among particle-attached groups [64]. The high abundance and amenability to culturing have contributed to the Betaproteobacteria being the best-studied group in freshwater lakes [61]. Venter et al. (2004) carried out a comprehensive genomic study of natural waters at an oligotrophic ocean site in the Sargasso Sea, and found an abundant bacterial distribution dominated by Proteobacteria of α, β and γ sub-classes, followed by Actinobacteria and Bacteroidetes [65]. These findings are similar to those from the fresh rainwater microbial community analyses in the current study as well as from the study conducted by Evans et al., 2009 in rainwater tanks [22]. The class Betaproteobacteria was the dominant and the most diverse group in freshwater mesocosm and this dominance is suggested to be associated with their ability to respond quickly to nutrient additions [66][67]. The predominance of Proteobacteria in fresh rainwater can also be suggestive of the generally clean oligotrophic nature of fresh rainwater.

Actinobacteria are Gram-positive bacteria representing the most abundant group (50–70%) of total bacteria in various freshwater habitats [68][69] were also found to be present in fresh rainwater. Previous studies have found that members of Actinobacteria contributed to glucose assimilation and heterotrophic nitrification [70][71] and played a key role in nutrient and energy cycling in aquatic habits [71]. Also most interestingly, the UV stress resistance has also been postulated to be one of the reasons for the success of the Actinobacteria in the upper waters of lakes, which often have high UV transparency. Warnecke and colleagues showed a significant positive relationship between Actinobacteria abundance and UV transparency in study on mountain lakes [68]. In addition to UV protection, many members of the Actinobacteria are capable of producing spores, allowing them to survive long periods of desiccation. Thus, strong UV protection and desiccation resistance via spore formation, together with the known small cell size of these organisms, would make the freshwater Actinobacteria particularly suitable for aerial dispersal, thereby explaining their ubiquitous representation in globally dispersed lakes. However, to date, these taxa have rarely been identified in any air samples [61]. This is the first study to-date to detect their presence in fresh rainwater suggesting their ability to survive in airborne particles that tend to be scavenged by rainwater during rain events [1]. This wet-deposition process enables the global distribution of the microorganisms of atmospheric origin in aquatic ecosystems [66].

Phylogenetic tree-based microbiome comparison of the fresh rainwater and reservoir water samples was performed by the neighbor-joining method of experimentally observed OTUs and reference type strains with each leaf representing an OTU displays counts of sequences observed in each sample (Fig. 7). Notable differences in the frequencies of the some of the bacterial genera were observed. Bradyrhizobium, Chitinophaga, Caulobacter and Ralstonia were found to be more abundant in reservoir water than in fresh rainwater whereas Curvibacter was found to be more abundant in fresh rainwater than in reservoir water. However, unlike taxonomic richness, the coverage of the fresh rainwater clone library (150 sequences from pooled DNA) as calculated based on the species OTU (97% 16 S rRNA gene sequence identity) by Good's Clone Coverage [42] was low (77%) as compared to the coverage of the reservoir water clone library (92%).

Figure 7. Taxonomic tree indicating the phylogenetic distribution of all bacterial groups in both rainwater and reservoir water samples.

The OTU with the greatest difference in relative abundance was associated with Curvibacter for fresh rainwater and with Ralstonia for reservoir water. Curvibacter and Ralstonia belong to the Burkholderiales order of Betaproteobacteria, under the taxa Comamonadaceae that is reported to be the most abundant typical freshwater fast growing and nutrient-loving group [31], [72]. Rainwater and reservoir samples formed separate groups in ordination analysis, but were not found to be significantly dissimilar (Adonis p-value  = 0.125) using OTU abundance metrics. The same diversity of bacteria being present in two different environmental samples suggests that the microbiological composition of the atmosphere and that of the hydrosphere are intimately linked with each other through wet and dry deposition mechanisms. The type of bacterial diversity present in fresh rainwater, the general abundance distribution, and the resemblance of the composition to that of reservoir, have indicated the likely existence of definable micro ecosystems in fresh rainwater and their impact on other aquatic systems. The functional operation of a stable micro-ecology, dominated by well-adapted core resident groups, may have implications with regard to the harvesting of rainwater and for management of water resources.


This study focused on the investigation of the microbial diversity of fresh rainwater and that of reservoir water with a concurrent assessment of bacterial counts and species profiles in a tropical region, characterized by abundant rainfall throughout the year, for the first time. The results of the study suggested that despite the presence of bacterial pathogens in rainwater and their wet deposition through rainfall, there was no significant increase in the microbial loading of the tropical reservoir water. It appears that the incoming bacterial loads from the rainwater entering the reservoir either undergo dilution, or are not sustained for a long period of time due to the influence of physical processes such as advection, dispersion and diffusion and other processes (competitive exclusion and nutrient change) that may act to regulate the survival of incoming bacteria. Thus, despite the presence of microbial pathogens, rainwater harvesting in large water catchment areas is a promising freshwater resource following treatment. These findings on the composition of aquatic microbial communities in the reservoir water and fresh rainwater by 16 s rRNA clone libraries indicated that the members of Betaproteobacteria dominated both the communities. The OTU with the greatest difference in the relative abundance for fresh rainwater was Curvibacter whereas for reservoir water it was Ralstonia. Actinobacter was detected in fresh rainwater, suggesting their presence in airborne particles as well. The fresh rainwater showed greater taxonomic richness than that of the reservoir water.

However, due to lower coverage in traditional cloning method, our understanding about the bacterial composition and phylogenetic diversity of the bacterial community in the freshwater is still limited. Hence, high-throughput molecular tools such as pyrosequencing and hybridization to PhyloChip array that are able to elucidate not only cultivable bacteria, but also viable-but-not-cultivable (VBNC) may extend and expand our understanding of the tropical freshwater reservoir microbial communities. More in-depth studies on the linkage between atmospheric inputs of microorganisms and the microbial diversity of freshwater resources are warranted. This research group has recently performed study of reservoir microbial diversity using PhyloChip microarrays at the same reservoir in the presence and absence of rain events (work in progress), which may add more insights into differences between the traditional cloning and metagenomics of freshwater bacterial diversity.

Supporting Information

Table S1.

Summary of physico chemical analyses of Fresh rainwater and reservoir water quality.



The authors gratefully acknowledge the support and contributions to Singapore-Delft Water Alliance, National University of Singapore (NUS) for the support provided to the pursuit of this study. Rajni Kaushik thanks SDWA for supporting her Ph.D. study. The technical assistance for sampling provided by Chen Sijing is gratefully acknowledged.

Author Contributions

Conceived and designed the experiments: RK RB. Performed the experiments: RK. Analyzed the data: RK RB HD. Contributed reagents/materials/analysis tools: RK RB. Contributed to the writing of the manuscript: RK RB HD.


  1. 1. Kaushik R, Balasubramanian R (2012) Assessment of bacterial pathogens in fresh rainwater and airborne particulate matter using Real-Time PCR. Atmos Environ 46: 131–139.
  2. 2. Womack AM, Bohannan BJM, Green JL (2010) Biodiversity and biogeography of the atmosphere. Philos Trans R Soc Lond B Biol Sci 365: 3645–3653.
  3. 3. Kaushik R, Balasubramanian R, de la Cruz AA (2012) Influence of air quality on the composition of microbial pathogens in fresh rainwater. Appl Environ Microbiol 78: 2813–2818.
  4. 4. Brookes JD, Antenucci J, Hipsey M, Burch MD, Ashbolt NJ, et al. (2004) Fate and transport of pathogens in lakes and reservoirs. Environ Int 30: 741–759.
  5. 5. Martin AR, Coombes PJ, Harrison TL, Dunstan RH (2010) Changes in abundance of heterotrophic and coliform bacteria resident in stored water bodies in relation to incoming bacterial loads following rain events. J Environ Monitor 12: 255–260.
  6. 6. Dimmick RL, Wolochow H, Chatigny MA (1979) Evidence that bacteria can form new cells in airborne particles. Appl. Environ. Microbiol 37: 924–927.
  7. 7. Ariya P, Nepotchatykh O, Ignatova O, Amyot M (2002) Microbiological degradation of atmospheric organic compounds. Geophys Res Lett 29: 1–4.
  8. 8. Amato P, Ménager M, Sancelme M, Laj P, Mailhot G, et al. (2005) Microbial population in cloud water at the Puy de Dôme: Implications for the chemistry of clouds. Atmos Environ 39: 4143–4153.
  9. 9. Kaushik R, Balasubramanian R (2013) Discrimination of viable from nonviable Gram-negative bacterial pathogens in airborne particles using propidium monoazide-assisted qPCR. Sci Total Environ 449: 237–243.
  10. 10. Auld H, MacIver D, Klaassen J (2004) Heavy rainfall and waterborne disease outbreaks: The Walkerton example. J Toxicol Environ Health A 67: 1879–1887.
  11. 11. Curriero FC, Patz JA, Rose JB, Lele S (2001) The association between extreme precipitation and waterborne disease outbreaks in the United States, 1948–1994. Am J Public Health 91: 1194–1199.
  12. 12. Ferguson C, Husman AMD, Altavilla N, Deere D, Ashbolt N (2003) Fate and transport of surface water pathogens in watersheds. Crit Rev Env Sci 33: 299–361.
  13. 13. Olson BH, Nagy LA (1984) Microbiology of potable water. Adv Appl Microbiol 30: 73–132.
  14. 14. Ward NR, Wolfe RL, Justice CA, Olson BH (1986) The identification of gram-negative, non-fermentative bacteria from water: problems and alternative approaches to identification. Adv Appl Microbiol 31: 293–365.
  15. 15. Edberg SC, Rice EW, Karlin RJ, Allen MJ (2000) Escherichia coli: the best biological drinking water indicator for public health protection. J Appl Microbiol 88: 106S–116S.
  16. 16. Sidhu J, Ahmed W, Toze S (2013) Application of Microbial Source Tracking Toolbox to Identify Sewage Contamination in Stormwater Run-off in Brisbane AWA Water. 40: 81–85.
  17. 17. LeChevallier MW, Babcock TM, Lee RG (1987) Examination and characterization of distribution system biofilms. Appl Environ Microbiol 53: 2714–2724.
  18. 18. Sartory DP (2004) Heterotrophic plate count monitoring of treated drinking water in the UK: a useful operational tool. Int J Food Microbiol 92: 297–306.
  19. 19. Oliver JD (2000) The public health significance of viable but nonculturable bacteria. In: R. RColwell and D. JGrimes (ed). Nonculturable microorganisms in the environment. ASM Press, Washington D.C., pp. 277–300.
  20. 20. Giovannoni SJ, Britschgi TB, Moyer CL, Field KG (1990) Genetic diversity in Sargasso Sea bacterioplankton. Nature 345: 60–63.
  21. 21. Szewzyk U, Szewzyk R, Manz W, Schleifer KH (2000) Microbiological safety of drinking water. Annu Rev Microbiol 54: 81–127.
  22. 22. Evans CA, Coombes PJ, Dunstan RH, Harrison T (2009) Extensive bacterial diversity indicates the potential operation of a dynamic micro-ecology within domestic rainwater storage systems. Sci Total Environ 407: 5206–5215.
  23. 23. Woese CR (1987) Bacterial evolution. Microbiol Rev 51: 221–271.
  24. 24. Höfle MG, Haas H, Dominik K (1999) Seasonal dynamics of bacterioplankton community structure in a eutrophic lake as determined by 5S rRNA analysis. Appl Environ Microbiol 65: 3164–3174.
  25. 25. Altmann D, Stief P, Amann R, De Beer D, Chramm A (2003) In situ distribution and activity of nitrifying bacteria in freshwater sediment. Environ Microbiol 5: 798–803.
  26. 26. Tringe SG, Hugenholtz P (2008) A renaissance for the pioneering 16S rRNA gene. Curr Opin Microbiol 11: 442–446.
  27. 27. Schwarzenbach RP, Egli T, Hofstetter TB, von Gunten U, Wehrli B (2010) Global Water pollution and human health. Annu Rev Env Resour 35: 109–136.
  28. 28. Huber JA, Mark Welch D, Morrison HG, Huse SM, Neal PR, et al. (2007) Microbial population structures in the deep marine biosphere. Science 318: 97–100.
  29. 29. Zhou J, Kang S, Schadt CW, Garten CT Jr (2008) Spatial scaling of functional gene diversity across various microbial taxa. Proc Natl Acad Sci USA 105: 7768–7773.
  30. 30. Hiorns WD, Methe BA, NierzwickiBauer SA, Zehr JP (1997) Bacterial diversity in Adirondack Mountain lakes as revealed by 16 S rRNA gene sequences. Appl Environ Microbiol 63: 2957–2960.
  31. 31. Zwart G, Hiorns WD, Methe B, van Agterveld MP, Huismans R, et al. (1998) Nearly identical 16 S rRNA sequences recovered from lakes in North America and Europe indicate the existence of clades of globally distributed freshwater bacteria. Syst Appl Microbiol 21: 546–556.
  32. 32. Wu X, Xi WY, Ye WJ, Yang H (2007) Bacterial community composition of a shallow hypertrophic freshwater lake in China, revealed by 16S rRNA gene sequences. FEMS Microbiol Ecol 61: 85–96.
  33. 33. Zehr JP, Ward BB (2002) Nitrogen cycling in the ocean: new perspectives on processes and paradigms. Appl Environ Microbiol 68: 1015–1024.
  34. 34. Mason OU, Di Meo-Savoie CA, Van Nostrand JD, Zhou JZ, Fisk MR, et al. (2009) Prokaryotic diversity, distribution, and insights into their role in biogeochemical cycling in marine basalts. ISME J 3: 231–242.
  35. 35. U.S. Environmental Protection Agency (USEPA) (2002) Method 1600: Enterococci in water by membrane filtration using membrane-Enterococcus Indoxyl-β-D-Glucoside Agar (mEI). EPA-821-R-02-022. Office of Water, Washington, D.C.
  36. 36. Weisburg WG, Barns SM, Pelletier DA, Lane DJ (1991) 16S ribosomal DNA amplification for phylogenetic study. J Bacteriol 173: 697–703.
  37. 37. Vergin KL, Rappe′ MS, Giovannoni SJ (2001) Streamlined method to analyze 16 S rRNA gene clone libraries. Biotechniques 30: 938–940.
  38. 38. DeSantis TZ, Hugenholtz P, Keller K, Brodie EL, Larsen N, et al. (2006) NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Res 34: W394–W399.
  39. 39. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, et al. (2009) Introducing mothur: Open-Source, Platform-Independent, Community-Supported Software for Describing and Comparing Microbial Communities. Appl Environ Microbiol 75: 7537–7541.
  40. 40. McDonald D, Price MN, Goodrich J, Nawrocki EP, DeSantis TZ, et al. (2012) An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME J 6: 610–618.
  41. 41. Cole JR, Wang Q, Cardenas E (2009) The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res 37: D141–D145.
  42. 42. Good IJ (1953) The Population Frequencies Of Species And The Estimation Of Population Parameters. Biometrika 40: 237–264.
  43. 43. Ludwig W, Strunk O, Westram R, Richter L, Meier H, et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res 32: 1363–1371.
  44. 44. Lozupone C, Hamady M, Knight R (2006) UniFrac—an online tool for comparing microbial community diversity in a phylogenetic context. BMC Bioinformatics 7: 371–385.
  45. 45. He Z, Xu M, Deng Y, Kang S, Kellogg L, et al. (2010) Metagenomic analysis reveals a marked divergence in the structure of belowground microbial communities at elevated CO2. Ecol Lett 13: 564–575.
  46. 46. Pisciotta JM, Rath DF, Stanek PA, Flanery DM, Harwood VJ (2002) Marine bacteria cause false-positive results in the Colilert-18 rapid identification test for Escherichia coli in Florida waters. Appl Environ Microbiol 68: 539–544.
  47. 47. Edberg SC, Allen MJ, Smith DB, Kriz NJ (1990) Enumeration of total coliforms and Escherichia coli from source water by the defined substrate technology. Appl Environ Microbiol 56: 366–369.
  48. 48. WHO (2006) Guidelines for drinking-water quality. Geneva: World Health Organization.
  49. 49. National Health and Medical Research Council (2004) Guidelines for drinking water quality in Australia. National Health and Medical Research Council (NHMRC & AWRC), Canberra, Australia.
  50. 50. Evans CA, Coombes PJ, Dunstan RH (2006) Wind, rain and bacteria: The effect of weather on the microbial composition of roof-harvested rainwater. Water Res 40: 37–44.
  51. 51. Turkum A, Pekey H, Pekey B, Tuncel G (2008) Investigating relationships between aerosol and rainwater compositions at different locations in Turkey. Atmos Res 89: 315–323.
  52. 52. Ahmed W, Richardson K, Sidhu J, Toze S (2012) Escherichia coli and Enterococcus spp. in rainwater tank samples: comparison of culture-based methods and 23 S rRNA gene quantitative PCR assays. Environ Sci Technol 46: 11370–11376.
  53. 53. Savichtcheva O, Okabe S (2006) Alternative indicators of fecal pollution: Relations with pathogens and conventional indicators, current methodologies for direct pathogen monitoring and future application perspectives. Water Res 40: 2463–2476.
  54. 54. Ahmed W, Sidhu J, Toze S (2013) Faecal Indicators and pathogens in Potable Rainwater tanks in Southeast Queensland. AWA Water 40: 88–92.
  55. 55. Merz J, Dangol PM, Dhakal MP, Dongol BS, Nakarmi G, et al. (2006) Rainfall–runoff events in a middle mountain catchment of Nepal. J Hydrol 331: 446–58.
  56. 56. Sargaonkar A (2006) Estimation of land use specific runoff and pollutant concentration for Tapi River Basin in India. Environ Monit Assess 117: 491–503.
  57. 57. Edwards RA, Rohwer F (2005) Viral metagenomics. Nat Rev Microbiol 3: 504–510.
  58. 58. Tseng CH, Chiang PW, Shiah FK, Chen YL, Liou JR, et al. (2013) Microbial and viral metagenomes of a subtropical freshwater reservoir subject to climatic disturbances. ISME J 7(12): 2374–86.
  59. 59. Amann RI, Ludwig W, Schleifer KH (1995) Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59: 143–169.
  60. 60. Meera V, Mansoor Ahammed M (2006) Water quality of rooftop rainwater harvesting systems: a review. J. Water Supply Res Technol.-AQUA 55 (8) 257–267.
  61. 61. Newton RJ, Jones SE, Eiler A, McMahon KD, Bertilsson S (2011) A guide to the natural history of freshwater lake bacteria. Microbiol Mol Biol Rev 75: 14–49.
  62. 62. Newton RJ, Kent AD, Triplett EW, McMahon KD (2006) Microbial community dynamics in a humic lake: differential persistence of common freshwater phylotypes. Environ Microbiol 8: 956–70.
  63. 63. Liu Z, Huang S, Sun G, Xu Z, Xu M (2012) Phylogenetic diversity, composition and distribution of bacterioplankton community in the Dongjiang River, China. FEMS Microb Ecol 80: 30–44.
  64. 64. Crump BC, Armbrust EV, Baross JA (1999) Phylogenetic analysis of particle-attached and free-living bacterial communities in the Columbia river, its estuary, and the adjacent coastal ocean. Appl Environ Microbiol 65: 3192–3204.
  65. 65. Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, et al. (2004) Environmental Genome Shotgun Sequencing of the Sargasso Sea. Science 304: 66–74.
  66. 66. Burkert U, Warnecke F, Babenzien D, Zwirnmann E, Pernthaler J (2003) Members of a readily enriched beta-proteobacterial clade are common in surface waters of a humic lake. Appl Environ Microbiol 69: 6550–6559.
  67. 67. Simek K, Hornak K, Jezbera J, Masin M, Nedoma J, et al. (2005) Influence of top-down and bottom-up manipulations on the R-BT065 subcluster of β-Proteobacteria an abundant group in bacterioplankton of a freshwater reservior. Appl Environ Microbiol 71: 2381–2390.
  68. 68. Warnecke F, Amann R, Pernthaler J (2004) Actinobacterial 16S rRNA genes from freshwater habitats cluster in four distinct lineages. Environ Microbiol 6: 242–253.
  69. 69. Lemke MJ, Lienau EK, Rothe J, Pagioro TA, Rosenfeld J, et al. (2009) Description of freshwater bacterial assemblages from the upper Parana River flood pulse system, Brazil. Microb Ecol 57: 94–103.
  70. 70. Brierley EDR, Wood M (2001) Heterotrophic nitrification in an acid forest soil: isolation and characterisation of a nitrifying bacterium. Soil Biol Biochem 33: 1403–1409.
  71. 71. Elifantz H, Malmstrom RR, Cottrell MT, Kirchman DL (2005) Assimilation of polysaccharides and glucose by major bacterial groups in the Delaware Estuary. Appl Environ Microbiol 71: 7799–7805.
  72. 72. Glockner FO, Zaichikov E, Belkova N, Denissova L, Pernthaler J, et al. (2000) Comparative 16S rRNA analysis of lake bacterioplankton reveals globally distributed phylogenetic clusters including an abundant group of Actinobacteria. Appl Environ Microbiol 66: 5053–5065.