ABSTRACT
Diverse bacterial assemblages were identified in a large, open stormwater drain (vernacular: nalah) built decades ago in a densely populated suburb of Delhi, India. Illumina-based next-generation sequencing (NGS) of 16S rRNA gene amplicons was conducted with metagenomic DNAs isolated from influent sewage water and sediment samples collected from Sahibabad drain, which now carries domestic and industrial wastes to downstream sewage treatment plants. Results are discussed with respect to diversity and adaptation to unique ecological niche(s) in these drains as well as the prevalence of potentially pathogenic bacteria. Recently, it has become a common practice to cover such drains with thick cement slabs to facilitate the construction of residential/commercial complexes. However, the impact of concrete covers on microbial communities that inhabit the Sahibabad drain is unknown. Results indicated that open drains with better aeration and exposure to sunlight contained microbes that likely enhance biodegradation in sewage water. The deposition of sediments along the course of the drain was dominated by methanogenic and sulphidogenic bacteria. The covering of the drain may have contributed to an increased abundance of anaerobic pathogens which settled in sediments and/or resuspended into sewage water. Such findings are important as the microbes active in sewage can impact public health and drain infrastructure.
HIGHLIGHTS
Characterization of microbiome associated with sediments and sewage water in a large urban drain (‘Nalah’) using NGS methods.
Reduced diversity of bacterial community in sewage water and sediment from covered parts of the drain.
Prevalence of pathogens observed in covered parts of the drain.
Functional characterization of bacterial community to identify potential microbial interactions occurring in the drain.
INTRODUCTION
Rapid urbanization has necessitated improvements in methods of sewage treatment and monitoring of disease-causing microbes (McLellan & Roguet 2019; LaMartina et al. 2021). Before the development of molecular techniques, plate-based counts of culturable pathogenic, coliform bacteria like E. coli, etc. were used to assess the pathogen load in wastewater samples and the efficacy of sewage treatment plants (Dudley et al. 1980). However, it is now evident that environmental samples, like sewage, harbour a rich diversity of microbes that include both culturable and non-culturable bacteria (Solden et al. 2016). In fact, the standard/benchmark bacteria used to assess wastewater samples are typically in the minority and are no longer considered suitable for evaluating sewage treatment plants (Frigon et al. 2013). Molecular methods like next-generation sequencing (NGS) of bacterial metagenomes have become the method of choice for such purposes (Solden et al. 2016). Several reports have provided a deeper understanding of the diversity and functional dexterity of the microbes in sewage samples, as well as the complex dynamics of microbial assemblages responding to aerobic and/or anaerobic digestors, constructed wetlands, etc. used in the treatment of sewage (Wang et al. 2021; Seth et al. 2024). Due to the suitability of metagenomic studies in deciphering the components and roles of microbes, including pathogenic bacteria, this approach has become de rigueur in studies on sewage wastewater and sludge.
In this study, the V3–V4 regions of 16S rRNA genes in complex microbial communities associated with sewage water and sewage sediment samples collected along the Sahibabad drain (nalah) were investigated by NGS. Similar studies have been made in raw sewage, and influent samples in many regions of the world including the United States, China and Brazil (Gao et al. 2016; Nascimento et al. 2018; LaMartina et al. 2021; Tyagi et al. 2023). The Sahibabad drain in western Uttar Pradesh is a part of the larger Ghaziabad drain system. It is approximately 35 km long. According to anecdotal reports, the Sahibabad drain was created in the 1950s from agricultural lands by digging extensive ditches with a width of several meters. It was created by the local urban planning authorities before the extensive development of housing and industrial real estate projects in the area. The locality currently inhabits 30,86,991 people and 396 industries (Nagar Nigam Ghaziabad 2023). It is unclear whether the Sahibabad drain was built as a rainwater conduit or for the movement of sewage to the Indirapuram sewage treatment plant (capacity 74 million litres daily (MLD)) downstream (Nagar Nigam Ghaziabad 2023). The Sahibabad drain passes through agricultural fields, residential complexes of multistoried buildings, offices, hotels, small-scale industries and cattle dairies. Some portions of the drain are covered by cemented blocks for reasons of aesthetics, civic concerns and purposes. The impact of these activities on the microbial communities in the drain is currently unclear. It may be noted that there is a court ruling against covering drains as it results in the generation of toxic gases like methane (Manoj Misra vs Union of India 2015). Therefore, it is important to understand the microbiome of sewage in the covered and open stretches not only to substantiate the court ruling but also for the public health hazard matters. Improved management and treatment of sewage is essential for achieving ‘Sustainable Development Goals’ (United Nations Sustainable Development Goals 2020) which includes reducing the volumes of untreated wastewater and improving water quality globally. Hence, a first-time study was undertaken to explore the microbes and environmental conditions existing in different parts of the Sahibabad urban drain (nalah) to establish a baseline assessment and snapshot of the microbial diversity, including potential pathogenic bacteria present in sewage water as well as sewage sediment deposits. This is important as untreated sewage in the drains contributes inoculum to the final influent for the downstream STP and production of activated sludge. Results obtained were discussed with relevance to microbial communities active in large, old drains and identification of potential health hazards in dealing with sewage management in cities undergoing rapid population and size expansion.
MATERIALS AND METHODS
Site description and sample collection
(a) Outline map of Delhi-NCR with study site (28°39.72′N 77°20.202′E) and (b) map of Sahibabad region showing sites of sample collection.
(a) Outline map of Delhi-NCR with study site (28°39.72′N 77°20.202′E) and (b) map of Sahibabad region showing sites of sample collection.
Isolation of genomic DNAs and 16S rRNA V3–V4 metagenomic sequencing
Total genomic DNAs were isolated in triplicates from each sample using 50 μl sewage water and 0.5 mg sewage sediment each, with a Qiagen Power Viral DNA kit (Cat. No.: 28000-50, Qiagen, Germany) after the vendor's recommendation. Three replicates for each sewage sample were pooled before being processed. The quality of the extracted DNAs was checked using agarose gel electrophoresis. The concentration of the isolated metagenomic DNAs was determined on a NanoDrop spectrophotometer (Thermo Scientific™ Multiskan™ GO Microplate Spectrophotometer). The isolated DNA samples were used for end-point PCR amplification using bacterial and archaeal 16S rRNA primers from the literature (Weisburg et al. 1991). Illumina MiSeq sequencing was used to examine the composition and diversity of the microbial communities in the sewage samples. The V3–V4 regions of the 16S rRNA gene were subsequently targeted using the Illumina MiSeq platform to generate paired-end reads (300 × 2 bp). Approximately 0.5 to 0.7 × 106 paired-end reads were generated per sample. The SRA files for the raw NGS data were submitted to the NCBI SRA database under Bioproject Accession Number PRJNA720111. The raw reads were imported into QIIME2 (Bolyen et al. 2019) ver. 2020.8 and denoised using the DADA2 plugin. Low-quality reads (Phred score < 30) and chimera sequences were removed. The filtered reads were merged and then clustered and aligned for the six samples. To characterize the taxonomic composition of the communities, 8,141 Operational Taxonomic Units (OTUs) were identified at a similarity threshold of 99%. The taxonomic assignment of these OTUs was done using the SILVA138 (Quast et al. 2013) database. Further, the OTU feature tables were used to compute rarefaction curves, and alpha diversity estimates including the Shannon Diversity index (Magurran 2004), and Chao1 index (Chao 1984) using phyloseq, Rarefy and VEGAN packages in R (R Core Team 2021).
Microbial community structure, physicochemical parameters, and multivariate analysis
Since the samples were collected from different parts of the same drain, the total relative abundance of phyla was first estimated using the total number of OTUs belonging to each phylum in all the samples. The relative abundances of microbial phyla were used to compute pairwise correlation among different samples using a data table package in R. These coefficients were calculated and represented as scatterplot matrix created using Hmisc, data table packages in R. Venn diagrams were created using ggplot2 packages in R to represent the number of shared and unique taxa in different samples. The per cent carbon, hydrogen, nitrogen and sulphur profiles in sediment samples were analyzed by CHNS Analyser (Elementar CHNS Analyzer Vario EL III). The levels of trace elements (heavy metals) in sewage water samples from covered and open drains were estimated using Atomic Absorption Spectroscopy (Shimadzu Atomic Absorption Spectrophotometer). The pH, electrical conductivity (EC), total dissolved solids (TDS) and total suspended solids (TSS) for the sewage samples were estimated using a Multiparameter Meter (Hanna Instruments Cat. No.: HI98194). Biological Oxygen Demand (BOD) and Chemical Oxygen Demand (COD) were estimated using the protocols described in literature (Allen et al. 1974; CPCB 2007). The physicochemical parameters including pH, TSS, TDS, COD, BOD, EC and metals were estimated using the sewage water samples. Major elements such as carbon, hydrogen, nitrogen, sulphur and other physicochemical properties were estimated using the sewage sediments. The differences in the community structures of different samples were analyzed using Principal Coordinate Analysis (PCoA) with Bray–Curtis distance estimates computed in R (R Core Team 2021). Canonical Correspondence Analysis and Redundancy Discriminant Analysis were done using environmental variables (pH, TDS and EC) and community composition data at various taxonomic levels in Canoco5 (ter Braak & Šmilauer 2012) to compute the Euclidean distances between the community in different samples. Parameters available for both sewage water and sediment sewage samples were included.
Functional annotation of microbial taxa
Different microbial taxa in the community were functionally annotated using metabolic information from MACADAM (Le Boulch et al. 2019) and MiDAS4 (Dueholm et al. 2022). The main functional groups studied were (i) hydrolytic bacteria, (ii) methanogenic bacteria and (iii) sulphidogenic bacteria. OTUs in each functional group were selected based on their roles described in databases and literature. The relative abundance of reads resembling probable functional OTUs of each group was calculated and compared among different samples as described in the literature (Selvarajan et al. 2018; Shi et al. 2018). The functional information inferred for different microorganisms was examined in different sewage water and sediment samples from open and covered parts of the drain and their likely roles in the sewage.
Detection of the pathogenic taxa
The identification of putative pathogens in the samples was done using a 16SPIP pipeline (Miao et al. 2017) and published reports in the literature. The test reports generated using 16SPIP for the sewage samples were summarized as ‘Abundance’ and ‘Diversity’ of pathogens in the samples. For pathogen detection using 16SPIP, the raw FASTQC files were imported into the 16SPIP pipeline. After the quality check and trimming of poor reads, the paired-end reads were merged using PEAR within 16SPIP. The merged reads were then aligned to the 16S reference database in ‘Sensitive Mode’ for pathogen detection. Subsequently, reads were mapped to the reference pathogen taxa at the species level (>99% identity), the genus level (>97%) and at the family level (>95%). The output test reports were generated and imported into MS-Excel 2013 for further calculations. Analysis of each sample was done individually to reduce the computational requirements. The abundance and diversity of putative pathogenic taxa in a sample were calculated as follows:
The abundance of putative pathogens in a sample = (Total number of reads matching to pathogenic taxa) × 100/ Total number of reads; and Diversity of putative pathogens in a sample = (Total number of pathogenic taxa identified) × 100/ Total number of taxa identified.
Network analysis of microbial community
Spearman's correlation coefficient values between the relative abundances of phyla present in the microbial community of the six samples were computed in R. The results are represented as a heatmap created using ggplot2. A correlation network based on Spearman's correlation coefficients was constructed with OTUs (Relative abundance > 0.001) in six sewage samples based on coefficients (p-value = 0.05) using microeco, and mecodev packages (Liu et al. 2021) in R. The optimal correlation cutoff was found to be at 0.43. Chordplot representation for the Spearman correlation network was generated for different taxonomic levels using the circlize package in R.
RESULTS AND DISCUSSION
Analysis of diversity estimates in sewage microbial communities
Details of NCBI SRA accession numbers; raw and filtered read counts obtained after 16S V3–V4 metagenome sequencing; total number of clustered OTUs and estimates of diversity indices for sewage samples
S. No. . | Sample name . | NCBI SRA accession numbers . | Raw reads . | Filtered reads . | Denoised reads . | Merged reads . | Non–chimeric reads . | Number of OTUs . | Shannon . | Chao1 . |
---|---|---|---|---|---|---|---|---|---|---|
1 | CSL1 | SAMN18633868 | 705564 | 429735 | 418103 | 388301 | 373237 | 2424 | 6.05 | 2424.23 |
2 | CSL2 | SAMN18633869 | 1639522 | 612692 | 575529 | 502648 | 466243 | 3529 | 6.45 | 3529.34 |
3 | OSL4 | SAMN18633870 | 811136 | 493669 | 477917 | 435164 | 413889 | 3540 | 6.70 | 3540.60 |
4 | CSS1 | SAMN18633871 | 984246 | 368484 | 348053 | 309670 | 295436 | 2722 | 6.39 | 2722.27 |
5 | CSS2 | SAMN18633872 | 1729471 | 635957 | 604192 | 539070 | 508119 | 3633 | 6.44 | 3634.54 |
6 | OSS4 | SAMN18633873 | 1061216 | 398576 | 375471 | 330809 | 314729 | 3111 | 6.38 | 3111.82 |
S. No. . | Sample name . | NCBI SRA accession numbers . | Raw reads . | Filtered reads . | Denoised reads . | Merged reads . | Non–chimeric reads . | Number of OTUs . | Shannon . | Chao1 . |
---|---|---|---|---|---|---|---|---|---|---|
1 | CSL1 | SAMN18633868 | 705564 | 429735 | 418103 | 388301 | 373237 | 2424 | 6.05 | 2424.23 |
2 | CSL2 | SAMN18633869 | 1639522 | 612692 | 575529 | 502648 | 466243 | 3529 | 6.45 | 3529.34 |
3 | OSL4 | SAMN18633870 | 811136 | 493669 | 477917 | 435164 | 413889 | 3540 | 6.70 | 3540.60 |
4 | CSS1 | SAMN18633871 | 984246 | 368484 | 348053 | 309670 | 295436 | 2722 | 6.39 | 2722.27 |
5 | CSS2 | SAMN18633872 | 1729471 | 635957 | 604192 | 539070 | 508119 | 3633 | 6.44 | 3634.54 |
6 | OSS4 | SAMN18633873 | 1061216 | 398576 | 375471 | 330809 | 314729 | 3111 | 6.38 | 3111.82 |
Rarefaction plot of the OTUs clustered in six sewage water and sediment samples with number of reads computed using R.
Rarefaction plot of the OTUs clustered in six sewage water and sediment samples with number of reads computed using R.
Analyses of taxonomic diversity in sewage microbial communities
(a) Histogram showing the relative abundance of microbial phyla found in the Sahibabad drain as a percentage of total reads belonging to each phylum in all the sewage samples and (b) scatterplot matrix computed for showing relationship among relative abundance estimates of microbial phyla in all the samples using R. The composition of bacterial communities of each sewage sample is shown with red bars in diagonal boxes from top left to bottom right. Relative abundance of phyla from different samples are plotted against each other in the individual scatterplots. The X-axis and Y-axis are adjusted to bins for relative abundance of phyla in the sample along the corresponding axes.
(a) Histogram showing the relative abundance of microbial phyla found in the Sahibabad drain as a percentage of total reads belonging to each phylum in all the sewage samples and (b) scatterplot matrix computed for showing relationship among relative abundance estimates of microbial phyla in all the samples using R. The composition of bacterial communities of each sewage sample is shown with red bars in diagonal boxes from top left to bottom right. Relative abundance of phyla from different samples are plotted against each other in the individual scatterplots. The X-axis and Y-axis are adjusted to bins for relative abundance of phyla in the sample along the corresponding axes.
At the genus level classification, based on the SILVA138 database, a core community of microbes was shared by all the sewage wastewater samples (545 genera) and sewage sediment samples (609 genera). Many classical genera known to be associated with sewage like Arcobacter, Acinetobacter, Aeromonas and Trichococcus were present in all the samples from the Sahibabad drain (McLellan & Roguet 2019). The total number of genera estimated in sewage samples collected from both open-drain sites was comparable. In the waste water samples from the open-drain sites, CSL2 had 851 taxa at genus level, whereas OSL4 contained 825 microbial genera. The trend was similar in the sediment samples with 874 genera in CSS2 and 862 in OSS4. The number of unique taxa at the genus and species levels was the lowest in the samples from the covered drain, CSL1 (Genus: 36, Species: 80) and CSS1 (Genus: 39, Species: 94), implying that covering of drains resulted in lowered diversity of microbial communities detected in the sewage wastewater and sediment samples. The wastewater sample CSL1 from the covered portion of the drain shared a few genera with open-drain samples CSL2 (74) and OSL4 (24). However, the wastewater samples from open drains (CSL2 and OSL4) shared a substantially larger number of genera (134) with each other. The wastewater and sediment samples from the covered drain site also shared fewer genera with each other (CSL1 and CSS1: 5) than the samples from the adjoining open-drain site (CSL2 and CSS2: 54). Similar trends were seen at the species level.
Diversity of major and minor phyla in sewage microbial communities
(a) Histogram(s) showing the relative abundance of major microbial phyla (relative abundance > 1%) in the sewage water (CSL1, CSL2, OSL4) and sediment samples (CSS1, CSS2, OSS4), (b) Stacked histogram comparing the abundance of minor microbial phyla (Relative abundance < 1%) in the sewage samples. The abundance of each phylum is shown here as a percentage of total number of reads belonging to minor phyla. Solid coloured circles indicate abundance of the corresponding minor phylum detected in different sewage samples.
(a) Histogram(s) showing the relative abundance of major microbial phyla (relative abundance > 1%) in the sewage water (CSL1, CSL2, OSL4) and sediment samples (CSS1, CSS2, OSS4), (b) Stacked histogram comparing the abundance of minor microbial phyla (Relative abundance < 1%) in the sewage samples. The abundance of each phylum is shown here as a percentage of total number of reads belonging to minor phyla. Solid coloured circles indicate abundance of the corresponding minor phylum detected in different sewage samples.
Despite their low relative abundance, a large diversity of minor phyla was observed in sewage samples (Figure 4(b)). The minor phylum Methylomirabilota showed high prevalence in samples from open drains, OSS4 (0.05%) > OSL4 (0.013%) > CSS2 (0.006%), CSL2 (0.004%) > CSS1 (0.002%) > CSL1 (0.001%). The phylum Cyanobacteria was also found in samples from open drains, OSL4 (0.19%) > CSL2 (0.11%), OSS4 (0.11%) > CSS2 (0.08%) while in samples from the covered part, it showed lower abundance (CSL1 – 0.06%; CSS1 – 0.05%). Several minor phyla including Deferribacterota (0.006%), Bdellovibrionota (0.114%), Crenarchaeota (0.006%), and Zixibacteria (0.022%) were abundant in the open-drain wastewater sample, OSL4. Archaeal phyla Euryarchaeota (CSS2 – 0.39%), Crenarchaeota (OSS4 – 0.05%) and Thermoplasmatota (CSS1 – 0.071%) were more prevalent in sediment samples. The Aerophobota, Deferrisomatota, Halanaerobiaeota and NB1-j phyla were exclusively found in sediments from this study indicating that sediment samples harboured a wide diversity of minor phyla.
The nuanced, taxonomic diversity of microbial communities observed in the sewage drain samples suggested a gamut of functional diversity inherent in the resident microbes. Among the major phyla in the dataset from this study, Proteobacteria, Bacteroidota and Firmicutes are established fermenting bacteria (Jin et al. 2018). Members of Desulfobacterota are involved in the reduction of sulphur-cycling intermediates, iron-oxidation, and aromatic hydrocarbon degradation (Murphy et al. 2021). Another major bacterial phylum Campilobacterota which showed lower prevalence in CSL1 and CSS1 samples (Figure 4(a)) can exhibit aerobic chemoheterotrophy, arsenate respiration and nitrogen fixation (Van Der Stel & Wösten 2019). Contrary to this, phylum Synergistota which had the highest relative abundance in samples from the covered drain (Figure 4(a)) includes taxa that are anaerobic, opportunistic pathogens and/or notorious for their function in fermentation reactions (Bhandari & Gupta 2012). Similar to the results of this study, the relative abundances of major microbial phyla are known to be variable in the sewage samples collected from different parts of the same drain (Jin et al. 2018). The functional information available for different minor phyla like Euryarchaeota, Bdellovibrionota, Crenarchaeota, Zixibacteria, and Gemmatimonadota in the MACADAM database indicated the occurrence of chemoheterotrophic bacteria which were abundant in open-drain sewage water and sediments (Figure 4(b)). Microbial phyla which were unique to sediment samples in this study have previously been detected in marine sediment-associated microbiomes (Liu et al. 2022). Various major and minor phyla observed in this study have also been observed in microbial communities reported previously in sewage samples (Jin et al. 2018; Tyagi et al. 2023). Oxygenic photoautotroph Cyanobacteria was the least prevalent in CSL1 (0.06%) and CSS1 (0.05%) samples from the covered drain site. Its abundance was more than ten times at the covered drain site as compared to sewage water and sediments from the open-drain site (Figure 4(b)). Functional annotation of other minor phyla using the MACADAM database showed members like Dadabacteria that can degrade organic matter, specifically microbial peptidoglycans and phospholipids. Some members of Aerophobota have fermentative and saccharolytic metabolisms (Liu et al. 2022). These taxa were more abundant in open drains and preferentially associated with sediment samples (Figure 4(b)). Samples from the open-drain site showed higher abundance of the phyla Cyanobacteria, Bacteroidota, Chloroflexi, Planctomycetota and Acidobacteriota (Figure 4). Factors like exposure to sunlight and air in open drains may offer a conducive environment for oxidation and phototrophic processes carried out by these taxa (Sepúlveda-Muñoz et al. 2023). Since the diversity of microbial taxa in drains is known to influence the natural biodegradation potential of sewage (Hernandez-Raquet et al. 2013), it is possible that enhanced degradation of complex compounds and toxicants in sewage is facilitated by the bacterial diversity in open drains.
Sewage microbial community and environmental factor analyses
(a) Principal coordinate analysis shows microbial community structure of different sewage samples obtained using R. (b) Redundancy analysis at phyla level with environmental variables in different sewage water (solid circles) and sewage sediment samples (solid squares). Samples from covered drain are shown in blue colour and open-drain samples are shown in red.
(a) Principal coordinate analysis shows microbial community structure of different sewage samples obtained using R. (b) Redundancy analysis at phyla level with environmental variables in different sewage water (solid circles) and sewage sediment samples (solid squares). Samples from covered drain are shown in blue colour and open-drain samples are shown in red.
RDA was also performed at the phylum level to determine relationships between bacterial community composition and environmental factors (Figure 5(b)). In the RDA analysis, pH, TDS, and EC (Figure 5(b), red arrows) formed acute angles, demonstrating that these parameters showed a close association. It can also be interpreted that pH had the most effect on the microbial populations in Sahibabad drain samples, followed by TDS and EC. Similar observations were revealed in CCA done with microbial communities from this study at the level of taxonomic families. Sewage sediment microbial communities were grouped away from the environmental factors. This suggested that change in environmental variables (analyzed here) corresponded to the dissimilarity in the composition of the sewage wastewater microbial communities detected in the Sahibabad drain. The community structures for sewage water and sediment samples were also considerably different. This observation confirmed the PCoA.
Differences in physicochemical characteristics are factors that contribute to differences between the microbial communities in wastewater and sediments (Lu et al. 2016). In this study, the pH of sewage water samples ranged from 7.75 ± 0.14 in CSL2 to 7.88 ± 0.02 in CSL1. TSS measured in CSL1 (15,930 ± 907 mg/L) was more than ten times than that present in CSL2 (1,450 ± 180.28 mg/L) and OSL4 (1,400 ± 52.92 mg/L). COD for CSL1(464 mg/L) was four times that in sewage water from adjacent open drain CSL2 (100 mg/L). BOD levels at CSL1(120 mg/L) were higher than in CSL2 (95 ± 15 mg/L). They were the highest from the industrial influent drain sample OSL4 (150 ± 15 mg/L) collected from Sampling Site II. The TDS and EC values showed similar trends OSL4 (TDS: 1,781 ± 73.37; EC: 3,565 ± 307.69) < CSL1 (TDS: 1,942 ± 167.01; EC: 3,833 ± 263.54) < CSL2 (TDS: 2,187 ± 180.85; EC: 4,379 ± 228.96). In this study, microbial communities of CSL1 and OSL4 samples were proximal in RDA and CCA since their physicochemical measurements, especially pH, BOD and EC showed similar trends (Figure 5). It was also observed that the sediments showed less TDS and EC as compared to sewage water. The levels of elements were very low in open-drain sediments CSS2 (carbon: 6.38%, hydrogen: 1.14%, nitrogen: 0.57%, sulphur: 1.7%) and OSS4 (carbon: 2.87%, hydrogen: 0.69%, nitrogen: 0.28%, sulphur: 0.44%). The CSS1 sample (covered drain) showed the highest percentage of carbon (21.86%), nitrogen (1.45%), hydrogen (2.89%) and sulphur (3.45%). Traces of heavy metals, like copper (0.01–0.03 ppm) and zinc (0.10–0.37 ppm), were also detected in sewage water samples collected from the Sahibabad drain, while levels of cadmium, chromium and lead were negligible. Thus, the measurements of the physicochemical properties of sewage water and sediments from covered and open drains indicated that COD, TSS, BOD and sulphur levels were higher in the CSL1 and CSS1 than in the open drains. Lower levels of BOD and COD in open-drain samples indicated lower concentrations of pollutants in open drains while increased TSS and EC in sewage water samples indicated a high contaminant load. This was important as the availability of substrates may influence the microbial groups present in the sewage samples from the Sahibabad drain. Consistent with the result of this study, previous reports have also shown that even slight changes in pH can reflect major effects on the bacterial communities in different environments (Nascimento et al. 2018). Variations in levels of COD and pH estimates have been employed to explain contrasts in the bacterial community structure of wastewater sludge systems (Gao et al. 2016). The differences in microbial community structure and environmental conditions observed in this study, albeit with a limited number of samples, likely reflected the varied ecological niche of the covered drain site. Similar results with an increase in pollutant levels after covering agricultural drains have been reported previously (Moghazy et al. 2007).
Functional diversity of genera in sewage microbial communities
Relative abundance of sequence reads indicating putative functional OTUs belonging to bacterial genera involved in (a) hydrolytic and fermentation, (b) sulphidogenic, and (c) methanogenic processes.
Relative abundance of sequence reads indicating putative functional OTUs belonging to bacterial genera involved in (a) hydrolytic and fermentation, (b) sulphidogenic, and (c) methanogenic processes.
On the other hand, OTUs resembling taxa involved in carbon and sulphur metabolism were more abundant in sediment samples. For instance, members of Desulfobacterota dominated the community of all sediment samples from the Sahibabad drain. Sulphur-reducing genera Desulfomicrobium, Smithella, Syntrophobacter and Unclassified members of Desulfobacteraceae, were most abundant in covered drain samples CSL1 and CSS1. The sulphidogenic community members were more abundant in sediment samples with respect to their corresponding sewage waters except in CSS1 (Figure 6(b)). CSL1 from the covered drain site showed the highest abundance of sulphidogenic community in all sewage waters. Likewise, methanogenic genera were more abundant in sewage sediments from all three sites as compared to sewage water (Figure 6(c)). Methanosaeta formed 2–4% of genera present in all samples from the Sahibabad drain. Methane-producing genera from archaeal phyla Methanobacterium, Methanolinea, Methanospirllium were prevalent in samples CSS1, CSS2 and OSS4 as compared to the sewage water from corresponding sites. In brief, the functional analysis of various taxa revealed the prevalence of genera involved in hydrolysis in open-drain sewage water samples and the methanogenic and sulphidogenic community dominated the sewage sediments in the drain.
Desulfomicrobium, Macellobacteroides, Methanosaeta, Smithella and Leptolinea were the most dominant bacterial taxa in the common core of the microbial community shared by all the sewage samples from the Sahibabad drain (Figure 6). Bacterial genera like Hypnocyclicus, Thauera, Geobacter, Trichlorobacter, and Azoarcus are known to be involved in the degradation of complex pollutants in sewage (Jin et al. 2018; Selvarajan et al. 2018). In this study, their abundance was higher in sewage water samples from open-drain sites (Figure 6(a)), suggesting efficient capabilities for natural/microbial degradation of sewage. Acidogenic and hydrolytic bacteria that were seen in sewage water samples from open-drain sites (Figure 6(a)) are known to enhance biodegradation and ameliorate pollutant loads, releasing by-products and elements that get deposited as concentrated sediments (Shi et al. 2018). The ability to identify and assign putative functions to microbes in sediments is critical as the release of corrosive methane and sulphide gases can deteriorate the drainage system (Eijo-Río et al. 2015). Hydrogen sulphide is a common by-product for sulphur-reducing bacteria. Production and release of hydrogen sulphide is more when oxygen is depleted (Eijo-Río et al. 2015). Its production would be higher in covered drains (Figure 6(b)) causing odour problems and corrosion of pipes. Blocking of drains by bulking of sediments would create unhygienic conditions. Moreover, damaged conduits can lead to contamination of water table and surface water with sewage. Such considerations suggest that covering old drains may be undesirable. Further research with a larger number of sewage samples from covered portions along the Sahibabad drain is required to investigate ramifications.
Co-occurrence patterns of microbial taxa in sewage microbial communities
Spearman's correlation coefficients for relative abundances of major and minor phyla in the microbial community of six sewage water and sediment samples. The correlation coefficients are coloured from red (positive correlation) to blue (negative correlation).
Spearman's correlation coefficients for relative abundances of major and minor phyla in the microbial community of six sewage water and sediment samples. The correlation coefficients are coloured from red (positive correlation) to blue (negative correlation).
Chordplot representation of the network of co-occurring bacterial taxa at the (a) phylum and (b) family level in sewage samples from this study. Network of co-occurring bacterial phyla in sewage water and sewage sediment samples was created using microeco package in R and Gephi with OTUs (RA > 0.001). The network was calculated based on Spearman correlations (p-value = 0.05, correlation cutoff = 0.43). Each sector of the circle represents one phylum. The width of each link represents the number of associations between the linked bacterial phyla. Families belonging to a phylum are denoted with circles of the same colour.
Chordplot representation of the network of co-occurring bacterial taxa at the (a) phylum and (b) family level in sewage samples from this study. Network of co-occurring bacterial phyla in sewage water and sewage sediment samples was created using microeco package in R and Gephi with OTUs (RA > 0.001). The network was calculated based on Spearman correlations (p-value = 0.05, correlation cutoff = 0.43). Each sector of the circle represents one phylum. The width of each link represents the number of associations between the linked bacterial phyla. Families belonging to a phylum are denoted with circles of the same colour.
The archaeal family Methanosaetaceae which is involved in acetoclastic and conventional methanogenesis was also associated with Anaerolineaceae (Figure 8(b)). Methanogens need such fermenting microorganisms for the production of their substantial metabolic educts (Morris et al. 2013). Likewise, Anaerolineaceae also showed an association with Bacteroidetes-Vadin HA17 which is known to be involved in rice straw degradation and methane production in degraded soils (Wei et al. 2019). The relative abundance of these methanogenic microbes was higher in sewage sediments as compared to sewage wastewater samples in this study (Figure 6(c)). Another interesting observation was that many taxa forming multiple associations (Desulfomicrobiaceae, Anaerolineaceae and Christensenellaceae) were the most abundant in sewage samples from covered drains (CSL1 and CSS1). Given that the metabolic interactions by these microbes are reported from environments with energetic constraints and anoxic conditions (Morris et al. 2013), it may be conjectured that due to the scarcity of oxygen as the electron acceptor, these anaerobic chemoheterotrophic taxa associate with fermenting microbes to complete degradation of large, complex compounds in covered drains which may be otherwise not possible. However, in open drains, the occurrence of such anaerobic microbes and their need for such associations is reduced. Similar interactions have been reported previously among microorganisms from complex environments (Morris et al. 2013; Wang et al. 2021; Üstüntürk-Onan et al. 2024). Further research with more samples is required to validate this correlation and inferred interactions within sewage microbial communities from the Sahibabad drain.
Detection of pathogenic microbial taxa
Histograms showing the (a) abundance and (b) diversity of pathogenic taxa present in the sewage water and sediment samples from this study identified using 16SPIP.
Histograms showing the (a) abundance and (b) diversity of pathogenic taxa present in the sewage water and sediment samples from this study identified using 16SPIP.
Pathogenic taxa associated with Anaerolineae, Dehalococcoidia, and Thermomicrobia classes from Chloroflexi have been linked to bacterial vaginosis, periodontitis, peri-implantitis, chronic kidney disease, and COPD exacerbations (Campbell et al. 2014) were abundant in samples from the covered drain site. Bifidobacterium and Collinsella are known opportunistic pathogens and are linked to dysbiosis-related disorders (Bag et al. 2017). These bacteria can modify the host bile acids to modulate the virulence and pathogenicity of enteric pathogens. Bifidobacterium had the highest abundance in CSL1 (1.44%). Collinsella aerofaciens was less abundant in CSL2 (0.72%) and CSS2 (0.81%) samples (open-drain site) as compared with samples CSL1 (0.86%), CSS1 (0.87%) from the adjacent closed drain site. Species of Desulfomicrobium are related to human periodontal disease and chronic inflammation (Kushkevych et al. 2021). This genus showed the highest abundance in sample CSL1 followed by CSS1 from the covered drain site. Gemmiger formicilis, known to play a role in the relapse of Crohn's disease (Ma et al. 2022), was more abundant in samples from the covered drain site than in open drains. Among all sewage water samples, Eubacterium tenue which is associated with bacteraemia (Lau et al. 2004) had the highest abundance in the CSL1 (1.09%) sample. Some methanogens observed to be more prevalent in samples from covered drains have been associated with oral and vaginal infections, obesity and digestive tract diseases (Khelaifia et al. 2013). An abundance of pathogens in sediments is a cause of public health concern as the concentration and abundance of pathogens increase when sediments are resuspended under the action of settling velocity of sediments, flow in the water column and other related factors (Wu et al. 2009). Results from this study also indicated that the sewage in the Sahibabad drain had pathogenic taxa distributed in sewage wastewater and sediments from open and covered drain sites (Figure 9). The pathogenic microbes are likely to be present in sewage wastewater that enters the downstream STPs and surface water, therefore, improved methods for the detection and quantification of potential pathogens in untreated sewage become extremely important.
CONCLUSION
Complex microbial communities were revealed in sewage wastewater and sediment samples from the Sahibabad drain. Analyses of sequence data from V3–V4 regions of 16S rRNA genes indicated taxonomic diversity and metabolic dexterity of constituents of the microbiome from sewage. Sewage waste water from open-drain sites contained microbes involved in the breakdown of complex substrates, while both sewage wastewater and sediment samples from the covered drain site contained dominant methanogenic and sulphidogenic microbes. Distinct co-occurrence patterns were observed in networks of microbial groups based on Spearman correlation coefficients which could be attributed to shared, tandem and/or complementary metabolic pathways facilitating resource utilization in the sewage environment. Potential pathogens were identified throughout the drain, whose prevalence in sediments and/or resuspension from sewage sediment into sewage wastewater could pose public health concerns. Anaerobic pathogens were dominant in samples from the covered drain site while sulphide-releasing bacteria were observed in all sediment samples from both covered and open-drain sites. The covered drain site examined in this study had lower microbial diversity that would reduce the potential for natural biodegradation of sewage, while open-drain sites contained microbial communities that were more likely to enhance the biodegradation of large substrates and biogeochemical nutrient cycling. Results from this study, albeit with a limited number of samples, emphasize the need for monitoring microbial communities in large drains for detection of pathogens and efficient sewage management; reduced deposition of sediments along the drain and use of alternatives for cemented drain covers. Long-term research is needed in order to understand the processes, seasonal fluxes and rates of natural, microbe-driven breakdown of sewage that accumulate from different influents. Such studies can also impact policies regarding emissions of noxious gases and unpleasant odours; restricting human/livestock mobility near large, open drains, as well as occupational health and safety of municipal workers engaged in cleaning and maintenance work under closed portions of large sewage drains. Cohesive efforts for monitoring sewage microbial communities with the cooperation of government and civic agencies will improve the sustainability of traditional Nalahs like the Sahibabad drain in densely populated urban areas.
ACKNOWLEDGEMENTS
The Ghaziabad Municipal Corporation, Uttar Pradesh, India, is thanked for funding the metagenomic analyses of drain water and sediment samples. MB acknowledges a PhD-ship (Senior Research Fellowship Ref File No.- 09/045/(1775)2020-EMR-I) from the Council for Scientific and Industrial Research, Government of India. The Delhi University Scientific Instrumentation Centre is thanked for assistance with physicochemical analyses. Members of the Plant Biotic Interactions Lab are thanked for their help during the experiments.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.