- Email: [email protected]
Shifts in microbial community in response to dissolved oxygen levels in activated sludge
Bioresource Technology 165 (2014) 257–264
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Shifts in microbial community in response to dissolved oxygen levels in activated sludge Trilok Chandra Yadav, Anshuman A. Khardenavis, Atya Kapley ⇑ Environmental Genomics Division, National Environmental Engineering Research Institute (CSIR-NEERI), Nagpur 440020, India
h i g h l i g h t s 1
Wastewater treatment was analyzed at three DO levels; 1, 2 and 4 mg l
.
Degradative efficiency was observed to vary between 60% and 65% at 4 ppm DO. Shifts in bacterial diversity were compared across different DO. Analytical and bioinformatics tools were used to assess degradative capacity.
a r t i c l e
i n f o
Article history: Received 10 January 2014 Received in revised form 28 February 2014 Accepted 3 March 2014 Available online 12 March 2014 Keywords: Activated biomass Amplicon library Dissolved oxygen Microbial diversity Wastewater treatment
a b s t r a c t This study evaluates the degradative efficiency of activated biomass collected from a Common Effluent Treatment Plant (CETP) under three different dissolved oxygen (DO) levels, 1, 2 and 4 mg l1. The change in bacterial diversity with reference to DO levels was also analyzed. Results demonstrate that degradative efficiency was the highest, when the reactor was maintained at 4 mg l1 DO, but amplicon library analysis showed a greater diversity of bacteria in the reactor maintained at 2 mg l1 DO. Bacteria belonging to the order Desulfuromonadales, Entomoplasmatales, Pasteurellales, Thermales and Chloroflexales have only been detected in this reactor. Ammonia and nitrate levels in all three reactors indicated efficient nitrification process. Results of this study offer new insights into understanding the performance of activated biomass vis-à-vis microbial diversity and degradative efficiency with reference to DO. This information would be useful in improving the efficiency of any wastewater treatment plant. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction The most common aerobic strategy for treatment of wastewater is the activated sludge process. Activated Sludge comprises of highly complex microbial biomass made of eukaryotes, bacteria, archaea, and viruses, in which bacteria are dominant and play an important role in the removal of organic pollutants (Rani et al., 2008; Yu and Zhang, 2012; Winkler et al., 2013). Parameters related to bacterial growth and degradation are mainly, the availability of nutrients and dissolved oxygen (DO). DO is a relative measure of the amount of oxygen that is dissolved in wastewater, and it usually fluctuates seasonally and varies with water temperature and altitude. DO levels govern the rate of degradation of the organics in aerobic growth physiology of microbial communities in any wastewater treatment plant (WWTP) and ⇑ Corresponding author. Address: Environmental Genomics Division, National Environmental Engineering Research Institute (CSIR-NEERI), Nehru Marg, Nagpur 440020, India. Tel.: +91 712 2249883. E-mail address: [email protected] (A. Kapley). http://dx.doi.org/10.1016/j.biortech.2014.03.007 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
also contribute to operational costs (Kapley et al., 2001; Wells et al., 2009). Different levels of DO can also affect the nitrification, denitrification, besides, aerobic reactions, behavior and activity of heterotrophic and autotrophic microorganisms. A low DO adversely affects treatment efficiencies of bulk wastewater volumes, especially in industrial effluents, coming from diverse sources. On the other hand, an excessive DO concentration will lead to unnecessary power consumption. In the aeration tank, the required aeration depends on the actual oxygen demand and the oxygen transfer efficiency. The actual oxygen demand is determined by the amount of pollutants oxidized and biomass produced, while the oxygen transfer efficiency is related to aeration devices, operational DO, temperature, sludge property, and MLSS concentration. Understanding the bacterial diversity present in a wastewater treatment plant and analyzing community shifts with change in operational parameters, is the key to develop successful treatment strategies. There are a large number of reported methods to study bacterial diversity; amplification and sequencing of 16S rRNA gene, Random Amplified Polymorphic DNA analysis (RAPD); (Kapley et al.,
258
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
2007; Li et al., 2012; Sharma et al., 2013), T-RFLP (Felföldi et al., 2010), PCR-DGGE (Lyautey et al., 2005; Yao et al., 2010; Liu et al., 2012, 2013). Each of these has been used successfully, but next generation sequence technologies have widened the scope of data generation and allow us to view this picture in greater detail. This study explores the change in microbial community profile at different DO levels in activated biomass collected from a Common Effluent Treatment Plant (CETP) and compares the change in treatment efficiency with change in DO. The microbial diversity of activated biomass was analyzed under three different DO levels to answer the following questions; (1) How much does the wastewater treatment efficiency vary with change in DO? and (2) How does DO affect the microbial community of the activated biomass? Keeping all operational parameters and nutrient levels constant, COD, ammonia and nitrate levels were monitored in all three reactors as a measure of treatment efficiency. 16S amplicon library was prepared and taxonomic analysis carried out to analyze community shifts based on DO variations. The distribution of bacterial phyla under different operational conditions can be linked to performance of wastewater treatment plant. This study aims to generate a knowledge base that can be used to enhance treatment efficiency in industrial wastewater treatment and define optimum DO conditions that correlate to maximum degradative efficiency. 2. Methods 2.1. Source of wastewater and activated sludge Wastewater and activated biomass used in this study was collected from a CETP, treating wastewater from pharma, chemical and dye industries. This CETP runs at a capacity 2500–3000 m3/ Day. Sampling of activated biomass was carried out from the aerator unit and both the biomass and wastewater (inlet) was immediately brought to the laboratory within 12 h. 2.2. Reactor design and setup The activated biomass collected from the CETP was added into three reactors having total volume of 5 L and working volume of 3 L each. The addition of biomass in each reactor was 3000 mg l1, maintaining the MLSS (mixed liquor suspended solids) concentration being operated at the CETP. Air inlet into the reactors was maintained with the help of a rotameter and the reactors were run at 1 mg l1, 2 mg l1 and 4 mg l1 DO and will be henceforth referred to as Reactor A, Reactor B and Reactor C respectively. The oxygen concentration in the reactors was monitored using an oxygen meter and probe (Yellow Springs Instrument Co., Model 85). All three reactors were fed with wastewater from the CETP and the HRT (hydraulic retention time) was maintained at 3 days to mimic conditions operating at the CETP. Wastewater was collected in bulk so as to maintain a constant influent throughout the period of study. A schematic diagram of the reactor setup can be seen in Supplementary Fig. 1. 2.3. Analytical methods COD, ammonia and nitrate were monitored every three days in keeping with HRT; samples were collected before the removal of treated wastewater and after addition of fresh wastewater. COD ammonia and nitrate were monitored for both influent and treated wastewater (effluent). 5 ml sample from each reactor was collected in triplicates and centrifuged at 7000g. The supernatant was filtered through 0.25 lm membrane filter. 500 ll of filtered sample was used to detect COD using COD detection kit (Merck Germany,
100–1500 mg l1), as per directions of the manufacturer. Levels of ammonia and nitrate were monitored according to APHA standard protocols (American Public Health Association). 2.4. Metagenomic DNA extraction 10 ml activated sludge samples were collected at 0 h and after 2 months from each reactor, centrifuged at 7000g at 4 °C. The samples were washed as per the protocol described earlier (Purohit et al., 2003) and DNA was prepared using the FastDNA SPIN Kit for soil (MP Biomedicals, USA) as per the instructions of the manufacturer. Metagenomic DNA was also prepared from the activated biomass sample that was collected from the CETP, before aliquoting it into the three reactors. This is referred to as ‘control DNA’. The individual DNA extracts were visualized using 1.0% gel electrophoresis, and the DNA concentrations and purities of the extracts were determined by microspectrophotometry (NanoDrop-1000, Thermo Scientific, USA). This sample was used for constructing the amplicon library. 2.5. Next generation sequencing of amplicon library The microbial community of the three reactors and control was identified by amplifying and sequence analysis of the V3 region of 16S rRNA gene from the metagenome. Primers used, target the surrounding conserved region and illumina sequences adapters and dual-index barcodes were added to the amplicon. The primers used to amplify the V3 region were; Forward primer sequence-50 CCTACGGGAGGCAGCAG 30 and Reverse primer sequence-50 ATTACCGCGGCTGCTGG 30 . Libraries were then normalized and sequenced on the MiSeq platform (Illumina). Sequence data was processed by read trimming and identification of V3 sequences, followed by filtering and assigning the operational taxonomic units (OTUs). The reads from filtered OTUs are processed using Quantitative Insights into Microbial Ecology (QIIME) program (Caporaso et al., 2010), to construct a representative sequence for each OTU. The representative sequence was aligned to the Greengenes core set reference databases using PyNAST program. 2.6. Analyzing sequence data using bioinformatics tools 2.6.1. Sequence pre-processing After applying quality parameter, mainly base quality, average base content, and GC distribution in the reads, there were total 7,66,619, 7,13,088 and 7,21,221 pair-end raw reads generated for reactors A, B and C respectively, while the control DNA showed 1,088,944 pair-end raw reads. Further applying specific filters as mentioned in Table 1 and 90–92% reads were obtained and it is used for downstream analysis. 2.6.2. Analysis using MG-RAST In order to study taxonomic abundance in the library, the obtained sequences were analyzed using Metagenomics-Rapid Annotation using Subsystem Technology server (MG-RAST), (Meyer et al., 2008) with default parameters under the accession number 4537850.3, 4545085.3, 4545086.3 and 4545087.3 for the control, reactor A, B and C respectively. Post-processing of pair end sequences are done by MG-RAST using their own QC pipeline, summarized in Supplementary Table 1. 2.6.2.1. Overall comparative taxonomic abundance. Based on the LCA (Lowest Common Ancestor) algorithm implemented in MG-RAST, comparative taxonomic tree study was performed with parameters of 90% identity, minimum alignment length 50 bp for phylum,
259
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264 Table 1 Analysis of reads generated in the amplicon library. The table describes the number of reads and filters used in this study. Sample
Total raw reads (no.)
Passed conserved region filter (no.)
Passed spacer filter (no.)
Passed read quality filter (no.)
Passed mismatch filter (no.)
Control Reactor A Reactor B Reactor C
1,088,944 766,619 713,088 721,221
10,33,681 7,19,639 6,58,215 6,78,097
10,32,933 7,19,014 6,57,656 6,77,486
10,31,861 7,18,951 6,57,544 6,77,420
10,07,153 (92.49%) 7,08,607 (92.43%) 6,43,042 (90.17%) 6,67,199 (92.50%)
class, order, family taxonomic level in all samples by targeting NCBI taxonomy database. 2.6.2.2. Statistical analysis. 2.6.2.2.1. Rarefaction curve and Principal Component Analysis. For analyzing the depth of sequencing performed for each sample to capture the most dominating taxa, rarefaction analysis was carried out using MG-RAST tools. The inter-relationships between the microbial diversity of activated sludge maintained at three different DO levels, compared to the control sample, was visualized using Principal Component Analysis (PCA) that was carried out to assess the variation among the samples using Paleontological Statistics software package for education and data analysis tool (PAST v3.01), (Hammer et al., 2001). PCA allows us to compare and analyze data that has a high degree of correlation within its multiple attributes like taxonomy. 2.6.2.2.2. Heatmap and Shannon index. Microbial abundance was mapped using heatmap modules in ‘‘R’’ statistical package. Shannon index values representing the diversity and evenness of a sample was calculated using with the formula
XR H ¼ i¼1 ðpi ln piÞ where pi is the proportion of characters belonging to the ith type of letter in the string of interest. Shannon’s index was calculated using the PAST v3.01 (Paleontological Statistics Version 3.0) software. 3. Results and discussion 3.1. Treatment efficiency in the reactors as monitored by COD removal The efficiency of the three reactors was analyzed as a measure of COD removal. Initial COD of the wastewater collected from the treatment plant was 3240 mg l1. Conditions being operated at
the CETP were mimicked in the lab with respect to MLSS, HRT and inlet wastewater. Fig. 1 demonstrates the DO dependent COD removal in all three reactors. It can be seen from the figure that Reactor C, operating at 4 mg l1 DO shows the highest efficiency of degradation, maintaining efficiency between 60% and 65% throughout the period of study (60 days). Reactor B, operating at 2 mg l1 DO demonstrates highest efficiency on day 30 and maintained degradative efficiency between 40% and 50%. Degradative efficiency of reactor A was the lowest, ranging from 15% to 25%. This study clearly demonstrates that COD removal rates depend on the dissolved oxygen levels available to the bacteria. Nitrogen levels in the wastewater are also a prime concern; not only do they lead to eutrophication of water bodies, but excessive ammonia in the wastewater influences the degradative capacity and effects microbial growth (Puigagut et al., 2005; Khardenavis et al., 2007; Rajagopal et al., 2013). Hence, the levels of ammonia and nitrate in the three reactors were also monitored. Fig. 2 demonstrates the status of the nitrification process under different DO levels, with reference to the levels of ammonia and nitrate. Removal of ammonia is roughly similar at 1 and 2 mg l1 DO, ranging between 40% and 45% in Reactor A and 45–50% in Reactor B. The highest efficiency of ammonia removal was observed when the DO was 4 mg l1; 70–75% in Reactor C. Ammonia is converted to nitrate in the nitrification process and Fig. 2 demonstrates that nitrate levels in the reactors are synergistic to the ammonia being degraded. This is suggestive of an efficient nitrification process. Table 2 lists the characteristics of both influent (collected at the CETP) and effluent (treated wastewater from the three separate reactors). At the initial stages of the study, a reactor operating at 8 mg l1 DO, was also set up, but the MLSS could not be maintained, hence was discontinued (data not shown). High bubbling rates caused the biomass to splutter and stick on the upper part of the reactor, disintegrating sludge flocs and thus causing drying and increased
Fig. 1. Degradative efficiency of three reactors maintained at different DO levels, as measured by COD removal. The reactors were operated for a period of 60 days and COD has been monitored every 15 days. COD of influent, effluent and treatment efficiency has been denoted in the figure.
260
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Fig. 2. Estimation of ammonia and nitrate measure by APHA protocol for a period of 60 days in each reactor. Removal efficiency of ammonia and percentage increase in nitrate has been represented as percentage.
Table 2 Characteristics of fresh and treated wastewater. Wastewater collected at the CETP and used to run the three bioreactors has been referred to as ‘influent’, while the treated wastewater from the three reactors has been referred to as ‘effluent’. Parameter
Effluent Influent (collected Reactor A Reactor B Reactor C from CETP)
HRT (days) MLSS (mg l1) SRT (days) COD (mg l1) Ammoniacal nitrogen (mg l1) Nitrate nitrogen (mg l1) pH
3 3000 15 3240 70.29 100 8.15
3 3000 15 312 25.65 246.86 7.63
3 3000 15 251 22.56 569.42 7.58
3 3000 15 142 12.53 588.22 7.59
⁄ Influent wastewater was collected and transported from CETP premises to the lab within 24 h. (Reactor A 1 mg l1 DO, Reactor B 2 mg l1 DO, Reactor C 4 mg l1 DO).
evaporation. These observations, coupled with results demonstrated in Fig. 1, suggest 2–4 mg l1 DO as optimum parameter to be maintained at a treatment plant.
3.2. Microbial diversity and richness In order to analyze the effect of DO levels on the microbial community of activated biomass, we compared the phylogenetic profiling of all three reactors with the control sample that represents activated biomass of the CETP. Amplicon libraries were constructed using V3 region primers and high throughput sequencing was carried out on the MiSeq platform. Bacterial 16S ribosomal RNA (rRNA) genes contain nine ‘‘hypervariable regions’’ (V1–V9) that demonstrate considerable sequence diversity among different bacteria. These variable regions are flanked by conserved regions that can be used to construct specific primers (Claesson et al., 2010). However, no single hypervariable region can differentiate among all bacteria and different regions show a better diversity for different groups (Chakravorty et al., 2007). This study did not aim to carry out a complete bacterial profile but to distinguishing all bacterial species to the genus level, which could be done using V3 region (Chakravorty et al., 2007). Hence, the amplicon library data presented here is from the V3 region of the 16S rRNA genes. The relative abundance of microbial diversity analyzed is demonstrated in Fig. 3. The figure shows data from phylum to family level based on relative abundance of microbial population. Only the top 15 taxonomic categories at class level and top 20 at order and family level have been summarized here. The taxonomic classes other than these results are categorized as ‘‘Others’’. The sequences with low similarity or no similarity; or where the V3
regions did not have any alignment hits against taxonomic database have been categorized as ‘‘Unknown’’. Sequences that were unclassified at a particular taxonomy have been labeled as ‘unclassified derived from phylum/order/class etc.’ Sequences that do not show homology in the NCBI database have been labeled as Unclassified. As can be seen in Fig. 3, in all four levels of classification, the unknown and unclassified diversity dominates, comprising of approximately 60% of the population. The phylum Actinobacteria was the next dominant taxonomy observed. The control sample demonstrated a higher percentage of anaerobic bacteria, e.g. Clostridiaceae (Family level) with 2.14% as compared to 0.6/0.7/ 0.8% in reactors A, B and C respectively.
3.3. Changes in microbial community with DO levels Fig. 4 demonstrates the change in bacterial diversity indicating differentiation only at class level while Fig. 5 depicts the taxonomic tree that describes differentiation from phylum to order level for all four amplicon libraries. The tree also demonstrates the abundance at each node. As can be seen in Fig. 4, the control sample shows a different distribution of bacterial classes when compared to the activated biomass being maintained at different DO levels. This figure suggests that alphaproteobacteria is sensitive to oxygen levels. Proteobacteria comprised of a (72.67%, 32.38%, 38.86% and 37.66%), b (11.72%, 47.41%, 37.22% and 38.23%), d (11.36%, 4.45%, 4.14% and 3.45%), c (1.82%, 10.36%, 12.22% and 13.09%) in control DNA, reactors A, B and C respectively. Fig. 5 demonstrates the presence of Acholeplasmatales, a facultative anaerobe. This order could only be detected in the control sample, but with oxygenation, this taxonomy was not observed. This suggests that the CETP was being maintained at very low DO. Rare bacteria, belonging to the order Rubrobacterales and Neisseriales that do not contain many members in the group were observed solely in the reactor operating at 1 mg l1 DO (Reactor A). The reactor operating at 2 mg l1 DO demonstrates the highest microbial diversity and bacteria belonging to the orders, Desulfuromonadales, Entomoplasmatales, Pasteurellales, Thermales and Chloroflexales have only been detected in this reactor. Vibrionales and Chlamydiales are only observed at higher DO levels and not detected at 1 or 2 mg l1 DO levels. What is noteworthy is that many of the bacteria detected here have also been reported in marine environments (Bolhuis and Stal, 2011). This is probably because of the salinity in the treatment plant; contributed to by the total dissolved solids (TDS) enriches the growth of marine bacteria. This study demonstrates the shifts in microbial community with different operational parameters and hence can be correlated to differential treatment efficiency. These observations can be extended in future studies
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
261
Fig. 3. Classification of bacterial diversity at Phylum, Class, Order and Family level of all three reactors in terms of relative abundance. The abundance has been represented in terms of percentage in total effective bacterial sequences in a sample.
Fig. 4. Abundances of different classes in Proteobacteria in all three reactors compared to control sample.
into identification of an indicator population that can be monitored to assess degradative efficiency in treatment plants. The most abundant phylum observed in this study across all reactors and previously reported in activated biomass are Actinobacteria, Proteobacteria, Firmicutes, Bacteroidetes, Deinococcus, Cyanobacteria, Acidobacteria, Nitrospira and Chloroflexi (LeCleir et al., 2004; Zang et al., 2008). Lee et al. (2002) showed the predominance of Proteobacteria in a WWTP with enhanced biological phosphorus removal, with and without nitrogen removal. Two classes of bacteria have been mainly reported in wastewaters containing ammoniacal nitrogen; ammonia-oxidizing Betaproteobacteria and Alphaproteobacteria (Manz et al., 1994; Wagner et al.,
1994; Amann et al., 1996; Kämpfer et al., 1996; Bond et al., 1999; Jiang et al., 2008). Gammaproteobacteria have been reported in previous studies (Nielsen et al., 1999; Crocetti et al., 2002; Liu et al., 2007). This study demonstrates a majority of Alpha and Beta proteobacteria in reactors B and C, while reactor A shows a higher percentage of Betaproteobacteria. At the Class level, Actinobacteria, Bacilli, Clostridia, Cytophaga, Deinococcus, Flavobacteria, Gammabacteria, Solibacteres and Actinomycetales, Clostridiales, Deinococcales, Flavobacteriales, Solibacterials, Coriobacterials dominate at 4 mg l1 DO. Table 3 lists the top 10 dominant genus in each sample. All reads could not be classified under definite taxonomy and hence
262
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Fig. 5. Comparison tree profile of all three reactors upto order level using LCA algorithm in MG-RAST tool with 90% identity and 50 bp minimum alignment length. Data is represented as follows; Reactor A (blue), Reactor B (green), Reactor C (red) and Control (olive green). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Table 3 Top 10 bacterial abundant genera in each reactor and control sample of activated sludge. Sequences that could not be assigned to a particular genus have been grouped according to the taxonomic levels of identification. Group 1 indicates unclassified sequences at class level, group 2 indicates unclassified sequences at order level, group 3 indicates unclassified sequences at family level, group 4 indicates unclassified sequences at genus level. All sequences other than the top ten have been grouped as group 5. Sr. no.
Control
% Relative abundance
Reactor A
% Relative abundance
Reactor B
% Relative abundance
Reactor C
% Relative abundance
1 2 3 4 5 6 7 8 9 10
Group 3 Mycobacterium Group 5 Group 4 Clostridium Hyphomicrobium Group 2 Tissierella Sphingomonas Desulfobacterium
45.9333270 17.3873886 8.14892061 6.47024567 5.08274872 5.03602481 2.12830791 1.89367263 1.87166499 1.416276172
Mycobacterium Group 2 Group 5 Group 3 Group 4 Clostridium Kribbella Deinococcus Group 1 Saccharopolyspora
41.229016 15.5879903 10.4142011 8.92307692 6.22046899 3.18124041 2.40631163 2.10387902 1.98378259 1.741836511
Mycobacterium Group 2 Group 5 Group 3 Group 4 Saccharopolyspora Clostridium Kribbella Deinococcus Pseudonocardia
43.3748711 14.6098357 9.42092364 8.29775626 4.41473193 3.30542012 3.15560674 2.72694995 2.58060046 1.823739792
Mycobacterium Group 2 Group 5 Group 3 Group 4 Clostridium Saccharopolyspora Myroides Deinococcus Kribbella
44.3009968 14.3852943 10.1807281 8.0387943 4.18236175 3.42582533 3.04121299 2.43931451 2.43852150 1.948438157
(Group 1 = unclassified sequences at Class level, Group 2 = unclassified sequences at order level, Group 3 = unclassified sequences at Family level, Group 4 = unclassified sequences at Genus level, Group 5 = all sequences other than top 10).
un-classified sequences appearing under each reactor have been grouped together as described in Table 3. The number of unclassified sequences decreases as we go to 1 mg l1 to 4 mg l1. This suggests that the microbial community is being enriched and reached to new ecological balance. Principal Component Analysis showed significant separation of the microbial community from the activated biomass that was collected at the CETP (Control) to the microbial community that
developed under different DO levels, despite originating from the same source (Fig. 6). Diversity of samples growing at 2 mg l1 and 4 mg l1 are close together (points B and C in the figure), while biomass growing at 1 mg l1 seems divergent (point A in the Figure). Detailed differentiation of the microbial community can be seen in the tree (Fig. 5) and the heatmap (Supplementary Fig. 2). The Shannon index values are 1.337, 1.180, 1.327 and 1.343 for control, 1 mg l1, 2 mg l1and 4 mg l1respectivly. The Results
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Fig. 6. PCA analysis of all 3 reactor amplicon libraries with control at phylum levels was conducted using the covariance matrix.
from rarefaction analysis indicated sufficient depth of data (Supplementary Fig. 3). The highest species diversity was observed in reactor B (2 mg l1 DO). 4. Conclusion This study demonstrates DO dependent degradative efficiency of activated biomass towards wastewater collected from CETP. Analysis of bacterial diversity was also shown to vary with different DO. The data generated could be used to select the optimum parameters required to run a WWTP efficiently, although pilot scale trials would be further required to take lab-scale results to the field. Acknowledgements The authors acknowledge the Council of Scientific and Industrial Research, India, CSIR-network project ESC-0108-MESER, for supporting this research. Trilok Chandra Yadav is grateful for the CSIR-UGC award of junior research fellowship. The management of the CETP is acknowledged for providing activated biomass and wastewater samples used in this study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2014. 03.007. References Amann, R., Snaidr, J., Wagner, M., Ludwig, W., Schleifer, K.H., 1996. In situ visualization of high genetic diversity in a natural microbial community. J. Bacteriol. 178, 3496–3500. Bolhuis, H., Stal, L.J., 2011. Analysis of bacterial and archaeal diversity in coastal microbial mats using massive parallel 16S rRNA gene tag sequencing. ISME J. 5, 1701–1712. Bond, P.L., Erhart, R., Wagner, M., Keller, J., Blackall, L.L., 1999. Identification of some of the major groups of bacteria in efficient and nonefficient biological phosphorus removal activated sludge systems. Appl. Environ. Microbiol. 65, 4077–4084.
263
Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., et al., 2010. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336. Chakravorty, S., Helb, D., Burday, M., Connell, N., Alland, D., 2007. A detailed analysis of 16S ribosomal RNA gene segments for the diagnosis of pathogenic bacteria. J. Microbiol. Methods 69, 330–339. Claesson, M.J., Wang, Q., O’Sullivan, O., Greene-Diniz, R., Cole, J.R., Ross, R.P., O’Toole, P.W., 2010. Comparison of two next-generation sequencing technologies for resolving highly complex microbiota composition using tandem variable 16S rRNA gene regions. Nucleic Acids Res. 38, e200. Crocetti, G.R., Banfield, J.F., Keller, J., Bond, P.L., Blackall, L.L., 2002. Glycogenaccumulating organisms in laboratory-scale and full-scale wastewater treatment processes. Microbiology 148, 3353–3364. Felföldi, T., Székely, A.J., Gorál, R., Barkács, K., Scheirich, G., András, J., Rácz, A., Márialigeti, K., 2010. Polyphasic bacterial community analysis of an aerobic activated sludge removing phenols and thiocyanate from coke plant effluent. Bioresour. Technol. 101, 3406–3414. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. Past: paleontological statistics software package for education and data analysis. Paleontol. Electrónica 4, 1– 9 (http://palaeo-electronica. org/2001_1/past/issue1_01. html). Jiang, X., Ma, M., Li, J., Lu, A., Zhong, Z., 2008. Bacterial diversity of active sludge in wastewater treatment plant. Front. Earth Sci. 15, 163–168. Kämpfer, P., Erhart, R., Beimfohr, C., Böhringer, J., Wagner, M., Amann, R., 1996. Characterization of bacterial communities from activated sludge: culturedependent numerical identification versus in situ identification using groupand genus-specific rRNA-targeted oligonucleotide probes. Microb. Ecol. 32, 101–121. Kapley, A., Prasad, S., Purohit, H.J., 2007. Changes in microbial diversity in fed-batch reactor operation with wastewater containing nitroaromatic residues. Bioresour. Technol. 98, 2479–2484. Kapley, A., Tolmare, A., Purohit, H.J., 2001. Role of oxygen in the utilization of phenol by Pseudomonas CF600 in continuous culture. J. Microbiol. Biotechnol. 17, 801– 804. Khardenavis, A.A., Kapley, A., Purohit, H.J., 2007. Simultaneous nitrification and denitrification by diverse Diaphorobacter sp. Appl. Microbiol. Biotechnol. 77, 403–409. LeCleir, G.R., Buchan, A., Hollibaugh, J.T., 2004. Chitinase gene sequences retrieved from diverse aquatic habitats reveal environment-specific distributions. Appl. Environ. Microbiol. 70, 6977–6983. Lee, N., La Cour Jansen, J., Aspegren, H., Henze, M.N.P.H., Nielsen, P.H., Wagner, M., 2002. Population dynamics in wastewater treatment plants with enhanced biological phosphorus removal operated with and without nitrogen removal. Water Sci. Technol. 46, 163–170. Li, H., Shen, T.T., Wang, X.L., Lin, K.F., Liu, Y.D., Lu, S.G., Gu, J.D., Wang, P., Lu, Q., Du, X.M., 2013. Biodegradation of perchloroethylene and chlorophenol cocontamination and toxic effect on activated sludge performance. Bioresour. Technol. 137, 286–293. Li, Y., Ying, Y., Zhao, D., Ding, W., 2012. Microbial community diversity analysis of Panax ginseng rhizosphere and non-rhizosphere soil using randomly amplified polymorphic DNA method. J. Genet. 2, 95–102. Liu, C., Ali, S.W., Guan, L.B., Yu, F.B., Li, S.P., Wong, M.H., 2012. Biotreatment of onitrobenzaldehyde manufacturing wastewater and changes in activated sludge flocs in a sequencing batch reactor. Bioresour. Technol. 104, 228–234. Liu, X.C., Zhang, Y., Yang, M., Wang, Z.Y., LV, W.Z., 2007. Analysis of bacterial community structures in two sewage treatment plants with different sludge properties and treatment performance by nested PCR-DGGE method. J. Environ. Sci. 19, 60–66. Lyautey, E., Lacoste, B., Ten-Hage, L., Rols, J.L., Garabetian, F., 2005. Analysis of bacterial diversity in river biofilms using 16S rDNA PCR-DGGE: methodological settings and fingerprints interpretation. Water Res. 39, 380–388. Manz, W., Wagner, M., Amann, R., Schleifer, K.H., 1994. In situ characterization of the microbial consortia active in two wastewater treatment plants. Water Res. 28, 1715–1723. Meyer, F., Paarmann, D., D’Souza, M., Olson, R., Glass, E.M., Kubal, M., Paczian, T., Rodriguez, A., Stevens, R., Wilke, A., Wilkening, J., Edwards, R.A., 2008. The metagenomics RAST server – a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinformatics 19, 386. Nielsen, A.T., Liu, W.T., Filipe, C., Grady, L., Molin, S., Stahl, D.A., 1999. Identification of a novel group of bacteria in sludge from a deteriorated biological phosphorus removal reactor. Appl. Environ. Microbiol. 65, 1251–1258. Puigagut, J., Salvadó, H., García, J., 2005. Short-term harmful effects of ammonia nitrogen on activated sludge microfauna. Water Res. 39, 4397–4404. Purohit, H.J., Kapley, A., Moharikar, A.A., Narde, G., 2003. A novel approach for extraction of PCR-compatible DNA from activated sludge samples collected from different biological effluent treatment plants. J. Microbiol. Methods 52, 315–323. Rajagopal, R., Massé, D.I., Singh, G., 2013. A critical review on inhibition of anaerobic digestion process by excess ammonia. Bioresour. Technol. 143, 632– 641. Rani, A., Porwal, S., Sharma, R., Kapley, A., Purohit, H.J., Kalia, V.C., 2008. Assessment of microbial diversity in effluent treatment plants by culture dependent and culture independent approaches. Bioresour. Technol. 99, 7098–7107. Sharma, G., Verma, H.N., Sharma, R., 2013. RAPD Analysis to study metagenome diversity in soil microbial community of arid zone plants. Proc. Natl. Acad. Sci. India Sec. B Biol. Sci. 83, 135–139.
264
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Wagner, M., Amann, R., Lemmer, H., Manz, W., Schleifer, K.H., 1994. Probing activated sludge with fluorescently labeled rRNA targeted oligonucleotides. Water Sci. Technol. 29, 15–23. Wells, G.F., Park, H.D., Yeung, C.H., Eggleston, B., Francis, C.A., Criddle, C.S., 2009. Ammonia-oxidizing communities in a highly aerated full-scale activated sludge bioreactor: betaproteobacterial dynamics and low relative abundance of Crenarchaea. Environ. Microbiol. 11, 2310–2328. Winkler, M.H., Kleerebezem, R., de Bruin, L.M.M., Verheijen, P.J.T., Abbas, B., Habermacher, J., van Loosdrecht, M.C.M., 2013. Microbial diversity differences within aerobic granular sludge and activated sludge flocs. Appl. Microbiol. Biotechnol. 97, 7447–7458.
Yao, Y., Guan, J., Tang, P., Jiao, H., Lin, C., Wang, J., Lu, Z., Min, H., Gao, H., 2010. Assessment of toxicity of tetrahydrofuran on the microbial community in activated sludge. Bioresour. Technol. 101, 5213–5221. Yu, K., Zhang, T., 2012. Metagenomic and metatranscriptomic analysis of microbial community structure and gene expression of activated sludge. PLoS One 7. http://dx.doi.org/10.1371/journal.pone.0038183. Zang, K., Kurisu, F., Kasuga, I., Furumai, H., Yagi, O., 2008. Analysis of the phylogenetic diversity of estrone-degrading bacteria in activated sewage sludge using microautoradiography–fluorescence in situ hybridization. Syst. Appl. Microbiol. 31, 206–214.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Shifts in microbial community in response to dissolved oxygen levels in activated sludge Trilok Chandra Yadav, Anshuman A. Khardenavis, Atya Kapley ⇑ Environmental Genomics Division, National Environmental Engineering Research Institute (CSIR-NEERI), Nagpur 440020, India
h i g h l i g h t s 1
Wastewater treatment was analyzed at three DO levels; 1, 2 and 4 mg l
.
Degradative efficiency was observed to vary between 60% and 65% at 4 ppm DO. Shifts in bacterial diversity were compared across different DO. Analytical and bioinformatics tools were used to assess degradative capacity.
a r t i c l e
i n f o
Article history: Received 10 January 2014 Received in revised form 28 February 2014 Accepted 3 March 2014 Available online 12 March 2014 Keywords: Activated biomass Amplicon library Dissolved oxygen Microbial diversity Wastewater treatment
a b s t r a c t This study evaluates the degradative efficiency of activated biomass collected from a Common Effluent Treatment Plant (CETP) under three different dissolved oxygen (DO) levels, 1, 2 and 4 mg l1. The change in bacterial diversity with reference to DO levels was also analyzed. Results demonstrate that degradative efficiency was the highest, when the reactor was maintained at 4 mg l1 DO, but amplicon library analysis showed a greater diversity of bacteria in the reactor maintained at 2 mg l1 DO. Bacteria belonging to the order Desulfuromonadales, Entomoplasmatales, Pasteurellales, Thermales and Chloroflexales have only been detected in this reactor. Ammonia and nitrate levels in all three reactors indicated efficient nitrification process. Results of this study offer new insights into understanding the performance of activated biomass vis-à-vis microbial diversity and degradative efficiency with reference to DO. This information would be useful in improving the efficiency of any wastewater treatment plant. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction The most common aerobic strategy for treatment of wastewater is the activated sludge process. Activated Sludge comprises of highly complex microbial biomass made of eukaryotes, bacteria, archaea, and viruses, in which bacteria are dominant and play an important role in the removal of organic pollutants (Rani et al., 2008; Yu and Zhang, 2012; Winkler et al., 2013). Parameters related to bacterial growth and degradation are mainly, the availability of nutrients and dissolved oxygen (DO). DO is a relative measure of the amount of oxygen that is dissolved in wastewater, and it usually fluctuates seasonally and varies with water temperature and altitude. DO levels govern the rate of degradation of the organics in aerobic growth physiology of microbial communities in any wastewater treatment plant (WWTP) and ⇑ Corresponding author. Address: Environmental Genomics Division, National Environmental Engineering Research Institute (CSIR-NEERI), Nehru Marg, Nagpur 440020, India. Tel.: +91 712 2249883. E-mail address: [email protected] (A. Kapley). http://dx.doi.org/10.1016/j.biortech.2014.03.007 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
also contribute to operational costs (Kapley et al., 2001; Wells et al., 2009). Different levels of DO can also affect the nitrification, denitrification, besides, aerobic reactions, behavior and activity of heterotrophic and autotrophic microorganisms. A low DO adversely affects treatment efficiencies of bulk wastewater volumes, especially in industrial effluents, coming from diverse sources. On the other hand, an excessive DO concentration will lead to unnecessary power consumption. In the aeration tank, the required aeration depends on the actual oxygen demand and the oxygen transfer efficiency. The actual oxygen demand is determined by the amount of pollutants oxidized and biomass produced, while the oxygen transfer efficiency is related to aeration devices, operational DO, temperature, sludge property, and MLSS concentration. Understanding the bacterial diversity present in a wastewater treatment plant and analyzing community shifts with change in operational parameters, is the key to develop successful treatment strategies. There are a large number of reported methods to study bacterial diversity; amplification and sequencing of 16S rRNA gene, Random Amplified Polymorphic DNA analysis (RAPD); (Kapley et al.,
258
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
2007; Li et al., 2012; Sharma et al., 2013), T-RFLP (Felföldi et al., 2010), PCR-DGGE (Lyautey et al., 2005; Yao et al., 2010; Liu et al., 2012, 2013). Each of these has been used successfully, but next generation sequence technologies have widened the scope of data generation and allow us to view this picture in greater detail. This study explores the change in microbial community profile at different DO levels in activated biomass collected from a Common Effluent Treatment Plant (CETP) and compares the change in treatment efficiency with change in DO. The microbial diversity of activated biomass was analyzed under three different DO levels to answer the following questions; (1) How much does the wastewater treatment efficiency vary with change in DO? and (2) How does DO affect the microbial community of the activated biomass? Keeping all operational parameters and nutrient levels constant, COD, ammonia and nitrate levels were monitored in all three reactors as a measure of treatment efficiency. 16S amplicon library was prepared and taxonomic analysis carried out to analyze community shifts based on DO variations. The distribution of bacterial phyla under different operational conditions can be linked to performance of wastewater treatment plant. This study aims to generate a knowledge base that can be used to enhance treatment efficiency in industrial wastewater treatment and define optimum DO conditions that correlate to maximum degradative efficiency. 2. Methods 2.1. Source of wastewater and activated sludge Wastewater and activated biomass used in this study was collected from a CETP, treating wastewater from pharma, chemical and dye industries. This CETP runs at a capacity 2500–3000 m3/ Day. Sampling of activated biomass was carried out from the aerator unit and both the biomass and wastewater (inlet) was immediately brought to the laboratory within 12 h. 2.2. Reactor design and setup The activated biomass collected from the CETP was added into three reactors having total volume of 5 L and working volume of 3 L each. The addition of biomass in each reactor was 3000 mg l1, maintaining the MLSS (mixed liquor suspended solids) concentration being operated at the CETP. Air inlet into the reactors was maintained with the help of a rotameter and the reactors were run at 1 mg l1, 2 mg l1 and 4 mg l1 DO and will be henceforth referred to as Reactor A, Reactor B and Reactor C respectively. The oxygen concentration in the reactors was monitored using an oxygen meter and probe (Yellow Springs Instrument Co., Model 85). All three reactors were fed with wastewater from the CETP and the HRT (hydraulic retention time) was maintained at 3 days to mimic conditions operating at the CETP. Wastewater was collected in bulk so as to maintain a constant influent throughout the period of study. A schematic diagram of the reactor setup can be seen in Supplementary Fig. 1. 2.3. Analytical methods COD, ammonia and nitrate were monitored every three days in keeping with HRT; samples were collected before the removal of treated wastewater and after addition of fresh wastewater. COD ammonia and nitrate were monitored for both influent and treated wastewater (effluent). 5 ml sample from each reactor was collected in triplicates and centrifuged at 7000g. The supernatant was filtered through 0.25 lm membrane filter. 500 ll of filtered sample was used to detect COD using COD detection kit (Merck Germany,
100–1500 mg l1), as per directions of the manufacturer. Levels of ammonia and nitrate were monitored according to APHA standard protocols (American Public Health Association). 2.4. Metagenomic DNA extraction 10 ml activated sludge samples were collected at 0 h and after 2 months from each reactor, centrifuged at 7000g at 4 °C. The samples were washed as per the protocol described earlier (Purohit et al., 2003) and DNA was prepared using the FastDNA SPIN Kit for soil (MP Biomedicals, USA) as per the instructions of the manufacturer. Metagenomic DNA was also prepared from the activated biomass sample that was collected from the CETP, before aliquoting it into the three reactors. This is referred to as ‘control DNA’. The individual DNA extracts were visualized using 1.0% gel electrophoresis, and the DNA concentrations and purities of the extracts were determined by microspectrophotometry (NanoDrop-1000, Thermo Scientific, USA). This sample was used for constructing the amplicon library. 2.5. Next generation sequencing of amplicon library The microbial community of the three reactors and control was identified by amplifying and sequence analysis of the V3 region of 16S rRNA gene from the metagenome. Primers used, target the surrounding conserved region and illumina sequences adapters and dual-index barcodes were added to the amplicon. The primers used to amplify the V3 region were; Forward primer sequence-50 CCTACGGGAGGCAGCAG 30 and Reverse primer sequence-50 ATTACCGCGGCTGCTGG 30 . Libraries were then normalized and sequenced on the MiSeq platform (Illumina). Sequence data was processed by read trimming and identification of V3 sequences, followed by filtering and assigning the operational taxonomic units (OTUs). The reads from filtered OTUs are processed using Quantitative Insights into Microbial Ecology (QIIME) program (Caporaso et al., 2010), to construct a representative sequence for each OTU. The representative sequence was aligned to the Greengenes core set reference databases using PyNAST program. 2.6. Analyzing sequence data using bioinformatics tools 2.6.1. Sequence pre-processing After applying quality parameter, mainly base quality, average base content, and GC distribution in the reads, there were total 7,66,619, 7,13,088 and 7,21,221 pair-end raw reads generated for reactors A, B and C respectively, while the control DNA showed 1,088,944 pair-end raw reads. Further applying specific filters as mentioned in Table 1 and 90–92% reads were obtained and it is used for downstream analysis. 2.6.2. Analysis using MG-RAST In order to study taxonomic abundance in the library, the obtained sequences were analyzed using Metagenomics-Rapid Annotation using Subsystem Technology server (MG-RAST), (Meyer et al., 2008) with default parameters under the accession number 4537850.3, 4545085.3, 4545086.3 and 4545087.3 for the control, reactor A, B and C respectively. Post-processing of pair end sequences are done by MG-RAST using their own QC pipeline, summarized in Supplementary Table 1. 2.6.2.1. Overall comparative taxonomic abundance. Based on the LCA (Lowest Common Ancestor) algorithm implemented in MG-RAST, comparative taxonomic tree study was performed with parameters of 90% identity, minimum alignment length 50 bp for phylum,
259
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264 Table 1 Analysis of reads generated in the amplicon library. The table describes the number of reads and filters used in this study. Sample
Total raw reads (no.)
Passed conserved region filter (no.)
Passed spacer filter (no.)
Passed read quality filter (no.)
Passed mismatch filter (no.)
Control Reactor A Reactor B Reactor C
1,088,944 766,619 713,088 721,221
10,33,681 7,19,639 6,58,215 6,78,097
10,32,933 7,19,014 6,57,656 6,77,486
10,31,861 7,18,951 6,57,544 6,77,420
10,07,153 (92.49%) 7,08,607 (92.43%) 6,43,042 (90.17%) 6,67,199 (92.50%)
class, order, family taxonomic level in all samples by targeting NCBI taxonomy database. 2.6.2.2. Statistical analysis. 2.6.2.2.1. Rarefaction curve and Principal Component Analysis. For analyzing the depth of sequencing performed for each sample to capture the most dominating taxa, rarefaction analysis was carried out using MG-RAST tools. The inter-relationships between the microbial diversity of activated sludge maintained at three different DO levels, compared to the control sample, was visualized using Principal Component Analysis (PCA) that was carried out to assess the variation among the samples using Paleontological Statistics software package for education and data analysis tool (PAST v3.01), (Hammer et al., 2001). PCA allows us to compare and analyze data that has a high degree of correlation within its multiple attributes like taxonomy. 2.6.2.2.2. Heatmap and Shannon index. Microbial abundance was mapped using heatmap modules in ‘‘R’’ statistical package. Shannon index values representing the diversity and evenness of a sample was calculated using with the formula
XR H ¼ i¼1 ðpi ln piÞ where pi is the proportion of characters belonging to the ith type of letter in the string of interest. Shannon’s index was calculated using the PAST v3.01 (Paleontological Statistics Version 3.0) software. 3. Results and discussion 3.1. Treatment efficiency in the reactors as monitored by COD removal The efficiency of the three reactors was analyzed as a measure of COD removal. Initial COD of the wastewater collected from the treatment plant was 3240 mg l1. Conditions being operated at
the CETP were mimicked in the lab with respect to MLSS, HRT and inlet wastewater. Fig. 1 demonstrates the DO dependent COD removal in all three reactors. It can be seen from the figure that Reactor C, operating at 4 mg l1 DO shows the highest efficiency of degradation, maintaining efficiency between 60% and 65% throughout the period of study (60 days). Reactor B, operating at 2 mg l1 DO demonstrates highest efficiency on day 30 and maintained degradative efficiency between 40% and 50%. Degradative efficiency of reactor A was the lowest, ranging from 15% to 25%. This study clearly demonstrates that COD removal rates depend on the dissolved oxygen levels available to the bacteria. Nitrogen levels in the wastewater are also a prime concern; not only do they lead to eutrophication of water bodies, but excessive ammonia in the wastewater influences the degradative capacity and effects microbial growth (Puigagut et al., 2005; Khardenavis et al., 2007; Rajagopal et al., 2013). Hence, the levels of ammonia and nitrate in the three reactors were also monitored. Fig. 2 demonstrates the status of the nitrification process under different DO levels, with reference to the levels of ammonia and nitrate. Removal of ammonia is roughly similar at 1 and 2 mg l1 DO, ranging between 40% and 45% in Reactor A and 45–50% in Reactor B. The highest efficiency of ammonia removal was observed when the DO was 4 mg l1; 70–75% in Reactor C. Ammonia is converted to nitrate in the nitrification process and Fig. 2 demonstrates that nitrate levels in the reactors are synergistic to the ammonia being degraded. This is suggestive of an efficient nitrification process. Table 2 lists the characteristics of both influent (collected at the CETP) and effluent (treated wastewater from the three separate reactors). At the initial stages of the study, a reactor operating at 8 mg l1 DO, was also set up, but the MLSS could not be maintained, hence was discontinued (data not shown). High bubbling rates caused the biomass to splutter and stick on the upper part of the reactor, disintegrating sludge flocs and thus causing drying and increased
Fig. 1. Degradative efficiency of three reactors maintained at different DO levels, as measured by COD removal. The reactors were operated for a period of 60 days and COD has been monitored every 15 days. COD of influent, effluent and treatment efficiency has been denoted in the figure.
260
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Fig. 2. Estimation of ammonia and nitrate measure by APHA protocol for a period of 60 days in each reactor. Removal efficiency of ammonia and percentage increase in nitrate has been represented as percentage.
Table 2 Characteristics of fresh and treated wastewater. Wastewater collected at the CETP and used to run the three bioreactors has been referred to as ‘influent’, while the treated wastewater from the three reactors has been referred to as ‘effluent’. Parameter
Effluent Influent (collected Reactor A Reactor B Reactor C from CETP)
HRT (days) MLSS (mg l1) SRT (days) COD (mg l1) Ammoniacal nitrogen (mg l1) Nitrate nitrogen (mg l1) pH
3 3000 15 3240 70.29 100 8.15
3 3000 15 312 25.65 246.86 7.63
3 3000 15 251 22.56 569.42 7.58
3 3000 15 142 12.53 588.22 7.59
⁄ Influent wastewater was collected and transported from CETP premises to the lab within 24 h. (Reactor A 1 mg l1 DO, Reactor B 2 mg l1 DO, Reactor C 4 mg l1 DO).
evaporation. These observations, coupled with results demonstrated in Fig. 1, suggest 2–4 mg l1 DO as optimum parameter to be maintained at a treatment plant.
3.2. Microbial diversity and richness In order to analyze the effect of DO levels on the microbial community of activated biomass, we compared the phylogenetic profiling of all three reactors with the control sample that represents activated biomass of the CETP. Amplicon libraries were constructed using V3 region primers and high throughput sequencing was carried out on the MiSeq platform. Bacterial 16S ribosomal RNA (rRNA) genes contain nine ‘‘hypervariable regions’’ (V1–V9) that demonstrate considerable sequence diversity among different bacteria. These variable regions are flanked by conserved regions that can be used to construct specific primers (Claesson et al., 2010). However, no single hypervariable region can differentiate among all bacteria and different regions show a better diversity for different groups (Chakravorty et al., 2007). This study did not aim to carry out a complete bacterial profile but to distinguishing all bacterial species to the genus level, which could be done using V3 region (Chakravorty et al., 2007). Hence, the amplicon library data presented here is from the V3 region of the 16S rRNA genes. The relative abundance of microbial diversity analyzed is demonstrated in Fig. 3. The figure shows data from phylum to family level based on relative abundance of microbial population. Only the top 15 taxonomic categories at class level and top 20 at order and family level have been summarized here. The taxonomic classes other than these results are categorized as ‘‘Others’’. The sequences with low similarity or no similarity; or where the V3
regions did not have any alignment hits against taxonomic database have been categorized as ‘‘Unknown’’. Sequences that were unclassified at a particular taxonomy have been labeled as ‘unclassified derived from phylum/order/class etc.’ Sequences that do not show homology in the NCBI database have been labeled as Unclassified. As can be seen in Fig. 3, in all four levels of classification, the unknown and unclassified diversity dominates, comprising of approximately 60% of the population. The phylum Actinobacteria was the next dominant taxonomy observed. The control sample demonstrated a higher percentage of anaerobic bacteria, e.g. Clostridiaceae (Family level) with 2.14% as compared to 0.6/0.7/ 0.8% in reactors A, B and C respectively.
3.3. Changes in microbial community with DO levels Fig. 4 demonstrates the change in bacterial diversity indicating differentiation only at class level while Fig. 5 depicts the taxonomic tree that describes differentiation from phylum to order level for all four amplicon libraries. The tree also demonstrates the abundance at each node. As can be seen in Fig. 4, the control sample shows a different distribution of bacterial classes when compared to the activated biomass being maintained at different DO levels. This figure suggests that alphaproteobacteria is sensitive to oxygen levels. Proteobacteria comprised of a (72.67%, 32.38%, 38.86% and 37.66%), b (11.72%, 47.41%, 37.22% and 38.23%), d (11.36%, 4.45%, 4.14% and 3.45%), c (1.82%, 10.36%, 12.22% and 13.09%) in control DNA, reactors A, B and C respectively. Fig. 5 demonstrates the presence of Acholeplasmatales, a facultative anaerobe. This order could only be detected in the control sample, but with oxygenation, this taxonomy was not observed. This suggests that the CETP was being maintained at very low DO. Rare bacteria, belonging to the order Rubrobacterales and Neisseriales that do not contain many members in the group were observed solely in the reactor operating at 1 mg l1 DO (Reactor A). The reactor operating at 2 mg l1 DO demonstrates the highest microbial diversity and bacteria belonging to the orders, Desulfuromonadales, Entomoplasmatales, Pasteurellales, Thermales and Chloroflexales have only been detected in this reactor. Vibrionales and Chlamydiales are only observed at higher DO levels and not detected at 1 or 2 mg l1 DO levels. What is noteworthy is that many of the bacteria detected here have also been reported in marine environments (Bolhuis and Stal, 2011). This is probably because of the salinity in the treatment plant; contributed to by the total dissolved solids (TDS) enriches the growth of marine bacteria. This study demonstrates the shifts in microbial community with different operational parameters and hence can be correlated to differential treatment efficiency. These observations can be extended in future studies
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
261
Fig. 3. Classification of bacterial diversity at Phylum, Class, Order and Family level of all three reactors in terms of relative abundance. The abundance has been represented in terms of percentage in total effective bacterial sequences in a sample.
Fig. 4. Abundances of different classes in Proteobacteria in all three reactors compared to control sample.
into identification of an indicator population that can be monitored to assess degradative efficiency in treatment plants. The most abundant phylum observed in this study across all reactors and previously reported in activated biomass are Actinobacteria, Proteobacteria, Firmicutes, Bacteroidetes, Deinococcus, Cyanobacteria, Acidobacteria, Nitrospira and Chloroflexi (LeCleir et al., 2004; Zang et al., 2008). Lee et al. (2002) showed the predominance of Proteobacteria in a WWTP with enhanced biological phosphorus removal, with and without nitrogen removal. Two classes of bacteria have been mainly reported in wastewaters containing ammoniacal nitrogen; ammonia-oxidizing Betaproteobacteria and Alphaproteobacteria (Manz et al., 1994; Wagner et al.,
1994; Amann et al., 1996; Kämpfer et al., 1996; Bond et al., 1999; Jiang et al., 2008). Gammaproteobacteria have been reported in previous studies (Nielsen et al., 1999; Crocetti et al., 2002; Liu et al., 2007). This study demonstrates a majority of Alpha and Beta proteobacteria in reactors B and C, while reactor A shows a higher percentage of Betaproteobacteria. At the Class level, Actinobacteria, Bacilli, Clostridia, Cytophaga, Deinococcus, Flavobacteria, Gammabacteria, Solibacteres and Actinomycetales, Clostridiales, Deinococcales, Flavobacteriales, Solibacterials, Coriobacterials dominate at 4 mg l1 DO. Table 3 lists the top 10 dominant genus in each sample. All reads could not be classified under definite taxonomy and hence
262
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Fig. 5. Comparison tree profile of all three reactors upto order level using LCA algorithm in MG-RAST tool with 90% identity and 50 bp minimum alignment length. Data is represented as follows; Reactor A (blue), Reactor B (green), Reactor C (red) and Control (olive green). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Table 3 Top 10 bacterial abundant genera in each reactor and control sample of activated sludge. Sequences that could not be assigned to a particular genus have been grouped according to the taxonomic levels of identification. Group 1 indicates unclassified sequences at class level, group 2 indicates unclassified sequences at order level, group 3 indicates unclassified sequences at family level, group 4 indicates unclassified sequences at genus level. All sequences other than the top ten have been grouped as group 5. Sr. no.
Control
% Relative abundance
Reactor A
% Relative abundance
Reactor B
% Relative abundance
Reactor C
% Relative abundance
1 2 3 4 5 6 7 8 9 10
Group 3 Mycobacterium Group 5 Group 4 Clostridium Hyphomicrobium Group 2 Tissierella Sphingomonas Desulfobacterium
45.9333270 17.3873886 8.14892061 6.47024567 5.08274872 5.03602481 2.12830791 1.89367263 1.87166499 1.416276172
Mycobacterium Group 2 Group 5 Group 3 Group 4 Clostridium Kribbella Deinococcus Group 1 Saccharopolyspora
41.229016 15.5879903 10.4142011 8.92307692 6.22046899 3.18124041 2.40631163 2.10387902 1.98378259 1.741836511
Mycobacterium Group 2 Group 5 Group 3 Group 4 Saccharopolyspora Clostridium Kribbella Deinococcus Pseudonocardia
43.3748711 14.6098357 9.42092364 8.29775626 4.41473193 3.30542012 3.15560674 2.72694995 2.58060046 1.823739792
Mycobacterium Group 2 Group 5 Group 3 Group 4 Clostridium Saccharopolyspora Myroides Deinococcus Kribbella
44.3009968 14.3852943 10.1807281 8.0387943 4.18236175 3.42582533 3.04121299 2.43931451 2.43852150 1.948438157
(Group 1 = unclassified sequences at Class level, Group 2 = unclassified sequences at order level, Group 3 = unclassified sequences at Family level, Group 4 = unclassified sequences at Genus level, Group 5 = all sequences other than top 10).
un-classified sequences appearing under each reactor have been grouped together as described in Table 3. The number of unclassified sequences decreases as we go to 1 mg l1 to 4 mg l1. This suggests that the microbial community is being enriched and reached to new ecological balance. Principal Component Analysis showed significant separation of the microbial community from the activated biomass that was collected at the CETP (Control) to the microbial community that
developed under different DO levels, despite originating from the same source (Fig. 6). Diversity of samples growing at 2 mg l1 and 4 mg l1 are close together (points B and C in the figure), while biomass growing at 1 mg l1 seems divergent (point A in the Figure). Detailed differentiation of the microbial community can be seen in the tree (Fig. 5) and the heatmap (Supplementary Fig. 2). The Shannon index values are 1.337, 1.180, 1.327 and 1.343 for control, 1 mg l1, 2 mg l1and 4 mg l1respectivly. The Results
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Fig. 6. PCA analysis of all 3 reactor amplicon libraries with control at phylum levels was conducted using the covariance matrix.
from rarefaction analysis indicated sufficient depth of data (Supplementary Fig. 3). The highest species diversity was observed in reactor B (2 mg l1 DO). 4. Conclusion This study demonstrates DO dependent degradative efficiency of activated biomass towards wastewater collected from CETP. Analysis of bacterial diversity was also shown to vary with different DO. The data generated could be used to select the optimum parameters required to run a WWTP efficiently, although pilot scale trials would be further required to take lab-scale results to the field. Acknowledgements The authors acknowledge the Council of Scientific and Industrial Research, India, CSIR-network project ESC-0108-MESER, for supporting this research. Trilok Chandra Yadav is grateful for the CSIR-UGC award of junior research fellowship. The management of the CETP is acknowledged for providing activated biomass and wastewater samples used in this study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2014. 03.007. References Amann, R., Snaidr, J., Wagner, M., Ludwig, W., Schleifer, K.H., 1996. In situ visualization of high genetic diversity in a natural microbial community. J. Bacteriol. 178, 3496–3500. Bolhuis, H., Stal, L.J., 2011. Analysis of bacterial and archaeal diversity in coastal microbial mats using massive parallel 16S rRNA gene tag sequencing. ISME J. 5, 1701–1712. Bond, P.L., Erhart, R., Wagner, M., Keller, J., Blackall, L.L., 1999. Identification of some of the major groups of bacteria in efficient and nonefficient biological phosphorus removal activated sludge systems. Appl. Environ. Microbiol. 65, 4077–4084.
263
Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., et al., 2010. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336. Chakravorty, S., Helb, D., Burday, M., Connell, N., Alland, D., 2007. A detailed analysis of 16S ribosomal RNA gene segments for the diagnosis of pathogenic bacteria. J. Microbiol. Methods 69, 330–339. Claesson, M.J., Wang, Q., O’Sullivan, O., Greene-Diniz, R., Cole, J.R., Ross, R.P., O’Toole, P.W., 2010. Comparison of two next-generation sequencing technologies for resolving highly complex microbiota composition using tandem variable 16S rRNA gene regions. Nucleic Acids Res. 38, e200. Crocetti, G.R., Banfield, J.F., Keller, J., Bond, P.L., Blackall, L.L., 2002. Glycogenaccumulating organisms in laboratory-scale and full-scale wastewater treatment processes. Microbiology 148, 3353–3364. Felföldi, T., Székely, A.J., Gorál, R., Barkács, K., Scheirich, G., András, J., Rácz, A., Márialigeti, K., 2010. Polyphasic bacterial community analysis of an aerobic activated sludge removing phenols and thiocyanate from coke plant effluent. Bioresour. Technol. 101, 3406–3414. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. Past: paleontological statistics software package for education and data analysis. Paleontol. Electrónica 4, 1– 9 (http://palaeo-electronica. org/2001_1/past/issue1_01. html). Jiang, X., Ma, M., Li, J., Lu, A., Zhong, Z., 2008. Bacterial diversity of active sludge in wastewater treatment plant. Front. Earth Sci. 15, 163–168. Kämpfer, P., Erhart, R., Beimfohr, C., Böhringer, J., Wagner, M., Amann, R., 1996. Characterization of bacterial communities from activated sludge: culturedependent numerical identification versus in situ identification using groupand genus-specific rRNA-targeted oligonucleotide probes. Microb. Ecol. 32, 101–121. Kapley, A., Prasad, S., Purohit, H.J., 2007. Changes in microbial diversity in fed-batch reactor operation with wastewater containing nitroaromatic residues. Bioresour. Technol. 98, 2479–2484. Kapley, A., Tolmare, A., Purohit, H.J., 2001. Role of oxygen in the utilization of phenol by Pseudomonas CF600 in continuous culture. J. Microbiol. Biotechnol. 17, 801– 804. Khardenavis, A.A., Kapley, A., Purohit, H.J., 2007. Simultaneous nitrification and denitrification by diverse Diaphorobacter sp. Appl. Microbiol. Biotechnol. 77, 403–409. LeCleir, G.R., Buchan, A., Hollibaugh, J.T., 2004. Chitinase gene sequences retrieved from diverse aquatic habitats reveal environment-specific distributions. Appl. Environ. Microbiol. 70, 6977–6983. Lee, N., La Cour Jansen, J., Aspegren, H., Henze, M.N.P.H., Nielsen, P.H., Wagner, M., 2002. Population dynamics in wastewater treatment plants with enhanced biological phosphorus removal operated with and without nitrogen removal. Water Sci. Technol. 46, 163–170. Li, H., Shen, T.T., Wang, X.L., Lin, K.F., Liu, Y.D., Lu, S.G., Gu, J.D., Wang, P., Lu, Q., Du, X.M., 2013. Biodegradation of perchloroethylene and chlorophenol cocontamination and toxic effect on activated sludge performance. Bioresour. Technol. 137, 286–293. Li, Y., Ying, Y., Zhao, D., Ding, W., 2012. Microbial community diversity analysis of Panax ginseng rhizosphere and non-rhizosphere soil using randomly amplified polymorphic DNA method. J. Genet. 2, 95–102. Liu, C., Ali, S.W., Guan, L.B., Yu, F.B., Li, S.P., Wong, M.H., 2012. Biotreatment of onitrobenzaldehyde manufacturing wastewater and changes in activated sludge flocs in a sequencing batch reactor. Bioresour. Technol. 104, 228–234. Liu, X.C., Zhang, Y., Yang, M., Wang, Z.Y., LV, W.Z., 2007. Analysis of bacterial community structures in two sewage treatment plants with different sludge properties and treatment performance by nested PCR-DGGE method. J. Environ. Sci. 19, 60–66. Lyautey, E., Lacoste, B., Ten-Hage, L., Rols, J.L., Garabetian, F., 2005. Analysis of bacterial diversity in river biofilms using 16S rDNA PCR-DGGE: methodological settings and fingerprints interpretation. Water Res. 39, 380–388. Manz, W., Wagner, M., Amann, R., Schleifer, K.H., 1994. In situ characterization of the microbial consortia active in two wastewater treatment plants. Water Res. 28, 1715–1723. Meyer, F., Paarmann, D., D’Souza, M., Olson, R., Glass, E.M., Kubal, M., Paczian, T., Rodriguez, A., Stevens, R., Wilke, A., Wilkening, J., Edwards, R.A., 2008. The metagenomics RAST server – a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinformatics 19, 386. Nielsen, A.T., Liu, W.T., Filipe, C., Grady, L., Molin, S., Stahl, D.A., 1999. Identification of a novel group of bacteria in sludge from a deteriorated biological phosphorus removal reactor. Appl. Environ. Microbiol. 65, 1251–1258. Puigagut, J., Salvadó, H., García, J., 2005. Short-term harmful effects of ammonia nitrogen on activated sludge microfauna. Water Res. 39, 4397–4404. Purohit, H.J., Kapley, A., Moharikar, A.A., Narde, G., 2003. A novel approach for extraction of PCR-compatible DNA from activated sludge samples collected from different biological effluent treatment plants. J. Microbiol. Methods 52, 315–323. Rajagopal, R., Massé, D.I., Singh, G., 2013. A critical review on inhibition of anaerobic digestion process by excess ammonia. Bioresour. Technol. 143, 632– 641. Rani, A., Porwal, S., Sharma, R., Kapley, A., Purohit, H.J., Kalia, V.C., 2008. Assessment of microbial diversity in effluent treatment plants by culture dependent and culture independent approaches. Bioresour. Technol. 99, 7098–7107. Sharma, G., Verma, H.N., Sharma, R., 2013. RAPD Analysis to study metagenome diversity in soil microbial community of arid zone plants. Proc. Natl. Acad. Sci. India Sec. B Biol. Sci. 83, 135–139.
264
T.C. Yadav et al. / Bioresource Technology 165 (2014) 257–264
Wagner, M., Amann, R., Lemmer, H., Manz, W., Schleifer, K.H., 1994. Probing activated sludge with fluorescently labeled rRNA targeted oligonucleotides. Water Sci. Technol. 29, 15–23. Wells, G.F., Park, H.D., Yeung, C.H., Eggleston, B., Francis, C.A., Criddle, C.S., 2009. Ammonia-oxidizing communities in a highly aerated full-scale activated sludge bioreactor: betaproteobacterial dynamics and low relative abundance of Crenarchaea. Environ. Microbiol. 11, 2310–2328. Winkler, M.H., Kleerebezem, R., de Bruin, L.M.M., Verheijen, P.J.T., Abbas, B., Habermacher, J., van Loosdrecht, M.C.M., 2013. Microbial diversity differences within aerobic granular sludge and activated sludge flocs. Appl. Microbiol. Biotechnol. 97, 7447–7458.
Yao, Y., Guan, J., Tang, P., Jiao, H., Lin, C., Wang, J., Lu, Z., Min, H., Gao, H., 2010. Assessment of toxicity of tetrahydrofuran on the microbial community in activated sludge. Bioresour. Technol. 101, 5213–5221. Yu, K., Zhang, T., 2012. Metagenomic and metatranscriptomic analysis of microbial community structure and gene expression of activated sludge. PLoS One 7. http://dx.doi.org/10.1371/journal.pone.0038183. Zang, K., Kurisu, F., Kasuga, I., Furumai, H., Yagi, O., 2008. Analysis of the phylogenetic diversity of estrone-degrading bacteria in activated sewage sludge using microautoradiography–fluorescence in situ hybridization. Syst. Appl. Microbiol. 31, 206–214.