Due to the important roles of archaea in wastewater treatment processes, archaeal communities have been studied extensively in various anaerobic reactors, but the knowledge of archaeal communities in full-scale activated sludge wastewater treatment plants (WWTPs) remains quite poor. In this study, 454-pyrosequencing was for the first time employed to investigate archaeal communities from 20 full-scale activated sludge WWTPs distributed at a 3,660-meter elevational scale in China. Results showed that archaeal communities from WWTPs were dominated by Methanosarcinales (84.6%). A core archaeal population (94.5%) composed of Methanosaeta, Methanosarcina, Methanogenium and Methanobrevibacter was shared among WWTPs. The elevational pattern of archaeal communities was observed in WWTPs, with an elevational threshold associated with archaeal community richness and structures at approximately 1,500 meters above sea level (masl). A declining trend in community richness with increasing elevation was observed at higher elevations, whereas no trend was presented at lower elevations. Spearman correlation analysis indicated that the archaeal community richness at higher elevations was associated with more environmental variables than that at lower elevations. Redundancy analysis indicated that wastewater variables were the dominant contributors to the variation of community structures at higher elevations, followed by operational variables and elevation.
Activated sludge is composed of a highly complex system, including eukaryotes (protozoa and fungi), bacteria, archaea, and viruses (You et al. 2009). The complex microbial community plays a crucial role in the removal of various pollutants in biological wastewater treatment processes. A complete understanding of the microbial community in activated sludge is important to the adjustment and optimization of wastewater treatment systems. Most previous studies focused on the ecological functions and characteristics of bacterial communities in activated sludge wastewater treatment plants (WWTPs) (Hashimoto et al. 2014), but overlooked the parallel characterization of archaeal communities.
Aerobic phase coupled to anoxic and anaerobic phases is the most widely used configuration process for carbon and nitrogen removal in activated sludge WWTPs. Aerobic ammonia-oxidizing archaea (AOA) was commonly indicated as one of the important components in the nitrogen removal process in WWTPs (Wei et al. 2011). Fredriksson et al. (2012) quantified the higher abundance of methanogenic archaea than bacteria in the anaerobic phase of activated sludge WWTPs. The observation implied the important activity of archaea in carbon removal in the anaerobic phase and the potential role of methanogenic archaea in the structure of activated sludge flocs, according to its crucial contribution to the formation and integrity of granular sludge in anaerobic bioreactors (Angenent et al. 2004). In addition, the production of methane may promote the growth of methanotrophic bacteria, which are involved in the nitrification and denitrification in WWTPs (Bodelier & Frenzel 1999). Due to these potentially important roles of archaea, it is necessary to understand the characteristics of archaeal community completely in activated sludge WWTPs.
In recent years, a few studies have paid attention to AOA in activated sludge (You et al. 2009; Wei et al. 2011), whereas much less focus has been put on the total archaeal community. To data, there are only three reported studies on the activated sludge archaeal communities in full-scale WWTPs. One study reported a simple archaeal community in two small WWTPs using denaturing gradient gel electrophoresis (DGGE) (Gray et al. 2002). The other two studies investigated the diversity and abundance of archaea in full-scale WWTPs using fluorescence in situ hybridization (FISH) and clone library methods (Sanchez et al. 2011; Fredriksson et al. 2012). However, the information on archaeal communities in activated sludge WWTPs remains quite poor. First, the low sequencing depth of these traditional methods applied in previous studies hindered the complete characterization of archaeal communities in activated sludge. Second, the previous studies on activated sludge archaeal communities were limited to one or two WWTPs. Third, no knowledge of the geographical characterization of archaeal communities in WWTPs has been described.
Besides conventional operational and wastewater variables, geographic distribution of WWTPs is also supported to be an influential variable of the microbial communities in WWTPs (Wang et al. 2012; Zhang et al. 2012). However, there has been no report on the association between elevation and microbial communities from WWTPs. In recent years, multiple WWTPs were constructed at high elevations, such as the Bolivia and Tibetan plateau above 3,000 meters above sea level (masl). However, many exhibit lower efficiency for the removal of contaminants than those at low elevations (Zabalaga et al. 2007). The function and performance of WWTPs are directly associated with the microbial communities in activated sludge. Thus, it is necessary to reveal the characteristics of microbial communities on the elevation scale. Elevational gradient is characterized by distinct changes in climate and biotic turnover over short elevation distances. It is indicated that natural archaea respond significantly in the community richness to elevational gradients (Singh et al. 2012). Yang et al. (2014) reported that the archaea in the Tibetan plateau show a larger variation in endemic genes than bacteria to adapt to the elevation-dependent environmental conditions. Therefore, we hypothesized that the activated sludge archaeal community in man-made WWTPs also varies with elevational gradients.
The objectives of this study were (1) to systematically characterize the archaeal diversity in activated sludge WWTPs via 454-pyrosequencing, (2) to test the archaeal community variability with elevation according to the above hypothesis, and (3) to evaluate the linkage of environmental variables and archaeal community.
MATERIALS AND METHODS
Activated sludge sampling and DNA extraction
Polymerase chain reaction (PCR) amplification and pyrosequencing
The V1 and V2 regions of the archaeal 16S rRNA gene were amplified using two forward primers A2Fa and N3F, coupled with one reverse primer A571R (Baker et al. 2003; Kan et al. 2011). Barcodes that allow sample multiplexing during pyrosequencing were incorporated between the 454 adapter and the forward primers. Pyrosequencing was performed using a Roche 454 FLX Titanium sequencer (Roche, Nutley, USA) at the Chinese National Human Genome Center (SinoGenoMax). In this study, 454-pyrosequencing of archaeal 16S rRNA genes generated 730,854 quality-filtered sequence reads, with an average of 4,350 sequences for each sample. The sequence library of each sample was normalized to the size of 3,627 sequences, which was the smallest sample, to conduct the statistical analyses for different samples at the same sequence depth (Figure S1, available with the online version of this paper). The obtained raw reads were deposited in the NCBI short-reads archive database (Accession number: SRX982227 and SRR3316543).
The raw reads were processed using Mothur. Raw reads were excluded if they were shorter than 200 bp or if there was a single base bias. PCR chimeras were removed using the chimera.uchime command implemented in Mothur. Similar sequences were clustered into operational taxonomic units (OTUs) with a 3% sequence dissimilarity cut-off. The phylogenetic taxonomy of remaining sequences was determined using the RDP classifier database with a confidence threshold of 80%. The comparison of the relative abundances of the most abundant OTUs among samples was conducted using the pheatmap package in R (v. 3.12, http://www.r-project.org/).
The relationships between environmental variables and archaeal community, which were characterized as community richness, evenness and structure, were performed separately. Alpha-diversity indices, including Shannon-H index and evenness index, were calculated using R software. Correlation coefficients between environmental variables and archaeal community richness were measured using the Spearman correlation method by SPSS 20.0 software. The similarity analysis of the OTU structures of samples was performed via clustering analysis based on the Bray-Curtis similarity using PAST 3.0 software. Redundancy analysis (RDA) was performed to evaluate the relationship between archaeal community structures and environmental variables with 999 permutations. The contributions of individual variables to the variance of archaeal communities were assessed via variance partitioning analysis. Environmental parameters were log-transformed for the analysis. A p-value <0.01 was considered significant.
RESULTS AND DISCUSSION
Archaeal diversity in activated sludge WWTPs
The systematic investigation of archaeal diversity in activated sludge was performed for the first time via 454 pyrosequencing. The 575,736 normalized archaeal sequences were grouped into 176 OTUs in total. The OTU richness, Shannon-H index, and evenness index are summarized in Table S3 (available with the online version of this paper). The results show that the archaeal diversity was much less than that of bacteria reported for municipal WWTPs at the same pyrosequencing depth, for the latter had at least 1,000 bacterial OTUs per sample (Zhang et al. 2012). The result was consistent with the finding that archaea commonly tend to have lower diversity than bacteria in the majority of environments (Cardinali-Rezende et al. 2012). We speculated that archaea might make use of the activated sludge environment in ways that are more restrictive than for bacteria. The speculation was supported by a previous study (Fredriksson et al. 2012), which confirmed that methanogenic archaea constantly existed in anoxic microenvironments of aerobic activated sludge by FISH, while bacteria occurred broadly in activated sludge.
Archaeal sequences were assigned to 3 phyla, 10 orders and 22 genera using the RDP classifier. The unclassified sequence portions in the total community increased from the phylum level (0.8%) to the genus level (26.5%). In general, methanogenic archaeal orders observed in this study were close to those from Althorpe (England) and Rya (Sweden) WWTPs investigated using cloning libraries and DGGE. However, at the genus level, the archaeal diversity observed here was higher than that in any other previous study on activated sludge WWTPs (Gray et al. 2002; Fredriksson et al. 2012).
Within Euryarchaeota, eight archaeal orders were revealed in activated sludge WWTPs. Acetoclastic methanogen Methanosarcinales, accounting for 84.6% of the total classified sequences, was the most dominant Euryarchaeota, followed by hydrogenotrophic methanogens Methanomicrobiales (6.2%), Methanobacteriales (4.2%), and Methanococcales (1.3%). The four orders were present in all samples, except LS, which did not contain Methanococcales. The other four orders only occurred in 2 to 11 WWTPs with low abundances (<1%). It was reported that hydrogenotrophic methanogen was predominant at high ammonia and chemical oxygen demand (COD) levels (>2.8 g NH4+ L−1 and >1 g COD L−1); otherwise, acetoclastic methanogen dominated in activated sludge (Regueiro et al. 2012; Fotidis et al. 2014). The low influent levels of ammonia and COD might contribute to the greater abundance of Methanosarcinales than that of hydrogenotrophic methanogens in this study (Table S2). Although Methanococcales was usually obtained from marine environments and wastewater-fed anaerobic digesters by previous investigations, the minor but stable proportion of Methanococcales among geographically different WWTPs suggests that it is a basic component of archaeal community in activated sludge. As a novel hydrogenotrophic methanogen, Methanocellales has only been found in anaerobic rice paddy soil (Sakai et al. 2008). In this study, Methanocellales was for the first time reported from activated sludge.
Methanosaeta and Methanosarcina are the only known acetoclastic methanogens. Methanosaeta was shown to greatly outcompete Methanosarcina in all samples. As reported in anaerobic reactors, the low acetate concentration and long solids retention time (SRT) in activated sludge yielded Methanosaeta dominance (McMahon et al. 2001; Conklin et al. 2006). All of the investigated WWTPs were operated with SRTs of at least 10 days, which were longer than the archaea doubling time of 6.5 days described in Althorpe WWTPs (Gray et al. 2002). In activated sludge bioreactors, low prevailing residual acetate concentrations, which were limited by the metabolic pathway of tricarboxylic acid cycle of aerobic micro-organisms (Zhang & Bryant 2011) in activated sludge, contributed to the suppression of Methanosarcina by Methanosaeta.
Filamentous Methanosaeta has been demonstrated to play important roles in the formation and maintenance of granular sludge in upflow anaerobic sludge blanket reactors (UASBR) (Zheng et al. 2006). However, the function of Methanosaeta in activated sludge remains poorly understood. Methanosarcina was usually observed in systems experiencing poor performance, indicating operational instability and acetate accumulation (McMahon et al. 2004). Thus, the low abundance of Methanosarcina might indicate the good performance of the WWTPs investigated in this study. Methanogenium, which uses cold shock proteins to help itself to adapt to cold environments, is usually detected from cold marine sediments (Kendall et al. 2007). This study for the first time reported that Methanogenium was also present in activated sludge. Methanobrevibacter, commonly detected in the rumen and faeces, plays an important role in the efficient digestion of polysaccharides (Federici et al. 2015; Nkamga et al. 2015). The constant presence of Methanobrevibacter in municipal WWTPs implied the potential influence of influent wastewater on activated sludge archaeal communities, due to the great input of human faeces into municipal wastewater.
It is noteworthy that AOA was only detected in SH and CQ WWTPs, represented by 16 raw reads in total. One possible reason was that the sequencing depth in this study was insufficient to detect it. The other possible reason was that AOA was really absent from the samples. The latter possibility was supported by the results of multiple other studies that the specific amplification of AOA ammonia monooxygenase genes usually failed in activated sludge from many municipal WWTPs (Zhang et al. 2011).
Variance in archaeal community richness with elevation
The relative abundance of individual archaeal orders and genera exhibited different associations with elevations above 1,500 masl (Figure S2, available with the online version of this paper). Among the most abundant orders, Methanosarcinales presented a significant and negative correlation in relative abundance with elevation (P < 0.001), while the relative abundance of Methanomicrobiales and Methanobacteriales was significantly and positively associated with elevations (P < 0.001). At the genus level, both Methanosaeta and Methanobacterium declined in their relative abundances with increasing elevations (P < 0.001). In contrast, the relative abundance of both Methanogenium and Methanobrevibacter was positively correlated with elevations (P < 0.001). Multiple studies reported the adaption of Methanogenium to cold based on its cold shock proteins in other environments (Saunders et al. 2003). Thus, the extreme temperature fluctuation in high elevations might contribute to the increase of Methanogenium in relative abundance with increased elevation. In all, the changes in the proportions of individual archaea taxa should partly result from the selection of elevation-dependent environmental variables to archaea in WWTPs.
The linear declining trend in archaeal community richness with increased elevation observed in this study was in agreement with the pattern observed from Mount Fuji, which displayed a mid-elevation ‘peak’ of archaeal community richness at 1,500 masl (Singh et al. 2012). This might suggest that the elevational threshold of 1,500 masl is shared in natural environments and human-controlled activated sludge systems. Increasing elevation is characterized by lower air pressure, stronger UV, and more extreme temperature fluctuations (Yang et al. 2014). These factors might put greater stress on the archaeal community above 1,500 masl than that below 1,500 masl. Thus, consistent with natural archaeal community, archaeal community richness in WWTPs might be partly affected by elevation-dependent environments. On the other hand, the influent wastewater from outside higher-elevation environment might also contribute to the variation of WWTP archaeal communities. Although some researchers reported that a majority of the archaea were from the recycled mature activated sludge (Fredriksson et al. 2012), there is no denying that the great variance among archaeal communities from geographically different WWTPs might originate from the long-term selective accumulation of partial archaea from influent archaea. This assumption needs to be demonstrated in future research.
Linkage between environmental variables and archaeal community structures from activated sludge
Significant elevational differences in archaeal community structure among WWTPs were also observed via clustering analysis. Figure S3 (available with the online version of this paper) shows that the archaeal communities from 20 WWTPs could be clustered into eight groups: Group I contains the sampled WWTPs located at 1,500–3,660 masl; Groups II to V contain the WWTPs located at 110–1,280 masl in western China; Groups VI to VIII contain all WWTPs below 60 masl. The grouping pattern of these WWTPs was significantly characterized by the three elevational gradients. A threshold of elevation associated with archaeal community structures was also obtained at 1,500 masl. The clusters of WWTPs below 1,500 masl did not agree with their elevational distances. For example, XA samples (386 masl) were more close to GY samples (1,280 masl) than TY samples (800 masl) on the clustering tree. Group II (110 masl) displayed a larger phylogenetic distance to group V (218 masl) than to groups III and IV (261–1,280 masl). However, the clustering relationships of WWTPs within group I almost corresponded to the elevational distances among WWTPs above 1,500 masl.
RDA showed that the explanatory percentages of elevation to the variance of archaeal community structures were 2% and 6% at lower and higher elevations, respectively. The explanatory percentage of variables to community structures corresponds to the influencing power of variables to community structures (Wang et al. 2012). Thus, the result suggested that the influence of elevation on archaeal community structures in WWTPs at higher elevations was much more than that at lower elevations. In addition, it is noteworthy that the explanatory power of operational and wastewater variables to archaeal community structures was much higher than that of elevation (averaging 4%) at both lower and higher elevations. This result indicates that although the archaeal community structure in activated sludge was significantly associated with elevation, it remains more influenced by operational and wastewater variables.
Scarce information on the relationships between environment variables and archaeal community structures in activated sludge was described in previous studies. However, multiple studies have demonstrated the important influences of environment variables on archaeal community structures in anaerobic reactors. It was reported that the fluctuations in the organic loading rate and F/M ratio affected the abundance and structure of archaeal communities at the species level in UASBR (Cardinali-Rezende et al. 2013). Jing et al. (2013) demonstrated the significant influence of HRT on the function and community of UASBR. HRT was determined to be the third influencing variable on the performance and community of UASBR reactors via the method of partitioning connection weights in the study of Shi et al. (2009). Conklin et al. (2006) reported that Methanosaeta was more abundant than Methanosarcina in anaerobic reactors with longer SRT. All WWTPs assessed in this study were performed with SRTs more than 10 days, which were longer than the archaea doubling time of 6.5 days as described in other studies (Gray et al. 2002). Therefore, the long SRT in activated sludge might contribute to the competition of individual methanogens.
The variance of archaeal community structure in activated sludge presented a weak association with removal efficiency of BOD (reBOD, 1%) at lower elevations, but significant and strong association with efBOD (18%) at higher elevations (Figure 6). On the other hand, the archaeal community richness also showed significantly positive association with multiple variables of WWTP performance, such as reCOD and reBOD, at higher elevations (Figure 5). These results implied that the archaeal community structure and richness in activated sludge might potentially influence the WWTP performance at higher elevations. Fredriksson et al. (2012) also reported that the relative abundance of Methanosaeta in activated sludge was significantly and negatively correlated with effluent water parameters. However, it cannot be concluded that archaea have an important and direct influence on WWTP performance. As the community structure of methanogens in bioreactors was greatly shaped by the metabolic products (e.g. acetate) of bacterial community (Zhang & Bryant 2011), the archaeal community structures might be interfered with by the variance of bacterial communities in activated sludge. Alternatively, since bacteria have been thought to be the important contributor to WWTP performance, the functional and taxonomic linkages of bacteria and archaea in wastewater treatment processes must be explored completely before determining their relevant contributions to WWTP performance.
This study for the first time investigated the archaeal diversity in full-scale activated sludge WWTPs at the elevation scale via 454-pyrosequencing. A core archaeal group composed of four genera dominated the archaeal community in activated sludge. Archaeal community richness and structure presented significant elevational patterns, with a threshold of elevation at approximately 1,500 masl. At higher elevations, a declining trend in community richness with increased elevation was observed. The community structure was more associated with wastewater variables at higher elevations than at lower elevations. This study provides new insights into the relationship between elevation and archaeal communities in activated sludge WWTPs. The results observed in this study would fundamentally aid in further researches on the manipulation and optimization of archaeal mediated processes in WWTPs at higher elevations.
This study was supported by National Natural Science Foundation of China (no. 91547105 and 51479066), Foundation for Innovative Research Groups of the National Natural Science Foundation of China (no. 51421006), Fundamental Research Funds for the Central Universities (no. 2016B10614), Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and Top-Notch Academic Programs Project of Jiangsu Higher Education Institutions (TAPP).