Bioresource Technology 151 (2014) 397–401
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Short Communication
Substrate type drives variation in reactor microbiomes of anaerobic digesters Wei Zhang a, Jeffrey J. Werner a,b, Matthew T. Agler a, Largus T. Angenent a,⇑ a b
Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY 14853, USA Chemistry Department, SUNY-Cortland, Cortland, NY 13045, USA
h i g h l i g h t s Substrate type drives variation of reactor microbiomes. Continuum exists and co-digestion microbiomes fall between two mixed substrate types. To link other environmental factors, many more samples must be included.
a r t i c l e
i n f o
Article history: Received 16 May 2013 Received in revised form 2 July 2013 Accepted 1 October 2013 Available online 10 October 2013 Keywords: Anaerobic digestion Reactor microbiomes 16S rRNA characterization Sanger sequencing Beta-diversity
a b s t r a c t The goal of this study was to obtain causative information about beta-diversity (differentiation between microbiomes) by comparing sequencing information between studies rather than just knowledge about alpha-diversity (microbiome richness). Here, published sequencing data were merged representing 78 anaerobic digester samples originating from 28 different studies for an overall comparison of beta-diversity (measured using unweighted UniFrac). It was found that digester microbiomes based on bacterial sequences clustered by substrate type, independent of the study of origin, and that this clustering could be attributed to distinct bacterial lineages. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Anaerobic digestion is advantageous due to its combined benefits of treating organic waste and producing bioenergy (Angenent et al., 2004; Lettinga, 1995). The open culture of complex, undefined microbial communities (referred to here as reactor microbiomes) of anaerobic digesters are necessary for robust performance (Kleerebezem and van Loosdrecht, 2007). Some work has been performed on the effect of external factors on anaerobic digestion performance and reactor microbiome structure. These external factors include the operating conditions: reactor configuration (Werner et al., 2011), staging (Angenent et al., 2002), mixing (Hoffmann et al., 2008), and organic loading rate (Chelliapan et al., 2011). However, a comprehensive analysis of how operating factors influence microbial community structure requires many sequencing samples and beta-diversity analysis (differentiation between microbiomes) (Werner et al., 2011). This also requires going beyond just reporting alpha-diversity (microbiome richness). Here, ⇑ Corresponding author. Tel.: +1 607 255 2480; fax: +1 607 257 4449. E-mail address: [email protected] (L.T. Angenent). 0960-8524/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2013.10.004
78 digester samples of full-length bacterial 16S rRNA clone libraries were surveyed by collecting sequences from published digester studies, representing a wide range of substrate types, and merged them to generate an overall comparison of the variation in reactor microbiomes.
2. Methods 2.1. Utilized sequences 19,674 Bacterial 16S rRNA gene sequences were collected from 28 studies from 1998 to present (Table 1). The sequences representing the bacterial portion of the reactor microbiome were focused on, because it has generally been observed to be more diverse and complex than archaeal populations in anaerobic digesters (Fernández et al., 1999). Studies were chosen based on the criteria that random clone libraries were picked and sequenced to produce near-full-length reads. This excludes the sequences that were generated by high-throughput sequencing platforms, such as 454 pyrosequencing and Illumina platforms, and therefore this
398
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401
Table 1 Characteristics of the 78 anaerobic digester samples from 28 studies. Number of samples
Number of reads
Substrate categories
Reactor types
Refs.
1 2 9 1 2 2 3 1 3 2 2 1 1 1 1 2 1 1 1 3 1 2 3 1 1 12 2 16
205 172 10,416 69 50 32 3167 47 38 42 76 68 37 102 67 43 29 28 20 21 56 33 47 114 39 479 57 4120
MW MW MW Manure Household Chemical Chemical Food Co-digestion Household Manure Household Household MW MW VFA VFA Food Chemical VFA Food VFA VFA Food VFA Beverage Beverage Manure
Stirred CSTR Stirred Stirred Plugflow Upflow Upflow Stirred CSTR Stirred Stirred Plugflow Stirred CSTR Plugflow Others Upflow Stirred Others CSTR CSTR CSTR Upflow Stirred Stirred Upflow Upflow ASBR
Chouari et al. (2005) Zhang et al. (2009) Rivière et al. (2009) Liu et al. (2009) Goberna et al. (2009) Enright et al. (2007) Perkins et al. (2011) Ariesyady et al. (2007) Wang et al. (2009) Levén et al. (2007) Cheon et al. (2008) Cheon et al. (2008) Cheon et al. (2008) Cheon et al. (2008) Cheon et al. (2008) Tatara et al. (2008) Satoh et al. (2007) Sasaki et al. (2007) Chen et al. (2009) Shigematsu et al. (2006) Klockea et al. (2007) Tan et al. (2007) McKeown et al. (2009) Feng et al. (2009) Weiss et al. (2009) Narihiro et al. (2009) Sekiguchi et al. (1998) GenBank accession numbers GQ132191–GQ135228 and GQ138118–GQ139199 (Werner et al. submitted)
Total 78
19,674
MW = municipal waste; VFA = volatile fatty acids.
work encompasses the work that has been performed with Sanger sequencing during the last one to two decades. 2.2. Sequence analysis Sequences were aligned to the Greengenes core alignment (DeSantis et al., 2006) and operational taxonomic units (OTUs) were picked at 97% identity using UCLUST (Edgar, 2010) in QIIME 1.3.0 (Caporaso et al., 2010). To account for the variation in the number of sequences per sample, 100 rarefactions of 50 sequences per sample were performed on the samples for which raw reads were available (i.e., samples for which all reads and not just OTU representatives were available on GenBank). For samples in which only OTU representative sequences were publicly available, rarefaction was not possible, and OTU content were combined for these samples with each of the rarefied-sample OTU tables after rarefaction and before analysis of beta-diversity. Phylogenetic distances between samples were calculated for each rarefaction using the unweighted UniFrac metric (Lozupone and Knight, 2005), and the resulting distances were averaged. Unweighted UniFrac was chosen as the beta-diversity metric because it ignores relative abundances and compares samples solely based on the evolutionary histories of their OTUs. This also removes some of the PCR biases that would be different between studies. The unweighted UniFrac distances were analyzed by principal coordinates analysis (PCoA). 3. Results and discussion The variation in unweighted UniFrac distances between all samples are shown in Fig. 1, and graphed as the first three axes from PCoA. This PCoA plot visualizes distances between samples by including as much variation as possible in lower-numbered axes, with each axis representing a different component of the var-
iation in between-sample distances. The first axis, which is the PC1 (7.2% of overall variation), primarily represents the bacterial community differences between samples of digesters that treated livestock wastes (referred to here as manure) and samples of all other digesters (Fig. 1A). Although the emphasis of the clustering along PC1 is in part weighted by the fact that there were a high number of swine manure digester samples (Table 1), the ordination along this axis demonstrated an important relationship among other samples containing manure. Three of the points labeled as manure came from other studies (Liu et al., 2009; Cheon et al., 2008) (see Fig. 2A for samples graphed by study of origin). In addition, the three samples labeled as ‘‘co-digestion’’ that also contained manure (Wang et al., 2009) were placed between the manure cluster and the remainder of the samples (Fig. 1A). This indicates that not only do manure digesters have distinctly different microbial communities from digesters treating other substrates but that there is a continuum of possible phylogenetic structures, depending on the specific substrate composition. Communities performing co-digestion lie along that continuum, sharing some phylogenetic components with both livestock manure digesters and others (e.g., food wastes). The variation represented by PC2 (5.4% variation) and PC3 (4.1% variation) shows that there were unique phylogenetic structure characteristics of digesters treating municipal biosolids (referred to here as municipal waste) (Fig. 1B). Also observed along these axes was that the digesters fed food waste and beverage waste clustered together with some overlap, and communities treating chemical waste were similar to each other. The clustering was not simply a confounding effect due to samples from within a study (Fig. 2B). The most striking variation along PC2, however, was the wide spread of samples fed substrates that were categorized as volatile fatty acids (VFAs). This included lab-scale reactors fed synthetic substrates composed of acetate, propionate, and n-
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401
399
Fig. 1. Variation in phylogenetic structure among 78 anaerobic digester samples, visualized as principal coordinates analysis (PCoA) of unweighted UniFrac distances: (A) PC1 vs PC2; and (B) PC2 vs PC3. Samples are colored by substrate type.
Fig. 2. Variation in phylogenetic structure among 78 anaerobic digester samples, visualized as principal coordinates analysis (PCoA) of unweighted UniFrac distances: (A) PC1 vs PC2 and (B) PC2 vs PC3. Sample points colored by study reference of origin, corresponding with references in the main text, and data listed in Table 1.
butyrate (Tatara et al., 2008; Satoh et al., 2007; Shigematsu et al., 2006; Tan et al., 2007), as well as a reactor fed vinegar (i.e., mostly acetate) (Narihiro et al., 2009). The primary community variation explained by axis PC2 was separation of select VFA-fed digesters and food waste digesters from the remainder of the samples. The variation captured by PC3 suggests a gradient from household/municipal waste to food/beverage/manure to industrial chemical waste (methanol, propanol, acetone, and terephthalic acid). However, the VFA-fed communities seem to span the overall variation in PC3. The separation of the VFA-fed communities along PC3 is explained by the study origin (Fig. 2B), suggesting that other unreported factors determined the placement of VFA experiments along this gradient. To identify the microbiome differences that resulted in the observed sample clustering by substrate type, the sample mapping on the phylogenetic tree was investigated using Topiary Explorer 0.9.1. Fig. 3 highlights the phylogenetic lineages that drove the separation of the two most distinct UniFrac sample clusters: livestock manure (Fig. 3A) and municipal waste (Fig. 3B). These included distinct lineages of Clostridia and Tenericutes that are unique to digesters treating manure, and Alphaproteobacteria, Betaproteobacteria, and Chloroflexi lineages that are unique to digesters treating municipal waste. Also included were other environmental factors, such as reactor configuration, pH, or temperature, in the comparative analysis, but data was scarce in the published reports. With the available data, obvious patterns separating reactor microbiomes based on the environmental factors were not observed. Certainly, it is known that these factors influence specific microbial taxa, but in an overall comparison of anaerobic digesters with a wide range of substrate types, it has been shown that it was substrate that mostly influenced the observed differences in phylogenetic structure. Pos-
sibly the very large effect of substrate type differences on microbiome composition masked the effect of other environmental factors. This brings up three important choices for future experimental work with anaerobic digesters: (i) community variation is anticipated across different substrate types to be of a greater magnitude than variation caused by differences in other operating conditions, and therefore the simultaneous analysis of substrate variation and other variables would require a particularly high sample number and independent analyses to elucidate deeper patterns. Thus, the future study will require both powerful statistics to analyze metadata in combination with sequences from 100s of samples that are obtained via a carefully planned experimental design that includes several bioreactors with time series; (ii) variations in digester bacterial communities are anticipated within a single substrate type to be relatively consistent in comparison to variation between substrate types. Thus, to ascertain the affect of environmental conditions on the beta-diversity, the future study will require a constant and identical substrate between bioreactors; and (iii) it is anticipated that reactor microbiomes are predictable rather than random. Thus, the experiments within future studies can be designed based on testable hypotheses. These three choices are especially relevant now as researchers have begun to use highthroughput, multiplexed 16S rRNA gene sequencing to survey many samples, using Illumina platforms.
4. Conclusion This study compared 19,764 near-full-length, bacterial 16S rRNA gene sequences from 28 different studies including 78 different digester samples. Substrate type driving variation of reactor
400
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401
Fig. 3. Maximum-likelihood phylogenetic tree (FastTree) of all 5,388 OTUs that were identified for the 78 anaerobic digester samples of this study, highlighting unique lineages that contributed to the UniFrac clustering of digester communities fed: (A) livestock manure (including co-digestion); and (B) municipal waste. Highlighted unique lineages are shown in red and purple, respectively, compared to grey branches for lineages of OTUs appearing in other samples. Consensus taxonomies were annotated using the Greengenes reference database, where taxonomic level is indicated in parentheses (c = class; p = phylum). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
microbiomes was observed across these different studies. This strong effect exists as a continuum for which microbiomes from co-digestion digesters fall between the two mixed substrate types. To link other environmental factors with microbiome composition, many more digester samples from within studies and substrate types must be included to overcome the masking of the strong substrate type driver. Acknowledgements W.Z. gratefully acknowledges the Chinese Scholarship Council for providing a research scholarship for her stay at Cornell University. The project was also supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, Grant number 2007-35504-18256. References Angenent, L.T., Zheng, D., Sung, S., Raskin, L., 2002. Microbial community structure and activity in a compartmentalized, anaerobic bioreactor. Water Environ. Res. 74, 450–461.
Angenent, L.T., Karim, K., Al-Dahhan, M.H., Wrenn, B.A., Domíguez-Espinosa, R., 2004. Production of bioenergy and biochemicals from industrial and agricultural wastewater. Trends Biotechnol. 22, 477–485. Ariesyady et al., 2007. Phylogenetic and functional diversity of propionate-oxidizing bacteria in an anaerobic digester sludge. Appl. Microbiol. Biotechnol. 5, 673– 683. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., ˇ a, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Fierer, N., Pen Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of highthroughput community sequencing data. Nat. Methods 7, 335–336. Chelliapan, S., Wilby, T., Yuzir, A., Sallis, P.J., 2011. Influence of organic loading on the performance and microbial community structure of an anaerobic stage reactor treating pharmaceutical wastewater. Desalination 271, 257–264. Chen et al., 2009. Characterization of active microbes in a full-scale anaerobic fluidized bed reactor treating phenolic wastewater. Microbes Environ. 24, 144– 153. Cheon, J., Hidaka, T., Mori, S., Koshikawa, H., Tsuno, H., 2008. Applicability of random cloning method to analyze microbial community in full-scale anaerobic digesters. J. Biosci. Bioeng. 106, 134–140. Chouari et al., 2005. Novel predominant archaeal and bacterial groups revealed by molecular analysis of an anaerobic sludge digester. Environ. Microbiol. 7, 1104– 1115. DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Huber, T., Dalevi, D., Hu, P., Andersen, G.L., 2006. Greengenes, a chimera-checked 16S
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401 rRNA gene database and workbench compatible with ARB. Appl. Environ. Microbiol. 72, 5069–5072. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. Enright et al., 2007. Temporal microbial diversity changes in solvent-degrading anaerobic granular sludge from low-temperature (15 °C) wastewater treatment bioreactors. Syst. Appl. Microbiol. 30, 471–482. Feng et al., 2009. Enhancement of waste activated sludge protein conversion and volatile fatty acids accumulation during waste activated sludge anaerobic fermentation by carbohydrate substrate addition: The affect of pH. Environ. Sci. Technol. 43, 4373–4380. Fernández, A., Huang, S., Seston, S., Xing, J., Hickey, R., Criddle, C., Tiedje, J., 1999. How stable is stable? Function versus community composition. Appl. Environ. Microbiol. 65, 3697–3704. Goberna et al., 2009. Effect of biowaste sludge maturation on the diversity of thermophilic bacteria and archaea in an anaerobic reactor. Appl. Environ. Microbiol. 75, 2566–2572. Hoffmann, R.A., Garcia, M.L., Veskivar, M., Karim, K., Al-Dahhan, M.H., Angenent, L.T., 2008. Effect of shear on performance and microbial ecology of completelystirred anaerobic digesters treating animal manure. Biotechnol. Bioeng. 100, 38–48. Kleerebezem, R., van Loosdrecht, M.C.M., 2007. Mixed culture biotechnology for bioenergy production. Curr. Opin. Biotechnol. 18, 207–212. Klockea et al., 2007. Microbial community analysis of a biogas-producing completely stirred tank reactor fed continuously with fodder beet silage as mono-substrate. Syst. Appl. Microbiol. 30, 139–151. Lettinga, G., 1995. Anaerobic digestion and wastewater treatment systems. Antonie van Leeuwenhoek 67, 3–28. Levén et al., 2007. Effect of process temperature on bacterial and archaeal communities in two methanogenic bioreactors treating organic household waste. FEMS Microbiol. Ecol. 59, 683–693. Liu, F.H., Wang, S.B., Zhang, J.S., Zhang, J., Yan, X., Zhou, H.K., Zhao, G.P., Zhou, Z.H., 2009. The structure of the bacterial and archaeal community in a biogas digester as revealed by denaturing gradient gel electrophoresis and 16S rDNA sequencing analysis. J. Appl. Microbiol. 106, 952–966. Lozupone, C., Knight, R., 2005. UniFrac: a new phylogenetic method for comparing microbial communities. Appl. Environ. Microbiol. 71, 8228–8235. McKeown et al., 2009. Psychrophilic methanogenic community development during long-term cultivation of anaerobic granular biofilms. ISME J. 3, 1231– 1242. Narihiro, T., Terada, T., Kikuchi, K., Iguchi, A., Ikaeda, M., Yamauchi, T., Shiraishi, K., Kamagata, Y., Nakamura, K., Sekiguchi, Y., 2009. Comparative analysis of
401
bacterial and archaeal communities in methanogenic sludge granules from upflow anaerobic sludge blanket reactors treating various food-processing, high-strength organic wastewaters. Microbes Environ. 24, 88–96. Perkins et al., 2011. Comparative 16S rRNA gene surveys of granular sludge from three upflow anaerobic bioreactors treating purified terephthalic acid (PTA) wastewater. Water Sci. Technol. 64, 1406–1412. Rivière et al., 2009. Towards the definition of a core of microorganisms involved in anaerobic digestion of sludge. ISME J. 3, 700–714. Sasaki et al., 2007. Microbial population in the biomass adhering to supporting material in a packed-bed reactor degrading organic solid waste. Appl. Microbiol. Biotechnol. 75, 941–952. Satoh, H., Miura, Y., Tsushima, I., Okabe, S., 2007. Layered structure of bacterial and archaeal communities and their in situ activities in anaerobic granules. Appl. Environ. Microbiol. 73, 7300–7307. Sekiguchi et al., 1998. Phylogenetic diversity of mesophilic and thermophilic granular sludges determined by 16S rRNA gene analysis. Microbiol. 144, 2655– 2665. Shigematsu, T., Era, S., Mizuno, Y., Ninomiya, K., Kamegawa, Y., Morimura, S., Kida, K., 2006. Microbial community of a mesophilic propionate-degrading methanogenic consortium in chemostat cultivation analyzed based on 16S rRNA and acetate kinase genes. Appl. Microbiol. Biotechnol. 72, 401–415. Tan, Y.Q., Shigematsu, T., Morimura, S., Kida, K., 2007. Effect of dilution rate on the microbial structure of a mesophilic butyrate-degrading methanogenic community during continuous cultivation. Appl. Microbiol. Biotechnol. 75, 451–465. Tatara, M., Makiuchi, T., Ueno, Y., Goto, M., Sode, K., 2008. Methanogenesis from acetate and propionate by thermophilic down-flow anaerobic packed-bed reactor. Bioresour. Technol. 99, 4786–4795. Wang, H., Lehtomäki, A., Tolvanen, K., Puhakka, J., Rintala, J., 2009. Impact of crop species on bacterial community structure during anaerobic co-digestion of crops and cow manure. Bioresour. Technol. 100, 2311–2315. Weiss et al., 2009. Investigation of factors influencing biogas production in a largescale thermophilic municipal biogas plant. Appl. Microbiol. Biotechnol. 84, 987– 1001. Werner, J.J., Knights, D., Garcia, M.L., Scalfone, N.B., Smith, S., Yarasheski, K., Cummings, T.A., Beers, A.R., Knight, R., Angenent, L.T., 2011. Bacterial community structures are unique and resilient in full-scale bioenergy systems. Proc. Natl. Acad. Sci. U.S.A. 108, 4158–4163. Zhang et al., 2009. Focused-pulsed sludge pre-treatment increases the bacterial diversity and relative abundance of acetoclastic methanogens in a full-scale anaerobic digester. Water Res. 43, 4517–4526.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Short Communication
Substrate type drives variation in reactor microbiomes of anaerobic digesters Wei Zhang a, Jeffrey J. Werner a,b, Matthew T. Agler a, Largus T. Angenent a,⇑ a b
Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY 14853, USA Chemistry Department, SUNY-Cortland, Cortland, NY 13045, USA
h i g h l i g h t s Substrate type drives variation of reactor microbiomes. Continuum exists and co-digestion microbiomes fall between two mixed substrate types. To link other environmental factors, many more samples must be included.
a r t i c l e
i n f o
Article history: Received 16 May 2013 Received in revised form 2 July 2013 Accepted 1 October 2013 Available online 10 October 2013 Keywords: Anaerobic digestion Reactor microbiomes 16S rRNA characterization Sanger sequencing Beta-diversity
a b s t r a c t The goal of this study was to obtain causative information about beta-diversity (differentiation between microbiomes) by comparing sequencing information between studies rather than just knowledge about alpha-diversity (microbiome richness). Here, published sequencing data were merged representing 78 anaerobic digester samples originating from 28 different studies for an overall comparison of beta-diversity (measured using unweighted UniFrac). It was found that digester microbiomes based on bacterial sequences clustered by substrate type, independent of the study of origin, and that this clustering could be attributed to distinct bacterial lineages. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Anaerobic digestion is advantageous due to its combined benefits of treating organic waste and producing bioenergy (Angenent et al., 2004; Lettinga, 1995). The open culture of complex, undefined microbial communities (referred to here as reactor microbiomes) of anaerobic digesters are necessary for robust performance (Kleerebezem and van Loosdrecht, 2007). Some work has been performed on the effect of external factors on anaerobic digestion performance and reactor microbiome structure. These external factors include the operating conditions: reactor configuration (Werner et al., 2011), staging (Angenent et al., 2002), mixing (Hoffmann et al., 2008), and organic loading rate (Chelliapan et al., 2011). However, a comprehensive analysis of how operating factors influence microbial community structure requires many sequencing samples and beta-diversity analysis (differentiation between microbiomes) (Werner et al., 2011). This also requires going beyond just reporting alpha-diversity (microbiome richness). Here, ⇑ Corresponding author. Tel.: +1 607 255 2480; fax: +1 607 257 4449. E-mail address: [email protected] (L.T. Angenent). 0960-8524/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2013.10.004
78 digester samples of full-length bacterial 16S rRNA clone libraries were surveyed by collecting sequences from published digester studies, representing a wide range of substrate types, and merged them to generate an overall comparison of the variation in reactor microbiomes.
2. Methods 2.1. Utilized sequences 19,674 Bacterial 16S rRNA gene sequences were collected from 28 studies from 1998 to present (Table 1). The sequences representing the bacterial portion of the reactor microbiome were focused on, because it has generally been observed to be more diverse and complex than archaeal populations in anaerobic digesters (Fernández et al., 1999). Studies were chosen based on the criteria that random clone libraries were picked and sequenced to produce near-full-length reads. This excludes the sequences that were generated by high-throughput sequencing platforms, such as 454 pyrosequencing and Illumina platforms, and therefore this
398
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401
Table 1 Characteristics of the 78 anaerobic digester samples from 28 studies. Number of samples
Number of reads
Substrate categories
Reactor types
Refs.
1 2 9 1 2 2 3 1 3 2 2 1 1 1 1 2 1 1 1 3 1 2 3 1 1 12 2 16
205 172 10,416 69 50 32 3167 47 38 42 76 68 37 102 67 43 29 28 20 21 56 33 47 114 39 479 57 4120
MW MW MW Manure Household Chemical Chemical Food Co-digestion Household Manure Household Household MW MW VFA VFA Food Chemical VFA Food VFA VFA Food VFA Beverage Beverage Manure
Stirred CSTR Stirred Stirred Plugflow Upflow Upflow Stirred CSTR Stirred Stirred Plugflow Stirred CSTR Plugflow Others Upflow Stirred Others CSTR CSTR CSTR Upflow Stirred Stirred Upflow Upflow ASBR
Chouari et al. (2005) Zhang et al. (2009) Rivière et al. (2009) Liu et al. (2009) Goberna et al. (2009) Enright et al. (2007) Perkins et al. (2011) Ariesyady et al. (2007) Wang et al. (2009) Levén et al. (2007) Cheon et al. (2008) Cheon et al. (2008) Cheon et al. (2008) Cheon et al. (2008) Cheon et al. (2008) Tatara et al. (2008) Satoh et al. (2007) Sasaki et al. (2007) Chen et al. (2009) Shigematsu et al. (2006) Klockea et al. (2007) Tan et al. (2007) McKeown et al. (2009) Feng et al. (2009) Weiss et al. (2009) Narihiro et al. (2009) Sekiguchi et al. (1998) GenBank accession numbers GQ132191–GQ135228 and GQ138118–GQ139199 (Werner et al. submitted)
Total 78
19,674
MW = municipal waste; VFA = volatile fatty acids.
work encompasses the work that has been performed with Sanger sequencing during the last one to two decades. 2.2. Sequence analysis Sequences were aligned to the Greengenes core alignment (DeSantis et al., 2006) and operational taxonomic units (OTUs) were picked at 97% identity using UCLUST (Edgar, 2010) in QIIME 1.3.0 (Caporaso et al., 2010). To account for the variation in the number of sequences per sample, 100 rarefactions of 50 sequences per sample were performed on the samples for which raw reads were available (i.e., samples for which all reads and not just OTU representatives were available on GenBank). For samples in which only OTU representative sequences were publicly available, rarefaction was not possible, and OTU content were combined for these samples with each of the rarefied-sample OTU tables after rarefaction and before analysis of beta-diversity. Phylogenetic distances between samples were calculated for each rarefaction using the unweighted UniFrac metric (Lozupone and Knight, 2005), and the resulting distances were averaged. Unweighted UniFrac was chosen as the beta-diversity metric because it ignores relative abundances and compares samples solely based on the evolutionary histories of their OTUs. This also removes some of the PCR biases that would be different between studies. The unweighted UniFrac distances were analyzed by principal coordinates analysis (PCoA). 3. Results and discussion The variation in unweighted UniFrac distances between all samples are shown in Fig. 1, and graphed as the first three axes from PCoA. This PCoA plot visualizes distances between samples by including as much variation as possible in lower-numbered axes, with each axis representing a different component of the var-
iation in between-sample distances. The first axis, which is the PC1 (7.2% of overall variation), primarily represents the bacterial community differences between samples of digesters that treated livestock wastes (referred to here as manure) and samples of all other digesters (Fig. 1A). Although the emphasis of the clustering along PC1 is in part weighted by the fact that there were a high number of swine manure digester samples (Table 1), the ordination along this axis demonstrated an important relationship among other samples containing manure. Three of the points labeled as manure came from other studies (Liu et al., 2009; Cheon et al., 2008) (see Fig. 2A for samples graphed by study of origin). In addition, the three samples labeled as ‘‘co-digestion’’ that also contained manure (Wang et al., 2009) were placed between the manure cluster and the remainder of the samples (Fig. 1A). This indicates that not only do manure digesters have distinctly different microbial communities from digesters treating other substrates but that there is a continuum of possible phylogenetic structures, depending on the specific substrate composition. Communities performing co-digestion lie along that continuum, sharing some phylogenetic components with both livestock manure digesters and others (e.g., food wastes). The variation represented by PC2 (5.4% variation) and PC3 (4.1% variation) shows that there were unique phylogenetic structure characteristics of digesters treating municipal biosolids (referred to here as municipal waste) (Fig. 1B). Also observed along these axes was that the digesters fed food waste and beverage waste clustered together with some overlap, and communities treating chemical waste were similar to each other. The clustering was not simply a confounding effect due to samples from within a study (Fig. 2B). The most striking variation along PC2, however, was the wide spread of samples fed substrates that were categorized as volatile fatty acids (VFAs). This included lab-scale reactors fed synthetic substrates composed of acetate, propionate, and n-
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401
399
Fig. 1. Variation in phylogenetic structure among 78 anaerobic digester samples, visualized as principal coordinates analysis (PCoA) of unweighted UniFrac distances: (A) PC1 vs PC2; and (B) PC2 vs PC3. Samples are colored by substrate type.
Fig. 2. Variation in phylogenetic structure among 78 anaerobic digester samples, visualized as principal coordinates analysis (PCoA) of unweighted UniFrac distances: (A) PC1 vs PC2 and (B) PC2 vs PC3. Sample points colored by study reference of origin, corresponding with references in the main text, and data listed in Table 1.
butyrate (Tatara et al., 2008; Satoh et al., 2007; Shigematsu et al., 2006; Tan et al., 2007), as well as a reactor fed vinegar (i.e., mostly acetate) (Narihiro et al., 2009). The primary community variation explained by axis PC2 was separation of select VFA-fed digesters and food waste digesters from the remainder of the samples. The variation captured by PC3 suggests a gradient from household/municipal waste to food/beverage/manure to industrial chemical waste (methanol, propanol, acetone, and terephthalic acid). However, the VFA-fed communities seem to span the overall variation in PC3. The separation of the VFA-fed communities along PC3 is explained by the study origin (Fig. 2B), suggesting that other unreported factors determined the placement of VFA experiments along this gradient. To identify the microbiome differences that resulted in the observed sample clustering by substrate type, the sample mapping on the phylogenetic tree was investigated using Topiary Explorer 0.9.1. Fig. 3 highlights the phylogenetic lineages that drove the separation of the two most distinct UniFrac sample clusters: livestock manure (Fig. 3A) and municipal waste (Fig. 3B). These included distinct lineages of Clostridia and Tenericutes that are unique to digesters treating manure, and Alphaproteobacteria, Betaproteobacteria, and Chloroflexi lineages that are unique to digesters treating municipal waste. Also included were other environmental factors, such as reactor configuration, pH, or temperature, in the comparative analysis, but data was scarce in the published reports. With the available data, obvious patterns separating reactor microbiomes based on the environmental factors were not observed. Certainly, it is known that these factors influence specific microbial taxa, but in an overall comparison of anaerobic digesters with a wide range of substrate types, it has been shown that it was substrate that mostly influenced the observed differences in phylogenetic structure. Pos-
sibly the very large effect of substrate type differences on microbiome composition masked the effect of other environmental factors. This brings up three important choices for future experimental work with anaerobic digesters: (i) community variation is anticipated across different substrate types to be of a greater magnitude than variation caused by differences in other operating conditions, and therefore the simultaneous analysis of substrate variation and other variables would require a particularly high sample number and independent analyses to elucidate deeper patterns. Thus, the future study will require both powerful statistics to analyze metadata in combination with sequences from 100s of samples that are obtained via a carefully planned experimental design that includes several bioreactors with time series; (ii) variations in digester bacterial communities are anticipated within a single substrate type to be relatively consistent in comparison to variation between substrate types. Thus, to ascertain the affect of environmental conditions on the beta-diversity, the future study will require a constant and identical substrate between bioreactors; and (iii) it is anticipated that reactor microbiomes are predictable rather than random. Thus, the experiments within future studies can be designed based on testable hypotheses. These three choices are especially relevant now as researchers have begun to use highthroughput, multiplexed 16S rRNA gene sequencing to survey many samples, using Illumina platforms.
4. Conclusion This study compared 19,764 near-full-length, bacterial 16S rRNA gene sequences from 28 different studies including 78 different digester samples. Substrate type driving variation of reactor
400
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401
Fig. 3. Maximum-likelihood phylogenetic tree (FastTree) of all 5,388 OTUs that were identified for the 78 anaerobic digester samples of this study, highlighting unique lineages that contributed to the UniFrac clustering of digester communities fed: (A) livestock manure (including co-digestion); and (B) municipal waste. Highlighted unique lineages are shown in red and purple, respectively, compared to grey branches for lineages of OTUs appearing in other samples. Consensus taxonomies were annotated using the Greengenes reference database, where taxonomic level is indicated in parentheses (c = class; p = phylum). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
microbiomes was observed across these different studies. This strong effect exists as a continuum for which microbiomes from co-digestion digesters fall between the two mixed substrate types. To link other environmental factors with microbiome composition, many more digester samples from within studies and substrate types must be included to overcome the masking of the strong substrate type driver. Acknowledgements W.Z. gratefully acknowledges the Chinese Scholarship Council for providing a research scholarship for her stay at Cornell University. The project was also supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, Grant number 2007-35504-18256. References Angenent, L.T., Zheng, D., Sung, S., Raskin, L., 2002. Microbial community structure and activity in a compartmentalized, anaerobic bioreactor. Water Environ. Res. 74, 450–461.
Angenent, L.T., Karim, K., Al-Dahhan, M.H., Wrenn, B.A., Domíguez-Espinosa, R., 2004. Production of bioenergy and biochemicals from industrial and agricultural wastewater. Trends Biotechnol. 22, 477–485. Ariesyady et al., 2007. Phylogenetic and functional diversity of propionate-oxidizing bacteria in an anaerobic digester sludge. Appl. Microbiol. Biotechnol. 5, 673– 683. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., ˇ a, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Fierer, N., Pen Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of highthroughput community sequencing data. Nat. Methods 7, 335–336. Chelliapan, S., Wilby, T., Yuzir, A., Sallis, P.J., 2011. Influence of organic loading on the performance and microbial community structure of an anaerobic stage reactor treating pharmaceutical wastewater. Desalination 271, 257–264. Chen et al., 2009. Characterization of active microbes in a full-scale anaerobic fluidized bed reactor treating phenolic wastewater. Microbes Environ. 24, 144– 153. Cheon, J., Hidaka, T., Mori, S., Koshikawa, H., Tsuno, H., 2008. Applicability of random cloning method to analyze microbial community in full-scale anaerobic digesters. J. Biosci. Bioeng. 106, 134–140. Chouari et al., 2005. Novel predominant archaeal and bacterial groups revealed by molecular analysis of an anaerobic sludge digester. Environ. Microbiol. 7, 1104– 1115. DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Huber, T., Dalevi, D., Hu, P., Andersen, G.L., 2006. Greengenes, a chimera-checked 16S
W. Zhang et al. / Bioresource Technology 151 (2014) 397–401 rRNA gene database and workbench compatible with ARB. Appl. Environ. Microbiol. 72, 5069–5072. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. Enright et al., 2007. Temporal microbial diversity changes in solvent-degrading anaerobic granular sludge from low-temperature (15 °C) wastewater treatment bioreactors. Syst. Appl. Microbiol. 30, 471–482. Feng et al., 2009. Enhancement of waste activated sludge protein conversion and volatile fatty acids accumulation during waste activated sludge anaerobic fermentation by carbohydrate substrate addition: The affect of pH. Environ. Sci. Technol. 43, 4373–4380. Fernández, A., Huang, S., Seston, S., Xing, J., Hickey, R., Criddle, C., Tiedje, J., 1999. How stable is stable? Function versus community composition. Appl. Environ. Microbiol. 65, 3697–3704. Goberna et al., 2009. Effect of biowaste sludge maturation on the diversity of thermophilic bacteria and archaea in an anaerobic reactor. Appl. Environ. Microbiol. 75, 2566–2572. Hoffmann, R.A., Garcia, M.L., Veskivar, M., Karim, K., Al-Dahhan, M.H., Angenent, L.T., 2008. Effect of shear on performance and microbial ecology of completelystirred anaerobic digesters treating animal manure. Biotechnol. Bioeng. 100, 38–48. Kleerebezem, R., van Loosdrecht, M.C.M., 2007. Mixed culture biotechnology for bioenergy production. Curr. Opin. Biotechnol. 18, 207–212. Klockea et al., 2007. Microbial community analysis of a biogas-producing completely stirred tank reactor fed continuously with fodder beet silage as mono-substrate. Syst. Appl. Microbiol. 30, 139–151. Lettinga, G., 1995. Anaerobic digestion and wastewater treatment systems. Antonie van Leeuwenhoek 67, 3–28. Levén et al., 2007. Effect of process temperature on bacterial and archaeal communities in two methanogenic bioreactors treating organic household waste. FEMS Microbiol. Ecol. 59, 683–693. Liu, F.H., Wang, S.B., Zhang, J.S., Zhang, J., Yan, X., Zhou, H.K., Zhao, G.P., Zhou, Z.H., 2009. The structure of the bacterial and archaeal community in a biogas digester as revealed by denaturing gradient gel electrophoresis and 16S rDNA sequencing analysis. J. Appl. Microbiol. 106, 952–966. Lozupone, C., Knight, R., 2005. UniFrac: a new phylogenetic method for comparing microbial communities. Appl. Environ. Microbiol. 71, 8228–8235. McKeown et al., 2009. Psychrophilic methanogenic community development during long-term cultivation of anaerobic granular biofilms. ISME J. 3, 1231– 1242. Narihiro, T., Terada, T., Kikuchi, K., Iguchi, A., Ikaeda, M., Yamauchi, T., Shiraishi, K., Kamagata, Y., Nakamura, K., Sekiguchi, Y., 2009. Comparative analysis of
401
bacterial and archaeal communities in methanogenic sludge granules from upflow anaerobic sludge blanket reactors treating various food-processing, high-strength organic wastewaters. Microbes Environ. 24, 88–96. Perkins et al., 2011. Comparative 16S rRNA gene surveys of granular sludge from three upflow anaerobic bioreactors treating purified terephthalic acid (PTA) wastewater. Water Sci. Technol. 64, 1406–1412. Rivière et al., 2009. Towards the definition of a core of microorganisms involved in anaerobic digestion of sludge. ISME J. 3, 700–714. Sasaki et al., 2007. Microbial population in the biomass adhering to supporting material in a packed-bed reactor degrading organic solid waste. Appl. Microbiol. Biotechnol. 75, 941–952. Satoh, H., Miura, Y., Tsushima, I., Okabe, S., 2007. Layered structure of bacterial and archaeal communities and their in situ activities in anaerobic granules. Appl. Environ. Microbiol. 73, 7300–7307. Sekiguchi et al., 1998. Phylogenetic diversity of mesophilic and thermophilic granular sludges determined by 16S rRNA gene analysis. Microbiol. 144, 2655– 2665. Shigematsu, T., Era, S., Mizuno, Y., Ninomiya, K., Kamegawa, Y., Morimura, S., Kida, K., 2006. Microbial community of a mesophilic propionate-degrading methanogenic consortium in chemostat cultivation analyzed based on 16S rRNA and acetate kinase genes. Appl. Microbiol. Biotechnol. 72, 401–415. Tan, Y.Q., Shigematsu, T., Morimura, S., Kida, K., 2007. Effect of dilution rate on the microbial structure of a mesophilic butyrate-degrading methanogenic community during continuous cultivation. Appl. Microbiol. Biotechnol. 75, 451–465. Tatara, M., Makiuchi, T., Ueno, Y., Goto, M., Sode, K., 2008. Methanogenesis from acetate and propionate by thermophilic down-flow anaerobic packed-bed reactor. Bioresour. Technol. 99, 4786–4795. Wang, H., Lehtomäki, A., Tolvanen, K., Puhakka, J., Rintala, J., 2009. Impact of crop species on bacterial community structure during anaerobic co-digestion of crops and cow manure. Bioresour. Technol. 100, 2311–2315. Weiss et al., 2009. Investigation of factors influencing biogas production in a largescale thermophilic municipal biogas plant. Appl. Microbiol. Biotechnol. 84, 987– 1001. Werner, J.J., Knights, D., Garcia, M.L., Scalfone, N.B., Smith, S., Yarasheski, K., Cummings, T.A., Beers, A.R., Knight, R., Angenent, L.T., 2011. Bacterial community structures are unique and resilient in full-scale bioenergy systems. Proc. Natl. Acad. Sci. U.S.A. 108, 4158–4163. Zhang et al., 2009. Focused-pulsed sludge pre-treatment increases the bacterial diversity and relative abundance of acetoclastic methanogens in a full-scale anaerobic digester. Water Res. 43, 4517–4526.
