The world is facing the challenges of energy shortage and climate change. Various innovation techniques are developing to produce and recover clean energy such as methane (Niu et al., 2014; Li et al., 2017b). Municipal solid waste (MSW) landfills are an important methane resource, representing 22% of global anthropogenic emissions (IPCC, 2013). Recently, Powell et al. (2016) investigated 850 landfills in the United States and found that the CH₄ emissions from landfills are largely underestimated (262 million tons vs. 121 million tons estimated by United States EPA), suggesting that more bioenergy is stored in landfills.

Methane production (methanogenesis) in solid waste decomposition is a typical methanogen-mediated and enzyme-catalyzed anaerobic digestion. Methanogenesis uses carbon such as acetic acid and carbon dioxide as the terminal electron acceptor to produce methane. The reaction involves a series of enzymes and coenzymes. Understanding the underlying mechanism of methanogen community dynamics and metabolic diversity during solid waste decomposition is critical for methane production and recovery.

Methanogens are the predominant archaea in landfills (Song et al., 2015a) and perform the terminal step of methane conversion (Garrity and Holt, 2001). According to the metabolic precursors consumed, methanogens are classified into three groups as acetoclastic methanogens (acetate), hydrogenotrophic methanogens (H₂, formate, and CO₂), and methylotrophic methanogens (methylated compounds) (Garrity and Holt, 2001; Vanwonterghem et al., 2016). Both acetoclastic and hydrogenotrophic methanogens in solid waste have been reported (Barlaz, 1997; Song et al., 2015a; De la Cruz et al., 2021; Zhao et al., 2021), and their abundance varied under different conditions. Few studies have investigated the methanogen community dynamics during solid waste decomposition. Monitoring changes in δ¹³CH₄ during the mesophilic fermentation of MSW showed that methanogenic metabolism changed with increasing incubation times (Qu et al., 2009). Conversely, automated ribosomal intergenic spacer analysis (ARISA) and fluorescence in situ hybridization (FISH) observation showed that the methanogen community is kept stable. Fei et al. (2015) studied the methanogen community changes at the methane production phase and found that Methanobacteriaceae was dominant throughout the phase. In contrast, we previously investigated the methanogen community in Chinese landfills and results showed that hydrogenotrophic methanogens Methanomicrobiales and/or Methanobacteriales were dominant (Song et al., 2015a). This discrepancy may reflect different components in solid waste in various regions. In addition, the study of Qu (Qu et al., 2009) only identified the methanogenesis phase and we only investigated methanogens in limited time points rather than the entire process.

Solid waste decomposition is a typical anaerobic digestion (AD), a multiphase complex biological process (Zhao et al., 2019). Methane production by AD depends on the stability and biological activity efficiency of the reaction. Core parameters such as volatile fatty acids (VFAs), pH, gas composition, and production have been used to indicate the stability of reaction (Boe et al., 2010; Polag et al., 2015). Most parameters, however, have deficiencies due to limited available conditions in responding to the actual condition. A stable carbon isotope shows potential for AD reaction monitoring with high-resolution and fast-responding characteristics (Polag et al., 2015). This technique will benefit from the succession of methanogens and methanogenic pathways during the entire process.

Methanogens’ metabolic function plays an essential role in methane production during solid waste decomposition. Methyl coenzyme M reductase, including two alpha (mcrA), two beta (mcrB), and two gamma (mcrG) subunits, catalyzes heterodisulfide formation between coenzyme M and coenzyme B and subsequent release of methane (Dhillon et al., 2005). Gene mcrA has been widely used to investigate methanogens in solid waste (Tang et al., 2016). However, methanogenic pathways and their associated functional genes during solid waste decomposition are not known. The recently developed PICRUSt using 16S rRNA homology analysis has shown great potential for metabolic function prediction (Langille et al., 2013).

In this study, the changes in methanogen community composition and functional genes relevant to methanogenic pathways during solid waste decomposition were investigated in a bioreactor by Illumina MiSeq sequencing and PICRUSt, respectively. The impact of the physiochemical parameters of solid waste and leachate on the methanogen community structure was then assessed by multivariate analysis. Study on the methanogen community dynamics and methanogenic pathways during solid waste decomposition not only provides direct evidence for methane production mechanism but also shows potential for the development of monitoring indicators in AD stability.

Landfills are the most commonly used site for MSW disposal. Large amounts of organic matter in landfills undergo a series of decomposition processes. The decomposition process has been modeled, and the three-stage (AP, ACP, and MP) MSW decomposition model is one of the most classic phase separation models (Barlaz et al., 1989; Kjeldsen et al., 2002). In this study, these three stages were successfully separated according to the physicochemical parameters of leachate (Liu et al., 2018). In addition, the values of OM, TN, BDM%, and nitrate decreased along the decomposition. There was 76% BDM lost by microorganism consumption. The combination of these physicochemical parameters’ variation showed that solid waste undergoes decomposition.

Landfills harbor highly diverse and dynamic microbial communities. Within them, the methanogen community plays a vital role in methane production. Obviously, the methanogen community responds to the variation in the decomposition condition. In this study, the methanogen community richness and diversity differed at three solid waste decomposition phases (e.g., ACP had the highest richness of methanogen community). Taxonomically, Methanomicrobiales, Methanosarcinales, and E2 were the three most dominant groups. Domination of Methanomicrobiales and Methanosarcinales was also detected in other studies on solid waste decomposition (Bareither et al., 2013; Fei et al., 2015; Song et al., 2015a), whereas the dominance of E2 was for the first time detected. This is probably due to most of the previous studies investigating the methanogen community in specific time points rather than the whole solid waste decomposition process. These three most dominant groups varied at the three phases of solid waste decomposition. Methanomicrobiales and Methanosarcinales were initially the dominant populations in AP, E2 became the dominant group in ACP, and Methanomicrobiales and Methanosarcinales dominated in MP again. High VFAs produced by high food waste (57.1%, mass percent) in this study probably cause the decrease in Methanomicrobiales and Methanosarcinales in ACP due to VFA inhabitation (Song et al., 2015a). E2 (affiliated to class Thermoplasmata) was abundant in ACP as well as in other AD (Godon et al., 1997; Leclerc et al., 2004), owing to their resistance ability on acidophilic and thermophilic conditions. Thermoplasmata in bovine rumen has been proved to be methylotrophic methanogens by metatranscriptomic analysis (Poulsen et al., 2013). Dominant methylotrophic methanogen E2 was detected in ACP of solid waste decomposition, highlighting the critical role of this group in acidogenesis. In addition, the methanogen community in solid waste differs from that in landfill leachate (Yang et al., 2021). For landfill leachate, E2 (32.2%) and Methanomicrobiales (35.3%) dominated in AP, E2 predominated in ACP (98.7%), and Methanomicrobiales predominated in MP (92.4%). The difference might be the microorganisms’ different dispersal in solid and liquid phases.

Studies on the methanogen community dynamics during methane production showed that hydrogenotrophic methanogens [Methanomicrobiales (Bareither et al., 2013) or Methanobacteriaceae (Fei et al., 2015)] were dominant throughout the phase, which is consistent with this study that hydrogenotrophic methanogen Methanomicrobiales was dominant in MP. However, our study showed that the methanogen community composition changed during solid waste decomposition. Two reasons might contribute to the difference. First, these two studies sampled in-place solid waste for approximately 3–4 months (Bareither et al., 2013) and 2–3 years (Fei et al., 2015), respectively. Therefore, the aerobic phase is absent due to AP occurring in a short term (usually 1–2 weeks) (Kjeldsen et al., 2002; Song et al., 2015b). Second, the abundance of food waste in tested solid waste was low (1.1%) or zero, much lower than that in this study (57.1%). In this study, AP was separated during solid waste decomposition based on previously published results (Kjeldsen et al., 2002) and under in situ bioreactor operation conditions. Abundances of Methanosarcinales and Methanomicrobiales were detected in solid waste samples of AP and MP. Acidic and thermal conditions in ACP favor the growth of archaea E2 (Godon et al., 1997; Leclerc et al., 2004), likely leading to abundant E2 occurring in solid waste.

Methanogen genera varied according to the variation of the methanogen order. Although methanogen Methanomicrobiales were dominant in AP and MP, their affiliated genera changed, showing that unclassified Methanomicrobiaceae was abundant in AP and Methanocorpusculum was dominant in MP. Hydrogenotrophic methanogens Methanomicrobiaceae and Methanocorpusculum have been reported in AD (Carballa et al., 2015) and molasses wastewater anaerobic reactor (McHugh et al., 2004) due to their tolerance to environmental stress. Acetoclastic methanogens such as Methanosarcina sp. have also been reported in AD (McHugh et al., 2003; Laloui-Carpentier et al., 2006; Langille et al., 2013), indicating that acetoclastic methanogenesis pathways extensively exist in AD. Genus vadinCA11 (affiliated to E2) was also abundant in AD (Godon et al., 1997; Leclerc et al., 2004), owing to its acidophilic and thermophilic adaption characters.

The 30 core top abundant OTUs clearly separated solid waste decomposition into three groups AP, ACP, and MP. At the same time, network analysis and hierarchical clustering analysis separated the methanogen community into three groups AP, ACP, and MP. The combination of these analyses showed that methanogen composition dramatically varied during solid waste and had unique patterns at each phase of decomposition.

RDA analysis showed that BOD₅ and nitrate of leachate were the significant impact factors in shaping the methanogen community. BOD₅ contains methane metabolic precursors, which are used with methanogen to produce methane. Nitrate has been found as the important factor associated with the methanogen community in landfill (Song et al., 2015a). However, the underlying mechanism of nitrate in driving the methanogen community is unknown. The link between methanogen community structure and important parameters (BOD₅ and nitrate) reflected that methanogen community variation accords with changes in physiochemical parameters.

Functional enzyme-encoding genes involved in acetoclastic, hydrogenotrophic, and methylotrophic pathways during solid waste decomposition were determined. For all detected genes in the acetoclastic pathway, AP and MP had higher hit reads than ACP, suggesting that the acetoclastic pathway was inhibited in acidogenesis. The high stress of acidogenesis such as high VFA inhabitation on acetoclastic methanogen has been observed in AD (Karakashev et al., 2006). Except FdhA, the rest of the five genes (FmdE, FTR, MCH, MTD, MER) of the hydrogenotrophic pathway were more abundant in MP than those in AP and ACP. This observation accords with the function of MP that methane production mainly occurs at this phase. The combination of the abundant genes’ involvement in the hydrogenotrophic pathway and our observation on dominant hydrogenotrophic methanogens in Chinese landfills (Song et al., 2015a) indicates that hydrogenotrophic methanogen and its methanogenic pathways significantly contribute to the methane production in solid waste decomposition. In theory, acetate and hydrogen contribute 67% and 33% of total methanogenesis, respectively (Conrad, 1999). However, increasing evidence showed that hydrogenotrophic methanogenesis contributes more CH₄ production than previously considered (Karakashev et al., 2006; Krakat et al., 2010; Nettmann et al., 2010), likely due to the higher stress tolerance ability of hydrogenotrophic methanogen than acetoclastic methanogen (Song et al., 2015a). The MtaA gene of the methylotrophic pathway was not detected in this study. The abundance of methylotrophic methanogen E2 in ACP might express other functions than methylotrophic methanogenesis. In response to the variation of the metabolic pathways and the McrG genes, involvement in methane production was inhibited in ACP.

In this study, methanogenesis functional enzyme-encoding genes were predicted based on 16S rRNA homology analysis, showing the potential for methanogenesis. The detected abundant hydrogenotrophic and acetoclastic methanogenesis functional genes in MP and AP were in response to the abundance of Methanomicrobiales and Methanosarcinales. Further studies based on RNA and protein high-throughput analysis are necessary to investigate the activated functions.

Methanogen community composition varied at AP, ACP, and MP, which is driven by BOD₅ and nitrate. Methanomicrobiales and Methanosarcinales were dominant in AP and MP, while E2 was abundant in ACP. The main methanogen orders and genera were significantly different during solid waste decomposition. The detected abundant hydrogenotrophic and acetoclastic methanogenesis functional genes in MP and AP are in response to the abundance of Methanomicrobiales and Methanosarcinales. The findings provide information on the methanogen community and methanogenesis functional genes in methane production in solid waste decomposition.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author/s.

SY, LS, and XP conceived of and designed the research. SY, LL, and RZ conducted the research. SY, LL, XP, and LS contributed to discussions and suggestions. SY analyzed the data. SY and LS wrote the manuscript. All authors read and approved the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.