Implementing metatranscriptomics to unveil the mechanism of bioaugmentation adopted in a continuous anaerobic process treating cow manure

(1)

University of Groningen

Implementing metatranscriptomics to unveil the mechanism of bioaugmentation adopted in a

continuous anaerobic process treating cow manure

Li, Yu; Zhao, Jing; Zhang, Zhenhua

Published in:

Bioresource Technology

DOI:

10.1016/j.biortech.2021.124962

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from

it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date:

2021

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Li, Y., Zhao, J., & Zhang, Z. (2021). Implementing metatranscriptomics to unveil the mechanism of

bioaugmentation adopted in a continuous anaerobic process treating cow manure. Bioresource

Technology, 330, 1-10. [124962]. https://doi.org/10.1016/j.biortech.2021.124962

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

Bioresource Technology 330 (2021) 124962

Available online 11 March 2021

Implementing metatranscriptomics to unveil the mechanism of

bioaugmentation adopted in a continuous anaerobic process treating

cow manure

Yu Li

a,*

_{, Jing Zhao}

a

_{, Zhenhua Zhang}

b

a_{Faculty of Science and Engineering, University of Groningen, Nijenborgh 4, 9747 AG Groningen, the Netherlands} b_{Department of Genetics, University Medical Center Groningen, Groningen, the Netherlands}

H I G H L I G H T S G R A P H I C A L A B S T R A C T •Bioaugmentation imposed a sustainable

enhancement on hydrolysis and methanogenesis.

•Bioaugmentation enhanced the pro-portions of cellulose-degrading bacteria. •Archaeal uncultured_ Bathyarchaeia was enriched in the bioaugmented reactor. •Cellulase and methyl-coenzyme M

reductase thrived in the bioaugmented reactor.

•Uncultured_ Bathyarchaeia was pro-posed as ‘acetogenic archaea’.

A R T I C L E I N F O Keywords: Anaerobic digestion Bioaugmentation Cellulase methyl-coenzyme M reductase Uncultured_ Bathyarchaeia A B S T R A C T

This study aimed to investigate the effect of bioaugmentation on microbial community and function in a continuous anaerobic process treating lignocellulosic cow manure. One reactor (Rb) received bioaugmentation

dosage for a certain period (d100-d170) and stopped afterward (d170-d220), while the same applied to the control (Rc) except sterilized bioaugmentation dosage was introduced. Samples were taken on day130, 170 and

220 from both reactors for metatranscriptomic analysis. The results underlined the promotive effect of bio-augmentation on indigenous microorganisms regarding hydrolysis and methanogenesis. Biobio-augmentation contributed to the enrichment of Clostridium, Cellvibrio, Cellulomonas, Bacillus, Fibrobacter, resulting in enhanced cellulase activity (Rb: 0.917–1.081; Rc: 0.551–0.677). Moreover, bioaugmentation brought Rb the prosperity of

uncultured_ Bathyarchaeia, a prominent archaeal group responsible for the improved methyl-coenzyme M

reductase activity, thus accelerated methanogenesis. Unique metabolic pathways (autotrophic carbon fixation and methanogenesis) in uncultured_ Bathyarchaeia broadened the horizon of its fundamental role as acetogens and methanogens in anaerobic digestion.

* Corresponding author.

E-mail address: yu.li@rug.nl (Y. Li).

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier.com/locate/biortech

https://doi.org/10.1016/j.biortech.2021.124962

(3)

1. Introduction

Anaerobic digestion (AD) acts as a viable technology to extract the residual value of agricultural wastes (i.e., crop residues and manure) generated by farm sectors. Within these criteria, researchers have exploited many attempts to improve the degradation of lignocellulosic components within agricultural wastes due to their recalcitrance in hydrolysis in AD. A similar situation also applies to cow manure (CM), the largest contributor in manure generation worldwide, which requires significant diminishment (Li et al., 2021). In this context, bio-augmentation has become a well-established biotechnological method that introduces functional-oriented microorganisms to modify or enhance the indigenous microbial guilds in AD. Driven by its environmental-friendly nature, researchers have developed single cul-ture, co-culcul-ture, and mixed three microbes for varying lignocellulosic components (including CM) in lab-scale trials (Ozbayram et al., 2020; Lee Jonathan et al., 2020). Most of them focused on batch experiments to understand the essential roles of injected microbes and obtained desirable methane yield improvement. However, to mimic the real application strategies in industrial AD installation, one should seek lone- term continuous/semi-continuous experiments to justify the feasibility of injected bioaugmentation dosage. Under such circumstances, the introduced pure culture competes with the indigenous microbes to metabolize organic matter of CM and face the wash-out stress induced by the daily fed-and-withdraw mode adopted in most running AD plants. As a result, the remaining continuous cases rendered limited enhance-ment compared with their batch counterparts, ranging from 0 to 7.5% (Martin-Ryals Ana et al., 2015; Tsapekos et al., 2017; Nielsen et al., 2007). Given that any remedy on bioaugmentation such as immobili-zation/membrane reactor will increase the overall running costs, alter-native modification such as tailor-made complex microbial consortium developed for the individual case seems more tempting, as suggested by

Tsapekos et al. (2017). Recently, our group has developed a complex microbial consortium growing on the fiber fraction of CM. The con-sortium was periodically applied as bioaugmentation dosage in a continuous reactor fed with CM to enhance the hydrolysis and, prob-ably, the methanogenesis (Li et al., 2020). Physiochemical analysis verified that enhanced cellulose removal, followed by improved methane yield, was obtained in the bioaugmented reactor. More importantly, such improvement could partially sustain, reflected by an 11% higher daily methane yield in the bioaugmented reactor compared with the control when bioaugmentation stopped (Table 1).

As a biological reinforcement method, the behaviour of the injected bioaugmentation dosage should be monitored to profile the microbial dynamic. Tsapekos et al. (2017) tracked the microbial community in a continuous digester using 16 s rRNA approaches. They concluded that the injected Clostridium thermocellum vanished after two hydrolytic retention times (HRT) and didn’t trigger any change in the indigenous community. Likewise, Lee Jonathan et al. (2020) reported that the introduction of mixed three microbes hardly shaped the original mi-crobial community. Unlike the studies mentioned above, the mimi-crobial consortium in our study was derived from the engineered reactor fed

with fiber instead of cellulose adopted for pure culture cultivation. We assumed that the bioaugmentation dosage was more robust and could reproduce itself in the host reactor.

Additionally, a metatranscriptomic approach was implemented in this study to deeply understand the entire metabolisms from the perspective of active enzymes. Indeed, one can rely on metagenomic information and the Kegg database to predict the potential metabolism (Villa Montoya Alejandra et al., 2019). More concrete and direct evi-dence can be captured from mRNA carrying the information required to encode relevant proteins (enzymes).

2. Materials and methods

2.1. Sample information

The samples analyzed in this study were collected in a previous study (Li et al., 2020). Briefly, three mesophilic (37˚C) continuous stirring tank reactors (Ra: bioaugmentation dosage provider; Rb: experimental reactor

receiving bioaugmentation dosage; Rc: control reactor receiving

steril-ized bioaugmentation dosage) were established in this study. All re-actors underwent a semi-continuous feeding mode, namely a certain amount of effluent was taken from the reactor before the supply of fresh substrate. The stirring rate in all reactors was set at 120 rpm by motor controllers. Hot water provided by a water bath was circulated through the two-layer water jacket of reactors to maintain the mesophilic con-dition. The working volume of the reactor was 3L. Accordingly, Ra was

initialized with inoculum obtained from an industrial AD plant treating manure. Ra’s feed was neutral detergent fiber derived from CM, while Rb

and Rc’s feed was CM. The HRT for R_a, R_b, and R_cwas 30, 25, and 25 days, respectively. The organic loading rate for the three reactors was 1 gVS/L/d. Before the start-up of the bioaugmentation experiment, Ra has

been stably running for 180 days, providing bioaugmentation dosage capable of utilizing recalcitrant fiber. Meanwhile, Rb and Rc have been

stably running for 100 days, with similar methane yields (Li et al., 2020). From day 100 to day 170, Rb received bioaugmentation dosage

from Ra, while sterilized bioaugmentation dosage was supplied to Rc.

From day 170 to day 220, bioaugmentation stopped, and Rb and Rc were

fed back with CM. On day 130 (when the effect of bioaugmentation emerged in Rb), 170 (the effect of bioaugmentation has lasted for almost

two HRT), and 220 (two HRT after bioaugmentation stopped), samples were taken from Rb and Rc for metatranscriptomic analysis. One sample

from Ra was carried throughout the bioaugmentation test to obtain the

microbial dynamic of bioaugmentation dosage. 2.2. Metatranscriptomic analysis

Once collected, samples were mixed with 100% ethanol at 1:1 (volume basis). Then, a phenol–chloroform isolation method was adopted to extract total RNA out of the biological samples. The extracted RNA initially went through a quality check regarding concentration, RIN, 23S/16S, and size. Additionally, the purity of the samples was tested by NanoDropTM. Total RNA samples were then treated with Ribo- Table 1

Characteristics regarding the bioaugmentation experiment (data was derived from Li et al., 2020).

Chemical

characteristics S Rbioaugmentation dosage) b: d150-d170 (active Rbioaugmentation stopped) b: d180-d220 (when Rbioaugmentation dosage) c: d150-d170 (sterilized Rc: d180-d220 (without bioaugmentation dosage) Daily methane yield

(mL/gVS/d) 92 ±5 179 ± 5 160 ± 2 144 ± 4 144 ± 4 Effluent Cellulose(mg/ g TS) – 75 ± 1 53 ± 3 88 ± 4 63 ± 3 Effluent Hemicellulose (mg/g TS) – 50 ± 5 37 ± 3 51 ± 2 40 ± 1 Total ammonia(mg/L) – 224 ± 8 234 ± 2 252 ± 10 252 ± 10 Free ammonia(mg/L) – 14 ± 1 13 ± 1 16 ± 2 16 ± 2

(4)

Bioresource Technology 330 (2021) 124962

3

off® rRNA Depletion Kit (Bacteria) to deplete rRNA. The remaining mRNA molecules were reversely synthesized into cDNA. Finally, a prokaryotic transcriptome library was constructed to obtain the tran-script sequence information from cDNA amplification.

2.3. Data analysis

The generated raw data of the Fastq-files were directly used for analysis. Specifically, rnaSpades de novo assembler was used to assemble contigs/transcripts (Nurk et al., 2013; Bushmanova et al., 2019). The contigs/transcripts were identified with a (modified) protein-database Uniprotkb/Trembl database (https://www.ebi.ac. uk/uniprot/download-center) using Diamond (Fast and sensitive pro-tein alignment using DIAMOND (Buchfink et al., 2015). Enzyme infor-mation and pathway inforinfor-mation (https://www.genome.jp/kegg/pathw ay.html) was added based on the EC numbers using a Python script (Placzek et al., 2016). Pathways were added only to key enzymes (key- enzymes are present in only one pathway). The contigs/transcripts were also identified against SILVA 132 LSU and SSU reference databases (https://www.arbsilva.de/no_cache/download/archive/release_132 /Exports/) with MEGABLAST (Altschul et al., 1990). SILVA taxonomy was obtained as a result. The EC number of cellulase (EC 3.2.1.4) was linked to species using the entry file downloaded from ENZYME hosts by Expasy (Gasteiger et al., 2003).

For comparison of the abundance of genes encoding for different enzymes, a student T-test was conducted in Excel with a threshold value of 0.05.

3. Results and discussion

3.1. Information of obtained sequences

Accordingly, approximately 60 million RNA sequences were ob-tained from each sample taken from Ra, Rb, and Rc (Table 2). The amino

acid sequences derived from the translated contigs were used to look for homologous proteins in the protein databases and link them to different metabolic pathways. Notably, the identified proteins in Rb and Rc were

assigned to eleven metabolic categories, as shown in Fig. 1. Proteins related to carbohydrate metabolism, amino acid metabolism, energy metabolism, and cofactors and vitamin metabolism were predominant. The predominance of these metabolisms was due to the characteristic of CM since carbohydrate (including lignocellulose) and crude proteins (nitrogenous compounds) are significant components in CM. Moreover, energy metabolism is ubiquitous among cells to maintain their meta-bolic activity, such as methanogenesis (Borrel et al., 2016). For proper cell growth, cofactors, vitamins, and amino acids are inevitably required as well, explaining the detection of relevant metabolism by transcript data (Fig. 1) (Heyer et al., 2019).

3.2. Bacterial phyla

Eleven phyla, which represent more than 1% in at least one sample,

are shown in Fig. 2. Several phyla were found ubiquitous, including Actinobacteria, Bacteroidetes, Chloroflex, Firmicutes, Proteobacteria, and Spirochaetes. Specifically, Actinobacteria, Bacteroidetes, and Fir-micutes were dominant in all samples, indicating their essential roles in utilizing polysaccharides in CM. Actinobacteria contain various genera involved in polysaccharides decomposition (such as cellulose and xylose), hydrolases secretion, and volatile fatty acids (VFAs) production (Chen et al., 2017; Barka et al., 2016). Similarly, microorganisms in Firmicutes are widely detected in lignocellulose-fed reactors and play vital roles in hydrolysis and acidogenesis (Lackner et al., 2020; Wei et al., 2020). Members of Bacteroidetes are broadly distributed among anaerobic habitats and are supposed to degrade proteins and carbohy-drates (Di Maria and Barratta, 2015). Being less abundant, Chloroflex, Proteobacteria, and Spirochaetes account for around 10% within the bacterial community (Fig. 2). Both Spirochaetes and Chloroflexi are suggested to degrade carbohydrates and proteins, while many Proteo-bacteria Proteo-bacteria are believed to grow on glucose and VFAs (Yi et al., 2014; Ariesyady et al., 2007). Actinobacteria gained its predominance in the reactor (Rb) receiving bioaugmentation dosage, reaching around

60% in different tested phases. Whereas in the control reactor (Rc), the

bacterial community was dominated mainly by Bacteroidetes and Fir-micutes, with Actinobacteria less abundant. Thus, the introduction of bioaugmentation dosage might boost the enrichment of Actinobacteria. 3.3. Bacterial genera

The obtained bacterial genera, which account for higher than 0.1% of the total sequences in at least one sample, are presented in Fig. 3A. Among them, bacterial genera involved in cellulose or hemicellulose degradation are the main focus of this study. Hence, the profiles of these genera are illustrated further (Fig. 3B). Initially, Acetivibrio, Actinotalea, Acinetobacter, Aeromicrobium, Bacteroides, Bacillus, Brachybacterium, Cellulomonas, Cellulosilyticum, Clostridium, Flavobacterium, Kocuria, Microbacterium, Micrococcus, Ruminiclostridium, Ruminococcus, Sangui-bacter, Treponema are potential cellulose-degraders (An et al., 2020; Dai et al., 2016; de Lima Brossi et al., 2016; Deng and Wang, 2017; Feng et al., 2017; Greening et al., 2019; Kavitha et al., 2020; Li et al., 2020; Liu et al., 2019; Miller David et al., 2011; Poulsen et al., 2016; Prze-mieniecki Sebastian et al., 2020; Tan et al., 2013; Wenzel et al., 2002; Xu et al., 2012; Zhivin et al., 2017). Due to a fiber-rich diet, these cellulose- degrading bacteria were enriched in Ra (over 50%). Among them,

Actinotalea (11%), Acetivibrio (10%), Bacillus (2%), Bacteroides (7%), Cellulomonas (4%), Clostridium (9%), Ruminiclostridium (5%), Treponema (1%) were predominant in Ra, while the rest of the cellulolytic genera

were less than 1%. More importantly, some of these genera were also enriched in Rb (16%), contributing to the thriving of indigenous

cellu-lolytic microbes (Rc:7%). Specifically, Bacillus, Cellulomonas,

Clos-tridium, and Ruminiclostridium were found higher in Rb than in Rc. In

other words, the injected bioaugmentation dosage might strengthen the activity of these four cellulolytic bacterial genera. To further elucidate the effect of bioaugmentation on cellulose degradation, cellulase (EC 3.2.1.4) were linked to microbial species. As shown in Table 3, some of the bacterial species which excrete cellulase were widely distributed among all treatments, such as Clostridium thermocellum, Clostridium cel-lulovorans, Cellvibrio japonicus, Fibrobacter succinogenes, Ruminiclostri-dium cellulolyticum, and RuminiclostriRuminiclostri-dium josui. The existence of these bacterial species may represent the initial cellulose-degrading capacity of the indigenous microorganisms in Rc. Whereas, some of the cellulase-

producing bacterial species were found exclusively in Ra and Rb,

including Clostridium sp., Clostridium longisporum, Cellulomonas fimi, Bacillus akibai, Bacillus cellulosilyticus, and Bacillus sp. Such observation compared well with the bacterial profile, where enriched Clostridium, Cellulomonas, and Bacillus were found in Rb (Fig. 3B). Therefore, we

demonstrated that the enhanced cellulase activity brought by bio-augmentation was responsible for improved cellulose degradation of CM.

Table 2

Metatranscriptomics results derived from the constructed cDNA libraries based upon the extracted mRNA from Rb and RC at different time points.

Sample RNA sequences RNA contigs Unique contigs

Bacterial Archaeal S 60,113,143 514,208 353,259 25,107 Rb-130 62,543,388 562,317 393,827 34,906 Rb-170 61,341,514 598,499 402,050 38,793 Rb-220 61,293,263 619,068 411,329 39,794 Rc-130 62,528,308 413,497 315,256 27,495 Rc-170 61,574,817 407,645 283,778 25,742 Rc-220 58,753,408 356,068 224,457 21,354 Y. Li et al.

(5)

3.4. Archaeal genera

The distribution pattern of the inhabited archaeal genera in both reactors is shown in Fig. 4. The predominant and ubiquitous archaeal genera, such as uncultured_ Bathyarchaeia (Phylum Bathyarchaeota), Methanoculleus, Methanosaeta, Methanolinea, Methanoregula, and Meth-anospirillum account for over 90% of the Archaea in both reactors, with different shares. Among them, uncultured_ Bathyarchaeia, Meth-anoculleus, Methanospirillum, and Methanosaeta thrived in both Rb and Rc

at different phases, highlighting their contribution to methane produc-tion. Remarkably, the proportion of acetoclastic Methanosaeta fluctuated little in Rb and Rc, suggesting that bioaugmentation may not intervene in

acetoclastic methanogenesis in this study (Gao et al., 2017). Whereas the proportions of hydrogenotrophic methanogens Methanoculleus and Methanospirillum differed significantly in Rb and Rc. Methanoculleus has

been frequently detected in manure and protein-rich reactors and could work with syntrophic bacteria (Li et al., 2020; Tian et al., 2019). While Methanospirillum is an efficient H2 utilizer probably associated with

lignocellulose degradation (Li et al., 2018). According to Nielsen et al. (2007), improved lignocellulose degradation could lead to more release of H2, which should have stimulated the growth of hydrogenotrophic

methanogens. In this sense, the decline of Methanoculleus and

Methanospirillum in Rb might imply that part of their methanogenic

metabolism was taken over by uncultured_ Bathyarchaeia. In other words, uncultured_ Bathyarchaeia might have a higher affinity with CO2 and H2

than Methanoculleus and Methanospirillum. Besides this study, uncultured_ Bathyarchaeia has been abundantly identified in other CM-fed AD sys-tems, underlying its fundamental role in the methanogenesis of AD treating CM (Dong et al., 2019; Wang et al., 2020). Till now, there have been 25 subgroups identified in Bathyarchaeota. These distinct bathy-archaeotal subgroups managed to adjust to marine and freshwater en-vironments (Zhou et al., 2018). Moreover, these strictly-anaerobic members can utilize 1) polymeric carbohydrates, 2) proteins, 3) fatty acids/aromatic compounds, 4) methane (or short-chain alkane) and methylated compounds, and/or 5) potentially other organic matter. Hence, the comprehensive metabolic capabilities of bathyarchaeotal members enable them to behave more than methanogens, but some-times acetogens interacting with acetoclastic methanogens (He et al., 2016). According to the genomic evidence, an energy conservation pathway named acetyl-coenzyme A-centralized heterotrophic pathway has been proposed to function in uncultured_ Bathyarchaeia, enabling its utilization of CO2 and H2 to autotrophic metabolism (Zhou et al., 2018).

As a result, in bioaugmentation tests, the strengthened uncultured_ Bathyarchaeia might compete to assimilate CO2 and H2, causing the

Fig. 1. Overall metabolic pathways identified in the microbial community of Ra (S), Rb (Bio-130, Bio-170, and Bio-220) and Rc (Con-130, Con-170 and Con-220).

(6)

5

Fig. 3. Predominant bacterial genera (A) and potential cellulose-degrading bacterial genera (B) obtained from Ra (S), Rb (Bio-130, Bio-170, and Bio-220) and Rc

(Con-130, Con-170 and Con-220) (Note: numbers in (a) indicate the relative abundance of certain bacterial genera, from low relative abundance to high relative abundance, the colour gradually transforms from green to yellow to red).

(7)

decline of other hydrogenotrophic methanogens (in our case Meth-anoculleus and Methanospirillum). As hypothesized in our previous research, uncultured_ Bathyarchaeia owned a distinguished metabolic potential among Archaea and was closely related to lignocellulose degradation and subsequent methanogenesis (Li et al., 2020). Accord-ingly, the introduced bioaugmentation dosage promoted the growth of uncultured_ Bathyarchaeia, reaching a predominant share (65%) even when the bioaugmentation stopped (Fig. 4). Admittedly, a high per-centage of uncultured_ Bathyarchaeia didn’t necessarily indicate a higher methane yield (Table 1). The introduction of bioaugmentation biomass enriched from a complex microbial pool (inoculum from an industrial AD plant treating manure) may present a sustained modification of the indigenous archaeal community. This modification was rarely shown in other continuous cases, where pure cultures were introduced (Martin- Ryals Ana et al., 2015; Tsapekos et al., 2017; Nielsen et al., 2007).

3.5. Metabolic pathway involved in AD of CM

Based on transcript data, the metabolic flows in AD of the lignocel-lulosic CM are demonstrated. For AD of fibrous waste, hydrolysis of recalcitrant lignocellulose is vital and regarded as the rate-limiting step. Hydrolysis relies on the collaborative action of complex arrays of en-zymes to decompose cellulose or hemicellulose. The metatranscriptomic analysis revealed that debranching enzymes such as acetylxylan esterase (EC 3.1.1.72), arabinofuranosidase (EC 3.2.1.185/3.2.1.55), and car-boxylesterase (EC 3.1.1.1), which eliminate the side-chain units of the xylan backbone, were actively expressed. Afterward, cellulose and hemicellulose are more accessible by hydrolytic enzymes. The

metatranscriptomic analysis also unveiled the activity of carbohydrate esterases (CE), which remove ester linkages in residual sugars to make their glycosidic linkages easily approachable by glycosidic hydrolases (EC 3.2.1.-) (Biely, 2012). Hence, the existence of debranching enzymes lays the foundation for subsequent cellulose/hemicellulose metabolism by varying microbial enzymes (i.e., cellulosomes). Accordingly, thor-ough decomposition of cellulose requires joint efforts from cellulose- related enzymes (cellulase, exo-glucanase, and β-glucosidase). Particu-larly, RNA transcripts coding for cellulase (EC 3.2.1.4) was found most abundant in Ra, verifying the promotive cellulose-degrading potential of

the bioaugmentation dosage (Table 4). Moreover, higher cellulase ac-tivity was identified in Rb than Rc (p<0.05) (Table 4). Enhanced

cellu-lase activity could guarantee an improved break-up of the internal bonds in the amorphous cellulose areas, thus exposing individual ends of polysaccharide chains. Then, glucan 1,4-β-glucosidase (EC 3.2.1.91) cleaves 2–4 blocks from the terminal of the exposed chains to release cellobiose. Its activity was found higher in Rb than Rc as well (p<0.05).

The followed-up β-glucosidase (EC 3.2.1.21) hydrolyzes the released cellobiose into individual glucose molecules for subsequent acido-genesis. The comparable activity of this enzyme, however, was obtained in Rb and Rc (p˃0.05). Therefore, bioaugmentation in this study boosted

mainly the activity of cellulase and exo-glucanase, while it had a minor effect on β-glucosidase. For hemicellulose (with xylan the principal component) degradation, xylanase (endo-1,4-β-xylanase, EC 3.2.1.8) is inevitably required to randomly cleave β-1,4 glycosidic linkages of the backbone structure of β-1,4-xylans, resulting in xylose and xylooligo-saccharides. Then, acetylxylan esterase (EC 3.1.1.72), α-D-glucuroni-dases (EC 3.2.1.1), xylan 1,4-β -xylosidase (EC 3.2.1.37), and α-L- Table 3

Cellulase excreted by certain microorganisms revealed by metatranscriptomics (Note: ‘+’ means the existence of cellulase possessed by bacterial microbes, ‘-’ means the absence of cellulase possessed by bacterial microbes, numbers in the parentheses suggest the frequency of the detected match between cellulase and the corre-sponding bacterial species).

Microorganisms S Rb-130 Rb-170 Rb-220 Rc-130 Rc-170 Rc-220 Clostridium thermocellum +(51) +(97) +(61) +(45) +(59) +(26) +(14) Clostridium sp. +(2) +(3) +(2) +(2) – – – Clostridium cellulovorans +(11) +(12) +(8) +(2) +(12) +(4) – Clostridium longisporum +(21) +(13) +(7) +(9) – – – Cellvibrio japonicus +(8) +(5) +(14) +(11) +(9) +(5) +(4) Cellulomonas fimi +(4) +(7) +(3) +(2) – – – Bacillus akibai +(2) +(3) +(8) +(4) – – – Bacillus cellulosilyticus +(5) +(9) +(6) +(5) – – – Bacillus sp. +(3) +(7) +(3) +(2) – – – Fibrobacter succinogenes +(6) +(3) +(5) +(2) +(4) +(3) +(2) Ruminiclostridium cellulolyticum +(12) +(23) +(9) +(14) +(20) +(7) +(3) Ruminiclostridium josui +(8) +(16) +(5) +(9) +(14) +(11) +(2)

Fig. 4. Relative abundance of Archaea known to be involved in methanogenesis from Ra (S), Rb (Bio-130, Bio-170, and Bio-220) and Rc (Con-130, Con-170 and

(8)

7

arabinofuranosidase (EC 3.2.1.55) collaboratively cleave the connec-tions among the xylose monomers and the xylan side chains for further utilization (Thapa et al., 2020). Despite the highest xylanase activity in Ra, comparable xylanase activity among Rb and Rc was obtained,

sug-gesting the limited enhancement of bioaugmentation on hemicellulose degradation (p˃0.05) (Table 1 and Table 4). Probably, the highly branched structure, low crystallinity, and heterogeneous nature of hemicellulose already make it easily degradable for hydrolytic microbes (Li et al., 2020). Collectively, the obtained transcript information on hydrolytic enzymes underlined the significance of the specifically enriched microbial consortia (complex microbes grow on fiber) to boost cellulolytic enzymes involved in the hydrolysis of recalcitrant lignocel-lulosic compounds in CM.

The transcripts also revealed enzymes responsible for converting hydrolytic products into acetyl-coenzyme A (acetyl-CoA), a critical intersection in the network of metabolic pathways. Briefly, two in-termediates β-D-fructose 6-phosphate and α-D-glucose 6-phosphate originating from cellulose/hemicellulose degradation are separately catalyzed into pyruvate via a series of enzymes (For cellulose-derived β-D-fructose 6-phosphate and α-D-glucose 6-phosphate: Hexokinase (EC 2.7.1.1), Glucokinase (EC 2.7.1.2), Glucose-6-phosphate isomerase (EC 5.3.1.9), 6-phosphofructokinase (EC 2.7.1.11), ADP-specific phos-phofructokinase (EC 2.7.1.146), Fructose-bisphosphate aldolase (EC 4.1.2.13), Glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12), Glyceraldehyde-3-phosphate dehydrogenase (NAD(P)(+)) (EC 1.2.1.59), Phosphoglycerate kinase (EC 2.7.2.3), Phosphoglycerate Table 4

Key enzymes involved in the metabolism of lignocellulose of CM and their activity (Important enzymes involved in hydrolysis of lignocellulose of CM are highlighted in red, enzymes involved in methanogenesis are highlighted in blue; DL: detection limitation).

EC Name S Bio130 Bio170 Bio220 Con130 Con170 Con220

1.2.1.12 Glyceraldehyde-3-phosphate dehydrogenase 0.089 0.068 0.062 0.099 0.081 0.111 0.067 1.1.1.49 Glucose-6-phosphate dehydrogenase (NADP(+)) 0.098 0.089 0.052 0.048 0.050 0.025 <DL

1.2.1.59 Glyceraldehyde-3-phosphate dehydrogenase (NAD(P)(+)) 0.159 0.095 0.052 0.053 0.035 0.061 <DL

1.2.7.1 Pyruvate synthase 0.137 0.187 0.214 0.139 0.097 0.312 0.001 2.2.1.1 Transketolase 0.293 0.141 0.145 0.222 0.162 0.244 0.186 2.7.1.11 6-phosphofructokinase 0.221 0.193 0.185 0.161 0.143 0.132 0.018 2.7.1.146 ADP-specific phosphofructokinase 0.011 0.014 0.010 0.005 0.002 0.005 0.001 2.7.1.1 Hexokinase 0.001 <DL 0.001 <DL <DL <DL <DL 2.7.1.2 Glucokinase 0.009 0.029 0.039 0.079 0.024 0.089 0.028 2.7.2.3 Phosphoglycerate kinase 0.400 0.369 0.422 0.549 0.311 0.412 0.151 2.7.1.40 Pyruvate kinase 0.112 0.09 0.081 0.105 0.066 0.107 0.073 3.1.1.31 6-phosphogluconolactonase 0.005 0.01 0.012 0.015 0.018 0.01 <DL 3.2.1.21 Beta-glucosidase 0.219 0.254 0.317 0.314 0.228 0.304 0.316 3.2.1.37 Xylan 1,4-beta-xylosidase 0.415 0.317 0.225 0.396 0.148 0.396 0.126 3.2.1.55 alpha-L-arabinofuranosidase 0.297 0.165 0.161 0.273 0.163 0.245 0.045 3.2.1.91 Cellulose 1,4-beta-cellobiosidase (non-reducing end) 0.358 0.179 0.151 0.122 0.047 0.038 0.055 3.2.1.4 Cellulase 2.03 1.081 0.969 0.917 0.677 0.665 0.511 3.2.1.8 Endo-1,4-beta-xylanase 3.93 1.417 1.381 1.278 1.47 1.17 0.971 4.1.2.13 Fructose-bisphosphate aldolase 0.718 0.351 0.442 0.452 0.377 0.308 0.101 4.1.2.14 2-dehydro-3-deoxy-phosphogluconate aldolase 0.011 0.011 0.003 0.005 0.01 0.01 <DL 4.2.1.11 Phosphopyruvate hydratase 0.706 0.783 0.785 0.713 0.525 0.630 0.282 4.2.1.12 Phosphogluconate dehydratase 0.001 0.001 <DL 0.001 0.001 0.001 <DL 5.1.3.1 Phosphoketolase 0.099 0.086 0.069 0.069 0.063 0.068 0.041 5.3.1.6 Ribose-5-phosphate isomerase 0.027 0.035 0.074 0.058 0.054 0.084 0.016 5.3.1.9 Glucose-6-phosphate isomerase 0.217 0.167 0.196 0.180 0.165 0.135 0.067 5.4.2.11 Phosphoglycerate mutase 0.063 0.05 0.04 0.051 0.018 0.071 0.056 1.1.1.35 3-hydroxyacyl-CoA dehydrogenase 0.013 0.007 0.012 0.026 0.003 0.003 <DL 1.12.98.1 Coenzyme F420 hydrogenase 0.073 0.306 0.442 0.264 0.009 0.065 0.089 1.12.98.2 5,10-methenyltetrahydromethanopterin hydrogenase <DL <DL <DL <DL <DL <DL <DL 1.17.1.10 Formate dehydrogenase 0.002 0.004 0.002 0.008 0.003 0.009 0.002 1.2.7.12 Formylmethanofuran dehydrogenase 0.002 <DL 0.001 0.002 <DL <DL <DL

1.2.7.4 Anaerobic carbon-monoxide dehydrogenase 0.093 0.256 0.296 0.259 0.143 0.564 0.215 1.5.1.20 Methylenetetrahydrofolate reductase (NAD(P)H) 0.087 0.055 0.053 0.062 0.039 0.092 <DL

1.5.1.5 Methylenetetrahydrofolate dehydrogenase (NADP(+)) 0.166 0.113 0.122 0.134 0.133 0.140 0.103 1.5.98.1 Methylenetetrahydromethanopterin dehydrogenase 0.296 0.461 0.354 0.389 0.320 0.392 0.314 1.5.98.2 5,10-methylenetetrahydromethanopterin reductase 0.433 0.900 0.886 1.152 0.754 1.197 0.932 2.1.1.245 5-methyltetrahydrosarcinapterin:corrinoid/iron-sulfur protein, Co-methyltransferase 0.032 0.027 0.048 0.071 0.020 0.014 0.064 2.1.1.246 [Methyl-Co(III) methanol-specific corrinoid protein]:coenzyme Mmethyltransferase <DL <DL <DL <DL <DL 0.002 <DL

2.1.1.247 [Methyl-Co(III) methylamine-specific corrinoid protein]:coenzyme

Mmethyltransferase 0.002 0.001 0.001 0.001 0.001

<DL <DL

2.1.1.248 [Methylamine–corrinoid protein] Co-methyltransferase 0.001 0.002 0.004 0.001 <DL 0.004 <DL

2.1.1.250 [Trimethylamine–corrinoid protein] Co-methyltransferase <DL 0.001 <DL <DL <DL <DL <DL

2.1.1.90 Methanol–corrinoid protein Co-methyltransferase 0.303 1.51 0.78 0.55 0.59 0.68 0.77 2.1.1.258 5-methyltetrahydrofolate:corrinoid/iron-sulfur protein, Co-methyltransferase 0.005 0.006 0.004 0.001 0.007 0.007 <DL

2.1.1.86 Tetrahydromethanopterin S methyltransferase 1.434 1.508 1.784 1.550 0.597 0.681 0.780 2.3.1.101 Formylmethanofuran–tetrahydromethanopterin N-formyltransferase 0.064 0.086 0.074 0.083 0.065 0.057 <DL

2.3.1.169 CO-methylating acetyl-CoA synthase 0.518 0.179 0.178 0.155 0.018 0.044 0.032 2.3.1.8 Phosphate acetyltransferase 0.059 0.056 0.068 0.085 0.021 0.014 0.034 2.3.1.9 Acetyl-CoA C-acetyltransferase 0.089 0.077 0.098 0.076 0.069 0.051 0.116 2.7.2.1 Acetate kinase 0.157 0.175 0.162 0.203 0.105 0.122 0.111 2.8.3.8 Acetate CoA-transferase 0.021 0.021 0.050 0.029 0.003 0.022 <DL 2.8.4.1 Coenzyme-B sulfoethylthiotransferase 9.461 2.526 2.808 1.515 0.818 0.891 0.605 3.5.4.27 Methenyltetrahydromethanopterin cyclohydrolase 0.041 0.055 0.044 0.046 0.041 0.036 <DL 3.5.4.9 Methenyltetrahydrofolate cyclohydrolase 0.003 0.018 0.026 0.053 0.051 0.047 <DL 4.2.1.17 Enoyl-CoA hydratase 0.031 0.088 0.070 0.103 0.084 0.052 0.072 6.2.1.1 Acetate–CoA ligase 2.192 1.833 1.928 1.739 1.037 1.190 0.749 6.3.4.3 formate-tetrahydrofolate ligase 0.154 0.21 0.173 0.181 0.177 0.41 0.057 Y. Li et al.

(9)

mutase (EC 5.4.2.11), Phosphopyruvate hydratase (EC 4.2.1.11), and Pyruvate kinase (EC 2.7.1.40); For hemicellulose-derived β-D-fructose 6- phosphate and α-D-glucose 6-phosphate: Transketolase (EC 2.2.1.1), Glucose-6-phosphate isomerase (EC 5.3.1.9), Glucose-6-phosphate de-hydrogenase (NADP(+)) (EC 1.1.1.49), 6-phosphogluconolactonase (EC 3.1.1.31), Phosphogluconate dehydratase (EC 4.2.1.12), and 2-dehydro- 3-deoxy-phosphogluconate aldolase (EC 4.1.2.14)). The generated py-ruvate was further catalyzed into acetyl-CoA via pypy-ruvate-ferredoxin oxidoreductase (EC 1.2.7.1). Throughout these processes, no apparent difference in enzymatic activity was identified in Rb and Rc, implying

indigenous microorganisms’ competence in fermenting easily- degradable carbohydrates of CM (p˃0.05) (Table 4).

Although protein metabolism is not the priority of this study, it de-serves illustration as it is a non-negligible contributor to methane pro-duction in AD of CM. Initially, protein in CM is hydrolyzed by varing peptidases (3.4.-.-) into different amino acids based on the sepecificity of peptidases. Subsequently, bunches of enzymes convert varying amino acids into pyruvate or acetyl-CoA (data not shown). Similar to ligno-cellulose metabolism, protein-derived pyruvate or acetyl-CoA is further catalyzed by enzymes that are involved in methanogenesis. Further-more, chemical characteristics indicated that bioaugmentation didn’t promote the degradation of protein of CM, which was understandable as the bioaugmentation dosage was not intentionally cultivated for protein degradation (Table 1). Such observation was also backed up by the transcript data, as no apparent difference of enzymatic activity was obtained in amino-acids metabolism (p˃0.05; data not shown).

Basically, AD is such a bio-process carrying out carbon fixation, with methane the end-product. Therefore, a clear demonstration of meta-bolism that occurs in methanogenesis will further justify bio-augmentation. In this study, enzymes for acetoclastic and hydrogenotrophic methanogenesis were active, indicating that both types of methanogens were involved in methane metabolism (Table 4). Through acetoclastic methanogenesis, the resultant acetate is ultimately converted into methane via consecutive steps (Phosphate acetyl-transferase (EC 2.3.1.8), Acetate kinase (EC 2.7.2.1), Acetate–CoA ligase (EC 6.2.1.1), CO-methylating acetyl-CoA synthase (EC 2.3.1.169), 5- methyltetrahydrosarcinapterin:corrinoid/iron-sulfur protein, Co- methyltransferase (EC 2.1.1.245), Tetrahydromethanopterin S methyl-transferase (EC 2.1.1.86), and Coenzyme-B sulfoethylthiomethyl-transferase (EC 2.8.4.1). While for hydrogenotrophic methanogenesis, a process named Wood–Ljungdahl (WL) pathway should be underlined, owing to its vital roles for energy generation and carbon fixation in Archaea. Initially, hydrogenotrophic methanogens (in this study Methanoculleus and Methanospirillum) stepwise reduce CO2 to the Methyl-H4MPT via

Methyl-Branch of the archaeal type WL pathway (MBWL: contains for-mylmethanofuran dehydrogenase (EC 1.2.7.12); forfor-mylmethanofuran– tetrahydromethanopterin N-formyltransferase (EC 2.3.1.101); meth-enyltetrahydromethanopterin cyclohydrolase (EC 3.5.4.27) methyl-enetetrahydromethanopterin dehydrogenase (EC 1.5.98.1) and 5,10- methylenetetrahydromethanopterin reductase (EC 1.5.98.2)). The methyl group of the formed Methyl-H4MPT is then transferred to

co-enzyme M, yielding methyl-S-CoM catalyzed by N5

-methyltetrahy-dromethanopterin (MTR; EC 2.1.1.86). Finally, methyl-S-CoM is reduced with coenzyme B to methane and a heterodisulfide (CoM–S–S–CoB) by methyl-coenzyme M reductase complex (MCR; EC 2.8.4.1). For potential metabolism of uncultured_ Bathyarchaeia in this study, it is noteworthy that the pure culture in Bathyarchaeota is still absent. Thus, for Bathyarchaeota members, most of their potential functions are inferred from the reconstructed genomes. Recently, the reconstruction of two Bathyarchaeota members (BA1 and BA2 with the completeness of their draft genomes 91.6% and 93.8%, respectively) has allowed researchers to deduce their potential methane-related meta-bolic capabilities (Evans Paul et al., 2015). Similar to hydrogenotrophic methanogens, BA1 and BA2 were proposed to perform MBWL. Mean-while, they hypothesized methylotrophic pathway (methanol: CoM methyltransferase (MTA) coupled with MCR) instead of the

hydrogenotrophic pathway as a methane-producing pathway due to the suspicion of the existence of genes encoding MTR. However, in this study, the enriched uncultured_ Bathyarchaeia (97%), methane produc-tion, as well as the detected activity of MTR in Ra revealed its potential

association of the archaeal WL pathway with methanogenesis, as pre-dicted by Borrel et al. (2016) (Table 1 and Table 4). Thus, one of the Bathyarchaeota groups’ functions in this study could be the same as hydrogenotrophic methanogens (Methanoculleus and Methanospirillum). Moreover, the highest activity of MCR was identified in Ra (9.461),

followed by Rb (1.515–2.808) and Rc (0.605–0.891), which compared

well with corresponding proportions of uncultured_ Bathyarchaeia (Fig. 4). As the last step of methane synthesis, the high activity potential of MCR might be beneficial to methane production, leading to improved methane yield in Rb than Rc (Table 1). Besides methanogenesis,

uncul-tured_ Bathyarchaeia might also contribute to acetogenesis by acting as ‘acetogenic archaea,’ reflected by the highest activity of CO-methylating acetyl-CoA synthase (EC 2.3.1.169) in Ra (Table 4). Probably, the energy

source of uncultured_ Bathyarchaeia is not limited to MTR (energy-saving complex), but acetyl-CoA, highlighting its versatile roles in AD (Evans Paul et al., 2015; He et al., 2016). To date, acetogens are well adapted to low-energy environments, using the simplest CO2-fixing pathway (the

WL pathway) for energy production and biosynthesis. On top of that, acetogenic archaea might have an energetic advantage over acetogenic bacteria. Since they do not have to invest ATP to activate formate, which could be a competitive advantage under energy-limiting conditions such as AD (Drake Harold et al., 2002). Hence, the existence of uncultured_ Bathyarchaeia, especially its prevalence in Ra and Rb may come from its

robustness metabolism in AD. In other words, uncultured_ Bathyarchaeia could act more than heterotrophic archaea but as autotrophic acetogens to obtain energy for self-growth. By this mechanism, we might explain the sustained predominance of uncultured_ Bathyarchaeia in Rb, which

otherwise might be outcompeted by indigenous microbes. Therefore, we proposed the fundamentally dual function (hydrogenotrophic and ace-togenic archaea) of uncultured_ Bathyarchaeia in this study, which might enrich the variety of metabolism of Bathyarchaeota members in the bio- system (Evans Paul et al., 2015; He et al., 2016; Borrel et al., 2016). In this context, uncultured_ Bathyarchaeia could also obtain energy (ATP) by converting acetyl-CoA to acetate (Fuchs, 2011). Thus, a higher abun-dance of transcripts coding for acetate kinase (EC 2.7.2.1) and phos-phate acetyltransferase (EC 2.3.1.8) in Rb than in Rc might back up the

suggested ‘acetogenic’ potential of uncultured_ Bathyarchaeia (Table 4) (p<0.05). Taken into account acetate-CoA ligase (EC 6.2.1.1), the high activity of these enzymes in Rb might suggest improved acetoclastic

methanogenesis as well (Table 4). Collectively, the injected uncultured_ Bathyarchaeia might enhance both hydrogenotrophic and acetoclastic methanogenesis, leading to the enhanced activity of MCR, which was in line with corresponding methane profiles (Table 1 and Table 4).

4. Conclusion

Extensive modification of indigenous microbial guilds was observed when bioaugmentation was conducted in a continuous process treating CM. High activity of cellulase was identified in the augmented reactor, which could explain the improved cellulose degradation. Dominant archaea shifted from Methanoculleus and Methanospirillum to uncultured_ Bathyarchaeia during bioaugmentation, most likely due to the high af-finity of CO2 and H2 of uncultured_ Bathyarchaeia. This study concluded

that an enhanced degradation of lignocellulose and methane yield of CM could be achieved simultaneously via complex microbial culture enriched from fiber fraction of CM. Such biological intervention can sustainably enhance AD of CM.

CRediT authorship contribution statement

Yu Li: Conceptualization, Methodology, Investigation, Data

(10)

9

Conceptualization, Methodology. Zhenhua Zhang: Methodology.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors gratefully acknowledge Gert-Jan Euverink for RNA sequencing support and Bert Geurkink for technical help on transcript data analysis.

Funding

Part of this work was supported by Chinese scholarship council (CSC).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.biortech.2021.124962.

References

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local

alignment search tool. J. Mol. Biol. 215 (3), 403–410.

An, Q., Cheng, J.-R., Wang, Y.-T., Zhu, M.-J., 2020. Performance and energy recovery of single and two stage biogas production from paper sludge: Clostridium thermocellum

augmentation and microbial community analysis. Renewable Energy 148, 214–222.

Ariesyady, H.D., Ito, T., Okabe, S., 2007. Functional bacterial and archaeal community structures of major trophic groups in a full-scale anaerobic sludge digester. Water

Res. 41 (7), 1554–1568.

Barka, E.A., Vatsa, P., Sanchez, L., Gaveau-Vaillant, N., Jacquard, C., Klenk, H.-P., Cl´ement, C., Ouhdouch, Y., van Wezel, G.P., 2016. Taxonomy, Physiology, and

Natural Products of Actinobacteria. Microbiol. Mol. Biol. Rev. 80 (1), 1.

Biely, P., 2012. Microbial carbohydrate esterases deacetylating plant polysaccharides.

Biotechnol. Adv. 30 (6), 1575–1588.

Borrel, G., Adam, P.S., Gribaldo, S., 2016. Methanogenesis and the Wood-Ljungdahl Pathway: An Ancient, Versatile, and Fragile Association. Genome Biology and

Evolution. 8 (6), 1706–1711.

Buchfink, B., Xie, C., Huson, D.H., 2015. Fast and sensitive protein alignment using

DIAMOND. Nat. Methods 12 (1), 59–60.

Bushmanova, E., Antipov, D., Lapidus, A., Prjibelski, A.D., 2019. rnaSPAdes: a de novo

transcriptome assembler and its application to RNA-Seq data. GigaScience 8 (9).

Chen, Y., Jiang, X., Xiao, K., Shen, N., Zeng, R.J., Zhou, Y., 2017. Enhanced volatile fatty acids (VFAs) production in a thermophilic fermenter with stepwise pH increase – Investigation on dissolved organic matter transformation and microbial community

shift. Water Res. 112, 261–268.

Dai, Y., Yan, Z., Jia, L., Zhang, S., Gao, L., 2016. The composition, localization and function of low-temperature-adapted microbial communities involved in methanogenic degradations of cellulose and chitin from Qinghai-Tibetan Plateau

wetland soils. J. Appl. Microbiol. 121 (1), 163–176.

de Lima Brossi, M.J., Jimenez, D.J., Cortes-Tolalpa, L., van Elsas, J.D., 2016. Soil-derived microbial consortia enriched with different plant biomass reveal distinct players

acting in lignocellulose degradation. Microbial Ecology. 71 (3), 616–627.

Deng, Y.-J., Wang, S.Y., 2017. Complex carbohydrates reduce the frequency of antagonistic interactions among bacteria degrading cellulose and xylan. FEMS

Microbiol. Lett. 364 (5).

Di Maria, F., Barratta, M., 2015. Boosting methane generation by co-digestion of sludge with fruit and vegetable waste: Internal environment of digester and methanogenic

pathway. Waste Manage. 43, 130–136.

Dong, L., Cao, G., Guo, X., Liu, T., Wu, J., 2019. Efficient biogas production from cattle manure in a plug flow reactor: A large scale long term study. Bioresour. Technol.

278, 450–455.

Drake Harold, L.H., Kusel, K., Matthies, C., 2002. Ecological consequences of the phylogenetic and physiological diversities of acetogens. Antonie van Leeuwenhoek.

81 (1–4), 203–213.

Evans Paul, N.P., Parks, D.H., Chadwick, G.L., Robbins, S.J., Orphan, V.J., 2015. Methane metabolism in the archaeal phylum Bathyarchaeota revealed by genome-

centric metagenomics. Science 350 (6259), 434–438.

Feng, L., Chen, K., Han, D., Zhao, J., Lu, Y.i., Yang, G., Mu, J., Zhao, X., 2017. Comparison of nitrogen removal and microbial properties in solid-phase denitrification systems for water purification with various pretreated lignocellulosic

carriers. Bioresour. Technol. 224, 236–245.

Fuchs, G., 2011. Alternative pathways of carbon dioxide fixation: insights into the early

evolution of life? Annu. Rev. Microbiol. 65, 631–658.

Gao, Y., Yang, A., Bao, J., Ma, R., Yan, L., 2017. Bioreactor performance and microbial community dynamics in a production-scale biogas plant in northeastern China. Int.

J. Agric. Biol. Eng. 10 (1), 191–201.

Gasteiger, E., Gattiker, A., Hoogland, C., Ivanyi, I., Appel, R.D., Bairoch, A., 2003. ExPASy: the proteomics server for in-depth protein knowledge and analysis. Nucleic

Acids Res. 31 (13), 3784–3788.

Greening, C., Geier, R., Wang, C., Woods, L.C., Morales, S.E., 2019. Diverse hydrogen production and consumption pathways influence methane production in ruminants.

ISME J. 13 (10), 2617–2632.

He, Y., Li, M., Perumal, V., Feng, X., Fang, J., Xie, J., Sievert, S.M., Wang, F., 2016. Genomic and enzymatic evidence for acetogenesis among multiple lineages of the archaeal phylum Bathyarchaeota widespread in marine sediments. Nat. Microbiol. 1

(6), 16035.

Heyer, R., Schallert, K., Siewert, C., Kohrs, F., Greve, J., Maus, I., Klang, J., Klocke, M., Heiermann, M., Hoffmann, M., Püttker, S., Calusinska, M., Zoun, R., Saake, G., Benndorf, D., Reichl, U., 2019. Metaproteome analysis reveals that syntrophy, competition, and phage-host interaction shape microbial communities in biogas

plants. Microbiome. 7 (1), 69.

Kavitha, S., Kannah, R.Y., Kasthuri, S., Gunasekaran, M., Pugazhendi, A., 2020. Profitable biomethane production from deligninfied rice straw biomass: the effect of

lignin, energy and economic analysis. Green Chem. 22 (22), 8024–8035.

Lackner, N., Wagner, A.O., Illmer, P., 2020. Effect of sulfate addition on carbon flow and microbial community composition during thermophilic digestion of cellulose. Appl.

Microbiol. Biotechnol. 104 (10), 4605–4615.

Lee Jonathan, T.E., Wang, Q., Lim, E.Y., Liu, Z., He, J., 2020. Optimization of bioaugmentation of the anaerobic digestion of Axonopus compressus cowgrass for

the production of biomethane. J. Cleaner Prod. 258.

Li, W., Khalid, H., Zhu, Z., Zhang, R., Liu, G., Chen, C., Thorin, E., 2018. Methane production through anaerobic digestion: Participation and digestion characteristics

of cellulose, hemicellulose and lignin. Appl. Energy 226, 1219–1228.

Li, Y., Zhao, J., Achinas, S., Zhang, Z., Krooneman, J., Euverink, G.J.W., 2020. The biomethanation of cow manure in a continuous anaerobic digester can be boosted via a bioaugmentation culture containing Bathyarchaeota. Sci. Total Environ. 745,

141042.

Li, Y., Zhao, J., Krooneman, J., Euverink, G.J.W., 2021. Strategies to boost anaerobic digestion performance of cow manure: Laboratory achievements and their full-scale

application potential. Sci. Total Environ. 755, 142940.

Liu, N., Fosses, A., Kampik, C., Parsiegla, G., Denis, Y., Vita, N., Fierobe, H.-P., Perret, S., 2019. In vitro and in vivo exploration of the cellobiose and cellodextrin phosphorylases panel in Ruminiclostridium cellulolyticum: implication for cellulose catabolism. Biotechnol. Biofuels 12 (1). https://doi.org/10.1186/s13068-019-1549- x.

Martin-Ryals Ana, A., Schideman, L., Li, P., Wilkinson, H., Wagner, R., 2015. Improving anaerobic digestion of a cellulosic waste via routine bioaugmentation with

cellulolytic microorganisms. Bioresour. Technol. 189, 62–70.

Miller David, A.D., Suen, G., Bruce, D., Copeland, A., Cheng, J.-F., 2011. Complete genome sequence of the cellulose-degrading bacterium Cellulosilyticum lentocellum.

J. Bacteriol. 193 (9), 2357–2358.

Nielsen, H.B., Mladenovska, Z., Ahring, B.K., 2007. Bioaugmentation of a two-stage thermophilic (68 degrees C/55 degrees C) anaerobic digestion concept for improvement of the methane yield from cattle manure. Biotechnol. Bioeng. 97 (6),

1638–1643.

Nurk, S., Bankevich, A., Antipov, D., Gurevich, A.A., Korobeynikov, A., 2013. Assembling single-cell genomes and mini-metagenomes from chimeric MDA products.

J. Comput. Biol. 20 (10), 714–737.

Ozbayram, E.G., Kleinsteuber, S., Nikolausz, M., 2020. Biotechnological utilization of animal gut microbiota for valorization of lignocellulosic biomass. Appl. Microbiol.

Biotechnol. 104 (2), 489–508.

Placzek, S., Schomburg, I., Chang, A., Jeske, L., Ulbrich, M., Tillack, J., Schomburg, D., 2016. BRENDA in 2017: new perspectives and new tools in BRENDA. Nucleic Acids

Res. 45 (D1), D380–D388.

Poulsen, H.V., Willink, F.W., Ingvorsen, K., 2016. Aerobic and anaerobic cellulase

production by Cellulomonas uda. Arch. Microbiol. 198 (8), 725–735.

Przemieniecki Sebastian, W.S., Kosewska, A., Ciesielski, S., Kosewska, O., 2020. Changes in the gut microbiome and enzymatic profile of Tenebrio molitor larvae biodegrading cellulose, polyethylene and polystyrene waste. Environmental Pollution. 256,

113265.

Thapa, S., Mishra, J., Arora, N., Mishra, P., Li, H., O′_{Hair, J., Bhatti, S., Zhou, S., 2020.}

Microbial cellulolytic enzymes: diversity and biotechnology with reference to lignocellulosic biomass degradation. Reviews in Environmental Science and Bio/

Technology. 19 (3), 621–648.

Tan, H., Deng, Q., Cao, L., 2013. Ruminant feces harbor diverse uncultured symbiotic

actinobacteria. World J. Microbiol. Biotechnol. 30 (3), 1093–1100.

Tian, H., Mancini, E., Treu, L., Angelidaki, I., Fotidis, I.A., 2019. Bioaugmentation strategy for overcoming ammonia inhibition during biomethanation of a protein-rich

substrate. Chemosphere 231, 415–422.

Tsapekos, P.P., Kougias, P.G., Vasileiou, S.A., Treu, L., Campanaro, S., 2017. Bioaugmentation with hydrolytic microbes to improve the anaerobic

biodegradability of lignocellulosic agricultural residues. Bioresour. Technol. 234,

350–359.

Villa Montoya Alejandra, C., da Silva Mazareli, R.C., Delforno, T.P., Centurion, V.B., Sakamoto, I.K. 2019. Hydrogen, alcohols and volatile fatty acids from the co- digestion of coffee waste (coffee pulp, husk, and processing wastewater) by applying autochthonous microorganisms. International Journal of Hydrogen Energy. 44(39), 21434-21450.

(11)

Wang, H., Li, J., Zhao, Y., Xu, C., Zhang, K., Li, J., Yan, L., Gu, J.-D., Wei, D., Wang, W., 2020. Establishing practical strategies to run high loading corn stover anaerobic digestion: Methane production performance and microbial responses. Bioresour. Technol. 310, 123364. https://doi.org/10.1016/j.biortech.2020.123364. Wei, Y., Yuan, H., Wachemo, A.C., Li, X., 2020. Anaerobic co-digestion of cattle manure

and liquid fraction of digestate (LFD) pretreated corn stover: Pretreatment process optimization and evolution of microbial community structure. Bioresour. Technol. 296, 122282. https://doi.org/10.1016/j.biortech.2019.122282.

Wenzel, M.M., Schonig, I., Berchtold, M., Kampfer, P., Konig, H., 2002. Aerobic and facultatively anaerobic cellulolytic bacteria from the gut of the termite Zootermopsis

angusticollis. J. Appl. Microbiol. 92 (1), 32–40.

Xu, Y., Ji, X., Chen, N., Li, P., Liu, W., 2012. Development of replicative oriC plasmids and their versatile use in genetic manipulation of Cytophaga hutchinsonii. Appl.

Microbiol. Biotechnol. 93 (2), 697–705.

Yi, J., Dong, B., Jin, J., Dai, X., Elias, D., 2014. Effect of Increasing total solids contents on anaerobic digestion of food waste under mesophilic conditions: Performance and

microbial characteristics analysis. PLoS ONE 9 (7), e102548.

Zhivin, O., Dassa, B., Moraïs, S., Utturkar, S.M., Brown, S.D., Henrissat, B., Lamed, R., Bayer, E.A., 2017. Unique organization and unprecedented diversity of the

Bacteroides (Pseudobacteroides) cellulosolvens cellulosome system. Biotechnol.

Biofuels 10 (1). https://doi.org/10.1186/s13068-017-0898-6.

Zhou, Z., Pan, J., Wang, F., Gu, J.-D., Li, M., 2018. Bathyarchaeota: globally distributed metabolic generalists in anoxic environments. FEMS Microbiol. Rev. 42 (5),