Obtainment of lignocellulose degradation microbial community: the effect of acid–base combination after restrictive enrichment

Microbial communities for bioconversion of lignocellulose have received widespread attention. Many cellulose-degrading microbial communities have been enriched from different sources. Combining two microbial communities with acidic and basic properties (acid–base combination) is a technique used alongside restricted enrichment culturing. Understanding how changes to microbial communities result in community’s structure and function is important for mechanistic reconstruction of microbiomes. In this study, we analyzed changes in microbial community structure to elucidate determination of the mechanisms of acid–base combination. We found that after restricted enrichment, the bacteria that were primarily retaining included not only those that decompose and utilize lignocellulose, such as Clostridium and Pseudomonas, but also synergistic microbiota such as Alkalobacillaceae. When the proportion of these two types of bacteria was unbalanced, the degradative ability of the microbial community was low, or pH changes of it did not compound regular changes, which may lead to the failure of restricted enrichment. Microbial communities were re-constituted by acid–base combination, whereby the degrading and synergistic strains were adjusted to a more appropriate proportion. The acid–base combination fixed the instability of microbial communities caused by the randomness of restrictive screening enrichment; it provided an effective method for obtaining high-quality lignocellulose-degrading microbial community.


Introduction
In nature, lignocellulose degradation often requires various microorganisms to cooperate as a microbial consortium (Tesfaw and Assefa 2014;Wang et al. 2020). Inspired by natural microbial consortia, artificial microbial consortia are created to address questions in scientific research and problems in industrial production (Xu and Yu 2021). Nevertheless, simple artificial consortia suffer from the shortage of characterized strains for synthetic consortia construction and the inability to co-culture microorganisms stably (Weiland 2010;Du et al. 2021). In contrast, undefined natural consortia originate from environmental microbial communities with an unknown number of constituents, exhibiting self-stability and low operational requirements (Jawed et al. 2019). Given the advantages of restrictive screen enrichment, such as the reduced metabolic burden and increased robustness to environment disturbances, many lignocellulose-degrading microbial communities are enriched by this process Cui et al. 2002;Hui et al. 2013;Haruta et al. 2002;Wang et al. 2011).
Despite these benefits, however, the uncontrollability of enrichment culturing requires many parallel experiments in order to obtain microbial communities with high lignocellulose degradability (Ferraro et al. 2020;Wang et al. 2021). Further, the degradative capability of initially suitable communities can often decrease in subsequent successive transferring. To overcome these challenges, an acid-base combination of microbial communities was used in the restrictive screen process (Haruta et al. 2002;Wang et al. Communicated by Erko Stackebrandt. 683 Page 2 of 12 2021). For acid-base variety, one microbial community with low pH was combined with another microbial community with high pH to obtain a novel community whose pH fluctuated within a specific range. The regularity of pH changes is very important for the microbial community. The products such as small molecular organic acids produced during the decomposition of lignocellulose quickly decreased the pH (< 6.5), and with the use of small molecular organic acids and other products, the pH quickly returned to neutral (7.0). If the pH does not decrease or the decrease is very slight (6.5-7.0), it indicates that the microbial community does not decompose the straw or decompose the straw very little. If the pH rises slowly and stays at 6.0-6.5 or even lower for a long time, it indicates that products such as small molecular organic acids have not been used in time, which is easy to inhibit the degrading bacteria and make the microbial community unstable. With the method of restricted culture and acid-base combination, many lignocellulose-degrading microbial communities, such as MC1 (Haruta et al. 2002), XDC-2 (Guo et al. 2010), and WCS-6 (Wang et al. 2011), were obtained. MC1, for example, was constructed from compost and is capable of effectively degrading various cellulosic materials, including rice straw Hua et al. 2014), filter paper (Kato et al. 2004), and cardboard (Yuan et al. 2012). In addition, it has been used for hydrolysis and acidification of agricultural waste (Yu et al. 2017), to enhance anaerobic digestion (Hua et al. 2016;Yuan et al. 2016), for concurrent saccharification and anaerobic digestion of Napier grass (Wen et al. 2015). After more than 40 successive generations of culture (Haruta et al. 2002) and approximately 20 years of research and application , the microbial community has remained stable with a high degradation efficiency. Therefore, in terms of practical application, restrictive screen enrichment and acid-base combination have achieved good results. At the same time, however, little research has been conducted into understanding the mechanisms of acid-base combinations. It is currently unknown how these communities form, particularly how acid-base combinations influence the microbial communities. With recent advances in sequencing, highthroughput approaches have made it possible to quantify the composition of microbial communities during this process (Estrela et al. 2021).
This study used compost as the inoculum source for the restrictive screen lignocellulosic degradation microbial community. The community structure changes were analyzed by high-throughput sequencing to understand the formation process and reveal the mechanisms of acid-base combinations after long-time enrichment cultures.

Inoculation and enrichment culturing
Our inoculum source was compost, sourced from the Zhuozhou Experimental Station, China Agricultural University. The compost pile was turned over once every 2 days during the thermophilic period (average 60 ℃) and once every 5 days after the temperature dropped to 35 ℃. Overall, the experiment lasted for 60 days.
Inoculum sources (5 g in fresh weight) were sampled during the ripening period (D1) and thermophilic period (D2). These were then inoculated into PCS medium with three replicates (A1, A2, and A3, and B1, B2, and B3, respectively). Cultures were maintained in 100 mL triangular bottles for 15 days in each generation.
The PCS medium was 0.5% peptone, 1% lignocellulose, 0.5% NaCl, 0.2% CaCO 3 , 0.1% yeast powder, pH natural (about 7.3) (Haruta et al. 2002). Rice straw was utilized as a carbon source, and small filter paper strips were added to the cultures. Those with decomposed filter paper strips were regarded as degradable and were chosen for the succession cultures. The inoculation was performed statically at 50 ℃ at a volume ratio of 5% (v/v).

Pretreatment of lignocellulosic materials and analysis of weight loss
The rice straw (harvested from Shangzhuang Experimental Station of China Agricultural University) was cut into approximately 10 cm length, soaked in 1% NaOH for 24 h, and washed with tap water until the pH value was between 7.0 and 8.0. Pieces were then dried at 50 ℃ and stored for future use. The weight loss of rice straw was tested as previously reported (Hua et al. 2014).

Acid-base combination of microbial community
Dynamic changes in pH were measured, communities with a slight pH decrease and a slow pH increase were selected. The pH of A1 and B2 decreased very slightly (6.5-7.0); it indicates that the microbial community decomposed the straw very little. The pH of A3 and B3 rose slowly and stayed at 6.0-6.5 for 9 days, it indicates that products such as small molecular organic acids have not been used in time, which was easy to inhibit the degrading bacteria and make the microbial community unstable. Based on those data, A1, B2, A3, and B3 were chosen for acid-base combination. 5 mL liquid of each community was extracted, taken, and combined to create a novel PCS medium. After merging, these new communities were named with the original Page 3 of 12 683 group numbers (A1A3, A1B3, B2A3, and B2B3), as shown in Fig. 1. Cultures of acid-base combination communities A1A3, A1B3, B2A3, and B2B3 were maintained in 100 mL triangular bottles for 15 days in each generation. After three generations, samples were taken for further analysis.
Sequences were filtered out with a quality value < 20. For each sample, the qualified reads were defined as a library and then down-sampled to the smallest library size, which was determined to compare the alpha diversity metrics, including Chao and Shannon (mothur v.1.30.1). After demultiplexing, the reads were assigned species-equivalent operational taxonomic units (OTUs) at 97% sequence similarity. The remaining OTUs were denominated at different classification levels according to the Silva database release 115 (http:// www. arb-silva. de). Raw sequences were archived in the GSA repository under accession number CRA007793 (for bacteria) and CRA007792 (for fungi).
Decreases in pH reflected rapid decomposition of cellulose, while increases in pH occurred with decreasing rates or cessation of cellulose degradation (Liu et al. 2006). pH of fermentation broth decreases were a prominent feature of the cellulosic materials degraded microbial community (Liu et al. 2006;. The pH of all six microbial communities decreased on the third day (Fig. 2C). The pH of A1 and B2 decreased to 6.6 and then began to rise rapidly to alkaline conditions, while A2 and B1 decreased to approximately 6.3. On the third day, A3 and B3 decreased to 6.1 and 6.3, respectively, and remained acidic until the 12th day. On the 15th day, the pH of A3 returned to neutral, while the pH of B3 remained acidic.
Microbial degradation of straw resulted in pH decreases in all treatments before rising to slightly alkaline conditions. Although A3 and B3 displayed faster degradation rates, the recovery time to slightly alkaline was longer than 12 days. Therefore, we selected B1 and A2 for further research and utilization. For A3 and B3, as the acidic products were not rapidly utilized, accumulation of these products may inhibit the degrading strains, thus inducing instability of the bacterial communities.
To study the acid-base combinations, A1 and B3 were pairwise combined with A3 and B3, respectively (Fig. 1). The new microbial communities A1A3, A1B3, A3B2, and B2B3 were created. The rice straw weight losses of A1A3, A1B3, A3B2, and B2B3 were 43.24 ± 6.73%, 32.18 ± 1.19%, 44.01 ± 2.93%, and 38.56 ± 2.83%, respectively. These were significantly greater than those of A1 and B2, but lower than those of A3 and B3, respectively (Fig. 2B). On the third day, the pH of A1A3, A1B3, A3B2, and B2B3 was 6, 6.5, 6.3, and 6.1, increasing to 7.1, 7.4, 8.2, and 7, on the 9th day, respectively (Fig. 2D). All returned to neutral and then became weakly alkaline. The combined microbial communities all followed the rule of pH fluctuating within 6-9 range, similar to other stable and efficient lignocellulose-decomposing microbial communities in previous reports (Liu et al. 2006). Acid-base combinations did not obtain a more efficient lignocellulosedecomposing microbial community, while the pH was more similar to a stable community.

Effect of inoculum source on the enrichment of microbial communities
Based on the T-test results, group A and group B's degradation ratio was not significantly different. However, the parallelity of group B was much higher than A. Therefore, both D1 and D2 had potential, while D2 was more suitable to screen highefficiency microbial communities (Fig. 3). Bacterial biodiversity was higher in the thermophilic period than in the ripening period. The bacteria composition of both compost samples was significantly different, which resulted in group A and group B also exhibiting significant differences. The inoculum source had a greater impact on restrictive screen results. In this study, composting during the thermophilic period was a better choice of inoculum source than during the ripening period.
Composting in the thermophilic period is often used as an inoculum source to screen cellulose-degrading bacteria (Tesfaw and Assefa 2014). For D2, the temperature of the compost body in the thermophilic period was higher, and organic matter decomposed rapidly. Cellulose materials decomposed more rapidly during this period, and bacteria with cellulose degradation abilities were also the most active. For D1, in the ripening period, the rapid decomposition of the organic matter ended.
The structure of the obtained microbial communities after restrictive screen culture Bacteria and Fungi community richness estimators including Chao and Shannon index are shown in Tables 1 and 2. A significant number of OTUs in D1 and D2 were eliminated (Fig. 4A, B). The number of retained OTUs in A1, A2, and A3 were 337, 310, and 418, respectively, with B1, B2, and B3 retaining 408, 392, and 353, respectively. In addition, some OTUs that were not detected in D1 and D2 were detected in our microbial communities. A1, A2, and B2 had 827, 807, and 713 OTUs that were not found in D1, while B1, B2, and B3 had 1128, 862, and 663 OTUs that were not identified in D2.
The retained bacteria (amount ratio more than 0.1%) are shown in Table 1. Bacteria such as Cellulosilyticum can degrade cellulose. EMSD5, Cellulosilyticum, alongside other bacteria and fungi, can synergistically degrade whole corn with cobs removed (Zhu et al. 2016). In the community SV79, which decomposes various lignocellulosic substrates to produce ethanol, Cellulosilyticum and Acetivibrio, Clostridium, Ruminococcus, and Sporomusa are the dominant bacteria (Zhao et al. 2014). Treponema primitia in the intestine of termites can degrade lignin (Lucey and Leadbetter 2014). Pseudomonas, Comamonas, and Lachnoclostridium can also decompose cellulose in a composite community (Liu et al. 2011;Xue et al. 2020;Zagrodnik et al. 2021).
Some retained bacteria could not use cellulose directly; however, they can exist in communities that decomposed lignocellulose, such as Alcaligenes. They can utilize organic acids and amino acids to produce ammonia and carbon dioxide. Alcaligenes faecalis have been reported in plants, soil, and other environments (Rosenberg 2014), and Alcaligenes sp TB is an aerobic denitrifying bacterium (Chen et al. 2016). Proteiniphilum is widely reported in anaerobic fermentation sludge, as it cannot decompose lignocellulose and cellobiose (Maspolim et al. 2015). It can, however, use peptone, arabinose, etc., and its metabolites include small molecular organic acids, CO 2, and H 2 (Langer et al. 2016). Treponema caldarium is a strictly anaerobic hyperthermia spirochete, which uses glucose, lactose, and other sugars as substrates, and has been reported in a complex community that degraded naphthalene (Koelschbach et al. 2017). These bacteria may assist in the elimination of acidic byproduct that inhibits cellulose-degrade strains.
Some bacteria, such as Sporomusa, Proteiniclasticum, Dethiosulfovibrionaceae, Desulfotomaculum, Desulfosporosinus, Pseudoxanthomonas, Nocardia, Alteromonadaceae, Erysipelotrichaceae, Novosphingobium, Propionigenium, Nonomuraea, Balneimonas, and Arcobacter which were found in all the six enriched communities, but not in D1 or D2. These newcomers may play specific roles in the microbial community and may have originated from the compost. Their abundance in compost may be extremely low and difficult to detect with high-throughput sequencing. After the growth environment changed, their abundance increased, allowing for detection with highthroughput sequencing. In addition, they may also have 683 Page 6 of 12 originated from the air as the PCS medium was not sterilized. Generally, the enriched microbial community needs to have a high anti-pollution capacity. It was not encouraged to be operated in a sterile environment such as an ultra-clean workbench, and thus other bacteria may have chance to join in Himanshu et al. (2017).
The abundance and composition of fungal OTUs from our communities were similar, although different from that of the compost (Fig. 4C). While many fungi were eliminated as a result of the restricted culture, some survived, though their OTUs were less abundant. Some may be involved in straw catabolism, such as Penicillium (Ogunyewo et al. 2020), while most fungi had little connection to straw decomposition, like Guehomyces pullulans (Nakagawa et al. 2006) and Hyphopichia burtonii (Groenewald and Smith 2010).

Effect of acid-base combination
The change of microbial community structure by an acidbase combination Among the obtained communities by acid-base combination, A1A3 had the lowest pH (6.0) on the 3rd day and had a greater degradation ratio (43.24 ± 6.73%). In contrast, A1B2 had the highest pH (6.5), and the degradation ratio was only 32.18 ± 1.19%. Therefore, analysis was focused on these two communities to elucidate the mechanism of acid-base combination.
A1A3 was combined within group A, and A1B3 was combined between group A and B. Before mixing, the dominant bacteria in A1 was Alcaligenes, with a ratio of 67% (Table 3; Fig. 5), which cannot decompose cellulose. Alcaligenes can use peptone and yeast extract powder in PCS medium and produce alkaline substances, inducing a pH increase. The relative abundance of Cellulosilyticum was less than 3.6%, other bacteria producing acid were much lower than Alcaligenes. The dominant bacteria of A3 were Cellulosilyticum, followed by Proteiniphilum, while the relative abundance of Alcaligenes was 11%. The dominant bacteria of B3 were Commonas, followed by Pseudomonas. The relative abundance of Lachnoclostridium, which may have the ability to decompose cellulose, was only 4%. The total proportion of Advenella, Alcaligenes, and Bordetella, which belong to Alkalobacillaceae, was 4.8%.
After the combination, the dominant bacterium of A1A3 was Comamonas, which was different from A1 and A3, and the proportion of Hydrogenispora increased to 12%. The database information may be hydrogenase poraethanolica, a strictly anaerobic bacteria producing hydrogen and ethanol . The relative abundance of Alcaligenes was 3%. After the combination, the dominant strain of A1B3 was Treponema, followed by Comamonas, Pseudomonas, Anaerophaga, and Alcaligenes.
The dominant bacteria varied greatly, whether it was the combination within one group or between two groups. After combination, the ratio of Cellulosilyticum decreased, the Pseudomonas ratio increased, and the Clostridium ratio was low. Those three genera all have the potential ability to decompose lignocellulose. Their total proportions in A1, A3, and B3 were 4.4%, 44.5%, and 14.5%, respectively, and their relative abundance in A1A3 and A1B3 was 10.9%, and 12.3%, respectively. The numerical relationship was A3 > A1A3 > A1, B3 > A1B3 > A1. A1A3 and A1B3 were significantly higher than A1 before the combination but lower than A3 and B3. The results were consistent with straw degradation. The pH of A1 was higher than 6.6 ( Fig. 2A), and the proportion of Alcaligenes in A1 was the highest corresponding. After the combination, the proportion of Alcaligenes in A1A3 and A1B3 is lower than that in A1, and their pH decreased below 6.5.

Change of fungi in the process of acid-base combination
Generally, all obtained fungal communities spatially overlapped, indicating a similar species diversity (Fig. 4C). Take A3B2, for example, the proportion of dominant fungi (1%) in the communities is shown in Fig. 6

Discussion
The screened bacteria were divided into three groups. The first group exhibited lignocellulose decomposition, including Clostridium, Cellulosilyticum, and Pseudomonas. The second group was likely to have auxiliary functions, such as Alcaligenes, and Pseudomonas (decomposing bacteria or auxiliary bacteria). The third group was likely not directly related to the decomposition of lignocellulose. Members of this group may have unknown functions or may be eliminated from the subculture in the future. While in the screening process, the randomness was apparent. The obtained microbial communities were significantly different with the same inoculum source and under the same culture conditions. For example, the community structure of A1 was significantly different from that of parallel A2 and A3, and the straw degradation rate was much lower than that of A2 and A3.
The acid-base combination aimed to obtain a microbial community with a stable and robust ability to decompose straw. The goal is to optimize the variety of different microbiota-producing acids and alkaline without damaging the synergistic relationship between microorganisms in nature. The pH of the obtained microbiota decreased rapidly with the decomposition of straw before returning to a slightly alkaline level.
The regular change of pH was closely related to the structure of the community. The composition of dominant bacteria varied greatly, whether within or between groups. The cellulose-degrading strains represented by Clostridium can metabolize straw to produce small molecular organic acids, reducing the pH of the solution. Helper strains represented by Alkalobacillus can metabolize Fig. 5 The dominant bacteria in the microbial communities A1, A3, B3, A1A3 and A1B3 (genus level). The quantitative proportion lower than 1% are combined as "low proportion" 683 Page 10 of 12 organic acids and amino acids to produce ammonia and carbon dioxide, increasing the solution's pH. When the auxiliary bacteria were dominant in the microbiota, their abundance was very high, such as A1, while the abundance of other strains was low. The decomposition ability of microbial bacteria was insufficient, and the pH was difficult to decline. When the number of auxiliary bacteria in the community was low, such as A3 and B3, pH only slowly rose back to slightly alkaline. After the acid-base combination, the microbial community was re-constituted. The degrading and auxiliary strains were adjusted to a more appropriate quantity and proportion, and unrelated strains were knocked out simultaneously, which changed the pH regularly. The acid-base combination was conducive to obtaining an efficient and stable microbial community and correcting microbial communities' reduced decomposition ability caused by a random process in the restricted culture process. Composting in the thermophilic period was a better choice of inoculum source than that of the ripening period. Many bacteria and most fungi were eliminated after the restrictive screening. Degrading strains, such as Clostridium and Pseudomonas, synergistic microbial such as Alkalobacillaceae were obtained. Microbial communities were re-constituted with an acid-base combination. The degrading and synergistic strains were adjusted to a more appropriate proportion, which changed the pH regularly. Acid-base combination fixed the instability of microbial communities caused by the randomness of restrictive screening enrichment. Combining these two techniques was highly conducive to obtaining efficient lignocellulose degradation microbial resources.