Bridging the soil carbon pool to microbial taxa and associated CAZyme profiles in grassland ecosystems

Bridging the soil carbon pool to microbial taxa and associated CAZyme profiles in grassland ecosystems | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 22 September 2025 V1 Latest version Share on Bridging the soil carbon pool to microbial taxa and associated CAZyme profiles in grassland ecosystems Authors : Xiaozhen Liu 0000-0003-4208-0961 [email protected] , Huiyun Liu , Yanling Yin , Xiaoxue Zhang , Yuanyuan Jing 0000-0002-4762-5433 , Ke Ma , Chongyin Wang , and Yafang Wang Authors Info & Affiliations https://doi.org/10.22541/au.175856337.71958541/v1 186 views 137 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract China’s grasslands are key to global carbon sequestration but face growing threats amid ecological fragility. Unravelling how grassland microbial communities and their CAZymes predict soil carbon pool stability is pivotal for clarifying carbon cycle mechanisms. We constructed comprehensive bacterial and archaeal MAGs from 570 Chinese grassland metagenomic samples. Actinobacteria had the highest CAZyme abundance, with f__70-9 members accounting for 16% of total CAZymes in o__Solirubrobacterales genomes. Approximately 90% of f__70-9 MAGs have both GH and GT genes，along with abundant Pfam domains and KOs implicated in carbohydrate metabolism, which may uniquely sustain enzyme activity in grassland ecosystems. Additionally, a study across nine Inner Mongolia sites showed TN and pH promoted MAOC accumulation, while biomass carbon inhibited it. Alkaline conditions were found to accelerate POC decomposition via actinomycete carboxylesterase genes (e.g., CE14). It was also found that microbial communities affect the stability of carbon pools by the group-specific regulation of carbon components. Microbial communities affect carbon pool stability via group-specific carbon component regulation; specifically, Actinomycetota promotes stable MAOC formation by secreting glycoside hydrolases (e.g., GH13). Overall, the study uncovers grassland bacteria/archaea diversity and their ecological functions, deepening understanding of microbial communities and their CAZymes in fragile ecosystems’ carbon cycles. 1.Introduction Grassland ecosystems play an important role in the global carbon cycle, as they store 34% of the global terrestrial carbon pool, with 90% of this carbon stored underground in the form of root biomass and soil organic carbon (SOC), thereby serving as a critical driver of soil carbon sequestration (Bai & Cotrufo 2022). China is home to the world’s second-largest grassland ecosystem, as its grassland area is 6–8% of the global total (equivalent to approximately 3.95 million square kilometres). Notably, these vast grasslands play a crucial role in global carbon sequestration, as they contribute 9% to 16% of the world’s total grassland carbon storage (Fan et al. 2008; Ni 2002). Thus, changes in SOC storage within China’s grassland could significantly impact global climate change and carbon cycles (Wang et al. 2011). In recent years, research focus has shifted from the chemical composition of SOC to its microbial regulatory mechanisms. Strong evidence indicates that the composition and functional traits of microbial communities directly influence SOC decomposition and accumulation. In addition, these communities play a crucial role in shaping the stability and dynamic fluctuations of carbon pools, underscoring their centrality in maintaining grassland carbon balance (He et al. 2023; Schmidt et al. 2011). SOC is mainly divided into particulate organic carbon (POC) and mineral-bound organic carbon (MAOC). POC is mainly derived from macromolecular polymers of partially decomposed plant debris (Bai & Cotrufo 2022; Li et al. 2023), whereas MAOC is mainly formed from simple molecular compounds of plants or microorganisms interacting with soil minerals (Cotrufo et al. 2019; Kleber et al. 2015). MAOC is more stable than POC and stays in soils for longer times, as its strong chemical bonds and physical protection limit microbial access and decomposition (Liu et al. 2025; Luo et al. 2022). However, the complexity of these processes limits our ability to draw definitive conclusions on the role of microorganisms in carbon loss and accumulation. The microbial regulation of soil carbon dynamics involves decomposing existing SOC and generating stable SOC via iterative biological processes, both of which induce carbon sequestration and shifts in the molecular composition of SOC (Liang et al. 2017; Qiu et al. 2023). Carbohydrate-active enzymes (CAZymes) play a key role in carbon cycles and have recently been used to explore the microbial functional response to C turnovers. Recent studies have shown that CAZyme-encoding gene abundance in microbial communities depends on vegetation type and soil properties. For instance, distinct CAZyme profiles have been observed across ecosystems, with grasslands differing from forests in the gene families linked to carbon degradation (Huang et al. 2024). Specific glycoside hydrolase (GH) and auxiliary activity (AA) enzymes have been identified as drivers of C kinetics, where GHs are primarily involved in polysaccharide decomposition and AAs are linked to lignin degradation (Lladó et al. 2019; López-Mondéjar et al. 2020). Notably, even within the same category of carbohydrate-active enzymes, their functional repertoire exhibits remarkable diversity. For instance, the cellulases and hemicellulases derived from glycoside hydrolase (GH) families serve as primary enzymatic drivers behind plant biomass degradation, whereas lysozymes and chitinases, also belonging to GH families, are implicated in the breakdown of bacterial and fungal necromass, respectively (Lombard et al. 2014; Žifčáková et al. 2017). This functional differentiation within GH families underscores the intricate specialisation of carbohydrate-active enzymes in mediating carbon flow across distinct trophic levels of soil organic matter. Grassland ecosystems, which are characterised with distinct vegetation compositions and management practices, exhibit unique patterns of organic matter input and microbial activity. These distinctive ecological features render the microbial processes mediated by CAZymes critically important. However, nowadays, the underlying mechanisms and regulatory dynamics of these processes remain poorly understood. Given the central role of grassland SOC stability in global carbon cycling and the mounting evidence regarding microbial regulation, decoding microbially mediated mechanisms of carbon pool stability is critical with regard to mitigating climate change. This study focuses on how microbial communities and their CAZymes predict and govern carbon pools, with three interlinked hypotheses: (i) environmental factors modulate carbon fractions through altering microbial structure and CAZyme abundance; (ii) specific microbial taxa drive MAOC formation and POC decomposition via signature CAZymes, thus directly shaping stability; (iii) the presence of uncharacterized lineages in grassland ecosystems that maintain activity to achieve resilience through unique CAZymes. These studies aim to enhance the understanding of carbon sequestration potential, improve the predictions of grassland responses to global change and guide sustainable management strategies to preserve carbon-rich ecosystems. 2.Materials and Methods 2.1 Field sites and soil sampling A total of nine sampling sites were selected for the study, and all nine sites are situated in the Inner Mongolia Autonomous Region. Table S1 provides specific information regarding these sampling locations. All sampling sites are characterised with the temperate continental monsoon climate. The annual average temperature varies between 2°C and 8°C across regions, with the growing season lasting 140–180 days and the frost-free period lasting 120–180 days annually. Precipitation is predominantly concentrated in summer (June to August), accounting for 60–75% of the annual total. In the past five years, meteorological data, such as temperature and precipitation, is from the National Meteorological Science Data Centre. The nine sampling points are SZ (Sunit Zuo Banner), DB (Darhan Muminggan United Banner), AT (Abag Town), EG (En Ge Bei Town), ET (En Ge Bei Town), KT (Kebul Town), ZB (Zhengxiangbai Banner), HN (Huretu Nauer Sumu) and ST (Shaogen Town) (Figure 1A) . Nine 80 cm × 80 cm sampling plots with consistent slope directions and relatively uniform vegetation characteristics were chosen at each sampling point, and they were separated by a distance of collected from the topsoil (0–10 cm) of each plot, and they were then mixed to form a composite sample. A portion of each soil sample was transferred to a 2-mL centrifuge tube, snap-frozen with liquid nitrogen and then sent for metagenomic sequencing. The remaining fresh soil was transported to the laboratory in an icebox. A part of the soil was stored at −80°C, and the other part was air-dried and sieved using a 0.2-mm sieve to remove roots and other debris. Then, it was subjected to analyses on physicochemical properties and enzyme activity. 3.1 Characteristics of SOC and C fractions SOC varied significantly in different sites, and the KT and HN values were significantly higher, reaching approximately 25.69 g/kg and 20.84 g/kg, respectively. POC was highest at 23.90 g/kg in KT, and MAOC was highest at 14.41 g/kg in HN. In other locations (ZB, ST, AT, SZ, etc.), the SOC, POC and MAOC concentrations were generally low. The different performances of POC and MAOC in different locations highlight the different factors affecting their distribution. For example, in HN, MAOC was higher than POC, indicating that minerals have a strong protective effect on organic carbon. Moreover, in KT, POC was dominant, indicating that there are different stabilisation mechanisms (Figure 1B and Table S2) . 3.2 Predictors for organic carbon and C fractions All SOC, POC and MAOC exhibited significant positive correlations with TN while exhibiting negative correlations with soil biomass carbon. SOC and POC were positively correlated with NH₄⁺-N and biomass nitrogen. SOC and MAOC displayed positive associations with pH but negative associations with S-β-GC (Figure 1C) . The results of the random forest model showed that SOC was closely related to TN, pH and NH 4 + -N and that biomass nitrogen, NO 3 - -N and biomass carbon were also the factors affecting SOC. Moreover, POC was found to be closely related to TN, pH, NH 4 + -N and biomass nitrogen. In addition, S-β-GC was found to be a factor affecting SOC. Moreover, MAOC was found to be closely related to TN, pH, S-β-GC and biomass carbon (Figure S1 and 1D) . These findings indicate that the availability of nitrogen may be accessed using microorganisms, thus promoting carbon accumulation. 3.3 Relationship between the microbiome and organic carbon fractions The α-diversity indices of microbial communities across nine grassland sites (i.e. EG, ET, KT, DB, SZ, ZB, AT, HN, ST) were evaluated using the Shannon index, Richness, Pielou’s evenness index and Chao1 index. ZB and AT had higher Shannon indices than KT, SZ, EG, DB and ST, showing better species richness and evenness. The Pielou index analysis showed that there was no significant difference in microbial community evenness between the HN and ZB groups, while the difference between the HN group and other groups (i.e. EG, ET, KT, DB, SZ, AT, ST) was significant. The analysis results of the Richness and Chao1 indices revealed significant disparities in species richness between the ET group and all other comparison groups, including EG, KT, DB, SZ, ZB, AT, HN and ST (Figure 2A and Table S3) . The differences in abundances across localities were evident in all surveyed species (Figure 2B). The canonical correspondence analysis (CCA) revealed that the grassland microbial community structure was significantly associated with environmental factors (R 2 = 0.8571, P = 0.0010). CCA1 and CCA2 biaxially explained 27.03% of the community environment association information. NH 4 + -N, NO 3 - -N and pH had a strong driving effect on community variation and could dominate community composition and distribution patterns by shaping niche differentiation (Figure 4C) . Moreover, the richness, evenness and abundance of the species were correlated with SOC. Some microbial groups, especially class and order microbial groups, were significantly positively correlated with SOC and MAOC, indicating that these groups may play an important role in SOC sequestration. On the contrary, species at each level were negatively correlated with pH, meaning that these groups may prefer acidic environments or be negatively affected by alkaline environments. The Mantel’s r value close to 0.6 indicated that there was a strong correlation between specific microbial groups and soil properties such as pH and MAOC, thus highlighting the potential impact of these taxa on soil fertility (Figure 4D) . The Pearson correlation quantifies the strength and direction of linear relationships, and the Mantel test evaluates the statistical significance of these correlations. It can be noted that the species at the class level have a great impact on the physical and chemical properties of soil. Thus, we further analysed the impact of the species at the phylum and class levels on soil carbon. SOC and MAOC exhibited positive correlations with the bacterial phyla Actinobacteria, Proteobacteria, Acidobacteria and Verrucomicrobia while showing negative correlations with the phylum Thaumarchaeota (Figure 4E) . SOC was positively correlated with CFGB10042, Nitrososphaeria and CFGB33437, indicating that these microbial groups may have significantly contributed to the dynamic changes of SOC through organic matter decomposition and carbon fixation. POC was positively correlated with Nitrososphaeria, CFGB10042 and CFGB80811, indicating that they may be involved in the POC cycle in soil ecosystems. MAOC was positively correlated with Thermoleophilia, CFGB77407 and Alphaproteobacteria, showing that Thermoleophilia and Alphaproteobacteria may play a potential role in promoting carbon mineralisation and increasing the MAOC pool. Moreover, biomass C was positively correlated with several microbial groups, including Alphaproteobacteria, Thermoleophilia and CFGB77407, indicating that these groups may have significantly influenced microbial carbon cycling processes and soil fertility status (Figure 4F) . 3.4 Microbial community structure and ecological function of key groups in different soil samples The Spearman correlation between microbial taxa at the species level was utilised to construct a symbiotic network from an ecological perspective, and the network topology characteristics were also evaluated. The co-occurrence network consisted of 270 nodes and 2498 edges with an average degree of 18.504. Actinobacteria, Proteobacteria, Thaumarchaeota and Firmicutes were found to be the dominant phyla, accounting for 41.85%, 35.19%, 6.3% and 5.19%, respectively. The detection of Thaumarchaeota revealed the potential role of archaea in the community, and the presence of unclassified bacteria suggested that there were still microbial groups that were not fully characterised in the research system (Figure 3A) . Compared with other sites, both ET and AT plots exhibited a higher number of nodes and edges, as well as higher average degrees. These network metrics showed that the microbial communities in ET and AT plots possess more potential participating units and a more complex system composition (Figure 3B, C, and D) . Moreover, the modularity showed that there were closely connected microbial subgroups in each group (Figure 3E and Table S4) . 3.5 S ummary of putative CAZyme genes in grassland GT, GH, CE and CBM together constitute 95% of CAZyme gene relative abundance. The abundance of AA and PL accounted for 5% of the total abundance of the CAZyme gene. Different carbohydrate-active enzyme categories showed different abundance distributions in each group. Both GH and GT were the core functional groups, and the high abundance of key enzymes indicates that they may have more advantages in carbohydrate conversion in relation to carbon cycles. ( Figure S2) . At the finer enzyme family level, GH13 was highest in the GH group, GT2 and GT4 were highest in the GT group, CBM48 was highest in the CBM group, CE14 and CE11 were highest in the CE group, PL1 was highest in the PL group and AA3 was highest in the AA group (Figure 4A) . The members of Actinomycetota, Acidobacteriota and Pseudomonadota were the highest contributors to the CAZyme gene’s enrichment. GH13, GT2, GT4, GT51 and CBM48 were distributed across above 50% of genera (Figure 4B) . We further analysed the correlation between the SOC fractions and carbohydrate enzymes using the Pearsonr test. SOC and POC showed positive correlations with GT2, GT55 and GT66, whereas they displayed negative correlations with GH2 and GH13. SOC and MAOC were negatively correlated with GH31, GT39, GH63, GH172, GH15 and GH2 (Figure 4C) . 3.6 Genomic properties and functional potential of grassland MAGs A total of 570 metagenomic samples were collected for analysis, among which 81 samples were obtained from this study, to offer a thorough overview of grassland microbial communities. A total of 1856 bacterial and archaeal MAGs were obtained based on the criteria of them, 509 high-quality MAGs with a completeness of >80% and a contamination of <10% were classified into 22 bacterial phyla and one archaeal phylum (Figure S3) . At the phylum level, 38.6% of the 474 bacterial MAGs were assigned to Actinomycetota, 21.9% were assigned to Pseudomonadota, 13.3% were assigned to Acidobacteriota, 5.9% were assigned to Bacteroidota, 4.6% were assigned to Gemmatimonadota, 3.6% were assigned to Chloroflexota and all the archaeal MAGs were assigned to Phylum Thermoproteota (Figure 5A and B) . The data included 474 bacterial and 35 archaeal species not previously reported in the GTDB database. Of these species, the following ranks could not be assigned compared with the GTDB database: family (5), genus (110) and species (466) (Figure 5C and D) , indicating that the predicted grassland MAGs were novel microbial taxa. Furthermore, a total of 1,828,761 protein-coding sequences were predicted using Prokka. An additional analysis of the carbohydrate active enzyme (CAZyme) spectrum was performed to further look into the details of the genomic potential for carbohydrate metabolism by searching against the CAZy database from individual MAGs. The 9,722 CAZy proteins included 5,073 glycosyl hydrolases (GH), 3,881 glycosyl transferases (GT), 904 carbohydrate-binding modules (CBM), 374 carboxyesterases (CE), 149 proteins with auxiliary activity (AA) and 95 polysaccharide lyases (PL) (Table S5) . These proteins were extensively present in the grassland microbial taxa. It was also found that Actinobacteria had the highest abundance of AA, CBM, GH, GT, CE and PL, accounting for approximately 48.5%, followed by Acidobacteria (18.2%) and Pseudomonas (17.4%). Nevertheless, the members of Acidobacteria only contained more GH and GT, accounting for 19.5% and 20.5%, while Pseudomonas contained more AA, CBM, CE and PL, accounting for 28.9%, 7.8%, 24.6% and 26.3% (Figure 5A and Table S6) . 3.7 The biological Functional Potential of f__70-9 The contributions of the members of Actinobacteria to CAZymes were further analysed by selecting the families with the highest enzyme content of 50%. Unexpectedly, it was found that the f__70-9 of the order Solirubrobacterales accounted for up to 16% (Figure S4) and that the f__70-9 families had not been excavated in the research of carbohydrate enzymes before. An in-depth analysis of 70-9 families was conducted. The results showed that GH and GT functional genes not only showed co-occurrence characteristics in 89% of the MAGs of this group, but they also accounted for the highest proportion and the most abundant number in all CAZymes (Figure S5) . In total, 32,868 (74.5%) of the 44,094 protein sequences in the 18 MAGs were annotated to the eggNOG database, revealing 2,375 unique PFAM domains and 1,843 unique Kos ( Table S7) . Among the PFAM domains, ABC_tran (601 occurrences), TetR_N (366) and Glycos_transf_2 (325) were most abundant. Notably, the five top 1% domains linked to carbohydrate metabolism were Glycos_transf_2 (325), Glyoxalase (210), Epimerase (197), Glycos_transf_1 (194) and Glyco_transf_4 (184), collectively totalling 1,110 occurrences. Similarly, the seven top 1% KOs that participated in the carbohydrate metabolism were K00059 (68), K12132 (67), K01784 (51), K01733 (46), K01895 (44), K01768 (41) and K00721 (40), accounting for 357 occurrences ( Figure 6A and B ). These findings demonstrate a strong correlation between this microbial phylum and carbohydrate metabolic processes. 4. Discussion Despite extensive research on grassland microorganisms, most studies have primarily focused on the community composition and function of microbial communities, such as those involved in nitrogen cycling and organic matter decomposition. With regard to grassland ecosystems, understanding the genetic basis of microorganisms and their potential in degrading complex carbohydrates is very important for elucidating carbon cycling processes. Thus, there is an urgent need to conduct in-depth research on the genomic assembly of grassland microorganisms in China and the exploration of the CAZymes they encode. 4.1 CAZyme-mediated carbon cycling processes by grassland microorganisms In this study, extensive data collection and analyses were performed on grassland metagenomes from 570 samples taken from nine provinces in China. While prior studies leveraging metagenome offered preliminary perspectives on the role of grassland systems in sustaining biological diversity, our research could advance these foundational findings by deepening the mechanistic understanding and delineating actionable pathways for sustainable exploration and utilisation. CAZymes are capable of degrading complex carbohydrates, such as plant residues and soil organic matter, therefore playing a crucial role in the carbon turnover process within grassland ecosystems (Lynd et al. 2002). We could identify six major functional families of GH, GT, CBM, AA, CE and PL in the assembled grassland MAGs (Garron & Henrissat 2019; Huang et al. 2018). Notably, GH13 and GT2 accounted for 32.4% of the total CAZyme gene abundance, constituting the core pathways of starch and cellulose degradation. This observation is aligned with recent findings, which demonstrated that carbohydrate metabolism genes dominate carbon cycling processes in global grassland ecosystems (Huang et al. 2024). Actinomycetes represent a major component of soil microbial communities and drive critical processes in organic matter cycling by the production of hydrolytic compounds (Remya & Vijayakumar n.d.). Several members contribute to degradation processes by harbouring cellulose utilisation-related GH genes, such as β-glucosidase, cellulase genes and Streptomyces, as main contributors. These organisms collectively show diverse cellulose utilisation functions and the coexistence of conservation and adaptive variation in soil carbon cycles as a whole (Berlemont & Martiny 2013). Regarding this study, it is worth noting that the number of CAZymes genes carried by the MAGs of Actinobacteria was significantly higher than that of other phyla. It was surprising to find that the six enzymes in Actinobacteria were concentrated in the f__70-9 families of Solirubrobacterales, while the 70-9 families had not yet been explored and studied. Thus, an in-depth analysis was performed in this study on the 70-9 family, and it was found that GH and GT functional genes not only co-occur in 89% of the genomes within this group but also constitute the highest proportion and exhibit the greatest abundance among all CAZymes. In addition, it was found that GH and GT have a synergistic effect, which is consistent with the previous results. (Fu et al. 2025). Fu et al. emphasised that the bacterial community exhibits genes related to the metabolite synthesis process (e.g. glycosyltransferases, GTs) while consuming genes related to extracellular enzyme production (e.g. glycoside hydrolases, GHs), thus resulting in a feebleness priming effect. This work for the first time demonstrates that the members of f__70-9 encoding a variety of CAZymes participate in carbohydrate metabolic processes in grassland ecosystems. There is reason to boldly speculate that 70-9 families have a significant impact on the functional balance of CAZymes and the carbon cycle function of microbial communities in grassland environments, providing a new key clue for understanding the microbial driving mechanism of grassland carbon pool stability. 4.2 Microbial communities as key regulators of carbon components This study could identify specific microbial groups closely related to SOC, POC and MAOC, and it could link the taxonomic composition to carbon pool dynamics. SOC was highly correlated with CFGB10042 and Nitrososphaeria, indicating that they play a dual role in organic matter decomposition and carbon fixation. Nitrosococcaceae is classified as a group of Thermoproteota, which is widely distributed in soil environments (Ren & Wang 2022). From the perspective of functional mechanism, Nitrosococcus may enhance carbon sequestration by coupling nitrification with organic mineralisation (Schmidt et al. 2011). In the first step of nitrification, it oxidises ammonia (NH₃) to hydroxylamine (NH₂OH) using ammonia monooxygenase (AMO), where electrons from this oxidation reaction enter the respiratory chain to fuel microbial metabolism (Wright & Lehtovirta-Morley 2023). POC comes from partially decomposed plant residues (Witzgall et al. 2021). After entering the soil, plant residues undergo a complex decomposition process under the action of microorganisms (Wickings et al. 2012). The part that has not been completely mineralised forms POC, which is related to Nitrososphaeria and CFGB10042, indicating that these groups may secrete enzymes targeting plant polysaccharides. In contrast, mineral-bound organic carbon stabilised by mineral interactions (Lavallee et al. 2020) is associated with Thermoleophilia and Alphaproteobacteria. Moreover, MAOC is not completely in the untouchable state of microorganisms, and there is a highly dynamic subset inside it. This subset may become an important reservoir for microorganisms and plants to obtain carbon sources and nutrients through a continuous turnover process (Jilling et al. 2018; Keiluweit et al. 2015). Thermoleophilia are known to have metabolic diversity in carbon utilisation (Peng et al. 2024) and may promote the formation of mineral-bound organic carbon by producing low molecular weight compounds bound to minerals, while Alphaproteobacteria may enhance mineral protection by secreting extracellular polymers. It is also worth noting that at the phylum level, Actinomycetota was significantly positively correlated with SOC and particulate organic carbon, further confirming their role as key decomposers in soil carbon cycles (Mitra et al. 2022). This result is in line with our findings that Actinobacteria have the highest abundance of carbohydrate-active enzymes, including glycoside hydrolases, which are involved in plant biomass degradation, thus supporting their key roles in particulate organic carbon turnover and SOC accumulation. 4.3 Spatial heterogeneity of SOC components and environmental drivers The inherent differences in stability and turnover rate between POC and MAOC profoundly affect the long-term carbon sequestration potential of soils and the dynamic balance of carbon pools (King et al. 2019). Moreover, the three pathways of plant carbon input may affect the formation and decomposition of POC and MAOC through different mechanisms, resulting in different degrees of SOC pools (Chen et al. 2021; Villarino et al. 2021). The obtained results showed that there were significant spatial variations in SOC, POC and MAOC in the nine grassland sites. Among them, the KT and HN sites had the highest SOC concentrations, reaching 25.69 g/kg and 20.84 g/kg, respectively. Specifically, the accumulation of POC in the KT plot mainly depended on the recent continuous input of plant litter (Cotrufo et al. 2015). In this region, dense vegetation cover may result in a large number of fresh plant residues entering the soil. These residues are not completely converted into small molecules under the initial decomposition of microorganisms; however, they remain in a relatively complete particle form, forming a carbon pool with POC as the main body (Even & Francesca Cotrufo 2024; Yin et al. 2025). When the input rate of plant litter exceeds the degradation rate of microorganisms, POC can be accumulated, its carbon pool activity is higher, the turnover cycle is relatively short and it is more susceptible to vegetation productivity and short-term environmental fluctuations (Chen et al. 2018b). In contrast, the persistence of MAOC in the HN plot was mainly realised through mineral–organic interactions. The clay minerals and iron and aluminium oxides in soil have a large number of active sites on the surface, which can be closely combined with the small molecular organic compounds produced by the microbial decomposition of plant residues through electrostatic adsorption and coordination exchange to form a stable mineral–organic complex (Kleber et al. 2021; Tong et al. 2024). This combination not only limits the contact and degradation of organic carbon by microorganisms through physical isolation but also enhances the anti-decomposition ability of organic carbon through chemical inertia. Thus, MAOC can be retained in soil for a long time and become a more stable carbon pool (Chen et al. 2020; Kleber et al. 2021). Therefore, the higher MAOC content in the HN plot suggests that soil mineral components may have a stronger carbon sequestration potential, and this stabilisation mechanism is more significantly regulated by intrinsic properties, such as soil texture and mineral composition. The random forest model and correlation analysis revealed that TN and pH were the core environmental factors regulating SOC, POC and MAOC in the SOC fractions. In terrestrial ecosystems, nitrogen-induced soil acidification is generally considered the main mechanism resulting in changes in soil microbial function, including inhibition of soil microbial biomass and respiration (Li et al. 2018; Zhang et al. 2020). In turn, soil acidification affects plant growth and biodiversity and is therefore considered an important factor in biogeochemical models (Li et al. 2018; Pan et al. 2020). The increase of nitrogen input not only affects microbial metabolism by alleviating nutrient limitation, but it is also more likely to change the bioavailability of soil carbon. However, many studies have shown that soil carbon utilisation is low under nitrogen addition, resulting in decreased microbial respiration and biomass (Chen et al. 2018a; Eberwein et al. 2015; Kamble et al. 2013). Specifically, in this study, the higher TN content in KT and HN plots may indirectly drive the metabolic activity of functional groups, such as Actinobacteria, by alleviating the nitrogen limitation of microorganisms, thus influencing the transformation and accumulation processes of carbon components. In addition, the negative correlation between pH and SOC components was further strengthened by the niche preference of Actinobacteria. It is well known that the decrease in soil pH has different effects on the competitive ability of bacteria and fungi. For example, in comparison with fungi, a low pH environment is more likely to weaken the competitive advantage of bacteria (Fernández-Calviño et al. 2011; Rousk et al. 2009). The pH conditions of the KT and HN plots in this study are just suitable for the proliferation of Actinobacteria, and their rich CAZymes libraries (such as GHs, GTs) can play a more efficient role in this environment, which not only promotes the degradation of plant residues into POC but also participates in the formation of MAOC. 5. Conclusion In this study, based on an analysis of 570 grassland metagenomic samples from nine Chinese provinces, the microbial-mediated carbon cycle mechanism, community regulation and spatial heterogeneity in grassland ecosystems were systematically revealed. In the CAZyme-mediated carbon cycle, six functional families, such as GH and GT, were identified. Among them, GH13 and GT2 constitute the core pathway of starch and cellulose degradation. The number of CAZyme genes carried by Actinobacteria was significantly higher than that of other groups. In particular, the GH and GT genes in the f__70-9 family of Solirubrobacterales were highly co-occurring and synergistic. For the first time, it was confirmed that this family was involved in grassland carbon and water metabolism, and it was speculated that it had an important impact on CAZyme function balance and community carbon cycle function. In terms of the microbial community regulation of carbon components, it was found that CFGB10042 and Nitrososphaeria are related to SOC and POC and that they may also participate in carbon cycles through decomposition and fixation. Moreover, it was found that Thermoleophilia and Alphaproteobacteria are associated with MAOC, promoting its stability by small molecule–mineral binding or secreting polymers, while Actinobacteria are a key decomposer of SOC and POC turnover due to its high abundance of CAZyme. In summary, this study provides new clues for understanding the microbial driving mechanism of grassland carbon pool stability and deepens the understanding of grassland carbon cycle processes and regulatory factors. Acknowledgments We thank Northwest A & F University and Grassland Research Institute of Chinese Academy of Agricultural Sciences for their technical support and guidance. Author contributions Huiyun Liu: conceptualization, data curation, formal analysis, investigation, methodology, software, visualization, writing – original draft, writing – review and editing. Yanling Yin: formal analysis, investigation, methodology, software, visualization, writing – review and editing. Xiaoxue Zhang: formal analysis, investigation, methodology, writing – review and editing. Ke Ma: methodology, software. Chongyin Wang: investigation. Yafang Wang: resources, writing – review and editing. Xiaozhen Liu: conceptualization, data curation, funding acquisition, investigation, project administration, resources, supervision, writing – review and editing. Conflicts of 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. Funding This study was supported by the Natural Science Foundation of Inner Mongolia (2024QN03028 to XL), the Agricultural Science and Technology Innovation Program of CAAS (CAAS-ASTIP-2024-IGR to XL), and the 2023 High Level Talents Project of Inner Mongolia (2024NMRC005 to XL). Data availability The raw sequencing reads of metagenomic sequencing were deposited into CNCB with BioProject accession number PRJCA039301. Supplementary data References： Bai, Y. & Cotrufo, M.F. (2022). Grassland soil carbon sequestration: current understanding, challenges, and solutions. Science , 377, 603–608.Berlemont, R. & Martiny, A.C. (2013). Phylogenetic distribution of potential cellulases in bacteria. Appl. Environ. Microbiol. , 79, 1545–1554.Chen, C., Hall, S.J., Coward, E. & Thompson, A. (2020). Iron-mediated organic matter decomposition in humid soils can counteract protection. Nat. Commun. , 11.Chen, H., Li, D., Zhao, J., Zhang, W., Xiao, K. & Wang, K. (2018a). Nitrogen addition aggravates microbial carbon limitation: evidence from ecoenzymatic stoichiometry. Geoderma , 329, 61–64.Chen, L., Fang, K., Wei, B., Qin, S., Feng, X., Hu, T., et al. (2021). Soil carbon persistence governed by plant input and mineral protection at regional and global scales. Ecol. Lett. , 24, 1018–1028.Chen, X., Xu, Y., Gao, H., Mao, J., Chu, W. & Thompson, M.L. (2018b). Biochemical stabilization of soil organic matter in straw-amended, anaerobic and aerobic soils. Sci. Total Environ. , 625, 1065–1073.Cotrufo, M.F., Ranalli, M.G., Haddix, M.L., Six, J. & Lugato, E. (2019). Soil carbon storage informed by particulate and mineral-associated organic matter. Nat. Geosci. , 12, 989–994.Cotrufo, M.F., Soong, J.L., Horton, A.J., Campbell, E.E., Haddix, M.L., Wall, D.H., et al. (2015). Formation of soil organic matter via biochemical and physical pathways of litter mass loss. Nat. Geosci. , 8, 776–779.Eberwein, J.R., Oikawa, P.Y., Allsman, L.A. & Jenerette, G.D. (2015). Carbon availability regulates soil respiration response to nitrogen and temperature. Soil Biol. Biochem. , 88, 158–164.Even, R.J. & Francesca Cotrufo, M. (2024). The ability of soils to aggregate, more than the state of aggregation, promotes protected soil organic matter formation. Geoderma , 442, 116760.Fan, J., Zhong, H., Harris, W., Yu, G., Wang, S., Hu, Z., et al. (2008). Carbon storage in the grasslands of China based on field measurements of above- and below-ground biomass. Clim. Change , 86, 375–396.Fernández-Calviño, D., Arias-Estévez, M., Díaz-Raviña, M. & Bååth, E. (2011). Bacterial pollution induced community tolerance (PICT) to Cu and interactions with pH in long-term polluted vineyard soils. Soil Biol. Biochem. , 43, 2324–2331.Fu, Y., Liu, W., Chen, Z., Redmile-Gordon, M., Liang, C., Tang, C., et al. (2025). Microbial metabolisms determine soil priming effect induced by organic inputs. Soil Biol. Biochem. , 209, 109885.Garron, M.-L. & Henrissat, B. (2019). The continuing expansion of CAZymes and their families. Curr. Opin. Chem. Biol. , 53, 82–87.He, P., Zhang, Y., Shen, Q., Ling, N. & Nan, Z. (2023). Microbial carbon use efficiency in different ecosystems: a meta‐analysis based on a biogeochemical equilibrium model. Global Change Biol. , 29, 4758–4774.Huang, L., Zhang, H., Wu, P., Entwistle, S., Li, X., Yohe, T., et al. (2018). dbCAN-seq: a database of carbohydrate-active enzyme (CAZyme) sequence and annotation. Nucleic Acids Research , 46, D516–D521.Huang, Q., Wang, B., Shen, J., Xu, F., Li, N., Jia, P., et al. (2024). Shifts in C-degradation genes and microbial metabolic activity with vegetation types affected the surface soil organic carbon pool. Soil Biol. Biochem. , 192, 109371.Jilling, A., Keiluweit, M., Contosta, A.R., Frey, S., Schimel, J., Schnecker, J., et al. (2018). Minerals in the rhizosphere: overlooked mediators of soil nitrogen availability to plants and microbes. Biogeochemistry , 139, 103–122.Kamble, P.N., Rousk, J., Frey, S.D. & Bååth, E. (2013). Bacterial growth and growth-limiting nutrients following chronic nitrogen additions to a hardwood forest soil. Soil Biol. Biochem. , 59, 32–37.Keiluweit, M., Bougoure, J.J., Nico, P.S., Pett-Ridge, J., Weber, P.K. & Kleber, M. (2015). Mineral protection of soil carbon counteracted by root exudates. Nat. Clim. Change , 5, 588–595.King, A.E., Congreves, K.A., Deen, B., Dunfield, K.E., Voroney, R.P. & Wagner-Riddle, C. (2019). Quantifying the relationships between soil fraction mass, fraction carbon, and total soil carbon to assess mechanisms of physical protection. Soil Biol. Biochem. , 135, 95–107.Kleber, M., Bourg, I.C., Coward, E.K., Hansel, C.M., Myneni, S.C.B. & Nunan, N. (2021). Dynamic interactions at the mineral–organic matter interface. Nat. Rev. Earth Environ. , 2, 402–421.Kleber, M., Eusterhues, K., Keiluweit, M., Mikutta, C., Mikutta, R. & Nico, P.S. (2015). Mineral–organic associations: formation, properties, and relevance in soil environments. In: Advances in Agronomy . Elsevier, pp. 1–140.Lavallee, J.M., Soong, J.L. & Cotrufo, M.F. (2020). Conceptualizing soil organic matter into particulate and mineral‐associated forms to address global change in the 21st century. Global Change Biol. , 26, 261–273.Li, Y., Sun, J., Tian, D., Wang, J., Ha, D., Qu, Y., et al. (2018). Soil acid cations induced reduction in soil respiration under nitrogen enrichment and soil acidification. Sci. Total Environ. , 615, 1535–1546.Li, Y., Zhang, X., Wang, B., Wu, X., Wang, Z., Liu, L., et al. (2023). Revegetation promotes soil mineral-associated organic carbon sequestration and soil carbon stability in the tengger desert, northern China. Soil Biol. Biochem. , 185, 109155.Liang, C., Schimel, J.P. & Jastrow, J.D. (2017). The importance of anabolism in microbial control over soil carbon storage. Nat. Microbiol. , 2.Liu, M., Zheng, S., Pendall, E., Smith, P., Liu, J., Li, J., et al. (2025). Unprotected carbon dominates decadal soil carbon increase. Nat. Commun. , 16, 2008.Lladó, S., Větrovský, T. & Baldrian, P. (2019). Tracking of the activity of individual bacteria in temperate forest soils shows guild-specific responses to seasonality. Soil Biol. Biochem. , 135, 275–282.Lombard, V., Golaconda Ramulu, H., Drula, E., Coutinho, P.M. & Henrissat, B. (2014). The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res. , 42, D490–D495.López-Mondéjar, R., Tláskal, V., Větrovský, T., Štursová, M., Toscan, R., Nunes Da Rocha, U., et al. (2020). Metagenomics and stable isotope probing reveal the complementary contribution of fungal and bacterial communities in the recycling of dead biomass in forest soil. Soil Biol. Biochem. , 148, 107875.Luo, R., Kuzyakov, Y., Zhu, B., Qiang, W., Zhang, Y. & Pang, X. (2022). Phosphorus addition decreases plant lignin but increases microbial necromass contribution to soil organic carbon in a subalpine forest. Global Change Biol. , 28, 4194–4210.Lynd, L.R., Weimer, P.J., Van Zyl, W.H. & Pretorius, I.S. (2002). Microbial cellulose utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. , 66, 506–577.Mitra, D., Mondal, R., Khoshru, B., Senapati, A., Radha, T.K., Mahakur, B., et al. (2022). Actinobacteria-enhanced plant growth, nutrient acquisition, and crop protection: advances in soil, plant, and microbial multifactorial interactions. Pedosphere , 32, 149–170.Ni, J. (2002). Carbon storage in grasslands of China. J. Arid. Environ. , 50, 205–218.Pan, S., Wang, Y., Qiu, Y., Chen, D., Zhang, L., Ye, C., et al. (2020). Nitrogen‐induced acidification, not N‐nutrient, dominates suppressive N effects on arbuscular mycorrhizal fungi. Global Change Biol. , 26, 6568–6580.Peng, X., Wang, S., Wang, M., Feng, K., He, Q., Yang, X., et al. (2024). Metabolic interdependencies in thermophilic communities are revealed using co-occurrence and complementarity networks. Nat. Commun. , 15.Qiu, J., Song, M., Wang, C., Dou, X., Liu, F., Wang, J., et al. (2023). Five-year warming does not change soil organic carbon stock but alters its chemical composition in an alpine peatland. Pedosphere , 33, 776–787.Remya, M. & Vijayakumar, R. (n.d.). Isolation and characterization of marine antagonistic actinomycetes from west coast of India.Ren, M. & Wang, J. (2022). Phylogenetic divergence and adaptation of nitrososphaeria across lake depths and freshwater ecosystems. ISME J. , 16, 1491–1501.Rousk, J., Brookes, P.C. & Bååth, E. (2009). Contrasting soil pH effects on fungal and bacterial growth suggest functional redundancy in carbon mineralization. Appl. Environ. Microbiol. , 75, 1589–1596.Schmidt, M.W.I., Torn, M.S., Abiven, S., Dittmar, T., Guggenberger, G., Janssens, I.A., et al. (2011). Persistence of soil organic matter as an ecosystem property. Nature , 478, 49–56.Tong, Y., Xiang, H., Jiang, J. & Chen, W. (2024). Interfacial interactions between minerals and organic matter: mechanisms and characterizations. Chemosphere , 359, 142383.Villarino, S.H., Pinto, P., Jackson, R.B. & Piñeiro, G. (2021). Plant rhizodeposition: A key factor for soil organic matter formation in stable fractions. Sci. Adv. , 7.Wang, S., Wilkes, A., Zhang, Z., Chang, X., Lang, R., Wang, Y., et al. (2011). Management and land use change effects on soil carbon in northern China’s grasslands: a synthesis. Agric. Ecosyst. Environ. , 142, 329–340.Wickings, K., Grandy, A.S., Reed, S.C. & Cleveland, C.C. (2012). The origin of litter chemical complexity during decomposition.Witzgall, K., Vidal, A., Schubert, D.I., Höschen, C., Schweizer, S.A., Buegger, F., et al. (2021). Particulate organic matter as a functional soil component for persistent soil organic carbon. Nat. Commun. , 12, 4115.Wright, C.L. & Lehtovirta-Morley, L.E. (2023). Nitrification and beyond: metabolic versatility of ammonia oxidising archaea. ISME J. , 17, 1358–1368.Yin, Q., Liu, Y., Li, J., Wu, J., Wang, Y., Zhang, H., et al. (2025). Soil organic carbon formation in grassland ecosystems: higher efficiency of roots than shoots and rhizodeposition. Soil Tillage Res. , 250, 106523.Zhang, X., Guo, J., Vogt, R.D., Mulder, J., Wang, Y., Qian, C., et al. (2020). Soil acidification as an additional driver to organic carbon accumulation in major chinese croplands. Geoderma , 366, 114234.Žifčáková, L., Větrovský, T., Lombard, V., Henrissat, B., Howe, A. & Baldrian, P. (2017). Feed in summer, rest in winter: microbial carbon utilization in forest topsoil. Microbiome , 5. Information & Authors Information Version history V1 Version 1 22 September 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords carbohydrate-active enzymes carbon fractions grassland carbon pool microbial community soil organic carbon Authors Affiliations Xiaozhen Liu 0000-0003-4208-0961 [email protected] Chinese Academy of Agricultural Sciences Grassland Research Institute View all articles by this author Huiyun Liu Northwest A&F University View all articles by this author Yanling Yin Shihezi University View all articles by this author Xiaoxue Zhang Chinese Academy of Agricultural Sciences Grassland Research Institute View all articles by this author Yuanyuan Jing 0000-0002-4762-5433 Chinese Academy of Agricultural Sciences Grassland Research Institute View all articles by this author Ke Ma Northwest A&F University View all articles by this author Chongyin Wang Chinese Academy of Agricultural Sciences Grassland Research Institute View all articles by this author Yafang Wang Northwest A&F University View all articles by this author Metrics & Citations Metrics Article Usage 186 views 137 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Xiaozhen Liu, Huiyun Liu, Yanling Yin, et al. Bridging the soil carbon pool to microbial taxa and associated CAZyme profiles in grassland ecosystems. Authorea . 22 September 2025. DOI: https://doi.org/10.22541/au.175856337.71958541/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175856337.71958541/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a003a604c8c88650',t:'MTc3OTUzNDk4Nw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();