Abstract: We analyzed the dynamic changes of the decomposition of fallen logs of Pinus koraiensis with different decay classes (ⅰ-ⅴ) and diameter classes (Ⅰ-Ⅲ) following the space-for-time substitution method. We further elucidated the decomposition mechanism by examining the interrelationships of chemical components, microbial communities, and related degradation enzyme activities during the decomposition process. The results showed that lignocellulose contents and the activities of related degrading enzymes were significantly varied across decay class and diameter class. The contents of lignin, cellulose, and hemicellulose generally showed a decreasing trend with increasing decay class. Compared with decay class ⅰ, lignin, cellulose, and hemicellulose contents in decay class ⅴ decreased by 17.2%-24.2%, 44.7%-64.7%, and 12.4%-45.2%, respectively, with cellulose showing the greatest decrease. Lignin content generally increased with increasing diameter class, whereas the diameter-related changes in cellulose and hemicellulose contents exhibited different patterns across decay classes. The activities of lignocellulose-degrading enzymes generally increased with increasing decay class. Compared with decay class ⅰ, the activities of cellulase, laccase, and β-glucosidase in decay class ⅴ increased by 160.1%-184.2%, 121.2%-337.9%, and 22.9%-185.1%, respectively. The variation patterns of degradation enzyme activities across diameter classes differed among decay classes and degradation enzyme types. Four functional modules were identified for bacteria and for fungi, respectively. Bacterial network Module 2 was identified as the key functional module for lignocellulose degradation, with Proteobacteria, Actinobacteria, Cyanobacteria, and Acidobacteria as the dominant phyla. The relative abundances of these phyla showed significant positive correlations with the activities of laccase, β-glucosidase, cellulase, and acidic xylanase. Modules 3 and 4 also showed significant positive correlations with some degradation enzyme activities. Module 1 showed no significant correlation with each degradation enzyme activity, indicating that multiple bacterial network modules synergistically participated in the decomposition of lignocellulose in fallen logs. All fungal network modules were dominated by Ascomycota and Basidiomycota, but none of them showed significant correlations with degrading enzyme activities. Bacterial community structure was significantly correlated with β-glucosidase, cellulase, and laccase activities, whereas fungal community structure was significantly correlated with β-glucosidase activity, nitrate content, ammonium content, and carbon-to-nitrogen ratio. Laccase, lignin peroxidase, β-glucosidase, and ammonium content were significantly correlated with changes in lignocellulose contents. In summary, key bacterial network modules played an important role in lignocellulose degradation, and the bacterial community drove the production of multifunctional enzymes for lignocellulose degradation. The fungal community showed no correlation with degrading enzyme activities, and its changes were primarily affected by substrate properties and nitrogen availability.

Key words: lignocellulose, microorganism, enzyme activity, co-occurrence network, redundancy analysis