Abstract:

Aims: Revealing the differences in species and metabolic functional groups of endophytic bacterial communities between tropical woody Fabaceae and Lauraceae plants, as well as their driving factors, contributes to understanding the adaptation and biodiversity maintenance mechanisms of tropical forests.
Method: In this study, Illumina Miseq sequencing platform was used to detect endophytic bacteria in Fabaceae and Lauraceae plants in the tropical mountain rainforest of Jianfengling in Hainan, and the FAPROTAX microbial geochemical cycle metabolic functional database was used to annotate the endophytic bacterial function.
Results: A total of 1,123 operational taxonomic units (OTUs) of endophytic bacteria belonging to 21 phyla, 36 classes, 51 orders, 92 families, and 160 genera were detected from four plant species, including two species (Ormosia semicastrata, O. balansae) of Fabaceae and two species (Cryptocarya chinensis, C. chingii) of Lauraceae. Among them, 600 OTUs were Proteobacteria, accounting for 57.17% of the total bacterial sequences, and 72 OTUs were Acidobacteria, accounting for 15.12%. The bacterial species of the Hymenobacter of Cytophagales, and Methylobacterium of Rhizobiales were the most abundant, with 37 and 27 OTUs, respectively. There were significant differences in endophytic bacterial species composition between Fabaceae and Lauraceae plants (ANOSIM: R = 0.5792, P = 0.004). The results of the environmental vector fitting analysis based on community non-parametric tests showed that the leaf potassium content and specific leaf area had the greatest impact on the species composition of endophytic bacterial communities. Endophytic bacteria with clear classification information, accounting for 54.63% total number of bacterial OTUs, were annotated to 28 metabolic functional groups. Of these functional groups, nitrogen fixation, aerobic chemoheterotrophy, cellulose degradation, methanol oxidation, methane oxidation, and urea degradation showed significantly higher relative abundance in non-legume Cryptocarya plants than in legume Ormosia plants. The results of non-metric multidimensional scaling analysis showed that bacterial metabolic functional groups were mainly influenced by specific leaf area and leaf phosphorus content.
Conclusion: The higher relative abundance of carbon and nitrogen metabolism functional groups in endophytic bacterial communities of non-legume plants in the Jianfengling tropical mountain rainforest may be one of their adaptive mechanisms to low effective nutrient soil environments.

Key words: woody legume, foliar endophytic bacteria, functional bacterial community, diversity, tropical rainforest