Biotic Interaction Underpins the Assembly Processes of the Bacterial Community Across the Sediment-Water Interface in a Subalpine Lake

Abstract

The sediment-water interface is the most active region for biogeochemical processes and biological communities in aquatic ecosystems. As the main drivers of biogeochemical cycles, the assembly mechanisms and the distribution characteristics of microbial communities at this boundary remain unclear. This study investigated the microbial communities across the sediment-water interface in a natural subalpine lake in China. The results indicated that the diversity of bacterial communities in middle sediment was significantly higher than that in overlying water and other sediments (p < 0.001). Pearson's correlation analysis indicated that the diversity was significantly influenced by biotic factors (e.g., diversity of fungus, protozoan and alga) and physicochemical parameters (e.g., total carbon, total organic carbon, nitrate, ammonium and pH) (p < 0.01). Null model analysis revealed that the homogeneous selection dominated the assembly of the bacteria community in sediment, whereas the heterogeneous selection dominated that in overlying water. The least squares path analysis showed that interactions between protozoa and bacteria had a greater impact on bacterial community assembly (p < 0.001). Important taxa influence the assembly by regulating biotic interactions. These findings provided a basis for understanding the importance of biotic interactions in maintaining subalpine lakes' ecosystems across the sediment-water interface.

Keywords: assembly process; biotic interaction; important taxa; protozoan; sediment–water interface.

Figure 1

Composition of bacterial communities at phylum (A), class (B) and genera (C) levels in overlying water (W), surface sediment (S1), middle sediment (S2) and bottom sediment (S3) in GH lake, as well as the composition of protozoan (D), fungal (E) and algal (F) communities at phylum levels.

Figure 2

The bacterial phylogenetic diversity (measured by the Faith index) (A), α-diversity (represented by the Shannon index (B) and Sobs index (C)) and its difference in bacterial communities at the overlying water (W), surface sediment (S1), middle sediment (S2) and bottom sediment (S3) in GH lake. “*” represents the degree of significance [p < 0.001 (***), p < 0.01 (**), p < 0.05 (*)].

Figure 3

Contributions of eukaryote community composition variables and physicochemical parameters to the diversity of bacterial communities based on correlation analysis in overlying water (W), surface sediment (S1), middle sediment (S2) and bottom sediment (S3). Colors represent Spearman correlations. Community composition variables in the model were a combination of microbial α-diversity, phylogenetic diversity (measured by Faith’s index), the first axis score of NMDS (NMDS1) and the second axis score of NMDS (NMDS2). “*” represents the degree of significance [p < 0.001 (***), p < 0.01 (**), p < 0.05 (*)].

Figure 4

Assembly processes of bacterial communities. (A) βNTI value based on a null model to determine the stochastic and deterministic processes of bacterial community assembly; (B) The relative contributions of five processes (heterogeneous selection; homogeneous selection; dispersal limitation; homogenizing dispersal; undominated process), based on βNTI and RC_Bray in assembly processes of bacterial communities.

Figure 5

(A) The positive-to-negative edges ratio of bacterial communities, bacterial communities with fungus (Bac-Fun) and protozoan (Bac-Pro) and alga (Bac-Alg) communities in W, S1, S2 and S3. (B) Robustness of bacterial communities, bacterial communities with fungus (Bac-Fun) and protozoan (Bac-Pro) and alga (Bac-Alg) communities in W, S1, S2 and S3. Different letters above the error bar indicate statistical difference among the robustness of four microbial communities (p < 0.05).

Figure 6

Effects of different variables on the assembly of bacterial communities (βNTI value) based on partial least squares path modeling. The red arrows represent positive pathways and the blue arrows indicate negative pathways. The positive-to-negative edges ratio (P/N) denoted biological factors; P/N-BP, P/N-BF and P/N-BA represent the positive-to-negative edges ratio of the bacterial community with protozoan, fungal and alga communities. Env represent the environmental heterogeneity (based on Euclidean Distance). The path coefficients are shown on the arrow. GOF, goodness of fit. “*” represents the degree of significance [p < 0.001 (***), p < 0.05 (*)].

Figure 7

Assembly processes of bacterial communities after removing important OTUs. (A) βNTI value based on the null model to determine the stochastic and deterministic processes of bacterial community assembly after removing important OTUs; (B) The relative contributions of five processes (heterogeneous selection; homogeneous selection; dispersal limitation; homogenizing dispersal; undominated process) based on βNTI and RC_Bray in assembly processes of bacterial communities after removing important OTUs.