Seasonal dynamics of free-living (FL) and particle-attached (PA) bacterial communities in a plateau reservoir

Abstract

In terms of lifestyle, bacterioplankton can be classified as free-living (FL) and particle-attached (PA) forms, and both play essential roles in biogeochemical cycling in aquatic ecosystems. Structure, distribution, and community assembly of FL and PA bacteria in plateau riverine waterbodies are largely unknown. Therefore, we explored the seasonal dynamics of FLand PA bacterial communities in the Wujiangdu reservoir, Yungui Plateau using 16S rRNA gene high-throughput sequencing. Results revealed there was a significant environmental heterogeneity in Wujiangdu reservoir seasonally. The dominant phylum was Actinomycetota for FL and Pseudomonadota for PA bacteria. Species richness and diversity was higher in autumn and winter compared to spring and summer. In general, PA diversity was greater than FL, but with some temporal variations. Species turnover was the major contributor to β-diversity of both FL and PA lifestyles, and significant differences were noticed between FL and PA bacterial community composition. Distinct co-occurrence network patterns implied that more connections exist between FL bacteria, while more complex PA networks were in parallel to their greater diversity and stronger interactions in biofilms on particles. Dispersal limitation was the major driving force for both FL and PA bacterial community assembly. Deterministic processes were of relatively low importance, with homogeneous selection for FL and heterogeneous selection for PA bacteria. Temperature was the most important environmental driver of seasonal bacterial dynamics, followed by nitrate for FL and Secchi depth for PA bacteria. This study allows for a better understanding of the temporal variability of different bacteria lifestyles in reservoirs in the vulnerable and rapidly changing plateau environment, facilitating further microbial research related to global warming and eutrophication.

Keywords: Wujiangdu reservoir; Yungui plateau; co-occurrence network; deterministic or stochastic processes; temporal bacterial dynamics.

Figure 1

Sampling locations in the Wujiangdu reservoir, China (5 sites: JK, TL, XT, XF and PY).

Figure 2

Seasonal variations of environmental variables. (A) Temp: water temperature; (B) TSS: total suspended solid; (C) SD: Secchi depth; (D) pH; (E) TN: total nitrogen; (F) TP: total phosphorus; and (G) PCA on all measured environmental factors.

Figure 3

Variations in relative abundance of (A) the top 10 phyla of FL and PA bacterial communities, and (B) across seasons in the Wujiangdu reservoir.

Figure 4

Seasonal variations of α-diversity of BC_FL and BC_PA in the Wujiangdu Reservoir (A) ASV richness; (B) Simpson; (C) Shannon; (D) Pielou’s Evenness; (E) ACE; (F) PD.

Figure 5

Seasonal variation in Bray-Curtis dissimilarity of BCFL and BCPA (A); β-diversity partitioning for BCFL and BCPA (B); and NMDS plot of BCFL (C) and BCPA (D).

Figure 6

Network analysis of BC_FL (A) and BC_PA (B); Zi-Pi plot of ASVs in BC_FL (C) and BC_PA (D) networks (the color of nodes refers to the phylum of the OTU; green edges: positive correlation; red edges: negative correlation; M1-M3 refers to modules in each network; Zi: within-module connectivity; Pi: among-module connectivity).

Figure 7

Relationship between community similarity (Bray-Curtis distance from the ASV table) of BC_FL and BC_PA related to (A,C) geographic distance and (B,D) environmental similarity; Bacterial community assembly processes for (E) BC_FL and (F) BC_PA. DL: dispersal limitation, HeS and HoS: heterogeneous and homogeneous selection, respectively.

Figure 8

RDA plot (A) BC_FL and (B) BC_PA and the percentage of variation explained by individual environmental variables for (C) BC_FL and (D) BC_PA.