Temporal patterns of bacterial communities in the Billings Reservoir system

Abstract

In this study, high-throughput sequencing of 16S rRNA amplicons and predictive PICRUSt functional profiles were used to perform a comprehensive analysis of the temporal bacterial distribution and metabolic functions of 19 bimonthly samples collected from July 2019 to January 2020 in the surface water of Billings Reservoir, São Paulo. The results revealed that most of the bacterial 16S rRNA gene sequences belonged to Cyanobacteria and Proteobacteria, which accounted for more than 58% of the total bacterial abundance. Species richness and evenness indices were highest in surface water from summer samples (January 2020), followed by winter (July 2019) and spring samples (September and November 2019). Results also showed that the highest concentrations of sulfate (SO₄^-2), phosphate (P), ammonia (NH₃), and nitrate (NO_3-) were detected in November 2019 and January 2020 compared with samples collected in July and September 2019 (P < 0.05). Principal component analysis suggests that physicochemical factors such as pH, DO, temperature, and NH₃ are the most important environmental factors influencing spatial and temporal variations in the community structure of bacterioplankton. At the genus level, 18.3% and 9.9% of OTUs in the July and September 2019 samples, respectively, were assigned to Planktothrix, while 14.4% and 20% of OTUs in the November 2019 and January 2020 samples, respectively, were assigned to Microcystis. In addition, PICRUSt metabolic analysis revealed increasing enrichment of genes in surface water associated with multiple metabolic processes rather than a single regulatory mechanism. This is the first study to examine the temporal dynamics of bacterioplankton and its function in Billings Reservoir during the winter, spring, and summer seasons. The study provides comprehensive reference information on the effects of an artificial habitat on the bacterioplankton community that can be used to interpret the results of studies to evaluate and set appropriate treatment targets.

Figure 1

Map showing sampling site locations in the Billings Reservoir in São Paulo. The map was generated using QGIS 3.22.0 (Redlands, CA, USA).

Figure 2

Seasonal variation of physicochemical properties of surface water samples from different sites of Billings reservoir during the study period (July 2019–January 2020).

Figure 3

Boxplot showing variation of physicochemical parameters (average water depth, temperature, dissolved oxygen (DO), pH, and turbidity) of surface water samples from different sites of Billings reservoir during the study period (July 2019–January 2020).

Figure 4

Boxplot showing variation of physicochemical parameters (sulfate (SO₄^–2), phosphate (P), ammonia (NH₃), and nitrate (NO₃)) of surface water samples from different sites of Billings reservoir during the study period (July 2019–January 2020).

Figure 5

Alpha diversity indices comparing the four sample groups. No significant differences were found between the groups.

Figure 6

Beta diversity computed by a principal coordinate analysis (PCoA) and PERMANOVA test and b pairwise comparisons of the bacterioplankton communities in the surface water of Billings reservoir during the study period (July 2019–January 2020).

Figure 7

Phylum level composition of bacterioplankton communities within the surface water of billings reservoir during the study period (July 2019–January 2020).

Figure 8

The distribution of the 10 most frequently detected bacterioplankton genera in all samples analyzed at four time points.

Figure 9

Boxplot showing the phylum relative abundance of the most abundant taxa of (A) Cyanobacteria, (B) Proteobacteria, (C) Actinobacteria, and (D) Bacteroidetes in four sample groups. Box borders represent the first and third quartiles, and the central lines represent the medians.

Figure 10

Principal component analysis (PCA) of the distribution of the 7 most abundant phyla (highlighted in color) relative to nine environmental variables in sample groups collected in A July 2019, B November 2019, and D January 2020. Samples are written in blue and indicated by a black dot.

Figure 11

Linear discriminant analysis effect size (LEfSe) of the five most affected eukaryotic phytoplankton bacteria with an LDA score higher than 2.0 and P values less than 0.05 in all sample groups.