Differences of bacterioplankton communities between the source and upstream regions of the Yangtze River: microbial structure, co-occurrence pattern, and environmental influencing factors

Abstract

The source area of the Yangtze River is located in the hinterland of the Qinghai-Tibet Plateau, which is known as the “Earth’s third pole.” It is the water conservation area and the natural barrier of the ecosystem of the Yangtze River basin. It is also the most sensitive area of the natural ecosystem, and the ecological environment is very fragile. Microorganisms play key roles in the biogeochemical processes of water. In this paper, the bacterioplankton communities in the source and upstream regions of the Yangtze River were studied based on 16S rRNA high-throughput sequencing, and their environmental influencing factors were further analyzed. Results showed that the upstream region had higher richness and diversity than the source region. The predominant bacterial phyla in the source and upstream regions were Proteobacteria, Firmicutes, and Actinobacteriota. The bacterial phyla associated with municipal pollution and opportunistic pathogen, such as Firmicutes and Actinobacteriota, were more abundant in the upstream. By contrast, distinct planktonic bacterial genera associated with mining pollution, such as Acidiphilium and Acidithiobacillus, were more abundant in the source region. The co-occurrence network showed that the interaction of bacterioplankton community is more frequent in the upstream. The bacterioplankton community compositions, richness, and functional profiles were affected by the spatial heterogeneity. Moreover, variation partitioning analysis further confirmed that the amount of variation in the source region independently explained by variables of altitude was the largest, followed by water nutrient. This paper revealed the spatial distribution of planktonic bacterial communities in the source and upstream regions of the Yangtze River and its correlation with environmental factors, providing information support for ensuring the health and safety of aquatic ecosystems in the Yangtze River Basin.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-024-01265-6.

Introduction

Rivers are an important component of the global water resource system. As the longest river in Asia and the third longest river in the world, the Yangtze River provides abundant freshwater resources. The Yangtze River, which is navigable for more than 2800 km and accounts for 80% of China’s inland waterway cargo volume, is known as China’s “golden waterway” and plays a vital role in the country’s economy [1]. The source region of the Yangtze River, located in the hinterland of the Tibetan Plateau, is well known as the “water tower of China” [2]. It is the world’s largest wetland ecosystem with the highest altitude and the most unique biodiversity characteristics. It is also the main supply station for freshwater resources in China and an important barrier for ecological security. Due to the particularity of geographical location, the source region is the most sensitive area of natural ecosystem and the ecological environment is relatively fragile [3]. The seasonal variation of temperature shows a unimodal pattern, the warm season with temperature greater than 0 °C is mainly concentrated in May to September every year, and the cold season with temperature less than 0 °C is October to April of the following year, and most of the areas are frozen during this period. The river reach from the source to Yichang city in Hubei province is known as the upper reaches of the Yangtze River, with numerous high mountains and valleys and an average gradient of 0.61‰ [4]. In addition, frequent human economic development activities have caused a series of ecological and environmental problems, such as desertification, climate warming and drought, severe snow disaster, and sharp decline in biodiversity [5]. In view of the unique geographical position and important economic status, it is of great significance to study the aquatic ecosystem in the source and upper reaches of the Yangtze River.

Bacteria are essential members of the river ecosystem and play a vital role in the biotransformation of organic matter and biogeochemical processes of nutrients [6, 7]. Moreover, bacterioplankton communities are considered to be the most diverse and abundant biogroup in the world [8], and they are extremely sensitive to environmental factors, such as hydrologic and geomorphic conditions [3], nutrient condition, and storm events [9, 10]. Previous studies found that anthropogenic activities also impacted bacterioplankton communities in freshwater ecosystems [11]. In order to adapt to different living environments, bacteria usually form a specific community structure to cope with various adverse effects [12]. Moreover, as it has received less attention than marine and lake microbiome, river microbiome is a hot topic in the current and future research [7]. Therefore, it is of great scientific value to elucidate the characteristics of phytoplankton communities and their responses to environmental pressure in unique plateau river environments.

Several previous researchers have documented bacterial communities in rivers and their responses to environmental factors. Ma et al. [13] found that the microbial population in higher latitudinal site was fewer than lower latitudinal one. Gao et al. [14] indicated that the bacterial community of sediment was mainly affected by seasonal properties, while the bacterial community of water were affected by both seasons and anthropogenic activities. Guo et al. [15] showed that bacterial abundance was significantly correlated with salinity, SO₄²⁻ and total organic carbon, while bacterial diversity was significantly correlated with SO₄²⁻ and total nitrogen. Some environmental indicators that are critical to the formation of microbial communities, such as nitrogen, phosphorus, and pH, are frequently examined [16, 17].

With the intensification of global climate change, the ecological environment of the Yangtze River has been significantly affected by climate warming in the past 10 years. The hydrological cycle and the spatial and temporal distribution pattern of water resources have changed, and the ecological environment has been threatened. This series of issues has become the focus of widespread attention in China and abroad. Several previous studies have documented the benthic animals, phytoplankton community structures, and hydrological components in the Yangtze River [5, 18, 19]. Currently, the importance of bacterioplankton community to the ecosystem and the threat from anthropogenic activities are the focus of aquatic ecosystem studies [7]. However, no study has yet been conducted about the spatial distribution and environmental influencing factors of bacterioplankton communities in the source and upstream regions of the Yangtze River.

Accordingly, in this study, we aimed to investigate the structure and potential function of bacterioplankton communities and their environmental influencing factors in the source and upstream regions of the Yangtze River. To this end, a series of sampling sites were selected from the source and upstream regions of the Yangtze River to evaluate the bacterioplankton communities by high-throughput sequencing. The main objectives of our study were to (i) describe the spatial distribution patterns of bacterioplankton community and (ii) determine the effects of environmental factors on bacterioplankton communities.

Materials and methods

Sample collection

The Yangtze River Basin research survey was conducted from August 2018 to May 2019. During this period, water samples were collected along the source and upstream regions of the Yangtze River. In this study, a total of 18 water samples were collected (Fig. 1), including ten sites at the source (Y1–Y10) and eight sites at the upstream (Y11–Y18). Detailed geographic location information (longitude and latitude coordinates) for these sites was recorded (Table S2). Two liters of surface water samples (0.5 m depth) was collected with a 2.5-L water sampler. At each site, water samples were collected in triplicate at 5-m intervals, and then mixed into one sample. One liter of water samples was filtered through a 20-μm mesh (Millipore Corporation, Billerica, MA, USA) to remove large particles or organisms and subsequently filtered through 0.22-μm polycarbonate membranes (Millipore Corporation, Billerica, MA, USA). The filters were frozen at − 80 °C until further processing. Additional water samples were temporarily stored in dark cooling boxes at 4 °C for physicochemical analyses.

Fig. 1.

Study area and sampling sites of this study. The source region contains Y1–Y10. The upstream region contains Y11–Y18

Physicochemical analysis

The temperature (T), pH, and dissolved oxygen (DO) were measured in situ during sampling using an YSI Pro2030 (YSI, Ohio, USA). Samples for total phosphorus (TP), total nitrogen (TN), permanganate index (COD_Mn), NH₄⁺-N, arsenic (As), and metal ions (Cr⁶⁺, Cu, Zn, Pb, Cd, and Hg) were analyzed by Qinghai Fishery Environment Monitoring Station. COD_Mn was measured using the acidic potassium permanganate oxidation method. TP, TN, and NH₄⁺-N of water were analyzed following national standard GB 11893–89, HJ 636–2012, and HJ 535–2009 [14, 20]. Cr⁶⁺ were determined by with a Double Beam UV–visible Spectrophotometer (TU-1901, T9CS) using the acidic potassium permanganate oxidation method. Samples were pre-digested with HNO₃ and H₂O₂ in the microwave Accelerated Reaction system (Mars5, CEM microwave Technology, USA) to detect and analyze Hg content. Nitric acid-sulfuric acid-perchloric acid pre-digested samples were used to detect As content. Then, the As and Hg were measured with an atomic fluorescence photometer (AFS-230E) [21]. The metal ions were determined with a graphite furnace flame atomic absorption spectrophotometer (ICE-3500, AA6300).

DNA extraction, PCR amplification, and sequencing

Microbial community genomic DNA was extracted from the water samples using the E.Z.N.A.® water DNA Kit (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s protocol. The quality and the quantity of final DNA were examined by agarose gel (1%) electrophoresis and a NanoDrop 2000 UV–vis spectrophotometer (Thermo Scientific, Wilmington, USA). The V3–V4 hypervariable regions of the bacterial 16S rRNA gene were amplified with primers 338F (50-ACTCCTACGGGAGGCAGCAG-30) and 806R (50-GGACTACHVGGGTWTCTAAT-30) by a thermocycler PCR system (GeneAmp 9700, ABI, USA). All DNA PCR reactions were carried out in a 50 μl reaction volume, containing 25 μl of 2 × Taq PCR mix, 3 μl Primer F (10 μM), 3 μl Primer FR (10 μM), 3 μl diluted DNA, and 16 μl double distilled H₂O. PCR amplification was initiated denaturation at 95 °C for 3 min, followed by 27 cycles at 95 °C for 30 s, 55 °C for 30 s, 72 °C for 45 s, and a final extension step at 72 °C for 10 min. The resulting PCR products were extracted from a 2% agarose gel, further purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) and quantified using QuantiFluor™-ST (Promega, USA) according to the manufacturer’s protocol. Purified amplicons were pooled in equimolar and paired-end sequenced (2 × 300) on an Illumina MiSeq platform (Illumina, San Diego, USA) according to the standard protocols by Majorbio Bio-Pharm Technology (Shanghai, China). The raw reads were deposited in the NCBI Sequence Read Archive database (Accession Number: PRJNA762667 and PRJNA835968).

Processing of sequencing data

Raw fastq files were demultiplexed, quality-filtered by Trimmomatic, and merged by FLASH with the following criteria: (i) The 300-bp reads were truncated at any site receiving an average quality score of < 20 over a 50-bp sliding window, and truncated reads shorter than 50 bp were discarded. Reads containing ambiguous characters were also discarded. (ii) Only overlapping sequences longer than 10 bp were assembled according to their overlapped sequence. The maximum mismatch ratio of the overlap region was 0.2. Reads that could not be assembled were discarded. (iii) Samples were distinguished according to the barcode and primers, and the sequence direction was adjusted as follows: exact barcode matching and two nucleotide mismatch in primer matching.

Operational taxonomic units (OTUs) with 97% similarity cutoff were clustered using the UPARSE version 7.1 [22, 23], and chimeric sequences were identified and removed. The taxonomy of each OTU representative sequence was analyzed by RDP Classifier version 2.2 [24] against the 16S rRNA database (SILVA SSU 132) using a confidence threshold of 0.7.

Data analyses

The sequences of each sample were randomly resampled to 28,392 reads on the minimum number of valid sequences to eliminate the influence of sequencing depth on subsequent analysis. According to the OTU information, Qiime (version v.1.9.1) was used to calculate the alpha diversity index, including Chao 1, ACE, Shannon, Simpson, and Good’s coverage. Beta diversity was analyzed with principal coordinates analysis (PCoA) and analysis of similarity (ANOSIM) based on the Bray–Curtis metric. Nonmetric multidimensional scaling was performed on the OTU data using Bray–Curtis distance matrices to examine the differences in bacterial community among the water samples. Co-occurrence patterns of bacterial communities was completed using the Wekemo Bioincloud (https://www.bioincloud.tech). The environmental factors were screened through a variance inflation factor (VIF) analysis to retain those with low collinearity. The VIF is a measure of collinearity among predictor variables within a multiple regression. It is calculated by taking the variance ratio of all betas of a given model divided by the variance of a single beta [25]. A canonical correspondence analysis (CCA) was performed to determine the environmental variables associated with the changes in the bacterial community structure using the vegan package. In this study, linear discriminant analysis effect size (LEfSe) analysis was performed to identify microbial biomarkers and functional differences with an alpha parameter of 0.05 and a linear discriminant analysis (LDA) threshold value of 2.0. Differences between two independent groups were evaluated using the Mann–Whitney U test. Tax4Fun was used to predict the microbial functions and metabolic pathways. Moreover, the functional profiles of the bacterioplankton communities were predicted using Tax4Fun from the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways [26].

Results

Environmental characteristics

The concentrations of heavy metals in river water are listed together with the other traditional physiochemical indices in Table S1. The elevation of the ten sampling sites in the source region of the Yangtze River ranged from 3517 to 4577 m, and the sites were generally considered high plateau areas. Spatial and temporal differentiation was not present in pH, which ranged from 7.8 to 8.66, indicating a weak alkalinity. Similarly, the DO contents did not show a clear spatial distribution. However, significantly (p < 0.01) higher values of T° (from 25.3 to 25.8) were registered during summer in the upstream of the Yangtze River compared with those (from 7.0 to 14.4) registered in the source of the Yangtze River.

From ten sampling sites in the source of the Yangtze River, except COD_Mn and As, environmental factors remained consistent, and there was no significant difference in other indicators (Mann–Whitney U test, p > 0.05). Overall, the Cu, Zn, Pb, Cd, and As concentrations were below the class I water quality according to the “Environmental Quality Standard for Surface Water” (GB3838-2002). The Cr⁶⁺ concentration satisfied class II, and the TN and Hg concentrations were Class III. TN ranged from 0.33 to 1.2 mg/L in the source region, and it was higher in Y7 and Y10 than in other sampling sites (p > 0.01). Of note, As was highest in Y1 (2.9 mg/mL, p < 0.01) than in other sampling sites. Collectively, the levels of these indices are closely influenced and connected by multiple anthropogenic activities.

Diversity and structure of the bacterioplankton community

Based on high-throughput sequencing, a total of 877,478 high-quality sequences (ranging from 30,319 to 73,887 per sample) were generated from 18 sampling sites, with an average sequence length of 426 bp (Table S3). Then, after random resampling, the reserved sequences were clustered into a total of 1162 OTUs. Good’s coverage ranged from 99.23 to 99.92% (Table S3). The rarefaction curves for combined samples are presented in Fig. S1. These results indicated that the majority of the microbial species present in the samples were detected.

Based on the OTU number, the water sample from Y14 was found to have the highest level of Chao 1 and ACE index, with a leading OTU number of 843; conversely, the samples from Y6 and Y17 displayed considerably lower level of Chao 1 and ACE index and consistently had the lowest number of OTUs (142 and 140, respectively; Table S3). Statistical analysis found no significant differences (Welch t-test, p > 0.05) between the groups in observed species and diversity (Simpson and Shannon index), but found significant differences (Welch t-test, p < 0.05) between richness (Chao 1and ACE). The upstream region had the highest Chao 1 richness (750.74 ± 351.16) and Shannon index (3.23 ± 1.68), which were higher than those in the source region (413.03 ± 159.38 and 2.78 ± 1.24) (Fig. 2; Table S3). In short, by linking the index values, we found that the bacterioplankton communities of the upstream were richer and more diverse than those of the source region.

Fig. 2.

Planktonic Bacterial diversity in water samples from Yangtze River. Significant differences between sampling sites areas are marked by different letters

The Venn diagram shows unique and shared bacterial OTUs between water samples in the source and upstream regions of the Yangtze River. As shown, the water samples from the source and upstream regions contained 610 and 592 unique OTUs, respectively, whereas 553 OTUs were commonly shared (Fig. S2). Multivariate statistical analyses were performed to compare the overall structure of the bacterioplankton communities in different sites. ANOSIM indicated that the bacterioplankton community structure in the source and upstream regions was significantly different (R = 0.7277, p = 0.001). The PCoA plot corroborated the ANOSIM results, which showed distinct separations of the bacterioplankton communities between source and upstream sampling sites, which showed that coordinate 1 and 2 explained 32.8 and 15.16% of variations, respectively (Fig. 3a). In addition, ANOSIM demonstrated distinct separation and significant difference (R = 0.8507, p = 0.001) in the bacterioplankton communities between the two regions (Fig. 3a). The hierarchical clustering tree on the OUT level determined the phylogenetic structure of the bacterioplankton communities (Fig. 3b). It clustered the sampling sites into two groups: Y1–Y10 were clustered into the first group representing the source region; Y11–18 were clustered into the second group and comprised the upstream region.

Fig. 3.

Principal coordinate analysis (PCoA) of the total samples based in Bray–Curtis distances (a). Hierarchical clustering tree of the eighteen samples taken along the upstream of the Yangtze River based on the Bray–Curtis distances (b)

Comparison of the bacterioplankton community structure between source and upstream samples

The phylogenetic classification of sequences from 18 samples was classified into 26 bacterial phyla, 55 classes, 125 orders, 237 families, and 695 genera. Across all sampling sites, the explanation for a large proportion of the phyla was as follows: Proteobacteria (1.03–95.94%) was the largest phylum, followed by Firmicutes (0–97.87%), Actinobacteriota (0.02–43.97%), and Chloroflexi (0–36.38%), which accounted for almost 90% of the total bacterial abundance in all sampling sites (Fig. 4a; Fig. S3a) and exhibited distinct composition difference at the class level. We further analyzed the composition of bacterioplankton communities at the class level. Gammaproteobacteria and Alphaproteobacteria were the dominant classes of Proteobacteria, and Bacilli and Clostridia were the main classes in Firmicutes, Actinobacteriota, and Chloroflexi (Fig. S4). Bacterioplankton communities varied significantly between the source and upstream regions. In the source region, the phylum with the highest average relative abundance was Proteobacteria (68.08%), followed by Firmicutes (19.43%), Chloroflexi (3.76%), and Bacteroidota (2.75%) (Fig. 4). The relative abundance of Firmicutes at Y1, Y7, and Y9 was higher than that at other sampling sites, with relative abundances of 39.66%, 26.42%, and 82.56%, respectively. The relative abundance of Chloroflexi at Y4, which mainly comprises the class of Chloroflexia (Fig. S4), was higher than that at other sampling sites, with a relative abundance of 36.38%. In comparison, the most prevalent phyla in the upstream region were Firmicutes (30.80%), followed by Proteobacteria (28.94%) and Actinobacteriota (24.08%) (Fig. 4). The relative abundances of Firmicutes at Y17 and Y18 were higher than those of other sites, with relative abundances of 97.87% and 84.08%, respectively.

Fig. 4.

The relative abundance of dominant phyla (a) and genus (b) in the bacterioplankton community in the source and upstream regions of the Yangtze River

The bacterioplankton communities between the source and upstream regions were significantly different at the genus level (Fig. 4b). Pseudomonas, Acinetobacter, Exiguobacterium, and Sphingomonas were shared in source and upstream regions. Acidiphilium, Acidithiobacillus, Romboutsia, and Herpetosiphon were unique in the source region, and Planomicrobium and CL500-29_marine were present only in upstream samples (Fig. S3 b–d). Pseudomonas (0.62–67.44%) was the most abundant genus within the source region. Remarkably, Acidiphilium (90.83%) was the most abundant genus in Y10, whereas only a few Acidiphilium OTUs were detected in all other sampling sites (Fig. 4b). Planomicrobium and CL500-29_marine were abundant in Y17 (75.47%), Y18 (77.47%), and Y11–Y14 (14.60–32.66%).

In the co-occurrence network analysis, the upstream shows a larger co-occurrence network than the source (Fig. 5a and b). Spearman correlation analysis showed that the source network had 50 nodes and 494 edges, whereas the upstream network had 50 nodes and 945 edges (Fig. 5). In general, taxa tend to co-occur (positive correlation, red line) rather than co-exclude (negative correlation, blue line). The values of topological parameters reflect the complexity of the network. Positive correlation represents symbiosis and parasitism, while negative correlation reflects predation and competition among bacteria [27]. The proportion of positive correlation in the upstream (88.25%) is significantly higher than that in the source (70%), whereas the proportions of negative correlation were 11.75% and 29% for upstream and source planktonic bacterial co-existence patterns, respectively. Taking all the correlations into account, the links between upstream bacteria were more complex than those between downstream bacteria, suggesting that the potential interactions in upstream planktonic bacterial networks were stronger. According to previous research [28], the higher the proportion of competition between bacterial species, the better the stability of the network under environmental interference. In the upper reaches of the Yangtze River, which has high bacterial richness and diversity, the interaction between bacterial species is also strong, which further complicates their networks. Therefore, the bacterial networks in the source region of the Yangtze River may be more susceptible to environmental changes than upstream. Proteobacteria and Firmicutes were highlighted as key taxa in the source, whereas Firmicutes and Actinobacteriota were highlighted as key taxa in the upstream. The Wilcoxon rank-sum test showed that Proteobacteria and Bdellovibrionota were significantly more prevalent in the source region than in the upstream region (p < 0.01). However, the abundances of Actinobacteriota and Acidobacteriota were significantly higher in the upstream region than in the source region (p < 0.01) (Fig. 5c).

Fig. 5.

Co-occurrence networks of bacterial communities based on correlation analysis: source (a); upstream (b). A connection stands for a strong (Spearman r > 0.4 or r <  − 0.4) and significant (P-value < 0.01) correlation. The lines between each pair of nodes represent positive (in red) and negative (in blue) interactions. The size of nodes is proportional to the link numbers of each node. Phylum types which had significant difference (P-value < 0.05) of relative abundance between the source and upstream regions (c). LEfSe identified the most differentially abundant taxa in specific sampling sites. Only taxa meeting an LDA significant threshold > 2 were shown in the figure (d)

We further confirmed the presence of different OTUs between two groups by LEfSe. The taxa with LDA values greater than 4 are shown in Fig. 5d. The significantly enriched microbial taxa in the source region were Proteobacteria (class, Gammaproteobacteria; order, Acetobacterales, Burkholderiales, and Pseudomonadales; family, Acetobacteraceae, Burkholderiaceae, and Pseudomonadaceae; genus, Methylobacterium-Methylorubrum, Acidiphilium, and Pseudomonas) and Chloroflexi (class, Chloroflexia). Meanwhile, Acidobacteriota, Cyanobacteria (order, Synechococcales; family, Cyanobiaceae; genus, Cyanobium_PCC-6307), Exiguobacterales (family, Exiguobacteraceae; genus, Exiguobacterium), and Actinobacteriota (class, Actinobacteria and Acidimicrobiia; order, Micrococcales and Microtrichales; family, Ilumatobacteraceae; genus, CL500-29_marine_group) were significantly enriched in the upstream region.

Effect of environmental parameters on bacterioplankton community

CCA analysis was used to interpret the contribution of environmental parameters on the variation of microbial community structures in the source and upstream regions of the Yangtze River (Fig. 6). To eliminate the strong collinearity among specific variables, only significant environmental parameters tested by VIF analysis were included in the model. Eight physicochemical parameters (altitude, TN, TP, COD_Mn, DO, NH₄⁺-N, Cr⁶⁺, and As) in source sampling sites (Fig. 6a) and three physicochemical parameters (T, pH, and DO) in upstream sampling sites (Fig. 6b) were taken into consideration to evaluate the relative contributions to bacterioplankton community and genera. In the source region, among all the environmental factors investigated, altitude (R² = 0.628, p = 0.02) apparently has the most significant influence on the bacterioplankton community, and most variances can be explained by TN (R² = 0.222, p = 0.426) and COD_Mn (R² = 0.273, p = 0.323). In terms of the correlation between environmental factors and genera, the results showed that TN contributed positively to Acidiphilium, Acidithiobacillus, and Romboutsia but contributed negatively to Pseudomonas and Herpetosiphon. Pseudomonas was positively correlated with all selected environmental factors except TN and DO. In the upstream region, the first and second axes explained 46.55% and 23.47% of the total variance, respectively (Fig. 6b). DO (R² = 0.932, p = 0.005) and T (R² = 0.818, p = 0.033) were the main environmental factors affecting the bacterioplankton community. Planomicrobium was the only core bacterium in Y17 and Y18, which was positively correlated with T and negatively correlated with DO and pH. Additionally, variation partitioning analysis was conducted by using the same physicochemical variables in the CCA. The environmental factors of the source region were categorized into four groups: altitude, heavy metal (Cr⁶⁺ and As), water nutrient (TN, TP, COD_Mn, DO, NH₄⁺-N), and pH (Fig. 6c). Moreover, the amount of variation explained by altitude (20.4%) was the largest, followed by water nutrient (17.15%), whereas heavy metals explained 3.17% of the variation. In the upstream region, the amount of variation explained by DO (60.28%) was the largest, whereas pH and T independently explained 9.73% and 8.38% of the variance, respectively (Fig. 6d).

Fig. 6.

Canonical correspondence analysis (CCA) of the bacterioplankton community, environmental parameters and water samples of the source (a) and upstream (b) regions of the Yangtze River. Variation partitioning between heavy metal, altitude, water nutrient, and pH on bacterial community structure of source region (c). Variation partitioning between DO, T, and pH on bacterial community structure of upstream region (d)

Functional prediction by Tax4Fun

A community prediction analysis using Tax4Fun was performed to determine the planktonic bacterial functions at different sites (Fig. 7). To gain more insights into the functional differences, we compared the relative abundance of KEGG at level 2 (Fig. 7a). The results of Tax4Fun revealed that planktonic bacterial assemblages had higher abundance related to carbohydrate metabolism, amino acid metabolism, membrane transport, signal transduction, metabolism of cofactors and vitamins, and energy metabolism. At KEGG level 3, a relative abundance greater than 1% of total detected pathways was defined as a dominant pathway. Compared with the upstream, the functional pathways related to two-component system, starch and sucrose metabolism, and amino sugar and nucleotide sugar metabolism were significantly enriched (LDA > 2) in the source samples (Fig. 7b). By contrast, the functional pathways involved in aminoacyl-tRNA biosynthesis; ribosome; glycine, serine, and threonine metabolism; oxidative phosphorylation; pyrimidine metabolism; and glyoxylate and dicarboxylate metabolism were significantly enriched (LDA > 2) in the source samples (Fig. 7b).

Fig. 7.

(a) The relative abundance of various dominant predicted functions of bacterial communities (> 1%) in different sites using Tax4Fun grouped into level 2 functional categories. (b) LEfSe identified the most differentially abundant pathways at level 3 between source (red) and upstream (blue) regions with LDA scores higher than 2

Discussion

The composition of bacterioplankton communities in the Yangtze River has been widely reported [15, 21, 29]. However, most of previous studies have focused on a certain reach of the Yangtze River, and the composition of bacterioplankton communities in the source region and the spatial distribution of the bacterioplankton communities along the Yangtze River remain largely unexplored. Here, we investigated the spatial distribution patterns of the bacterioplankton community and its response to the environmental factors in the source and upstream regions of the Yangtze River.

The aquatic bacterioplankton community is a dynamic complex system with numerous interacting microorganisms, which are controlled by various scale dependencies and stochastic forces [30]. In our study, the bacterioplankton abundance in the source region (an average Chao 1 index of 413.03) was lower than those in the upstream region (an average Chao 1 index of 750.74) and also lower than those previously reported in water bodies from Songhua River and Yellow River (average Chao 1 indices of 2886.73 and 522, respectively) [31, 32]. Compared with the Songhua River and Yellow River, the bacterial diversity indices have the same trend as the Chao 1 index [31, 32]. These results showed that the source region has relatively low bacterial abundance and diversity, which might be attributed to high altitude and low temperature in the source plateau region. Another reason may be that higher solar UV radiation and lower concentration of atmospheric in the source region of the Yangtze River will affect the composition of microorganisms in the atmosphere and surface soil, and then indirectly affect the microorganisms in the source area of the Yangtze River through rainwater flowing into the river. Wu et al. reported that the molecular structure of DNA and RNA can be destroyed by high UV intensity, which can lead to low microbial diversity in plateau rivers [33]. In addition, the index pattern of low Chao 1 and Shannon is consistent with that of other plateau rivers, such as the Huangshui River and Heihe River [25, 34], which further verified our results. Contrary to alpha diversity indices, the similarity of microbial communities was higher in the source region compared with the upstream region. Compared with the upstream sites, the source sampling sites were geographically farther apart but the bacterioplankton communities were more similar, indicating that the spatial heterogeneity of the bacterioplankton communities in the source region was significantly less.

Our results showed that microbial communities fluctuated greatly in terms of taxonomic composition and spatial distribution. In general, Proteobacteria was the most dominant phylum, followed by Firmicutes, Actinobacteriota, Chloroflexi, and Bacteroidota, which constituted nearly 92% of the total microbiota in all sampling sites. The dominant planktonic bacterial phyla are ubiquitous in the high-elevation (plateau) and low-elevation (plain) environments according to previous reports [35–37]. Among them, Proteobacteria was the most abundant in all samples in the source region and mainly consisted of Gammaproteobacteria and Alphaproteobacteria, whereas Firmicutes was the most abundant in the upstream region. The next most abundant phyla were Firmicutes in the source region and Proteobacteria and Actinobacteriota in the upstream region. Gammaproteobacteria predominated in all source region samples, followed by Alphaproteobacteria and Clostridia. However, although Gammaproteobacteria and Alphaproteobacteria can be found in freshwater and seawater, they are more abundant in seawater [38]. Gammaproteobacteria are widely distributed in eutrophic and oligotrophic environments [39]. This discovery may be due to the salt mineral resources, such as gypsum and calcium in Glauber’s salt mine, which are widely distributed along the banks of the source region of the Yangtze River. Firmicutes and Actinobacteriota are prevalent in animal rumen [40], fecal discharge [41], and polluted sites [42]. Numerous Firmicutes species are associated with human activities. The significantly higher abundance of these organisms in the upstream region of the Yangtze River is probably a result of industry and domestic waste-water discharge during urbanization. Notably, the abundance of Firmicutes made up almost all of the abundance in Y17 and Y18 (97.87 and 84.08%). This may be because the two sampling sites are close to the Three Gorges Dam, which is a tourist attraction and is greatly influenced by human activities. Moreover, the pollutant deposition in front of the dam enriches Firmicutes. Many bacteria within the phylum Bacteroidota have been separated, and it is proved to be critical to the organic matter mineralization in the environment [43]. A previous study also found that the Bacteroidota phylum was closely related to the eutrophication of water bodies [34]. In this study, the abundance of Bacteroidota was higher in the source region compared with other samples, indicating that the water body in the source region was in eutrophication state. Complementarily, the co-occurrence network showed that Actinobacteriota, Acidobacteriota, Patescibacteria, and Planctomycetota were four keystone taxa exclusively associated with the upstream region, whereas Proteobacteria, Firmicutes, Chloroflexi, Deinococcota, Bacteroidota, and Cyanobacteria were six key phyla associated with both regions. The bacterial networks were more intricate in the upstream than in the source, which indicates that the interaction of bacterial community is more frequent in the upstream.

At the genus level, the microbial composition was different between the source and upstream regions. Pseudomonas (26.86%), Acidiphilium (10%), and Acidithiobacillus (5.46%) had the highest abundances in the source region, whereas Planomicrobium (20.10%), CL500-29_marine_group (13.33%), and Acinetobacter (10.94%) were dominant in the upstream. Pseudomonas, a potential pathogen, has been recognized for bioremediation purposes such as degradation of hydrocarbons, pesticides, and dyes [44, 45]. Pseudomonas had the highest abundance in Y5 and Y6, accounting for 64.15% and 67.44%, respectively, indicating that these sampling sites are possibly associated with the discharge of untreated effluents from the industry and pose potential health risks. Acidiphilium and Acidithiobacillus are heterotrophic acidophiles, which are the most abundant heterotrophs in mineral-leaching environments or acidic aquatic environments [46, 47]. However, the pH of the source region was weak alkaline, which was not suitable for heterotrophic acidophiles. Qiao found that there are different levels of Zn, Pb, and Cd contamination in sediments related to lead–zinc deposits in the vicinity of the area, which are higher than the background value under natural conditions [48]. Apart from its unique hydrology, biology, climate, and landform, the study area had notable geological and metallogenic conditions [25]. We hypothesized that these acidophilic bacteria in the source region originated from sediments along the river and were influenced by strong winds and rainfall. Members of the genus Planomicrobium have been isolated from fermented seafood, marine mud, Antarctic samples, intertidal sediments, and glacier and marine solar salterns [49]. These are extreme, hypersaline, and humid environments, indicating the high tolerance of the strain toward its environment [50]. Planomicrobium has significant activity for lipase, protease, and cellulose biodegradation, highlighting the importance of the functions of ecosystem services that this genus provides [51, 52]. CL500-29_marine_group was found primarily in marine ecosystems and was confirmed to exist in freshwater rivers and lakes by Zwart [53]. Furthermore, CL500-29_marine_group can effectively use different forms of carbon-based compounds [54] and has been shown to play a dominant role in water quality purification [55]. Bacteria of the genus Acinetobacter belong to the highly diverse class Gammaproteobacteria [56]. Some Acinetobacter species are identified as opportunistic pathogens in humans, which can easily adapt to diverse environments, including natural environment and hospital settings [57].

Spatial distance and environmental variables are the two main factors that drive microbial distribution patterns in rivers. In this study, the bacterioplankton communities in the source and upstream regions were clustered into two groups. The bacterioplankton communities in rivers can be influenced by many factors such as nutrient concentration [58], DO [45], and heavy metals [48]. Consistent with previous studies [59, 60], our findings showed that altitude, TN, and COD_Mn were the main factors affecting the bacterioplankton communities in the source region. Zhang et al. [25] found that altitude was a significant environmental factor to explain the variation of bacterioplankton communities in plateau rivers, implying the important role of altitude to plateau rivers. Altitude affects the composition of planktonic bacteria mainly by influencing physicochemical indexes and geographical isolation of water. Specifically, the change of altitude leads to the change of physical and chemical indexes of the water body, including water temperature and dissolved oxygen. Secondly, geographic isolation between habitats caused by altitude hinders the dispersal of bacteria between habitats, resulting in differences in bacterial community [27]. TN was thought to affect the microbial community by affecting the availability of these elements and further influencing the microbial biomass [61]. Our results showed that DO explained the largest portion of the variation of microbial communities in upstream region. The present study also confirmed that pH could also affect the structure of bacterioplankton community. Moreover, pH was previously considered to be a major factor affecting the formation of bacterioplankton community [62]. The low correlation between pH and the bacterioplankton communities in the source region might be attributed to the fact that the pH range of previous studies (pH 4–9) was larger than that for source samples (pH 8.27–8.66). Nutrient composition may be an important factor affecting microbial community structure [63]. While only DO in upstream samples was found to be significantly correlated with the bacterioplankton community (p = 0.005), TN and DO from source samples showed non-significant correlation, with p values of 0.426 and 0.736, respectively. The bioavailable forms of phosphorus, nitrogen, and carbon might be more utilized by microorganisms and might show closer associations with the composition of bacterioplankton community [15]. Therefore, the response of microbial communities to specific environmental factors, such as bioavailable forms of phosphorus, nitrogen, and carbon, needs to be further studied.

Understanding the functional characteristics of bacterioplankton communities is of great significance to the study of ecosystem processes and community assembly mechanisms [25]. In this study, the microbial functions involved in carbohydrate metabolism, amino acid metabolism, and membrane transport were enriched in both source and upstream samples. These pathways, as the metabolic pathways of core resources, are potential drivers for microbial community structure and function in rivers [64]. The function of the two-component system was the highest in the source samples, whereas aminoacyl-tRNA biosynthesis was more abundant in the upstream samples. The two-component system is a common stimulus–response mechanism that allows bacteria to perceive and respond to variable environmental conditions through a series of phosphorylation reactions [65]. Furthermore, the two-component system is beneficial for bacteria to maintain, adapt, and protect themselves against a series of environmental stressors [61]. The high function of the two-component system in the source region may be due to the harsh environment in the plateau, such as low temperature, hypoxia, and strong UV radiation. A previous study proposed that the inhibition of Aminoacyl-tRNA biosynthesis is an effective antimicrobial strategy [66], which may be due to the higher abundance of Acinetobacter, an opportunistic pathogen, in upstream samples. Our results indicated that water in different habitats could increase the relative abundance of metabolic activities associated with their characteristics to help them resist environmental changes.

Conclusion

In conclusion, this paper is the first to investigate the spatial distribution and influencing factors of bacterioplankton communities in the source and upstream regions of the Yangtze River. In general, the bacterioplankton communities between two areas were different, and obvious spatial patterns of the bacterial community were observed. Higher richness and diversity were found in the upstream region. The dominant planktonic bacterial groups at the phylum level were Proteobacteria, Firmicutes, Actinobacteria, and Chloroflexi. Distinct planktonic bacterial genera associated with mining pollution, such as Acidiphilium and Acidithiobacillus, were more abundant in the source region. By contrast, the planktonic bacterial phyla associated with municipal pollution and opportunistic pathogen, such as Firmicutes and Actinobacteriota, were more abundant in the upstream region. The co-occurrence network showed that the interaction of bacterioplankton community is more frequent in the upstream region compared with that in the source region. The bacterial communities in the source region were mainly affected by altitude, whereas those in the upstream region were affected by dissolved oxygen and temperature. However, due to the difficulty of upstream sample collection, the influence of environmental factors on the bacterioplankton community is not sufficiently studied. In the future, more large-scale scientific investigations should be carried out to comprehensively explore the impact of human activities and environmental changes on the aquatic ecosystem of the Yangtze River Basin.

Supplementary information

Below is the link to the electronic supplementary material.

Author contribution

Qianqian Zhang was responsible for the design, data analysis, and writing. Jinyong Zhang performed statistical analyses. Juan Zhao, Shenglong Jian, Guojie Wang, and Hongtao Guan conducted sampling and the lab work. Shuyi Wang and Jicheng Yang conducted the experimental work. Lijian Ouyang put valuable suggestions to the review comments. Zhenbing Wu conducted the experimental work, reviewed, and edited the manuscript. Aihua Li obtained the funding. All authors read and reviewed the manuscript.

Funding

This work was supported by grants from the Weifang Xiashan Reservoir Management Service Centre (SDZTXS-20211115), National Natural Science Foundation of China (No. 32073023), Qinghai KLPA project (2019/2020), Key Project of Scientific and Technological Innovation of Hubei Province (2018ABA101), Wuhan Science and Technology Project (No. 2019020701011480), and the Bijie Talent Team of Biological Protection and Ecological Restoration in Liuchong River Basin (No. 202112).

Data Availability

The raw reads were deposited in the NCBI Sequence Read Archive database (Accession Number: PRJNA762667 and PRJNA835968).

Declarations

The work described has not been submitted elsewhere for publication. All the authors listed have approved the manuscript that is enclosed. All applicable national and institutional guidelines for the care and use of animals were followed. We believe that the highlights of this manuscript will make it interesting to general readers of your journal.

Conflict of interest

The authors declare no competing interests.