Changes in Freeze-Thaw Environments in a Cold Lake: Eliciting New Insights into the Activity and Composition of Bacterial Communities

Abstract

This study explored the dynamics of bacterial community composition, response to environmental factors, and co-occurrence network models across different habitats of Wuliangsuhai Lake during the glacial period. Water quality analysis and high-throughput sequencing were performed at 14 monitoring sites within the lake. Additionally, a co-occurrence network between the two bacterial operational taxonomic unit (OTU)-OTUs was established. The results indicated significant differences in water quality indices, namely total nitrogen (TN), chemical oxygen demand (COD), total dissolved solids (TDS), salinity (SAL), chlorophyll-a (Chl.a), and electrical conductivity (EC), between the ice bodies of Wuliangsuhai Lake and subglacial water. Although there were no significant differences in α diversity across various media, substantial differences were observed in β diversity. The VIF and RDA analyses revealed that lake water quality factors significantly affected the microbial community structure and COD and TDS had the highest explanation for the community composition change. Network analysis demonstrated that competition dominated the bacterial community in water bodies with higher complexity and stability and ice body bacteria exhibited more reciprocal relationships and weaker resistance to external environmental disturbances. The co-occurrence network demonstrated a modular structure in the external environment, with g_Flavobacterium, f_Arcobacteraceae, and g_Sphingobacteriaceae being the main keystone species. Investigating the habitat heterogeneity of lake bacterial communities and identifying major groups and key species using molecular ecological network models and their topological effects can provide a theoretical foundation for monitoring and assessing the structural stability of lake ecosystems in cold regions.

1. Introduction

Lakes are crucial natural resources that exhibit sensitivity to alterations in their surrounding habitats [1], significantly enhancing the biodiversity of the planet and exhibiting more intricate ecological structures and functions [2]. Microorganisms, ubiquitous on Earth, play a vital role in their geochemical life cycle and their impact on environmental preservation, governance, and ecological stability is indispensable [3]. The characteristics of plankton include a short life cycle, susceptibility to change, and sensitivity to alterations in the physical, chemical, and other factors of water. Its structural attributes can reflect the resilience of the water ecosystem to pollution stress and are often regarded as indicators of water ecosystem quality.

To examine potential interactions among microbial groups, co-occurrence patterns of bacterial groups can be analyzed via network structures to address the intricate spatial or temporal structures of microorganisms [4]. Numerous microorganisms establish ecological communities through interspecific interactions, including mutualism and competition, and are intricately related to their surrounding environment [5]. Co-occurrence network analysis can identify potential keystone species and their interactions (antagonistic or cooperative), revealing core species within microbial communities [6]. Therefore, exploring core species and key taxa in molecular ecological networks, identifying niche overlap or separation, and elucidating the essential functions of microbiota across distinct habitats can further demonstrate the mechanisms underlying phenotypic variations among samples. Recent studies have focused on microbiome network analysis. For example, Wahdan et al. analyzed the molecular ecological network of bacteria and fungi in wheat straw, revealing greater stability under extreme conditions during the early decomposition stages, with rhizobia serving as the core species, facilitating the decomposition process [7]. Similarly, Zhang et al. evaluated community interactions based on bacterial symbiotic networks in wetland water samples and sediments, highlighting reciprocal cooperation as the main link between bacterial communities in wetland environments. Keystone species can enhance ecosystem stability, whereas variations exist among habitats and microorganisms at the principal connecting points within modules can serve as nitrogen sources for the growth of other microorganisms [8].

Wulangsuhai Lake, situated in Wulat Front Banner, Bayannur City, Inner Mongolia Autonomous Region, is the largest natural wetland at the same latitude in China and the largest freshwater lake in the Yellow River Basin of Inner Mongolia. In addition, it serves as a discharge site for domestic sewage, agricultural irrigation, and industrial wastewater from the Hetao Irrigation District. This lake is a typical artificial control structure in cold and arid regions and is important for mitigating pollution in the lower Yellow River, controlling soil and water erosion, and enhancing the basin environment. The presence of the 0 ℃ isotherm in Qinling makes it different from southern lakes, potentially leading to freezing, wherein nutrients and pollutants migrate beneath the ice, altering the bacterial habitat conditions. This forms a mutual feeding mechanism within the ice–water–bacteria ecological process, inducing water–ice body bacterial heterogeneity and restructuring the bacterial community based on microbial metabolism. Currently, research on ice-free lakes has explored bacterial structure and metabolism in high-altitude arid regions and most studies on ice-covered lakes have primarily examined bacterial diversity and function in subglacial water bodies and sediments. Studies on the ice-sealing period of Wuliangsuhai Lake have concentrated on the heat flux at the ice–water interface [9,10], primary productivity [11,12], and comprehensive pollution indices [13]. There have been a few reports on planktonic community composition and its response to environmental factors, with even fewer studies on microbial co-occurrence networks in diverse habitats during the ice period. Therefore, there is an urgent need to utilize Wuliangsuhai Lake as a research area, with a specific focus on microorganisms in various physical states of ice and water. This involves employing high-throughput sequencing technology and constructing molecular ecological network models to analyze bacterial community structures and co-occurrence patterns across various habitats. This study aimed to address the following scientific questions. (1) What are the variations in microbial composition across diverse habitat systems? (2) How do environmental factors influence the microbial community response? (3) What are the species niches and network complexities within the co-occurrence network of water–ice microorganisms? (4) Do core species exist within the co-occurrence network of water and ice microbes and, if present, what are the disparities?

2. Materials and Methods

2.1. Overview of the Study Area

Wuliangsuhai Lake (40°47′–41°03′ N, 108°43′–108°57′ E) (Figure 1) is a river-trace lake formed by diversion of the Yellow River. The primary source of replenishment is farmland runoff and sewage wastewater originating from the extensive agricultural irrigation area of Hetao upstream of the drainage channel, with the effluent ultimately flowing into the Yellow River. Situated in the temperate continental monsoon climate zone, it experiences an annual average temperature of 6.6 ℃ and an average rainfall of 224 mm. The lake typically freezes in late November each year and begins thawing from the end of March to the beginning of April the following year, with an ice period spanning 4–5 months. It stretches longitudinally from north to south and is narrower from east to west, measuring 35–40 km in length and 5–10 km in width. The average water depth is less than 2.5 m, covering an area of 325.31 km².

2.2. Sample Collection

Based on lake topography, hydrology, and ice period, seven sampling points were evenly distributed across the lake area to collect water and ice samples. The sampling locations were categorized into three sections: lake inlet (I12, J11), middle (M12, P11, P9), and outlet (S6, V3). Ice samples (HB) and water samples (BS) were collected in January 2022. At each sampling point, icicles were cut into three layers and combined to form a single ice sample, which was then placed in a 3 L pre-sterilized plastic bottle. Another water sampler retrieved a 3 L water sample from beneath the ice, with both water and ice sampling procedures repeated three times at each designated point. The water samples and ice meltwater were individually divided into two equal portions, which were then separately filtered using 0.45 μm and 0.20 μm filter membranes. The former filtrate was utilized for physicochemical factor analysis, while the latter, containing concentrated biological material, was placed in sterilized centrifuge tubes using high-temperature sterilized tweezers and stored at an ultra-low temperature of −80 ℃. Subsequently, the total DNA was extracted. To minimize the error, the physical and chemical index test results represent the average of three measurements, with an error margin of less than 5%.

2.3. Analysis of Sample Environmental Factors

The pH, EC, oxidation–reduction potential (ORP), SAL, and TDS index values of each sample were measured using a YSI portable multi-parameter water quality analyzer. The TN, total phosphorus (TP), dissolved total phosphorus (DTP), dissolved inorganic phosphorus (DIP), ammonium nitrogen (NH₃-N), COD, and Chl.a contents were determined according to the recommended alkaline potassium persulfate digestion-ultraviolet spectrophotometry, molybdenum-antimony resistance spectrophotometry, phosphomolybdenum blue spectrophotometry, Nessler’s reagent spectrophotometry, permanganate index method, and acetone extraction spectrophotometry procedures outlined in the Water and Wastewater Monitoring Analysis method (fourth edition) [14].

2.4. DNA Extraction and PCR Amplification

Environmental sample DNA was extracted using the Fast DNA™ Spin Kit for Soil (MP Biomedicals, California, USA), followed by confirmation of the presence of the extracted genomic DNA through 1% agarose gel electrophoresis and spectrophotometry to ensure adequate concentration. Amplification of the highly variable region of the 16S rRNA gene was conducted using primers 338F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). Randomly selecting representative samples for the pre-experiment ensured that products with suitable concentrations could be amplified to the experimental concentration by the majority of the samples within the minimum number of cycles. Subsequently, all the samples were tested under formal experimental conditions, with three replicates per sample. The PCR products of each sample were subjected to 2% agarose gel electrophoresis after mixing in equal volumes. PCR products were recovered using the AxyPrep DNA gel recovery kit and eluted with biological buffer Tris_HCl before further detection via 2% agarose electrophoresis. The PCR products were quantified using the QuantiFluor™-ST blue fluorescence quantification system (Promega), according to the initial electrophoresis quantification results. Subsequently, the amplicons were combined in equal molar quantities, based on the sequencing volume requirements of each sample. Finally, sequencing was performed using the Illumina MiSeq platform by Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). (Shanghai Majorbio Bio-pharm Technology Co., Ltd.) and the original bacterial sequencing data were deposited in the NCBI database under the registered sequence number PRJNA998176.

2.5. Bioinformatics Analysis and Data Processing

FASTP (v.0.19.6) was used for data pre-processing, quality control, and deduplication. FLASH (v.1.2.11) was employed for pair-end sequence concatenation with a confidence threshold of 0.7. The RDP classifier was applied for Bayesian computation using the Silva (R.138) database to classify the OTU at a 97% similarity level. Sequences were flattened based on the minimum sample sequence and chloroplast and mitochondrial sequences were removed for comparison. QIIME (v.1.9.1) was used to generate a taxonomic abundance table, which was subsequently subjected to statistical analysis.

Species composition analysis was conducted using R (v.3.3.1) and involved counting and mapping. Mouthur (v.1.30.2) facilitated α diversity analysis, with differences in richness and the diversity index between groups assessed via a non-parametric Mann–Whitney U test. Non-metric multidimensional scaling (NMDS) analysis was performed using the vegan software package in R (v. 3.3.1) and discussed the β diversity. The scipy package in Python was employed for Wilcoxon rank sum tests to validate inter-group differences in species variation analysis, with multiple FDR tests conducted and confidence intervals calculated via bootstrapping at a 95% confidence level. Analysis of similarities (ANOSIM), non-parametric multivariate analysis of variance (Adonis), and partial least squares discriminant analysis (PLS-DA), supported by the vegan and mixOmics packages, enabled linear discrimination of sample groups and identified key species variables differentiating between groups. Additionally, environmental factors with variance inflation factor (VIF) values exceeding 10 were filtered through multiple VIF analysis. RDA analysis and mapping, implemented using the vegan package, revealed the relationship between microorganisms and environmental factors.

Data analysis of the physical and chemical factors of the water body was conducted using SPSS (v.26.0) software. Before statistical analysis, the Schapiro–Wilke (S–W) test results were calculated to determine whether the sample distributions adhered to normality. Based on the normality results, a difference analysis was performed using paired sample t-tests and double paired sample t-tests with non-parametric tests. Significance was assessed using the Sig and asymptotic results from t-tests and Mann–Whitney U tests.

2.6. Microbial Co-Occurrence Network Architecture

Using Spearman’s rank correlation, a co-occurrence network was constructed using the Networkx package in Python. The network retained the top 100 species in total abundance at the OTU taxonomic level, with a correlation coefficients absolute value of ≥0.5 and p-value < 0.05 for the screened OTU data. Gephi (v.0.10.1) was used for data visualization. The average degree, diameter, graph density, modularity measure, average clustering coefficient, average path length, and centrality index were calculated to assess the network topology. Within-module connectivity (Zi) and among-module connectivity (Pi) were calculated to identify module hubs (Zi > 2.5; Pi ≤ 0.62), connectors (Zi ≤ 2.5; Pi > 0.62), network hubs (Zi > 2.5; Pi > 0.62), and peripheral nodes (Zi ≤ 2.5; Pi ≤ 0.6), where all nodes within the module hub area were designated as key species [15]. Module hubs, connectors, and network hubs represent pivotal nodes. OTUs were identified within the key node region and were characterized by high degree values (>5) and low betweenness centrality values (<1000), indicating significant co-occurrence relationships [16]. Through network and topological attribute analyses, relationships within or between sample groups were delineated, facilitating the comprehensive and efficient extraction of species information from complex data.

3. Results

3.1. Physical and Chemical Characteristics of Water and Ice Bodies

Principal component analysis (PCA) was conducted on the primary environmental factors of the water and ice bodies at each sampling point during the ice-sealing period in Wuliangsuhai Lake (Supplementary Material Figure S1). The analysis revealed environmental heterogeneity in the index values of water and ice bodies (PERMANOVA, p = 0.002, R² = 0.943). The comparison via an independent sample T-test highlighted significant differences in TN, EC, TDS, SAL, COD, and Chl.a concentrations between lake water and ice bodies, while other physical and chemical indices demonstrated no significant variations. Specifically, NH₃-N, TN, DTP, TDS, SAL, COD, and Chl.a concentrations in water were notably higher than those in ice, whereas TP, DIP, pH, EC, and ORP levels in water were lower than those in ice (Figure 2).

3.2. Bacterial Community Composition

The sequencing results revealed 351,568 gene sequences derived from 14 water and ice samples collected from seven stations. After categorization, 5289 OTUs were annotated, representing 61 phyla, 172 classes, 414 orders, 697 families, 1360 genera, and 2495 species. Species with a relative abundance exceeding 1% were deemed dominant, whereas those below this threshold were classified as others. The dominant phyla within the two groups were 10 and 11, respectively, with the common dominant phyla being Proteobacteria, Actinobacteriota, Bacteroidota, Firmicutes, Campilobacterota, Chloroflexi, Verrucomicrobiota, and Patescibacteria. Additionally, the dominant bacteria in the water were Acidobacteriota and Desulfobacterota, whereas those prevalent in ice bodies included Cyanobacteria, Planctomycetota, and Gemmatimonadota (Figure 3).

At the class classification level, both groups contained 18 and 15 dominant bacteria. The detected common dominant bacteria included Gammaproteobacteria, Bacteroidia, Actinobacteria, Alphaproteobacteria, Campylobacteria, Clostridia, Verrucomicrobiae, Acidimicrobiia, Anaerolineae, and Bacilli. The distinct bacteria found in the water were Saccharimonadia, Gracilibacteria, Desulfobacteria, Parcubacteria, Omnitrophia, Vicinamibacteria, Desulfitobacteriia, and KD4-96. In the ice body, unique bacteria included Longimicrobia, unclassified_p_Proteobacteria, Chloroflexia, Thermoleophilia, and Cyanobacteria (Supplementary Material Figure S2).

At the order classification level, the analysis revealed 37 and 31 dominant bacteria within the two groups. Among the top 10 orders, the shared dominant bacteria included Burkholderiales, Micrococcales, Rhodobacterales, Flavobacteriales, Frankiales, Corynebacteriales, Campylobacterales, and Sphingobacteriales. Unique to the water samples were Chitinophagales and Pem15, whereas Sphingomonadales and Rhizobiales were specific to the ice samples (Supplementary Material Figure S3).

At the family classification level, both groups exhibited dominant bacterial counts of 45 and 53, respectively. Among the top 10 families, both groups shared dominant members, which encompassed Flavobacteriaceae, Comamonadaceae, Arcobacteraceae, Burkholderiaceae, Microbacteriaceae, Rhodobacteraceae, Sporichthyaceae, and Sphingobacteriaceae. Notably, families exclusive to water bodies comprised Alcaligenaceae and Mycobacteriaceae, whereas those endemic to ice bodies included Sphingomonadaceae and Nocardiaceae (Supplementary Material Figure S4).

The water bacterial community exhibited 51 dominant genera, with the top 10 in high relative abundance being Rhodoluna, Rhodoferax, Pseudorhodobacter, Polynucleobacter, unclassified_f_Rhodobacteraceae, Flavobacterium, Limnohabitans, Mycobacterium, Pedobacter, and unclassified_f_Alcaligenaceae. In contrast, the ice body bacterial community comprised 64 dominant genera, with the top 10 in high relative abundance, including Flavobacterium, Rhodoferax, norank_f_Arcobacteraceae, Cupriavidus, Polaromonas, Pedobacter, Pseudarcobacter, Rhodoluna, Rhodococcus, and Limnohabitans (Figure 4).

A Wilcoxon test was performed to analyze the relative abundance of dominant groups in the water and ice bacterial communities at both phylum and genus levels. The results indicated significant differences between the two communities in the relative abundance of 4 out of the 11 dominant phyla. Specifically, Bacteroidota, Campilobacterota, and Gemmatimonadota exhibited higher relative abundances in the ice bacterial communities, whereas Patescibacteria showed higher prevalence in the water bacterial community (Figure 5a). At the genus level, significant differences in relative abundance were observed among the 15 dominant genera out of the 64 dominant bacterial genera, distinguishing the water and ice bacterial communities. Notably, the relative abundances of seven dominant genera in water bacterial communities exceeded those in ice bacterial communities, including Polynucleobacter, Rhodoluna, Pseudorhodobacter, unclassified_f_Alcaligenaceae, Legionella, norank_f_norank_o_PeM15, and norank_f_norank_o_Candidatus_Kaiserbacteria. Conversely, the relative abundance of eight dominant bacterial genera was higher in the ice body bacterial community than the water bacterial community, including Flavobacterium, Pseudarcobacter, Sphingomonas, Pseudomonas, norank_f_T34, Massilia, Planococcus, and Aquaspirillum_arcticum_group (Figure 5b).

3.3. Structural Diversity of Bacteria

First, α diversity was computed for multiple points, followed by a comparison between water and ice bodies using the Wilcoxon rank-sum test. The Shannon index of the bacterial community (water = 4.0734, ice = 4.6807), Simpson diversity index (water = 0.03624, ice = 0.056585), ACE abundance index (water = 1908.3, ice = 2089.2), and Chao abundance index (water = 1649, ice = 1725.9) were not significantly different (p > 0.05; Figure 6a–d).

NMDS analysis was performed using planktonic bacterial community data from various habitats within Wuliangsuhai Lake to explore spatial heterogeneity. The NMDS results revealed significant alterations in the β diversity of bacterial communities between water and ice bodies (p = 0.001; Figure 6e). The bacterial communities in water and ice bodies exhibited aggregation characteristics. Conversely, these communities demonstrated distinct separations from each other. This suggests that the composition of bacterial communities differed between water bodies and ice bodies during the ice period. Furthermore, the differences between the water and ice bodies surpassed the variance within each group, with the divergence within the ice body being greater than that within the water body. This observation was supported by the PLS-DA (Figure 6f). In the Adonis analysis, the Bray-Curtis semi-metric was employed to partition the total variance. Physicochemical factors such as COD, SAL, TDS, and EC exhibited greater explanatory power for the distinctions among samples from different habitats. Concurrently, variations in grouping also contributed to the differences observed between the samples (p < 0.05; Supplementary Material Table S1).

3.4. Environmental Drivers of Bacterial Community Structure

The significance of environmental factors on the bacterial community was analyzed using planktonic bacterial community data. Following the variance inflation factor (VIF) expansion analysis, the driving factors (TN and SAL) with VIF values exceeding 10 were excluded. Furthermore, RDA axes 1 and 2 had interpretation rates of 25.99% and 16.74%, respectively. Collectively, physicochemical indices accounted for 42.73% of the total interpretation of community composition (Figure 7). The water and ice bodies exhibited dispersion, with discernible spatial differences. Notably, TDS and COD emerged as significant factors affecting changes in the bacterial community structure, with the hierarchy of environmental factors affecting different habitats being COD > TDS > EC > Chl. a > ORP > TP > NH₃-N > DTP > pH > DIP.

3.5. Co-Occurrence Patterns of Bacteria

Using the top 100 species with total abundance at the OTU classification level, a co-occurrence network map of bacterial communities in water and ice bodies was established to investigate microorganism interrelationships across different habitats (Figure 8). Distinct nodes represent different OTUs, whereas varied colors denote affiliations with different phyla. Node size was positively correlated with the degree of connectivity. The line color between nodes signifies their relationship: purple indicates a positive correlation and blue indicates a negative correlation. Edge weight represents the absolute value of the correlation coefficient, thicker lines denote stronger correlations between species, and multiple lines indicate closer connections between species. The water bacterial community comprised 99 nodes and 364 edges, whereas the ice body contained 98 nodes and 521 edges, with positive edges accounting for 56.59% and 72.74%, respectively (Supplementary Material Table S2).

Comparatively, the bacterial community of the ice body exhibited a higher average degree, graph density, and clustering coefficient than those of the water body. However, the bacterial community of the water body displayed a higher degree of modularity and average path length than those of the ice body. Based on topological properties, most of the nodes in both water and ice bodies were classified as peripheral nodes, with a significantly higher proportion in ice bodies (water = 54.55%, ice = 69.39%). Moreover, the number of bacterial communities in the connector area was higher in water (45.45%) than in ice (30.61%) (Supplementary Material Figure S5). Bacteroides (water = 22.22%, ice = 40%), Proteobacteria (water = 26.67%, ice = 43.33%), and Actinobacteria (water = 26.67%, ice = 43.33%) were the three most abundant phyla in connectors. Among the 23 key OTUs in the water bacterial network, 6 belonged to Bacteroidota, 6 belonged to Actinobacteriota, 4 belonged to Campilobacterota, 2 belonged to Firmicutes, and 5 belonged to Proteobacteria. In the ice bacterial network, 2 belonged to Actinobacteriota, 5 to Bacteroidota, 1 to Campilobacterota, 1 to Cyanobacteria, and 9 to Proteobacteria. OTU1290 emerged as a common keystone species, whereas OTU12 and OTU2866 exhibited the lowest abundances (Supplementary Material Table S3).

4. Discussion

4.1. Habitat Heterogeneity

This study revealed that physical and chemical indices such as TN, NH3-N, COD, TDS, and SAL were higher in water bodies than in ice bodies, which is consistent with findings from both domestic and foreign lake studies [17,18,19]. This observation can be attributed to the impact of temperature on the solubility of surface water pollutants during the ice-sealing period, which leads to an accelerated precipitation rate of pollutants as the water temperature decreases. Additionally, a concentration disparity between water and pollutant molecules at the water–ice interface facilitated the transition of some pollutants from the solid to the liquid phase, resulting in a concentration effect and a subsequent decline in concentration [20]. Simultaneously, during the growth stage of the ice sheet [21], nutrients were redistributed, leading to a rapid increase in nutrient concentration in the subglacial water, while nutrient consumption decreased during the ice cover period. Consequently, there was an elevation in the concentrations of nutrients and organic matter, facilitating the metabolic growth of phytoplankton and algae and resulting in high levels of COD and Chl.a in the water.

4.2. Distribution and Composition of Bacterial Communities

At the phylum level, the main bacterial groups constituting the community structure within the water and ice bodies of Wuliangsuhai Lake were Proteobacteria, Actinobacteriota, and Bacteroidota. These findings align with both domestic and international research on river and lake ecosystems during glacial periods, which consistently identified these groups as the dominant bacterial constituents [22,23,24,25]. Patescibacteria exhibited a high abundance in the bacterial community of the water body owing to a decrease in the DO concentration, which remained at extremely low levels during the extended ice period in Wuliangsuhai Lake. Notably, Patescibacteria can be predominantly detected in anoxic or low-oxygen environments [26].

At the class level, Gammaproteobacteria, Bacteroidia, Actinobacteria, and Alphaproteobacteria were predominant in both water and ice bodies, consistent with the findings of Dong et al. regarding the Bailang River Estuary in winter [27]. Saccharimonadia, a member of the phylum Patescibacteria, demonstrated high specificity. Its presence can be attributed to the higher pollution intensity in the water body than in the ice body. Biochemical pollutants in water, often in the dissolved form, facilitate the prevalence of Saccharimonadia, which is known to play a significant role in sewage bacteria and the degradation of sugar compounds [28,29]. Moreover, water has a higher concentration of ammonia nitrogen than ice. Saccharimonadia is associated with ammonia nitrogen removal [30], resulting in a higher concentration of Saccharimonadia in water.

At the order level, Burkholderiales was the first bacterial group in both water and ice bodies. This group is rich in catabolic enzymes capable of degrading refractory pollutants under diverse environmental conditions and is associated with the degradation of microplastics and aromatic compounds [31]. Furthermore, the surface water of Wuliangsuhai Lake harbors a significant quantity of microplastics. During the ice period, the slow growth of ice allows particles in water to passively merge, resulting in a lower microplastic density, making it easier for them to become embedded in the ice body [32]. This phenomenon may also explain the changes in abundance.

At the family level, Comamonadaceae emerges as a dominant and subdominant bacterial family in both water and ice and is known for its resilience in cold environments. Extensive research has highlighted its abundance and prevalence in freshwater ecosystems, establishing it as a key bacterial group [33,34]. After the cessation of autumn irrigation in the Hetao Irrigation District, significant quantities of microplastics, primarily composed of polyethylene, were transported into Wuliangsuhai Lake via agricultural drainage channels [35]. As the surface of the lake froze, these particles became embedded in the ice sheet and Comamonadaceae was identified as the primary bacteria associated with microplastic degradation [36,37]. Additionally, differences in abundance are attributed to variations in nutrient salts during the ice period, particularly nitrate and ammonia nitrogen from agricultural wastewater, which serve as key drivers of microbial growth [38]. Comamonadaceae, which functions as denitrifying bacteria, efficiently reduces N₂O and nitrite to nitrogen via anaerobic processes, making it an effective agent for denitrification [39].

Saline-tolerant Geodermatophilaceae, discovered in desert environments [40], exhibit robust resilience to low-nutrient conditions [41]. The Hetao Irrigation area, characterized by severe soil salinization in China, encompasses Wuliangsuhai Lake, a pivotal water concentration zone within the irrigation region. As freezing occurs, nutrient elements, such as nitrogen and phosphorus, concentrated in subglacial water deplete the nutrient levels in the ice body. This phenomenon explains the presence of Geodermatophilaceae in various habitats during the glaciation. This notable difference in abundance arises from fine sand particles originating from nearby deserts accumulating on the ice sheet of Wuliangsuhai Lake. Solar radiation melts sediments and snow on the ice surface, generating liquid water during the day that refreezes at night. Furthermore, previous studies have indicated the presence of Geodermatophilaceae in snow dust [42].

At the genus level, Rhodoluna and Flavobacterium emerged as the most prevalent bacterial genera in both the water and ice bodies. Rhodoluna, which belongs to the family Microbacteriaceae within the phylum Actinobacteriota, is commonly found in freshwater lakes [43]. Flavobacterium can hydrolyze various organic compounds and thrives in shallow freshwater lakes with oligotrophic and eutrophic nutrients in cold regions, dominating bacterial populations in ice bodies [44,45]. The specificity of Rhodoluna can be attributed to its dominance in saline environments [46]. Additionally, the abundance of Actinobacteriota and Microbacteriaceae in the water body of Wuliangsuhai Lake surpasses that in the ice body (Figure 3; Supplementary Material Figure S4) and the water body exhibits significantly higher salinity than the ice body (Figure 2), potentially accounting for this disparity.

4.3. Diversity of Bacterial Communities and Responses to Environmental Factors

When investigating microbial diversity in the environment, the richness and diversity of the microbial community can be reflected through single-sample diversity analysis. In this study, the Chao, Ace, Shannon, and Simpson indices were adopted to reflect community richness and diversity. No significant differences were observed in the richness and diversity of planktonic bacterial communities across the different habitats of Wuliangsuhai Lake during the freeze-up period (Figure 6a–d). The higher richness of the ice body aligned with the findings of Li et al. [47]. However, the diversity variance may result from the distinct hydrometeorological characteristics of Dali Lake, including recharge sources, drainage methods, and the low temperature of the ice body. These factors expand the niche availability of cold-tolerant bacteria.

The NMDS analysis and PLS-DA results indicated a higher similarity in the community composition between water and ice. However, due to varying water quality and nutrient characteristics across different areas of Wuliangsuhai Lake, significant spatial heterogeneity was observed at sample sites J11 (drained inlet), P11 (middle of the lake), and I12 (algal lake district). Moreover, the microflora composition exhibited notable variability under diverse environmental conditions (Figure 6f). To explore the differences between water and ice bodies, ANOSIM analysis of inter-group disparities yielded a P-value of 0.007, confirming that differences in bacterial community composition among different habitat types surpassed those within the same group. This validated the meaningfulness of such grouping. Simultaneously, the formation of the ice sheet had a concentration effect on nutrient elements, resulting in a “downward shift” of nutrient salts and elevating nutrient concentrations in subglacial water [48]. This phenomenon can also affect community diversity in various habitats. Given the substantial environmental variations among sampling points in Wuliangsuhai Lake, the response of flora in water and ice to environmental factors can be expected to differ (Figure 7). Previous studies have demonstrated that TN, TP, pH, DO, NH₃-N, DTP, DIP, ORP, COD, and Chl.a levels during the ice period are driving factors influencing microbial community distribution in lakes and rivers [49,50], which is consistent with our findings. However, the RDA results indicated that COD and TDS were significant factors affecting spatial differences. The COD concentration was positively correlated with the organic matter pollution level in water, serving as a measure of water organic content and providing ample nutrition for bacterial growth. A notable correlation exists between COD concentration and bacterial composition diversity in urban lakes and rivers entering lakes [51,52]. These water bodies receive significant pollutant loads from industrial and human activities. The similarity between the findings of the study on Wuliangsuhai Lake and urban lakes may be caused by Wuliangsuhai Lake being the sole area in the Hetao Irrigation District engaged in agricultural and industrial production, domestic sewage discharge, and various pollutant discharges. Furthermore, TDS serves as a salinization index in lakes, significantly influencing the microbial community in Wuliangsuhai Lake during the ice period. Research has indicated that elevated TDS concentrations can restrict microbial metabolic activity, increase water osmotic pressure, and induce plasmolysis and cell inactivation, consequently reducing biodiversity. Microbial secretion of extracellular polymers responds to this, further inhibiting microbial activity [53]. Generally, nitrogen concentration can directly affect the composition of planktonic bacteria, providing crucial nutrients and energy for microbial life. Despite more than 80% of the total nitrogen in Wuliangsuhai Lake originating from ammonia nitrogen [54], the results of this study indicate that ammonia nitrogen had no significant effect on lake bacterial community changes. The explanatory power of ammonia nitrogen for the bacterial community was 0.0653% in water and −5.0327% in ice. Such discrepancies could be due to an ecological water supply project in Wuliangsuhai Lake, which can alter pollution sources from eutrophic and unknown (hydrating factors) to saline and unknown (hydrating factors), with the most significant improvement in the NH₃-N content.

4.4. Co-Occurrence Networks of Bacterial Communities

The biological co-occurrence network serves as a reflection of potential interactions between species and the foundational community structure, with positive and negative connections indicating interactive relationships [55]. Bacterial networks in various habitats are primarily characterized by cooperative relationships, with water bodies exhibiting a higher proportion of negative correlation links among bacterial communities (Supplementary Material Table S2), suggesting more competition among bacterial populations in water bodies. The intensification of competition enhances network stability [56]. Consequently, the bacterial community structure in water exhibited greater stability, potentially reflecting greater ecosystem stability. Short paths facilitate the rapid transmission of local disturbances throughout the network, thereby altering its structure and function [56]. The ice bacterial network exhibited a shorter average path and stronger reciprocity, indicating a greater susceptibility to external environmental interference and reduced community stability. Furthermore, the bacterial networks in water bodies demonstrated high average clustering coefficients, average paths, and modularity, indicative of a more complex network structure and intensified population interactions. Studies have indicated a positive correlation between network complexity and stability [57], aligning with the greater stability observed in bacterial networks within water bodies. Both the water and ice networks exhibit modularity values exceeding 0.4, confirming their modular structures. Connectors play vital roles between and within modules, with a higher abundance observed in water networks than in ice, suggesting greater connectivity between bacterial modules in water. Moreover, the bacterial community network within ice may become increasingly fragile because key species maintain their structural integrity and possess unique roles within the microbial community. The removal of these species results in significant alterations in microbiome function and structure [58], leading to substantial changes in the network structure. This disruption may break the co-evolutionary relationships between species, rendering the network increasingly fragile and prone to collapse [59]. Co-occurrence networks revealed relationships between individual populations and ecosystems, with key species emerging simultaneously, implying potential cooperation between core bacterial groups in water and ice bodies to adapt to environmental changes. Among these key species, OTU5115, OTU4829, OTU1249, and OTU5716 belonged to the genus Flavobacterium, OTU1290 belonged to Arcobacteraceae, and OTU3946 belonged to Sphingorhabdus. Studies have indicated that Flavobacterium exhibits a pronounced preference for cold environments, thriving at 4 °C, and is commonly found in polar lakes, as well as cold environments such as streams and rivers. Flavobacterium produces a variety of cold-active enzymes that facilitate the dissolution of algal bacteria and degradation of biological macromolecules [60]. Arcobacteraceae have been linked to human excrement in sewage [61] and have also been detected in sewage sediments [62]. Sphingorhabdus has been identified as a core microorganism in microplastic environments [63]. The climatic conditions and drainage patterns of Wuliangsuhai Lake during the ice period may provide an environmental niche for these OTUs to contribute to water purification in bacterial communities across various habitats or serve as indicators for monitoring and diagnosing water quality. Furthermore, low-abundance OTUs were identified as keystone species, indicating that the potential for screened microorganisms to become keystone species and alter community composition is not solely dependent on their spatial abundance; dominant species do not necessarily equate to keystone species.

In this study, OTU analysis was conducted by selecting sequences with over 97% similarity to representative sequences, resulting in OTU tables. However, the sensitivity of the OTU approach to the similarity threshold might mask sequences with sequencing errors, leading to inaccurate abundance estimations for certain OTUs. Additionally, a loose similarity threshold may obscure genuine sequence variation. To address these issues, future research could employ more precise clustering algorithms such as DADA2 or unoise2. Furthermore, the limited number of sampling points in this study may affect the statistical significance of indicators. To enhance scientific rigor, future studies should incorporate samples from different habitat media using horizontal or vertical stratification. Moreover, while this study primarily focused on the geographical distribution pattern of microbial communities, it lacks both quantitative and qualitative discussions on the evolutionary mechanisms governing these communities. Future investigations should delve into the spatial heterogeneity of microbial community assembly mechanisms, which is a central topic in microbial ecology.

5. Conclusions

(1) The dominant bacterial compositions in both water and ice bodies were similar, albeit with significantly different abundances. The main bacterial phyla in both environments were Proteobacteria, Actinobacteriota, and Bacteroidota, with the common dominant classes being Gammaproteobacteria, Bacteroidia, Actinobacteria, and Alphaproteobacteria. Burkholderiales were predominantly abundant in both water and ice bodies, with Comamonadaceae being the dominant and subdominant family. Notably, Rhodoluna and Flavobacterium were the most dominant bacterial genera in various habitats. Significant differences were observed among Patescibacteria, Saccharimonadia, Geodermatophilaceae, and Rhodoluna at different taxonomic levels;

(2) There were no significant differences in the diversity and abundance indices between the water and ice bodies and bacterial diversity and abundance were lower in the water body than in the ice body. The variation among bacterial samples in the ice bodies exceeded that in the water bodies and the disparity between the two outweighed the differences within each group. COD and TDS emerged as the primary environmental parameters influencing the composition and structure of bacterial communities in water and ice environments;

(3) The water bacterial network exhibited numerous negative correlation connections, indicating intense interspecies competition and contributing to its complexity and stability. Keystone species were observed across various bacterial co-occurrence networks and potentially cooperated to address external environmental challenges. Flavobacterium, Arcobacteraceae, and Sphingorhabdus emerged as the key species in the co-occurrence network, likely influencing water quality and serving as biological indicators for monitoring and assessing water ecosystems.

Author Contributions

C.F.: writing–original draft, methodology, validation, formal analysis, writing–review and editing, and data curation. J.L.: funding acquisition and methodology. Y.J.: supervision and funding acquisition. Z.T.: test and methods analysis. Z.Z.: resources and writing—review and editing. Y.H.: sample collection and test. Y.L.: sample collection and test. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material and further inquiries can be directed to the corresponding authors.

Acknowledgments

We extend our sincere gratitude to the development team behind R Software, whose tools played a pivotal role in our data analysis endeavors. Additionally, we express our appreciation to the R User Support team for their invaluable assistance in resolving the specific technical issues encountered along the way.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures

Figure 1. Distribution of sampling stations in Wuliangsuhai Lake. (a) Geographical location of Wuliangsuhai Lake in China and distribution diagram of sampling points in the lake. (b) Photographs of the sampling process and collection of water and ice samples.

Figure 2. Comparison of basic physicochemical indices of ice−water media in lakes (W: water; I: ice body). Comparisons between the two groups were performed using Student’s t-test. The value on the horizontal line in the figure is the p−value and ns represents no difference.

Figure 3. Analysis of the dominant bacterial community composition in lake water and ice bodies. (a) Dominant bacteria in water at the phylum level. (b) Dominant bacteria in ice bodies at the phylum level.

Figure 4. Composition of dominant bacteria in water and ice bodies. (a) Top 20 dominant bacterial genera in water bodies. (b) Top 20 dominant bacterial genera in ice bodies.

Figure 5. Differences between groups of dominant bacteria in the lake water and ice bodies. (a) Differences in the dominant bacterial phyla. (b) Differences in the dominant bacterial genera. Two comparisons were made using the Wilcoxon test: * p < 0.05 and ** p < 0.01.

Figure 6. Comparative analysis of bacterial community diversity in lake water and ice bodies: (a) Shannon diversity index, (b) Simpson diversity index, (c) ACE diversity index, (d) Chao diversity index, (e) NMDS analysis of water and ice bodies, and (f) PLS−DA analysis.

Figure 7. RDA analysis of bacterial community composition and environmental factors.

Figure 8. Co-occurrence network of bacteria in the water and ice bodies. (a) Interspecific interactions among water bacteria. (b) Interspecific interactions among ice bacteria.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d16060311/s1, Figure S1: Principal component analysis of physicochemical factors of Wuliangsuhai Lake during the glacial period; Figure S2: Analysis of the dominant bacterial community composition in lake water and ice bodies. (a) Dominant bacteria in water at the class level. (b) Dominant bacteria in ice body at the class level; Figure S3: Analysis of the dominant bacterial community composition in lake water and ice bodies. (a) Dominant bacteria in water at the order level. (b) Dominant bacteria in ice body at the order level; Figure S4: Analysis of the dominant bacterial community composition in lake water and ice bodies. (a) Dominant bacteria in water at the family level. (b) Dominant bacteria in ice body at the family level; Figure S5: Topological effect of co-occurrence network of bacteria in lake water and ice bodies; Table S1: Results of permutation multivariate analysis of variance for physicochemical factors with grouping; Table S2: Network statistical indices; Table S3: Key OTU information of bacteria networks in lake water and ice bodies.

References

1. Brönmark, C.; Hansson, L.A. Environmental issues in lakes and ponds: Current state and perspectives. Environ. Conserv.; 2002; 29, pp. 290-307. [DOI: https://dx.doi.org/10.1017/S0376892902000218]

2. Davidson, T.A.; Jeppesen, E. The role of palaeolimnology in assessing eutrophication and its impact on lakes. J. Paleolimnol.; 2013; 49, pp. 391-410. [DOI: https://dx.doi.org/10.1007/s10933-012-9651-0]

3. Wang, Y.; Bao, G. Diversity of prokaryotic microorganisms in alkaline saline soil of the Qarhan Salt Lake area in the Qinghai–Tibet Plateau. Sci. Rep.; 2022; 12, 3365. [DOI: https://dx.doi.org/10.1038/s41598-022-07311-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35233041]

4. Barberán, A.; Bates, S.T.; Casamayor, E.O.; Fierer, N. Using network analysis to explore co-occurrence patterns in soil microbial communities. ISME J.; 2011; 6, pp. 343-351. [DOI: https://dx.doi.org/10.1038/ismej.2011.119] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21900968]

5. Hirano, H.; Takemoto, K. Difficulty in inferring microbial community structure based on co-occurrence network approaches. BMC Bioinform.; 2019; 20, 329. [DOI: https://dx.doi.org/10.1186/s12859-019-2915-1] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31195956]

6. Berry, D.; Widder, S. Deciphering microbial interactions and detecting keystone species with co-occurrence networks. Front. Microbiol.; 2014; 5, 219. [DOI: https://dx.doi.org/10.3389/fmicb.2014.00219] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24904535]

7. Wahdan, S.F.M.; Ji, L.; Schädler, M.; Wu, Y.-T.; Sansupa, C.; Tanunchai, B.; Buscot, F.; Purahong, W. Future climate conditions accelerate wheat straw decomposition alongside altered microbial community composition, assembly patterns, and interaction networks. ISME J.; 2022; 17, pp. 238-251. [DOI: https://dx.doi.org/10.1038/s41396-022-01336-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36352255]

8. Zhang, L.; Hong, W.; Pan, Z.; Fang, W.; Shen, Z.; Cai, H. Wastewater treatment effectiveness is facilitated by crucial bacterial communities in the wetland ecosystem. Sci. Total Environ.; 2023; 857, 159375. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.159375] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36240933]

9. Lu, P.; Cao, X.; Li, G.; Huang, W.; Leppäranta, M.; Arvola, L.; Huotari, J.; Li, Z. Mass and Heat Balance of a Lake Ice Cover in the Central Asian Arid Climate Zone. Water; 2020; 12, 2888. [DOI: https://dx.doi.org/10.3390/w12102888]

10. Huo, J.; Shi, X.; Zhao, S.; Sun, B.; Li, G.; Yu, H.; Wang, S. Dynamics of water temperature and interfacial heat flux in lake ice and under-ice layers. Hydrol. Curr. Res.; 2023; 54, pp. 855-868. [DOI: https://dx.doi.org/10.2166/nh.2023.013]

11. Shi, X.; Yu, H.; Zhao, S.; Sun, B.; Liu, Y.; Huo, J.; Wang, S.; Wang, J.; Wu, Y.; Wang, Y. et al. Impacts of environmental factors on Chlorophyll-a in lakes in cold and arid regions: A 10-year study of Wuliangsuhai Lake, China. Ecol. Indic.; 2023; 148, 110133. [DOI: https://dx.doi.org/10.1016/j.ecolind.2023.110133]

12. Yu, H.; Shi, X.; Zhao, S.; Sun, B.; Liu, Y.; Arvola, L.; Li, G.; Wang, Y.; Pan, X.; Wu, R. et al. Primary productivity of phytoplankton and its influencing factors in cold and arid regions: A case study of Wuliangsuhai Lake, China. Ecol. Indic.; 2022; 144, 109545. [DOI: https://dx.doi.org/10.1016/j.ecolind.2022.109545]

13. Shi, R.; Zhao, J.; Shi, W.; Song, S.; Wang, C. Comprehensive Assessment of Water Quality and Pollution Source Apportionment in Wuliangsuhai Lake, Inner Mongolia, China. Int. J. Environ. Res. Public Health; 2020; 17, 5054. [DOI: https://dx.doi.org/10.3390/ijerph17145054] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32674384]

14. The Editorial Committee of ‘Water and Wastewater Monitoring and Analysis Method’ of the State Environmental Protection 323 Administration. Water and Wastewater Monitoring and Analysis Method; 4th ed. China Environmental Science Press: Beijing, China, 2022.

15. Zhong, L.; Wu, T.; Ding, J.; Xu, W.; Yuan, F.; Liu, B.F.; Zhao, L.; Li, Y.; Ren, N.Q.; Yang, S.S. Co-composting of faecal sludge and carbon-rich wastes in the earthworm’s synergistic cooperation system: Performance, global warming potential and key microbiome. Sci. Total Environ.; 2023; 857, 159311. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.159311] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36216047]

16. Li, Y.; Wu, H.; Shen, Y.; Wang, C.; Wang, P.; Zhang, W.; Gao, Y.; Niu, L. Statistical determination of crucial taxa indicative of pollution gradients in sediments of Lake Taihu, China. Environ. Pollut.; 2019; 246, pp. 753-762. [DOI: https://dx.doi.org/10.1016/j.envpol.2018.12.087] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30623831]

17. Wu, R.; Liu, Y.; Zhang, S.; Shi, X.; Zhao, S.; Lu, J.; Kang, X.; Wang, S.; Wu, Y.; Arvola, L. Characterization of nitrogen and phosphorus at the ice-water-sediment interface and the effect of their migration on overlying water quality in Daihai Lake (China) during the freezing period. Sci. Total Environ.; 2023; 893, 164863. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.164863] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37321494]

18. Pieters, R.; Lawrence, G.A. Effect of salt exclusion from lake ice on seasonal circulation. Limnol. Oceanogr.; 2009; 54, pp. 401-412. [DOI: https://dx.doi.org/10.4319/lo.2009.54.2.0401]

19. Wu, R.; Zhang, S.; Liu, Y.; Shi, X.; Zhao, S.; Kang, X.; Quan, D.; Sun, B.; Arvola, L.; Li, G. Spatiotemporal variation in water quality and identification and quantification of areas sensitive to water quality in Hulun lake, China. Ecol. Indic.; 2023; 149, 110176. [DOI: https://dx.doi.org/10.1016/j.ecolind.2023.110176]

20. Rokkan Iversen, K.; Primicerio, R.; Larsen, A.; Egge, J.K.; Peters, F.; Guadayol, O.; Jacobsen, A.; Havskum, H.; Marrase, C. Effects of small-scale turbulence on lower trophic levels under different nutrient conditions. J. Plankton Res.; 2009; 32, pp. 197-208. [DOI: https://dx.doi.org/10.1093/plankt/fbp113]

21. Yang, F.; Cen, R.; Feng, W.; Zhu, Q.; Leppäranta, M.; Yang, Y.; Wang, X.; Liao, H. Dynamic simulation of nutrient distribution in lakes during ice cover growth and ablation. Chemosphere; 2021; 281, 130781. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2021.130781]

22. Zhong, M.; Capo, E.; Zhang, H.; Hu, H.; Wang, Z.; Tian, W.; Huang, T.; Bertilsson, S. Homogenisation of water and sediment bacterial communities in a shallow lake (lake Balihe, China). Freshw. Biol.; 2022; 68, pp. 155-171. [DOI: https://dx.doi.org/10.1111/fwb.14016]

23. Bowman, J.S.; Rasmussen, S.; Blom, N.; Deming, J.W.; Rysgaard, S.; Sicheritz-Ponten, T. Microbial community structure of Arctic multiyear sea ice and surface seawater by 454 sequencing of the 16S RNA gene. ISME J.; 2011; 6, pp. 11-20. [DOI: https://dx.doi.org/10.1038/ismej.2011.76] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21716307]

24. Cai, M.; Wang, B.C.; Han, J.B.; Yang, J.; Zhang, X.Y.; Guan, X.Y.; Jiang, H.C. Microbial difference and its influencing factors in ice-covered lakes on the three poles. Environ. Res.; 2024; 252 Pt 1, 118753. [DOI: https://dx.doi.org/10.1016/j.envres.2024.118753]

25. Sułowicz, S.; Bondarczuk, K.; Ignatiuk, D.; Jania, J.A.; Piotrowska-Seget, Z. Microbial communities from subglacial water of naled ice bodies in the forefield of Werenskioldbreen, Svalbard. Sci. Total Environ.; 2020; 723, 138025. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.138025] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32213417]

26. He, C.; Keren, R.; Whittaker, M.L.; Farag, I.F.; Doudna, J.A.; Cate, J.H.D.; Banfield, J.F. Genome-resolved metagenomics reveals site-specific diversity of episymbiotic CPR bacteria and DPANN archaea in groundwater ecosystems. Nat. Microbiol.; 2021; 6, pp. 354-365. [DOI: https://dx.doi.org/10.1038/s41564-020-00840-5] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33495623]

27. Dong, W.; Cui, Z.; Zhao, M.; Li, J. Seasonal and Spatial Variations of Bacterial Community Structure in the Bailang River Estuary. J. Mar. Sci. Eng.; 2023; 11, 825. [DOI: https://dx.doi.org/10.3390/jmse11040825]

28. Nielsen, P.H.; Kragelund, C.; Seviour, R.J.; Nielsen, J.L. Identity and ecophysiology of filamentous bacteria in activated sludge. FEMS Microbiol. Rev.; 2009; 33, pp. 969-998. [DOI: https://dx.doi.org/10.1111/j.1574-6976.2009.00186.x]

29. Kindaichi, T.; Yamaoka, S.; Uehara, R.; Ozaki, N.; Ohashi, A.; Albertsen, M.; Nielsen, P.H.; Nielsen, J.L.; Stams, A. Phylogenetic diversity and ecophysiology of Candidate phylum Saccharibacteria in activated sludge. FEMS Microbiol. Ecol.; 2016; 92, fiw078. [DOI: https://dx.doi.org/10.1093/femsec/fiw078] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27090759]

30. Remmas, N.; Melidis, P.; Zerva, I.; Kristoffersen, J.B.; Nikolaki, S.; Tsiamis, G.; Ntougias, S. Dominance of candidate Saccharibacteria in a membrane bioreactor treating medium age landfill leachate: Effects of organic load on microbial communities, hydrolytic potential and extracellular polymeric substances. Bioresour. Technol.; 2017; 238, pp. 48-56. [DOI: https://dx.doi.org/10.1016/j.biortech.2017.04.019]

31. Morya, R.; Salvachúa, D.; Thakur, I.S. Burkholderia: An Untapped but Promising Bacterial Genus for the Conversion of Aromatic Compounds. Trends Biotechnol.; 2020; 38, pp. 963-975. [DOI: https://dx.doi.org/10.1016/j.tibtech.2020.02.008]

32. Obbard, R.W.; Sadri, S.; Wong, Y.Q.; Khitun, A.A.; Baker, I.; Thompson, R.C. Global warming releases microplastic legacy frozen in Arctic Sea ice. Earth’s Future; 2014; 2, pp. 315-320. [DOI: https://dx.doi.org/10.1002/2014EF000240]

33. Moon, K.; Kang, I.; Kim, S.; Kim, S.J.; Cho, J.C. Genomic and ecological study of two distinctive freshwater bacteriophages infecting a Comamonadaceae bacterium. Sci. Rep.; 2018; 8, 7989. [DOI: https://dx.doi.org/10.1038/s41598-018-26363-y] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29789681]

34. Newton, R.J.; Jones, S.E.; Eiler, A.; McMahon, K.D.; Bertilsson, S. A Guide to the Natural History of Freshwater Lake Bacteria. Microbiol. Mol. Biol. Rev.; 2011; 75, pp. 14-49. [DOI: https://dx.doi.org/10.1128/MMBR.00028-10] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21372319]

35. Mao, R.; Hu, Y.; Zhang, S.; Wu, R.; Guo, X. Microplastics in the surface water of Wuliangsuhai Lake, northern China. Sci. Total Environ.; 2020; 723, 137820. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.137820]

36. Taipale, S.J.; Vesamäki, J.; Kautonen, P.; Kukkonen, J.V.K.; Biasi, C.; Nissinen, R.; Tiirola, M. Biodegradation of microplastic in freshwaters: A long-lasting process affected by the lake microbiome. Environ. Microbiol.; 2022; 25, pp. 2669-2680. [DOI: https://dx.doi.org/10.1111/1462-2920.16177]

37. Gong, M.; Yang, G.; Zhuang, L.; Zeng, E.Y. Microbial biofilm formation and community structure on low-density polyethylene microparticles in lake water microcosms. Environ. Pollut.; 2019; 252, pp. 94-102. [DOI: https://dx.doi.org/10.1016/j.envpol.2019.05.090]

38. Zhang, L.; Fang, W.; Li, X.; Gao, G.; Jiang, J. Linking bacterial community shifts with changes in the dissolved organic matter pool in a eutrophic lake. Sci. Total Environ.; 2020; 719, 137387. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.137387]

39. Heylen, K.; Gevers, D.; Vanparys, B.; Wittebolle, L.; Geets, J.; Boon, N.; De Vos, P. The incidence of nirS and nirK and their genetic heterogeneity in cultivated denitrifiers. Environ. Microbiol.; 2006; 8, pp. 2012-2021. [DOI: https://dx.doi.org/10.1111/j.1462-2920.2006.01081.x]

40. He, Z.; Wang, Y.; Bai, X.; Chu, M.; Yi, Y.; Zhu, J.; Gu, M.; Jiang, L.; Zhang, Z. Bacterial Community Composition and Isolation of Actinobacteria from the Soil of Flaming Mountain in Xinjiang, China. Microorganisms; 2023; 11, 489. [DOI: https://dx.doi.org/10.3390/microorganisms11020489]

41. Essoussi, I.; Ghodhbane-Gtari, F.; Amairi, H.; Sghaier, H.; Jaouani, A.; Brusetti, L.; Daffonchio, D.; Boudabous, A.; Gtari, M. Esterase as an enzymatic signature of Geodermatophilaceaeadaptability to Sahara desert stones and monuments. J. Appl. Microbiol.; 2010; 108, pp. 1723-1732. [DOI: https://dx.doi.org/10.1111/j.1365-2672.2009.04580.x]

42. Dastrup, D.B.; Carling, G.T.; Collins, S.A.; Nelson, S.T.; Fernandez, D.P.; Tingey, D.G.; Hahnenberger, M.; Aanderud, Z.T. Aeolian dust chemistry and bacterial communities in snow are unique to airshed locations across northern Utah, USA. Atmos. Environ.; 2018; 193, pp. 251-261. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2018.09.016]

43. Glöckner, F.O.; Zaichikov, E.; Belkova, N.; Denissova, L.; Pernthaler, J.; Pernthaler, A.; Amann, R. Comparative 16S rRNA Analysis of Lake Bacterioplankton Reveals Globally Distributed Phylogenetic Clusters Including an Abundant Group of Actinobacteria. Appl. Environ. Microbiol.; 2000; 66, pp. 5053-5065. [DOI: https://dx.doi.org/10.1128/AEM.66.11.5053-5065.2000] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11055963]

44. Michaud, L.; Caruso, C.; Mangano, S.; Interdonato, F.; Bruni, V.; Lo Giudice, A. Predominance of Flavobacterium, Pseudomonas, andPolaromonaswithin the prokaryotic community of freshwater shallow lakes in the northern Victoria Land, East Antarctica. FEMS Microbiol. Ecol.; 2012; 82, pp. 391-404. [DOI: https://dx.doi.org/10.1111/j.1574-6941.2012.01394.x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22512730]

45. Brinkmeyer, R.; Knittel, K.; Jürgens, J.; Weyland, H.; Amann, R.; Helmke, E. Diversity and Structure of Bacterial Communitiesin Arctic versus Antarctic Pack Ice. Appl. Environ. Microbiol.; 2003; 69, pp. 6610-6619. [DOI: https://dx.doi.org/10.1128/AEM.69.11.6610-6619.2003] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/14602620]

46. Selivanova, E.A.; Poshvina, D.V.; Khlopko, Y.A.; Gogoleva, N.E.; Plotnikov, A.O. Diversity of Prokaryotes in Planktonic Communities of Saline Sol-Iletsk lakes (Orenburg Oblast, Russia). Microbiology; 2018; 87, pp. 569-582. [DOI: https://dx.doi.org/10.1134/S0026261718040161]

47. Li, W.B.; Yang, X.; Tian, Y.N.; Du, L. Differences of plankton community structure in ice-water in Dari Lake under frozen condition. Environ. Sci.; 2021; 42, pp. 2276-2286. [DOI: https://dx.doi.org/10.13227/j.hjkx.202009218] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33884797]

48. Fang, Y.; Changyou, L.; Leppäranta, M.; Xiaonghong, S.; Shengnan, Z.; Chengfu, Z. Notable increases in nutrient concentrations in a shallow lake during seasonal ice growth. Water Sci. Technol.; 2016; 74, pp. 2773-2783. [DOI: https://dx.doi.org/10.2166/wst.2016.433] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27997388]

49. Zhu, C.; Zhang, J.; Nawaz, M.Z.; Mahboob, S.; Al-Ghanim, K.A.; Khan, I.A.; Lu, Z.; Chen, T. Seasonal succession and spatial distribution of bacterial community structure in a eutrophic freshwater Lake, Lake Taihu. Sci. Total Environ.; 2019; 669, pp. 29-40. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2019.03.08] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30877958]

50. Shen, Z.; Xie, G.; Zhang, Y.; Yu, B.; Shao, K.; Gao, G.; Tang, X. Similar assembly mechanisms but distinct co-occurrence patterns of free-living vs. particle-attached bacterial communities across different habitats and seasons in shallow, eutrophic Lake Taihu. Environ. Pollut.; 2022; 314, 120305. [DOI: https://dx.doi.org/10.1016/j.envpol.2022.120305]

51. Zhang, H.; Wang, Y.; Chen, S.; Zhao, Z.; Feng, J.; Zhang, Z.; Lu, K.; Jia, J. Water Bacterial and Fungal Community Compositions Associated with Urban Lakes, Xi’an, China. Int. J. Environ. Res. Public Health; 2018; 15, 469. [DOI: https://dx.doi.org/10.3390/ijerph15030469]

52. Shao, K.; Yao, X.; Wu, Z.; Jiang, X.; Hu, Y.; Tang, X.; Xu, Q.; Gao, G. The bacterial community composition and its environmental drivers in the rivers around eutrophic Chaohu Lake, China. BMC Microbiol.; 2021; 21, 179. [DOI: https://dx.doi.org/10.1186/s12866-021-02252-9]

53. Yuan, W.; Su, X.; Cui, G.; Wang, H. Microbial community structure in hypolentic zones of a brine lake in a desert plateau, China. Environ. Earth Sci.; 2016; 75, 1132. [DOI: https://dx.doi.org/10.1007/s12665-016-5914-9]

54. Wan, F.; Zhang, F.; Zheng, X.; Xiao, L. Study on Ecological Water Demand and Ecological Water Supplement in Wuliangsuhai Lake. Water; 2022; 14, 1262. [DOI: https://dx.doi.org/10.3390/w14081262]

55. Newman, M.E.J. The structure and function of complex networks. Siam Rev.; 2003; 45, pp. 167-256. [DOI: https://dx.doi.org/10.1137/S003614450342480]

56. Shen, Z.; Yu, B.; Shao, K.; Gao, G.; Tang, X. Warming reduces microeukaryotic diversity, network complexity and stability. Environ. Res.; 2023; 238, 117235. [DOI: https://dx.doi.org/10.1016/j.envres.2023.117235]

57. Yuan, M.M.; Guo, X.; Wu, L.; Zhang, Y.; Xiao, N.; Ning, D.; Shi, Z.; Zhou, X.; Wu, L.; Yang, Y. et al. Climate warming enhances microbial network complexity and stability. Nat. Clim. Chang.; 2021; 11, pp. 343-348. [DOI: https://dx.doi.org/10.1038/s41558-021-00989-9]

58. Banerjee, S.; Schlaeppi, K.; van der Heijden, M.G.A. Keystone taxa as drivers of microbiome structure and functioning. Nat. Rev. Microbiol.; 2018; 16, pp. 567-576. [DOI: https://dx.doi.org/10.1038/s41579-018-0024-1]

59. Allesina, S.; Bodini, A.; Pascual, M. Functional links and robustness in food webs. Philos. Trans. R. Soc. Lond. B Biol. Sci.; 2009; 364, pp. 1701-1709. [DOI: https://dx.doi.org/10.1098/rstb.2008.0214]

60. Bernardet, J.F.; Bowman, J.P. The Genus Flavobacterium; Springer: New York, NY, USA, 2006; [DOI: https://dx.doi.org/10.1007/0-387-30747-8_17]

61. Collado, L.; Inza, I.; Guarro, J.; Figueras, M.J. Presence of Arcobacter spp. in environmental waters correlates with high levels of fecal pollution. Environ. Microbiol.; 2008; 10, pp. 1635-1640. [DOI: https://dx.doi.org/10.1111/j.1462-2920.2007.01555.x]

62. Morales Medina, W.R.; Eramo, A.; Fahrenfeld, N.L. Metabolically Active Prokaryotes and Actively Transcribed Antibiotic Resistance Genes in Sewer Systems: Implications for Public Health and Microbially Induced Corrosion. Microb. Ecol.; 2021; 83, pp. 583-595. [DOI: https://dx.doi.org/10.1007/s00248-021-01775-y]

63. Di Pippo, F.; Venezia, C.; Sighicelli, M.; Pietrelli, L.; Di Vito, S.; Nuglio, S.; Rossetti, S. Microplastic-associated biofilms in lentic Italian ecosystems. Water Res.; 2020; 187, 116429. [DOI: https://dx.doi.org/10.1016/j.watres.2020.116429]