Rivers play a vital role in maintaining global biodiversity and the overall stability of aquatic ecosystems [1, 2]. However, increasing urbanization is leading to widespread community disassembly and posing an overwhelming threat to riverine biodiversity [3–5]. For instance, extensive human settlement along rivers alters significant hydrological conditions that influence the community structure and function of riverine communities [6–8]. Anthropogenic activities affect water quality through nutrient inputs, flow modifications, and alteration of watershed-scale networks between riverine and upland habitats, thereby hindering the sustainable provision of ecosystem services [7, 9–11]. Urbanization-induced changes in land use types [12] and eutrophication in aquatic environments [13] have accelerated non-random species losses, promoted the recruitment of tolerant species, and ultimately led to a reorganization of ecological communities [14, 15]. Therefore, understanding how urbanization reshapes aquatic communities and their underlying ecological mechanisms is a prerequisite for environmental assessment, river ecosystem restoration and biodiversity conservation [3, 16, 17].
Previous studies have shown that microbial communities are an essential part of nutrient cycling and organic pollutant degradation in aquatic ecosystems [18]. Moreover, they also play an important role in the functioning of ecosystems by converting nutrients and providing the basal food source for higher trophic levels [19]. In natural ecosystems, microbial communities can be broadly categorized into habitat generalists and specialists based on indices such as Shannon diversity, abundance, and ecological niche widths [20, 21]. Generally, generalists are more environmentally tolerant and are more common in different habitats [22], whereas specialists are more restricted to specific habitats due to their lower environmental tolerance and susceptibility to environmental changes [23, 24]. However, conflict studies suggest that microbial generalists may be more sensitive to environmental changes than specialists [25, 26]. Therefore, the question remains open as to how different environmental conditions affect the assembly of microbial generalists and specialists.
It is generally accepted that the assembly of microbial communities under various conditions can be explained by either ecological niche theory (deterministic processes) or neutral theory (stochastic processes) in aquatic ecosystems [27, 28]. The assembly processes of planktonic microbial communities mainly include homogeneous selection, heterogeneous selection, dispersal limitation, homogenizing dispersal and ecological drift [29]. For instance, the community assembly of generalists and specialists in the sedimentary habitats of Lake Tibetan was been found to be dominated by stochastic processes [30]. However, study [31] pointed out that the community assembly of specialists in upland lakes is controlled only by deterministic processes, which is consistent with the results of studies on generalists and specialists in lakes and reservoirs at different latitudes in eastern China [32]. Study [33] found that the assembly of generalists in the plastic cycle of freshwater lakes is determined by deterministic processes, while the assembly of specialists is determined by stochastic processes. This implies that the community assembly of generalists and specialists shows significant difference between different habitats with temporal and environmental variables [34]. Stochastic processes that shaped microbial communities across different seasons and scales were discovered for river ecosystems [35]. However, quantifying bacterioplankton community assembly in riverine ecosystems with different levels of urbanization, especially to distinguish between generalist and specialist communities, is much less appreciated.
Recently, the continued development of co-occurrence networks has provided a powerful tool for unraveling the complex interactions between microorganisms [36, 37]. In a co-occurring network, highly connected modular structures are considered as important ecological units [38], while the nodes of topological features are used to determine the potential importance of microorganisms in the network [39]. As well as the robustness of the network, the resistance of the network to the loss of some nodes is also determined [40]. A study of lake sediment microbes on the Tibetan Plateau found that habitat specialists were more closely connected between modules compared to generalists [30]. Another study indicated that the network structure of specialists was more complex compared to generalists in subtropical bays [41]. Furthermore, comparing topological features at the network level helps to better understand the differences in co-occurrence patterns between generalist and specialist subcommunities [33].
The Yangtze River, the third longest and third richest river system in the world, is being destroyed by human expansion [42]. Over the past three decades, over 15 large dams have been built along the main branch of the upper Yangtze River, including the Three Gorges Dam, the largest dam in the world [43, 44], to support sustainable economic growth in China. The various responses of microbial diversity and community assembly processes to damming effects have been extensive investigated [43, 45–47]. Since more than 400 million residents live in the river basin [48], there was a large immigration to Chongqing city due to the Three Gorges Reservoir (TGR) project, resulting in rapid urbanization of the city [49]. By the end of 2011, at least 31.4 million people lived in Chongqing Municipality [50]. Rapid urbanization in the tributaries of the river basin [51–53] leads to severe eutrophication and water quality degradation [54–56]. For instance, aquatic microbial diversity and community structure may be vulnerable to the spatial variability of physicochemical and biotic parameters during urbanization [57], and microbial communities in subtropical coastal ecosystems are less resistant to environmental changes caused by urbanization than that caused by reclamation [58]. However, it is not clear how plankton microbial diversity and community assembly processes will respond to urbanization effects. Therefore, quantifying the effects of urbanization on local environmental conditions will improve the understanding of microbial community patterns arising from the interplay between regional land use and local environmental conditions.
Given that urbanization may constitute a strong environmental filter, understanding vital ecological processes that govern microbial community assembly in rivers under human pressure clearly provides new perspectives for biodiversity conservation and sustainable watershed management. Here, 16S rRNA high- throughput sequencing was used to examine bacterial communities from 18 sampling sites in the upper TGR in summer (August) and winter (December) to ask: i) whether bacterial community structure and ecological functions show significant differences between them have different levels of urbanization? ii) How does the assembly of generalist and specialist species subcommunities alter at different seasons with different levels of urbanization? iii) How does urbanization affect the stability and complexity of interaction networks?