Phosphorus concentrations as an indicator of regime transition
This and previous studies showed that the lake water displays significant environmental differences along the regime transition from the macrophyte-dominated regime (MDR) to the phytoplankton-dominated regime (PDR) [12,49,50]. Certain environmental parameters have been suggested to be responsible for the observed differences in bacterioplankton community, such as pH  and nutrient enrichment , as well as the occurrence of phytoplankton [3,52], which are frequently associated with eutrophication. The mechanisms of the regime shift can involve multiple drivers as well as interactions of macrophytes, phytoplankton, nutrients, and herbivorous waterfowl . Here, we focused on the phosphorus concentration because it is commonly targeted to mitigate eutrophication and it is proposed in previous studies that regime shifts from MDR to PDR in shallow lakes are mostly induced by increasing nutrient concentrations, especially phosphorus [2,52,53]. In this study we found that almost all of the environmental parameters at six sampling sites across the transition from MDR to PDR were significantly linearly correlated with TP, especially turbidity (Fig. S4). Turbidity is a measure of water transparency, which is mainly influenced by concentrations of phytoplankton and re-suspended sediment particles [13,14]. The wind-driven process of sediment re-suspension is much stronger in PDR-Core sites than in MDR-Core sites in Taihu Lake because the submersed macrophyte communities in MDR have a calming effect on the water column and sediment turbulence [8,54], which is consistent with the results found in this study.
Compositional dissimilarity of bacterioplankton communities differed along regime transition
Our results indicate that the bacterioplankton community was signiﬁcantly correlated with environmental factors at the regime transition, especially TP, which is regarded as a limiting nutrient in the eutrophic freshwater lakes that contribute to regime shift [55-57]. A significant difference between bacterioplankton community in lake ecosystems with disparate environmental variables has also been widely observed [13,58,59]. The alpha diversity of bacterioplankton communities along the regime transition from MDR to PDR was observed to decrease and linearly correlate with TP signiﬁcantly, which is consistent with the observation that bacterial diversity was negatively affected by phytoplankton blooms . In general, when the evenness of bacterioplankton community decreases with the regime shift, the diversity does so too, indicating that a particular species becomes more dominant with the regime transition . Also, several taxonomic groups displayed substantial changes in abundance as the regime shifted (Fig. 2 and Fig. S7). Regime shifts provide a mechanism for selection, as abundant OTUs become rare or even more abundant due to their sensitivity to changes in the environmental conditions [1,3]. Intriguingly, these changes were occasionally non-linear, suggesting that taxa had different optima along the MDR-PDR gradient. For a number of taxa, e.g., Acidobacteria and Cyanobacteria, the relationship to TP is well described by quadratic functions, which means that the taxon has an optimal value for TP and declines in abundance below and above that value, as opposed to a gradual change of bacterioplankton communities along the environmental gradient along the regime shift from MDR to PDR. It has been demonstrated that Betaproteobacteria tends to be the dominant group in relatively oligotrophic lakes [12,13], while Firmicutes, which appears to be involved in DOC degradation [62,63], often become dominant in eutrophic lakes alongside blooms of Cyanobacteria . Our results, that Betaproteobacteria and Firmicutes were negative and positive linearly correlated with TP, are consistent with these findings. Therefore, those consistently variable OTUs might be used as discriminant taxa for the different regimes.
Changes in assembly processes of bacterioplankton communities were attributed to regime transition
Although the deterministic processes were observed to dominate in shaping the bacterioplankton communities in both the MDR-Core and PDR-Core sites, the deterministic process in MDR-Core sites were observed to be stronger than in PDR-Core sites. According to other experimental studies, competition among species could be reduced in more productive environments, hence weakening selection and strengthening the stochasticity of bacterial communities [5,65]. Previous work has also shown that the degree of determinism increases in extreme or low resource environments for the assembly process of microbial community [66-69]. The stronger deterministic process in MDR-Core sites may be due to lower nutrient concentrations compared to PDR-Core-sites, leading to relatively stronger selection due to increased competition for resources and less diverse resources . As to the decreased contribution of deterministic processes in the edge sites of MDR and PDR, there could be two reasons. On the one hand, edge sites could provide a more favorable living environment because of the higher nutrient concentrations compared to MDR-Core sites and lower algal toxin pressure compared to PDR-Core sites . On the other hand, water disturbance driven by strong wind in the edge sites of the two regimes in Lake Taihu could also have resulted in an increase of the disturbance of environmental conditions and the passive dispersal of species, hence increasing the stochasticity of bacterioplankton community [65,70]. While the comparably strong deterministic process in PDR-Core sites may be due to the algal toxin pressure that comes from Cyanobacteria bloom , but weakened by the extremely abundant resources . Of the investigated ecological processes, variable selection contributed the most to the assembly of bacterioplankton community (Fig. S9b). In freshwater lakes, both community structure and the assembly processes of bacterioplankton have been shown to be intensively inﬂuenced by nutrient loading [5,49] as well as submerged macrophytes [8,12]. The heterogeneous pattern of the bacterioplankton community assembly processes in this study could therefore be attributed to differences in nutrient loading along the regime shift from MDR to PDR (Fig. 5).
Regime transition revealed by network clusters and hub species
In ecological systems, coexistence is potentially supported by niche processes like environmental ﬁltering, as the different filters lead to establishment of different communities [17,40]. Species that share similar ecological niches may compete or cooperate to resist environmental pressure when resources are scarce or under environmental stress [71,72]. Compared to the MDR-core sites, the Cyanobacteria bloom caused by high concentration of TP in PDR-Core sites strengthen selection due to its algal toxin [60,62]. The modularity of the site-specific co-occurrence network varied along the regime shift (Table 1). Previous studies have shown the existence of environmentally driven modules [73,74]. In this study we also found that closely related taxa tended to be positively interconnected and clustered together (Fig. 4). Moreover, we found several hub species in the three network clusters, which are likely indicators of the effect of the regime shift on the bacterioplankton community. The first and third cluster contain taxa that dominate the MDR or PDR regime, respectively. These are also the taxa that consistently increase or decrease across the environmental gradient (Fig. 2). For instance, Firmicutes dominating cluster 3 and the PDR regime increase with TP while Betaproteobacteria dominating cluster 1 and the MDR regime decrease with TP. The taxa in the second cluster show more subtle patterns. Actinobacteria first increase and then decrease with TP, thus differing in their response from the Firmicutes, whereas Gammaproteobacteria increase with TP but then saturate, in contrast to both the Firmicutes and the Actinobacteria. Thus, cluster 2 groups taxa which benefit from increased TP, but only up to a point. Given that clusters in microbial co-occurrence networks may represent different niches , the TP-correlated clusters indicate high similarity in species' co-occurrence patterns along the regime shift and may be summarized by the hub species in these clusters (Fig. 5).