Does genome size drive the pH-related shifts in bacterial biodiversity within forest soils?

Over the last decade, the mechanisms underlying changes in bacterial biodiversity in forest soils based on 16S rRNA sequencing have gained considerable attention in ecological and geographical research (Bahram et al., 2018;Piton et al., 2023). Soil pH plays a pivotal role in the development of soil bacterial biodiversity (Ramoneda et al., 2023;Luan et al., 2023;He et al., 2023). To explore the mechanism underlying the changes in bacterial biodiversity in forest soils with varying pH, researchers have evaluated bacterial 16S diversity. Wang et al. (2023) analysed the 16S and bacterial metagenomic data from soil samples obtained at 12 forest sites in China and concluded that on a pH gradient, the mismatch between taxonomic and functional diversity may be attributed to changes in genome size. That is, a small community of large-genome taxa could harbour higher functional diversity than a large community of small-genome taxa. Biodiversity in resource-limited conditions with acidic pH environments (commonly found at low latitudes) should be lower than that in resource-abundant conditions with neutral pH environments (commonly found at high latitudes). Yet this proposed decoupling stands in contrast to the theoretical expectation that genome size and biodiversity exhibit a positive correlation under natural ecological conditions (Lynch & Conery, 2003;Joyce, 2002;Pellicer et al., 2023). 2020), all of whom utilized identical sampling sites, methods, and experimental conditions. Since the measurements and primary analyses were conducted by the same research group, the datasets are highly comparable. All statistical analyses were performed in IBM SPSS Statistics 27.0, and line graph plotting was conducted using OriginPro 2022 (v9.9.0.225).Forest soil pH in China peaks near 40°N (Zheng et al., 2022). While Wang et al. reported a positive linear correlation, our analysis revealed a nonlinear relationship, particularly near neutral pH (Wang et al., 2023). Previous studies have also shown that if Wang et al. could add more sampling sites in the north of 40 °N, the unimodal curve could be more clearly observed (Song, 2023). In our study, bacterial biodiversity showed a significant unimodal relationship with latitude. (Figure S1a). Furthermore, unimodal relationships between bacterial functional diversity, genome size, and carbohydrate-active enzyme gene diversity and latitude were observed through metagenomic annotation and quantification of diversity indices based on richness (Figure S1b-d); all plot visualizations were generated in OriginPro 2022. Our work aligns with previous studies showing a unimodal pattern of bacterial biodiversity around 40°N (Liu et al., 2020). Therefore, the interpretation presented by Wang et al. may be affected by the fact that they only selected three sites north of 40°N, which may have been insufficient to clearly reveal the unimodal pattern (Wang et al., 2023).Meanwhile, both bacterial functional diversity and genome size display a significant decreasing -increasing relationship with latitude, also turning near 40°N (Figure S1b-d).Bacterial biodiversity is primarily linked to latitude, whereas functional diversity and genome size correlate more closely with pH. Thus, the observed large genomes and high functional diversity in acidic soils likely reflect the prevalence of such soils in climatically favorable low-latitude forests, not a direct stress response, leading to the earlier misinterpretation.Fungi, particularly mycorrhizal fungi, are widely recognized as key drivers of microbial diversity in forest soils of China due to their central role in nutrient cycling and cannot be overlooked (Chen et al., 2022). We found that latitude indirectly affects fungal diversity by positively influencing plant nitrogen and phosphorus reuptake rates (NR/PR), and NR/PR reflects N/P limitation (Du et al., 2020). A possible explanation for this pattern is that when plants are limited by phosphorus, they become more reliant on symbiotic fungi for nutrient supply (Zheng et al., 2022). This may create a cascading effect on the dependence of the plants on symbiotic fungi, thereby shaping the latitudinal variation pattern of fungal diversity S2a), no such correlation exists for functional diversity. Fungal diversity and bacterial carbohydrate-active enzyme gene diversity (Figure S2b) and genome size (Figure S2c) were analyzed for pairwise correlations via SPSS Statistics 27.0, with diversity metrics based on richness for all microbial indices. In contrast to the taxonomic pattern, fungal diversity is negatively correlated with the diversity of bacterial carbohydrate-active enzyme genes (Figure S2b). This observational pattern is consistent with the dominant role of mycorrhizal fungi in the carbon cycle, where they directly exchange carbon, nitrogen, and phosphorus with plant roots (Chen et al., 2022; Lebreton et al., 2021). Their symbiotic advantage suggests they may occupy key carbon-metabolism niches, potentially contributing to competitive exclusion of bacteria from these functions (Song et al., 2023). However, the nutrients processed and released by fungi into the soil create new ecological niches, which we interpret as consistent with a "trickle-down effect" that promotes bacterial taxonomic diversity and may account for the observed positive correlation (Jacoby & Kopriva, 2019). Notably, fungal diversity shows no linkage to bacterial genome size (Figure S2c), suggesting that these putative fungal-driven processes operate largely independently of genome size evolution in bacteria.To verify whether our inference is consistent with the actual situation, we used structural equation modelling in R 4.4.1 (packages lavaan v0.6-11 and piecewiseSEM v2.1.0) to examine the ecological processes suggested by (H1) Wang et al. and (H2) this study (Figure 1a,b). The results showed that H2 had a higher degree of explanation for bacterial diversity than H1. Because in the collinearity detection, latitude and pH not only did not show a significant linear correlation (R²=0.213, P=0.131), but the VIF was 1.27, which is less than 5, which indicates a collinear relationship. Therefore, we combined the two models into one for verification and found that the effect size of the relationship between genome size and functional diversity was very small, less than 0.1 (Figure 1c). This confirms our assertion that although soil pH had a significant impact on genome size, genome size had a very small impact on functional diversity, and therefore was not the main driver of bacterial diversity in the context of large-scale forest soil pH gradients in China. Specifically, the ratio of the absolute value of the parameter estimate to the standard error for the pathway from bacterial genome size to functional diversity is approximately 0.336 (Supplementary Material S1), which is far below the 1.96 threshold for the 95% confidence interval, indicating that the observed weak effect of this pathway is merely due to random error rather than a true causal relationship. Notably, this pathway is the only one with a non-significant effect among all pathways affecting bacterial taxonomic diversity (S.16S), which effectively excludes systematic error as a potential confounding factor. We recognize that such genome size effects may be scale dependent and context specific, and are not intended to imply they are universally negligible in other ecological systems or spatial scales. In conclusion, the verification results of the structural equation model indicate that H2 was clearly more consistent with the actual situation than H1, as the p-values derived from both the chi-square test and Fisher's C test all demonstrated that the fit of H2 was far superior to that of H1, leading to this conclusion, indicating that the variation in bacterial biodiversity in forest soils in China with pH is not driven by genome size, but mainly by geographical and fungal factors.This study demonstrates that pH-driven variations in forest soil bacterial biodiversity across China are not governed by genome size, as previously suggested, but are primarily influenced by geographical and fungal factors. The earlier conclusion-that the decoupling of bacterial taxonomic and functional diversity stems from changes in genome size-likely overlooked the confounding effects of site heterogeneity and large-scale geographical influences. To enhance the robustness of future research, we propose: 1) accounting for geographical and fungal variables; 2) ensuring the selection of representative sampling sites; and 3) employing transect-based or controlled experiments to better isolate environmental drivers.