Critical transition of soil microbial diversity and composition triggered by plant rhizosphere effects

Abstract

Over the years, microbial community composition in the rhizosphere has been extensively studied as the most fascinating topic in microbial ecology. In general, plants affect soil microbiota through rhizodeposits and changes in abiotic conditions. However, a consensus on the response of microbiota traits to the rhizosphere and bulk soils in various ecosystems worldwide regarding community diversity and structure has not been reached yet. Here, we conducted a meta-analysis of 101 studies to investigate the microbial community changes between the rhizosphere and bulk soils across various plant species (maize, rice, vegetables, other crops, herbaceous, and woody plants). Our results showed that across all plant species, plant rhizosphere effects tended to reduce the rhizosphere soil pH, especially in neutral or slightly alkaline soils. Beta-diversity of bacterial community was significantly separated between into rhizosphere and bulk soils. Moreover, r-strategists and copiotrophs (e.g. Proteobacteria and Bacteroidetes) enriched by 24-27% in the rhizosphere across all plant species, while K-strategists and oligotrophic (e.g. Acidobacteria, Gemmatimonadete, Nitrospirae, and Planctomycetes) decreased by 15-42% in the rhizosphere. Actinobacteria, Firmicutes, and Chloroflexi are also depleted by in the plant rhizosphere compared with the bulk soil by 7-14%. The Actinobacteria exhibited consistently negative effect sizes across all plant species, except for maize and vegetables. In Firmicutes, both herbaceous and woody plants showed negative responses to rhizosphere effects, but those in maize and rice were contrarily enriched in the rhizosphere. With regards to Chloroflexi, apart from herbaceous plants showing a positive effect size, the plant rhizosphere effects were consistently negative across all other plant types. Verrucomicrobia exhibited a significantly positive effect size in maize, whereas herbaceous plants displayed a negative effect size in the rhizosphere. Overall, our meta-analysis exhibited significant changes in microbial community structure and diversity responding to the plant rhizosphere effects depending on plant species, further suggesting the importance of plant rhizosphere to environmental changes influencing plants and subsequently their controls over the rhizosphere microbiota related to nutrient cycling and soil health.

Keywords: K-strategists; meta-analysis; microbiome; r-strategists; rhizosphere.

Figure 1

Global distribution of rhizosphere and bulk data used in this meta-analysis, including 861 data points from 101 publications. Six groups of plant species are shown in the legends, with sample sizes for each group given in parentheses.

Figure 2

Effect size of soil pH (A) and plant species effect on soil pH (B). The bulk soil pH was grouped ranging from 5 (pH< 5), 6 (5 ≤ pH<6), 7 (6 ≤ pH<7), 8 (7 ≤ pH<8) and 9 (8 ≤ pH<9). Plant species including maize, rice, vegetables, herbaceous and woody plants. Data are expressed as the mean effect size (d₊₊) with bias-corrected 95% confidence intervals. Percentage changes for means and observation numbers for the category are given in parentheses. Asterisk indicates P< 0.05.

Figure 3

Effects of plant species on bacterial alpha-diversity, i.e., Richness and Shannon indices, grouped by the plant species. Data is expressed as the mean effect size (d₊₊) with bias-corrected 95% confidence intervals. Percentage changes for means and observation numbers for the category are given in parentheses. Asterisk indicates P< 0.05.

Figure 4

The relative abundance of major bacterial taxas at the phylum levels (A), the principal coordinate analysis (PCoA) of microbial beta-diversity based on Bray-Curtis similarity distances between rhizosphere and bulk soils (B) and among plant species (C).

Figure 5

Plant rhizosphere effects on major bacterial phyla (mean proportion > 5%), i.e., Bacteriodetes and Proteobacteria, grouped by the plant species including maize, rice, vegetables, other crops, herbaceous and woody plants. Data is expressed as the mean effect size (d₊₊) with bias-corrected 95% confidence intervals. Percentage changes for means and observation numbers for the category are given in parentheses. Asterisk indicates P< 0.05.

Figure 6

Plant rhizosphere effects on major bacterial phyla (mean proportion > 5%), i.e., Acidobacteria, Actinobacteria, Firmicutes and Chloroflexi, grouped by the plant species including maize, rice, vegetables, other crops, herbaceous and woody plants. Data is expressed as the mean effect size (d₊₊) with bias-corrected 95% confidence intervals. Percentage changes for means and observation numbers for the category are given in parentheses. Asterisk indicates P< 0.05.

Figure 7

Plant rhizosphere effects on other bacterial phyla (median proportion of 1-5%), i.e., Gemmatimonadete, Nitrospirae, Planctomycetes and Verrucomicrobia, grouped by the plant species including maize, rice, vegetables, other crops, herbaceous and woody plants. Data is expressed as the mean effect size (d₊₊) with bias-corrected 95% confidence intervals. Percentage changes for means and observation numbers for the category are given in parentheses. Asterisk indicates P< 0.05.

Figure 8

Plant rhizosphere effects on major fungal phyla (mean proportion > 5%), i.e. Ascomycota and Basidiomycota, grouped by the plant species including maize, rice, vegetables, other crops, herbaceous and woody plants. Data is expressed as the mean effect size (d₊₊) with bias-corrected 95% confidence intervals. Percentage change for means and observation numbers for the category are given in parentheses. Asterisk indicates P< 0.05.