Rhizosphere-associated soil microbiome variability in Verticillium wilt-affected Cotinus coggygria

Abstract

Introduction: Verticillium wilt is the most devastating soil-borne disease affecting Cotinus coggygria in the progress of urban landscape construction in China.

Methods: To assess the variability of the rhizosphere-associated soil microbiome in response to Verticillium wilt occurrence, we investigated the microbial diversity, taxonomic composition, biomarker species, and co-occurrence network of the rhizosphere-associated soil in Verticillium wilt-affected C. coggygria using Illumina sequencing.

Results: The alpha diversity indices of the rhizosphere bacteria in Verticillium wilt-affected plants showed no significant variability compared with those in healthy plants, except for a moderate increase in the Shannon and Invsimpson indices, while the fungal alpha diversity indices were significantly decreased. The abundance of certain dominant or crucial microbial taxa, such as Arthrobacter, Bacillus, Streptomyces, and Trichoderma, displayed significant variations among different soil samples. The bacterial and fungal community structures exhibited distinct variability, as evidenced by the Bray-Curtis dissimilarity matrices. Co-occurrence networks unveiled intricate interactions within the microbial community of Verticillium wilt-affected C. coggygria, with greater edge numbers and higher network density. The phenomenon was more evident in the fungal community, showing increased positive interaction, which may be associated with the aggravation of Verticillium wilt with the aid of Fusarium. The proportions of bacteria involved in membrane transport and second metabolite biosynthesis functions were significantly enriched in the diseased rhizosphere soil samples.

Discussion: These findings suggested that healthy C. coggygria harbored an obviously higher abundance of beneficial microbial consortia, such as Bacillus, while Verticillium wilt-affected plants may recruit antagonistic members such as Streptomyces in response to Verticillium dahliae infection. This study provides a theoretical basis for understanding the soil micro-ecological mechanism of Verticillium wilt occurrence, which may be helpful in the prevention and control of the disease in C. coggygria from the microbiome perspective.

Keywords: Cotinus coggygria; Illumina sequencing; Verticillium wilt; plant health; rhizosphere; soil microbiome.

Figure 1

Boxplots of the alpha diversity indices of bacterial (A) and fungal (B) OTUs in the Cotinus coggygria soil microbiome under different conditions. Different lowercase letters denoted significant differences at p-values < 0.05 levels among different soil samples. RHS, rhizosphere soil from healthy plants; RPS, rhizosphere soil from Verticillium wilt-affected plants; NRHS, non-rhizosphere soil from healthy plants; NRPS, non-rhizosphere soil from Verticillium wilt-affected plants; LS, bulk soil in the C. coggygria forest.

Figure 2

Relative abundances of the dominant bacterial and fungal taxa in the Cotinus coggygria soil microbiome under different conditions. (A) Heatmap representing the relative abundance of the top 30 bacterial taxa at the genus level, (B) average relative abundance of bacterial genera with significant differences among different soil samples, (C) heatmap representing the relative abundance of the top 30 fungal taxa at the genus level, and (D) average relative abundance of fungal genera with crucial features or significant differences among the soil samples. The x-axis represented samples, and the y-axis represented relative abundance. The error bar represented the standard deviation of mean values. Different lowercase letters showed significant differences in abundance among different soil samples based on the LSD test (p < 0.05). RHS, rhizosphere soil from healthy plants; RPS, rhizosphere soil from Verticillium wilt-affected plants; NRHS, non-rhizosphere soil from healthy plants; NRPS, non-rhizosphere soil from Verticillium wilt-affected plants; LS, bulk soil in the C. coggygria forest.

Figure 3

Principal coordinate analysis (PCoA) of bacterial (A) and fungal (B) communities of the Cotinus coggygria soil microbiome under different conditions. The PCoA plots were based on the Bray–Curtis dissimilarity of bacterial and fungal communities at the OTU levels according to the ANOSIM test. RHS, rhizosphere soil from healthy plants; RPS, rhizosphere soil from Verticillium wilt-affected plants; NRHS, non-rhizosphere soil from healthy plants; NRPS, non-rhizosphere soil from Verticillium wilt-affected plants; LS, bulk soil in the C. coggygria forest.

Figure 4

Linear discriminant analysis effect size (LEfSe) of bacterial taxa in the Cotinus coggygria soil microbiome under different conditions. The non-parametric factorial Kruskal–Wallis test with p < 0.05 and logarithmic LDA score > 3.5 was used to identify the significantly enriched bacterial taxa (using the all-against-all comparisons parameter). Different colors depicted different soil samples, while circles from inside to outside showed phylogenetic levels from phylum to genera. The sizes of the circles were proportional to the mean relative abundance of each taxon. The yellow circles represented the absence of significantly different taxa. Genera with a relative abundance of less than 0.1% were not included.

Figure 5

Linear discriminant analysis effect size (LEfSe) of fungal taxa in the Cotinus coggygria soil microbiome under different conditions. The non-parametric factorial Kruskal–Wallis test with p < 0.05 and logarithmic LDA score > 3.5 was used to identify the significantly enriched fungal taxa (using the all-against-all comparisons parameter). The notes were the same as Figure 4.

Figure 6

Co-occurrence networks constructed based on bacterial and fungal communities in the Cotinus coggygria soil microbiome under different conditions. (A) Bacterial networks of soil samples from healthy plants, BHS, (B) bacterial networks of soil samples from Verticillium wilt-affected plants, BPS, (C) fungal networks of soil samples from healthy plants, FHS, and (D) fungal networks of soil samples from Verticillium wilt-affected plants, FPS. The nodes were colored based on phyla, and each node size was proportional to the degree of centrality. A connection denotes a strong (Spearman’s ρ > 0.8 or < −0.8, red or blue edges) and significant (p < 0.05) correlation. Potential keystone taxa were shown with bold genera names.

Figure 7

Proportions of bacterial functions in the Cotinus coggygria soil microbiome under different conditions based on KEGG and Bugbase databases. (A,B) KEGG ortholog function for bacteria in the rhizosphere (A) and non-rhizosphere (B) soil samples, (C) bacteria function profiles in different soil samples based on Bugbase prediction. RHS, rhizosphere soil from healthy plants; RPS, rhizosphere soil from Verticillium wilt-affected plants; NRHS, non-rhizosphere soil from healthy plants; NRPS, non-rhizosphere soil from Verticillium wilt-affected plants; LS, bulk soil in C. coggygria forest.