The effect of crop rotation on soil-borne diseases is a representative case of plant-soil feedback in the sense that plant disease resistance is influenced by soils with different cultivation histories. This study examined the microbial mechanisms inducing the differences in the clubroot (caused by Plasmodiophora brassicae pathogen) damage of Chinese cabbage (Brassica rapa subsp. pekinensis) after the cultivation of different preceding crops. It addresses two key questions in crop rotation: changes in the soil bacterial community induced by the cultivation of different plants and the microbial mechanisms responsible for the disease-suppressive capacity of Chinese cabbage. Twenty preceding crops from different plant families showed significant differences in the disease damage, pathogen density, and bacterial community composition of the host plant. Structural equation modelling revealed that the relative abundance of four key bacterial orders in Chinese cabbage roots can explain 85% and 70% of the total variation in pathogen density and disease damage, respectively. Notably, the relative dominance of Bacillales and Rhizobiales, which have a trade-off relationship, exhibited predominant effects on pathogen density and disease damage. The disease-suppressive soil legacy effects of preceding crops are reflected in compositional changes in key bacterial orders, which are intensified by the bacterial community network.

Clubroot (Plasmodiophora brassicae) is an important soilborne disease that causes severe damage to cruciferous crops in China. This study aims to compare the differences in chemical properties and microbiomes between healthy and clubroot-diseased soils. To reveal the difference, we measured soil chemical properties and microbial communities by sequencing 18S and 16S rRNA amplicons. The available potassium in the diseased soils was higher than in the healthy soils. The fungal diversity in the healthy soils was significantly higher than in the diseased soils. Ascomycota and Proteobacteria were the most dominant fungal phylum and bacteria phylum in all soil samples, respectively. Plant-beneficial microorganisms, such as Chaetomium and Sphingomonas, were more abundant in the healthy soils than in the diseased soils. Co-occurrence network analysis found that the healthy soil networks were more complex and stable than the diseased soils. The link number, network density, and clustering coefficient of the healthy soil networks were higher than those of the diseased soil networks. Our results indicate that the microbial community diversity and network structure of the clubroot-diseased soils were different from those of the healthy soils. This study is of great significance in exploring the biological control strategies of clubroot disease.