作物轮作
微生物群
农林复合经营
环境科学
作物
环境资源管理
业务
环境规划
生物
农学
生物信息学
作者
Jiaqing Wu,Yixiang Liu,Hai Yu,Feifei Fan,Xiahong He,Youyong Zhu,Dong Yang,Min Yang,Shusheng Zhu
标识
DOI:10.1016/j.xplc.2025.101502
摘要
Crops leave a soil legacy with altruistic effects for subsequent crops but not for themselves. While research has focused on improvements in soil physicochemical properties and the suppression of non-host pathogens, the altruistic microbiome and its assembly mechanism driven by root exudates remain largely unknown. Here, we identified altruistic but self-detrimental phenomena when garlic was rotated with other crops based on meta-analysis and in vivo experiments. Studies utilizing a globally adopted garlic-pepper rotation system demonstrated density-dependent enrichment of key microbial taxa, especially the Penicillium genus, which supports the healthy growth of non-Allium plants but exhibits pathogenicity toward garlic. Furthermore, we found that garlic roots stably secrete diallyl disulfide (DADS) into soil, imposing reactive oxygen species (ROS) stress in the rhizosphere and reshaping the microbial community, particularly suppressing ROS-sensitive pathogens while enriching ROS-tolerant beneficial microorganisms. As a result, Penicillium allii, with strong oxidative stress tolerance, survives and accumulates in the highly stressful garlic rhizosphere environment, thereby playing an "altruistic but self-detrimental" role in the rotation system. In addition, preliminary field experiments showed that co-application of DADS with P. allii could enhance stable colonization of P. allii, promoting sustainable management of soil-borne diseases and improving yield. In summary, this study reveals that garlic root exudate DADS triggers ROS-mediated selection pressure, enriching stress-tolerant P. allii and establishing an "altruistic" microbiome succession mechanism in crop-rotation systems. This mechanism enables targeted soil-borne disease management through plant-driven microbial community engineering.
科研通智能强力驱动
Strongly Powered by AbleSci AI