Drought-induced shifts in cowpea rhizoplane bacterial communities across different vegetative and reproductive stages

ORCID

Bheemanahalli: https://orcid.org/0000-0002-9325-4901

MSU Affiliation

College of Agriculture and Life Sciences; Department of Plant and Soil Sciences; Institute for Genomics, Biocomputing and Biotechnology

Creation Date

2025-11-19

Abstract

The increasing prevalence of drought poses significant challenges to global food security, necessitating a deeper understanding of plant-microbiome interactions which help crop production. This study investigated the dynamics of drought stress-induced changes in rhizosphere-associated bacterial communities of two cowpea (Vigna unguiculata L.) genotypes (EpicSelect4 and UCR369) across four growth stages. Community-level physiological profiling using Biolog EcoPlate analysis revealed that drought reduced rhizosphere microbial metabolic activity (carbon substrate utilization) in both genotypes, but UCR369 maintained higher metabolic capability than EpicSelect4 across growth stages. Further, integration of amplicon metagenomics and physiological data showed that drought significantly altered rhizoplane bacterial communities in cowpea, with distinct genotype-specific responses. There was a decline in Alpha diversity under drought, while community composition shifted based on genotype. Beta diversity results revealed that genotype and drought significantly influenced microbial community structure across growth stages. Proteobacteria dominated the root zone of the EpicSelect4 genotype, while UCR369 showed an increase in Actinobacteria under drought conditions. Redundancy analysis revealed that soil enzyme activities (β-glucosidase and N-acetyl-glucosaminidase) and physiological traits werecorrelated significantly with microbial community shifts. Interpretable machine learning approach identified Actinobacteriota and Cyanobacteria as the key biomarkers enriched under drought, with genera such as Streptomyces and Ensifer potentially contributing to drought tolerance. The Random Forest model coupled with SHapley Additive exPlanations (SHAP) values demonstrated high predictive accuracy for identifying drought-related biomarkers, aligning with DeSeq2 analysis results. These models provided insights into the potential contributions of specific microbial taxa to cowpea drought tolerance, offering a promising avenue for developing microbiome-based strategies to improve crop resilience and sustainability under drought conditions.

Publication Date

9-1-2025

Publication Title

Plant Stress

Publisher

Elsevier

Link to Full Text

Share

COinS
 

Digital Object Identifier (DOI)

https://doi.org/10.1016/j.stress.2025.100915