Genotypic variation in maize leaf senescence after ear removal is caused by excess accumulation of photosynthates

Abstract Leaf senescence is a key developmental process influencing photosynthesis, carbon allocation, and crop productivity, all tightly regulated by source–sink balance. In maize (Zea mays), abscisic acid (ABA) signaling and non-structural carbohydrate (NSC) partitioning have been proposed as major determinants of senescence timing under sink limitation. Through multi-year, multi-genotype analyses, we refine the current understanding by identifying starch-induced chloroplast disruption as the primary trigger of sink-removal-induced senescence, whereas ABA acts mainly as a secondary, modulatory signal. We also reveal a respiration-linked metabolic compensation mechanism that stabilizes leaf function in genotypes with delayed senescence. In early-senescing genotypes, ear removal induced significant chlorophyll loss, anthocyanin accumulation, and severe sugar/starch overaccumulation that disrupted chloroplast ultrastructure, coinciding with reduced leaf protein contents and impaired photosynthesis. By contrast, genotypes with delayed senescence maintained pigment stability, displaying enhanced dark respiration, balanced carbohydrate metabolism, and chloroplast integrity. Shading experiments further confirmed that starch overaccumulation, rather than ABA elevation, is the dominant driver of sink-removal-induced senescence. Metabolomic analyses showed that early-senescing genotypes accumulate soluble sugars and phosphoenolpyruvate (PEP)-pathway intermediates, whereas genotypes with delayed senescence exhibited reduced tricarboxylic acid (TCA) cycle activity and elevated aspartate-family amino acids. Transcriptomics revealed downregulation of sucrose transporters (e.g., ZmSUT7) and starch degradation genes (e.g., ZmBmy3) alongside upregulation of anthocyanin and carbohydrate metabolism genes in early-senescing genotypes. Co-expression network analysis further identified hub genes associated with carbohydrate and amino acid metabolism, linking transcriptional regulation to genotype-specific metabolic adjustments. Together, these findings refine current source–sink regulatory models and offer potential physiological and metabolic targets for improving source–sink coordination and enhancing yield resilience in maize.