Tissue-specific metabolomic profiling reveals cultivar-dependent drought tolerance mechanisms in rice (Oryza sativa L.)

Drought stress threatens rice production globally, causing yield reductions of 50–70%. This study employed gas chromatography–mass spectrometry (GC-MS)-based metabolomics to investigate drought responses in four Egyptian rice cultivars (Giza 177, Giza 178, Sakha 104, Sakha 108) exposed to 15% polyethylene glycol stress for 14 days. Sakha 104 demonstrated the highest drought tolerance index (82.7%). Metabolomic analysis identified 114 polar metabolites in leaves and 97 in roots, representing amino acids, organic acids, sugars, and related compounds involved in osmotic adjustment. Principal component analysis (PCA) showed greater metabolic adaptability in roots (70.4% variance) than leaves (53.9%). Variable importance in projection analysis identified 40 metabolites with VIP > 1.0, among which 13 priority biomarkers (VIP > 1.6) showed the strongest discriminatory capacity, including amino acids, organic acids, sugars, and secondary metabolites. Hierarchical clustering revealed leaf metabolomes grouped by treatment, while root metabolomes clustered by cultivar, indicating tissue-specific adaptation strategies. Cultivar-specific signatures included leucine and L-serine in Giza 177, increased adenosine-5’-monophosphate in Sakha 104, higher oxalic acid and spermine in Sakha 108, and shikimic acid in Giza 178 roots. Venn diagram analysis identified only four metabolites universally expressed in leaves and three in roots. Strong correlations between metabolomic profiles and physiological traits confirmed the relevance of the biomarkers. Root tissues exhibited roughly 20 times more metabolic diversity than leaves, supporting their role as primary stress sensors. These findings identify practical metabolic biomarkers for breeding programs and shed light on the biochemical mechanisms underlying drought tolerance in Egyptian rice germplasm.