Characterization of Changes in Active Ingredients and Mining of Key Metabolites in Bletilla striata under Shading and Drought Stresses

Abstract

Shading and drought are considered crucial abiotic stress factors that limit the normal growth of plants. Under natural conditions, the quality of Bletilla striata pseudobulbs (BP), a Chinese traditional medicinal crop, is often affected by the dual stresses of shading and drought. However, the relationship and mechanism of the interaction between the two stress factors in B. striata remain unclear. In this study, we examined the changes in photosynthetic properties and active ingredients of B. striata under shading (L), drought (W), and shading-drought dual stresses (LW). We aimed to explore the metabolite mechanism that led to these changes using GC-MS-based non-targeted metabolomics techniques. The results indicated a significant reduction in the polysaccharide content of BP under W and LW treatments compared to the control (CK). The total phenol content was significantly reduced under L treatment, while the total flavonoid content did not change significantly under the three stresses. The significant increase in militarine content under all three stresses implies that B. striata may modulate its biosynthesis in response to different environmental stresses. Transpiration rate and stomatal conductance were reduced, amino acid expression was up-regulated, and carbohydrate expression was down-regulated in B. striata under L treatment. The net photosynthesis rate, stomatal conductance, and transpiration rate exhibited significant reductions, and the tuber metabolic disorder marker Homocysteine increased and organic acid content as well under W treatment. The net photosynthetic rate, transpiration rate, stomatal conductance, and water use efficiency of B. striata were further reduced under LW compared with single stress, which is in agreement with the “trade-off theory”. Pseudobulb metabolite changes, in combination with the results of the two single stresses, showed an up-regulation of amino acids and disaccharide compounds and a down-regulation of monosaccharide compounds. A support vector machine model (SVM) was used to screen 10 marker metabolites and accurately predict the changes in active ingredient content through an artificial neural network model (ANN). The results suggest that an appropriate stress environment can enhance the content of the target active ingredients based on cultivation goals.