Abstract
INTRODUCTION: Quinolizidine alkaloids, such as matrine and sophocarpine, enriched in Sophora davidii seeds, demonstrate notable anticancer properties. However, the biosynthetic pathway of these alkaloids remains incompletely elucidated, and the expression patterns of key enzyme genes involved in this pathway require further investigation. Quantitative real-time PCR (qRT-PCR) serves as a highly sensitive method for gene expression analysis, yet selecting appropriate reference genes is crucial to ensure the accuracy and reliability of results. METHODS: Ten candidate reference genes (18S, ACT13, RL15B, RL74, RLA2, RL182, RL291, EF1-α, EF1G, and YLS8) were evaluated for their expression stability in Sophora davidii seeds collected at five distinct developmental stages post-flowering, characterized by significant morphological changes. Five computational tools-GeNorm, NormFinder, BestKeeper, ΔCt, and RefFinder-were employed to comprehensively analyze the stability of these genes. RESULTS: Among the candidate genes, EF1G and RL291 exhibited the highest expression stability, whereas RL182 proved unsuitable as a reference gene. Validation experiments confirmed that normalization using stable reference genes (e.g., EF1G and RL291) yielded accurate quantification of target gene expression. DISCUSSION: This study identifies EF1G and RL291 as optimal reference genes for qRT-PCR analysis during Sophora davidii seed development, addressing a critical methodological gap in alkaloid biosynthesis research. These findings underscore the necessity of rigorous reference gene validation to ensure reliable gene expression data. The results advance our understanding of quinolizidine alkaloid biosynthesis and highlight the broader importance of reference gene selection in plant molecular studies.