Flavonoid synthases (FNSs) are key enzymes catalyzing the conversion of flavanones to flavonoids, yet their functions in citrus remain functionally uncharacterized. In this study, we identified three FNSII genes in the citrus genome. Phylogenetic analysis revealed that citrus FNSII genes share the closest evolutionary distance with apple FNSII genes. Chromosomal localization demonstrated that the three FNSII genes are distributed across two out of nine chromosomes. Gene structure analysis indicated that the majority of motifs within these three FNSII genes are highly conserved. We cloned a gene called CitFNSII-1 from citrus. Transient overexpression of CitFNSII-1 in citrus leaves significantly increased flavonoid content, while simultaneous virus-induced silencing of CitFNSII-1 led to synchronously and significantly reduced gene expression levels and flavonoid content in citrus seedlings. Through the Agrobacterium rhizogenes-mediated genetic transformation system, overexpression of CitFNSII-1 was found to markedly enhance flavonoid accumulation in hairy roots, whereas knockout of CitFNSII-1 resulted in a significant decrease in flavonoid content in hairy roots. Further experiments verified an interaction between CitFNSII-1 and the Chalcone isomerase-1 (CHI-1) protein. The results demonstrated that the flavonoid accumulation patterns of CHI-1 and CitFNSII-1 are highly similar. In conclusion, this study advances the understanding of the flavonoid biosynthesis pathway in citrus and provides a theoretical foundation for molecular breeding strategies in citrus.

(1) Reverse transcription quantitative real-time PCR (RT-qPCR) requires reliable reference genes for accurate data normalization; however, optimal reference genes for the economically and ecologically valuable timber species Phoebe zhennan remain uncharacterized; (2) Here, we selected nine candidate reference genes derived from full-length transcriptome sequencing to evaluate their expression stability across abiotic (drought) and biotic (Colletotrichum fructicola infection) stresses. Transcript abundance was analyzed via RT-qPCR using four distinct algorithms (Delta Ct, geNorm, NormFinder, and BestKeeper), with RefFinder used to reconcile analytical discrepancies and generate a definitive consensus ranking; (3) Our analysis showed that expression stability is highly context-dependent: CYP20-1 and HSP70-1 were the most stable reference genes under drought stress, whereas Actin-101 and Actin constituted the optimal pair under disease stress. For cross-condition assessments, Actin-101 and β-Tubulin served as the most reliable baseline combination. Subsequent empirical validation quantifying stress-responsive transcripts demonstrated a significant positive correlation between RT-qPCR relative expression and corresponding RNA-seq data (drought: R = 0.80; disease: R = 0.76); (4) This study identifies and validates the first set of reference genes for P. zhennan, providing a foundation for accurate gene expression analysis in this species, which is crucial for understanding its response to environmental stresses.