Transcriptome-informed selection and validation of reference genes for RT-qPCR normalization in Cuphea hookeriana across tissues and under low-temperature stress

Reverse transcription quantitative PCR (RT-qPCR) is widely used to quantify gene expression, but reliable normalization requires reference genes with stable expression in the species, tissues, and experimental conditions being examined. Cuphea hookeriana is an ornamental shrub whose wider landscape application is limited by poor cold tolerance. However, no validated reference genes have been reported for RT-qPCR normalization in this species. In this study, twelve candidate reference genes were selected from a C. hookeriana leaf transcriptome generated during a 5 °C low-temperature time course and deposited in the NCBI Sequence Read Archive under BioProject accession number PRJNA1117973. The candidates comprised elongation factor 1-alpha ( EF1α ), DnaJ heat shock protein ( DNAJ ), cyclophilin ( CYP ), glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ), actin ( ACT ), 18 S ribosomal RNA ( 18 S rRNA ), eukaryotic initiation factor 2 ( EIF2 ), tubulin ( TUB ), protein phosphatase 2 A ( PP2A ), acyl carrier protein ( ACP ), ubiquitin ( UBQ ), and ubiquitin-conjugating enzyme ( UBC ). Their expression stability was evaluated in roots, stems, leaves, and flowers and in leaves exposed to 5 °C for 0, 12, 24, 36, and 48 h. The amplification efficiencies ranged from 92.41% to 104.50%, with correlation coefficients of 0.992–0.999. geNorm pairwise-variation analysis indicated that three reference genes were required for the different-tissue dataset, whereas two were sufficient for the low-temperature-treatment and pooled datasets. Accordingly, the validation analysis used ACP , UBC , and EF1α for tissues and 18 S rRNA and EIF2 for low-temperature-treated samples. One-way ANOVA of ΔCq values showed significant differences among tissues for CSLD and among treatment times for MYB11 . Normalization using the least stable single genes, TUB or UBQ , produced group-dependent relative-expression estimates that differed significantly from those obtained using the geNorm-recommended combinations. This study identifies condition-specific reference-gene combinations for RT-qPCR analysis in C. hookeriana . The geometric mean of ACP , UBC , and EF1α is recommended for comparisons among tissues; 18 S rRNA and EIF2 are recommended for samples collected during the 5 °C treatment; and EIF2 and ACT are recommended for the pooled dataset containing all tissue and low-temperature-treatment samples.