Enhancing Transcriptional Data Reliability in Fish Oogenesis Using cDNA-Based Normalization

Abstract

Transcriptomic analyses are widely used to elucidate the molecular mechanisms driving gametogenesis and reproduction in fish, yet their accuracy depends heavily on appropriate normalization of gene expression data. Conventional approaches that rely on single or multiple reference genes are problematic during teleost oogenesis, as profound structural and physiological remodeling of the ovary challenges the assumption that commonly used reference transcripts remain stable. In this study, we assessed by qPCR the transcriptional variability of four widely used reference genes (actb, ef-1, gapdh, and 18S rRNA) throughout the oogenic cycle of the thicklip grey mullet (Chelon labrosus), using geNorm and NormFinder analyses, and we additionally evaluated total cDNA concentration as an alternative normalization factor. To examine the performance and interpretive consequences of each normalization strategy, we compared expression patterns of key steroidogenic genes (star, cyp19a1a, and cyp11b) normalized by individual reference genes, combinations of reference genes, or total cDNA concentration. All evaluated reference genes displayed notable transcriptional variability across oogenesis, confirming their limited suitability as sole internal controls. In contrast, normalization approaches integrating multiple reference genes and/or total cDNA concentration consistently provided greater stability and more reliable biological interpretation. These results support a refined and more robust normalization framework for transcriptional analyses in fish ovaries, particularly during stages of extensive tissue remodeling. Our findings demonstrate cDNA-based normalization is straightforward, rapid, and easy to implement across laboratories, providing a practical alternative for achieving accurate, reproducible transcript quantification in fish ovary studies.