Integrated transcriptomic analysis of milk-derived RNA associated with divergent milk yield phenotypes in buffalo

Buffaloes are the second-largest milk-producing animals, and their milk possesses significant nutritional and health benefits. However, the relatively low milk yield severely limited the development of buffalo dairy products. Here, to explore molecular features associated with divergent milk yield phenotypes, we performed an integrated transcriptomic analysis of milk-derived RNA from high-yield (HY, 3804.16 ± 486.23 kg/lactation) and low-yield (LY, 1928.47 ± 110.28 kg/lactation) Murrah buffaloes during mid-lactation. We identified 17,735 mRNAs, 2,641 long non-coding RNAs (lncRNAs), 6,668 circular RNAs (circRNAs), and 578 microRNAs (miRNAs). Comparative analyses revealed that buffaloes with higher milk yield exhibited coordinated upregulation of genes involved in oxidative phosphorylation, glucose metabolism, cellular metabolism, and protein synthesis, reflecting an enhanced cellular bioenergetic and biosynthetic state in milk-associated cells. In contrast, transcripts associated with signal transduction, transporter activity, and apoptosis-related processes showed reduced expression in the HY group. Integration of mRNA and non-coding RNA profiles enabled the construction of putative competing endogenous RNA (ceRNA) networks, highlighting SNX25 and DHPR as candidate genes potentially associated with milk yield-related cellular pathways. These findings suggest that high-yielding buffaloes exhibit transcriptional features consistent with enhanced metabolic and biosynthetic activity in milk-associated cells. This study provides an exploratory multi-layer transcriptomic framework describing molecular features associated with high milk yield phenotypes in buffaloes and offers candidate targets for future functional validation and genetic improvement strategies.