The development of secondary hair follicles in cashmere goats directly influences the yield and quality of cashmere, and hair follicle morphogenesis is a complex biological process involving multiple signaling pathways and regulatory factors. In recent years, the regulatory role of long non-coding RNAs in hair follicle development has gradually been uncovered. Based on a previously established transcriptome database of embryonic skin tissue from cashmere goats, this study ultimately identified lncRNA MRPS28, which exhibits significant differential expression during the morphogenesis of secondary hair follicles. Our results indicate that lncRNA MRPS28 is a non-coding RNA transcribed from the intron region of the MRPS28 gene. Additionally, it can inhibit the proliferation and migration of dermal fibroblasts while promoting their apoptosis. Further mechanistic studies have shown that lncRNA MRPS28 acts as a competing endogenous RNA, sponging chi-miR-145-5p, thereby relieving its inhibitory effect on the target gene NUAK1. This, in turn, inhibits the formation of dermal papilla structures and ultimately affects the morphogenesis of secondary hair follicles during the embryonic period. This study provides a novel perspective on dissecting the regulatory network of hair follicle development in cashmere goats and offers potential theoretical insights for breeding new strains of cashmere goats.

Non-coding negative regulation is an important but incompletely resolved component of gene-expression control in livestock. We conducted a narrative review using PubMed, Web of Science Core Collection, and Scopus, prioritizing primary studies in pigs, cattle, sheep, goats, and poultry and using model-organism studies only for directly relevant mechanistic or methodological evidence. We distinguish sequence-defined silencers, context-dependent enhancer-silencer elements, functional microRNA-binding sites, long non-coding RNA-associated regulation, and candidate repressive chromatin regions, and evaluate representative livestock studies using a four-tier framework comprising candidate annotation, molecular functional evidence, endogenous causal evidence, and animal-level validation. Current evidence is strongest for a small number of loci, whereas many of the livestock candidates reviewed here are supported primarily by association, chromatin state, or in vitro assays. We further assess how multi-omics data can prioritize candidates and assign target genes, while emphasizing that reporter activity, repressive chromatin marks, and population association do not establish endogenous causality. Progress will depend on locus-specific perturbation, rescue experiments, relevant animal phenotyping, and inheritance testing. This evidence-centered framework provides a focused basis for interpreting negative regulatory variation in livestock and identifying robust targets for future genetic improvement.