Omics & gene networks

Long non-coding RNA dynamics in the liver transcriptome of dairy cattle with divergent feed efficiencies under defined dietary regimes

---

Abstract

Long non-coding RNAs (lncRNAs) are regulatory elements ≥200 nucleotides that regulate gene expression through multiple mechanisms, generating splice variants with distinct functions. Implicated in metabolic and immune pathways, they may influence complex traits including feed efficiency (FE) in livestock. The objective of this study was to identify and characterize lncRNAs associated with FE in Holstein and Jersey cows, focusing on differential expression (DE) and relationships with residual feed intake (RFI) across diets. RNA-Sequencing data (GEO: GSE92398) from liver biopsies of 9 Holstein (4 high-, 5 low-RFI [kg DMI/d]) and 10 Jersey cows (5 high-, 5 low-RFI), selected from a population of 200 and fed in a crossover design, were analyzed using FEELnc. Differential expression was assessed with DESeq2 (FDRC) or high-concentrate (HC) diets, and C vs. HC within either low- or high-RFI animals. Holstein cattle exhibited 4 DE lncRNAs under the C diet, 3 under HC, 35 in low-RFI, and 2 in high-RFI animals. Similarly, Jerseys showed 5 DE lncRNAs under C, 26 under HC, 115 in low-RFI, and 3 in high-RFI animals. A notable lncRNA, MSTRG.10056 (FDR=0.011), linked to fatty acid ω-oxidation and mapped to ENSBTAG00000000229, exhibited extensive alternative splicing and, together with MSTRG.4827 (FDR=0.013), is associated with protein-coding genes previously identified as DE in low-RFI animals. In Jersey, several lncRNAs were linked to protein-coding genes, including GIMAP8 and SLC1A2, with corresponding mRNA isoforms also DE. Linear regression showed a negative association between GIMAP8 expression and its lncRNA (β = -2.3) while a positive association was observed for SLC1A2 (β = 0.19). Overall, differences in the extent of DE observed in independently analyzed Holstein and Jersey datasets suggest breed-associated transcriptional patterns, with low-RFI animals showing greater network complexity and responsiveness to dietary treatments. No significant DE lncRNAs were shared across breeds in the independent analyses. However, considering all expressed lncRNAs in a network context, MSTRG.6142 was identified in both breeds through Weighted Gene Co-expression Network Analysis (WGCNA), with expression showing moderate correlations with RFI (Holstein: -0.31; Jersey: -0.34) and diet (Holstein: -0.54; Jersey: 0.33), and explaining 9.6-28.7% of variance (r²) in expression-trait correlations across Holstein and Jersey cattle. Weighted Gene Co-expression Network Analysis run per breed detected 12 and 11 modules in Holstein and Jersey, with hub lncRNAs showing moderate to high module membership (0.28-0.55). These preliminary findings highlight lncRNAs associated with FE‑related transcriptional patterns within each breed, which could serve as potential candidates for further validation to improve FE in dairy cattle.

Keywords: 2026

How to Cite:

Muhammad, F., Asselstine, V., Baes, C., Miglior, F., Schenkel, F. & Cánovas, Á., (2026) “Long non-coding RNA dynamics in the liver transcriptome of dairy cattle with divergent feed efficiencies under defined dietary regimes”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286490. doi: https://doi.org/10.31274/wcgalp.24027

Rights: 1