Metabolic gene expression differences in emaciated and moderately conditioned horses: RNA-seq analysis of postmortem skeletal muscle

Abstract

Emaciated (E) horses maintain different metabolic profiles compared with moderately conditioned (M) horses, however, pathway knowledge in affected tissues is limited. When gene expressions across varied tissue types were assessed using RT-qPCR for gene activity related to glucose transport, protein synthesis, and fatty acid oxidation, no differential expression was identified. Euthanasia with sodium barbiturates has been cited to limit gene expression postmortem in RT-qPCR assays, leading to an identified need for alternative gene expression assessment postmortem. To investigate the hypothesis that metabolic-associated gene regulation differs between E and M, RNA sequencing (RNA-seq) assessment was employed. Eight horses (aged 20 ± 6 yr, mean ± SD) were split into 2 groups based on BCS assessed by 3 independent reviewers (1–9 scale, 1 = emaciated, 9 = obese): 4 M (BCS 4–5, 4.75 ± 0.43) and 4 E (BCS 1–2, 1.25 ± 1.43). After euthanasia via sodium barbiturate overdose, middle gluteal skeletal muscle samples were collected from each horse, flash frozen in liquid nitrogen and stored at −80°C until RNA-seq assessment. Differential expression analysis was performed using the limma-voom method due to its reliable inference in experiments with limited numbers of replicates (R 4.4.2). Principal component analysis and correlation analysis between samples revealed differences in gene expression profiles between emaciated and moderately conditioned samples. Differential expression analysis for emaciated versus moderately conditioned animals identified 817 differentially expressed genes (DEGs), with 243 upregulated and 574 downregulated (FDR adjusted P < 0.05). Furthermore, this data recapitulates RT-qPCR findings previously reported specific to glucose, protein, and fatty acid metabolism where no differential expression in RNA-seq results were identified. Functional enrichment analysis discovered overrepresented metabolic pathways in DEGs (FDR adjusted P < 0.05), including glucose metabolism, oxidative phosphorylation, and ATP synthesis, as well as mitochondrial functions. Issues identified while using RT-qPCR in postmortem samples could be modulated by using RNA-seq, and preliminary results indicate biologically relevant discoveries are possible even in cases where expectations for specific genes are not met. Due to unique, tissue-specific metabolic pathways, continued RNA-seq analysis of equine tissues, both in antemortem and postmortem samples, may allow a better understanding of metabolic differences between E and M horses.