Identification and Expression Profiling of Novel microRNAs in Pig Fetal Skeletal Muscle

MicroRNAs (miRNAs) are a class of noncoding RNAs known to post-transcriptionally regulate gene expression through binding with target mRNAs, ultimately affecting a multitude of biological processes and phenotypes. It has been documented that miRNAs influence skeletal muscle development; however lack of miRNA annotation in pigs hinders understanding of molecular mechanisms underlying this process. We sought to identify novel miRNAs in fetal longissimus dorsi (LD) muscle and compare expression of these miRNAs at 41 days gestation (dg) and 70 dg (n=3 per stage), representing primary and secondary fetal myogenesis. Total RNA was isolated from LD samples of fetuses obtained from Yorkshire x Landrace gilts. Small-RNA sequencing was performed on the Illumina HiSeq 4000 platform, generating 30-60 million 1x50 reads per sample. High-quality reads were aligned to the S. scrofa reference genome (v11.1), and mapping and prediction of novel miRNAs was performed using miRDeep2. Predicted miRNAs with significant randfold p-values, miRDeep2 scores >7, and total read counts per million ≥1 for each sample were retained. Annotated human miRNAs with ≤2 mismatches with common and stage-specific novel miRNAs were found using miRBase. Differential expression analysis was performed on novel miRNAs using DESeq2. TargetScan was used to find conserved targets of human miRNAs with sequence identity to differentially expressed (DE) pig miRNAs. At 41 dg and 70 dg, 83 and 73 novel miRNAs were predicted in at least two samples, respectively. Of these, 59 were common to both stages. We identified 10 DE miRNAs (|log2 fold change|>1 and adj.p