| Porcine skeletal muscle is a highly heterogeneous tissue, of which longissimus doris muscle (LDM) and psoas major muscle (PMM), as two skeletal muslces with high economic value and distinct physiological features, have been the emphasis and difficulty of porcine molecular breeding and improvement. Long non-coding RNA (lncRNA), especially the circular RNA (circRNA), has been emerging as the important new members of regulatory non-coding RNA and become a hot field for worldwide biology researchers. However, the identification and functional study of porcine lncRNA (including circRNA), in contrast to human and model animals, have been considerably lagged. In this study, we reconstructed the transcriptome of porcine LDM and PMM using RNA-seq approach, and identified and analyzed the genomic features of mRNA, lncRNA and circRNA for each sequencing library using our in-house bioinformatic pipeline. The main points are indicated as follows:The myofiber cross-sectional area of LDM (2.82 × 103 μm2) was significantly larger than that of PMM (1.91× 103 μm2)(P< 0.05). The content of type-Ⅱb fiber (fast-switch glycolytic fiber, MyHC4 gene) in PMM wassignificantly lower than that of LDM (P< 0.001), while The content of type-I fiber (fast-switch oxidative fiber, MyHC7 gene) in PMM were significantly higher than its counterpart in LDM (P< 0.001). The expression level of LDHA, a key enzyme of glycolytic metabolism in muscle, wassignificant lower in PMM than that of LDM (P< 0.001). In contrast, the expression level of PDH1, a key enzyme of oxidative metabolism in muscle, was significant higher in PMM than that of LDM (P< 0.05). In quantitative real-time PCR (Q-PCR) analysis of mtDNA copy number, the mtDNA copy number of PMM (1.13 × 103 per cell) was significant higher than that of the LDM (5.03×102 per cell)(P< 0.05). We performed largescale cDNA sequencing experiments between LDM and PMM fromthree biological repeats using the Illumina paired-end RNA-seq approach, resulting in a total of ～465 M (million) paired-end reads of the 100 nucleotide (nt) length. This yielded 46.50 giga bases (Gb) of sequence, and about 96.30% of the reads passed initial quality thresholds-(Illumila protocol) and were deemed as clean reads for the subsequent analyses.Using Cufflinks assembler, we obtained 101,800 transcripts for each sequencing library on average, which covered 64.11%(35,054/54,675) and 85.86%(26,262/30,586) of Refseq and Ensembl transcripts, respectively. Using Scripture assembler, we obtained 50,886 transcripts for each sequencing library on average, which covered 37.9%(20,721/54,675) and 67.02%(20,499/30,586) of Refseq and Ensembl transcripts, respectively.In the sequencing libraries of LDM and PMM, we identified 40,148 reliable mRNA transcripts, including 15,106 known and 25,042 novel mRNA transripts. We found that a total of 40,060 mRNA transcripts were co-expressed in LDM and PMM, while 58 and 30 mRNA transcripts were specifically expressed in LDM and PMM, respectively.In this study, we detected 21,936 and 22,954 alternative splicing events in LDM and PMM, respectively. We also detected 19,514 SNVs and 2,131 INDELs in all six sequencing libraries, of which 9,391 SNVs and 1,023 INDELs were found in LDM, while 10,123 SNVs and 1,108 INDELs were found in PMM.In the expression analysis,217 differentially expressed (DE) mRNA were identified between LDM and PMM. The 73 LDM-enriched mRNA were found to be mainly related to ’Glycolysis’,’Glycolysis/Gluconeogenesis’and’Glucose catabolic process’, whereas the 144 PMM-enriched mRNA were mainly linked to ’ECM-receptor interaction’,’Skeletal system development’and’Muscle organ development’.A total of 326 lncRNA were identified in the six sequencing libraries, including 297 Intergenic,23 Intronic and 6 Antisense lncRNA. A total of 317 (97.24%) lncRNA were found to be co-expressed in LDM and PMM, while 5 (1.53%) and 4 (1.23%) lncRNA were observed to be specifically expressed in LDM and PMM, respectively.The lncRNA identified in this study have similar genomic features with the known lincRNA. However, in contrast to the coding mRNA genes, The lncRNA identified in this study have much lower exon number, shorter transcript length, shorter ORF length and lower protein coding potency.A total of 12 lncRNA were identified to be differentially expressed (DE) between LDM and PMM, including 1 Antisense and 11 Intergenic lncRNA. The functional enrichment analysis with correlated coding genes showed that these DE lncRNA were significantly related to’Glycolysis’and’Oxidative phosphorylation’of skeletal muscle.A total of 56 circRNA were identified in this study, of which 30 (57.57%) were located in Scaffold sequence of pig genome,6 (10.71%) were located in chromesome 13,14 (25.00%) were located in chromesome 14,3 (5.36%) "were located in chromesome 15,2 (3.57%) were located in chromesome 18 and 1 (1.79%) were located in chromesome x. Sequence analysis showed that there were large amount of miRNA target sites in the circRNA. |
PDF Full Text Request