Transcriptional profiling identifies differential expression of long non-coding RNAs in Jo-1 associated and inclusion body myositis

Abstract

Myositis is characterised by muscle inflammation and weakness. Although generally thought to be driven by a systemic autoimmune response, increasing evidence suggests that intrinsic changes in the muscle might also contribute to the pathogenesis. Long non-coding RNAs (lncRNAs) are a family of novel genes that regulate gene transcription and translation. To determine the potential role of lncRNAs, we employed next generation sequencing to examine the transcriptome in muscle biopsies obtained from two histologically distinct patient populations, inclusion body myositis (IBM) and anti-Jo-1-associated myositis (Jo-1). 1287 mRNAs and 1068 mRNAs were differentially expressed in the muscle from Jo-1 and IBM patients, respectively. Pathway analysis showed the top canonical pathway in both Jo-1 and IBM was oxidative phosphorylation and mitochondrial dysfunction. We identified 731 known and 325 novel lncRNAs in the muscles biopsies. Comparison with controls showed 55 and 46 lncRNAs were differentially expressed in IBM and Jo-1 myositis, respectively. Of these, 16 lncRNAs were differentially expressed in both IBM and Jo-1 myositis and included upregulated H19, lncMyoD and MALAT1. Given that these are known to regulate muscle proliferation and differentiation, we speculate that changes in lncRNAs might contribute to the phenotypic changes in Jo-1 and IBM myositis.

Figure 1

Muscle biopsy analysis. (a) Description of bioinformatics analysis pathway, (b) Muscle histology where A-C shows normal muscle tissue from one of the controls. D-F shows classical features of IBM in one of the biopsies. G–I are sections from a muscle biopsy from one of the patients with Jo-1 myositis. A,D and G were stained with HE and B,E and H with modified Gomori trichrome. C,F and I were immunolabelled for β₂-microglobulin. Inset in F shows inclusions immunolabelled for ubiquitin. The scale bar represents 50 μm, (c) Principle component analysis (PCA) of the mRNA data from control, IBM and Jo-1 myositis, (d) Non-supervised hierarchical cluster analysis of the mRNA data from control, IBM and Jo-1 myositis.

Figure 2

Analysis of mRNA and lncRNA expression data in control, IBM and Jo-1 myositis. (a) Venn diagram showing overlap between differentially expressed mRNAs in IBM and Jo-1 myositis, (b) Changes in levels of β₂-microglobulin (B2M), CD74 and IGF2 mRNA expression were confirmed by qRT-PCR. The data are presented as the mean ± SEM of 4–5 patients where *p < 0.05 and ***p < 0.001 (Mann-Whitney U-test). (c) Pathway analysis of differentially expressed mRNAs in IBM and Jo-1 myositis, (d) Classification of Gencode and novel lncRNA in skeletal muscle and, (e) List of most highly expressed lncRNA in control skeletal muscle.

Figure 3

Characterisation of differentially expressed lncRNAs in control, IBM and Jo-1 myositis. (a) Venn diagram showing overlap between differentially expressed lncRNAs in IBM and Jo-1 myositis, (b) Classification of differentially expressed lncRNAs in IBM and Jo-1 myositis, (c) Correlation between differentially expressed lncRNAs and that of the nearest protein coding mRNA, (d) Correlation between differential lncRNA expression in IBM and Jo-1 myositis, (e) PhastCons analysis of the conservation of the differentially expressed lncRNA species in human and mouse cells compared with the exon, intronic and untranslated regions (UTRs) of protein coding genes and (f) Pie chart showing the % distribution of repeat sequences in the various sub-populations of lncRNAs across all 4 cell types with SINES = short interspersed nuclear elements, LINES = long interspersed nuclear elements and LTR = long terminal repeat.

Figure 4

Differential expression of H19, MALAT1, NEAT1 and hLncMYoD in control, IBM and Jo-1 myositis. (a) View from the Integrated Genome Viewer (IGV) of aligned sequence data for H19, MALAT1, NEAT1 and hLncMYoD. Red and blue blocks represent reads aligned to the positive and negative strand of the DNA, respectively. The dark gray histogram represents the sum of the aligned sequencing reads along the genome, with the scale stated on the top right corner, (b) Changes in levels of H19, MALAT1, NEAT1 and hLncMYoD were confirmed using qRT-PCR the data are the mean ± SEM of 5 patients where *p < 0.05 (Mann-Whitney U-test).

Figure 5

Differential expression of novel lncRNAs in control, IBM and Jo-1 myositis. View from the Integrated Genome Viewer (IGV) of aligned sequence data for novel lncRNA that are up-regulated (a,c,e) and down-regulated (b,d,f) in both IBM and Jo-1 myositis (a,b), IBM alone (c,d) and Jo-1 alone (e,f).