RNA-Seq of Dysferlinopathy patients reveals differential gene for Limb-Girdle and Miyoshi subtypes

[Speaker] John Hoon Rim:1
[Co-author] Ha Young Shin:2, Kyeong-Jee Cho:1, Heon Yung Gee:1
1:Department of Pharmacology, Yonsei University College of Medicine, Korea, 2:Department of Neurology, Yonsei University College of Medicine, Severance Hospital, Korea

Background: Recessively inherited mutations in the dysferlin gene (DYSF) are known to cause a group of muscular dystrophies called dysferlinopathies. Among dysferlinopathies, limb-girdle muscular dystrophy type 2B (LGMD2B) is characterized by progressive weakness of the proximal lower limb girdle muscles, whereas Miyoshi myopathy (MM) mostly affects the distal muscle groups of the limb girdle. Since genetic mutational spectrums of both subtypes overlap and appear not to be discriminatory, we applied the transcriptomic analysis to investigate differentially expressed genes responsible for two different subtypes.

Methods: RNA sequencing (RNA-seq) from vastus lateralis muscle biopsy samples in 6 dysferlinopathy patients with confirmed DYSF mutations (i.e. 3 LGMD2B and 3 MM patients) and 3 normal control individuals were performed by Macrogen (Seoul, Korea). RNA-seq data were mapped and analyzed by CLC Genomics Workbench v.9.0.1 software. Gene ontology (GO) enrichment analyses were carried out using DAVID software, and further visualized using REViGO program. Protein-protein interaction (PPI) analysis was performed by in silico algorithms including STRING and Gene Mania. Gene prioritization analysis was also performed using ToppGene Suite and Endeavour.

Results: A total of 22 and 12 genes were exclusively up-regulated in LGMD2B and MM compared to control, respectively. While 42 genes were commonly down-regulated in both subtypes, a total of 18 and 119 genes were exclusively down-regulated in LGMD2B and MM compared to control, respectively. For down-regulated genes, distribution of major representative GO terms in LGMD2B and MM were different; [response to organic substrate] in LGMD2B whereas [cellular carbohydrate metabolism] in MM was the most enriched GO term, respectively. Furthermore, PPI analysis for down-regulated genes in MM using two algorithms commonly identified parvin-beta encoded by PARVB as interacting with dysferlin. When down-regulated genes in MM were additionally analyzed using gene prioritization tools, PARVB was one of the top candidate genes in association with DYSF.

Conclusions: We expect that PARVB as well as other genes revealed in study would cause distinctive pathophysiological environments for MM and LGMD2B. Our data might serve as the roadmap for the discovery of novel gene responsible for classifying subtypes in dysferlinopathy using transcriptomic analysis.

Advanced Search