Transcriptomic Analysis of Inflammatory Cardiomyopathy Identifies Molecular Signatures of Disease and Informs in silico Prediction of a Network-Based Rationale for Therapy

Kamayani Singh, Hai Fang, Graham Davies, Benjamin Wright, Helen Lockstone, Richard O. Williams, Daniela Ciháková, Julian C. Knight, Shoumo Bhattacharya

Research output: Contribution to journalArticlepeer-review

Abstract

Inflammatory cardiomyopathy covers a group of diseases characterized by inflammation and dysfunction of the heart muscle. The immunosuppressive agents such as prednisolone, azathioprine and cyclosporine are modestly effective treatments, but a molecular rationale underpinning such therapy or the development of new therapeutic strategies is lacking. We aimed to develop a network-based approach to identify therapeutic targets for inflammatory cardiomyopathy from the evolving myocardial transcriptome in a mouse model of the disease. We performed bulk RNA sequencing of hearts at early, mid and late time points from mice with experimental autoimmune myocarditis. We identified a cascade of pathway-level events involving early activation of cytokine and chemokine-signaling pathways that precede leucocyte infiltration and are followed by innate immune, antigen-presentation, complement and cell-adhesion pathway activation. We integrated these pathway events into a network-like representation from which we further identified a 50-gene subnetwork that is predominantly induced during the course of autoimmune myocardial inflammation. We developed a combinatorial attack strategy where we quantify network tolerance to combinatorial node removal to determine target-specific therapeutic potential. We find that combinatorial attack of Traf2, Nfkb1, Rac1, and Vav1 disconnects 80% of nodes from the largest network component. Two of these nodes, Nfkb1 and Rac1, are directly targeted by prednisolone and azathioprine respectively, supporting the idea that the methodology developed here can identify valid therapeutic targets. Whereas Nfkb1 and Rac1 removal disconnects 56% of nodes, we show that additional removal of Btk and Pik3cd causes 72% node disconnection. In conclusion, transcriptome profiling, pathway integration, and network identification of autoimmune myocardial inflammation provide a molecular signature applicable to the diagnosis of inflammatory cardiomyopathy. Combinatorial attack provides a rationale for immunosuppressive therapy of inflammatory cardiomyopathy and provides an in silico prediction that the approved therapeutics, ibrutinib and idelalisib targeting Btk and Pik3cd respectively, could potentially be re-purposed as adjuncts to immunosuppression.

Original languageEnglish (US)
Article number640837
JournalFrontiers in immunology
Volume12
DOIs
StatePublished - Mar 5 2021

Keywords

  • autoimmune
  • diagnosis
  • myocarditis
  • network
  • therapy
  • transcriptome

ASJC Scopus subject areas

  • Immunology and Allergy
  • Immunology

Fingerprint

Dive into the research topics of 'Transcriptomic Analysis of Inflammatory Cardiomyopathy Identifies Molecular Signatures of Disease and Informs in silico Prediction of a Network-Based Rationale for Therapy'. Together they form a unique fingerprint.

Cite this