Myositis Autoantigen Expression Correlates With Muscle Regeneration but Not Autoantibody Specificity

Abstract

Objective: Although more than a dozen myositis-specific autoantibodies (MSAs) have been identified, most patients with myositis are positive for a single MSA. The specific overexpression of a given myositis autoantigen in myositis muscle has been proposed as initiating and/or propagating autoimmunity against that particular autoantigen. The present study was undertaken to test this hypothesis.

Methods: In order to quantify autoantigen RNA expression, RNA sequencing was performed on muscle biopsy samples from control subjects, MSA-positive patients with myositis, regenerating mouse muscles, and cultured human muscle cells.

Results: Muscle biopsy samples were available from 20 control subjects and 106 patients with autoantibodies recognizing hydroxymethylglutaryl-coenzyme A reductase (n = 40), signal recognition particles (n = 9), Jo-1 (n = 18), nuclear matrix protein 2 (n = 12), Mi-2 (n = 11), transcription intermediary factor 1γ (n = 11), or melanoma differentiation-associated protein 5 (n = 5). The increased expression of a given autoantigen in myositis muscle was not associated with autoantibodies recognizing that autoantigen (all q > 0.05). In biopsy specimens from both myositis muscle and regenerating mouse muscles, autoantigen expression correlated directly with the expression of muscle regeneration markers and correlated inversely with the expression of genes encoding mature muscle proteins. Myositis autoantigens were also expressed at high levels in cultured human muscle cells.

Conclusion: Most myositis autoantigens are highly expressed during muscle regeneration, which may relate to the propagation of autoimmunity. However, factors other than overexpression of specific autoantigens are likely to govern the development of unique autoantibodies in individual patients with myositis.

Figure 1:

Mean RNA expression levels of the different myositis autoantigens (columns) in muscle biopsies from control patients and patients from the different autoantibody groups (rows). For example, SRP54 (first column) is expressed at uniformly high levels in all autoantibody groups and not higher in anti-SRP positive patients (cell with a white circle in the first column). Of note, the anti-MDA5 autoantigen (IFIH1) is highly expressed in patients with dermatomyositis (anti-Mi2, -NPX2, -TIF1γ, and -MDA5), moderately expressed in patients with anti-Jo1 autoantibodies and expressed at low levels in immune-mediated necrotizing myositis (anti-SRP and anti-HMGCR). White circles highlight the expression levels of each autoantigen in its corresponding autoantibody group. TPM: Transcripts Per Kilobase Million.

Figure 2:

Matrices show correlation coefficients for the RNA expression of myositis autoantigens (rows) versus the expression of genes found in T cells (CD3E, CD4, and CD8A), macrophages (CD14), regenerating muscle cells (NCAM1, MYOG, MYOD1, PAX7, MYH3, MYH8), and mature muscle cells (ACTA1, MYH1, and MYH2)(columns) in (a) myositis muscle biopsies and (b) regenerating mouse muscles (all the time points were pooled for the analysis). Red denotes a positive correlation and blue denotes a negative correlation.

Figure 3:

Evolution of the RNA expression levels (log₂[TPM+1]) of the different myositis autoantigens (rows) as human skeletal muscle myoblasts (HSMM) differentiate into myotubes and during the regeneration of mouse muscle following injury with cardiotoxin (CTX). In the HSMM model, proliferating myoblasts are placed in differentiation media at day 0 and allowed to differentiate into myotubes over the next 6 days. In the mouse injury model, the tibialis anterior muscle is injected with CTX at day 0 and the muscle is allowed to regenerate for as long as 28 days. Vertical lines indicate the 95% confidence interval for each value. CTX day 0 corresponds to the contralateral (uninjured) tibialis anterior muscle.