Pak, Katherine

ABSTRACT Autoantibody internalization has been implicated in autoimmune disease pathogenesis, yet its mechanisms, and generality across different diseases, cell types, and affected tissues remain poorly defined. Using bulk RNA sequencing, we identified reproducible, autoantibody-specific transcriptomic signatures consistent with autoantigen dysfunction in muscle biopsies from patients with anti-Mi2 dermatomyositis and anti-PM/Scl scleromyositis across independent cohorts. Electroporation of purified patient IgG into primary cultures of healthy cells was sufficient to induce the corresponding transcriptomic programs in vitro . Direct immunofluorescence demonstrated immunoglobulin internalization into subcellular compartments matching the localization of the autoantigen in different affected tissues. Spatial transcriptomic analyses revealed that antibody-secreting cells translocated cytoplasmic material (i.e., immunoglobulin RNA) into adjacent affected cells expressing autoantibody-specific transcripts. The disease-specific transcripts were present not only in muscle fibers, but also in other cells, including macrophages, endothelial cells, and fibroblasts. Autoantibody-induced transcriptomic programs were associated with cell damage and autoantibody-specific reactive inflammatory programs, including activation of type I interferon and TGF-ß1 signaling in anti-Mi2 dermatomyositis and activation of type II interferon in anti-PM/Scl scleromyositis. Antibody internalization was also observed in different tissues from patients with other autoimmune diseases, including anti-U1RNP mixed connective tissue disease, anti-Ku overlap syndrome, and anti-Scl70 systemic sclerosis. Together, these findings establish autoantibody internalization as a shared pathogenic mechanism across diverse autoimmune diseases, providing a unifying framework for conditions driven by autoantibodies against intracellular antigens.

ABSTRACT Objectives T‐cell large granular lymphocytic leukemia (T‐LGLL), reported in up to 58% of inclusion body myositis (IBM) patients, is a rare leukemia of cytotoxic or less commonly helper T cells. The range of myopathies in T‐LGLL and the impact of coexisting T‐LGLL in IBM are not well understood. Our objectives are to investigate the spectrum of myopathies in patients with T‐LGLL and compare the clinical, serological, pathological, and transcriptomic features of IBM patients with and without T‐LGLL. Methods We reviewed two Mayo Clinic cohorts: (1) T‐LGLL patients evaluated for myopathies (2003–2018), and (2) IBM patients tested for T‐LGLL via T‐cell receptor gene rearrangement or flow cytometry (2016–2022). We also compared transcriptomic profiles of IBM muscle biopsies with and without T‐LGLL. Results Of 447 T‐LGLL patients, 13 (2.9%) had myopathies, IBM being the most common ( n = 7). Of 43 IBM patients, 9 (20.9%) had T‐LGLL, with 5 diagnosed prior to the onset of weakness. Clinical and pathological features were largely similar between IBM patients with and without T‐LGLL, except for milder finger flexor weakness and more frequent neutropenia (55.6% vs. 0%, p < 0.001) in the T‐LGLL group. However, transcriptomic analysis of muscle tissues revealed a more immunologically active profile, with upregulation of T‐cell and macrophage markers, elevated levels of IFN‐γ gene and IFN‐γ–inducible genes, heightened cytokine activity, and enhanced immunoglobulin production in IBM patients with T‐LGLL. Interpretation IBM is the most common myopathy in T‐LGLL patients. While clinico‐sero‐pathological features were largely similar, transcriptomic differences suggest IBM with T‐LGLL may represent a distinct, more immune‐driven subtype.

Abstract Objectives Granulomatous myositis (GM) is defined by focal collections of activated macrophages that fuse to form multinucleated cells that aggregate into granulomas within skeletal muscle. This study aimed to elucidate the pathophysiology of GM by defining its specific transcriptomic profile. Methods Bulk RNA sequencing was performed on 722 muscle biopsies, including 38 from patients with GM, other myopathies and healthy comparators. Spatial transcriptomics and immunohistochemistry assays were used to identify the location of specific transcripts and proteins within the tissue. Results We conducted a systematic analysis of the inflammatory pathways implicated in GM. This analysis revealed that GM muscle biopsies are characterized by elevated levels of IFNγ, IFNγ-inducible genes and proinflammatory cytokine genes including IL1B, TNF and TGFB1. Additionally, we identified a specific gene signature in GM, comprising 1293 genes that were significantly upregulated and 256 significantly downregulated. These genes were strongly correlated with transcriptomic markers of disease activity. Using this gene signature, support vector machine models accurately identified GM, achieving an area under the curve of 99.6% (99.0–99.9%) and an accuracy of 98.6% (98.2–98.9%). Among the identified genes, CHIT1 emerged as the most significantly upregulated gene, with its expression strongly correlating with disease severity. Furthermore, CHIT1 was detected at both the RNA and protein levels in granulomas and giant cells. Conclusion GM is transcriptomically characterized by a strong IFNγ signature and overexpression of proinflammatory cytokines, including IL1B, TNF and TGFB1. Additionally, it exhibits a unique transcriptomic profile, including CHIT1, which correlates with disease activity.

Abstract Objective Myositis is a heterogeneous family of inflammatory myopathies. We sought to define the differential expression of cytokines, cytokine receptors and immune checkpoint genes in muscle biopsies from patients with different forms of myositis in order to characterize patterns of inflammation in each. Methods Bulk RNA-sequencing was performed on muscle biopsy samples from 669 patients, including 105 with DM, 80 with immune-mediated necrotizing myopathy (IMNM), 65 with anti-synthetase syndrome, 53 with IBM, 19 with anti-PM/Scl myositis, 310 with other inflammatory or genetic myopathies and 37 controls with normal tissue (NT). Myositis clinical groups and autoantibody subgroups were analysed separately. Expression data were analysed for 338 genes encoding cytokines, cytokine receptors and immune checkpoints. Myositis group-specific genes were identified from this list by finding genes that were specifically differentially expressed in one group compared with all samples and compared with NT (α &lt; 0.001). Results IBM patients had the most differentially overexpressed genes (71) among all clinical groups, including 37 that were IBM-specific. Among the top genes were several involved in type 1 inflammation, including CCL5, CXCR3, CCR5, CXCL9 and IFNG. Anti-Jo1 and anti-PM/Scl patients exhibited differential overexpression of a similar set of genes, while DM patients exhibited differential overexpression of a different set of genes involved in type 1 inflammation. IMNM patients had the least number of differentially overexpressed genes with no predominant inflammatory pattern. Conclusion Each myositis clinical group and autoantibody subgroup had differentially overexpressed inflammatory mediators, including a strong type 1 inflammatory gene signature in IBM.