Next Article in Journal
CPF Induces GC2spd Cell Injury via ROS/AKT/Efcab6 Pathway
Previous Article in Journal
Cannabidiol Effects on Depressive-like Behavior and Neuroinflammation in Female Rats Exposed to High-Fat Diet and Unpredictable Chronic Mild Stress
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Cells
Volume 14
Issue 12
10.3390/cells14120939
cells-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Muscle Spatial Transcriptomic Reveals Heterogeneous Profiles in Juvenile Dermatomyositis and Persistence of Abnormal Signature After Remission

by
Margot Tragin
1,
Séverine A. Degrelle
2,
Baptiste Periou
3,
Brigitte Bader-Meunier
4,5,
Christine Barnerias
5,6,
Christine Bodemer
5,7,
Isabelle Desguerre
5,6,
Mathieu Paul Rodero
8,†,
François Jérôme Authier
3,6,9,† and
Cyril Gitiaux
5,6,8,10,*,†
1
Bioinformatics Facilities Imagine Institute, SFR Necker, 75015 Paris, France
2
Inovarion, 75005 Paris, France
3
Service d’Histologie, AP-HP, Hôpital Henri Mondor, 4010 Créteil, France
4
Department of Pediatric Immunology-Hematology and Rheumatology, Necker Enfants Malades Hospital, AP-HP Centre—Paris University, 75015 Paris, France
5
Hôpital Necker-Enfants Malades, 149 Rue de Sèvres, 75015 Paris, France
6
Reference Centre for Neuromuscular Diseases “Nord-Est-Ile de France”, FILNEMUS, 75015 Paris, France
7
Department of Dermatology, Necker Enfants Malades Hospital, AP-HP Centre—Paris University, 75015 Paris, France
8
Laboratoire de Chimie et de Biochimie Pharmacologiques et Toxicologiques, Université Paris Cité, 75006 Paris, France
9
INSERM, IMRB, University Paris Est Créteil, 94010 Créteil, France
10
Department of Pediatric Clinical Neurophysiology, Necker Enfants Malades Hospital, AP-HP Centre—Paris University, 75015 Paris, France
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Cells 2025, 14(12), 939; https://doi.org/10.3390/cells14120939
Submission received: 6 May 2025 / Revised: 7 June 2025 / Accepted: 16 June 2025 / Published: 19 June 2025
Browse Figures
Versions Notes

Abstract

This study aimed to investigate the spatial heterogeneity of molecular signature in the muscle of juvenile dermatomyositis (JDM) patients before and after treatment. Unsupervised reference-free deconvolution of spatial transcriptomics and standardized morphometry were performed in two JDM muscle biopsies with different clinical severity at disease onset and compared to healthy muscle. Identified signatures were scored in two additional JDM muscle biopsies from the same patient before and after remission. Disappearance of the normal muscle signature mostly corresponding to mitochondrial biology was observed in JDM. Three pathological transcriptomic signatures were isolated, related to “myofibrillar stress”, “muscle remodeling” and “interferon signaling” signatures. The “myofibrillar stress signature” was prominent in the most severe biopsy while the “muscle remodeling” signature was mostly present in the biopsy from the patient with good outcome. These signatures unveiled genes not previously associated with JDM including ANKRD1 and FSLT1 for “myofibrillar stress” and “muscle remodeling” signatures, respectively. Post-treatment analysis of muscle after two years remission showed a persistence of pathological signatures. This pilot study of JDM muscle identified spatially distributed pathological signatures that persist after remission. This work paves the way for a better understanding of the pathophysiology in affected muscle and the identification of biomarkers that predict relapse.

1. Introduction

Juvenile dermatomyositis (JDM) is a rare and heterogeneous autoimmune and inflammatory myopathy characterized by specific skin and muscle involvement. Other organ involvements, particularly of the lungs and gastrointestinal tract, are associated with mortality in some cases. Although modern treatment has improved the outcome of JDM, cohort studies suggest that only 50% of patients achieve remission off therapy [1,2]. Muscle-specific autoantibodies (MSA) help to stratify patients into different phenotypes. Their role extends from mere diagnostic dimensions to prognostic and therapeutic ones and their involvement in the pathophysiology of JDM by disrupting their intracellular targets in now recognized [3]. JDM pathophysiology is characterized by a strong and selective upregulation of type I interferon (IFN-I, i.e., IFNα/β) in the circulation and tissue which is attributed to complex interplay between genetic and environmental risk factors [4,5]. In muscle tissue, the preferential activation of IFN-I pathway is a major contributor to myopathological changes and is associated with a multifocal capillary loss, muscle ischaemia and perifascicular myofibre atrophy. Altered mitochondrial biology also occurs in JDM and may be part of the pathological loop that drives IFN-I production [6]. Deregulation of IFN-I pathway is interconnected with endothelial dysfunction and vasculopathy and is associated with the poorer prognosis of JDM [7]. The nature of the triggering events and the link between microvascular dysfunction and IFN-I upregulation remains unresolved [8,9]. Early identification of patients with more severe diseases requiring specific treatment remains a major unmet medical need. Such identification remains hampered by the poor understanding of the cellular and molecular mechanisms underlying the variability of disease progression. Here, we took advantage of multi-cellular pixel resolution spatial transcriptomic (ST) profiling of muscle using a reference-free deconvolution approach in JDM to describe spatial transcriptomic signatures at disease onset. Furthermore, ST showed that abnormal signatures persist after remission, paving the way for understanding of disease relapse in JDM.

2. Materials and Methods

2.1. Patients and Muscle Biopsies

Patients were recruited in the reference centers for Rare Pediatric Inflammatory Diseases (RAISE/FAI2R) and for Rare Neuromuscular Diseases (Nord-Est-Ile de France/FILNEMUS). JDM patients were classified according to conventional clinico-pathological ENMC criteria [10]. For all patients, serum was screened for MSAs and myositis-associated antibodies at disease onset, using an immunodot assay (Euro-line Autoimmune Inflammatory Myopathies 16 Ag; Euroimmun). Open Deltoid biopsies from JDM patients were performed at the time of the diagnosis in the context of diagnostic work-up. The clinical and demographic characteristics of the JDM patients are shown in Table 1. Normal control biopsy samples were obtained from the paravertebral muscle of patients undergoing spinal arthrodesis. The two healthy control samples were from female donors, who were aged 10 and 15 years old respectively. All biopsy samples were blindly reviewed for clinical data using the severity scoring tool for muscle biopsy evaluation in JDM patients [11].

2.2. Methods for Histology, Immunohistochemistry and Morphometric Analysis

See Supplementary Information.

2.3. Visium Spatial Gene Expression Library Construction

Spatial gene expression libraries were generated following the 10X Genomics Visium Spatial Gene Expression protocol (User Guide, CG000239 Rev F). Muscle 10 µm-cryo-sectionned were placed onto Visium glass slides. The slides were transferred to 37 °C for 1 min before immersion and fixation in methanol at −20 °C for 30 min. After fixation, sections were dried by adding 500 μL isopropanol for 1 min then air-dried, stained with Mayer’s Hematoxylin/Eosin and dried again for 5 min at 37 °C. Imaging was performed using a Zeiss Axio Imager.D1/D2 at 10× magnification, and the images were processed using ZEISS ZEN 3.7 software. After imaging, the slide was transferred to the Slide Cassette. The tissue sections then were permeabilized using the permeabilization enzyme at 37 °C (incubation time was determined following tissue permeabilization optimization), after which it was removed and each well washed with 0.1 × SSC. A reverse transcription mix containing RT Reagent, Template Switch Oligo, Reducing Agent B, RT Enzyme D, and Nuclease-free water were added, and the slide was incubated at 53 °C for 45 min. After removal, 0.08M KOH was added to each well and incubated at room temperature (RT) for 5 min, then washed with EB (Qiagen, Venlo, The Netherlands). Next, a second strand mix containing Second Strand Reagent, Second Strand Primer, and Second Strand Enzyme was added and incubated at 65 °C for 15 min. After washing with EB, 0.08M KOH was added to each well and incubated for 10 min at RT. Into each tube of an 8-tube strip, 5 μL Tris (1 M, pH 7.0) was added, followed by 35 μL of sample from the wells. cDNA amplification was carried out using a master mix containing Amp mix and cDNA primers. The number of cycles was determined using qPCR with the addition of KAPA SYBR FAST (Sigma-Aldrich, Burlington, MA, USA). The samples were purified using SPRIselect (Beckman Coulter, Brea, CA, USA) at 0.6×. Quality control was performed on all cDNA samples using an Agilent TapeStation Screen Tape, and concentrations were determined using a High Sensitivity Qubit assay (Thermo-Fisher, Waltham, MA, USA). Next, fragmentation was carried out by incubating the samples at 32 °C for 5 min and then 65 °C for 30 min with EB and a fragmentation mix containing Fragmentation Buffer and Fragmentation Enzyme. After fragmentation, a purification using SPRIselect at 0.6× and 0.8× was performed before adaptor ligation. An adaptor ligation mix containing Ligation Buffer, DNA Ligase, and Adaptor Oligos were added and incubated for 15 min at 20 °C, followed by purification using SPRIselect at 0.8×. For indexing, Amp mix, and an individual dual index (Kit TT Set A) were added to each sample, and the indexing protocol was run in a thermocycler. The number of cycles for indexing was determined by the cDNA input calculated from the previous quality control. After indexing, a final purification was performed using SPRIselect at 0.6× and 0.8×. Quality control was performed on all samples using an Agilent TapeStation Screen Tape, and concentrations were determined using a High Sensitivity Qubit assay (Thermo-Fisher). Finally, the libraries were sequenced in paired-end 150 bp mode on an Illumina NextSeq 500 system. The sequencing depth was determined by the percentage of capture areas covered by each muscle sample.

2.4. Visium Data Processing and Analysis

Illumina reads of each area were aligned on the GRCh38-2020-A (Ensembl 98) reference and counted using the spaceranger count function from the 10× genomics software Spaceranger 2.1.1. Areas corresponding to the same experiment were integrated using spacerange aggregate function with default parameters. Exploration of integrated data using the 10× Genomics software Loupe Browser 7.0.1 allows to annotate and discard spots outside and/or associated with low quality areas of the biopsies (e.g., edges, folded slice). All further analyses and visualization were performed under R (version 4.4.2). Reference-free deconvolution was performed on the count matrix using the STdeconvolve v1.4.0 R package [12]. Spots with a minimum of 100 detected genes and informative genes reaching the restrictCorpus default parameters were considered. The number of signatures was fixed a priori to K = 5, corresponding to a local minimum of perplexity (Online Supplemental Figure S1). Then, a list of marker genes (up and downregulated) was defined in each of the 5 signatures based on the following log2FC formula [12]. Enrichment tests using ReactomePA v 1.46.0 R package [13] and Ingenuity Pathway Analysis (IPA, QIAGEN Inc., https://digitalinsights.qiagen.com/IPA (accessed on 5 May 2025)), were performed on marker genes list. For IPA, pathway network with significant p-values (p < 0.05) was generated [14]. In a next step, the correspondence of spatial transcriptomics data from JDM patient biopsies sampled before and after treatment with the transcriptomics signatures (i.e., list of marker genes) were investigated using a directional (upSet and downSet options) Singscore v. 1.20.0 R package [15,16]. Singscore ranges from −1 to 1, a spot which singscore is 1 means that its transcriptome matches the tested signature. Empirical Pvalue were estimated for each scored spot using 1000 permutations with generateNull() and getPvals() functions from Singscore v. 1.20.0 R package. Estimated Pvalue corresponds to the results of an enrichment test of the signatures in each spot. Dimensional reduction and t-SNE (t-distributed stochastic neighbor embedding) calculation were performed using M3C v.1.22.0 R package [17]. Data visualization was done with ggplot2 v. 3.5.0 R package.

3. Results

3.1. Spatial Transcriptomic Analysis Reveals Specific and Spatialized Muscle Signatures in JDM as Compared to Healthy Muscle

To define the transcriptomic signature of the different areas of affected muscle in JDM, we performed 10X Genomics Visium Spatial Gene Expression analysis on the muscle obtained at the diagnosis of an acute NXP2+ JDM with severe muscle involvement (muscle biopsy severity score: 20/27, VAS: 7/10), an indolent, slowly progressive MSA-negative JDM (muscle biopsy severity score: 13/27, VAS: 4/10), and healthy muscle (HM). Morphometric analysis of the myofibre size (cross-sectional area), of fibrosis and capillary density showed a more severe overall atrophy in indolent MSA-negative JDM, whereas fibrosis and capillary loss were similar in both JDM and normal muscle tissue (Figure 1A–L). On average, 80 million RNA sequences were produced per biopsy ([44 × 106; 135 × 106], SD: 48 × 106), covering a total of 2071 usable spots. The capture of 8800 to 90,600 (SD 17,750) reads per spot allowed the detection of a mean of 336 ([252; 398] SD: 75.4) genes per spot, corresponding to approximately 14,000 genes detected per entire biopsy. The mean number of genes per spot was consistent across the three biopsies (acute NXP2+: 398, indolent MSA-negative: 358, HC: 252). T-SNE performed on global aggregated data showed that spots from both muscles of JDM patients clustered together, away from the healthy muscle spots (Figure 2A). In accordance with the clinical and pathological severity, the distribution of spots in the t-SNE shows a gradient, with spots from the indolent MSA-negative JDM muscle mostly positioned between spots from the healthy muscle and spots from the acute NXP2+ JDM muscle. The reference-free deconvolution approach allowed (i) to identify 5 transcriptional profiles (i.e., molecular signatures) (Online Supplemental Figure S1), in which a list of specifically up or down-regulated transcripts were identified (Online Supplemental Table S1) and (ii) to estimate the percentage of each molecular signature in each spot of each biopsy. Signature 1 was prominent in normal muscle (99.3%), signature 2 was less expressed and similar between the three conditions, signature 3 was prominent in acute NXP2+ JDM (78.8%) while signature 5 was prominent in indolent MSA-negative JDM muscle (65.3%). Signature 4 was associated with both JDM (acute NXP2+ JDM 9.2%; indolent MSA-negative JDM 14.2%) (Figure 2B,C).
This analysis refined the transcriptomic mapping on the three biopsies and showed that the transcriptome of JDM patients is spatially distributed, with regionalized specific signatures, whereas healthy biopsy showed a homogeneous transcriptome in each spot.

3.2. Muscles from Acute and Indolent JDM Could Be Distinguished by the Expression of a Predominant “Myofibrillar Stress” Signature or an “Muscle Remodeling” Signature Respectively but Shared “IFN Signaling” Signature

Using the Reactome database, we identified significantly enriched pathway for each of the 5 lists of genes previously identified (Figure 3A,C,E,G, Online Supplementary Table S2). Signature 1, which was mostly present in healthy muscle, was mainly characterized by higher expression of genes related to “citric acid (TCA) cycle and respiratory electron transport”. Indeed, ATP6, ATP8 and NDA5 were among the most highly expressed genes in healthy muscle compared to JDM. Signature 3, which was mostly present in the acute NXP2+ JDM, was associated with the “muscle contraction” pathway, with ACTC1, MYH8 and MHY3 among the top over expressed genes compared to the other biopsies. Signature 5, which is mostly present in the biopsy of indolent MSA-negative JDM was associated with the “Extracellular matrix organization” pathway. The top overexpressed genes from this signature are associated with extracellular matrix composition (exp COL6A3/COL15A1), with mitochondrial biology (exp NDUFB5/NDUFB9/NDUFA7) and with other genes of interest (exp FSTL1). Interestingly, signature 4 which was common between the two JDM biopsies was strongly associated with the “interferon signaling” pathway, notably with ISG15, IFIT1 and IFI27 among the most overrepresented transcripts. Signature 2 was almost absent in all biopsies (less than 1%). It was characterized by the overexpression of genes shared with signature 1 (Online Supplemental Figure S2 and Table S2). For this reason, we decided not to include it for the rest of the analysis. Interestingly, network analysis performed with IPA showed very comparable results for Signature 1, 3 and 4 (Figure 3B,D,F,H,). However, the most abundant network for signature 5 is composed of genes from the mitochondrial respiratory chain (NDUFB9, B5 and A7) as well as genes of interest not previously reported in muscle from JDM (for exp FSTL1 and TAGLN2) confirming the complexity of this signature 5 likely corresponding to muscle remodeling (Figure 3H). Based on these results, we relabeled signature 1 as “healthy muscle signature”, signature 3 as “myofibrillar stress signature”, signature 4 as “interferon signature” and signature 5 as “muscle remodeling signature”.

3.3. Spatialized Transcriptomic Pathological Signatures Are Associated with Myopathological Changes

Each Visium spot was assigned to the molecular signature showing the maximal percentage after deconvolution for both cases of JDM (Figure 4A–C). Myofibre changes in JDM muscles were delineated on JDM muscle biopsies (Figure 4D,E). Then, blinded to the Visium spatial transcriptomic data, the delineated areas were superimposed on Visium spots. Fisher test showed that the associations between pathological changes and signatures were significant (p = 0.0004998). We observed that ischemic changes and the perifascicular atrophy were preferentially associated with the “myofibrillar stress” signature whereas the global atrophy was associated with the “muscle remodelling” and “IFN signatures”. Moderate myopathic changes were preferentially associated with “muscle remodelling” and “healthy muscle” signatures (Figure 4F). The transcriptomic profiles reflected the respective histopathological patterns observed in the two JDM muscles, namely ischemic injuries in acute NXP2+ JDM with microinfarcts, myosinolysis and perifascicular atrophy, and chronic moderate myopathic changes in indolent, slowly progressive MSA-negative JDM with myofiber size variation and internalized nuclei.

3.4. Persistence of Myofibrillar Stress Signature and Muscle Remodeling Signature in the Muscle of a Patient in Two-Years Remission After JAK Inhibitor Treatment

We then investigated the correspondence of spatial transcriptomic data from two additional muscle biopsies obtained at diagnosis and after two years of remission from the same NXP2+ JDM patient as compared to healthy muscle with the transcriptomic signatures identified above (Figure 5A–C). At disease onset the patient presented with a very severe ischemic myopathy characterized by massive endomysial capillary loss (muscle severity score: 26/27, VAS: 8/10). The second muscle biopsy performed 42 months later and after 22 months of JAKi treatment showed almost no abnormality: Moderate residual changes (myofibre size irregularity, rounded appearance and centronucleation of myofibres), mild endomysial fibrosis and complete restoration of endomysial microvascular bed (muscle severity score: 3/27, VAS: 1/10). To quantify the change in signature expression, we calculated a Singscore for each spot of each biopsy representing the relative expression of each signature within this spot. Compared to healthy muscle, the “healthy muscle” signature was almost absent in biopsy performed at diagnosis, while all three pathological signatures “myofibrillar stress”, “interferon” and “muscle remodeling” were present. Following JAKi treatment, we observed a significant increase in the healthy signature in most spots. Conversely, we observed a decrease in all three pathological signatures. Interestingly, only the interferon score returned to levels comparable to healthy muscle, but unexpectedly both the “myofibrillar stress” and “muscle remodeling” signature persisted in the biopsy (Figure 5D).

4. Discussion

This study represents the first application of spatial transcriptomics to JDM muscle biopsies using the Visium Spatial Gene Expression technology unveiling the disappearance of normal mitochondrial muscle signature substituted by pathological signatures depending on clinical severity in JDM. Furthermore, abnormal disease signatures persist while patients are on remission. Our understanding of the pathophysiology of JDM is based on studies conducted over the past decades, initially focusing on muscle vasculopathy and more recently on the role of IFN-I upregulation. In particular, we have little insight into the chronology of events taking place in the affected muscle. To date, omics studies in DM/JDM have focused on either circulating cells or bulk muscle and single cell/nuclei RNA sequencing. These studies have identified distinct gene expression patterns in myositis associated with different autoantibodies [18,19] questioning the pathological significance of these autoantibodies [20] or have addressed the mechanisms contributing to the abnormal IFN-I signature in the patients [6,21].
The recent development of spatial transcriptomics constitutes a breakthrough in the understanding of pathological mechanisms at the tissue level and has recently been applied to several neuromuscular disorders in animal models [22,23] and in human myositis, including sarcoidosis myositis [24], inclusion body myositis [25] and dermatomyositis [26]. Here, we have applied this strategy to muscle biopsies from juvenile dermatomyositis. Visium technology allows the investigation of 55 μm circular spots (i.e., subcellular resolution), in which mRNA from several biological cell types could be captured [27]. We chose herein a reference-free deconvolution [12], which allows the identification of molecular signatures based only on the information present in the spatial transcriptomics dataset and prevents us from missing myofibres, and others cell types affected in JDM muscle [27].
This approach allowed us to identify four physiological and pathological transcriptomic signatures with biological relevance that are heterogeneously distributed within patient’s biopsies. Indeed, in patient presenting with severe and acute NXP2+ JDM, the model of deconvolution identified an almost complete disappearance of the healthy transcriptomic signature in muscle tissue substituted by a prominent pathological “myofibrillar stress” signature. We observed that this pathological signature within the muscle tissue was preferentially associated with ischaemic changes in myofibres including ischaemic punch-out vacuoles and microinfarcts. ANKRD1 was the most upregulated gene from the myofibrillar stress signature along with genes encoding for sarcomeric proteins essential for muscle contraction. ANKRD1 encodes for a pleotropic transcriptional coactivator highly conserved across mammals and is involved in various biological processes, including cardiac muscle development [28] and remodeling following injury or stress [29]. Interactions between this protein and the sarcomeric proteins myopalladin and titin suggest that it is involved in the myofibrillar stretch-sensor system [30]. Its overexpression can lead to heart diseases, such as diastolic dysfunction and heart failure [31]. More recently ANKRD1 was identified as a mesenchymal-specific transcriptional coactivator driving fibroblast activation in cancer but also in other conditions such as hypertrophic scarring and idiopathic pulmonary fibrosis [32]. ANKRD1 is also activated in renal ischaemia–reperfusion injury (IRI) models in vivo and in vitro [33]. Altogether, these findings highlight the potential role of ANKRD1 as a biomarker or therapeutic target for pathological tissue regeneration in JDM. As vasculopathy is central to the pathogenesis of juvenile dermatomyositis [34], the role of ANKRD1 as a mediator of myofibre response to abnormal blood flow may be further investigated to gain insight into the pathophysiology of myofibre ischaemic changes that occur in severe JDM.
Conversely, in indolent, slowly progressive, MSA-negative JDM, the “muscle remodelling signature” was mostly present. This signature is composed of a mixture of genes encoding for proteins involved in extracellular matrix remodelling, mitochondrial biology, angiogenesis and immune functions. We identified differentially expressed genes that are involved in this pathological signature associated with myofibre changes mostly consisting of a global atrophy. Among these, some genes that have never been described in JDM muscle may be of particular interest. FSTL1 encodes for a small glycoprotein secreted by skeletal muscle and myocardium and have been implicated in various physiological and pathological processes, including inflammation, wound healing, and cancer progression [35]. The cardioprotective role of FSTL1 has been extensively studied in recent years, although its mechanisms of action remain elusive [36]. In muscle tissue, secretion of FSTL1 by myogenic cells promotes ischaemia-induced revascularisation by activating Akt-endothelial nitric oxide synthase-dependent signaling within endothelial cells [37]. In immune diseases, FSTL1 has a dual function during inflammatory process (anti-inflammatory factor in the acute phase, then pro-inflammatory effect in chronic diseases), probably due to the activation of different signalling pathways [38,39]. Therefore, FSTL1 may be of interest as a novel non-invasive biomarker or as a new therapeutic target in JDM [40].
The interferon signature was present in the biopsies of all active patients confirming the central role of type I interferon upregulation in JDM pathophysiology. While type I interferon signature has been reported in the muscle of JDM patients by several studies [4,6,41], it is worth noting that this signature is very heterogeneously distributed within the biopsies. It suggests that type I interferon may not be homogeneously produced in affected muscle or that type I interferon responding cells may not be homogeneously distributed in the muscle.
A major challenge in the management of JDM patients is the strategy of tapering treatment following good clinical response. Indeed, approximately 50% of patients will achieve complete clinical remission after first, second- or third-line treatment. Disease relapse can occur even after several years of disease remission [42,43,44]. Therefore, we need to better understand what the drivers of this risk are and identify biomarkers to better define our tapering strategies in JDM patients. In this study we had the opportunity to study the muscle of a patient performed after 2 years of complete remission following treatment with Ruxolitinib. At the time of her second biopsy, the patient had a normal muscle strength and almost normal muscle biopsy. This was confirmed at the molecular level by the strong reduction of all three pathological signatures compared to the first biopsy. However, only the interferon signature normalised while both “myofibrillar stress” and the “muscle remodelling” signatures persisted, suggesting that despite clinical and histological remission, disease signatures persisted.
The persistence of the “muscle remodelling” signature could reflect the process towards complete muscle regeneration to restore a normal muscle function but concomitant “myofibrillar stress” signature suggests that the disease persists and could explain the risk of disease relapse even after a long period of clinical remission [42]. The complete disappearance of the interferon signature is consistent with many previous publications linking interferon protein and signature to disease activity [45]. This finding suggests the involvement of additional, interferon-independent pathways in JDM pathophysiology and highlights the potential need to develop complementary therapeutic strategies targeting these persistent molecular abnormalities.
This study has several limitations. First, we did not combine single-nucleus RNA sequencing with spatial transcriptomics from patient muscle biopsies to map cell-type-specific drivers underlying JDM pathogenesis. Second, the study is limited by a small sample size and our results should be confirmed in a larger cohort of JDM muscle tissues with different MSA subtypes and outcomes to refine correlations between histological/transcriptomic data and to decipher the spatial/temporal chronology of the muscle lesions. Third, this study is descriptive, and the observed correlations should not imply causality without complementary investigations.

5. Conclusions

This study represents the first application of spatial transcriptomics to JDM muscle biopsies using the Visium Spatial Gene Expression technology suggesting the disappearance of normal mitochondrial muscle signature substituted by pathological signatures depending on clinical severity in JDM. Furthermore, abnormal disease signatures persist while patients are on remission. These pilot study provide promising insights into possible mechanisms of myofibre injuries and disease relapse after remission in JDM.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cells14120939/s1, Figure S1: Selection of appropriate number of transcriptomics signatures K a priori; Figure S2: Ten first enriched Reactome pathways for signature 2; Table S1: List of gene markers for each of 5 signatures; Table S2: Indolent and acute JDM muscles display distinct transcriptomic signatures (all results).

Author Contributions

M.T., S.A.D. and B.P. conducting experiments; C.G., F.J.A. and M.P.R. designing research studies; M.T., S.A.D., B.P., B.B.-M., C.B. (Christine Barnerias), C.B. (Christine Bodemer) and I.D. acquiring data; M.T., S.A.D., B.P., C.G., M.P.R. and F.J.A. analyzing data; C.G., M.T., F.J.A. and M.P.R. writing the manuscript; C.G., F.J.A. and M.P.R. equally contributed to this work. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by a research grant from Agence Nationale de la Recherche (ANR) via project JDMINF2 (ANR-21-CE17-0025) and RHU CARMMA (ANR-15-RHUS-0003), and by a research grant from Association Française contre les Myopathies (AFM), Pole Translamuscle (project number 2946).

Institutional Review Board Statement

This study was performed in line with the principles of the Declaration of Helsinki. In accordance with current French legislation and hospital research ethics committee (Approval #12-009 at CPP Ile-de-France IX).

Informed Consent Statement

Parents or legal representatives gave their written informed consent for the participation of the children to the study.

Data Availability Statement

The data that support the findings of this study are openly available in BioStudies at http://doi.org/studies/S-BSST1845, reference number S-BSST1845.

Acknowledgments

The authors thank all the Children and their families who have contributed to this research. The authors also thank Jules Gilet who initiated the initial muscle spatial transcriptomic experiment.

Conflicts of Interest

The authors have no relevant financial or non-financial interests to disclose. The author Séverine Degrelle was employed by the company Inovarion. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Dabbak, I.; Rodero, M.P.; Aeschlimann, F.A.; Authier, F.-J.; Bodemer, C.; Quartier, P.; Bondet, V.; Charuel, J.-L.; Duffy, D.; Gitiaux, C.; et al. Efficacy and Tolerance of Corticosteroids and Methotrexate in Patients with Juvenile Dermatomyositis: A Retrospective Cohort Study. Rheumatology 2022, 61, 4514–4520. [Google Scholar] [CrossRef]
  2. Maccora, I.; Brunner, H.I.; Cassedy, A.; Altaye, M.; Quinlan-Waters, M.; Lovell, D.J.; Grom, A.; Angeles-Han, S.T. Achieving Medication-Free Remission in Patients With Juvenile Dermatomyositis. ACR Open Rheumatol. 2025, 7, e11751. [Google Scholar] [CrossRef] [PubMed]
  3. Stenzel, W.; Goebel, H.-H.; Bader-Meunier, B.; Gitiaux, C. Inflammatory Myopathies in Childhood. Neuromuscul. Disord. 2021, 31, 1051–1061. [Google Scholar] [CrossRef] [PubMed]
  4. Gallay, L.; Mouchiroud, G.; Chazaud, B. Interferon-Signature in Idiopathic Inflammatory Myopathies. Curr. Opin. Rheumatol. 2019, 31, 634–642. [Google Scholar] [CrossRef]
  5. Ward, J.M.; Ambatipudi, M.; O’Hanlon, T.P.; Smith, M.A.; de Los Reyes, M.; Schiffenbauer, A.; Rahman, S.; Zerrouki, K.; Miller, F.W.; Sanjuan, M.A.; et al. Shared and Distinctive Transcriptomic and Proteomic Pathways in Adult and Juvenile Dermatomyositis. Arthritis Rheumatol. 2023, 75, 2014–2026. [Google Scholar] [CrossRef]
  6. Wilkinson, M.G.L.; Moulding, D.; McDonnell, T.C.R.; Orford, M.; Wincup, C.; Ting, J.Y.J.; Otto, G.W.; Restuadi, R.; Kelberman, D.; Papadopoulou, C.; et al. Role of CD14+ Monocyte-Derived Oxidised Mitochondrial DNA in the Inflammatory Interferon Type 1 Signature in Juvenile Dermatomyositis. Ann. Rheum. Dis. 2023, 82, 658–669. [Google Scholar] [CrossRef] [PubMed]
  7. Besançon, A.; Brochard, K.; Dupic, L.; Gitiaux, C.; Delville, M.; Krid, S.; Quartier, P.; Saire, E.; Salomon, R.; Talbotec, C.; et al. Presentations and Outcomes of Juvenile Dermatomyositis Patients Admitted to Intensive Care Units. Rheumatology 2017, 56, 1814–1816. [Google Scholar] [CrossRef]
  8. Gitiaux, C.; De Antonio, M.; Aouizerate, J.; Gherardi, R.K.; Guilbert, T.; Barnerias, C.; Bodemer, C.; Brochard-Payet, K.; Quartier, P.; Musset, L.; et al. Vasculopathy-Related Clinical and Pathological Features Are Associated with Severe Juvenile Dermatomyositis. Rheumatology 2016, 55, 470–479. [Google Scholar] [CrossRef]
  9. Gitiaux, C.; Kostallari, E.; Lafuste, P.; Authier, F.-J.; Christov, C.; Gherardi, R.K. Whole Microvascular Unit Deletions in Dermatomyositis. Ann. Rheum. Dis. 2012, 72, 445–452. [Google Scholar] [CrossRef]
  10. Mammen, A.L.; Allenbach, Y.; Stenzel, W.; Benveniste, O.; ENMC 239th Workshop Study Group. 239th ENMC International Workshop: Classification of Dermatomyositis, Amsterdam, The Netherlands, 14–16 December 2018. Neuromuscul. Disord. NMD 2020, 30, 70–92. [Google Scholar] [CrossRef]
  11. Varsani, H.; Charman, S.C.; Li, C.K.; Marie, S.K.N.; Amato, A.A.; Banwell, B.; Bove, K.E.; Corse, A.M.; Emslie-Smith, A.M.; Jacques, T.S.; et al. Validation of a Score Tool for Measurement of Histological Severity in Juvenile Dermatomyositis and Association with Clinical Severity of Disease. Ann. Rheum. Dis. 2013, 74, 204–210. [Google Scholar] [CrossRef] [PubMed]
  12. Miller, B.F.; Huang, F.; Atta, L.; Sahoo, A.; Fan, J. Reference-Free Cell Type Deconvolution of Multi-Cellular Pixel-Resolution Spatially Resolved Transcriptomics Data. Nat. Commun. 2022, 13, 2339. [Google Scholar] [CrossRef]
  13. Yu, G.; He, Q.-Y. ReactomePA: An R/Bioconductor Package for Reactome Pathway Analysis and Visualization. Mol. Biosyst. 2016, 12, 477–479. [Google Scholar] [CrossRef] [PubMed]
  14. Chaigne, B.; Clary, G.; Le Gall, M.; Dumoitier, N.; Fernandez, C.; Lofek, S.; Chafey, P.; Moinzadeh, P.; Krieg, T.; Denton, C.P.; et al. Proteomic Analysis of Human Scleroderma Fibroblasts Response to Transforming Growth Factor-ß. Proteom. Clin. Appl. 2019, 13, e1800069. [Google Scholar] [CrossRef]
  15. Foroutan, M.; Bhuva, D.D.; Lyu, R.; Horan, K.; Cursons, J.; Davis, M.J. Single Sample Scoring of Molecular Phenotypes. BMC Bioinform. 2018, 19, 404. [Google Scholar] [CrossRef]
  16. Bhuva, D.D.; Cursons, J.; Davis, M.J. Stable Gene Expression for Normalisation and Single-Sample Scoring. Nucleic Acids Res. 2020, 48, e113. [Google Scholar] [CrossRef] [PubMed]
  17. John, C.R.; Watson, D.; Russ, D.; Goldmann, K.; Ehrenstein, M.; Pitzalis, C.; Lewis, M.; Barnes, M. M3C: Monte Carlo Reference-Based Consensus Clustering. Sci. Rep. 2020, 10, 1816. [Google Scholar] [CrossRef]
  18. Pinal-Fernandez, I.; Casal-Dominguez, M.; Derfoul, A.; Pak, K.; Miller, F.W.; Milisenda, J.C.; Grau-Junyent, J.M.; Selva-O’Callaghan, A.; Carrion-Ribas, C.; Paik, J.J.; et al. Machine Learning Algorithms Reveal Unique Gene Expression Profiles in Muscle Biopsies from Patients with Different Types of Myositis. Ann. Rheum. Dis. 2020, 79, 1234–1242. [Google Scholar] [CrossRef]
  19. Kim, H.; Gunter-Rahman, F.; McGrath, J.A.; Lee, E.; de Jesus, A.A.; Targoff, I.N.; Huang, Y.; O’Hanlon, T.P.; Tsai, W.L.; Gadina, M.; et al. Expression of Interferon-Regulated Genes in Juvenile Dermatomyositis versus Mendelian Autoinflammatory Interferonopathies. Arthritis Res. Ther. 2020, 22, 69. [Google Scholar] [CrossRef]
  20. Pinal-Fernandez, I.; Muñoz-Braceras, S.; Casal-Dominguez, M.; Pak, K.; Torres-Ruiz, J.; Musai, J.; Dell’Orso, S.; Naz, F.; Islam, S.; Gutierrez-Cruz, G.; et al. Pathological Autoantibody Internalisation in Myositis. Ann. Rheum. Dis. 2024, 83, 1549–1560. [Google Scholar] [CrossRef]
  21. Chen, X.; Lian, D.; Zeng, H. Single-Cell Profiling of Peripheral Blood and Muscle Cells Reveals Inflammatory Features of Juvenile Dermatomyositis. Front. Cell Dev. Biol. 2023, 11, 1166017. [Google Scholar] [CrossRef] [PubMed]
  22. Patsalos, A.; Halasz, L.; Oleksak, D.; Wei, X.; Nagy, G.; Tzerpos, P.; Conrad, T.; Hammers, D.W.; Sweeney, H.L.; Nagy, L. Spatiotemporal Transcriptomic Mapping of Regenerative Inflammation in Skeletal Muscle Reveals a Dynamic Multilayered Tissue Architecture. J. Clin. Investig. 2024, 134, e173858. [Google Scholar] [CrossRef] [PubMed]
  23. Heezen, L.G.M.; Abdelaal, T.; van Putten, M.; Aartsma-Rus, A.; Mahfouz, A.; Spitali, P. Spatial Transcriptomics Reveal Markers of Histopathological Changes in Duchenne Muscular Dystrophy Mouse Models. Nat. Commun. 2023, 14, 4909. [Google Scholar] [CrossRef]
  24. Lequain, H.; Dégletagne, C.; Streichenberger, N.; Valantin, J.; Simonet, T.; Schaeffer, L.; Sève, P.; Leblanc, P. Spatial Transcriptomics Reveals Signatures of Histopathological Changes in Muscular Sarcoidosis. Cells 2023, 12, 2747. [Google Scholar] [CrossRef] [PubMed]
  25. Wischnewski, S.; Thäwel, T.; Ikenaga, C.; Kocharyan, A.; Lerma-Martin, C.; Zulji, A.; Rausch, H.-W.; Brenner, D.; Thomas, L.; Kutza, M.; et al. Cell Type Mapping of Inflammatory Muscle Diseases Highlights Selective Myofiber Vulnerability in Inclusion Body Myositis. Nat. Aging 2024, 4, 969–983. [Google Scholar] [CrossRef]
  26. Abad, C.; Pinal-Fernandez, I.; Guillou, C.; Bourdenet, G.; Drouot, L.; Cosette, P.; Giannini, M.; Debrut, L.; Jean, L.; Bernard, S.; et al. IFNγ Causes Mitochondrial Dysfunction and Oxidative Stress in Myositis. Nat. Commun. 2024, 15, 5403. [Google Scholar] [CrossRef]
  27. Kleino, I.; Frolovaitė, P.; Suomi, T.; Elo, L.L. Computational Solutions for Spatial Transcriptomics. Comput. Struct. Biotechnol. J. 2022, 20, 4870–4884. [Google Scholar] [CrossRef]
  28. Huang, C.; Zhong, Q.; Lian, W.; Kang, T.; Hu, J.; Lei, M. Ankrd1 Inhibits the FAK/Rho-GTPase/F-Actin Pathway by Downregulating ITGA6 Transcriptional to Regulate Myoblast Functions. J. Cell. Physiol. 2024, 239, e31359. [Google Scholar] [CrossRef]
  29. Boskovic, S.; Marín Juez, R.; Stamenkovic, N.; Radojkovic, D.; Stainier, D.Y.; Kojic, S. The Stress Responsive Gene Ankrd1a Is Dynamically Regulated during Skeletal Muscle Development and Upregulated Following Cardiac Injury in Border Zone Cardiomyocytes in Adult Zebrafish. Gene 2021, 792, 145725. [Google Scholar] [CrossRef]
  30. Arimura, T.; Bos, J.M.; Sato, A.; Kubo, T.; Okamoto, H.; Nishi, H.; Harada, H.; Koga, Y.; Moulik, M.; Doi, Y.L.; et al. Cardiac Ankyrin Repeat Protein Gene (ANKRD1) Mutations in Hypertrophic Cardiomyopathy. J. Am. Coll. Cardiol. 2009, 54, 334–342. [Google Scholar] [CrossRef]
  31. Piroddi, N.; Pesce, P.; Scellini, B.; Manzini, S.; Ganzetti, G.S.; Badi, I.; Menegollo, M.; Cora, V.; Tiso, S.; Cinquetti, R.; et al. Myocardial Overexpression of ANKRD1 Causes Sinus Venosus Defects and Progressive Diastolic Dysfunction. Cardiovasc. Res. 2020, 116, 1458–1472. [Google Scholar] [CrossRef] [PubMed]
  32. Mazzeo, L.; Ghosh, S.; Di Cicco, E.; Isma, J.; Tavernari, D.; Samarkina, A.; Ostano, P.; Youssef, M.K.; Simon, C.; Dotto, G.P. ANKRD1 Is a Mesenchymal-Specific Driver of Cancer-Associated Fibroblast Activation Bridging Androgen Receptor Loss to AP-1 Activation. Nat. Commun. 2024, 15, 1038. [Google Scholar] [CrossRef]
  33. Han, S.; Guo, J.; Kong, C.; Li, J.; Lin, F.; Zhu, J.; Wang, T.; Chen, Q.; Liu, Y.; Hu, H.; et al. ANKRD1 Aggravates Renal Ischaemia–reperfusion Injury via Promoting TRIM25-Mediated Ubiquitination of ACSL3. Clin. Transl. Med. 2024, 14, e70024. [Google Scholar] [CrossRef]
  34. Papadopoulou, C.; Hong, Y.; Krol, P.; Al Obaidi, M.; Pilkington, C.; Wedderburn, L.R.; Brogan, P.A.; Eleftheriou, D. The Vasculopathy of Juvenile Dermatomyositis: Endothelial Injury, Hypercoagulability, and Increased Arterial Stiffness. Arthritis Rheumatol. 2021, 73, 1253–1266. [Google Scholar] [CrossRef]
  35. Mattiotti, A.; Prakash, S.; Barnett, P.; van den Hoff, M.J.B. Follistatin-like 1 in Development and Human Diseases. Cell. Mol. Life Sci. 2018, 75, 2339–2354. [Google Scholar] [CrossRef]
  36. Maruyama, S.; Nakamura, K.; Papanicolaou, K.N.; Sano, S.; Shimizu, I.; Asaumi, Y.; van den Hoff, M.J.; Ouchi, N.; Recchia, F.A.; Walsh, K. Follistatin-like 1 Promotes Cardiac Fibroblast Activation and Protects the Heart from Rupture. EMBO Mol. Med. 2016, 8, 949–966. [Google Scholar] [CrossRef]
  37. Ouchi, N.; Oshima, Y.; Ohashi, K.; Higuchi, A.; Ikegami, C.; Izumiya, Y.; Walsh, K. Follistatin-like 1, a Secreted Muscle Protein, Promotes Endothelial Cell Function and Revascularization in Ischemic Tissue through a Nitric-Oxide Synthase-Dependent Mechanism. J. Biol. Chem. 2008, 283, 32802–32811. [Google Scholar] [CrossRef] [PubMed]
  38. Murakami, K.; Tanaka, M.; Usui, T.; Kawabata, D.; Shiomi, A.; Iguchi-Hashimoto, M.; Shimizu, M.; Yukawa, N.; Yoshifuji, H.; Nojima, T.; et al. Follistatin-Related Protein/Follistatin-like 1 Evokes an Innate Immune Response via CD14 and Toll-like Receptor 4. FEBS Lett. 2012, 586, 319–324. [Google Scholar] [CrossRef] [PubMed]
  39. Chaly, Y.; Fu, Y.; Marinov, A.; Hostager, B.; Yan, W.; Campfield, B.; Kellum, J.A.; Bushnell, D.; Wang, Y.; Vockley, J.; et al. Follistatin-like Protein 1 Enhances NLRP3 Inflammasome-Mediated IL-1β Secretion from Monocytes and Macrophages. Eur. J. Immunol. 2014, 44, 1467–1479. [Google Scholar] [CrossRef]
  40. Li, X.; Fang, Y.; Jiang, D.; Dong, Y.; Liu, Y.; Zhang, S.; Guo, J.; Qi, C.; Zhao, C.; Jiang, F.; et al. Targeting FSTL1 for Multiple Fibrotic and Systemic Autoimmune Diseases. Mol. Ther. 2021, 29, 347–364. [Google Scholar] [CrossRef]
  41. Hou, C.; Durrleman, C.; Periou, B.; Barnerias, C.; Bodemer, C.; Desguerre, I.; Quartier, P.; Melki, I.; Rice, G.I.; Rodero, M.P.; et al. From Diagnosis to Prognosis: Revisiting the Meaning of Muscle ISG15 Overexpression in Juvenile Inflammatory Myopathies. Arthritis Rheumatol. 2021, 73, 1044–1052. [Google Scholar] [CrossRef] [PubMed]
  42. Rodero, M.P.; Pelleau, S.; Welfringer-Morin, A.; FJDM Study Group; Duffy, D.; Melki, I.; Bader-Meunier, B.; FJDM Study Group. Onset and Relapse of Juvenile Dermatomyositis Following Asymptomatic SARS-CoV-2 Infection. J. Clin. Immunol. 2022, 42, 25–27. [Google Scholar] [CrossRef] [PubMed]
  43. Kim, S.; El-Hallak, M.; Dedeoglu, F.; Zurakowski, D.; Fuhlbrigge, R.C.; Sundel, R.P. Complete and Sustained Remission of Juvenile Dermatomyositis Resulting from Aggressive Treatment. Arthritis Rheum. 2009, 60, 1825–1830. [Google Scholar] [CrossRef] [PubMed]
  44. Muramatsu, K.; Ujiie, H.; Yokozeki, M.; Tsukinaga, I.; Ito, M.; Shikano, T.; Suzuki, A.; Tozawa, Y.; Kobayashi, I. Recurrence of Juvenile Dermatomyositis 8 Years after Remission. JAAD Case Rep. 2017, 3, 29–32. [Google Scholar] [CrossRef]
  45. Rodero, M.P.; Decalf, J.; Bondet, V.; Hunt, D.; Rice, G.I.; Werneke, S.; McGlasson, S.L.; Alyanakian, M.-A.; Bader-Meunier, B.; Barnerias, C.; et al. Detection of Interferon Alpha Protein Reveals Differential Levels and Cellular Sources in Disease. J. Exp. Med. 2017, 214, 1547–1555. [Google Scholar] [CrossRef]
Cells 14 00939 g001
Figure 1. Detailed automated morphometric analysis of the muscle biopsies used to determine spatial transcriptomic signatures in JDM. Immunofluorescence staining of healthy muscle (upper panel: (A,E,I)), acute NXP2+ JDM (middle panel: (B,F,J)) and indolent MSA-negative JDM (lower panel: (C,G,K)). Representative images processed by an ImageJ macro (first column (A–C)) of myofibres areas (from ≤ 400 to > 7000 μm2) performed using a muscle tissue section stained for laminin to delineate myofibres; (second column (E–G)) of collagen VI immunofluorescence staining as marker of fibrosis; (third column (I–K)) of vessel number (CD31+) per myofibre (from 0 to > 9) performed using a muscle tissue section double-stained for laminin and CD31. Tool for the ImageJ software (version 1.54p 17) allowing for automated morphometry of (D) myofibre cross-sectional areas (H), of fibrosis and (L) endomysial vascular density assessed by the number of capillary associated with each myofibre. Bar graphs: median with range shown. Statistical analysis: chi-square (D,L); Kruskall Wallis allowed by Dunn’s multiple comparison test (H). JDM, juvenile dermatomyositis. NS: non-significant. p values ** ≤ 0.01 *** ≤ 0.01.
Figure 1. Detailed automated morphometric analysis of the muscle biopsies used to determine spatial transcriptomic signatures in JDM. Immunofluorescence staining of healthy muscle (upper panel: (A,E,I)), acute NXP2+ JDM (middle panel: (B,F,J)) and indolent MSA-negative JDM (lower panel: (C,G,K)). Representative images processed by an ImageJ macro (first column (A–C)) of myofibres areas (from ≤ 400 to > 7000 μm2) performed using a muscle tissue section stained for laminin to delineate myofibres; (second column (E–G)) of collagen VI immunofluorescence staining as marker of fibrosis; (third column (I–K)) of vessel number (CD31+) per myofibre (from 0 to > 9) performed using a muscle tissue section double-stained for laminin and CD31. Tool for the ImageJ software (version 1.54p 17) allowing for automated morphometry of (D) myofibre cross-sectional areas (H), of fibrosis and (L) endomysial vascular density assessed by the number of capillary associated with each myofibre. Bar graphs: median with range shown. Statistical analysis: chi-square (D,L); Kruskall Wallis allowed by Dunn’s multiple comparison test (H). JDM, juvenile dermatomyositis. NS: non-significant. p values ** ≤ 0.01 *** ≤ 0.01.
Cells 14 00939 g001
Cells 14 00939 g002
Figure 2. Spatial transcriptomic analysis reveals specific and spatialized muscle signatures in JDM as compared to healthy muscle. T-distributed stochastic neighbor embedding (t-SNE) dimensional reduction of 2071 transcriptomes (spots) of the three biopsies (A). De novo deconvolution of transcriptomes (spots) in 5 signatures. Healthy muscle (first row), acute NXP2+ JDM (middle row) and indolent MSA-negative JDM (last row). Signature 1 is mostly associated with healthy muscle, signature 3 with acute NXP2+ JDM and signature 5 with indolent MSA-negative JDM. Signature 4 is present in both JDM and signature 2 is not specific and present among the three conditions (B). Bar graph of the percentages of each signature in the three conditions (C).
Figure 2. Spatial transcriptomic analysis reveals specific and spatialized muscle signatures in JDM as compared to healthy muscle. T-distributed stochastic neighbor embedding (t-SNE) dimensional reduction of 2071 transcriptomes (spots) of the three biopsies (A). De novo deconvolution of transcriptomes (spots) in 5 signatures. Healthy muscle (first row), acute NXP2+ JDM (middle row) and indolent MSA-negative JDM (last row). Signature 1 is mostly associated with healthy muscle, signature 3 with acute NXP2+ JDM and signature 5 with indolent MSA-negative JDM. Signature 4 is present in both JDM and signature 2 is not specific and present among the three conditions (B). Bar graph of the percentages of each signature in the three conditions (C).
Cells 14 00939 g002
Cells 14 00939 g003
Figure 3. Indolent MSA negative and acute NXP2+ JDM muscles display distinct transcriptomic signatures. Enriched Reactome pathways (A,C,E,G) and IPA analysis (B,D,F,H) were performed on differentially expressed genes from signatures 1, 3, 4 and 5 respectively. Complete list of pathways and results for signature 2 are available in Supplemental Figure S2. For Reactome pathways analysis, only the ten most significant pathways are shown. The count represents the number of genes from marker genes list, that are found in a specific Gene Ontology (GO) term. The GeneRatio is the proportion of marker genes associated with a particular GO term relative to the total number of our marker genes in the lists. For IPA analysis, genes up- and down-regulated are respectively highlighted in green and red. The different shapes represent various types of molecules and their relationships according to the IPA knowledge base (http://ingenuity.com/ (accessed on 8 January 2025)).
Figure 3. Indolent MSA negative and acute NXP2+ JDM muscles display distinct transcriptomic signatures. Enriched Reactome pathways (A,C,E,G) and IPA analysis (B,D,F,H) were performed on differentially expressed genes from signatures 1, 3, 4 and 5 respectively. Complete list of pathways and results for signature 2 are available in Supplemental Figure S2. For Reactome pathways analysis, only the ten most significant pathways are shown. The count represents the number of genes from marker genes list, that are found in a specific Gene Ontology (GO) term. The GeneRatio is the proportion of marker genes associated with a particular GO term relative to the total number of our marker genes in the lists. For IPA analysis, genes up- and down-regulated are respectively highlighted in green and red. The different shapes represent various types of molecules and their relationships according to the IPA knowledge base (http://ingenuity.com/ (accessed on 8 January 2025)).
Cells 14 00939 g003
Cells 14 00939 g004
Figure 4. Spatialized transcriptomic pathological signatures are associated with myofibre injuries. Visium spots using 10X Genomics software Loupe Browser 7.0.1 in muscle cryosection from Acute NXP2+ (A JDM) (A) and Indolent MSA-negative (I JDM) (B) JDM, and healthy muscle (HM) (C). Each Visium spot was assigned to the molecular signature showing the maximal percentage after deconvolution. (D,E): Delineation of myopathological changes on HE stained muscle cryosection from Acute (D) and Indolent (E) JDM. Each area was delineated according to prominent histological feature and annoted as followed: (1), Normal or subnormal muscle tissue; (2–4), Interstitial tissue: 2, connective tissue; 3, artery; and 4, vein; (5) Myoaponevrotic junction; (6) Perifascicular atrophy; (7–9) Ischemic changes composed of ischaemic punch-out vacuoles and microinfarcts consisting of small foci of contiguous necrotic or regenerating fibres: 7, marked; 8, moderate; and 9, slight; (10) Overall atrophy; (11) Moderate myopathic changes (nuclear internalisation, atrophy); (12) Interstitium with focal collagenolysis; (13) Vasculitis; Areas excluded correspond to artefactually injured area not included in ST analysis. (F) Distribution of transcriptomic signatures in the different areas of interest (Fischer test, p < 0.05).
Figure 4. Spatialized transcriptomic pathological signatures are associated with myofibre injuries. Visium spots using 10X Genomics software Loupe Browser 7.0.1 in muscle cryosection from Acute NXP2+ (A JDM) (A) and Indolent MSA-negative (I JDM) (B) JDM, and healthy muscle (HM) (C). Each Visium spot was assigned to the molecular signature showing the maximal percentage after deconvolution. (D,E): Delineation of myopathological changes on HE stained muscle cryosection from Acute (D) and Indolent (E) JDM. Each area was delineated according to prominent histological feature and annoted as followed: (1), Normal or subnormal muscle tissue; (2–4), Interstitial tissue: 2, connective tissue; 3, artery; and 4, vein; (5) Myoaponevrotic junction; (6) Perifascicular atrophy; (7–9) Ischemic changes composed of ischaemic punch-out vacuoles and microinfarcts consisting of small foci of contiguous necrotic or regenerating fibres: 7, marked; 8, moderate; and 9, slight; (10) Overall atrophy; (11) Moderate myopathic changes (nuclear internalisation, atrophy); (12) Interstitium with focal collagenolysis; (13) Vasculitis; Areas excluded correspond to artefactually injured area not included in ST analysis. (F) Distribution of transcriptomic signatures in the different areas of interest (Fischer test, p < 0.05).
Cells 14 00939 g004
Cells 14 00939 g005
Figure 5. Persistence of “myofibrillar stress” signature and “muscle remodeling” signature in the muscle of a patient in two-years remission after JAK inhibitor treatment. Directionnal singscore and associated Pvalue (purple color scale) calculated by permutation tests of marker genes lists for lists for the 4 normal and pathological muscle spatial transcriptomic signatures (“healthy muscle”, myofibrillar stress”, “IFN signaling” and “muscle remodeling”). Singscores in each spatial transcriptomic spot: JDM biopsy at disease onset (A), JDM biopsy after JAKi treatment from the same patient (B), healthy muscle biopsy (C). Distribution of singscores values (from −1 to 1) in the three muscle biopsies (D).
Figure 5. Persistence of “myofibrillar stress” signature and “muscle remodeling” signature in the muscle of a patient in two-years remission after JAK inhibitor treatment. Directionnal singscore and associated Pvalue (purple color scale) calculated by permutation tests of marker genes lists for lists for the 4 normal and pathological muscle spatial transcriptomic signatures (“healthy muscle”, myofibrillar stress”, “IFN signaling” and “muscle remodeling”). Singscores in each spatial transcriptomic spot: JDM biopsy at disease onset (A), JDM biopsy after JAKi treatment from the same patient (B), healthy muscle biopsy (C). Distribution of singscores values (from −1 to 1) in the three muscle biopsies (D).
Cells 14 00939 g005
Table 1. Demographics and clinical characteristics of the three JDM patients (four muscle samples) included in the study. JDM1: acute NXP2+ JDM; JDM2: indolent, slowly progressive MSA-negative JDM; JDM3 onset: acute NXP2+ JDM at disease onset; JDM3 treated: NXP2+ JDM after two years of remission; MMT: Manual Muscle Testing; CMAS: Childhood Myositis Assessment Scales (CMAS).
Table 1. Demographics and clinical characteristics of the three JDM patients (four muscle samples) included in the study. JDM1: acute NXP2+ JDM; JDM2: indolent, slowly progressive MSA-negative JDM; JDM3 onset: acute NXP2+ JDM at disease onset; JDM3 treated: NXP2+ JDM after two years of remission; MMT: Manual Muscle Testing; CMAS: Childhood Myositis Assessment Scales (CMAS).
Sample IDJDM1JDM2JDM3 OnsetJDM3 Treated
Age at sample, years1351315
SexFFFF
Disease duration at time of biopsy, months14124
Myositis specific antibodyNXP20NXP2NXP2
Disease activity at time of biopsy:
CMAS (0–52)230/46252
MMT (0–80)47NR3880
CK U/L30201500862100
Characteristic skin rash1110
Gottron’s papules1110
Telangiectasiae1010
Subcutaneous limb edema1010
Skin ulcerations0000
Interstitial lung disease0000
Cardiovascular involvement0010
Gastrointestinal involvement0010
Muscle biopsy score data *
Inflammatory domain (0–12)84120
Muscle fibre domain (0–10)7792
Vascular domain (0–3)3130
Connective tissu domain (0–2)2121
Total biopsy score (0–27)2013263
Histopathology visual analogue score (VAS) (0–10)7481
* Data generated using validated JDM muscle biopsy score from Varsani et al. [11].
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Tragin, M.; Degrelle, S.A.; Periou, B.; Bader-Meunier, B.; Barnerias, C.; Bodemer, C.; Desguerre, I.; Rodero, M.P.; Authier, F.J.; Gitiaux, C. Muscle Spatial Transcriptomic Reveals Heterogeneous Profiles in Juvenile Dermatomyositis and Persistence of Abnormal Signature After Remission. Cells 2025, 14, 939. https://doi.org/10.3390/cells14120939

AMA Style

Tragin M, Degrelle SA, Periou B, Bader-Meunier B, Barnerias C, Bodemer C, Desguerre I, Rodero MP, Authier FJ, Gitiaux C. Muscle Spatial Transcriptomic Reveals Heterogeneous Profiles in Juvenile Dermatomyositis and Persistence of Abnormal Signature After Remission. Cells. 2025; 14(12):939. https://doi.org/10.3390/cells14120939

Chicago/Turabian Style

Tragin, Margot, Séverine A. Degrelle, Baptiste Periou, Brigitte Bader-Meunier, Christine Barnerias, Christine Bodemer, Isabelle Desguerre, Mathieu Paul Rodero, François Jérôme Authier, and Cyril Gitiaux. 2025. "Muscle Spatial Transcriptomic Reveals Heterogeneous Profiles in Juvenile Dermatomyositis and Persistence of Abnormal Signature After Remission" Cells 14, no. 12: 939. https://doi.org/10.3390/cells14120939

APA Style

Tragin, M., Degrelle, S. A., Periou, B., Bader-Meunier, B., Barnerias, C., Bodemer, C., Desguerre, I., Rodero, M. P., Authier, F. J., & Gitiaux, C. (2025). Muscle Spatial Transcriptomic Reveals Heterogeneous Profiles in Juvenile Dermatomyositis and Persistence of Abnormal Signature After Remission. Cells, 14(12), 939. https://doi.org/10.3390/cells14120939

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop