Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Identification of key biomarkers associated with immune cells infiltration for myocardial injury in dermatomyositis by integrated bioinformatics analysis

Identification of key biomarkers associated with immune cells infiltration for myocardial injury... Background Dermatomyositis (DM) is an acquired autoimmune disease that can cause damage to various organs, including the heart muscle. However, the mechanisms underlying myocardial injury in DM are not yet fully understood. Methods In this study, we utilized publicly available datasets from the Gene Expression Omnibus (GEO) database to identify hub-genes that are enriched in the immune system process in DM and myocarditis. Weighted gene co- expression network analysis ( WGCNA), differentially expressed genes (DEGs) analysis, protein–protein interaction (PPI), and gene ontology (GO) analysis were employed to identify these hub-genes. We then used the CIBERSORT method to analyze immune cell infiltration in skeletal muscle specimens of DM and myocardium specimens of myocarditis respectively. Correlation analysis was performed to investigate the relationship between key genes and infiltrating immune cells. Finally, we predicted regulatory miRNAs of hub-genes through miRNet and validated their expression in online datasets and clinical samples. Results Using integrated bioinformatics analysis, we identified 10 and 5 hub-genes that were enriched in the immune system process in the database of DM and myocarditis respectively. The subsequent intersections between hub-genes were IFIT3, OAS3, ISG15, and RSAD2. We found M2 macrophages increased in DM and myocarditis com- pared to the healthy control, associating with the expression of IFIT3, OAS3, ISG15, and RSAD2 in DM and myocar- ditis positively. Gene function enrichment analysis (GSEA) showed that IFIT3, OAS3, ISG15, and RSAD2 were mainly enriched in type I interferon (IFN) signaling pathway, cellular response to type I interferon, and response to type I interferon. Finally, we verified that the expression of miR-146a-5p was significantly higher in the DM with myocardial injury than those without myocardial injury (p = 0.0009). Yue Zhang, Linwei Shan, and Dongyu Li contributed equally to this work. Qiang Wang and Lei Zhou contributed equally to this work. *Correspondence: Qiang Wang jerrytortoise@163.com; zhoulei@njmu.edu.cn Lei Zhou jerrytortoise@163.com; zhoulei@njmu.edu.cn Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 2 of 12 Conclusion Our findings suggest that IFIT3, OAS3, ISG15, and RSAD2 may play crucial roles in the underlying mecha- nism of myocardial injury in DM. Serum miR-146a-5p could be a potential biomarker for myocardial injury in DM. Keywords Dermatomyositis, Myocardial injury, Bioinformatic, Immune cells infiltration, Biomarker Background Methods Dermatomyositis (DM) is an acquired autoimmune dis- Data sources ease in which skeletal muscle is targeted by the immune Series matrix files and platform information of GSE1551, system, which is a subset of idiopathic inflammatory GSE48280, GSE5370, GSE128470, GSE1145, GSE35182, myopathy (IIM) [1]. DM is characterized primarily by and GSE147517 were obtained from the National muscle inflammation, proximal muscle weakness, and Center Biotechnology Information Gene Expression cutaneous involvement. Additionally, the disease may Omnibus (NCBI-GEO) (https:// www. ncbi. nlm. nih. gov/ present with extra-muscular symptoms affecting vari - geo). We selected 13 DM patients and 10 normal indi- ous organs including the heart, joints, lungs, and gas- viduals, 5 DM patients, and 5 normal individuals, and 5 trointestinal tract [2]. DM patients and 4 normal individuals from GSE1551, Several studies have identified cardiovascular GSE48280, and GSE5370, respectively, and the samples involvement in DM is a feared threat to prognosis and a were all skeletal muscle biopsy specimens. We merged frequent cause of death in several studies [3–5]. Due to the GSE1551, GSE48280, and GSE5370 and used the the structural and functional similarities between skel- SVA software package to correct the bath. Then, we used etal and cardiac muscle, it is supposed that what affects principal component analysis (PCA) [10] to evaluate the skeletal muscle may be related to damage to cardiac results of the correction. Finally, we obtained a normal- muscle. In autopsy studies, myocarditis was the most ized gene expression matrix file containing 42 samples common pathologic features [6, 7], and previous imag- (23 DM patients, and 19 normal individuals). We chose ing studies also indicate inflammation as an underlying samples of myocardium obtained from 11 normal indi- pathology [8]. In different animal models of IIM, the viduals and 7 inflammatory cardiomyopathy patients due myocardium shows an inflammation similar to that of to viral myocarditis in GSE1145. In the section of valida- skeletal muscle morphologically [9]. tion, we chose GSE128470 which includes 12 samples of Despite the accumulating evidence linking DM and muscle obtained from normal individuals and 12 samples cardiovascular involvement, the extent of this link and from DM patients. Furthermore, we used GSE35182 to the underlying mechanisms are not yet fully under- validate the expression of genes between chronic myocar- stood. Moreover, specific recommendations for the ditis and normal mice; there are 6 mice per group. Finally, treatment of patients with DM and cardiovascular we validated the predicted target miRNAs expression in involvement are lacking. Thus, it is essential to explore GSE147517 comprising 5 myocarditis and 5 normal indi- the molecular characteristics and mechanisms of myo- viduals serum specimens. cardial injury in DM before developing screening rec- ommendations and treatment strategies for patients with DM and myocardial injury. Over the past decades, Construction of weighted gene co‑expression network gene microarray technology together with integrated analysis bioinformatic analyses has been performed to provide The WGCNA package in R was utilized to build a co- tremendous assistance in identifying novel key genes expression network targeting the top 5000 genes with related to various diseases. In this study, we identified median absolute deviation [11, 12]. The R function pick - the co-expression genes related to the immune sys- SoftThreshold was used to calculate the soft threshold - tem process between DM and myocarditis, explored ing power β, to which co-expression similarity is raised the association between these genes and immune cells to calculate adjacency. Then, we converted the adjacency infiltrating skeletal muscle or myocardium, and pre - into a topological overlap matrix (TOM) to measure the dicted regulatory miRNAs of the hub-genes. Finally, we network connectivity of genes. Genes with similar pat- verified our results through online datasets and clinical terns were clustered into the same modules (minimum samples. With the above approaches, it is hoped that size = 30) using average linkage hierarchical cluster- our results may provide a preliminary insight into the ing, which were represented by branches and different mechanism of myocardial injury in DM and a search colors of the cluster tree, constructed module relation- for possible biomarkers. ships, calculation of the correlation between gene mod- ules and phenotypes, and the modules related to clinical Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 3 of 12 traits were identified. Gene significance (GS) and module Prediction of potential hub‑gene related target miRNAs membership (MM) were calculated to relate modules to We used miRNet (www. mirnet. ca/) [25], a tool that inte- clinical traits. Finally, the highly correlated module was grates data from 11 different miRNA databases, to pre - analyzed to explore its core genes and potential roles. dict regulatory miRNAs of the common hub-genes. Subjects Identification of differentially expressed genes A total of 10 DM patients were enrolled from the First The limma package in R (Version 4.0.3) software was Affiliated Hospital of Nanjing Medical University utilized to screen DEGs between inflammatory car - between January 2020 and January 2021. We divided diomyopathy patients and normal controls [13, 14]. The these patients into two groups: DM with myocardial differentially expressed genes were screened under the injury (n = 5) and DM without myocardial injury (n = 5) condition of |log FC|> 1 and adjusted P value < 0.05. based on the results of the cardiac magnetic resonance examination. The inclusion criteria were as follows: (1) Evaluation of immune cell infiltration age > 18  years old, (2) all the patients fulfilled the 1975 CIBERSORT [15] is a deconvolution algorithm for ana- Bohan and Peter criteria for dermatomyositis [26], and lyzing gene expression data and uses a gene expres- (3) patients who underwent cardiac magnetic reso- sion signature for characterizing the proportion of each nance examination during hospitalization. The exclu - immune cell type. We performed immune infiltration sion criteria were as follows: (1) interstitial lung disease, by using the CIBERSORT.R script downloaded from (2) previous history of cardiac disease, (3) tumor, (4) the CIBERSORT website (https:// ciber sortx. stanf ord. renal insufficiency, and (5) surgery within the six previ - edu/). We used the original CIBERSORT gene signature ous months. Ethical approval was obtained for our sin- file LM22, which defines 22 immune cell subtypes, to gle-center cross-sectional study, and the need to obtain calculate the proportion of 22 immune cells in DM and informed consent was waived (2020-SR-228). inflammatory cardiomyopathy patients. Then, we used the “ggplot2” package [16] to draw violin diagrams to Sample collection and RNA isolation visualize the differences compared to normal controls in Venous blood samples were collected into EDTA tubes immune cell infiltration. We also calculated the Spear - from DM patients within 24  h before and after the car- man correlation coefficient between identified hub-genes diac magnetic resonance examination. Serum was with infiltrating macrophages M2. obtained at room temperature of 3000 rpm for 5 min and frozen at –  80  °C for further use. Isolation of total RNA Functional annotation including miRNA was performed from serum using the Metascape (http:// Metas cape. org/ gp/ index. html) is a free miRNeasy Serum/Plasma Advanced Kit (Qiagen, Ger- web-based analytics tool for comprehensive gene annota- mantown, MD, USA) according to the manufacturer’s tion and analysis resources, combining a GO and KEGG protocol. During the extraction, 3.5 μL of miRNeasy pathway enrichment analysis search to leverage over 40 8 Serum/Plasma Spike-In Control (1.6 × 10 copies/μL of independent knowledge bases [17–19]. To understand the C. elegans miR-39 miRNA mimic) was added to each the function of these core genes and DEGs, we used sample as an internal control. RNA quantity and quality metascape to perform GO and KEGG pathway enrich- were evaluated using the Nanodrop ND-2000 spectro- ment analyses. photometer (Nanodrop Technologies, Wilmington, DE). Directly after isolation, RNA was subjected to the reverse Protein–protein interaction network analysis transcription process. Search Tool for the Retrieval of Interacting Genes (STRING) (Version 11.3, https:// cn. string- db. org/) is a Reverse transcription reaction and real‑time quantitative useful online tool dedicated to analyzing the functional PCR (RT‑qPCR) protein association networks [20–22]. The core genes The expression levels of miR-27b-3p, miR-130a-3p, miR- and DEGs were mapped to the STRING database, and 1-3p, miR-133a-3p, miR-16-5p, and miR-146a-5p were only the experimentally validated interactions with a measured using the Bulge-Loop miRNA qRT-PCR combined score > 0.4 were selected as significant. Sub - Starter Kit (one RT primer and a pair of qPCR primers sequently, the PPI network was visualized by Cytoscape for each set) designed specifically by RiboBio (Guang - software (version 3.5.1) (www. cytos cape. org/) [23]. The zhou, China) in accordance with the manufacturer’s plug-in cytoHubba in Cytoscape was used to screen the instructions. The average expression levels of serum hub-genes from the PPI network, and in our study, the miRNAs were normalized against cel-miR-39 (Qiagen, top ten genes were identified as hub-genes [24]. Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 4 of 12 Germantown, MD, USA). The fold-change (FC) for each Cardiovascular Imaging, Calgary, Canada). The presence miRNA relative to the control was calculated using the of LGE was defined with a signal intensity level increase −ΔΔCT 2 method [27]. The efficiency was tested before of more than five S.D. of remote myocardium on the all −ΔΔCT using the 2 method, and the results were collected short-axis contrast images from base to apex [28]. ECV from the same experiment in three replicates. was calculated by the following equation: ECV = (1— 1 1 post contrast T1 myo native T1 myo HCT) . 1 1 CMR post contrast T1 blood blood T1 myo Determination of hematocrit (HCT) and calculation All participants underwent CMR imaging on a 3.0  T of ECV were completed within 24  h after CMR scan- whole-body scanner (MAGNETOM Skyra, Siemens ning. ECV is a marker of myocardial tissue remodeling Healthcare, Erlangen, Germany) with an 18-chan- and provides a physiologically intuitive unit of measure- nel phase-array coil using ECG gating. Late gadolinium ment. Normal ECV values of 24% ± 3 (3.0  T) have been enhancement (LGE) images were acquired 8–15  min reported in healthy individuals [29]. In our study, we after intravenous administration of gadolinium-DTPA divide patients with LGE positive and ECV > 25% into (Magnevist, Bayer, Berlin, Germany) at a dose of DM with myocardial injury group [30]. 0.2  mmol/kg in short-axis stack using a phase-sensitive inversion-recovery (PSIR) gradient echo sequence. Typi- Statistical analysis cal parameters of motion-corrected basal, mid, and apical All statistical analyses were performed in the R language level of LV short-axis Modified Look-Locker inversion- (Version 4.0.3). All statistical tests were bilateral, and recovery (MOLLI) T1 mapping sequence with a 5(3)3 adjusted P value < 0.05 was statistically significant. scheme before and 15–20 min after intravenous contrast agent injection. The native and post-contrast T1 values of the myocar - Results dium were measured on a region of interest at the myo- Bioinformatic analysis workflow cardial septum of mid-ventricular short-axis slice by two Our workflows are shown in Fig.  1. First, we found key experienced blinded investigators in consensus with modules related to DM using WGCNA and discovered CMR using commercial software (CVI42, Circle hub-genes associated with the immune system process Fig. 1 Flowchart of this study Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 5 of 12 Fig. 2 Screening of hub-genes related to the immune system process in DM. a PCA before the batch correction of three datasets. b PCA after the batch correction of three datasets. c Analysis of the scale-free fit index for various soft-threshold powers; the red line was set at 0.90. d Analysis of the mean connectivity for various soft-threshold powers. e Clustering dendrogram of genes in the co-expression network. f Clustering of all modules, the red line indicates the height cutoff (0.25). g Cluster of merged modules. h Identification of weighted gene co-expression network modules associated with DM. i The MM versus GS scatter plot of the turquoise module. j Functional annotation of the 120 hub-genes involved in the turquoise module. k The major PPI network analysis of the top 10 hub-genes from 120 hub-genes through cytoHubba software through PPI and GO/KEGG enrichment analysis of people) was obtained. We calculated the median abso- core genes in key modules. Second, we searched DEGs lute deviation for each gene, sorted the values from between myocarditis and normal myocardium, and the large to small, and then selected the top 5000 genes same PPI and GO/KEGG enrichment analysis was used for WGCNA. The expression data map of these 5000 to screen for hub-genes associated with the immune genes was constructed into a gene co-expression net- system process in myocarditis. The common hub-genes work using the WGCNA package in R software. By set- related to immune processes in both DM and myocar- ting the soft-threshold power as 6 (scale-free R2 = 0.92, ditis were obtained by intersecting the above two sets slope = -2.15; Fig .  2c, d) and cut height as 0.25, we of hub-genes. The function of these common hub-genes acquired 12 modules (Fig.  2e–g), genes that cannot be and their relationship with immune infiltrating cells were included in any module were added to the grey module then investigated. Finally, we used miRNet to predict reg- and rejected in subsequent analyses. ulatory miRNAs of the common hub-genes and validated From the heatmap of module-trait correlations, we their expression in online datasets and clinical samples. found that the turquoise module was the most highly correlated with DM (correlation coefficient = 0.61, Identification and analysis of key module of DM by WGCNA p = 2e − 05; Fig. 2h). We analyzed the correlation between To construct a gene co-expression network, the data module membership and gene significance in the tur - of three series matrix files were downloaded from the quoise module. The results showed that module mem - GEO database. A principal component analysis was bership in the turquoise module (r = 0.84, p = 1e − 200) performed to visualize the grouping of read counts was significantly correlated with gene significance for and identify batch effects. Then, we merged these three DM. The turquoise module contained 954 genes; 120 datasets into one dataset and corrected the batch using genes were identified as core genes with high MM (> 0.8) the SVA software package. Figure  2b shows that the and GS (> 0.2) values (Fig.  2i). We found that the core inter-batch variations are effectively removed after data genes were the most enriched in the immune system normalization. Finally, a normalized gene expression process through GO and KEGG enrichment analysis on matrix file containing 42 samples (23 DM, 19 healthy the metascape website (Fig. 2j). In the major PPI network Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 6 of 12 Fig. 3 Screening of hub-genes related to the immune system process in inflammatory cardiomyopathy. a The volcano plot of DEGs in inflammatory cardiomyopathy. b Heat map clustering of the DEGs between normal controls and inflammatory cardiomyopathy patients. c The major PPI network analyzing of top 10 hub-genes from downregulated genes. d The major PPI network analyzing of top 10 hub-genes from upregulated genes. e Functional annotation of the 20 hub-genes contained in c and d. f A Venn diagram showing the number of commonly expressed genes between hub-genes related to the immune system process in inflammatory cardiomyopathy and DM analysis of top 10 hub-genes from 120 hub-genes through related to the immune system process in inflammatory the cytoHubba software, the shade of the node’s color cardiomyopathy and hub-genes in DM using the VennDi- reflects the degree of connectivity (Fig.  2k); all 10 hub- agram package yielded 4 common DEGs: IFIT3, OAS3, genes were enriched in the immune system process. ISG15, and RSAD2 (Fig. 3f ). Identification and analysis of DEGs in myocarditis database Evaluation of immune cell infiltration and immune cell We identified 570 DEGs composed of 253 upregulated correlation analysis and 317 downregulated genes by using the limma pack- CIBERSORT analytical tool calculated the fractions of age. The volcano plot (Fig.  3a) showed DEGs between 22 types of leukocyte subpopulations in myocardial tis- inflammatory cardiomyopathy and normal controls. We sue and skeletal muscle samples respectively, including imported 253 upregulated and 317 downregulated genes naïve B cells, memory B cells, plasma B cells, CD8 + T into STRING database to construct the PPI network cells, CD4 + naïve T cells, CD4 + memory resting T complex respectively; we then used cytoHubba App in cells, CD4 + memory activated T cells, follicular helper cytoscape to examine hub-genes based on the “degree” T cells, regulatory T cells (Tregs), gamma delta T cells, algorithm. The genes ISG15, IFIT3, XAF1, RSAD2, IGF1, resting natural killer (NK) cells, activated NK cells, OAS3, IFI44, SAMD9L, IFI44L, and TLR3 were the top monocytes, M0, M1, and M2 macrophages, resting and ten upregulated genes (Fig.  3d), and the genes EGFR, activated myeloid dendritic cells, resting and activated CDH1, WDTC1, NGF, BYSL, CCL2, TGFB1, SOCS3, mast cells, eosinophils, and neutrophils. The violin plot POLR1A, and NOL6 were the top ten downregulated of the immune cell infiltration difference showed that M1 genes (Fig.  3c). GO and KEGG enrichment analyses of macrophages in the DM were higher than that in the con- the above 20 hub-genes were performed using the metas- trol group, while M2 macrophages in both the DM and cape (Fig. 3e). Among them, IFIT3, OAS2, ISG15, XAF1, the inflammatory cardiomyopathy group were higher and RSAD2 were enriched in the immune system pro- than that in the control group (Fig.  4). Then, the spear - cess. The subsequent intersection between hub-genes man correlation coefficient between hub-genes and the Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 7 of 12 Fig. 4 Immune cell infiltration analysis. a Heat map of relative proportions of 22 infiltrated immune cells in patients with DM. b Violin chart of the abundance of each type of immune cell infiltration in DM and control groups. c The correlation analysis of hub-genes and M2 macrophages in DM. d Heat map of relative proportions of 22 infiltrated immune cells in patients with inflammatory cardiomyopathy. e Violin chart of the abundance of each type of immune cell infiltration in inflammatory cardiomyopathy and control groups. f The correlation analysis of hub-genes and M2 macrophages in inflammatory cardiomyopathy infiltration level of the immune cell was calculated. As a the common hub-genes and identified DEGs between the result, M2 macrophages were positively associated with two clusters. As shown in Fig.  5c, d, DEGs were mainly the expression of IFIT3, OAS3, ISG15, and RSAD2 in enriched in type I interferon signaling pathway, cellu- patients with inflammatory cardiomyopathy and dermat - lar response to type I interferon, and response to type I omyositis, respectively (Fig. 4c, f ). interferon. Prediction and validation of potential miRNAs targeting Validation hub‑genes with GEO databases hub‑genes To further validate the expression of IFIT3, OAS3, ISG15, We applied the miRNet database to screen the targeted and RSAD2 in myocardial tissue and skeletal muscle tis- miRNAs of ISG15, IFIT3, RSAD2, and OAS3. As illus- sue, we selected GSE128470 and GSE35182 as testing trated in Fig.  6a, a total of 122 miRNAs were predicted, datasets. As shown in Fig.  5a, the expression levels of 23 with association with three or more genes, six of which IFIT3, OAS3, ISG15, and RSAD2 were verified in myo - were confirmed to be upregulated in serum exosomes of cardial tissue between myocarditis mice and normal con- patients with myocarditis than normal control (Fig. 6b). trol. Then, the expression levels of IFIT3, OAS3, ISG15, and RSAD2 were also significantly higher in DM skeletal Validation of miRNAs in serum samples muscle tissues than in the normal controls (Fig. 5b). To verify the clinical application potential of miR-27b-3p, miR-130a-3p, miR-1-3p, miR-133a-3p, miR-16-5p, and Functional enrichment analysis of IFIT3, OAS3, ISG15, miR-146a-5p, comparative analysis of these 6 micro- and RSAD2 RNAs between DM with myocardial injury and DM To further explore the potential function of the com- without myocardial injury was performed. The base - mon hub-genes, GO/KEGG enrichment analysis was line demographic and clinical characteristics of 10 DM performed. We divided the samples from the GSE128470 patients were shown in Table 1. The level of miR-146a-5p dataset into a high-expression group and a low-expres- expression was statistically significantly higher in the sion group according to the median expression level of Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 8 of 12 Fig. 5 Validation and functional enrichment analysis of hub-genes in GEO databases. a Validation of hub-genes in GSE35182. b Validation of hub-genes in GSE128470. c The enriched biological process, cell component, molecular function, and KEGG pathways of DEGs (DEGs between two clusters divided by the expression of common hub-genes). d Circos plot to indicate the relationship between genes and biological process terms DM with myocardial injury group than the other group Autopsy studies have revealed that myocarditis is the (p = 0.0009) (Fig. 6c). predominant pathological feature of myocardial injury in DM, with the immune system thought to play a pivotal role in the development of both conditions. Based on the Discussion above two reasons, we used an integrated bioinformat- Although the cause of DM pathogenesis remains unclear, ics analysis to screen hub-genes related to the immune previous studies have established the existence of myo- system process in DM and myocarditis respectively. The cardial injury in DM. Autopsy studies indicate that intersection of these two sets of genes represents a group myocarditis is the most frequent pathological manifes- of common hub-genes that are believed to be linked to tation [3, 31]. Some serological biomarkers may serve myocardial injury in DM. This approach has been suc - as a screening tool for myocardial injury in DM such as cessfully applied in a variety of biological contexts to cTNI and GDF-15 [32, 33]. Despite this, little research identify common risk genes and mechanisms associated has explored the genetic underpinnings of myocardial with multiple disease phenotypes [34–36]. In our study, injury in DM. Our study has identified several common the similarity of immune cell infiltration in the myocar - hub-genes that are associated with both the immune dium and skeletal muscle could also in turn suggest that process of myocarditis and DM. These genes may play a immune system processes may play a role in the process role in the onset of myocardial injury in DM. By identify- of myocardial injury in dermatomyositis. ing miRNAs that target these genes and validating their We finally identified 4 common hub-genes related potential, we have found miR-146a-5p to be a promising to the immune process of myocarditis and DM: IFIT3, biomarker for detecting myocardial injury in DM. OAS3, ISG15, and RSAD2. Gene function enrichment analysis showed these genes were mainly enriched Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 9 of 12 Fig. 6 Screening and validation of potential miRNAs targeting hub-genes. a An Interaction network of four hub-genes and potential miRNAs-targeted. b The volcano plot of DE-miRNAs between myocarditis and normal control in GSE147517. c Validation of miR-146a-5p expression in the serum of patients with DM systemic sclerosis [37–39]. Several previous studies Table 1 The characteristics of DM patients between two groups with and without myocardial injury have confirmed that the type I IFN signaling pathway plays a prominent role in DM, and the expression lev- DM with DM without p value els of it is associated with DM activity [40–45]. Cassius myocardial myocardial injury injury (n = 5) (n = 5) et al. reported type I interferon signature was also to be highly expressed in the MDA5 + DM subtype [43]. In Age, year 47.80 ± 6.61 52.00 ± 9.43 0.441 fact, a recent study suggests that inhibition of the type Female (n, %) 5 (100%) 5 (100%) 1 I IFN signaling pathway may reduce cardiovascular risk Disease duration, 6.20 ± 4.66 3.20 ± 2.77 0.251 in SLE patients [46]. month Given that immune cells play an essential role in the cTNT, ng/L 80.85 ± 14.95 168.21 ± 125.36 0.347 process of myocardial injury in DM, we sought to inves- CKMB, U/L 70.94 ± 128.15 218.36 ± 190.35 0.189 tigate the infiltration of immune cells in both DM and Pro-BNP, pg/ml 83.98 ± 70.58 198.93 ± 164.78 0.246 myocarditis patients. We found that T cells and mac- rophages comprise the majority of infiltrated immune cells in the skeletal muscle of DM, which is consistent in type I interferon (IFN) signaling pathway, cellular with a recent study [47]. Furthermore, we observed an response to type I interferon, and response to type I increase in both M1 and M2 macrophages in DM patients interferon. The maladaptive immune response created compared to healthy controls. Prior studies also revealed by type I IFN signaling is upregulated in many autoim- macrophages infiltration in the muscle is involved in mune diseases such as systemic lupus erythematosus the development and progression of DM [48, 49] and is (SLE), rheumatoid arthritis, Sjogren’s syndrome, and associated with disease severity [50]. Increasing evidence Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 10 of 12 showed that immune cell infiltration in the myocardium sample size will be needed in future studies to verify the has adverse effect on heart function recently [51–53]. role of miR-146a-5p as a biomarker predicting myocar- In the myocardium, macrophages are one of the most dial injury in DM. important cardiac immune cells and the central regulator of immune systems. In the past, macrophages were clas- sified into M1 and M2 types by their surface molecules. Conclusion Further, research indicated that M1 macrophages have a Our study identified 4 common hub-genes related to the pro-inflammatory phenotype with anti-pathogen activ - immune system process of myocarditis. We speculated ity while M2 macrophages promote anti-inflammatory that these genes may play a role in the process of myocar- effects and tissue repair responses [54]. In animal models dial injury in DM. Serum miR-146a-5p could be a poten- of experimental autoimmune myocarditis and viral myo- tial biomarker to predict myocardial injury in DM. carditis, the acute phase of myocarditis is dominated by pro-inflammatory macrophages, while the chronic phase Abbreviations is dominated by M2 macrophages [55–57]. The activation DM Dermatomyositis of the autoimmune system will eventually lead to exces- IIM Idiopathic inflammatory myopathy NCBI-GEO National C enter Biotechnology Information Gene Expression sive accumulation and transformation of macrophages, Omnibus resulting in myocardial inflammation and fibrosis [58, PCA Principal component analysis 59]. In this study, the M2 macrophages of myocardi- WGCNA Weighted gene co-expression network analysis TOM Topological overlap matrix tis patients increased while M1 macrophages were not GS Gene significance statistically different from normal controls, which may MM Module membership be due to the fact that myocardium specimens in the STRING Search Tool for the Retrieval of Interacting Genes FC Fold-change selected dataset were in the chronic phase of myocarditis. LGE Late gadolinium enhancement Considering the similarities of immune cell infiltration in DEGs Differentially expressed genes the myocardium and skeletal muscle, we speculated that PPI Protein-protein interaction GO Gene ontology the disorder in macrophages might play a potentially sig- GSEA Gene function enrichment analysis nificant role in the process of myocardial injury in DM. IFN Type I interferon Previous studies have shown that miRNAs can be used Acknowledgements as markers in a variety of cardiovascular diseases [60]. We thank all the patients and healthy donors involved in the study. miR-146a-5p is an important regulator of the immune response and inflammation [61, 62] and is abundant in Authors’ contributions Study conception and design was performed by YZ, LS, DL, XS, YZ, QW, and immune cells and the heart [63, 64]. It has been impli- LZ. Data collection was performed by YZ, LS, DL, MD, XS, and YZ. YZ, LS, DL, Y T, cated in cardiac hypertrophy, ischemia/reperfusion WQ, JD, and MD analyzed and interpreted the data. All authors were involved injury, peripartum cardiomyopathy, doxorubicin toxic- in drafting the article or making critical revisions for important intellectual content, and the authors read and approved the submitted manuscript. ity, diabetic cardiomyopathy, and atherosclerosis [65–70]. The increased presence of circulating miR-146a-5p has Funding been reported in patients with spontaneous coronary This work was supported by the National Natural Science Foundation of China (81970723). artery dissection, aortic dissection, and acute coronary syndromes [71–73]. Our study found that serum miR- Availability of data and materials 146a-5p was significantly elevated in DM patients with The data used and/or analyzed in the current study are available from the cor- responding author on reasonable request. myocardial injury than without myocardial injury, sug- gesting the potential of miR-146a-5p as a biomarker for Declarations assessing myocardial injury in DM. There were certain limitations in our study. First, our Ethics approval and consent to participate study was based on bioinformatics analysis from public This study was approved by the medical ethics committee of The First Affili- ated Hospital of Nanjing Medical University. Ethical approval was obtained datasets, which may not fully reflect the actual situation. for our single-center cross-sectional study, and the need to obtain informed Secondly, due to the difficulty of obtaining cardiac sam - consent was waived (2020-SR-228). ples from DM patients, we analyzed the gene sets of myo- Consent for publication carditis and DM separately. Further in  vitro and in  vivo Not applicable. experiments are needed to confirm the role of common hub-genes in DM with myocardial injury. Thirdly, we Competing interests The authors declare no competing interests. searched only one dataset containing myocardial speci- mens in myocarditis patients, so we used murine myo- cardium for our subsequent validation. Finally, a larger Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 11 of 12 Author details customizable protein-protein networks, and functional characteriza- Department of Cardiology, The First Affiliated Hospital of Nanjing Medical tion of user-uploaded gene/measurement sets. Nucleic Acids Res. University, Nanjing, China. Department of Rheumatology, The First Affiliated 2021;49(D1):D605–12. Hospital of Nanjing Medical University, Nanjing, China. Department of Radiol- 22. von Mering C, Huynen M, Jaeggi D, Schmidt S, Bork P, Snel B. STRING: a ogy, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. database of predicted functional associations between proteins. Nucleic Acids Res. 2003;31(1):258–61. Received: 11 January 2023 Accepted: 20 April 2023 23. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–504. 24. Lin CY, Chin CH, Wu HH, Chen SH, Ho CW, Ko MT. Hubba: hub objects analyzer–a framework of interactome hubs identification for network References biology. Nucleic Acids Res. 2008;36( Web Server issue):W438-443. 1. DeWane ME, Waldman R, Lu J. Dermatomyositis: clinical features and 25. Fan Y, Siklenka K, Arora SK, Ribeiro P, Kimmins S, Xia J. miRNet - dissecting pathogenesis. J Am Acad Dermatol. 2020;82(2):267–81. miRNA-target interactions and functional associations through network- 2. Sena P, Gianatti A, Gambini D. Dermatomyositis: clinicopathological corre- based visual analysis. Nucleic Acids Res. 2016;44( W1):W135-141. lations. G Ital Dermatol Venereol. 2018;153(2):256–64. 26. Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two parts). N 3. Zhang L, Wang GC, Ma L, Zu N. Cardiac involvement in adult polymyositis Engl J Med. 1975;292(7):344–7. or dermatomyositis: a systematic review. Clin Cardiol. 2012;35(11):686–91. 27. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using 4. Kariyanna PT, Jayarangaiah A, Mahmood A, Hare M, Taklalsingh N, McFar- real-time quantitative PCR and the 2(-Delta Delta C( T )) Method. Methods. lane IM. Rapid progression of heart failure in a patient with idiopathic 2001;25(4):402–8. inflammatory myopathy. Am J Med Case Rep. 2018;6(8):157–60. 28. Bondarenko O, Beek AM, Hofman MB, Kuhl HP, Twisk JW, van Dockum WG, 5. Yafasova A, Diederichsen LP, Schou M, Sun G, Torp-Pedersen C, Gislason Visser CA, van Rossum AC. Standardizing the definition of hyperenhance - GH, Fosbol EL, Kober L, Butt JH. Increased long-term risk of heart failure ment in the quantitative assessment of infarct size and myocardial viabil- and other adverse cardiac outcomes in dermatomyositis and polymyosi- ity using delayed contrast-enhanced CMR. J Cardiovasc Magn Reson. tis: Insights from a nationwide cohort. J Intern Med. 2021;290(3):704–14. 2005;7(2):481–5. 6. Denbow CE, Lie JT, Tancredi RG, Bunch TW. Cardiac involvement in 29. Yu L, Sun J, Sun J, Li J, Dong Y, Zhou X, Greiser A, Han Y, Zhang Q, Xie Q, polymyositis: a clinicopathologic study of 20 autopsied patients. Arthritis et al. Early detection of myocardial involvement by T1 mapping of cardiac Rheum. 1979;22(10):1088–92. MRI in idiopathic inflammatory myopathy. J Magn Reson Imaging. 7. Haupt HM, Hutchins GM. The heart and cardiac conduction system in 2018;48(2):415–22. polymyositis-dermatomyositis: a clinicopathologic study of 16 autopsied 30. Haaf P, Garg P, Messroghli DR, Broadbent DA, Greenwood JP, Plein S. patients. Am J Cardiol. 1982;50(5):998–1006. Cardiac T1 Mapping and Extracellular Volume (ECV ) in clinical practice: a 8. Diederichsen LP, Simonsen JA, Diederichsen AC, Hvidsten S, Hougaard comprehensive review. J Cardiovasc Magn Reson. 2016;18(1):89. M, Junker P, Sondergaard K, Lundberg IE, Tvede N, Sandgaard NC, et al. 31. Sultan SM, Ioannou Y, Moss K, Isenberg DA. Outcome in patients with idi- Cardiac abnormalities in adult patients with polymyositis or dermatomy- opathic inflammatory myositis: morbidity and mortality. Rheumatology ositis as assessed by noninvasive modalities. Arthritis Care Res (Hoboken). (Oxford). 2002;41(1):22–6. 2016;68(7):1012–20. 32. Chen F, Peng Y, Chen M. Diagnostic approach to cardiac involvement in 9. Prisco F, Papparella S, Paciello O. The correlation between cardiac and idiopathic inflammatory myopathies. Int Heart J. 2018;59(2):256–62. skeletal muscle pathology in animal models of idiopathic inflammatory 33. Qiu M, Sun X, Qi X, Liu X, Zhang Y, Zhang N, Lu F, Liu W, Changjing F, myopathies. Acta Myol. 2020;39(4):313–9. Wang Q, et al. The diagnostic value of GDF-15 for myocardial involve- 10. Ringner M. What is principal component analysis? Nat Biotechnol. ment in idiopathic inflammatory myopathy. Rheumatology (Oxford). 2008;26(3):303–4. 2021;60(6):2826–33. 11. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation 34. Sezin T, Vorobyev A, Sadik CD, Zillikens D, Gupta Y, Ludwig RJ. Gene network analysis. BMC Bioinformatics. 2008;9:559. expression analysis reveals novel shared gene signatures and candidate 12. Zhang B, Horvath S. A general framework for weighted gene co-expres- molecular mechanisms between pemphigus and systemic lupus erythe- sion network analysis. Stat Appl Genet Mol Biol. 2005;4:Article17. matosus in CD4(+) T cells. Front Immunol. 1992;2017:8. 13. Smyth GK. Linear models and empirical bayes methods for assessing dif- 35. Zou R, Zhang D, Lv L, Shi W, Song Z, Yi B, Lai B, Chen Q, Yang S, Hua P. Bio- ferential expression in microarray experiments. Stat Appl Genet Mol Biol. informatic gene analysis for potential biomarkers and therapeutic targets 2004;3:Article3. of atrial fibrillation-related stroke. J Transl Med. 2019;17(1):45. 14. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma pow- 36. Zhu Y, Ding X, She Z, Bai X, Nie Z, Wang F, Wang F, Geng X. Explor- ers differential expression analyses for RNA-sequencing and microarray ing shared pathogenesis of Alzheimer’s disease and type 2 diabetes studies. Nucleic Acids Res. 2015;43(7):e47. mellitus via co-expression networks analysis. Curr Alzheimer Res. 15. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, Hoang CD, 2020;17(6):566–75. Diehn M, Alizadeh AA. Robust enumeration of cell subsets from tissue 37. Hill CL, Zhang Y, Sigurgeirsson B, Pukkala E, Mellemkjaer L, Airio A, Evans expression profiles. Nat Methods. 2015;12(5):453–7. SR, Felson DT. Frequency of specific cancer types in dermatomyositis and 16. Ito K, Murphy D. Application of ggplot2 to Pharmacometric Graphics. CPT polymyositis: a population-based study. Lancet. 2001;357(9250):96–100. Pharmacometrics Syst Pharmacol. 2013;2:e79. 38. Chen LY, Cui ZL, Hua FC, Yang WJ, Bai Y, Lan FH. Bioinformatics analy- 17. Tripathi S, Pohl MO, Zhou Y, Rodriguez-Frandsen A, Wang G, Stein DA, sis of gene expression profiles of dermatomyositis. Mol Med Rep. Moulton HM, DeJesus P, Che J, Mulder LC, et al. Meta- and orthogonal 2016;14(4):3785–90. integration of influenza “OMICs” data defines a role for UBR4 in virus bud- 39. Siddiqi KZ, Zinglersen AH, Iversen KK, Rasmussen NS, Nielsen CT, ding. Cell Host Microbe. 2015;18(6):723–35. Jacobsen S. A cluster of type II interferon-regulated genes associates with 18. Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, disease activity in patients with systemic lupus erythematosus. J Autoim- Benner C, Chanda SK. Metascape provides a biologist-oriented resource mun. 2022;132:102869. for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. 40. Baechler EC, Bilgic H, Reed AM. Type I interferon pathway in adult and 19. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illu- juvenile dermatomyositis. Arthritis Res Ther. 2011;13(6):249. mina sequence data. Bioinformatics. 2014;30(15):2114–20. 41. Gitiaux C, Latroche C, Weiss-Gayet M, Rodero MP, Duffy D, Bader-Meunier 20. Vizcaino JA, Deutsch EW, Wang R, Csordas A, Reisinger F, Rios D, Dianes B, Glorion C, Nusbaum P, Bodemer C, Mouchiroud G, et al. Myogenic JA, Sun Z, Farrah T, Bandeira N, et al. ProteomeXchange provides glob- progenitor cells exhibit type I interferon-driven proangiogenic proper- ally coordinated proteomics data submission and dissemination. Nat ties and molecular signature during juvenile dermatomyositis. Arthritis Biotechnol. 2014;32(3):223–6. Rheumatol. 2018;70(1):134–45. 21. Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, Doncheva NT, Legeay M, Fang T, Bork P, et al. The STRING database in 2021: Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 12 of 12 42. Ladislau L, Suarez-Calvet X, Toquet S, Landon-Cardinal O, Amelin D, Depp 63. Feng B, Chen S, McArthur K, Wu Y, Sen S, Ding Q, Feldman RD, Chakrabarti M, Rodero MP, Hathazi D, Duffy D, Bondet V, et al. JAK inhibitor improves S. miR-146a-mediated extracellular matrix protein production in chronic type I interferon induced damage: proof of concept in dermatomyositis. diabetes complications. Diabetes. 2011;60(11):2975–84. Brain. 2018;141(6):1609–21. 64. Shimada BK, Yang Y, Zhu J, Wang S, Suen A, Kronstadt SM, Jeyaram A, 43. Cassius C, Amode R, Delord M, Battistella M, Poirot J, How-Kit A, Lepel- Jay SM, Zou L, Chao W. Extracellular miR-146a-5p induces cardiac innate letier C, Jachiet M, de Masson A, Frumholtz L, et al. MDA5(+) dermato- immune response and cardiomyocyte dysfunction. Immunohorizons. myositis is associated with stronger skin type I interferon transcriptomic 2020;4(9):561–72. signature with upregulation of IFN-kappa transcript. J Invest Dermatol. 65. Lin G, Huang J, Chen Q, Chen L, Feng D, Zhang S, Huang X, Huang Y, Lin 2020;140(6):1276-1279 e1277. Q. miR-146a-5p mediates intermittent hypoxia-induced injury in H9c2 44. Kuriyama Y, Shimizu A, Kanai S, Oikawa D, Tokunaga F, Tsukagoshi H, Ishi- cells by targeting XIAP. Oxid Med Cell Longev. 2019;2019:6581217. kawa O. The synchronized gene expression of retrotransposons and type 66. Shu L, Zhang W, Huang G, Huang C, Zhu X, Su G, Xu J. Troxerutin attenu- I interferon in dermatomyositis. J Am Acad Dermatol. 2021;84(4):1103–5. ates myocardial cell apoptosis following myocardial ischemia-reperfusion 45. Buang N, Tapeng L, Gray V, Sardini A, Whilding C, Lightstone L, Cairns TD, injury through inhibition of miR-146a-5p expression. J Cell Physiol. Pickering MC, Behmoaras J, Ling GS, et al. Type I interferons affect the 2019;234(6):9274–82. metabolic fitness of CD8(+) T cells from patients with systemic lupus 67. Dong M, Chen D, Zhu Y, Yang S, Kumar S, Zhang R, Zhou Y, Yang Z, Zheng erythematosus. Nat Commun. 2021;12(1):1980. N, Zhu T, et al. Impaired regulation of MMP2/16-MLCK3 by miR-146a-5p 46. Casey KA, Smith MA, Sinibaldi D, Seto NL, Playford MP, Wang X, Carlucci increased susceptibility to myocardial ischemic injury in aging mice. PM, Wang L, Illei G, Yu B, et al. Modulation of cardiometabolic disease Cardiovasc Res. 2022:cvac104. markers by type I interferon inhibition in systemic lupus erythematosus. 68. Oh JG, Watanabe S, Lee A, Gorski PA, Lee P, Jeong D, Liang L, Liang Y, Arthritis Rheumatol. 2021;73(3):459–71. Baccarini A, Sahoo S, et al. miR-146a suppresses SUMO1 expression 47. Yin R, Wang G, Zhang L, Li T, Liu S. Dermatomyositis: immunological and induces cardiac dysfunction in maladaptive hypertrophy. Circ Res. landscape, biomarkers, and potential candidate drugs. Clin Rheumatol. 2018;123(6):673–85. 2021;40(6):2301–10. 69. Wang X, Ha T, Liu L, Zou J, Zhang X, Kalbfleisch J, Gao X, Williams D, Li C. 48. Ragusa F. Dermatomyositis and MIG. Clin Ter. 2019;170(2):e142–7. Increased expression of microRNA-146a decreases myocardial ischaemia/ 49. Jiang T, Huang Y, Liu H, Xu Q, Gong Y, Chen Y, Hu X, Han Z, Gao M. reperfusion injury. Cardiovasc Res. 2013;97(3):432–42. Reduced miR-146a promotes REG3A expression and macrophage migra- 70. Cheng HS, Besla R, Li A, Chen Z, Shikatani EA, Nazari-Jahantigh M, tion in polymyositis and dermatomyositis. Front Immunol. 2020;11:37. Hammoutene A, Nguyen MA, Geoffrion M, Cai L, et al. Paradoxical sup - 50. Peng QL, Zhang YL, Shu XM, Yang HB, Zhang L, Chen F, Lu X, Wang GC. pression of atherosclerosis in the absence of microRNA-146a. Circ Res. Elevated serum levels of soluble CD163 in polymyositis and dermato- 2017;121(4):354–67. myositis: associated with macrophage infiltration in muscle tissue. J 71. Zhelankin AV, Stonogina DA, Vasiliev SV, Babalyan KA, Sharova EI, Doludin Rheumatol. 2015;42(6):979–87. YV, Shchekochikhin DY, Generozov EV, Akselrod AS. Circulating extracel- 51. Carrillo-Salinas FJ, Ngwenyama N, Anastasiou M, Kaur K, Alcaide P. Heart lular miRNA analysis in patients with stable CAD and acute coronary inflammation: immune cell roles and roads to the heart. Am J Pathol. syndromes. Biomolecules. 2021;11(7):962. 2019;189(8):1482–94. 72. Lozano-Prieto M, Adlam D, Garcia-Guimaraes M, Sanz-Garcia A, Vera-Tome 52. Lafuse WP, Wozniak DJ, Rajaram MVS. Role of cardiac macrophages on P, Rivero F, Cuesta J, Bastante T, Baranowska-Clarke AA, Vara A, et al. cardiac inflammation, fibrosis and tissue repair. Cells. 2020;10(1):51. Differential miRNAs in acute spontaneous coronary artery dissection: 53. Kawada JI, Takeuchi S, Imai H, Okumura T, Horiba K, Suzuki T, Torii Y, pathophysiological insights from a potential biomarker. EBioMedicine. Yasuda K, Imanaka-Yoshida K, Ito Y. Immune cell infiltration landscapes 2021;66:103338. in pediatric acute myocarditis analyzed by CIBERSORT. J Cardiol. 73. Fang L, Ellims AH, Moore XL, White DA, Taylor AJ, Chin-Dusting J, Dart AM. 2021;77(2):174–8. Circulating microRNAs as biomarkers for diffuse myocardial fibrosis in 54. Italiani P, Boraschi D. From monocytes to M1/M2 macrophages: pheno- patients with hypertrophic cardiomyopathy. J Transl Med. 2015;13:314. typical vs. functional differentiation. Front Immunol. 2014;5:514. 55. Anzai A, Mindur JE, Halle L, Sano S, Choi JL, He S, McAlpine CS, Chan CT, Publisher’s Note Kahles F, Valet C, et al. Self-reactive CD4(+) IL-3(+) T cells amplify autoim- Springer Nature remains neutral with regard to jurisdictional claims in pub- mune inflammation in myocarditis by inciting monocyte chemotaxis. J lished maps and institutional affiliations. Exp Med. 2019;216(2):369–83. 56. Clemente-Casares X, Hosseinzadeh S, Barbu I, Dick SA, Macklin JA, Wang Y, Momen A, Kantores C, Aronoff L, Farno M, et al. A CD103(+) conventional dendritic cell surveillance system prevents development of overt heart failure during subclinical viral myocarditis. Immunity. 2017;47(5):974-989 e978. 57. Hou X, Chen G, Bracamonte-Baran W, Choi HS, Diny NL, Sung J, Hughes D, Won T, Wood MK, Talor MV, et al. The cardiac microenvironment instructs divergent monocyte fates and functions in myocarditis. Cell Rep. 2019;28(1):172-189 e177. 58. Wolf AA, Yanez A, Barman PK, Goodridge HS. The ontogeny of monocyte subsets. Front Immunol. 2019;10:1642. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : 59. Bajpai G, Schneider C, Wong N, Bredemeyer A, Hulsmans M, Nahrendorf M, Epelman S, Kreisel D, Liu Y, Itoh A, et al. The human heart contains fast, convenient online submission distinct macrophage subsets with divergent origins and functions. Nat thorough peer review by experienced researchers in your field Med. 2018;24(8):1234–45. 60. Barwari T, Joshi A, Mayr M. MicroRNAs in cardiovascular disease. J Am Coll rapid publication on acceptance Cardiol. 2016;68(23):2577–84. support for research data, including large and complex data types 61. Sheedy FJ, O’Neill LA. Adding fuel to fire: microRNAs as a new class of • gold Open Access which fosters wider collaboration and increased citations mediators of inflammation. Ann Rheum Dis. 2008;67 Suppl 3:iii50-55. 62. Taganov KD, Boldin MP, Chang KJ, Baltimore D. NF-kappaB-dependent maximum visibility for your research: over 100M website views per year induction of microRNA miR-146, an inhibitor targeted to signal- ing proteins of innate immune responses. Proc Natl Acad Sci U S A. At BMC, research is always in progress. 2006;103(33):12481–6. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Arthritis Research & Therapy Springer Journals

Identification of key biomarkers associated with immune cells infiltration for myocardial injury in dermatomyositis by integrated bioinformatics analysis

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/identification-of-key-biomarkers-associated-with-immune-cells-pr02apet7c

References (84)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
eISSN
1478-6362
DOI
10.1186/s13075-023-03052-4
Publisher site
See Article on Publisher Site

Abstract

Background Dermatomyositis (DM) is an acquired autoimmune disease that can cause damage to various organs, including the heart muscle. However, the mechanisms underlying myocardial injury in DM are not yet fully understood. Methods In this study, we utilized publicly available datasets from the Gene Expression Omnibus (GEO) database to identify hub-genes that are enriched in the immune system process in DM and myocarditis. Weighted gene co- expression network analysis ( WGCNA), differentially expressed genes (DEGs) analysis, protein–protein interaction (PPI), and gene ontology (GO) analysis were employed to identify these hub-genes. We then used the CIBERSORT method to analyze immune cell infiltration in skeletal muscle specimens of DM and myocardium specimens of myocarditis respectively. Correlation analysis was performed to investigate the relationship between key genes and infiltrating immune cells. Finally, we predicted regulatory miRNAs of hub-genes through miRNet and validated their expression in online datasets and clinical samples. Results Using integrated bioinformatics analysis, we identified 10 and 5 hub-genes that were enriched in the immune system process in the database of DM and myocarditis respectively. The subsequent intersections between hub-genes were IFIT3, OAS3, ISG15, and RSAD2. We found M2 macrophages increased in DM and myocarditis com- pared to the healthy control, associating with the expression of IFIT3, OAS3, ISG15, and RSAD2 in DM and myocar- ditis positively. Gene function enrichment analysis (GSEA) showed that IFIT3, OAS3, ISG15, and RSAD2 were mainly enriched in type I interferon (IFN) signaling pathway, cellular response to type I interferon, and response to type I interferon. Finally, we verified that the expression of miR-146a-5p was significantly higher in the DM with myocardial injury than those without myocardial injury (p = 0.0009). Yue Zhang, Linwei Shan, and Dongyu Li contributed equally to this work. Qiang Wang and Lei Zhou contributed equally to this work. *Correspondence: Qiang Wang jerrytortoise@163.com; zhoulei@njmu.edu.cn Lei Zhou jerrytortoise@163.com; zhoulei@njmu.edu.cn Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 2 of 12 Conclusion Our findings suggest that IFIT3, OAS3, ISG15, and RSAD2 may play crucial roles in the underlying mecha- nism of myocardial injury in DM. Serum miR-146a-5p could be a potential biomarker for myocardial injury in DM. Keywords Dermatomyositis, Myocardial injury, Bioinformatic, Immune cells infiltration, Biomarker Background Methods Dermatomyositis (DM) is an acquired autoimmune dis- Data sources ease in which skeletal muscle is targeted by the immune Series matrix files and platform information of GSE1551, system, which is a subset of idiopathic inflammatory GSE48280, GSE5370, GSE128470, GSE1145, GSE35182, myopathy (IIM) [1]. DM is characterized primarily by and GSE147517 were obtained from the National muscle inflammation, proximal muscle weakness, and Center Biotechnology Information Gene Expression cutaneous involvement. Additionally, the disease may Omnibus (NCBI-GEO) (https:// www. ncbi. nlm. nih. gov/ present with extra-muscular symptoms affecting vari - geo). We selected 13 DM patients and 10 normal indi- ous organs including the heart, joints, lungs, and gas- viduals, 5 DM patients, and 5 normal individuals, and 5 trointestinal tract [2]. DM patients and 4 normal individuals from GSE1551, Several studies have identified cardiovascular GSE48280, and GSE5370, respectively, and the samples involvement in DM is a feared threat to prognosis and a were all skeletal muscle biopsy specimens. We merged frequent cause of death in several studies [3–5]. Due to the GSE1551, GSE48280, and GSE5370 and used the the structural and functional similarities between skel- SVA software package to correct the bath. Then, we used etal and cardiac muscle, it is supposed that what affects principal component analysis (PCA) [10] to evaluate the skeletal muscle may be related to damage to cardiac results of the correction. Finally, we obtained a normal- muscle. In autopsy studies, myocarditis was the most ized gene expression matrix file containing 42 samples common pathologic features [6, 7], and previous imag- (23 DM patients, and 19 normal individuals). We chose ing studies also indicate inflammation as an underlying samples of myocardium obtained from 11 normal indi- pathology [8]. In different animal models of IIM, the viduals and 7 inflammatory cardiomyopathy patients due myocardium shows an inflammation similar to that of to viral myocarditis in GSE1145. In the section of valida- skeletal muscle morphologically [9]. tion, we chose GSE128470 which includes 12 samples of Despite the accumulating evidence linking DM and muscle obtained from normal individuals and 12 samples cardiovascular involvement, the extent of this link and from DM patients. Furthermore, we used GSE35182 to the underlying mechanisms are not yet fully under- validate the expression of genes between chronic myocar- stood. Moreover, specific recommendations for the ditis and normal mice; there are 6 mice per group. Finally, treatment of patients with DM and cardiovascular we validated the predicted target miRNAs expression in involvement are lacking. Thus, it is essential to explore GSE147517 comprising 5 myocarditis and 5 normal indi- the molecular characteristics and mechanisms of myo- viduals serum specimens. cardial injury in DM before developing screening rec- ommendations and treatment strategies for patients with DM and myocardial injury. Over the past decades, Construction of weighted gene co‑expression network gene microarray technology together with integrated analysis bioinformatic analyses has been performed to provide The WGCNA package in R was utilized to build a co- tremendous assistance in identifying novel key genes expression network targeting the top 5000 genes with related to various diseases. In this study, we identified median absolute deviation [11, 12]. The R function pick - the co-expression genes related to the immune sys- SoftThreshold was used to calculate the soft threshold - tem process between DM and myocarditis, explored ing power β, to which co-expression similarity is raised the association between these genes and immune cells to calculate adjacency. Then, we converted the adjacency infiltrating skeletal muscle or myocardium, and pre - into a topological overlap matrix (TOM) to measure the dicted regulatory miRNAs of the hub-genes. Finally, we network connectivity of genes. Genes with similar pat- verified our results through online datasets and clinical terns were clustered into the same modules (minimum samples. With the above approaches, it is hoped that size = 30) using average linkage hierarchical cluster- our results may provide a preliminary insight into the ing, which were represented by branches and different mechanism of myocardial injury in DM and a search colors of the cluster tree, constructed module relation- for possible biomarkers. ships, calculation of the correlation between gene mod- ules and phenotypes, and the modules related to clinical Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 3 of 12 traits were identified. Gene significance (GS) and module Prediction of potential hub‑gene related target miRNAs membership (MM) were calculated to relate modules to We used miRNet (www. mirnet. ca/) [25], a tool that inte- clinical traits. Finally, the highly correlated module was grates data from 11 different miRNA databases, to pre - analyzed to explore its core genes and potential roles. dict regulatory miRNAs of the common hub-genes. Subjects Identification of differentially expressed genes A total of 10 DM patients were enrolled from the First The limma package in R (Version 4.0.3) software was Affiliated Hospital of Nanjing Medical University utilized to screen DEGs between inflammatory car - between January 2020 and January 2021. We divided diomyopathy patients and normal controls [13, 14]. The these patients into two groups: DM with myocardial differentially expressed genes were screened under the injury (n = 5) and DM without myocardial injury (n = 5) condition of |log FC|> 1 and adjusted P value < 0.05. based on the results of the cardiac magnetic resonance examination. The inclusion criteria were as follows: (1) Evaluation of immune cell infiltration age > 18  years old, (2) all the patients fulfilled the 1975 CIBERSORT [15] is a deconvolution algorithm for ana- Bohan and Peter criteria for dermatomyositis [26], and lyzing gene expression data and uses a gene expres- (3) patients who underwent cardiac magnetic reso- sion signature for characterizing the proportion of each nance examination during hospitalization. The exclu - immune cell type. We performed immune infiltration sion criteria were as follows: (1) interstitial lung disease, by using the CIBERSORT.R script downloaded from (2) previous history of cardiac disease, (3) tumor, (4) the CIBERSORT website (https:// ciber sortx. stanf ord. renal insufficiency, and (5) surgery within the six previ - edu/). We used the original CIBERSORT gene signature ous months. Ethical approval was obtained for our sin- file LM22, which defines 22 immune cell subtypes, to gle-center cross-sectional study, and the need to obtain calculate the proportion of 22 immune cells in DM and informed consent was waived (2020-SR-228). inflammatory cardiomyopathy patients. Then, we used the “ggplot2” package [16] to draw violin diagrams to Sample collection and RNA isolation visualize the differences compared to normal controls in Venous blood samples were collected into EDTA tubes immune cell infiltration. We also calculated the Spear - from DM patients within 24  h before and after the car- man correlation coefficient between identified hub-genes diac magnetic resonance examination. Serum was with infiltrating macrophages M2. obtained at room temperature of 3000 rpm for 5 min and frozen at –  80  °C for further use. Isolation of total RNA Functional annotation including miRNA was performed from serum using the Metascape (http:// Metas cape. org/ gp/ index. html) is a free miRNeasy Serum/Plasma Advanced Kit (Qiagen, Ger- web-based analytics tool for comprehensive gene annota- mantown, MD, USA) according to the manufacturer’s tion and analysis resources, combining a GO and KEGG protocol. During the extraction, 3.5 μL of miRNeasy pathway enrichment analysis search to leverage over 40 8 Serum/Plasma Spike-In Control (1.6 × 10 copies/μL of independent knowledge bases [17–19]. To understand the C. elegans miR-39 miRNA mimic) was added to each the function of these core genes and DEGs, we used sample as an internal control. RNA quantity and quality metascape to perform GO and KEGG pathway enrich- were evaluated using the Nanodrop ND-2000 spectro- ment analyses. photometer (Nanodrop Technologies, Wilmington, DE). Directly after isolation, RNA was subjected to the reverse Protein–protein interaction network analysis transcription process. Search Tool for the Retrieval of Interacting Genes (STRING) (Version 11.3, https:// cn. string- db. org/) is a Reverse transcription reaction and real‑time quantitative useful online tool dedicated to analyzing the functional PCR (RT‑qPCR) protein association networks [20–22]. The core genes The expression levels of miR-27b-3p, miR-130a-3p, miR- and DEGs were mapped to the STRING database, and 1-3p, miR-133a-3p, miR-16-5p, and miR-146a-5p were only the experimentally validated interactions with a measured using the Bulge-Loop miRNA qRT-PCR combined score > 0.4 were selected as significant. Sub - Starter Kit (one RT primer and a pair of qPCR primers sequently, the PPI network was visualized by Cytoscape for each set) designed specifically by RiboBio (Guang - software (version 3.5.1) (www. cytos cape. org/) [23]. The zhou, China) in accordance with the manufacturer’s plug-in cytoHubba in Cytoscape was used to screen the instructions. The average expression levels of serum hub-genes from the PPI network, and in our study, the miRNAs were normalized against cel-miR-39 (Qiagen, top ten genes were identified as hub-genes [24]. Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 4 of 12 Germantown, MD, USA). The fold-change (FC) for each Cardiovascular Imaging, Calgary, Canada). The presence miRNA relative to the control was calculated using the of LGE was defined with a signal intensity level increase −ΔΔCT 2 method [27]. The efficiency was tested before of more than five S.D. of remote myocardium on the all −ΔΔCT using the 2 method, and the results were collected short-axis contrast images from base to apex [28]. ECV from the same experiment in three replicates. was calculated by the following equation: ECV = (1— 1 1 post contrast T1 myo native T1 myo HCT) . 1 1 CMR post contrast T1 blood blood T1 myo Determination of hematocrit (HCT) and calculation All participants underwent CMR imaging on a 3.0  T of ECV were completed within 24  h after CMR scan- whole-body scanner (MAGNETOM Skyra, Siemens ning. ECV is a marker of myocardial tissue remodeling Healthcare, Erlangen, Germany) with an 18-chan- and provides a physiologically intuitive unit of measure- nel phase-array coil using ECG gating. Late gadolinium ment. Normal ECV values of 24% ± 3 (3.0  T) have been enhancement (LGE) images were acquired 8–15  min reported in healthy individuals [29]. In our study, we after intravenous administration of gadolinium-DTPA divide patients with LGE positive and ECV > 25% into (Magnevist, Bayer, Berlin, Germany) at a dose of DM with myocardial injury group [30]. 0.2  mmol/kg in short-axis stack using a phase-sensitive inversion-recovery (PSIR) gradient echo sequence. Typi- Statistical analysis cal parameters of motion-corrected basal, mid, and apical All statistical analyses were performed in the R language level of LV short-axis Modified Look-Locker inversion- (Version 4.0.3). All statistical tests were bilateral, and recovery (MOLLI) T1 mapping sequence with a 5(3)3 adjusted P value < 0.05 was statistically significant. scheme before and 15–20 min after intravenous contrast agent injection. The native and post-contrast T1 values of the myocar - Results dium were measured on a region of interest at the myo- Bioinformatic analysis workflow cardial septum of mid-ventricular short-axis slice by two Our workflows are shown in Fig.  1. First, we found key experienced blinded investigators in consensus with modules related to DM using WGCNA and discovered CMR using commercial software (CVI42, Circle hub-genes associated with the immune system process Fig. 1 Flowchart of this study Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 5 of 12 Fig. 2 Screening of hub-genes related to the immune system process in DM. a PCA before the batch correction of three datasets. b PCA after the batch correction of three datasets. c Analysis of the scale-free fit index for various soft-threshold powers; the red line was set at 0.90. d Analysis of the mean connectivity for various soft-threshold powers. e Clustering dendrogram of genes in the co-expression network. f Clustering of all modules, the red line indicates the height cutoff (0.25). g Cluster of merged modules. h Identification of weighted gene co-expression network modules associated with DM. i The MM versus GS scatter plot of the turquoise module. j Functional annotation of the 120 hub-genes involved in the turquoise module. k The major PPI network analysis of the top 10 hub-genes from 120 hub-genes through cytoHubba software through PPI and GO/KEGG enrichment analysis of people) was obtained. We calculated the median abso- core genes in key modules. Second, we searched DEGs lute deviation for each gene, sorted the values from between myocarditis and normal myocardium, and the large to small, and then selected the top 5000 genes same PPI and GO/KEGG enrichment analysis was used for WGCNA. The expression data map of these 5000 to screen for hub-genes associated with the immune genes was constructed into a gene co-expression net- system process in myocarditis. The common hub-genes work using the WGCNA package in R software. By set- related to immune processes in both DM and myocar- ting the soft-threshold power as 6 (scale-free R2 = 0.92, ditis were obtained by intersecting the above two sets slope = -2.15; Fig .  2c, d) and cut height as 0.25, we of hub-genes. The function of these common hub-genes acquired 12 modules (Fig.  2e–g), genes that cannot be and their relationship with immune infiltrating cells were included in any module were added to the grey module then investigated. Finally, we used miRNet to predict reg- and rejected in subsequent analyses. ulatory miRNAs of the common hub-genes and validated From the heatmap of module-trait correlations, we their expression in online datasets and clinical samples. found that the turquoise module was the most highly correlated with DM (correlation coefficient = 0.61, Identification and analysis of key module of DM by WGCNA p = 2e − 05; Fig. 2h). We analyzed the correlation between To construct a gene co-expression network, the data module membership and gene significance in the tur - of three series matrix files were downloaded from the quoise module. The results showed that module mem - GEO database. A principal component analysis was bership in the turquoise module (r = 0.84, p = 1e − 200) performed to visualize the grouping of read counts was significantly correlated with gene significance for and identify batch effects. Then, we merged these three DM. The turquoise module contained 954 genes; 120 datasets into one dataset and corrected the batch using genes were identified as core genes with high MM (> 0.8) the SVA software package. Figure  2b shows that the and GS (> 0.2) values (Fig.  2i). We found that the core inter-batch variations are effectively removed after data genes were the most enriched in the immune system normalization. Finally, a normalized gene expression process through GO and KEGG enrichment analysis on matrix file containing 42 samples (23 DM, 19 healthy the metascape website (Fig. 2j). In the major PPI network Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 6 of 12 Fig. 3 Screening of hub-genes related to the immune system process in inflammatory cardiomyopathy. a The volcano plot of DEGs in inflammatory cardiomyopathy. b Heat map clustering of the DEGs between normal controls and inflammatory cardiomyopathy patients. c The major PPI network analyzing of top 10 hub-genes from downregulated genes. d The major PPI network analyzing of top 10 hub-genes from upregulated genes. e Functional annotation of the 20 hub-genes contained in c and d. f A Venn diagram showing the number of commonly expressed genes between hub-genes related to the immune system process in inflammatory cardiomyopathy and DM analysis of top 10 hub-genes from 120 hub-genes through related to the immune system process in inflammatory the cytoHubba software, the shade of the node’s color cardiomyopathy and hub-genes in DM using the VennDi- reflects the degree of connectivity (Fig.  2k); all 10 hub- agram package yielded 4 common DEGs: IFIT3, OAS3, genes were enriched in the immune system process. ISG15, and RSAD2 (Fig. 3f ). Identification and analysis of DEGs in myocarditis database Evaluation of immune cell infiltration and immune cell We identified 570 DEGs composed of 253 upregulated correlation analysis and 317 downregulated genes by using the limma pack- CIBERSORT analytical tool calculated the fractions of age. The volcano plot (Fig.  3a) showed DEGs between 22 types of leukocyte subpopulations in myocardial tis- inflammatory cardiomyopathy and normal controls. We sue and skeletal muscle samples respectively, including imported 253 upregulated and 317 downregulated genes naïve B cells, memory B cells, plasma B cells, CD8 + T into STRING database to construct the PPI network cells, CD4 + naïve T cells, CD4 + memory resting T complex respectively; we then used cytoHubba App in cells, CD4 + memory activated T cells, follicular helper cytoscape to examine hub-genes based on the “degree” T cells, regulatory T cells (Tregs), gamma delta T cells, algorithm. The genes ISG15, IFIT3, XAF1, RSAD2, IGF1, resting natural killer (NK) cells, activated NK cells, OAS3, IFI44, SAMD9L, IFI44L, and TLR3 were the top monocytes, M0, M1, and M2 macrophages, resting and ten upregulated genes (Fig.  3d), and the genes EGFR, activated myeloid dendritic cells, resting and activated CDH1, WDTC1, NGF, BYSL, CCL2, TGFB1, SOCS3, mast cells, eosinophils, and neutrophils. The violin plot POLR1A, and NOL6 were the top ten downregulated of the immune cell infiltration difference showed that M1 genes (Fig.  3c). GO and KEGG enrichment analyses of macrophages in the DM were higher than that in the con- the above 20 hub-genes were performed using the metas- trol group, while M2 macrophages in both the DM and cape (Fig. 3e). Among them, IFIT3, OAS2, ISG15, XAF1, the inflammatory cardiomyopathy group were higher and RSAD2 were enriched in the immune system pro- than that in the control group (Fig.  4). Then, the spear - cess. The subsequent intersection between hub-genes man correlation coefficient between hub-genes and the Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 7 of 12 Fig. 4 Immune cell infiltration analysis. a Heat map of relative proportions of 22 infiltrated immune cells in patients with DM. b Violin chart of the abundance of each type of immune cell infiltration in DM and control groups. c The correlation analysis of hub-genes and M2 macrophages in DM. d Heat map of relative proportions of 22 infiltrated immune cells in patients with inflammatory cardiomyopathy. e Violin chart of the abundance of each type of immune cell infiltration in inflammatory cardiomyopathy and control groups. f The correlation analysis of hub-genes and M2 macrophages in inflammatory cardiomyopathy infiltration level of the immune cell was calculated. As a the common hub-genes and identified DEGs between the result, M2 macrophages were positively associated with two clusters. As shown in Fig.  5c, d, DEGs were mainly the expression of IFIT3, OAS3, ISG15, and RSAD2 in enriched in type I interferon signaling pathway, cellu- patients with inflammatory cardiomyopathy and dermat - lar response to type I interferon, and response to type I omyositis, respectively (Fig. 4c, f ). interferon. Prediction and validation of potential miRNAs targeting Validation hub‑genes with GEO databases hub‑genes To further validate the expression of IFIT3, OAS3, ISG15, We applied the miRNet database to screen the targeted and RSAD2 in myocardial tissue and skeletal muscle tis- miRNAs of ISG15, IFIT3, RSAD2, and OAS3. As illus- sue, we selected GSE128470 and GSE35182 as testing trated in Fig.  6a, a total of 122 miRNAs were predicted, datasets. As shown in Fig.  5a, the expression levels of 23 with association with three or more genes, six of which IFIT3, OAS3, ISG15, and RSAD2 were verified in myo - were confirmed to be upregulated in serum exosomes of cardial tissue between myocarditis mice and normal con- patients with myocarditis than normal control (Fig. 6b). trol. Then, the expression levels of IFIT3, OAS3, ISG15, and RSAD2 were also significantly higher in DM skeletal Validation of miRNAs in serum samples muscle tissues than in the normal controls (Fig. 5b). To verify the clinical application potential of miR-27b-3p, miR-130a-3p, miR-1-3p, miR-133a-3p, miR-16-5p, and Functional enrichment analysis of IFIT3, OAS3, ISG15, miR-146a-5p, comparative analysis of these 6 micro- and RSAD2 RNAs between DM with myocardial injury and DM To further explore the potential function of the com- without myocardial injury was performed. The base - mon hub-genes, GO/KEGG enrichment analysis was line demographic and clinical characteristics of 10 DM performed. We divided the samples from the GSE128470 patients were shown in Table 1. The level of miR-146a-5p dataset into a high-expression group and a low-expres- expression was statistically significantly higher in the sion group according to the median expression level of Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 8 of 12 Fig. 5 Validation and functional enrichment analysis of hub-genes in GEO databases. a Validation of hub-genes in GSE35182. b Validation of hub-genes in GSE128470. c The enriched biological process, cell component, molecular function, and KEGG pathways of DEGs (DEGs between two clusters divided by the expression of common hub-genes). d Circos plot to indicate the relationship between genes and biological process terms DM with myocardial injury group than the other group Autopsy studies have revealed that myocarditis is the (p = 0.0009) (Fig. 6c). predominant pathological feature of myocardial injury in DM, with the immune system thought to play a pivotal role in the development of both conditions. Based on the Discussion above two reasons, we used an integrated bioinformat- Although the cause of DM pathogenesis remains unclear, ics analysis to screen hub-genes related to the immune previous studies have established the existence of myo- system process in DM and myocarditis respectively. The cardial injury in DM. Autopsy studies indicate that intersection of these two sets of genes represents a group myocarditis is the most frequent pathological manifes- of common hub-genes that are believed to be linked to tation [3, 31]. Some serological biomarkers may serve myocardial injury in DM. This approach has been suc - as a screening tool for myocardial injury in DM such as cessfully applied in a variety of biological contexts to cTNI and GDF-15 [32, 33]. Despite this, little research identify common risk genes and mechanisms associated has explored the genetic underpinnings of myocardial with multiple disease phenotypes [34–36]. In our study, injury in DM. Our study has identified several common the similarity of immune cell infiltration in the myocar - hub-genes that are associated with both the immune dium and skeletal muscle could also in turn suggest that process of myocarditis and DM. These genes may play a immune system processes may play a role in the process role in the onset of myocardial injury in DM. By identify- of myocardial injury in dermatomyositis. ing miRNAs that target these genes and validating their We finally identified 4 common hub-genes related potential, we have found miR-146a-5p to be a promising to the immune process of myocarditis and DM: IFIT3, biomarker for detecting myocardial injury in DM. OAS3, ISG15, and RSAD2. Gene function enrichment analysis showed these genes were mainly enriched Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 9 of 12 Fig. 6 Screening and validation of potential miRNAs targeting hub-genes. a An Interaction network of four hub-genes and potential miRNAs-targeted. b The volcano plot of DE-miRNAs between myocarditis and normal control in GSE147517. c Validation of miR-146a-5p expression in the serum of patients with DM systemic sclerosis [37–39]. Several previous studies Table 1 The characteristics of DM patients between two groups with and without myocardial injury have confirmed that the type I IFN signaling pathway plays a prominent role in DM, and the expression lev- DM with DM without p value els of it is associated with DM activity [40–45]. Cassius myocardial myocardial injury injury (n = 5) (n = 5) et al. reported type I interferon signature was also to be highly expressed in the MDA5 + DM subtype [43]. In Age, year 47.80 ± 6.61 52.00 ± 9.43 0.441 fact, a recent study suggests that inhibition of the type Female (n, %) 5 (100%) 5 (100%) 1 I IFN signaling pathway may reduce cardiovascular risk Disease duration, 6.20 ± 4.66 3.20 ± 2.77 0.251 in SLE patients [46]. month Given that immune cells play an essential role in the cTNT, ng/L 80.85 ± 14.95 168.21 ± 125.36 0.347 process of myocardial injury in DM, we sought to inves- CKMB, U/L 70.94 ± 128.15 218.36 ± 190.35 0.189 tigate the infiltration of immune cells in both DM and Pro-BNP, pg/ml 83.98 ± 70.58 198.93 ± 164.78 0.246 myocarditis patients. We found that T cells and mac- rophages comprise the majority of infiltrated immune cells in the skeletal muscle of DM, which is consistent in type I interferon (IFN) signaling pathway, cellular with a recent study [47]. Furthermore, we observed an response to type I interferon, and response to type I increase in both M1 and M2 macrophages in DM patients interferon. The maladaptive immune response created compared to healthy controls. Prior studies also revealed by type I IFN signaling is upregulated in many autoim- macrophages infiltration in the muscle is involved in mune diseases such as systemic lupus erythematosus the development and progression of DM [48, 49] and is (SLE), rheumatoid arthritis, Sjogren’s syndrome, and associated with disease severity [50]. Increasing evidence Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 10 of 12 showed that immune cell infiltration in the myocardium sample size will be needed in future studies to verify the has adverse effect on heart function recently [51–53]. role of miR-146a-5p as a biomarker predicting myocar- In the myocardium, macrophages are one of the most dial injury in DM. important cardiac immune cells and the central regulator of immune systems. In the past, macrophages were clas- sified into M1 and M2 types by their surface molecules. Conclusion Further, research indicated that M1 macrophages have a Our study identified 4 common hub-genes related to the pro-inflammatory phenotype with anti-pathogen activ - immune system process of myocarditis. We speculated ity while M2 macrophages promote anti-inflammatory that these genes may play a role in the process of myocar- effects and tissue repair responses [54]. In animal models dial injury in DM. Serum miR-146a-5p could be a poten- of experimental autoimmune myocarditis and viral myo- tial biomarker to predict myocardial injury in DM. carditis, the acute phase of myocarditis is dominated by pro-inflammatory macrophages, while the chronic phase Abbreviations is dominated by M2 macrophages [55–57]. The activation DM Dermatomyositis of the autoimmune system will eventually lead to exces- IIM Idiopathic inflammatory myopathy NCBI-GEO National C enter Biotechnology Information Gene Expression sive accumulation and transformation of macrophages, Omnibus resulting in myocardial inflammation and fibrosis [58, PCA Principal component analysis 59]. In this study, the M2 macrophages of myocardi- WGCNA Weighted gene co-expression network analysis TOM Topological overlap matrix tis patients increased while M1 macrophages were not GS Gene significance statistically different from normal controls, which may MM Module membership be due to the fact that myocardium specimens in the STRING Search Tool for the Retrieval of Interacting Genes FC Fold-change selected dataset were in the chronic phase of myocarditis. LGE Late gadolinium enhancement Considering the similarities of immune cell infiltration in DEGs Differentially expressed genes the myocardium and skeletal muscle, we speculated that PPI Protein-protein interaction GO Gene ontology the disorder in macrophages might play a potentially sig- GSEA Gene function enrichment analysis nificant role in the process of myocardial injury in DM. IFN Type I interferon Previous studies have shown that miRNAs can be used Acknowledgements as markers in a variety of cardiovascular diseases [60]. We thank all the patients and healthy donors involved in the study. miR-146a-5p is an important regulator of the immune response and inflammation [61, 62] and is abundant in Authors’ contributions Study conception and design was performed by YZ, LS, DL, XS, YZ, QW, and immune cells and the heart [63, 64]. It has been impli- LZ. Data collection was performed by YZ, LS, DL, MD, XS, and YZ. YZ, LS, DL, Y T, cated in cardiac hypertrophy, ischemia/reperfusion WQ, JD, and MD analyzed and interpreted the data. All authors were involved injury, peripartum cardiomyopathy, doxorubicin toxic- in drafting the article or making critical revisions for important intellectual content, and the authors read and approved the submitted manuscript. ity, diabetic cardiomyopathy, and atherosclerosis [65–70]. The increased presence of circulating miR-146a-5p has Funding been reported in patients with spontaneous coronary This work was supported by the National Natural Science Foundation of China (81970723). artery dissection, aortic dissection, and acute coronary syndromes [71–73]. Our study found that serum miR- Availability of data and materials 146a-5p was significantly elevated in DM patients with The data used and/or analyzed in the current study are available from the cor- responding author on reasonable request. myocardial injury than without myocardial injury, sug- gesting the potential of miR-146a-5p as a biomarker for Declarations assessing myocardial injury in DM. There were certain limitations in our study. First, our Ethics approval and consent to participate study was based on bioinformatics analysis from public This study was approved by the medical ethics committee of The First Affili- ated Hospital of Nanjing Medical University. Ethical approval was obtained datasets, which may not fully reflect the actual situation. for our single-center cross-sectional study, and the need to obtain informed Secondly, due to the difficulty of obtaining cardiac sam - consent was waived (2020-SR-228). ples from DM patients, we analyzed the gene sets of myo- Consent for publication carditis and DM separately. Further in  vitro and in  vivo Not applicable. experiments are needed to confirm the role of common hub-genes in DM with myocardial injury. Thirdly, we Competing interests The authors declare no competing interests. searched only one dataset containing myocardial speci- mens in myocarditis patients, so we used murine myo- cardium for our subsequent validation. Finally, a larger Zhang  et al. Arthritis Research & Therapy (2023) 25:69 Page 11 of 12 Author details customizable protein-protein networks, and functional characteriza- Department of Cardiology, The First Affiliated Hospital of Nanjing Medical tion of user-uploaded gene/measurement sets. Nucleic Acids Res. University, Nanjing, China. Department of Rheumatology, The First Affiliated 2021;49(D1):D605–12. Hospital of Nanjing Medical University, Nanjing, China. Department of Radiol- 22. von Mering C, Huynen M, Jaeggi D, Schmidt S, Bork P, Snel B. STRING: a ogy, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. database of predicted functional associations between proteins. Nucleic Acids Res. 2003;31(1):258–61. Received: 11 January 2023 Accepted: 20 April 2023 23. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–504. 24. Lin CY, Chin CH, Wu HH, Chen SH, Ho CW, Ko MT. Hubba: hub objects analyzer–a framework of interactome hubs identification for network References biology. Nucleic Acids Res. 2008;36( Web Server issue):W438-443. 1. DeWane ME, Waldman R, Lu J. Dermatomyositis: clinical features and 25. Fan Y, Siklenka K, Arora SK, Ribeiro P, Kimmins S, Xia J. miRNet - dissecting pathogenesis. J Am Acad Dermatol. 2020;82(2):267–81. miRNA-target interactions and functional associations through network- 2. Sena P, Gianatti A, Gambini D. Dermatomyositis: clinicopathological corre- based visual analysis. Nucleic Acids Res. 2016;44( W1):W135-141. lations. G Ital Dermatol Venereol. 2018;153(2):256–64. 26. Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two parts). N 3. Zhang L, Wang GC, Ma L, Zu N. Cardiac involvement in adult polymyositis Engl J Med. 1975;292(7):344–7. or dermatomyositis: a systematic review. Clin Cardiol. 2012;35(11):686–91. 27. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using 4. Kariyanna PT, Jayarangaiah A, Mahmood A, Hare M, Taklalsingh N, McFar- real-time quantitative PCR and the 2(-Delta Delta C( T )) Method. Methods. lane IM. Rapid progression of heart failure in a patient with idiopathic 2001;25(4):402–8. inflammatory myopathy. Am J Med Case Rep. 2018;6(8):157–60. 28. Bondarenko O, Beek AM, Hofman MB, Kuhl HP, Twisk JW, van Dockum WG, 5. Yafasova A, Diederichsen LP, Schou M, Sun G, Torp-Pedersen C, Gislason Visser CA, van Rossum AC. Standardizing the definition of hyperenhance - GH, Fosbol EL, Kober L, Butt JH. Increased long-term risk of heart failure ment in the quantitative assessment of infarct size and myocardial viabil- and other adverse cardiac outcomes in dermatomyositis and polymyosi- ity using delayed contrast-enhanced CMR. J Cardiovasc Magn Reson. tis: Insights from a nationwide cohort. J Intern Med. 2021;290(3):704–14. 2005;7(2):481–5. 6. Denbow CE, Lie JT, Tancredi RG, Bunch TW. Cardiac involvement in 29. Yu L, Sun J, Sun J, Li J, Dong Y, Zhou X, Greiser A, Han Y, Zhang Q, Xie Q, polymyositis: a clinicopathologic study of 20 autopsied patients. Arthritis et al. Early detection of myocardial involvement by T1 mapping of cardiac Rheum. 1979;22(10):1088–92. MRI in idiopathic inflammatory myopathy. J Magn Reson Imaging. 7. Haupt HM, Hutchins GM. The heart and cardiac conduction system in 2018;48(2):415–22. polymyositis-dermatomyositis: a clinicopathologic study of 16 autopsied 30. Haaf P, Garg P, Messroghli DR, Broadbent DA, Greenwood JP, Plein S. patients. Am J Cardiol. 1982;50(5):998–1006. Cardiac T1 Mapping and Extracellular Volume (ECV ) in clinical practice: a 8. Diederichsen LP, Simonsen JA, Diederichsen AC, Hvidsten S, Hougaard comprehensive review. J Cardiovasc Magn Reson. 2016;18(1):89. M, Junker P, Sondergaard K, Lundberg IE, Tvede N, Sandgaard NC, et al. 31. Sultan SM, Ioannou Y, Moss K, Isenberg DA. Outcome in patients with idi- Cardiac abnormalities in adult patients with polymyositis or dermatomy- opathic inflammatory myositis: morbidity and mortality. Rheumatology ositis as assessed by noninvasive modalities. Arthritis Care Res (Hoboken). (Oxford). 2002;41(1):22–6. 2016;68(7):1012–20. 32. Chen F, Peng Y, Chen M. Diagnostic approach to cardiac involvement in 9. Prisco F, Papparella S, Paciello O. The correlation between cardiac and idiopathic inflammatory myopathies. Int Heart J. 2018;59(2):256–62. skeletal muscle pathology in animal models of idiopathic inflammatory 33. Qiu M, Sun X, Qi X, Liu X, Zhang Y, Zhang N, Lu F, Liu W, Changjing F, myopathies. Acta Myol. 2020;39(4):313–9. Wang Q, et al. The diagnostic value of GDF-15 for myocardial involve- 10. Ringner M. What is principal component analysis? Nat Biotechnol. ment in idiopathic inflammatory myopathy. Rheumatology (Oxford). 2008;26(3):303–4. 2021;60(6):2826–33. 11. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation 34. Sezin T, Vorobyev A, Sadik CD, Zillikens D, Gupta Y, Ludwig RJ. Gene network analysis. BMC Bioinformatics. 2008;9:559. expression analysis reveals novel shared gene signatures and candidate 12. Zhang B, Horvath S. A general framework for weighted gene co-expres- molecular mechanisms between pemphigus and systemic lupus erythe- sion network analysis. Stat Appl Genet Mol Biol. 2005;4:Article17. matosus in CD4(+) T cells. Front Immunol. 1992;2017:8. 13. Smyth GK. Linear models and empirical bayes methods for assessing dif- 35. Zou R, Zhang D, Lv L, Shi W, Song Z, Yi B, Lai B, Chen Q, Yang S, Hua P. Bio- ferential expression in microarray experiments. Stat Appl Genet Mol Biol. informatic gene analysis for potential biomarkers and therapeutic targets 2004;3:Article3. of atrial fibrillation-related stroke. J Transl Med. 2019;17(1):45. 14. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. limma pow- 36. Zhu Y, Ding X, She Z, Bai X, Nie Z, Wang F, Wang F, Geng X. Explor- ers differential expression analyses for RNA-sequencing and microarray ing shared pathogenesis of Alzheimer’s disease and type 2 diabetes studies. Nucleic Acids Res. 2015;43(7):e47. mellitus via co-expression networks analysis. Curr Alzheimer Res. 15. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, Hoang CD, 2020;17(6):566–75. Diehn M, Alizadeh AA. Robust enumeration of cell subsets from tissue 37. Hill CL, Zhang Y, Sigurgeirsson B, Pukkala E, Mellemkjaer L, Airio A, Evans expression profiles. Nat Methods. 2015;12(5):453–7. SR, Felson DT. Frequency of specific cancer types in dermatomyositis and 16. Ito K, Murphy D. Application of ggplot2 to Pharmacometric Graphics. CPT polymyositis: a population-based study. Lancet. 2001;357(9250):96–100. Pharmacometrics Syst Pharmacol. 2013;2:e79. 38. Chen LY, Cui ZL, Hua FC, Yang WJ, Bai Y, Lan FH. Bioinformatics analy- 17. Tripathi S, Pohl MO, Zhou Y, Rodriguez-Frandsen A, Wang G, Stein DA, sis of gene expression profiles of dermatomyositis. Mol Med Rep. Moulton HM, DeJesus P, Che J, Mulder LC, et al. Meta- and orthogonal 2016;14(4):3785–90. integration of influenza “OMICs” data defines a role for UBR4 in virus bud- 39. Siddiqi KZ, Zinglersen AH, Iversen KK, Rasmussen NS, Nielsen CT, ding. Cell Host Microbe. 2015;18(6):723–35. Jacobsen S. A cluster of type II interferon-regulated genes associates with 18. Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, disease activity in patients with systemic lupus erythematosus. J Autoim- Benner C, Chanda SK. Metascape provides a biologist-oriented resource mun. 2022;132:102869. for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. 40. Baechler EC, Bilgic H, Reed AM. Type I interferon pathway in adult and 19. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illu- juvenile dermatomyositis. Arthritis Res Ther. 2011;13(6):249. mina sequence data. Bioinformatics. 2014;30(15):2114–20. 41. Gitiaux C, Latroche C, Weiss-Gayet M, Rodero MP, Duffy D, Bader-Meunier 20. Vizcaino JA, Deutsch EW, Wang R, Csordas A, Reisinger F, Rios D, Dianes B, Glorion C, Nusbaum P, Bodemer C, Mouchiroud G, et al. Myogenic JA, Sun Z, Farrah T, Bandeira N, et al. ProteomeXchange provides glob- progenitor cells exhibit type I interferon-driven proangiogenic proper- ally coordinated proteomics data submission and dissemination. Nat ties and molecular signature during juvenile dermatomyositis. Arthritis Biotechnol. 2014;32(3):223–6. Rheumatol. 2018;70(1):134–45. 21. Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, Doncheva NT, Legeay M, Fang T, Bork P, et al. The STRING database in 2021: Zhang et al. Arthritis Research & Therapy (2023) 25:69 Page 12 of 12 42. Ladislau L, Suarez-Calvet X, Toquet S, Landon-Cardinal O, Amelin D, Depp 63. Feng B, Chen S, McArthur K, Wu Y, Sen S, Ding Q, Feldman RD, Chakrabarti M, Rodero MP, Hathazi D, Duffy D, Bondet V, et al. JAK inhibitor improves S. miR-146a-mediated extracellular matrix protein production in chronic type I interferon induced damage: proof of concept in dermatomyositis. diabetes complications. Diabetes. 2011;60(11):2975–84. Brain. 2018;141(6):1609–21. 64. Shimada BK, Yang Y, Zhu J, Wang S, Suen A, Kronstadt SM, Jeyaram A, 43. Cassius C, Amode R, Delord M, Battistella M, Poirot J, How-Kit A, Lepel- Jay SM, Zou L, Chao W. Extracellular miR-146a-5p induces cardiac innate letier C, Jachiet M, de Masson A, Frumholtz L, et al. MDA5(+) dermato- immune response and cardiomyocyte dysfunction. Immunohorizons. myositis is associated with stronger skin type I interferon transcriptomic 2020;4(9):561–72. signature with upregulation of IFN-kappa transcript. J Invest Dermatol. 65. Lin G, Huang J, Chen Q, Chen L, Feng D, Zhang S, Huang X, Huang Y, Lin 2020;140(6):1276-1279 e1277. Q. miR-146a-5p mediates intermittent hypoxia-induced injury in H9c2 44. Kuriyama Y, Shimizu A, Kanai S, Oikawa D, Tokunaga F, Tsukagoshi H, Ishi- cells by targeting XIAP. Oxid Med Cell Longev. 2019;2019:6581217. kawa O. The synchronized gene expression of retrotransposons and type 66. Shu L, Zhang W, Huang G, Huang C, Zhu X, Su G, Xu J. Troxerutin attenu- I interferon in dermatomyositis. J Am Acad Dermatol. 2021;84(4):1103–5. ates myocardial cell apoptosis following myocardial ischemia-reperfusion 45. Buang N, Tapeng L, Gray V, Sardini A, Whilding C, Lightstone L, Cairns TD, injury through inhibition of miR-146a-5p expression. J Cell Physiol. Pickering MC, Behmoaras J, Ling GS, et al. Type I interferons affect the 2019;234(6):9274–82. metabolic fitness of CD8(+) T cells from patients with systemic lupus 67. Dong M, Chen D, Zhu Y, Yang S, Kumar S, Zhang R, Zhou Y, Yang Z, Zheng erythematosus. Nat Commun. 2021;12(1):1980. N, Zhu T, et al. Impaired regulation of MMP2/16-MLCK3 by miR-146a-5p 46. Casey KA, Smith MA, Sinibaldi D, Seto NL, Playford MP, Wang X, Carlucci increased susceptibility to myocardial ischemic injury in aging mice. PM, Wang L, Illei G, Yu B, et al. Modulation of cardiometabolic disease Cardiovasc Res. 2022:cvac104. markers by type I interferon inhibition in systemic lupus erythematosus. 68. Oh JG, Watanabe S, Lee A, Gorski PA, Lee P, Jeong D, Liang L, Liang Y, Arthritis Rheumatol. 2021;73(3):459–71. Baccarini A, Sahoo S, et al. miR-146a suppresses SUMO1 expression 47. Yin R, Wang G, Zhang L, Li T, Liu S. Dermatomyositis: immunological and induces cardiac dysfunction in maladaptive hypertrophy. Circ Res. landscape, biomarkers, and potential candidate drugs. Clin Rheumatol. 2018;123(6):673–85. 2021;40(6):2301–10. 69. Wang X, Ha T, Liu L, Zou J, Zhang X, Kalbfleisch J, Gao X, Williams D, Li C. 48. Ragusa F. Dermatomyositis and MIG. Clin Ter. 2019;170(2):e142–7. Increased expression of microRNA-146a decreases myocardial ischaemia/ 49. Jiang T, Huang Y, Liu H, Xu Q, Gong Y, Chen Y, Hu X, Han Z, Gao M. reperfusion injury. Cardiovasc Res. 2013;97(3):432–42. Reduced miR-146a promotes REG3A expression and macrophage migra- 70. Cheng HS, Besla R, Li A, Chen Z, Shikatani EA, Nazari-Jahantigh M, tion in polymyositis and dermatomyositis. Front Immunol. 2020;11:37. Hammoutene A, Nguyen MA, Geoffrion M, Cai L, et al. Paradoxical sup - 50. Peng QL, Zhang YL, Shu XM, Yang HB, Zhang L, Chen F, Lu X, Wang GC. pression of atherosclerosis in the absence of microRNA-146a. Circ Res. Elevated serum levels of soluble CD163 in polymyositis and dermato- 2017;121(4):354–67. myositis: associated with macrophage infiltration in muscle tissue. J 71. Zhelankin AV, Stonogina DA, Vasiliev SV, Babalyan KA, Sharova EI, Doludin Rheumatol. 2015;42(6):979–87. YV, Shchekochikhin DY, Generozov EV, Akselrod AS. Circulating extracel- 51. Carrillo-Salinas FJ, Ngwenyama N, Anastasiou M, Kaur K, Alcaide P. Heart lular miRNA analysis in patients with stable CAD and acute coronary inflammation: immune cell roles and roads to the heart. Am J Pathol. syndromes. Biomolecules. 2021;11(7):962. 2019;189(8):1482–94. 72. Lozano-Prieto M, Adlam D, Garcia-Guimaraes M, Sanz-Garcia A, Vera-Tome 52. Lafuse WP, Wozniak DJ, Rajaram MVS. Role of cardiac macrophages on P, Rivero F, Cuesta J, Bastante T, Baranowska-Clarke AA, Vara A, et al. cardiac inflammation, fibrosis and tissue repair. Cells. 2020;10(1):51. Differential miRNAs in acute spontaneous coronary artery dissection: 53. Kawada JI, Takeuchi S, Imai H, Okumura T, Horiba K, Suzuki T, Torii Y, pathophysiological insights from a potential biomarker. EBioMedicine. Yasuda K, Imanaka-Yoshida K, Ito Y. Immune cell infiltration landscapes 2021;66:103338. in pediatric acute myocarditis analyzed by CIBERSORT. J Cardiol. 73. Fang L, Ellims AH, Moore XL, White DA, Taylor AJ, Chin-Dusting J, Dart AM. 2021;77(2):174–8. Circulating microRNAs as biomarkers for diffuse myocardial fibrosis in 54. Italiani P, Boraschi D. From monocytes to M1/M2 macrophages: pheno- patients with hypertrophic cardiomyopathy. J Transl Med. 2015;13:314. typical vs. functional differentiation. Front Immunol. 2014;5:514. 55. Anzai A, Mindur JE, Halle L, Sano S, Choi JL, He S, McAlpine CS, Chan CT, Publisher’s Note Kahles F, Valet C, et al. Self-reactive CD4(+) IL-3(+) T cells amplify autoim- Springer Nature remains neutral with regard to jurisdictional claims in pub- mune inflammation in myocarditis by inciting monocyte chemotaxis. J lished maps and institutional affiliations. Exp Med. 2019;216(2):369–83. 56. Clemente-Casares X, Hosseinzadeh S, Barbu I, Dick SA, Macklin JA, Wang Y, Momen A, Kantores C, Aronoff L, Farno M, et al. A CD103(+) conventional dendritic cell surveillance system prevents development of overt heart failure during subclinical viral myocarditis. Immunity. 2017;47(5):974-989 e978. 57. Hou X, Chen G, Bracamonte-Baran W, Choi HS, Diny NL, Sung J, Hughes D, Won T, Wood MK, Talor MV, et al. The cardiac microenvironment instructs divergent monocyte fates and functions in myocarditis. Cell Rep. 2019;28(1):172-189 e177. 58. Wolf AA, Yanez A, Barman PK, Goodridge HS. The ontogeny of monocyte subsets. Front Immunol. 2019;10:1642. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : 59. Bajpai G, Schneider C, Wong N, Bredemeyer A, Hulsmans M, Nahrendorf M, Epelman S, Kreisel D, Liu Y, Itoh A, et al. The human heart contains fast, convenient online submission distinct macrophage subsets with divergent origins and functions. Nat thorough peer review by experienced researchers in your field Med. 2018;24(8):1234–45. 60. Barwari T, Joshi A, Mayr M. MicroRNAs in cardiovascular disease. J Am Coll rapid publication on acceptance Cardiol. 2016;68(23):2577–84. support for research data, including large and complex data types 61. Sheedy FJ, O’Neill LA. Adding fuel to fire: microRNAs as a new class of • gold Open Access which fosters wider collaboration and increased citations mediators of inflammation. Ann Rheum Dis. 2008;67 Suppl 3:iii50-55. 62. Taganov KD, Boldin MP, Chang KJ, Baltimore D. NF-kappaB-dependent maximum visibility for your research: over 100M website views per year induction of microRNA miR-146, an inhibitor targeted to signal- ing proteins of innate immune responses. Proc Natl Acad Sci U S A. At BMC, research is always in progress. 2006;103(33):12481–6. Learn more biomedcentral.com/submissions

Journal

Arthritis Research & Therapy – Springer Journals

Published: Apr 28, 2023

Keywords: Dermatomyositis; Myocardial injury; Bioinformatic; Immune cells infiltration; Biomarker

There are no references for this article.