Region Specific miRNA–mRNA Networks in Gray and White Matter Lesions of Progressive Multiple Sclerosis

ABSTRACT

Objective

Multiple sclerosis (MS) is a neurodegenerative demyelinating disease of the central nervous system. This study aimed to identify micro‐RNA (miRNA)–mRNA regulatory networks underlying region‐specific molecular mechanisms in white matter and gray matter lesions in progressive MS.

Methods

Global miRNA and mRNA expression profiling were previously performed on white matter and gray matter lesion tissues from postmortem progressive MS brains. These datasets were integrated using Ingenuity Pathway Analysis (IPA) to identify enriched canonical pathways based on experimentally validated miRNA target genes. Functional effects of selected miRNAs were evaluated in primary oligodendrocyte progenitor cells (OPCs) following miRNA mimic transfection using immunocytochemistry and qPCR.

Results

Nine miRNAs were commonly dysregulated in both White Matter lesions (WMLs) and Gray Matter lesions (GMLs). Pathway analysis identified two major downstream pathways—cellular senescence and EIF2 signaling—with opposing predicted activation states (activated in White Matter lesions and inhibited in Gray Matter lesions). Among the shared miRNAs, miR‐30a and miR‐100 exhibited opposing expression patterns, with downregulation in GMLs and upregulation in WMLs, and both were predicted to regulate each pathway. Corresponding target genes of these miRNAs were significantly dysregulated in MS lesions. In primary Oligodendrocyte Progenitor Cells, overexpression of miR‐30a or miR‐100 significantly reduced proliferation and impaired differentiation, accompanied by decreased expression of their respective target genes.

Interpretation

These findings identify miRNAs commonly dysregulated across white matter and gray matter lesions in progressive multiple sclerosis. Furthermore, this study also identifies distinct miRNA–mRNA networks and miRNA‐driven regulatory mechanisms that contribute to region‐specific impairment of oligodendrocyte maturation and remyelination.

1. Introduction

Multiple sclerosis (MS) is a chronic autoimmune neurodegenerative disorder of the central nervous system characterized by demyelination in the brain and spinal cord. Although remyelination can occur during the early stages of disease, this repair process progressively fails as MS advances [1]. Historically, MS was considered a predominantly white matter (WM) disease, with WM lesions (WMLs) viewed as the principal pathological hallmark and closely associated with immune cell infiltration [2, 3]. However, accumulating evidence indicates that gray matter (GM) pathology is also a major contributor to disease progression [2]. Cortical involvement and gray matter lesions (GMLs) are strongly associated with clinical manifestations such as cognitive impairment and neurological disability, and although inflammation is less pronounced in GM than in WM, it remains an important driver of MS progression [4]. Importantly, early cortical involvement is linked to more rapid disease progression, highlighting the urgent need to define GM‐specific disease mechanisms [5]. Cortical lesions also have a higher propensity for remyelination compared to WMLs [6]. Therefore, elucidating the critical molecular differences in cellular microenvironment between GM and WM that shape tissue‐specific mechanisms of demyelination may help explain regional differences in remyelination capacity between GMLs and WMLs and inform the development of targeted, compartment‐specific therapeutic strategies.

MicroRNAs (miRNAs) are small non‐coding RNAs (~21–23 nucleotides) that regulate post‐transcriptional gene expression by binding to miRNA response elements (MREs) in the 3′ untranslated region (UTR) of target mRNAs [7]. Several miRNAs, including miR‐219 and miR‐338, are critical regulators of oligodendrocyte progenitor cell (OPC) differentiation and myelination [8, 9, 10]. In the context of MS, numerous studies have reported dysregulated miRNA expression within MS lesions [11, 12, 13], and our group has previously characterized lesion‐specific miRNA alterations in progressive MS across both WM and GM compartments [14, 15]. Despite these advances, it remains unclear how differences in the lesion microenvironment (WM versus GM) shape miRNA‐driven regulatory programs that influence OPC behavior and create region‐specific barriers to remyelination. In this study, we identify distinct miRNA–mRNA regulatory networks and downstream pathways that differentiate WM and GM lesion biology and provide mechanistic insight into remyelination failure in MS.

2. Materials and Methods

2.1. Human Subjects and Brain Tissue Collection

Human brain tissues were obtained from patients with multiple sclerosis (MS) and non‐neurological control subjects through the Cleveland Clinic Brain Tissue Procurement Program. All procedures were approved by the Cleveland Clinic Institutional Review Board and tissue collection was conducted under a rapid autopsy protocol. Written informed consent was obtained from donors or their next of kin prior to tissue procurement. Clinical and demographic characteristics of the cases included in this study have been reported previously [14, 15, 16].

2.2. Global miRNA Expression Profiling

WM and GM miRNA expression profiling was performed using MS tissues and previously reported by us [14, 15]. The WM miRNA dataset comprised tissue obtained from WM of five progressive MS patients and five age‐matched healthy control subjects, with paired samples of normal appearing white matter (NAWM) and WMLs analyzed in MS cases. The GM miRNA dataset included tissue collected from normal appearing gray matter (NAGM) and GMLs from five progressive MS patients, along with GM tissue from five healthy control subjects. All tissue processing, miRNA extraction, and profiling procedures were conducted as described in the original publications [14, 15].

Differentially expressed miRNAs were identified independently for GM and WM datasets using a nonparametric (Mann–Whitney) test and corrected for false discovery rate (FDR) (p < 0.05) [14, 15]. miRNAs with low overall expression (raw count < 500 across samples) were excluded. Differentially expressed miRNAs were subsequently cross‐compared to identify miRNAs that were GM‐specific (n = 21), WM‐specific (n = 40), or shared between both compartments (n = 9) (Figure 1A–C).

FIGURE 1.

Dysregulated miRNAs and their target genes shared between GMLs and WMLs in MS. (A,B) Heatmaps showing significantly upregulated and downregulated miRNAs identified exclusively in WMLs, exclusively in GMLs, or shared between lesion types. (C) miRNAs differentially expressed in both GMLs and WMLs. Seven miRNAs displayed concordant regulation across compartments, whereas miR‐30a and miR‐100 were downregulated in GMLs but upregulated in WMLs. (D) The nine miRNAs commonly dysregulated in GMLs and WMLs are associated with 464 experimentally validated target genes. Target gene expression was assessed using gene‐expression microarray datasets from progressive MS brains. (E) Heatmaps showing expression of experimentally‐validated miRNA target genes in lesions and matched normal‐appearing brain regions.

2.3. miRNA Target Gene Identification and Pathway Analysis

To assess the functional impact of the nine miRNAs commonly dysregulated in GM and WM, experimentally validated mRNA targets were identified using the miRTargetLink2.0 database [17]. This analysis yielded 464 target genes regulated by at least two of the nine shared miRNAs. These genes were queried against previously published GM‐specific and WM‐specific microarray datasets [18, 19] to determine expression patterns in lesions relative to matched normal‐appearing tissue (Figure 1E). Pathway enrichment analysis was performed using QIAGEN Ingenuity Pathway Analysis (IPA). Core analyses were conducted separately for GM and WM target gene datasets using default parameters, restricted to human genes and experimentally validated interactions. Canonical pathways were ranked based on statistical significance of overlap in target genes and the genes associated with a pathway (p‐value), and predicted activation state (Z‐score) of the pathway (Figure 2A,B). Genes shared between enriched GM and WM pathways were further analyzed to identify miRNA–gene regulatory interactions, retaining only miRNA–gene pairs exhibiting reciprocal expression patterns. Networks were visualized using Cytoscape [20] and compared between GM and WM datasets.

FIGURE 2.

IPA of target genes regulated by miRNAs dysregulated in GMLs and WMLs. (A,B) Top enriched canonical pathways and predicted activation states in WM and GM, with activated pathways shown in orange and inhibited pathways in blue. (C) Venn diagram showing overlap between genes involved in the cellular senescence pathway and experimentally validated targets of the nine shared miRNAs. (D,E) miRNA–mRNA interaction networks with reciprocal expression patterns within the cellular senescence pathway in WM (D) and GM (E). (F) Venn diagram illustrating overlap between genes in the EIF2 signaling pathway and experimentally validated miRNA target genes. (G,H) miRNA–mRNA interaction networks within the EIF2 signaling pathway showing inverse expression relationships in WM (G) and GM (H).

2.4. Primary Mouse OPC Cultures and Transfections

Primary OPCs were isolated from postnatal day 6–7 (P6–P7) C57BL/6J wild‐type mice (The Jackson Laboratory, strain: #000664). Animals were housed under standard conditions with a 12‐h light/dark cycle and access to food and water ad libitum. OPCs were isolated using an immunopanning protocol as previously described [21, 22]. CD140a‐positive OPCs were maintained in DMEM‐SATO‐based proliferation medium supplemented with PDGF‐AA (20 ng/mL), NT‐3 (2 ng/mL), forskolin (4.2 μg/mL), and CNTF (10 ng/mL). For differentiation experiments, cultures were switched to differentiation medium (DMEM‐SATO supplemented with forskolin and T3 [40 ng/mL]) at 60%–70% confluence.

Primary OPCs were seeded onto coverslips in 24‐well plates and transfected at 60%–70% confluence with miRNA mimics (miR‐30a or miR‐100; mirVana, Thermo Fisher Scientific) or negative control mimics using Lipofectamine RNAiMAX according to the manufacturer's instructions. Cells were maintained in transfection medium for 48 h prior to analysis.

2.5. Quantitative Real‐Time PCR

Total RNA, including small RNAs, was isolated using the miRNeasy Mini Kit (Qiagen). miRNAs were reverse‐transcribed using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems), and miR‐30a and miR‐100 expression were quantified using TaqMan MicroRNA Assays (Catalog number: #4427975) with U6 small nuclear RNA as the endogenous control. For mRNA analysis, cDNA was synthesized using the High‐Capacity cDNA Reverse Transcription Kit (Applied Biosystems), and target gene expression was quantified using TaqMan Gene Expression Assays with Gapdh as the endogenous control. The assay IDs of all miRNAs and genes used for qPCR validation are listed in Table S1. All reactions were run in triplicate, and relative expression was calculated using the 2^−ΔΔCT method. Statistical analyses were performed using GraphPad Prism 11.

2.6. Immunocytochemistry

OPCs were fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X‐100 in phosphate‐buffered saline (PBS), followed by blocking in 5% normal goat serum and 1% bovine serum albumin (BSA) for 1 h at room temperature. Cells were incubated with primary antibodies (Rat anti‐NG2: MAB6699, 1:100, R&D Systems; Rabbit anti‐Ki67: ab15580, 1:250, abcam; Rat anti‐MBP: MAB386, 1:100, Miltenyi Biotec) overnight at 4°C, followed by Alexa Fluor‐conjugated secondary antibodies for 1 h at room temperature. Nuclei were counterstained with DAPI and coverslips were mounted using ProLong Gold Antifade Mountant. Imaging was performed using confocal microscopy (Leica‐Artemis) at the Cleveland Clinic Research Imaging Core.

2.7. Statistical Analysis

Statistical analyses were performed using GraphPad Prism 11. For miRNA and mRNA expression analyses, comparisons between lesion tissue and matched normal‐appearing regions were performed using Welch's t‐test. For in vitro experiments, quantitative data are presented as mean ± SEM, and statistical significance was assessed using unpaired two‐tailed t‐tests unless otherwise indicated. All experiments were performed using tissue or cells derived from biologically‐independent samples, and p < 0.05 was considered statistically significant.

3. Results

3.1. Distinct miRNA Signatures in WMLs and GMLs of Progressive MS

To determine whether miRNA regulation differs between lesion compartments, we compared miRNA expression profiles of GMLs and WMLs from progressive MS patient brains [14, 15]. In the GM dataset, the analysis identified 496 expressed miRNAs, of which 30 were significantly differentially expressed in GMLs relative to NAGM (12 upregulated and 18 downregulated; p < 0.05). In parallel, profiling of WMLs detected 362 miRNAs, with 49 showing significant differential expression compared with NAWM (23 upregulated and 26 downregulated; p < 0.05).

Most differentially expressed miRNAs were compartment‐restricted, with 81.6% unique to WM (n = 40) and 70.0% unique to GM (n = 21). Heatmaps depicting these miRNAs, categorized as ‘GM‐only’, ‘WM‐only’, or ‘GM & WM’, are shown in Figure 1A,B. These data indicate substantial compartment‐specific differences in miRNA dysregulation between WMLs and GMLs in progressive MS.

Although there were 30 miRNAs dysregulated in GMLs and 49 miRNAs dysregulated in the WMLs, only 9 miRNAs were common between both GMLs and WMLs (Figure 1C). To identify mechanisms that could be contributing to the difference in WM and GM lesion biology, we focused on these 9 miRNAs. Seven of these miRNAs exhibited concordant expression (similar direction) changes between GMLs and WMLs. In contrast, two miRNAs—miR‐30a and miR‐100—demonstrated opposing expression patterns, being downregulated in GMLs and upregulated in WMLs. This bidirectional regulation suggests that miR‐30a and miR‐100 may be differentially modulated by lesion microenvironment and supports their candidacy as mediators of region‐specific molecular responses in MS.

3.2. Shared miRNAs Target Overlapping but Differentially‐Regulated Gene Networks in GMLs and WMLs

Individual miRNAs can regulate multiple gene products within a pathway, and groups of miRNAs may coordinately target genes within shared molecular networks. To define the downstream targets of the nine miRNAs commonly dysregulated in GMLs and WMLs, we used the miRTargetLink2.0 database to identify experimentally validated mRNA targets [17]. This analysis yielded 464 genes targeted by at least two of the nine miRNAs.

To assess whether these shared miRNA targets were differentially regulated across lesion compartments, we examined gene expression profiles from the database of mRNA expression profiling of GMLs and WMLs [18, 19].

Of these 464 mRNA targets, 453 were detected in GMLs, with 175 being upregulated and 278 downregulated relative to NAGM. In WMLs, 389 target genes were detected, with 189 being upregulated and 200 downregulated compared to NAWM (Figure 1D). Among detected targets, 63 genes were significant in GMLs (20 upregulated, 43 downregulated; p < 0.05) while 89 genes were significantly changed in WMLs (48 upregulated, 41 downregulated; p < 0.05) (Figure 1E). These findings indicate that miRNAs shared between GMLs and WMLs converge on target genes that are differentially regulated across lesion compartments.

3.3. miRNA‐Driven Pathway Engagement Predicts Opposite Regulation of Cellular Senescence and EIF2 Signaling

To identify shared and divergent downstream pathways associated with miRNA target genes in GMLs and WMLs, we performed Core Analysis using QIAGEN IPA on the compartment‐specific gene (GML and WML) datasets. Canonical pathways were ranked based on significance of enrichment (p < 1 × 10⁻⁶) and predicted activation state as determined by IPA Z‐scores (Figure 2A,B). The pathways that were predicted to be most significantly enriched were predominantly related to cellular growth, proliferation, and development, integrated stress response (ISR), and cell‐cycle regulation.

Among these, cellular senescence and EIF2 signaling emerged as prominent pathways in both GM and WM datasets. Notably, while the same sets of genes were enriched in both compartments, the predicted activation states differed: both pathways were predicted to be activated in WMLs and inactivated in GMLs (Figure 2A,B). These findings suggest that compartment‐specific differences in pathway engagement are driven primarily by divergent expression patterns of shared target genes rather than distinct gene composition.

We next examined miRNA–gene interactions within the senescence and EIF2 signaling pathways. We identified 24 genes from the experimentally validated targets of the 9 common miRNAs that were involved in the senescence pathway (Figure 2C). Given the typically inverse relationship between miRNA and target gene expression, we filtered for miRNA–gene pairs exhibiting reciprocal expression patterns within each compartment (Figure 2D,E). Comparably, 21 genes from our list of experimentally‐validated targets were involved in the EIF2 pathway (Figure 2F). We shortlisted the miRNA–mRNA pairs exhibiting opposing expression patterns (Figure 2G,H). Interestingly, both pathway‐specific interaction networks included miR‐30a and miR‐100, the two miRNAs previously identified as showing opposing expression trends between GMLs and WMLs (Figure 2D,E,G,H). Consistent with their bidirectional regulation across compartments, miR‐30a and miR‐100 were associated with inverse regulation of multiple genes within both pathways. Together, these data indicate that differential miRNA expression (miR‐30a and miR‐100) may contribute to contrasting regulation of cellular senescence and EIF2 signaling in GMLs versus WMLs during progressive MS.

3.4. miR‐30a and miR‐100 Exert Inhibitory Effects on OPC Proliferation and Differentiation

Among the nine miRNAs dysregulated in both WMLs and GMLs, miR‐30a and miR‐100 displayed opposing expression patterns, with upregulation in WMLs and downregulation in GMLs. Both miRNAs were linked to the most differentially regulated downstream pathways—EIF2 signaling and cellular senescence—which likewise showed opposing predicted activation states in WMLs and GMLs. These findings implicate miR‐30a and miR‐100 in region‐specific molecular responses in progressive MS.

To define the mechanisms by which these miRNAs influence oligodendrocyte lineage biology, we first identified experimentally validated targets of miR‐30a using miRTargetLink2.0. Candidate genes were interrogated in single‐cell RNA‐sequencing datasets from MS brains (GSE180759) using BioTuring BBrowserX, focusing on genes enriched in the oligodendrocyte lineage [23, 24] (Figure 3A). This analysis identified three targets: ELOVL5 and NCAM1 in WM and LIFR in GM. Consistent with miR‐30a dysregulation, LIFR mRNA was increased in GMLs (Figure 3B), whereas ELOVL5 and NCAM1 were reduced in WMLs (Figure 3B–D). UMAP visualization of 66,432 nuclei from MS lesion snRNA‐sequencing data (GSE180759) [23] confirmed expression of all three genes within oligodendrocyte lineage populations by UMAP visualization (Figure 3EI–EIII).

FIGURE 3.

miR‐30a and miR‐100 regulate oligodendrocyte lineage genes dysregulated in MS lesions and inhibit OPC proliferation and differentiation. (A) Experimentally‐validated miR‐30a target genes expressed in oligodendrocyte lineage cells identified using single‐cell RNA‐sequencing datasets. (B–D) Differential expression of miR‐30a target genes LIFR, ELOVL5, and NCAM1 in MS lesions relative to matched normal‐appearing tissue. (E) UMAP visualization of single‐nucleus RNA‐sequencing data as reported by Absinta et al. [23], with expression of miR‐30a target genes shown across oligodendrocyte lineage clusters (EI–EIII). (F) qPCR confirmation of miR‐30a overexpression in primary OPCs following mimic transfection. (G–I) qPCR analysis showing reduced expression of miR‐30a target genes in OPCs transfected with miR‐30a compared with mimic control (p < 0.05). (J,K) Representative immunofluorescence images of NG2⁺ OPCs (green) and Ki67⁺ proliferating cells (red), with quantification of NG2⁺Ki67⁺ OPCs normalized to total DAPI⁺ nuclei. (L,M) Representative immunofluorescence images of MBP⁺ differentiated oligodendrocytes (green) with quantification normalized to total DAPI⁺ nuclei. (N) Identification of MKLN1 as a key miR‐100 target expressed in oligodendrocyte lineage cells. (O) Differential expression of MKLN1 in GMLs compared with NAGM in MS tissue. (P) UMAP visualization showing MKLN1 expression across cell‐type clusters in the Absinta et al. dataset [23]. (Q) qPCR confirmation of miR‐100 overexpression in primary OPCs following mimic transfection. (R) qPCR analysis showing reduced MKLN1 expression in OPCs transfected with miR‐100. (S,T) Representative immunofluorescence images and quantification of NG2⁺Ki67⁺ OPCs following miR‐100 transfection. (U,V) Representative immunofluorescence images and quantification of MBP⁺ oligodendrocytes following miR‐100 transfection, normalized to total DAPI⁺ nuclei.

To assess functional consequences of miR‐30a dysregulation, primary mouse OPCs were transfected with a miR‐30a mimic. qPCR confirmed robust miR‐30a overexpression (Figure 3F) accompanied by significant downregulation of Lifr, Elovl5, and Ncam1 (Figure 3G–I), indicating direct regulation in OPCs. Immunocytochemical analysis performed 48 h post‐transfection revealed a significant reduction in the proportion of Ki67‐positive proliferating OPCs (Figure 3J,K) and a decrease in the number of MBP‐positive differentiated oligodendrocytes (Figure 3L,M).

We next examined miR‐100, which exhibited a similar pattern of compartment‐specific dysregulation and pathway association. Using the same analytical pipeline, we identified MKLN1 as a key experimentally validated miR100 target enrich‐ed in oligodendrocyte lineage cells (Figure 3N). UMAP visualization confirmed MKLN1 expression across relevant cell populations (Figure 3P), and transcriptomic analysis demonstrated increased MKLN1 expression in GMLs relative to NAGM (Figure 3O). Transfection of primary OPCs with a miR‐100 mimic resulted in efficient miR‐100 overexpression and marked suppression of Mkln1 expression (Figure 3Q,R). Similar to miR‐30a, miR‐100 overexpression significantly reduced OPC proliferation and differentiation, as evidenced by fewer Ki67‐positive cells and reduced numbers of MBP‐positive oligodendrocytes (Figure 3S–V).

Collectively, these data identify miR‐30a and miR‐100 to constrain OPC proliferation and differentiation. The compartment‐specific dysregulation of miRNAs in MS lesions correlating with differential engagement of cellular senescence and EIF2 signaling pathways indicates miRNA‐mediated regulatory programs may be underlying the potentially divergent effects on repair capacity between WM and GM.

4. Discussion

Since miRNAs control post‐transcriptional gene regulation, they can serve as important molecular markers for studying MS progression. Numerous studies have examined miRNA expression in different biological samples from MS patients, including serum, plasma, cerebrospinal fluid, and immune cells [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35]. However, very few studies have analyzed miRNA expression directly from MS brain tissue, and most of these have focused exclusively on WMLs [11, 15, 36].

Beyond these WM‐focused analyses, we have previously identified miRNAs altered exclusively in GMLs that correlate with measures of atrophy and may serve as potential biomarkers of GM demyelination [14]. Despite these insights, a significant knowledge gap remains in the systematic comparison of miRNA expression patterns and their downstream regulatory pathways across GM and WM. Initial efforts to address these differences include work by Fritsche et al. [37], who performed miRNA profiling across MS GMLs and WMLs and identified miRNAs enriched in GM that regulate synaptic and neuronal proteins such as synaptotagmin‐7. While that study focused on miRNAs involved in GM‐specific regulatory pathways, the broader interplay between GM and WM lesion microenvironments remains largely unaddressed.

Our study focuses on a distinct group of 9 miRNAs dysregulated across both WM and GM. In this study, we found that the same miRNAs could have different downstream gene expression and pathway signatures because of the lesion microenvironment. Examination of the gene targets of these miRNAs uncovered key cellular pathways—such as cellular senescence and ISR mediated by EIF2 signaling—that are predicted to be differentially engaged in GMLs and WMLs. Specifically, these pathways were predicted to be activated in WM but suppressed or inactivated in GM. These observations align with histopathological findings showing increased cellular senescence in demyelinated MS lesions [38]. Moreover, recent spatial and single‐cell transcriptomic studies have demonstrated that chronic inflammation promotes the accumulation of senescent‐like glial cells more prominently in WMLs than in GMLs [39]. Together, these studies support our conclusion that cellular senescence in MS is not a uniform process but instead is strongly influenced by the lesion microenvironment.

In addition to cellular senescence, EIF2 signaling emerged as another highly dysregulated canonical pathway. The ISR is mediated by phosphorylation of EIF2, which induces stress‐responsive genes while transiently reducing global protein synthesis, thereby promoting cell survival under chronic stress [40]. Recent studies have shown that EIF2‐mediated stress signaling can enhance oligodendrocyte survival and promote remyelination in demyelinating disease models [41, 42]. These findings provide a functional framework supporting our results, which indicate that EIF2‐mediated ISR is predicted to be selectively activated in WMLs as an adaptive response to inflammatory and metabolic stress. In contrast, our data predict suppression of EIF2 signaling in GMLs, suggesting that oligodendrocytes in GM may fail to mount an effective stress response. These molecular differences provide rationale for deeper investigation into regional disparities in oligodendrocyte survival and remyelination capacity.

A key outcome of our study is the identification of regulatory miRNAs dysregulated in both WMLs and GMLs whose target gene networks are enriched for cellular senescence and EIF2 signaling pathways. From this group of nine miRNAs, we identified miR‐30a and miR‐100 as particularly noteworthy due to their opposing expression patterns between WM and GM: both were downregulated in GMLs and upregulated in WMLs. This unique bidirectional regulation suggested a role for these miRNAs in mediating region‐specific molecular responses. This hypothesis was further supported by our in vitro studies, which demonstrated that overexpression of miR‐30a or miR‐100 inhibited both OPC proliferation and differentiation. In vitro validation confirms that the overexpression of miR‐30a and miR‐100 may contribute to the regulation of a network of genes essential for OPC homeostasis and differentiation. The downregulation of ELOVL5 (lipid metabolism), NCAM1 (cell adhesion), and LIFR (growth factor signaling) provides a clear mechanistic framework for why OPCs in the WML microenvironment fail to proliferate and achieve terminal maturation. These results, however, provide rationale for more detailed investigation into individual markers of the cell senescence and ISR response pathways underlying differential remyelination in MS lesions. This study has several limitations. First, the sample size of postmortem progressive MS tissue (n = 5 per compartment) is relatively small, reflecting limited tissue availability and potentially reducing statistical power and generalizability. We partially tried to address this by integrating independent datasets and functional validation; however, validation in larger cohorts is needed. Second, we acknowledge that the analyses were performed on bulk tissue where differences in cellular composition between WMLs and GMLs may influence expression profiles. We tried to address the cell specificity through integration of our data with a previously published snRNA‐seq dataset (GSE180759). Finally, functional studies presented here relied on miRNA overexpression, which models increased expression in WMLs but does not fully address reduced expression in GMLs. Future studies involving complementary loss‐of‐function studies and direct target validation will be required to confirm causal mechanisms.

In summary, our findings demonstrate that beyond histopathological differences, WMLs and GMLs in progressive MS exhibit distinct miRNA‐driven molecular programs. These results highlight the importance of considering region‐specific miRNA regulation when developing therapeutic strategies aimed at enhancing remyelination and tissue repair.

Author Contributions

A.S., A.T., T.N., N.K.R., H.C. performed experiments and data analysis. A.S. and R.D. wrote the manuscript. All authors have seen and approved the submitted manuscript.

Funding

National Institute of Neurological Disorders and Stroke (NS123546).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: miRNAs and gene TaqMan assays IDs and catalog numbers used for the RT‐qPCR validation study.

Funding Statement

This work was funded by National Institute of Neurological Disorders and Stroke grant NS123546.

Data Availability Statement

The previously published miRNAs microarray data [14, 15], as well as all raw data included in the current manuscript will be available from the corresponding author (duttar@ccf.org).