MyD88 Deficiency Protects Mice From Experimental Autoimmune Encephalomyelitis by Influencing Both Dendritic Cells and T Cells

TL;DR

MyD88 deficiency in mice reduces autoimmune inflammation by affecting dendritic cells and T cells, offering a potential treatment target for multiple sclerosis.

Contribution

The study reveals MyD88's dual role in dendritic cells and T cells during autoimmune neuroinflammation.

Findings

MyD88 deficiency impairs dendritic cell maturation and cytokine production.

MyD88 deficiency reduces Th1 and Th17 cell differentiation and T cell activation.

MyD88 is highly expressed in dendritic cells and CD4+ T cells in multiple sclerosis patients.

Abstract

Dendritic cells (DCs) play a central role in both the development and maintenance of adaptive immunity by their ability to prime and regulate T cell function. These interactions between DCs and T cells are crucial to the pathogenesis of multiple sclerosis (MS) and its animal model, experimental autoimmune encephalomyelitis (EAE). Myeloid differentiation primary response protein 88 (MyD88) signalling is pivotal in the pathogenesis of MS and EAE; however, its specific contributions across various cell types in the context of these conditions remain inadequately understood. In this study, we reanalysed single‐cell RNA sequencing data from MS patients and revealed significant upregulation of MYD88 in DCs and CD4+ T cells isolated from PBMCs of MS patients. Single‐cell RNA sequencing analysis revealed that during the peak phase of EAE, Myd88 is highly expressed in moDCs and pDCs compared to cDCs. Notably, the absence of Myd88 in DCs resulted in significantly reduced interactions with T cell clusters. Our in vivo and in vitro results showed that while MyD88 deficiency did not affect lymphocyte production in the thymus, it resulted in impaired Th1 and Th17 cell differentiation and diminished T cell activation. Mechanistically, MyD88−/− DCs exhibited impaired maturation and a reduced production of pro‐inflammatory cytokines, such as IL‐6, TNF‐α, and IL‐12, which may be linked to their role in directing Th1 and Th17 cell differentiation. Our findings suggest that MyD88 is essential for the priming of inflammatory T cells and the activation of DCs, and their interactions with T cells, underscoring its role in neuroinflammation. This study highlights the potential therapeutic implications of targeting MyD88 pathways in MS and related disorders.

This study investigates the role of MyD88 in dendritic cells (DCs) and T cells during experimental autoimmune encephalomyelitis (EAE). We found that MyD88 is highly expressed in DCs and CD4+ T cells in people with multiple sclerosis, and its deficiency impairs DC maturation, reduces pro‐inflammatory cytokine production, weakens DC–T cell interactions, and limits Th1/Th17 differentiation, highlighting its critical role in driving autoimmune neuroinflammation and its potential as a therapeutic target.

Figures

MyD88 is involved in the pathogenesis of MS and EAE. (A) Heatmap of MYD88 gene expression in DCs and CD4+ T cells derived from PBMC samples collected from MS patients and healthy controls using the GEO dataset (GSE138266). (B) Western blotting analysis for MYD88 expression from the spinal cord of CFA mice and EAE mice (n = 6/group) during the onset of disease. (C) Active EAE was induced in the MyD88+/+ mice and MyD88−/− mice (n = 15/group). Clinical disease was monitored for 16 days. Three independent experiments have been performed. (D) Flow cytometric quantification of antigen‐specific cytokine production by CD4+ T cells. Splenocytes from MyD88+/+ mice and MyD88−/− mice on day 10 post EAE induction were cultured with MOG35–55 peptide (50 μg/mL). After 18 h, the cells were harvested, and intracellular staining for IFN‐γ and IL‐17A was performed (n = 3–6/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05, and *p < 0.001. Data are presented as mean ± SEM.

MyD88 regulates inflammatory T cell development both in vivo and in vitro. (A) Flow cytometric quantification of IFN‐γ, IL‐17A and GM‐CSF producing CD4+ T cells in the spleen of MyD88+/+ and MyD88−/− mice on day 10 post EAE induction (n = 3–6/group). (B) Heatmap of MyD88 gene expression among different Th cells derived from the public RNA‐seq dataset GSE206304. (C) Representative histogram showing MyD88 deletion attenuated the expansion in cell size and granularity of TCR‐stimulated CD4+ T cells based on their FSC and SSC profiles after 3 days (n = 3/group). (D) Flow cytometric quantification of cytokine‐producing CD4+ T cells (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. p < 0.01, and p < 0.0001. Data are presented as mean ± SEM.

MyD88 expression is required for DC maturation. (A) Flow cytometric quantification of the total CD11c+ DCs in the spleen of MyD88+/+ and MyD88−/− mice on day 15 post EAE induction (n = 5/group). (B) Flow cytometric quantification of CD86, CD80, and CD40 on the surface of DCs in the spleen (n = 5/group). (C) DCs were isolated from naïve MyD88+/+ and MyD88−/− mice and stimulated with either LPS, CpG DNA, or MOG35‐55 for 24 h. The maturation of DCs was assessed by measuring MFI of CD40, CD80, and CD86 (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05, p < 0.01, *p < 0.001, and p < 0.0001. Data are presented as mean ± SEM.

DC subset composition in naïve and EAE spleen and expression of Myd88. (A) UMAP projection of all 1211 DCs presenting 4 different subset clustering in naive versus EAE spleen. (B) Bubble plot of annotated markers for DC subsets. (C) Stacked bar plot of DC subset proportions. (D) Bubble plot of Myd88 expression in DC subsets from EAE spleen. (E) UMAP plot of EAE‐affected MyD88+ and MyD88− DCs. (F) Stacked bar plot of the proportion of EAE‐affected MyD88+ and MyD88− DCs.

Cellular communication analysis between MyD88+ and MyD88− DCs and T cells in EAE. (A) Cell communication diagram. (B) Interaction heatmap of cell communication. (C) Bubble plot of cell communication. (D) Chord diagram showing preferential interactions between T cell and MyD88+ DC clusters.

Deletion of MyD88 reprograms the gene expression profile of DCs. (A) Volcano plot of RNA‐seq gene counts in DCs from MyD88+/+ and MyD88−/− mice on day 10 post EAE induction (n = 4/group). (B) Gene set enrichment analysis showing changed pathways from differentially expressed genes between MyD88+/+ and MyD88−/− DCs. (C) Altered gene expression associated with cytokines and chemokine expression (n = 4/group). (D) qPCR analysis of cytokine expression in the DCs of MyD88+/+ and MyD88−/− mice on day 10 post EAE induction (n = 4‐5/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05 and *p < 0.001. Data are presented as mean ± SEM.

DC‐MyD88 regulates Th cell differentiation by influencing the production of inflammatory cytokines. (A) Schematic illustration of the DC‐T cell coculture system. (B) Intracellular cytokine staining was performed to analyze lineage‐specific cytokine expression, and representative flow plots are shown. (C) Flow cytometric quantification of IFN‐γ‐ and IL‐17A‐producing CD4+ T cells from the DC‐T cell coculture system in the presence or absence of IL‐12p70 or IL‐6 in vitro (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05 and p < 0.01. Data are presented as mean ± SEM. Schematics were created using BioRender (https://biorender.com).

Full text

Introduction

1

Multiple sclerosis (MS) is the most common demyelinating disease of the central nervous system (CNS) that involves many different types of immune cells [1, 2]. Dendritic cells (DCs) serve as primary antigen‐presenting cells in both MS and its animal model, experimental autoimmune encephalomyelitis (EAE). They initiate and amplify the response of autoreactive T cells that target CNS myelin components. Upon activation, these myelin‐specific T cells migrate to the CNS, where they interact with CNS‐infiltrating DCs presenting their specific antigens. This interaction leads to the reactivation of T cells by CNS‐resident DCs, triggering the release of inflammatory mediators that recruit additional immune cells, thereby exacerbating the autoimmune assault on myelin sheaths [3]. Furthermore, DCs within the CNS can directly drive the differentiation of naïve T cells into pathogenic Th17 cells, which contribute to the demyelination and neurodegeneration observed in EAE [4, 5]. However, our understanding of the factors regulating these interactions in EAE remains limited.

Myeloid differentiation primary response protein 88 (MyD88) is a multifunctional protein associated with various autoimmune diseases [6, 7, 8, 9, 10, 11, 12, 13, 14]. Numerous lines of evidence have established the critical role of MyD88 signalling in the development of EAE. Notably, the increase in Myd88 mRNA levels in the CNS has been documented during both the onset and peak stages of EAE, indicating its importance in disease pathogenesis [14]. Additionally, MyD88^−/−^ mice exhibit complete resistance to EAE [13, 14, 15, 16]. These studies have demonstrated defects in Th17 cell priming due to the impaired functions of both MyD88^−/−^ DCs and MyD88^−/−^ CD4^+^ T cells. In particular, MyD88^−/−^ DCs produced significantly lower levels of cytokines such as IL‐6 and IL‐23p40, which are essential for directing Th17 differentiation [15]. While these studies elucidate the roles of MyD88 in EAE development, the extent to which the protective phenotype observed in MyD88^−/−^ mice involves interactions between DCs and CD4^+^ T cells is not completely understood.

Here, we found that Myd88 mRNA levels were enriched in DCs and CD4^+^ T cells from PBMCs of MS patients. Our in vivo results provide evidence that MyD88 deficiency impaired Th17 cell development and DC function. Utilizing single‐cell RNA sequencing, we investigated the complex cellular composition of the spleen during the peak phase of EAE, identifying 4 distinct subclusters of DCs: cDC1, cDC2, moDC, and pDC. Notably, Myd88 expression was found to be particularly high in moDCs and pDCs and non‐Myd88‐expressing DCs showed diminished interactions with T cell clusters. Furthermore, MyD88 regulates the secretion of inflammatory mediators from DCs, which are crucial for effective T cell polarisation and the maintenance of the autoimmune response. Overall, these findings underscore the critical role of MyD88 in mediating immune responses in MS and EAE and suggest potential therapeutic approaches for modulating MyD88 signalling in autoimmune diseases.

Materials and Methods

2

Mice

2.1

MyD88^+/+^ and MyD88^−/−^ mice were obtained from Professor Peiris, Joseph Sriyal Malik at the University of Hong Kong, which were originally provided by Dr. Shizuo Akira (Osaka University, Osaka, Japan). 6‐ to 8‐week‐old animals of both sexes were used in this study. Genotyping details are shown in Figure S1A and the knockout of MyD88 was verified in splenocytes by western blotting (Figure S1B). The 2D2 TCR transgenic mice, C57BL/6‐Tg (Tcra2D2, Tcrb2D2) 1Kuch/J (RRID: IMSR_JAX: 006912), were acquired from the Jackson Laboratory (Bar Harbour, ME). All breeders and experimental mice were housed under specific pathogen‐free conditions following the Institutional Laboratory Animal Research Unit Guidelines of the City University of Hong Kong. All animal experiments conducted in this study were approved by the Animal Ethics Committee of the City University of Hong Kong (AN‐STA‐00000146).

Induction and Evaluation of Active EAE

2.2

To induce active EAE, 6‐ to 8‐week‐old MyD88^+/+^ and MyD88^−/−^ mice were subcutaneously immunised with 200 μL of 2 mg/mL MOG_35‐55_ peptide (MEVGWYRSPFSRVVHLYRNGK) (Thermo Scientific), which was emulsified with 2 mg/mL of Mycobacterium tuberculosis H37RA (BD Bioscience, 231141) in incomplete Freund's adjuvant (BD Bioscience, 263910), at bi‐sites on the lower flank area. All mice then received intraperitoneal injections of 400 ng pertussis toxin (Calbiochem, 516560) on day 0 and day 2 after immunisation. Clinical signs of EAE were monitored daily and scored as previously described [17]. At 10‐ and 16‐days post‐immunisation, mice from both groups were sacrificed, and spleens, spinal cords, and thymuses were collected for analysis. Our study included both male and female mice, with no significant gender differences noted in the results.

Antigen Recall Assay

2.3

To measure peripheral T cell antigen recall response during EAE, splenocytes from MyD88^+/+^ and MyD88^−/−^ mice induced for EAE (day 10) were collected and homogenised into single cell suspensions, and erythrocytes were lysed. Splenocytes (5 × 10^5) were seeded in 96‐well round bottom plates in a total volume of 200 μL per well and pulsed with MOG_35‐55_ peptide (50 μg/mL) in the presence of GolgiStop (BD Biosciences, 554724). Cells were stimulated for 18 h at 37°C and 5% CO_2_ atmosphere. Flow cytometric analyses were performed for detecting cytokine (IFN‐γ, IL‐17A, and GM‐CSF)‐producing cells using a BD Celesta flow cytometer (BD Biosciences). All data were analysed using FlowJo software (Version 10.8.1, BD Biosciences).

RNA Isolation and Quantitative PCR

2.4

Splenic CD11c^+^ cells from MyD88^+/+^ and MyD88^−/−^ mice on day 10 post EAE induction were enriched with the EasySep Mouse CD11c Positive Selection Kit (Stemcell, 18 780). The RNA was extracted using RNAiso plus (Takara, 9109) and then reverse‐transcribed into cDNA using HiScript III All‐in‐one RT SuperMix (Vazyme, R333‐01). qPCR was conducted using Taq Pro Universal SYBR qPCR Master Mix (Vazyme, Q712) on a QuantStudio 7 Pro Dx RealTime PCR System (Thermo Fisher). The results are presented as relative expression values normalised to Actb. The following primer pairs were used: Actb (forward 5′‐GTGACGTTGACATCCGTAAAGA‐3′ and reverse 5′‐GCCGGACTCATCGTACTCC‐3′), Il6 (forward 5′‐ACTTCCAGCCAGTTGCCTTCTTG‐3′ and reverse 5′‐TGGTCTGTTGTGGGTGGTATCCTC‐3′), Il23 (forward 5′‐AGCGGGACATATGAATCTACTAAGAGA‐3′ and reverse 5′‐GTCCTAGTAGGGAGGTGTGAAGTTG‐3′), Tgfb1 (forward 5′‐TACCATGCCAACTTCTGTCTGGGA‐3′ and reverse 5′‐ATGTTGGACAACTGCTCCACCTTG‐3′), Tnfa (forward 5′‐AAAGGACACCATGAGCACGGAAAG‐3′ and reverse 5′‐CGCCACGAGCAGGAATGAGAAG‐3′), Il1b (forward 5′‐TTCAGGCAGGCAGTATCACTC‐3′ and reverse 5′‐GAAGGTCCACGGGAAAGACAC‐3′), Il12b (forward 5′‐GACCATCACTGTCAAAGAGTTTCTAGAT‐3′ and reverse 5′‐AGGAAAGTCTTGTTTTTGAAATTTTTTAA‐3′), Il10 (forward 5′‐AGGCGCTGTCATCGATTTCT‐3′ and reverse 5′‐GGCCTTGTAGACACCTTGGTC‐3′).

Bulk RNA‐Sequencing of Sorted DCs

2.5

Splenic CD11c^+^ cells from MyD88^+/+^ and MyD88^−/−^ mice on day 10 post EAE induction were enriched with the EasySep Mouse CD11c Positive Selection Kit (Stemcell, 18780) for bulk RNA‐sequencing analysis. Prior to downstream analysis, the quality of the sequencing data was assessed using “FastQC” and “MultiQC,” and the mean Phred quality score of each sample was above 36.5. The “STAR” aligner was then used to map the reads to the reference genome, and “featureCounts” was employed to generate gene expression matrices for both groups. To investigate the differences between the MyD88^+/+^ and MyD88^−/−^ groups, the R package “DESeq2” was utilised. Differential expression analysis was performed, filtering out genes with an adjusted p‐value below 0.05 and a fold change exceeding 2.

Single‐Cell RNA Sequencing and Data Processing

2.6

The single‐cell suspension obtained from the spleens of wild‐type mice on day 17 post EAE induction was sent for 10× Genomics single‐cell sequencing. Library was prepared in accordance with the 10× Genomics protocol (10× Genomics, Pleasanton, CA). The analysis of scRNA‐seq data was conducted using Seurat (version 5.3.0) and SCP (version 0.4.2). The filtering criteria encompassed the total number of genes, the percentage of mitochondrial RNA, and the count of unique molecular identifiers (UMIs). A standard data integration workflow was executed using Seurat. After integration, the data underwent scaling, principal component analysis (PCA), dimensionality reduction via UMAP, and clustering for identifying and visualising clusters. Cell markers were filtered based on the following criteria: “p < 0.05, 0.5 < pct.1 < 1, pct.2 < 0.5, > 0.25 log‐fold,” derived from the FindAllMarkers function, which was utilised to categorise different cell groups.

Cell–Cell Communication Analysis

2.7

The CellChat package (version 2.2.0) was employed to infer, analyse, and visualise cell–cell communication among different immune cells, referencing the CellChat database (https://github.com/sqjin/CellChat). For the analysis of cellular interactions, expression levels corresponding to the total number of reads were computed and aligned with a consistent set of coding genes across all transcriptomes. Expression values were averaged for each cell sample or single‐cell cluster. Differential analysis of intercellular communication involved calculating and comparing the information flow for each signalling pathway, defined as the probabilities of communication between all pairs of cell populations within the inferred network.

Immunoblotting

2.8

Spinal cord or spleen tissues from MyD88^+/+^ and MyD88^−/−^ mice on day 10 post EAE induction were homogenised and directly lysed in RIPA buffer (Solarbio, R0020) supplemented with PMSF (Solarbio, P0100) on ice for 30 min before denaturation at 105°C for 5 min. Protein samples were separated via 12% SDS‐PAGE, blotted onto nitrocellulose membranes, blocked with 5% nonfat milk PBST (PBS with Tween 20) and incubated with the following primary antibodies: anti‐MyD88 (ABclonal, A22600) and anti‐ACTB (ABclonal, AC026). Primary antibodies were detected with horseradish peroxidase‐labelled goat anti‐rabbit (ABclonal, AS014) or anti‐mouse (ABclonal, AS003) and visualised using a SuperFemto ECL Chemiluminescence Kit (Vazyme, E423) and detected using a digital ChemiDoc imaging system (BioRad).

T Cell Activation and T Cell Differentiation

2.9

Fresh CD4^+^ T cells were purified from naïve MyD88^+/+^ or MyD88^−/−^ mouse splenocytes by negative selection using a CD4^+^ T Cell Isolation Kit (Miltenyi Biotec, 130–104–454, purity > 95%) as described in the manufacturer's protocols. T cell activation assay was conducted as previously described [18]. Briefly, purified T cells were activated with 2 μg/mL plate‐bound anti‐CD3 (Invitrogen, 145‐2C11) and 1 μg/mL anti‐CD28 (Invitrogen, 37.51) antibodies in complete RPMI 1640 medium for 3 days. For Th cell differentiation, purified T cells were activated and polarised into various Th subsets using the following cytokines and antibodies. For Th1 cells: 15 ng/mL rmIL‐12 (Peprotech, 210–12), 10 ng/mL rhIL‐2 (Peprotech, 212–12), and 5 μg/mL anti‐IL‐4 (Biolegend, 504122). For Th17 cells: 50 ng/mL rmIL‐6 (Peprotech, 216–16), 5 ng/mL rhTGF‐β1 (Peprotech, 100‐21C), 10 μg/mL anti‐IL‐4 (Biolegend, 504122), and 10 μg/mL anti‐IFN‐γ (Biolegend, 505834).

Flow Cytometry

2.10

Spleens, spinal cords, and thymuses from MyD88^+/+^ and MyD88^−/−^ mice were homogenised into single‐cell suspensions and passed through a 70‐μm strainer. Surface staining was conducted at 4°C using fluorescently labelled monoclonal antibodies against mouse CD45, CD3, CD4, CD8α, CD11b, CD11c, CD40, CD80, and CD86 (Biolegend). Intracellular staining was conducted using fluorescently labelled monoclonal antibodies against mouse FoxP3 (eBioscience, 11–5773‐82) and the FoxP3 Fix/Perm Buffer Set (Biolegend, 421403). To detect the expression of IL‐17A, IFN‐γ, and GM‐CSF in CD4^+^ T cells, splenocytes were first treated with a cell stimulation cocktail with Brefeldin A (Biolegend, 423303) for 4 h at 37°C. Cells were stained with Ghost Dye Violet 510 (Tonbo Reagents, SKU 13–0870‐T100) to assess cell viability. After surface staining, cells were fixed with Perm‐Fix solution (BD Biosciences, 51‐2090KZ) at 4°C and intracellular cytokines were stained for 30 min at 4°C. All sample acquisition was performed with a BD Celesta flow cytometer from BD Biosciences and further analysed with the FlowJo software (version 10.8.1, BD Biosciences).

Evaluation of DC Activation

2.11

To assess the maturation status of DCs, isolated splenic DCs from naïve MyD88^+/+^ or MyD88^−/−^ mice were primed with LPS, CpG DNA, or MOG_35‐55_ for 24 h at 37°C. DCs were cultured in RPMI 1640 complete medium at a concentration of 2 × 10^5^ cells/mL and were analyzed for their activation state by assessing the expression levels of CD40, CD80, and CD86 using flow cytometry.

DC and T Cell Co‐Culture

2.12

DCs were isolated from the spleens of naïve MyD88^+/+^ and MyD88^−/−^ mice. Naïve MOG_35‐55_‐specific CD4^+^ T cells were purified from the spleens of 2D2 mice. MyD88^+/+^ and MyD88^−/−^ DCs were pulsed with 20 μg/mL MOG_35‐55_ for 30 min and cocultured with naïve 2D2 T cells at a ratio of 1:10 in the presence of anti‐IL‐4 (10 μg/mL, for Th1), or rhTGF‐β1 (3 ng/mL) + anti‐IL‐4 (10 μg/mL) + anti‐IFN‐γ (15 μg/mL, for Th17). To evaluate the impact of DC‐secreted inflammatory factors in T cell differentiation, MyD88^−/−^ DCs were supplemented with IL‐12p70 (for Th1) or IL‐6 (for Th17). After 4 days of co‐culture, T cells underwent intracellular cytokine staining and flow cytometry analysis to identify Th1 cells (CD4^+^ IFN‐γ^+^) and Th17 cells (CD4^+^ IL‐17A^+^).

Histopathology

2.13

Spinal cord tissues from EAE mice (day 15) were dissected and fixed in 4% PFA for 48 h. Tissue sections were flattened, dehydrated, processed for paraffin embedding and cut into 5 μm sections for haematoxylin and eosin (H&E) staining. For detection of demyelination and leukocytic infiltration, spinal cord samples were processed and stained with Luxol Fast Blue (Roth) and Cresyl Violet (Sigma).

Database Analysis

2.14

The bulk mRNA sequencing data for Th0, Th1, pathogenic Th17, and non‐pathogenic Th17 cells were obtained from the Gene Expression Omnibus database (GSE206304 [19]). Publicly available MS patient PBMC scRNA‐sequencing data were downloaded under accession code GSE138266 [20]. Furthermore, published scRNA‐sequencing data of naïve murine spleens were accessed and downloaded from public repositories under the accession codes GSE222784 [21] and GSE186158 [22].

Statistical Analysis

2.15

The results are presented as mean ± standard error of the mean (SEM). Unpaired Student's t‐test using GraphPad Prism 9 (GraphPad Software, San Diego, CA) was used to assess the statistical significance of differences between experimental groups. Significance levels are indicated in the figures as follows: p < 0.05 (noted as ), p < 0.01 (), p < 0.001 (), and p < 0.0001 (****). The figure legends provide the number of independent experiments or mice per group used in the respective studies.

Results

3

MyD88 Is Involved in the Pathogenesis of MS and EAE

3.1

To study the role of MyD88 in MS/EAE development, we first examined whether MyD88 level is altered in MS/EAE. We reanalysed the scRNA‐seq data from PBMCs from MS patients and a control cohort of patients with idiopathic intracranial hypertension [20]. The MYD88 transcript was significantly upregulated in DCs and CD4^+^ T cells in the PBMCs of MS patients (Figure 1A), suggesting a potential involvement of MYD88 expression in these cells in the pathogenesis of MS. We further found that the MYD88 protein expression in the spinal cord was significantly increased at the onset stage of EAE (Figure 1B). To directly investigate the role of MyD88 in the development of EAE, we induced EAE in the MyD88^−/−^ mice using the MOG_35‐55_ emulsified in complete Freund's adjuvant followed by administration of pertussis toxin. In consistency with previous studies [13, 14, 15, 16], all MyD88^−/−^ mice demonstrated complete resistance to EAE development (Figure 1C). At the peak stage of EAE, the mice were sacrificed for evaluation of immune cell infiltration and demyelination in the spinal cord. MyD88^−/−^ mice were protected from the infiltration of inflammatory lymphocytes and myeloid cells, particularly CD4^+^ T cells, microglia (CD11b^+^ CD45^low^), macrophages (CD11b^+^ CD45^hi^), and MHC II^+^ CD11c^+^ populations (Figure S2A). Consistently, spinal cord tissues from MyD88^−/−^ mice exhibited no signs of inflammation or demyelination (Figure S2B). We further investigated whether this protection against EAE was due to impaired induction of peripheral antigen‐specific CD4^+^ effector T cell responses. To explore this, we isolated splenocytes from MyD88^+/+^ and MyD88^−/−^ mice on day 10 post EAE induction and stimulated them with MOG_35–55_ peptide (50 μg/mL) to assess antigen‐specific cytokine production. Our results indicated that MyD88^−/−^ T cells produced minimal levels of IFN‐γ, IL‐17A, and GM‐CSF (Figure 1D), indicating an essential role of MyD88 in T cell priming during the induction phase of EAE. Together, these findings put forth a role for MyD88 in MS/EAE pathogenesis.

*MyD88 is involved in the pathogenesis of MS and EAE. (A) Heatmap of MYD88 gene expression in DCs and CD4+ T cells derived from PBMC samples collected from MS patients and healthy controls using the GEO dataset (GSE138266). (B) Western blotting analysis for MYD88 expression from the spinal cord of CFA mice and EAE mice (n = 6/group) during the onset of disease. (C) Active EAE was induced in the MyD88+/+ mice and MyD88−/− mice (n = 15/group). Clinical disease was monitored for 16 days. Three independent experiments have been performed. (D) Flow cytometric quantification of antigen‐specific cytokine production by CD4+ T cells. Splenocytes from MyD88+/+ mice and MyD88−/− mice on day 10 post EAE induction were cultured with MOG35–55 peptide (50 μg/mL). After 18 h, the cells were harvested, and intracellular staining for IFN‐γ and IL‐17A was performed (n = 3–6/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05, and **p < 0.001. Data are presented as mean ± SEM.

MyD88 Regulates Inflammatory T Cell Development Both in Vivo and in Vitro

3.2

Previous studies from our lab [23] and others [24, 25] have shown altered T cell development in EAE and MS; we hypothesised that MyD88 deficiency‐mediated protection against EAE may be related to impaired T cell development in these mice. However, we found comparable frequencies of thymocyte subpopulations between MyD88^+/+^ and MyD88^−/−^ mice (Figure S3A,B), indicating no abnormalities in thymocyte development due to MyD88 deficiency. Additionally, the frequencies of CD4^+^ IFN‐γ^+^, CD4^+^ IL‐17A^+^, and CD4^+^ FoxP3^+^ cells in the spleen were similar in both groups (Figure S4A), suggesting that MyD88 deficiency did not influence the peripheral T cell homeostasis. To further investigate the impact of MyD88 deficiency on inflammatory T cell development during the onset of EAE, we harvested spleens from EAE mice on day 10 post‐immunisation and assessed inflammatory T cell development. MyD88^−/−^ mice exhibited a reduced percentage of Th17 cells among total CD4^+^ T cells compared to MyD88^+/+^ mice, while the proportions of CD4^+^ T cells producing IFN‐γ and GM‐CSF remained equivalent (Figure 2A). Although it has been proposed that Tregs could contribute to the inadequate T cell priming observed in MyD88^−/−^ mice [26], our findings indicated that the frequency of Tregs was actually lower in the spleen of MyD88^−/−^ mice compared with MyD88^+/+^ mice during EAE (Figure S4B). These results imply that the impaired T cell priming in MyD88^−/−^ mice is unlikely to be mediated by Tregs, but rather reflects a deficiency in the initiation of adaptive immune responses. To understand the role of MyD88 in Th cell differentiation, we compared the gene expression profile of MyD88 across different Th cell types using the public dataset GSE206304 [19]. We found that Th1 and pathogenic Th17 cells expressed upregulated levels of Myd88 (Figure 2B). To investigate the role of T cell‐MyD88 in regulating T cell activation, we isolated naïve CD4^+^ T cells from MyD88^+/+^ and MyD88^−/−^ mice and compared their primary responses to TCR stimulation with anti‐CD3 and anti‐CD28. MyD88^−/−^ CD4^+^ T cells showed reduced expansion in cell size and granularity upon TCR stimulation (Figure 2C). To further evaluate whether MyD88 has a direct effect on Th cell differentiation, we cultured MyD88^+/+^ and MyD88^−/−^ CD4^+^ T cells under Th1 and Th17 skewing conditions. Our results showed that MyD88^−/−^ CD4^+^ T cells showed a substantial defect in IFN‐γ and IL‐17A production under Th1 and Th17 cell‐polarising conditions (Figure 2D). Together, these data indicate that MyD88 is required for inflammatory T cell development both in vivo and in vitro.

*MyD88 regulates inflammatory T cell development both in vivo and in vitro. (A) Flow cytometric quantification of IFN‐γ, IL‐17A and GM‐CSF producing CD4+ T cells in the spleen of MyD88+/+ and MyD88−/− mice on day 10 post EAE induction (n = 3–6/group). (B) Heatmap of MyD88 gene expression among different Th cells derived from the public RNA‐seq dataset GSE206304. (C) Representative histogram showing MyD88 deletion attenuated the expansion in cell size and granularity of TCR‐stimulated CD4+ T cells based on their FSC and SSC profiles after 3 days (n = 3/group). (D) Flow cytometric quantification of cytokine‐producing CD4+ T cells (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. **p < 0.01, and ***p < 0.0001. Data are presented as mean ± SEM.

MyD88 Expression Is Required for DC Maturation

3.3

In autoimmune diseases, DCs regulate tissue pathology by priming and differentiating pathogenic and regulatory T cells, reactivating autoimmune T cells in target tissues, and controlling epitope spreading [27, 28]. Alterations in DC function can lead to an imbalance in immune responses and contribute to immune pathologies such as systemic lupus erythematosus, inflammatory bowel disease, and MS [29, 30, 31, 32]. Therefore, we speculated that the MyD88 deficiency‐mediated protection against EAE could also be due to impaired DCs' ability to initiate and amplify autoreactive CD4^+^ T cells in peripheral compartments. To test this, we first compared the effects of MyD88 deletion on the development and maturation of DCs during the peak stage of EAE. Although the total number and frequency of CD11c^+^ DCs in the spleen was comparable between MyD88^+/+^ and MyD88^−/−^ mice (Figure 3A), MyD88^−/‐^DCs exhibited significantly reduced expression of the co‐stimulatory molecules CD40, CD80, and CD86 (Figure 3B). Since CpG‐DNA and LPS are potent adjuvants for activating autoreactive encephalitogenic T cells [33, 34], and the induction of EAE requires DCs to effectively phagocytose and process the MOG_35‐55_ peptide, we then assessed the maturation status of MyD88^−/−^ DCs upon stimulation with LPS, CpG DNA, and MOG_35‐55_ peptide. Our results indicated that, unlike LPS and MOG_35‐55_ peptide, CpG DNA induces DC maturation in a MyD88‐dependent manner (Figure 3C). Taken together, these data indicate that MyD88 deficiency hinders DC maturation.

*MyD88 expression is required for DC maturation. (A) Flow cytometric quantification of the total CD11c+ DCs in the spleen of MyD88+/+ and MyD88−/− mice on day 15 post EAE induction (n = 5/group). (B) Flow cytometric quantification of CD86, CD80, and CD40 on the surface of DCs in the spleen (n = 5/group). (C) DCs were isolated from naïve MyD88+/+ and MyD88−/− mice and stimulated with either LPS, CpG DNA, or MOG35‐55 for 24 h. The maturation of DCs was assessed by measuring MFI of CD40, CD80, and CD86 (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05, **p < 0.01, ***p < 0.001, and ***p < 0.0001. Data are presented as mean ± SEM.

Single‐Cell RNA Sequencing Uncovers Composition of DC Subsets and Myd88 Expression in naïve and EAE Mice

3.4

Understanding the heterogeneity of DC subpopulations is crucial for elucidating their roles in T cell activation during EAE. To explore these heterogeneities, we conducted single‐cell RNA sequencing on immune cells from the spleens of wild‐type mice at the peak stage of EAE and integrated this data with publicly available single‐cell RNA sequencing from naïve spleens [21, 22]. Overall, 46506 cells passing the quality control were eligible for subsequent analysis. Based on the expression of canonical genes, the clusters were annotated into 12 major immune cell types: dendritic cells, neutrophils, CD4 T cells, CD8 T cells, DN T cells, Tregs, B cells, NK cells, plasma cells, monocytes, macrophages, and mast cells (Figure 4A). To gain insight into the heterogeneity of functional subtypes of DCs in EAE, we performed unsupervised re‐clustering using the UMAP algorithm, identifying 4 subclusters—cDC1, cDC2, moDC, and pDC—comprising 599 cells from naïve mice and 612 cells from EAE mice (Figure 4A), based on high expression of canonical marker genes (Figure 4B). Specifically, we observed an increase in cDC1 and moDC following EAE induction (Figures 4C), with Myd88 being highly expressed in moDCs and pDCs compared to the cDC1 and cDC2 clusters (Figures 4D).To further investigate the role of MyD88 in these DC subtypes, we selected all DC clusters from EAE samples in the single‐cell dataset and categorised those expressing MyD88 at levels greater than 0 as MyD88^+^ DCs, while those lacking MyD88 expression were designated as MyD88^−^ DCs (Figure 4E). Similar to findings in tumour studies [35], the proportion of MyD88^−^ DC subpopulations was significantly higher than MyD88^+^ subpopulations in EAE (Figure 4F).

DC subset composition in naïve and EAE spleen and expression of Myd88. (A) UMAP projection of all 1211 DCs presenting 4 different subset clustering in naive versus EAE spleen. (B) Bubble plot of annotated markers for DC subsets. (C) Stacked bar plot of DC subset proportions. (D) Bubble plot of Myd88 expression in DC subsets from EAE spleen. (E) UMAP plot of EAE‐affected MyD88+ and MyD88− DCs. (F) Stacked bar plot of the proportion of EAE‐affected MyD88+ and MyD88− DCs.

Cell–Cell Communication Analysis Reveals Reduced Interaction Between MyD88

− DCs and T Cells

3.5

To investigate the implications of MyD88‐mediated cross‐talk between DCs and T cells in EAE, we performed a ligand‐receptor network analysis comparing MyD88^+^ and MyD88^−^ DCs (sender cells) with T cell subpopulations (receiver cells) from the spleens of EAE‐affected mice. Our findings revealed that MyD88^−^ DCs exhibited decreased interaction strength and fewer connections with T cell subpopulations compared to MyD88^+^ DCs (Figure 5A,B). Notably, MyD88^+^ DCs demonstrated stronger interactions in ligand‐receptor pairs such as Tnf‐Tnfrsf1b, Sirpb1a/Sirpb1b‐Cd47, Selplg‐Sell, and (Itgav+Itgb1)‐Adgre5, indicating that DC‐MyD88 significantly influences T cell inflammatory responses, adhesion, and migration (Figure 5C,D). Additionally, MyD88^+^ DCs appear to preferentially engage Tregs through the Entpd1‐Adora2a axis. Moreover, the absence of MyD88 in DCs is predicted to impair the binding of H2‐t22 and H2‐m3 to Cd8a and Cd8b1 on CD8 T cells, suggesting that MyD88 is pivotal for CD8 T cell activation by regulating MHC molecules that present antigens to CD8 T cells via the CD8 co‐receptors (Figure 5C,D). Collectively, these results underscore the critical role of MyD88^+^ DCs in facilitating T cell responses during EAE.

Cellular communication analysis between MyD88+ and MyD88− DCs and T cells in EAE. (A) Cell communication diagram. (B) Interaction heatmap of cell communication. (C) Bubble plot of cell communication. (D) Chord diagram showing preferential interactions between T cell and MyD88+ DC clusters.

Deletion of Myd88 Reprograms DC Gene Expression

3.6

To gain a deeper understanding of the potential effects of MyD88 deletion on DC function, we examined the global transcriptomic profiles of splenic DCs from MyD88^+/+^ and MyD88^−/−^ mice on day 10 post EAE induction. RNA sequencing analysis uncovered substantial genome‐wide alterations in transcript levels. MyD88 deficiency in DCs significantly induced (n = 122) or reduced (n = 214) gene expression, including reduced Myd88 mRNA expression (2.07‐fold reduction, p = 4.79E‐11) (Figure 6A). Subsequent pathway enrichment analysis identified the impacted Th1 and Th17 cell differentiation pathway in MyD88^−/−^ DCs (Figure 6B). Previous studies have shown that the expression of many genes, such as Il6, Il23, Tgfb1, Tnfa, Il1b, Il12b, and Il10, plays a crucial role in promoting DC maturation or regulating their capacities in Th cell differentiation [36, 37, 38]. Our RNA sequencing data and subsequent qRT‐PCR analysis revealed the impaired cytokine gene expression in MyD88^−/−^ DCs (Figure 6C,D).

*Deletion of MyD88 reprograms the gene expression profile of DCs. (A) Volcano plot of RNA‐seq gene counts in DCs from MyD88+/+ and MyD88−/− mice on day 10 post EAE induction (n = 4/group). (B) Gene set enrichment analysis showing changed pathways from differentially expressed genes between MyD88+/+ and MyD88−/− DCs. (C) Altered gene expression associated with cytokines and chemokine expression (n = 4/group). (D) qPCR analysis of cytokine expression in the DCs of MyD88+/+ and MyD88−/− mice on day 10 post EAE induction (n = 4‐5/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05 and **p < 0.001. Data are presented as mean ± SEM.

DC‐MyD88 Regulates Th Cell Differentiation by Modulating the Production of Inflammatory Cytokines

3.7

To further investigate how MyD88 regulates polarisation of CD4^+^ T cells into specific Th cell subsets, we established a co‐culture model of DCs and naïve CD4^+^ T cells expressing a transgenic T cell receptor specific for the MOG_35‐55_ peptide (2D2 cells) (Figure 7A). We pulsed MyD88^+/+^ and MyD88^−/−^ DCs with MOG_35‐55_ peptide and co‐cultured them with naïve 2D2 CD4^+^ T cells. Our data suggest that MyD88^−/−^ DCs exhibit impaired ability in polarising naïve CD4^+^ T cells into Th1 and Th17 subsets (Figure 7B,C). Importantly, the additional supplementation of IL‐12p70 and IL‐6 restored the impaired differentiation of Th1 and Th17 cells in our co‐culture system (Figure 7B,C). Collectively, these data highlight that MyD88 is essential for DCs to produce proinflammatory cytokines and subsequent regulation of Th1/Th17 cell differentiation.

*DC‐MyD88 regulates Th cell differentiation by influencing the production of inflammatory cytokines. (A) Schematic illustration of the DC‐T cell coculture system. (B) Intracellular cytokine staining was performed to analyze lineage‐specific cytokine expression, and representative flow plots are shown. (C) Flow cytometric quantification of IFN‐γ‐ and IL‐17A‐producing CD4+ T cells from the DC‐T cell coculture system in the presence or absence of IL‐12p70 or IL‐6 in vitro (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05 and *p < 0.01. Data are presented as mean ± SEM. Schematics were created using BioRender (https://biorender.com).

Discussion

4

MyD88 signalling is crucial for innate immune responses to various stimuli [39]. For instance, MyD88 signalling in CD11c^+^ cells is essential for activating intestinal DCs and inducing colonic Th17/Th1 cell development during C. rodentium infection [40]. Additionally, surgical trauma and material implantation trigger necrocytosis, leading to high mobility group protein‐1 release, which activates DCs via the TLR4/MyD88/NF‐κB [41]. Moreover, MyD88^−/−^ CD4^+^ T cells show an intrinsic inability to secrete IL‐17A [42, 43]. Notably, MyD88^−/−^ mice exhibit complete resistance to EAE [13, 14, 15, 16]. While prior studies have established the involvement of MyD88 in initiating T cell responses [42], its impact on DC function and subsequent T cell priming in neuroinflammation remains inadequately understood. This study explored the role of MyD88 in regulating DC and T cell functions in the EAE model.

The immunostimulatory functions of DCs are closely linked to their maturation status. Tolerogenic DCs dampen T cell activity and suppress immune responses in autoimmune conditions, while mature DCs enhance immune functions in pathogenic scenarios. However, the signalling mechanisms that drive DC maturation require further investigation. Although previous research has shown that MyD88 expression on DCs does not hinder their maturation in autoimmune myocarditis [7], our current study reveals that MyD88 deficiency adversely affects the activation state of DCs and their capacity to stimulate CD4^+^ T cell differentiation. This discrepancy in DC functionality between the EAE and autoimmune myocarditis model may stem from varying types of stimuli, such as different antigens, immune environments, and tissue‐specific factors that influence TLR/MyD88 signalling pathways. Our single‐cell RNA sequencing analysis also indicated that MyD88^+^ DCs demonstrate enhanced interactions with both CD4^+^ and CD8^+^ T cells, highlighting their role in facilitating autoimmunity. Recent research demonstrated that bone marrow‐derived DCs from EAE mice expressed increased levels of MyD88 [44]. Similarly, we observed upregulation of MYD88 in DCs and CD4^+^ T cells within the PBMCs of MS patients, suggesting a potential role for MyD88 in the pathogenesis of the disease. In line with prior study [15], we also found diminished Il6 expression in MyD88^−/−^ DCs. Moreover, expression levels of Il12b and Tnfa were reduced in these cells. Importantly, we directly demonstrated that the lack of IL‐12 and IL‐6 in MyD88^−/−^ DCs leads to impaired differentiation of Th1 and Th17 cells.

While the roles of MyD88 in the innate immune system are well established, its potential functions in the adaptive immune system remain less understood. Increasing evidence highlights the important roles of MyD88 in T cells, yet its specific functions in this context are still debated. Early studies showed that T cells require MyD88 signalling to facilitate inflammation. For instance, MyD88 expression in T cells is essential for combating Toxoplasma gondii [45] and is necessary for T cell effector functions in the development of inflammatory bowel disease [46]. Furthermore, the MyD88‐dependent signalling pathway in CD4^+^ T cells promotes proliferation and strengthens humoral immune responses [47]. Conversely, more recent studies indicate that MyD88 may limit T cell activation. For example, MyD88 acts downstream of the TCR to restrict CD4^+^ T cell activation during cardiac adaptation to stress [48], and the deletion of MyD88 in T cells enhances anti‐tumour activity in melanoma [49]. While MyD88 has been shown to limit T cell activation and effector functions in some contexts, its role may differ based on specific signals, tissue environments, or disease states that can either promote or restrict T cell responses, potentially leading to the observed differences in activation and differentiation outcomes in MyD88^−/−^ CD4^+^ T cells. In our research, we found that MyD88^−/−^ CD4^+^ T cells exhibited diminished activation following TCR stimulation and demonstrated impaired ability to differentiate into Th1 and Th17 cells.

The pathogenesis of MS and EAE involves a diverse array of cell types and signalling pathways. However, the specific role of MyD88 in mediating these interactions across different cell types remains unclear. Notably, MyD88 signalling is critical not only for various immune cell types, including DCs [50, 51, 52], macrophages [53], T cells [42, 45, 48, 54], and B cells [6, 55], but also for CNS development, structure, and function. In neurons, MyD88 drives chemokine production to recruit protective leukocytes [56, 57], and its deficiency leads to structural and behavioural abnormalities [58]. Astrocytic MyD88 modulates inflammatory responses and synaptic function [59, 60], while microglial MyD88 controls polarisation [61, 62], phagocytosis [63], acute neuronal toxicity [64], neuroinflammation [65, 66], and remyelination [67]. These findings suggest that additional cell types, encompassing both immune and non‐immune cells, may also require MyD88 signalling to initiate the induction of EAE. Future studies should utilse conditional knockout models to validate the impact of MyD88 on EAE pathology across various cell populations.

Accumulating evidence positions MyD88 as a promising therapeutic target in several autoimmune diseases such as EAE [44, 68, 69, 70, 71], lupus [72], psoriasis [73, 74], and rheumatoid arthritis [75]. Therapeutic strategies aimed at inhibiting MyD88 signalling include peptide‐based inhibitors, small‐molecule compounds, RNA interference approaches, and immunomodulatory agents such as vitamin D_3_. Some of these inhibitors have advanced to clinical evaluation, notably the small‐molecule Enpatoran (M5049), which is currently undergoing phase II trials for systemic lupus erythematosus [76]. Mechanistically, these inhibitors can function by disrupting MyD88 dimerization, suppressing MyD88‐mediated signalling, or promoting the differentiation of tolerogenic DCs [44, 77]. Collectively, these findings underscore the translational potential of targeting the MyD88 pathway in the treatment of autoimmune and inflammatory disorders.

Conclusions

5

This study highlights the multifaceted roles of MyD88 in both CD4^+^ T cells and DCs during the pathogenesis of EAE. The resistance observed in MyD88^−/−^ mice is attributed to defects in immune cell priming and cytokine production, underscoring the necessity of MyD88 signalling in neuroinflammatory diseases. Future investigations should focus on exploring the therapeutic implications of modulating MyD88 pathways in the treatment of MS and related disorders.

Author Contributions

Wen Si: writing – original draft, methodology, formal analysis, conceptualization. Gaochen Zhu: methodology, visualisation, formal analysis. Qianling Jiang: methodology. Xin Ma: methodology. Guan Yang: writing – review and editing, supervision, resources, project administration, funding acquisition, conceptualization.

Funding

This work was supported by the National Natural Science Foundation of China, 82371350.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: MyD88 deletion in mice. (A) Genotyping PCR was conducted to identify WT (+/+), MyD88 heterozygous (+/−) and homozygous (−/−) knockout mice. (B) Immunoblotting was performed to evaluate the expression levels of MyD88 in the spleen of MyD88^+/+^ and MyD88^−/−^ mice. Figure S2: MyD88 is essential for the induction of EAE. (A) Flow cytometric analysis of lymphocyte and myeloid cell frequencies in the spinal cord (n = 4/group). (B) Representative histological sections of spinal cords from MyD88^+/+^ and MyD88^−/−^ mice stained for H&E and luxol fast blue and counterstained with cresyl echt violet solution at disease peak (day 16) (n = 5/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05 and **p < 0.01. Data are presented as mean ± SEM. Figure S3: The impact of MyD88 deficiency on T cell development in the thymus and peripheral lymphocyte profile. (A) Flow cytometric analysis was employed to examine CD4^+^ and CD8^+^ single‐positive thymocytes, as well as CD4^+^CD8^+^ double‐positive and CD4^−^CD8^−^ double‐negative thymocytes in MyD88^+/+^ and MyD88^−/−^ mice (n = 3–4/group). (B) Frequencies of T cell subsets in the spleen of MyD88^+/+^ and MyD88^−/−^ mice (n = 4/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. Data are presented as mean ± SEM. Figure S4: Peripheral T cell subtypes in the spleen of MyD88^+/+^ and MyD88^−/−^ mice. (A) Flow cytometric quantification of CD4^+^ IFN‐γ^+^ and CD4^+^ IL‐17A^+^ T cells in the spleen of naïve MyD88^+/+^ and MyD88^−/−^ mice (n = 4/group). (B) Flow cytometric quantification of CD4^+^ FoxP3^+^ Tregs in the spleen of EAE (day 10)‐induced MyD88^+/+^ and MyD88^−/−^ mice (n = 3–5/group). Statistical analysis was conducted using the two‐tailed unpaired Student's t‐test. *p < 0.05. Data are presented as mean ± SEM.