Myeloid ZNRF1 suppresses autoimmune demyelination and neuroinflammation by regulating MHC-II-mediated T cell activation

Abstract

See More

Multiple sclerosis (MS) is an autoimmune and neuroinflammation disease characterized by axonal damage, inflammatory demyelination, and neurodegeneration. However, the precise mechanisms underlying MS pathogenesis remain largely unclear. Here, we identify ZNRF1, an E3 ubiquitin ligase, as a critical regulator of experimental autoimmune encephalomyelitis (EAE), a murine model that recapitulates the autoimmune demyelination features of MS. Mice lacking ZNRF1 exhibit increased susceptibility to EAE progression. Notably, ZNRF1 depletion in peripheral myeloid cells, but not in microglia, leads to enhanced immune cell infiltration into the central nervous system, resulting in demyelination and exacerbated disease severity. The heightened EAE severity in Znrf1-deficient mice is associated with increased polarization of Th1 and Th17 cells, elevated antigen-specific T helper cell proliferation, and amplified immune responses. Furthermore, following EAE induction, macrophages from Znrf1-deficient mice display elevated surface expression of MHC class II (MHC-II) molecules. Collectively, our findings suggest that ZNRF1 in peripheral myeloid cells plays a suppressive role in neuroinflammation by regulating MHC-II surface expression, thereby controlling antigen-specific T-cell proliferation and activation.

Introduction

Multiple sclerosis (MS) is a chronic autoimmune disorder of the central nervous system (CNS) that primarily targets myelinated axons [12, 13, 23, 57]. The immune system erroneously attacks the myelin sheath surrounding nerve fibers, leading to inflammation, demyelination, and the formation of lesions (plaques), the pathological hallmark of MS, with in gray and white matter of the brain, spinal cord, and optic nerve [26, 66]. Although the etiology of MS remains poorly understood, both genetic and environmental factors are believed to contribute to its development [2, 23, 35, 63]. The pathophysiology of MS involves intricate interactions between the innate and adaptive immune systems, with peripheral immune cells such as B cells, T cells, and myeloid cells (e.g., dendritic cells (DCs) and macrophages), as well as CNS-resident cells including microglia and astrocytes, playing pivotal roles [1, 5, 7, 19, 23, 33, 38, 41, 55, 62, 73, 75]. However, despite extensive research, the precise mechanisms underlying MS pathogenesis remain elusive.

CD4⁺ T cells are central to the pathogenesis of MS [5, 19, 30, 45, 53]. Autoreactive CD4⁺ T cells activate B cells and differentiate into various subsets of T helper (Th) cells, which orchestrate immune responses. Naive T cell activation requires antigen presentation by antigen-presenting cells (APCs), such as B cells, macrophages, and DCs [61]. In MS patients, elevated numbers of interferon (IFN)-γ-producing CD4⁺ T (Th1) cells, interleukin-17 (IL-17)-producing CD4⁺ T (Th17) cells, and cytotoxic CD8⁺ T cells are found in both peripheral blood and the CNS, contributing to oligodendrocyte and neuronal damage [52, 53]. Activated CD4⁺ T cells proliferate, differentiate into specific Th-cell subsets, and generate memory T cells [19, 69], which are categorized into central memory T (T_CM) cells and effector memory T (T_EM) cells [6, 69]. In MS patients, autoreactive CD4⁺ T_EM cells are enriched, whereas T_CM cells are reduced in peripheral blood compared to healthy individuals [44], highlighting the importance of T-cell subset balance in MS pathophysiology.

APCs critically shape T-cell responses in MS. Monocytes differentiate into macrophages that express major histocompatibility complex class II (MHC-II) and costimulatory molecules such as CD80 and CD86 and secrete proinflammatory cytokines that drive CNS demyelination [1, 29, 41, 87, 89]. Additionally, monocyte infiltration into lesions correlates with disease severity during early MS [1, 41, 80]. DCs, as professional APCs, initiate both innate and adaptive immune responses [25, 62, 75]. They are broadly categorized into conventional DCs (cDCs), which express high levels of MHC-II to present antigen, and plasmacytoid DCs (pDCs), which primarily produce robust type I interferons in response to viral infections [5, 25]. Notably, cDCs are elevated in the blood of patients with secondary progressive MS (SPMS) compared to relapsing–remitting MS (RRMS) and healthy controls [5, 38]. APCs fine-tune T-cell activation through stimulatory and inhibitory pathways: MHC-II promotes CD4⁺ T-cells activation, whereas inhibitory molecules such as programmed cell death protein-1 (PD-1) and its ligand programmed cell death ligand-1 (PD-L1), as well as Fas and its ligand FasL regulate T-cell apoptosis and tolerance. Dysregulation of these pathways are associated with MS pathogenesis [9, 20, 24, 28, 42, 43, 48, 64, 82, 84].

Microglia, the resident macrophages of the CNS, also play dual roles in MS pathology [87, 92]. During early disease stages, activated microglia contribute to the formation of demyelinating lesions by secreting proinflammatory cytokines and chemokines, which recruit T cells into the CNS, thereby accelerating MS pathophysiology [36, 59, 76, 92]. Conversely, at later stages, microglia mediate neuroprotection through clearance of myelin debris, secretion of anti-inflammatory cytokines and neuroprotective factors, and enhancement of antioxidant responses that promote remyelination [9, 46, 79]. These context-dependent roles highlight the functional plasticity of microglia in MS progression.

ZNRF1, an E3 ubiquitin ligases, contains zinc and ring finger domains essential for its E3 ligase activity [3, 4]. ZNRF1 also possesses an N-myristoylated domain and is localized to the endosome–lysosome compartments [4]. Under oxidative stress conditions, such as neuronal apoptosis and axonal degeneration, ZNRF1 activity is enhanced by epidermal growth factor receptor (EGFR)-mediated phosphorylation at tyrosine 103, leading to ubiquitination and degradation of protein kinase B (AKT) and promoting Wallerian degeneration [85, 86]. Beyond neuronal function, ZNRF1 plays significant roles in innate immune regulation. Our recent work shows that ZNRF1 ubiquitinates and degrades caveolin-1 through AKT-GSK3β signaling, thereby promoting Toll-like receptor 4 (TLR4)-mediated inflammatory responses essential for host defense against gram-negative bacteria [47]. Additionally, we have demonstrated that c-Src-activated ZNRF1 negatively regulates TLR3 signaling by facilitating lysosomal degradation of TLR3 [50]. ZNRF1 also targets EGFR for degradation, modulating its downstream signaling [74]. These findings highlight the multifaceted role of ZNRF1 in both neuronal and immune system regulation. Given this background, we hypothesize that ZNRF1 may influence MS pathogenesis through its effects on immune cell function and neuroinflammation. In the present study, we investigate the role of ZNRF1 in various immune cells involved in EAE. Our findings suggest that ZNRF1 in peripheral myeloid cells, but not microglia, controls the surface expression of MHC-II, thereby modulating antigen-specific T-cell proliferation and activation. These results underscore a previously unrecognized immunosuppressive role for ZNRF1 in myeloid cells during neuroinflammatory responses.

Materials and methods

Mice

Znrf1^F/F and Znrf1^−/− mice were previously described [47, 50]. To generate a myeloid cell lineage with Znrf1 deletion (referred to as Znrf1^Δmye), Znrf1^F/F mice were crossed with LysM-Cre mice (004781; The Jackson Laboratory). To create a CNS tissue-resident microglial lineage with Znrf1 deletion (Znrf1^Δglia), Znrf1^F/F mice were crossed with Cx3cr1-CreERT2-EYFP mice (021160; The Jackson Laboratory). Microglial Znrf1 gene deletion was induced by intraperitoneally injecting the mice with tamoxifen (0.135 mg/g body weight) for 5 consecutive days. A subsequent waiting period of 4–6 weeks was allowed to ensure replenishment of peripheral wild-type macrophages. All animals were housed in a pathogen-free animal facility, and all animal experiments were performed in accordance with animal welfare guidelines, approved by the Institutional Animal Care and Use Committee of the College of Medicine, National Taiwan University (approval number: 20180450 and 20230070).

Antibodies and reagents

CD4-BV711 (#563726; clone: RM4-5; dilution, 1:1000), IL-17A-BV421 (#563354; clone, TC11-18H10; dilution, 1:500), H-2 Kb/H-2Db-BV480 (#746258; clone, C3H.SW; dilution, 1:400), RORγt-PE (#562607; clone, Q31-378; dilution, 1:100) and Brilliant Stain Buffer Plus (#566385) were from BD Pharmingen (Basel, Switzerland). CD80-PE-Cy7 (#60–0801; clone:16-10A1; dilution, 1:400) and Ly-6G-violetFluor 450 (#75–1276; clone, 1A8; dilution, 1:200) were obtained from Cytek Biosciences (San Diego, CA). CD11b-PE-Cy7 (#101216; clone, M1/70; dilution, 1:100), CD11b-BV510 (#101263; clone, M1/70; dilution, 1:800), CD11c-APC (#117310; clone, N418; dilution, 1:1000), CD11c-PE (#117308; clone, N418; dilution, 1:1000), CD172a-PE-Dazzle 594 (#144016; clone, P84; dilution, 1:200), CD178 (FasL)-PE (#106605; clone, MFL3; dilution, 1:100), CD19-PE (#152407; clone, 1D3; dilution, 1:1000), CD25-BV421 (#102043; clone, PC61; dilution, 1:500), CD274 (B7-H1, PD-L1)-BV421 (#124315; clone, 10F.9G2; dilution, 1:200), CD3-APC (#100322; clone, 145-2C11; dilution, 1:200), CD3-PE (#100205; clone, 17A2; dilution, 1:1000), CD44-APC (#103012; clone, IM7; dilution, 1:1000), CD45-FITC (#103108; clone, 30-F11; dilution, 1:1000), CD40-APC/Cyanine 7 (#124638; clone, 3/23; dilution, 1:100), CD45-BV421 (#103147; clone, 30-F11; dilution, 1:1000), CD45-PerCP-Cy5.5 (#103132; clone, 30-F11; dilution, 1:800), CD45R/B220-FITC (#103206; clone, RA3-6B2; dilution, 1:500), CD45R/B220-Alexa Fluor 700 (#103232; clone, RA3-6B2; dilution, 1:100), CD62L-BV605 (#104438; clone, MEL-14; dilution, 1:1000), CD69-PE (#104507; clone, H1.2F3; dilution, 1:500), CD8a-PE (#100708; clone, 53–6.7; dilution, 1:1000), CD86-PE/Cy 5 (#105015; clone, GL-1; dilution, 1:400), Gr1 (Ly-6G/Ly-6C)-PE (#108407; clone, RB6-8C5; dilution, 1:1000), F4/80-BV711 (#123147; clone, BM8; dilution, 1:100), I-A/I-E (MHC-II)-BV605 (#107639; clone, M5/114.15.2; dilution, 1:400), IFN-γ-APC (#505810; clone, XMG1.2; dilution, 1:500), Ly-6C-FITC (#128006; clone, HK1.4; dilution, 1:400), T-bet-BV605 (#644817; clone, 4B10, dilution, 1:100) and XCR1-BV785 (#148225; clone, ZET; dilution, 1:100) were purchased from Biolegend (San Diego, CA).

Establishment of the Experimental Autoimmune Encephalomyelitis (EAE) model

To establish a mouse model of EAE, mice aged 8–12 weeks were immunized with 200 µg of MOG_35–55 (sequence: MEVGWYRSPFSRVVHLYRNGK; #SC1208; GeneScript, Piscataway, NJ) in 100 µL of 1X Dulbecco’s PBS (DPBS; Gibco, Waltham, MA), mixed with 100 µL of Freund’s complete adjuvant (CFA; #263810; BD/DIFICO, Franklin Lakes, NJ), containing 4 mg of Mycobacterium tuberculosis H37Ra (#231141, BD/DIFCO). Mice received two subcutaneous injections of 100 µL of MOG_35-55/CFA emulsion on day 0. Additionally, they were intraperitoneally injected with 200 ng of pertussis toxin (#180, List Biological Laboratories, Campbell, CA) in 500 µL of 1X DPBS on days 0, 2, and 7. The mice were monitored for weight changes and clinical symptoms, which were scored using the following scale: grade 0, normal with no observable neurological defect; grade 0.5, partial paralysis of the tail; grade 1, complete paralysis of the tail; grade 2, impaired righting reflex; grade 2.5, impaired movement coordination; grade 3, abnormal gait; grade 3.5, partial paresis of the hindlimbs; grade 4, paresis affecting both hindlimbs; grade 4.5, able to paddle with both hindlimbs, but neither hindlimb able to move forward past the hip joint; grade 5, paralysis of one hindlimb; grade 5.5, complete paralysis of both hindlimbs; grade 6, complete paralysis of both hindlimbs with labored breathing; grade 6.5, paralysis of both hindlimbs with partial paralysis of forelimbs or moribund state; grade 7, death. Intermediate values (in increments of 0.25) were used to reflect gradual symptom progression between these defined points.

Isolation of CNS-infiltrating cells

Mice were perfused with 40 mL of ice-cold 1X DPBS to remove blood from circulation, followed by the collection of brain and spinal cord tissues. The collected tissues were minced and digested enzymatically for 45 min at 37 °C with rotation at 250 RPM in a shaker using a solution of 0.1 mg/mL collagenase/dispase (#10269638001, Roche, Basel, Switzerland) and 0.75 mg/mL DNase I (#10104159001; Roche) in Roswell Park Memorial Institute (RPMI) medium (Gibco) containing 2% fetal bovine serum (FBS; Corning, Woodland, CA), 1% non-essential amino acids (NEAA; Gibco), 1% L-glutamine (Sartorius, Beit Haemek, Israel), 1% penicillin–streptomycin (Sartorius), and 1% sodium pyruvate (Sartorius). The resulting cell suspension was sieved through a 40-µm cell strainer to remove debris and stopped reactions with 10% FBS complete RPMI medium. After centrifugation at 800 × g for 5 min at room temperature, the cell pellet was resuspended in 7 mL of 30% Percoll (#17089101; GE Healthcare, Little Chalfont, UK) layered on top of 4 mL of 70% Percoll. The gradient was centrifuged at 800 × g for 20 min at room temperature without applying the brake. The cells were collected from the interphase between the 30% and 70% Percoll layers, washed with DPBS, and resuspended in complete RPMI medium on ice for subsequent flow cytometric analysis.

Immunohistochemistry and Luxol Fast Blue staining

Mice were perfused with 40 mL of ice-cold 1X DPBS, followed by 40 mL of ice-cold 4% paraformaldehyde (PFA; Alfa Aesar, Lancashire, UK). Spinal cord tissues were collected and fixed in 2% PFA overnight at 4 °C. Subsequently, the tissues were embedded in paraffin for sectioning. The tissue sections were stained using hematoxylin and eosin (H&E) for general histological assessment and Luxol Fast Blue to evaluate myelin integrity.

Serial 4 µm paraffin-embedded tissue sections were deparaffinized using EZ Prep solution (Ventana Medical Systems, Inc., Tucson, AZ). The slides were incubated with anti-CD3 antibody (#790–4341; clone, 2GV6; Ventana Medical Systems, Tucson, Arizona), anti-CD68 antibody (#ab125212; polyclonal; Abcam, Cambridge, UK), anti-IFNγ antibody (#BS-0480R; polyclonal, Bioss, Boston, USA), and anti-IL-17A antibody (#BS-1183R; polyclonal, Bioss, Boston, USA) at a 1:100 dilution for 120 min using the automated Ventana Benchmark XT system (Ventana Medical Systems). Immunolabeling was visualized using Ultraview DAB Detection Kit (Ventana Medical Systems) according to the manufacturer’s protocol. All sections were counterstained with hematoxylin provided by Ventana reagents. The tissues were imaged using TissueFAXS microscopy (TissueGnostics, Vienna, Austria). Image projection and analysis were conducted with TissueFAXS software (TissueGnostics, Vienna, Austria), Fuji Image J (National Institutes of Health, USA), and QuPath version 0.6.0 [8].

Isolation of splenocytes and lymph node cells

To prepare a single-cell suspension from spleen tissues, the spleen was pressed through a 40-µm strainer (Falcon, Glendale, AZ) using the plunger of a 5-mL syringe. The resulting cell suspension was centrifuged at 800 × g for 3 min at room temperature. Red blood cells were lysed by incubating the suspension in 10 mL of 1X red blood cell (RBC) lysis buffer (0.15 M ammonium chloride, 10 mM potassium bicarbonate, and 0.1 mM EDTA disodium salt dihydrate; pH 7.2) for 5 min at room temperature. Following this, the suspension was centrifuged at 400 × g for 3 min at room temperature. To remove debris, the cells were resuspended in 10 mL of 1X DPBS and gently pipetted up and down using a 10-ml serological pipette (SPL Life Sciences, Korea). After centrifugation at 400 × g for 3 min at room temperature, dead cells were removed using the Dead Cell Removal Kit (#130–090-101; MACS, Bergisch Gladbach, Germany) in conjunction with MS separation columns (#130–042-201; MACS).

For lymph node cell isolation, draining lymph nodes were collected and pooled from the brachial, axillary, and inguinal lymph regions. The lymph nodes were then mechanically dissociated by pressing them through a 70-µm strainer (Falcon) using the plunger of a 3-mL syringe. The resulting cell suspension was centrifuged at 400 × g for 3 min at room temperature. The pellet was then incubated in 5 mL of 1X RBC lysis buffer for 5 min at room temperature to lyse red blood cells. After a second centrifugation at 400 × g for 3 min, the pellet was resuspended in 5 mL of 1X DPBS and gently pipetted up and down with a 5-ml serological pipette (SPL Life Sciences). Following a final centrifugation at 400 × g for 3 min at room temperature, cells were resuspended in 5 mL of complete RPMI medium and stored on ice until further use.

Flow cytometry: staining and analysis

Single-cell suspensions (2 × 10⁶) were washed twice with 1X DPBS and stained with either Fixable Viability Stain 780 (#565388; BD Biosciences; dilution 1:1000) or Zombie NIR (#423106; Biolegend; dilution 1:1000) for 10 min on ice. After centrifugation at 400 × g for 3 min, the cells were washed twice with flow cytometry buffer (1X DPBS containing 2% FBS and 0.05% sodium azide). Fc receptors were blocked by incubating the cells with purified anti-mouse CD16/32 antibodies (FcγRIII; #101302; Biolegend; dilution 1:100) for 5 min on ice. The cells were then stained with fluorophore-conjugated antibodies for 10 min on ice. Following two additional washes with flow cytometry buffer, the cells were fixed with 2% PFA at 4 °C for 20 min. After fixation, the cells were washed twice with 1X DPBS and stored in 1X DPBS containing 2 mM EDTA at 4 °C. For MOG_35–55 TCR-independent intracellular cytokine staining, splenocytes and lymph node cells (2 × 10⁶) were incubated with 50 ng/mL phorbol-12-myristate-13-acetate (#P8139; Sigma-Aldrich, St. Louis, MO), 500 ng/mL ionomycin (#I0634; Sigma-Aldrich), and 5 µg/mL brefeldin A (BFA, Biolegend) for 4 h at 37 °C. After surface staining, the cells were fixed and permeabilized using the Cytofix/Cytoperm Fixation/Permeabilization Solution Kit (#554714; BD Biosciences) and incubated with intracellular antibodies overnight at 4 °C. The cells were then washed with 1 × Cytoperm solution and stored at 4 °C in 1X DPBS containing 2 mM EDTA. Flow cytometry data were acquired using Cytek Aurora, Cytek Northern Lights, BD LSR II and BD FACSLyric flow cytometry systems. Data analysis was performed using FlowJo software (BD Biosciences).

Proliferation Assay

CD4⁺ T cell from 2D2 TCR mice (006912; The Jackson Laboratory) were purified from splenocytes (10 × 10⁶) and labeled with 5 µM carboxyfluorescein succinimidyl ester (CFSE; #65–0850-84, eBioscience, San Diego, CA) in 1X DPBS at room temperature for 10 min in the dark. The reaction was quenched by adding 5 mL of ice-cold complete RPMI medium, followed by a 5-min incubation at 4 °C. Afterward, the cells were centrifuged at 400 × g for 5 min and immediately washed twice with 5 mL of ice-cold complete RPMI. CD11b⁺ cells were purified from the splenocytes of mice post-EAE induction. CFSE labeled CD4⁺ T cells from 2D2 TCR were co-cultured with CD11b⁺ cells in 48-well plates for 3 days. After the co-culture period, the cells were stained with surface markers, fixed, and stored at 4 °C in 1X DPBS containing 2 mM EDTA, followed by flow cytometric analysis.

Spinal cord tissue labeling and imaging

Spinal cord tissue labeling and imaging were conducted as described previously [14]. Briefly, mouse blood vessels were labeled by vessel painting via cardiac perfusion with Wheat Germ Agglutinin-Alexa Fluor 488 conjugate (30 mg/g body weight; #W11261, Invitrogen, Carlsbad, CA), followed by 4% PFA perfusion fixation. Afterward, the spinal cord was harvested, and vibratome Sects. (350 mm) of the tissue were post-fixed in 4% PFA at 15 °C for 30 min [51, 70]. The spinal cord was washed with 1X DPBS and storage in 0.1% PFA at 4 °C. The fixed tissues were then immersed in 2% Triton X-100 (Sigma-Aldrich) solution overnight at 15 °C for permeabilization. The spinal cord was then rinsed in 1X DPBS, followed by blocking with a buffer containing 2% Triton X-100, 10% normal goat serum, and 0.02% sodium azide in 1X DPBS for 2 h at 15 °C. Two primary antibodies were used to immuno-label the spinal cord: rabbit anti-lymphatic endothelial hyaluronan receptor 1 (Lyve-1; lymphatic endothelial marker; #ab14917, Abcam) and Alexa Fluor 647-conjugated anti-mouse CD3 antibody (#100209; clone, 17A2; Biolegend). The spinal cord tissues were incubated with the primary antibody, diluted in the dilution buffer (1:100, 0.25% Triton X-100, 1% normal goat serum, and 0.02% sodium azide in 1XDPBS) for 2 days at 15 °C to label CD3⁺ T cells and cellular structures in the spinal cord. Alexa Fluor 546-conjugated goat anti-rabbit IgG secondary antibody (#A11010; Thermo, Waltham, MA, USA; dilution, 1:200) was used to visualize the immunostained structures. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; #D9542; Sigma-Aldrich). The labeled spinal cord tissues were then immersed overnight in the optical-clearing solution with a high refractive index (RapiClear 1.52 solution, SunJin Lab, Hsinchu, Taiwan), followed by an additional day of immersion in fresh RapiClear solution. The tissues were imaged using transmitted light and confocal microscopy (LSM800; Carl Zeiss, Jena, Germany). Image projection and analysis were conducted with LSM Image Browser (Carl Zeiss) and ZEN 3.2 software (Carl Zeiss).

Immunoblotting

Murine microglial cells were purified using Percoll gradient as described previously and washed twice with 1X DPBS. The cells were resuspended in ice-cold radioimmunoprecipitation assay buffer (RIPA) buffer (50 mM Tris–HCl, 150 mM NaCl, 1% SDS, 1% sodium deoxycholate, and 1% Triton X-100; pH 7.5) containing protease inhibitors (2 µg/ml Aprotinin, 1 mg/ml Benzamidine, 1 µg/ml Pepstatin A, and 2 µg/ml Leupeptins; Sigma-Aldrich) and were sonicated with three 30 s bursts, separated by 1-min intervals, followed by incubation on ice for 30 min. After centrifugation at 14,000 × g for 10 min at 4 °C, the supernatant was transferred into a new eppendorf tube. SDS protein sample buffer (5X; 0.25 M Tris–HCl, 4% SDS, 16% 2-mercaptoethanol, 30% glycerol, and 0.002% bromophenol blue, pH 6.8) was added, and the mixture was incubated at 98 °C for 15 min. Protein samples were stored at −20 °C until further use in immunoblotting. Protein samples were separated by 12% SDS-PAGE in 1X running buffer (25 mM Tris–HCl, 190 mM glycine, and 0.1% SDS, pH 8.3) at 40–80 V. The separated proteins were transferred onto a polyvinylidene fluoride (PVDF) membrane (Merck Millipore, Billerica, MA) using 1X transfer buffer (25 mM Tris–HCl, 190 mM glycine, and 20% Methanol, pH 8.3) at 100 V for 2.5 h. The membrane was blocked with 10% non-fat milk in 1X TBST (20 mM Tris–HCl, 150 mM NaCl, and 0.1% Tween-20, pH 7.5) for 1 h and washed with 1X TBST four times for 15 min each. The membranes were incubated overnight at 4 °C with primary antibodies. After four additional washes with 1X TBST (15 min each), the membranes were incubated with HRP-conjugated secondary antibodies diluted in 10% milk for 2 h at room temperature. Following four final washes with 1X TBST (15 min each), signals were detected using an enhanced chemiluminescence (ECL) Western blotting substrate.

Statistical analysis

GraphPad Prism 8.0 software was used for data analysis. Results are presented as mean ± SD. Statistical significance was determined using the Wilcoxon matched-pairs signed-rank test for EAE clinical scores and body weight changes. For two-group comparisons, statistical significance was assessed using the unpaired, two-tailed Student’s t-test. P values < 0.05 were considered statistically significant.

Results

Systemic depletion of ZNRF1 in mice promotes EAE progression and neuroinflammation

ZNRF1, originally identified as a regulator during nerve injury, is known to participate in oxidative stress-induced neuronal death and axonal degeneration [3, 85, 86]. However, its role in neuroinflammation remains unclear. To address this, we explored the involvement of ZNRF1 in the pathogenesis of experimental autoimmune encephalomyelitis (EAE), a widely used mouse model that recapitulates certain aspects of the inflammatory cascade in human MS [18]. Wild-type (Znrf1^+/+) and Znrf1^−/− mice were immunized with MOG_35-55 to induce EAE. Compared to wild-type controls, Znrf1^−/− mice exhibited significantly higher clinical scores after EAE induction (Fig. 1a). Additionally, the body weights in Znrf1^−/− mice were changed, inversely correlating with enhanced clinical severity of EAE (Fig. 1b). This was corroborated by flow cytometric analysis, which revealed significantly increased numbers of infiltrating leucocytes (CD45⁺), CD4⁺ T cells (CD45⁺CD4⁺), cytotoxic T cells (CD45⁺CD8⁺), and myeloid cells (CD45⁺CD11b⁺) in the spinal cords of Znrf1^−/− mice at both day 18 (Fig. 1c) and day 30 (Fig. S3b) following EAE induction. At the peak (day 18) and late stages (day 30) of the disease, spinal cords were harvested and analyzed using hematoxylin and eosin (H&E) and Luxol fast blue (LFB) staining. H&E staining showed significantly greater immune cell infiltration in the spinal cords of Znrf1^−/− mice (Fig. 1d and Fig. S3c). Consistently, LFB staining also revealed increased demyelination in the spinal cord of Znrf1^−/− mice compared to Znrf1^+/+ controls (Fig. 1d and Fig. S3c). Immunohistochemistry (IHC) analysis further confirmed elevated levels of infiltrating CD3⁺ T cells and macrophages (CD68⁺) in the spinal cords of Znrf1^−/− mice at both day 18 and day 30 post-EAE induction (Fig. 1e and Fig. S3c). In addition, IHC revealed increased infiltration of Th17 cells (IL-17A⁺) and a trend toward higher numbers of Th1 cells (IFNγ⁺) in the spinal cords of Znrf1^−/− mice at the same time points (Fig. 1f and Fig. S3c). Together, these data demonstrate that systemic ZNRF1 deficiency exacerbates EAE progression and neuroinflammation.

Fig. 1

Systemic deletion of ZNRF1 promotes EAE pathogenesis and neuroinflammation in mice. Age-matched female Znrf1^+/+ and Znrf1^−/− mice were immunized with 200 μg of MOG_35-55 peptide emulsified in Complete Freund’s Adjuvant (CFA), followed by intraperitoneal (i.p.) injection of 200 ng pertussis toxin (PTX) to induce EAE. a Clinical scores of EAE progression were monitored and recorded daily (N = 40). b Body weights of immunized mice were measured and normalized to their respective weights at day 8 post-immunization (N = 40). c At day 18 post-EAE induction, spinal cords from female Znrf1^+/+ and Znrf1^−/− mice were harvested to prepare single-cell suspensions. Cells were stained with antibodies against CD45, CD4, CD8, and CD11b, followed by flow cytometry analysis to quantify leucocytes (CD45⁺), T helper (Th) cells (CD45⁺CD4⁺), cytotoxic T cells (CD45⁺CD8⁺), and myeloid cells (CD45⁺CD11b⁺) (N = 6). Representative flow plots and quantified cell numbers of infiltrating immune cells in the CNS are presented. d-f Spinal cord tissue sections from Znrf1^+/+ and Znrf1^−/− mice were collected and analyzed at day 18 post-EAE induction. d Tissue sections were stained with hematoxylin and eosin (H&E) to assess inflammatory cell infiltration and with luxol fast blue (LFB) to evaluate white matter demyelination. e and f Tissue sections were subjected to immunohistochemistry (IHC) using antibodies against CD3 (T cells) and CD68 (macrophages) (e) or IFNγ (Th1 cells) and IL-17A (Th17 cells) (f). d-f Scale bars: 200 μm (whole spinal cord sections) and 100 μm (enlarged region #1 and #2). g Flow cytometric analysis of naïve CD4⁺ T cells (CD4⁺CD44⁻CD62L⁺), central memory CD4⁺ T (T_CM) cells (CD4⁺CD44⁺CD62L⁺), and effector memory CD4⁺ T (T_EM) cells (CD4⁺CD44⁺CD62L⁻) in the spleens and draining lymph nodes (DLNs) of Znrf1^+/+ (N = 4) and Znrf1^−/− mice (N = 4) at day 18 post-EAE induction. h Flow cytometric analysis of naïve CD8⁺ T cells (CD8⁺CD44⁻CD62L⁺), CD8⁺ T (T_CM) cells (CD8⁺CD44⁺CD62L⁺), and effector memory CD8⁺ T (T_EM) cells (CD8⁺CD44⁺CD62L⁻) in the spleens and DLNs of Znrf1^+/+ (n = 4) and Znrf1^−/− mice (N = 4) at day 18 post-EAE induction. i-k Flow cytometric analysis of Th1 (CD4⁺IFNγ⁺) cells (i), Th17 (CD4⁺IL-17A⁺) cells (j), and cytotoxic T (CD8⁺IFNγ⁺) cells (k) in the spleens and DLNs of Znrf1^+/+ (n = 3) and Znrf1^−/− mice (N = 3) at day 18 post-EAE induction. (a and b) Data are presented as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, determined by the Wilcoxon matched-pairs signed-rank test. d-f Representative spinal cord sections and quantification of infiltrating immune cells or demyelination in Znrf1^+/+ and Znrf1^−/− mice are shown. g-k Representative flow cytometry plots and quantified T cell populations in the spleens and DLNs. c-k Data are presented as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, determined by the unpaired Student’s t-test

ZNRF1 deficiency enhances T cell activation and differentiation in peripheral lymphoid organs

Given the central role of CD4⁺ T cells in MS pathogenesis [5, 16, 45], we evaluated whether ZNRF1 influences T cell development and activation by analyzing T cell populations in the spleen and thymus of Znrf1^+/+ and Znrf1^−/− mice. T cell development appeared unaffected by ZNRF1 deficiency, as no differences were observed in the frequencies of T cell subsets in either the spleen or thymus (Fig. S1a -S1f). Upon TCR stimulation, Znrf1^−/− T cells exhibited comparable proliferation to Znrf1^+/+ T cells (Fig. S2a), and no differences were observed in IL-2 or IFNγ production between the two genotypes (Fig. S2b). Furthermore, activation markers, including CD25 and CD69, on CD4⁺ T cells were similar between Znrf1^+/+ and Znrf1^−/− mice (Fig. S2d-e), and the frequency of Th1 cells (CD4⁺IFNγ⁺) remained unchanged (Fig. S2c). Similarly, central memory T cell markers CD44 and CD62L on CD4⁺ T cells did not differ between Znrf1^+/+ and Znrf1^−/− mice (Fig. S2f). We also assessed CD4⁺ T cell polarization toward Th1 and Th17 lineages and found that ZNRF1 did not alter either pathway (Fig. S2g and S2h). These findings suggest that ZNRF1 is dispensable for T cell development, activation, and Th1/17 polarization under both homeostatic and stimulatory conditions.

We then investigated how ZNRF1 suppresses EAE induction by analyzing T cell activation in peripheral lymphoid organs, including the spleen and draining lymph nodes (DLNs), at the peak and chronic stages of the disease. Compared to Znrf1^+/+ mice, naïve CD4⁺ T (CD44⁻CD62L⁺) cells were decreased, while effectors memory CD4⁺ T (CD44⁺CD62L⁻) cells in both spleens and DLNs of Znrf1^−/− mice were significantly increased at day 18 after EAE induction (Fig. 1g). Similarly, at day 30 post-immunization, naïve CD4⁺ T cells were decreased, whereas central memory CD4⁺ T (T_CM) cells and effectors memory CD4⁺ T (T_EM) cells were also elevated in peripheral lymphoid organs of Znrf1^−/− mice (Fig. S3e). In addition, naïve CD8⁺ T (CD8⁺CD44⁻CD62L⁺) cells in the spleen of Znrf1^−/− mice, compared to Znrf1^+/+ mice, were significantly decreased at both day 18 (Fig. 1h) and day 30 (Fig. S3f) post-EAE induction. Conversely, CD8⁺ T_CM cells (CD8⁺CD44⁺CD62L⁺) and CD8⁺ T_EM cells (CD8⁺CD44⁺CD62L⁻) were markedly increased in the DLNs of Znrf1^−/− mice at both time points after EAE induction (Fig. 1h and S3f). These data suggest that ZNRF1 deficiency enhances T cell activation and differentiation in peripheral lymphoid organs. We further examined the polarization of effector T cell subsets-Th1 (CD4⁺IFNγ⁺), Th17 (CD4⁺IL-17A⁺), and cytotoxic CD8⁺ T (CD8⁺IFNγ⁺) cells in Znrf1^+/+ and Znrf1^−/− mice in both spleens and DLNs during EAE progression. At baseline, the frequencies of Th1, Th17, and cytotoxic CD8⁺ T cells were comparable between naïve Znrf1^+/+ and Znrf1^−/− mice (Fig. S2i-S2k). Following EAE induction, Th1 cells were significantly increased in both spleens and DLNs of Znrf1^−/− mice compared to Znrf1^+/+ controls at both day 18 and day 30 (Fig. 1i and S3h). Th17 cells were significantly increased in DLNs of Znrf1^−/− mice at day 30, and slightly elevated in Znrf1^−/− spleen at day 18 after immunization (Fig. 1j and S3i). Furthermore, the cytotoxic CD8⁺ T cell population was significantly increased only in the spleens of Znrf1^−/− mice at day 30 after EAE (Fig. 1k and S3j). These data suggest that ZNRF1 deficiency promotes effector T cell activation, differentiation, and polarization during EAE.

ZNRF1 deficiency in microglia does not affect EAE pathogenesis

Besides T cells, myeloid cells, including macrophages and DCs, as well as microglia (the resident macrophages in the CNS), contribute to the pathogenesis of EAE [21]. To investigate whether microglial ZNRF1 influences EAE, we crossed Znrf1^F/F mice with Cx3cr1-CreERT2 transgenic mice [27, 90] to generate Znrf1^F/F: Cx3cr1-CreERT2 mice (Fig. 2a). Microglial Znrf1 was deleted by intraperitoneal tamoxifen injection, followed by a 30-day waiting period (referred to hereafter as Znrf1^Δglia) (Fig. 2b). Given that the half-life of myeloid-derived macrophages is approximately four weeks, while that of microglia exceed six months [27, 72, 90], myeloid-derived macrophages were replaced by newly undeleted macrophages derived from bone marrow progenitors, whereas the long-lived Znrf1-deleted microglia persisted. To confirm the efficiency of Znrf1 deletion in microglia, we assessed ZNRF1 protein levels in microglia isolated from wild-type and Znrf1^Δglia mice by immunoblotting, and observed a significant reduction in ZNRF1 protein levels in Znrf1-deleted microglia (Fig. 2c). By contrast, ZNRF1 protein levels in bone marrow-derived macrophages (BMDMs) and F4/80⁺ splenic macrophages were comparable between wild-type and Znrf1^Δglia mice (Fig. 2c). We then induced EAE in both control and Znrf1^Δglia mice. Upon EAE induction, Znrf1^F/F and Znrf1^Δglia mice exhibited similar clinical scores and body weight changes (Fig. 2d and e). Consistent with these observations, H&E staining revealed comparable immune cell infiltration in the spinal cords of Znrf1^F/F and Znrf1^Δglia mice at both day 18 and day 30 after EAE induction (Fig. 2f and h). LFB staining also showed no exacerbated demyelination in Znrf1^Δglia spinal cords (Fig. 2f and h). IHC analysis further demonstrated similar infiltration of CD3⁺ T cells and CD68⁺ macrophages in the spinal cords of both groups at both time points (Fig. 2g and i). Flow cytometric analysis confirmed that the numbers of infiltrating leucocytes, CD4⁺ T helper (Th) cells, cytotoxic T cells, and myeloid cells were comparable in the spinal cords of both groups (Fig. 2j and k). Together, these data indicate that microglia ZNRF1 does not significantly contribute to the pathogenesis of EAE.

Fig. 2

ZNRF1 deficiency in microglia does not affect EAE pathogenesis and neuroinflammation in mice. a Schematic diagram illustrating the tamoxifen-inducible Cx3cr1-Cre (Cx3cr1-CreERT) allele, the Znrf1^Flox allele, and the deleted Znrf1 allele (Znrf1^Δglia) after Cre-mediated recombination driven by tamoxifen (TAM) administration. ERT, estrogen receptor; IRES, internal ribosome entry site; YFP, yellow fluorescent protein. b Schematic diagram depicting the generation of Znrf1^Δglia mice. 5-week-old Znrf1^Flox/Flox:Cx3cr1-CreERT mice were i.p. injected with TAM (0.135 mg/g body weight) daily for five days, followed by a 4-week waiting period to allow the replacement of myeloid cells. c Microglia, BMDMs, and splenic F4/80⁺ cells were isolated from Znrf1^F/F and Znrf1^Δglia mice at 4–5 weeks post-TAM administration and subjected to immunoblotting analysis for ZNRF1 expression. (d-i) Age-matched female Znrf1^F/F and Znrf1^Δglia mice were immunized with 200 μg of MOG_35-55 peptide emulsified in CFA, followed by i.p. injection of 200 ng PTX to induce EAE. Clinical scores of EAE progression were monitored and recorded daily (N = 15) (d). Body weights of immunized mice were measured and normalized to their respective weights at day 8 post-immunization (e). Spinal cords from Znrf1^F/F and Znrf1^Δglia mice were collected at day 18 (f and g) and day 30 (h and i) after EAE induction. (f and h) Tissue sections were subjected to H&E and LFB staining. (g and i) Tissue sections were subjected to IHC using antibodies against CD3 (T cells) and CD68 (macrophages). (f-i) Scale bars: 200 μm (whole spinal cord sections) and 100 μm (enlarged region #1 and #2). (j-k) Single-cell suspensions were prepared from spinal cords of female Znrf1^F/F and Znrf1^Δglia mice at day 18 (j) and day 30 (k) post-EAE induction, followed by flow cytometry analysis of leucocytes (CD45⁺), Th cells (CD45⁺CD4⁺), cytotoxic T cells (CD45⁺CD8⁺), and myeloid cells (CD45⁺CD11b⁺) (N = 6). Representative flow cytometry plots and quantified numbers of infiltrating immune cells in the CNS are presented. (d and e) Data are presented as mean ± SD. ns, not significant. Statistical significance was determined by the Wilcoxon matched-pairs signed-rank test. (f-i) Representative spinal cord sections and quantification of infiltrating immune cells or demyelination in Znrf1^F/F and Znrf1^Δglia mice are shown. (d-k) Data are presented as mean ± SD. ns, not significant. Statistical significance was determined by the unpaired Student’s t-test

ZNRF1 deficiency in myeloid cells aggravates EAE

To investigate the role of ZNRF1 in myeloid cells during EAE, we crossed Znrf1^F/F mice with LysMCre knock-in mice [15, 78] to generate myeloid cell-specific ZNRF1-deficient (hereafter referred to as Znrf1^Δmye) mice and induced EAE in both control and Znrf1^Δmye mice. Unlike Znrf1^Δglia mice, Znrf1^Δmye mice developed markedly more severe EAE phenotypes, exhibiting elevated clinical scores and more pronounced changes in body weight (Fig. 3a and b). Flow cytometric analysis corroborated these results, showing significantly increased infiltrating leucocytes, CD4⁺ T cells, and myeloid cells, but not cytotoxic CD8⁺ T cells, in spinal cords of Znrf1^Δmye mice at both day 18 (Fig. 3c) and day 30 (Fig. S5a) after EAE induction. Notably, Znrf1^Δmye mice also exhibited significantly elevated Th1 and Th17 cells (Fig. 3d).

Fig. 3

ZNRF1 deficiency in myeloid cells promotes EAE pathogenesis and neuroinflammation in mice. Age-matched female Znrf1^F/F and Znrf1^Δmye mice were immunized with 200 μg of MOG_35-55 peptide emulsified in CFA, followed by i.p. injection of 200 ng PTX to induce EAE. a Clinical scores of EAE progression were monitored and recorded daily (N = 42). b Body weights of immunized mice were measured and normalized to their respective weights at day 8 post-immunization. c At day 18 post-EAE induction, spinal cords from female Znrf1^F/F and Znrf1^Δmye mice were harvested to prepare single-cell suspensions. Cells were stained with antibodies against CD45, CD4, CD8, and CD11b, followed by flow cytometry analysis to quantify leucocytes (CD45⁺), Th cells (CD45⁺CD4⁺), cytotoxic T cells (CD45⁺CD8⁺), and myeloid cells (CD45⁺CD11b⁺) (N = 6). d Flow cytometry analysis of Th1 (CD4⁺IFNγ⁺) and Th17 (CD4⁺IL-17A⁺) cells in the spinal cords of Znrf1^F/F (N = 5) and Znrf1^Δmye mice (N = 5) at day 18 post-EAE induction. e–g Spinal cord tissue sections from Znrf1^F/F and Znrf1^Δmye mice were collected and quantified at day 18 post-EAE induction. Tissue sections were subjected to H&E and LFB staining (e), and IHC staining using antibodies against CD3 (T cells) and CD68 (macrophages) (f) or IFNγ (Th1 cells) and IL-17A (Th1 cells) (g). Scale bars: 200 μm (whole spinal cord sections) and 100 μm (enlarged region #1 and #2). h-i Whole-mount imaging of the anterior (h) and posterior (i) spinal cord was performed in Znrf1^F/F and Znrf1^Δmye mice after EAE induction. At day 18 post-EAE induction, Znrf1^F/F and Znrf1^Δmye mice were perfused with wheat germ agglutinin (WGA) to label blood vessels (red). Spinal cords were collected and immunostained with antibodies against CD3 (green) to detect CD3⁺ T cells, lymphatic vessel endothelial hyaluronan receptor 1 (Lyve1) (magenta) for lymph vessels, and DAPI (white) for nuclei. Tissue clearing was performed as described in the METHODS section. Confocal images from whole-mount spinal cords and enlarged regions are shown. Scale bars: 2000 μm (whole-mount anterior spinal cord for Znrf1^F/F and Znrf1^Δmye mice and whole-mount posterior spinal cord for Znrf1^Δmye mice), 1000 μm (whole-mount posterior spinal cord for Znrf1^F/F mice), 200 μm (enlarged region #1), 100 μm (enlarged region #2 and #3). (a,b) Data are presented as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, determined by the Wilcoxon matched-pairs signed-rank test. (c and d) Representative flow cytometry plots and quantified numbers of infiltrating immune cells in the CNS are shown. Data are presented as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, determined by the unpaired Student’s t-test. (e–g) Representative spinal cord sections and quantification of infiltrating immune cells or demyelination in Znrf1^F/F and Znrf1^Δmye mice. Data are presented as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, determined by the unpaired Student’s t-test

Consistent with these findings, H&E staining revealed increased immune cell infiltration in spinal cords of Znrf1^Δmye mice (Fig. 3e and S5b). In addition, LFB staining showed more extensive demyelination in Znrf1^Δmye spinal cords (Fig. 3e and S5b). IHC analysis further confirmed greater infiltration of CD3⁺ T cells and CD68⁺ macrophages in Znrf1^Δmye mice at both day 18 and day 30 post-induction (Fig. 3f and S5b). IHC analysis also demonstrated increased infiltration of Th1 cells (IFNγ⁺) in Znrf1^Δmye mice at both time points (Fig. 3g and S5b), while Th17 cells (IL-17A⁺) were significantly increased at day 30 and modestly elevated at day 18 post-induction (Fig. 3g and S5b). Inflammatory lesions in spinal cords typically develop randomly following EAE induction, with immune cells infiltrating within these lesions. To visualize the localization of infiltrating CD3⁺ T cells in the spinal cords, we performed whole-mount spinal cord immunostaining with anti-CD3 antibodies and subjected the samples to tissue clearing as described in the Methods. Consistent with the above findings, we observed increased CD3⁺ T cell infiltration in both anterior and posterior regions of the spinal cord in Znrf1^Δmye mice at day 18 post-EAE induction (Fig. 3h and i). Together, these data indicate that ZNRF1 in myeloid cells suppresses EAE progression and associated neuroinflammation.

ZNRF1 deficiency in myeloid cells increases T cell priming in peripheral lymphoid organs after EAE induction

To further investigate how ZNRF1 deficiency in myeloid cells enhances EAE progression, we analyzed T cell priming in peripheral lymphoid tissues (spleen and DLNs) after EAE induction. During the early phase (day 9) of EAE induction, CD69 expression on CD4⁺ T cells was significantly increased in both spleens and DLNs of Znrf1^Δmye mice (Fig. S4a). At the same time point, Th1 and Th17 cell populations were also significantly elevated (Fig. S4d and S4e). We next examined various CD4⁺ T cell populations in these tissues. Our results revealed that naïve CD4⁺ T cells (CD44⁻CD62L⁺) were significantly reduced in both spleens and DLNs of Znrf1^Δmye mice at day 18 and 30 after EAE induction, whereas CD4⁺ T_CM cells remained comparable to control (Fig. 4a and S5c). In contrast, CD4⁺ T_EM cells were significantly increased in both spleens and DLNs of Znrf1^Δmye mice at day 18 after immunization (Fig. 4a). We also examined CD8⁺ T cell subsets. Compared to Znrf1^F/F mice, CD8⁺ T_CM cells were significantly increased in DLNs of Znrf1^Δmye mice at both day 18 and 30 (Fig. 4b and S5d). In addition, CD8⁺ T_EM cells were slightly increased in DLNs at day 18 and in spleens at day 30 in Znrf1^Δmye mice following EAE induction (Fig. 4b and S5d).

Fig. 4

ZNRF1 deficiency in myeloid cells enhances T cell differentiation in peripheral lymphoid organs following EAE induction. a Flow cytometry analysis of naïve CD4⁺ T cells (CD4⁺CD44⁻CD62L⁺), CD4⁺ T_CM cells (CD4⁺CD44⁺CD62L⁺), and CD4⁺ T_EM cells (CD4⁺CD44⁺CD62L⁻) in the spleens and DLNs of Znrf1^F/F (N = 4) and Znrf1^Δmye mice (N = 4) at day 18 post-EAE induction. b Flow cytometry analysis of naïve CD8⁺ T cells (CD8⁺CD44⁻CD62L⁺), CD8⁺ T_CM cells (CD8⁺CD44⁺CD62L⁺), and effector memory CD8⁺ T_EM cells (CD8⁺CD44⁺CD62L⁻) in the spleens and DLNs of Znrf1^F/F (N = 4) and Znrf1^Δmye mice (N = 4) at day 18 post-EAE induction. (c-e) Flow cytometry analysis of Th1 cells (CD4⁺IFNγ⁺) (c), Th17 cells (CD4⁺IL-17A⁺) (d), and cytotoxic T cells (CD8⁺IFNγ⁺) (e) in the spleens and DLNs of Znrf1^F/F (N = 3) and Znrf1^Δmye mice (N = 3) at day 18 post-EAE induction. Representative flow cytometry plots and quantified numbers of T cell populations in the spleen and DLNs are shown. Data are displayed as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, determined by the unpaired Student’s t-test

We further investigated the differentiation of effector T cell subsets (Th1, Th17, and cytotoxic CD8⁺ T cell) in both spleen and DLN of Znrf1^F/F and Znrf1^Δmye mice during EAE. Th1 cells were significantly increased in DLNs at day 18 and in spleens at day 30 post-EAE induction in Znrf1^Δmye mice (Fig. 4c and S5e). However, Th17 cells were elevated in both spleens and DLNs of Znrf1^Δmye mice at both day 18 and day 30 (Fig. 4d and S5f). Moreover, CD8⁺ cytotoxic T cells were significantly increased in the DLNs, but not the spleen, of Znrf1^Δmye mice at both time points (Fig. 4e and S5g). Taken together, these results suggest that ZNRF1 deficiency in myeloid cells promotes robust T cell activation and differentiation during EAE.

ZNRF1 regulates antigen-specific T cell proliferation and activation after EAE induction

Antigen-activated T cells are known to be primed by antigen-presenting cells (APCs) in peripheral lymphoid organs before migrating to the CNS [31, 58]. Based on the results shown above, we hypothesized that ZNRF1 in myeloid cells regulates APC function and thereby contributes to antigen-specific T cell responses during EAE. To test this hypothesis, we first co-cultured carboxyfluorescein succinimidyl ester (CFSE)-labeled CD4⁺ T cells from the spleens of 2D2 TCR transgenic mice with CD11b⁺ myeloid cells isolated from either un-immunized Znrf1^F/F or Znrf1^Δmye mice in the presence or absence of MOG_35-55 for 3 days (Fig. S6a). MOG_35-55 stimulation led to significantly increased proliferation of CD4⁺ T cells co-cultured with Znrf1^Δmye myeloid cells compared with Znrf1^F/F controls (Fig. S6d). We next repeated the co-culture using CFSE-labeled CD4⁺ T cells from 2D2 mice and CD11b⁺ myeloid cells isolated from immunized Znrf1^F/F and Znrf1^Δmye mice at day 12 post-EAE, again in the presence or absence of MOG_35-55 for 3 days (Fig. 5a). Consistent with the earlier experiments, MOG_35-55 stimulation induced significantly greater proliferation of CD4⁺ T cells co-cultured with Znrf1^Δmye myeloid cells compared with controls (Fig. 5b). Additionally, the populations of activated T cells (CD4⁺CD25⁺) and CD4⁺ T_CM cells were significantly elevated in Znrf1^Δmye cultures following MOG_35-55 stimulation (Fig. 5c and d). We also measured cytokine levels in the co-culture supernatants. IFNγ levels were significantly increased in Znrf1^Δmye co-cultures after MOG_35-55 stimulation, whereas IL-17A levels were only slightly elevated (Fig. 5e). By contrast, the levels of pro-inflammatory cytokines, including granulocyte–macrophage colony-stimulating factor (GM-CSF), IL-1β, IL-6, and tumor necrosis factor-α (TNFα), were comparable between Znrf1^F/F and Znrf1^Δmye cultures following MOG_35-55 stimulation (Fig. 5f). Together, these findings demonstrate that ZNRF1 suppresses myeloid cells-mediated antigen-specific T cell proliferation and activation.

Fig. 5

ZNRF1 deficiency in myeloid cells enhances T cell proliferation and activation in peripheral lymphoid organs after EAE induction. a Schematic diagram of the experimental setup. Znrf1^F/F and Znrf1^Δmye mice were immunized with MOG_35–55 without PTX administration. On day 12 post-immunization, CD11b⁺ cells were isolated from the spleens of immunized Znrf1^F/F and Znrf1^Δmye mice using magnetic selection. CD4⁺ T cells were isolated from the spleens of naive 2D2 mice via magnetic selection and labeled with 5 μM carboxyfluorescein succinimidyl ester (CFSE). CD11b⁺ and CD4⁺ cells were co-cultured at a 10:1 ratio in the absence or presence of 50 μg/mL MOG_35–55 for 3 days, followed by flow cytometry analysis. b Flow cytometric analysis of CD4⁺ T cell proliferation in co-cultures of CD11b⁺ cells from Znrf1^F/F (N = 7) and Znrf1^Δmye (N = 7) mice with naive 2D2 CD4⁺ T cells for 3 days, either in the absence or presence of 50 µg/ml MOG_35-55. Representative flow plots (top) and quantification of the percentage of proliferating CD4⁺ T cells (bottom) are shown. c Flow cytometric analysis of activated T cells (CD4⁺CD25⁺) among total T cells in Znrf1^F/F (N = 7) and Znrf1^Δmye (N = 7) mice 3 days after co-culture. d Flow cytometric analysis of CD4⁺ T_CM cells (CD4⁺CD44⁺CD62L⁺) among total T cells in Znrf1^F/F (N = 7) and Znrf1^Δmye (N = 7) mice 3 days after co-culture. e–f Cytokine levels in supernatants from myeloid cells-T cell cocultures for 3 days were determined by multiplex immunoassay. All data are presented as mean ± SD. ns, not significant. *P < 0.05, **P < 0.01, determined by the unpaired Student’s t-test

ZNRF1 deficiency upregulates MHC-II on macrophages after EAE induction

Our findings suggest that ZNRF1 deficiency in myeloid cells leads to more severe EAE progression, likely through enhanced antigen-specific T cell proliferation, activation, and differentiation. APCs are known to regulate T cell activation and functions through various surface molecules, including MHC-I, MHC-II, CD40, PD-L1, FasL, and CD80/CD86 [9, 48, 84]. To determine whether ZNRF1 deficiency affects the expression of these regulatory molecules, we performed flow cytometric analysis on spleen-derived myeloid subsets at day 12 post-EAE. We defined and quantified distinct myeloid APC cell subsets, including neutrophils (CD45⁺CD11c⁻B220⁻F4/80⁻ CD11b⁺Ly6G⁺Ly6C⁻), monocytes (CD45⁺CD11c⁻B220⁻F4/80⁻CD11b⁺Ly6G⁻Ly6C⁺), macrophages (CD45⁺Ly6G⁻Ly6C⁻CD11c⁻B220⁻CD11b⁺F4/80⁺), dendritic cells (DCs; CD45⁺Ly6G⁻Ly6C⁻CD11b⁻F4/80⁻CD11c⁺), conventional DCs (cDC; CD45⁺Ly6G⁻Ly6C⁻CD11b⁻F4/80⁻CD11c⁺B220⁻), plasmacytoid DCs (pDC; CD45⁺Ly6G⁻Ly6C⁻CD11b⁻F4/80⁻CD11c⁺B220⁺), cDC1(CD45⁺Ly6G⁻Ly6C⁻CD11b⁻F4/80⁻CD11c⁺B220⁻CD172a⁻XCR1⁺), and cDC2 (CD45⁺Ly6G⁻Ly6C⁻CD11b⁻F4/80⁻CD11c⁺B220⁻CD172a⁺XCR1⁻) (Fig. S7a). We found no significant differences in the percentage of these leukocytes and myeloid APC cell subsets between Znrf1^+/+ vs. Znrf1^−/− mice (Fig. 6a) or Znrf1^F/F vs. Znrf1^Δmye mice (Fig. 6b) at day 12 post-EAE.

Fig. 6

ZNRF1 deficiency in myeloid cells alters macrophage surface MHC-II expression following EAE induction. a, b Dot plots showing the percentages of CD45⁺cells, neutrophils, monocytes, macrophages, DCs, cDCs, pDCs, cDC1s, and cDC2s within total live cells (left) and within CD45⁺ cells (right) in the spleens from Znrf1^+/+ (N = 4) versus Znrf1^−/− (N = 4) mice (a) or Znrf1^F/F (N = 4) versus Znrf1^Δmye (N = 4) mice (b) at day 12 post-EAE induction. c, d Flow cytometric analysis of surface MHC-II expression on macrophages, DCs, and cDCs in the spleens of Znrf1^+/+ (N = 4) versus Znrf1^−/− (N = 4) mice (c) or Znrf1^F/F (N = 4) versus Znrf1^Δmye (N = 4) mice (d) at day 12 post-EAE induction. e Flow cytometric analysis of surface and total MHC-II expression on macrophages in the spleens of Znrf1^F/F (N = 5) versus Znrf1^Δmye (N = 5) mice at day 12 post-EAE induction. The right panels show the quantified mean fluorescence intensity (MFI) of MHC-II. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, determined by the unpaired Student’s t-test

Given that MHC-II is crucial for CD4⁺ T cell activation, we measured its surface expression on these myeloid cells. Compared to Znrf1^+/+ mice, Znrf1^−/− macrophages displayed increased MHC-II surface expression, whereas other myeloid APC subtypes did not, at day 12 after EAE induction (Fig. 6c and S8a). Similarly, Znrf1^Δmye macrophages, but not other myeloid APC subtypes, showed upregulated MHC-II surface levels compared to Znrf1^F/F controls at both day 9 and day 12 (Fig. 6d, S7c, and S9a). By contrast, MHC-I levels, which are important for CD8⁺ T cell activation, remained unchanged across all myeloid APC subtypes in both Znrf1^+/+ vs. Znrf1^−/− (Fig. S8e) and Znrf1^F/F vs. Znrf1^Δmye mice (Fig. S9e). Furthermore, surface levels of co-stimulatory molecules (CD40 and CD80/86) and inhibitory molecules (PD-L1 and FasL) were comparable between Znrf1^+/+ vs. Znrf1^−/− (Fig. S8b-S8d and S10a-S10b) and Znrf1^F/F vs. Znrf1^Δmye mice at day 12 post-EAE induction (Fig. S9b-S9d and S10c-S10d). To determine whether ZNRF1 regulates MHC-II protein expression or its surface localization, we assessed both surface and total MHC-II expression on macrophages from Znrf1^F/F and Znrf1^Δmye mice at day 12 post-EAE induction. Surface expression of MHC-II, but not MHC-I, FasL, or PD-L1, was significantly increased in Znrf1^Δmye macrophages compared with Znrf1^F/F controls (Fig. 6e and Fig. S11). However, total MHC-II levels in macrophages were comparable between Znrf1^F/F mice and Znrf1^Δmye groups (Fig. 6e). These findings indicate that ZNRF1 selectively regulates MHC-II surface expression on macrophages during EAE.

In summary, our work reveals that ZNRF1, an E3 ubiquitin ligase, plays a critical role in suppressing the pathogenesis of EAE. Specifically, ZNRF1 deficiency in myeloid cells, but not microglia, leads to exacerbated disease, driven by enhanced antigen-specific T cell activation, proliferation, and Th1/Th17 polarization. Elevated MHC-II surface expression on ZNRF1-deficient macrophages likely underlies this enhanced T cell priming. These findings highlight ZNRF1 as a novel immune checkpoint regulator with potential relevance to MS pathogenesis.

Discussion

Multiple sclerosis (MS) is a common chronic inflammatory disease of the CNS, characterized demyelination that forms plaques in the brain and spinal cord, accompanied by infiltration of immune cells [83, 88]. The entry of peripheral immune cells, including macrophages, DCs, T cells, and B cells, into the CNS through a disrupted blood–brain barrier (BBB), along with the activation of resident microglia and astrocytes, collectively contributes to inflammation, demyelination, and neuronal damage [37, 65, 83, 88]. However, the precise mechanisms driving neuroinflammation in MS remain incompletely understood. In this study, we demonstrate that ZNRF1 plays a protective role in EAE progression, specifically in peripheral macrophages rather than microglia. ZNRF1 deficiency in myeloid cells led to increased immune cell infiltration into the CNS and more severe EAE clinical symptoms upon MOG_35-55 immunization. Furthermore, the absence of ZNRF1 in myeloid-derived macrophages enhanced the activation and differentiation of T cells in peripheral lymphoid organs. These data collectively indicate that ZNRF1 regulates MHC-II surface expression in myeloid cells to prevent excessive T cell-mediated neuroinflammation during EAE.

Accumulated evidences indicate that CD4⁺ T cells play a critical role in MS pathogenesis and are closely associated with disease severity [45, 73]. Th1 and Th17 subsets are particularly important in driving MS progression in humans as well as in the murine EAE model, which mimics aspects of MS [45]. Our data suggest that ZNRF1 does not directly involve in CD4⁺ T cell regulations to affect EAE progression. T cell maturation in the thymus and peripheral lymphoid organs appeared normal, and no defects in T cell activation or polarization were observed in Znrf1^−/− mice. Additionally, the similar disease severity observed in both Znrf1^−/− mice and Znrf1^Δmye mice further suggests that ZNRF1’s immunoregulatory role in the EAE model is mediated through myeloid cells, which in turn regulate T cell function.

While T cells are central to MS pathology, other immune cell types, including monocyte-derived macrophages and microglia, also play essential roles in disease progression [68]. Microglia, as CNS-resident macrophages, are key players in CNS development, immune surveillance, and repair [5, 22, 54]. They exert both pro-inflammation and neuroprotective effects. For example, microglia phagocytose myelin debris and secrete neuroprotective and anti-inflammatory factors that promote remyelination during EAE. However, they also contribute to neuroinflammation by releasing proinflammatory cytokines and chemokines, thereby exacerbating demyelination and lesion formation [32]. Monocyte-derived macrophages exert similar dual roles in EAE. They infiltrate the CNS, contribute to axonal degeneration, act as antigen-presenting cells, and drive pathogenic Th1 and Th17 responses [11, 67, 77, 91]. In MS lesions, upregulated expression of HLA and CD86 on macrophages has been observed [93]. Moreover, in the EAE model, the activation of both macrophages and microglia correlates with disease progression [36]. These data suggest that both microglia and peripheral macrophages contribute to MS and EAE progression. However, recent studies suggest that the relative contributions of microglia and monocyte-derived macrophages to MS pathogenesis are context-dependent, varying by disease phase and CNS compartment. In the early stages of EAE, macrophages and DCs exhibit stronger antigen-presenting capabilities than microglia. In contrast, in the later phases of EAE, microglia become key APCs stimulating T cells [68]. Microglia are particularly proinflammatory in the early phases of MS but play protective roles in later stages, promoting repair through debris clearance, neuroprotective factors and anti-inflammatory cytokines secretion, and antioxidant production [9, 46, 79]. Supporting this dual role, PLX5622-mediated microglia depletion delays EAE onset but does not affect the chronic disease phase [60]. Conversely, depletion of macrophages reduces CNS damage and disease severity in rat EAE models [10, 34]. Molecular pathways governing macrophages versus microglial function in EAE are being elucidated. For example, Gasdermin D (GSDMD), which triggers pyroptosis, and Signal Transducer and Activator of Transcription 3 (STAT3), a transcription factor in the Janus kinase (JAK)/STAT family, contribute to EAE through peripheral myeloid cells, but not microglia [14, 49]. Cbl-b, a Cbl family E3 ubiquitin ligase, also acts in monocyte-derived macrophages, but not microglia, to inhibit IL-6 production and Th17 differentiation, thus limiting EAE progression [89, 91].

Our study adds to this growing body of evidence by showing that ZNRF1 in peripheral macrophages, but not microglia, regulates EAE progression. Consistently, ZNRF1 deficiency in myeloid cells, but not microglia, led to an increase in pathogenic Th1 and Th17 responses. In contrast, comparable infiltration of Th1 and Th17 cells was observed between wild-type and Znrf1^Δglia mice, suggesting that the lack of phenotype in Znrf1^Δglia mice is unlikely due to the absence of other immune APCs required for T cell priming. Furthermore, surface expression of MHC-II was significantly upregulated on peripheral macrophages in Znrf1^−/− mice, while co-stimulatory molecules such as CD40, CD80, or CD86 remained unchanged. Additionally, myeloid cells isolated from immunized Znrf1^Δmye mice enhanced the activation and differentiation of CD4⁺ T cells from 2D2 transgenic mice ex vivo. Since MHC-II is essential for antigen presentation to CD4⁺ T cells [71, 81], and MHC-II alleles are strongly associated with MS susceptibility [9, 16, 17], our findings suggest that ZNRF1 is a key regulator of MHC-II surface expression, and, by extension, of T cell activation and neuroinflammation.

The surface expression of MHC-II on myeloid cells is known to increase during inflammation [9, 16, 17]. Our previous studies have shown that ZNRF1 modulates inflammatory responses mediated by TLR4 and TLR3 through distinct mechanisms [47, 50]. ZNRF1 regulates TLR4-driven immune responses by targeting caveolin-1, a major component of caveolae, for proteasomal degradation. On the other hand, ZNRF1 directly ubiquitinates TLR3 to regulate its lysosomal trafficking and termination of downstream signaling. In addition, ZNRF1 activity can also be modulated by receptor-mediated tyrosine kinases, thereby influencing downstream responses [50, 74]. Our data suggest that ZNRF1 suppresses EAE pathogenesis by maintaining homeostasis levels of cell surface MHC-II on macrophages. MHC-II is trafficked to the cell surface via endosomal-lysosomal compartments following synthesis [39, 40, 71]. E3 ubiquitin ligases, such as MARCHI, have been shown to restrict MHC-II level by ubiquitinating its β chain and promoting its degradation [56, 71]. As ZNRF1 is also an E3 ubiquitin ligase, it remains to be determined whether it regulates MHC-II trafficking to the cell surface through ubiquitination. Our findings highlight ZNRF1 as a key regulator of EAE development through its control of MHC-II surface expression on myeloid cells. Further investigations are warranted to elucidate the underlying mechanisms by which inflammatory signals activate ZNRF1 and how ZNRF1, in turn, modulates MHC-II surface expression during EAE.

In summary, our findings demonstrate that ZNFR1 in peripheral macrophages, but not microglia, plays a crucial immunosuppressive role in autoimmune neuroinflammation and EAE pathogenesis. ZNRF1 maintains the homeostasis of surface MHC-II expression on macrophages, thereby restraining CD4⁺ T cell activation and differentiation. This process attenuates T cell-driven neuroinflammation and demyelination in the CNS during EAE (Fig. 7). These insights identify ZNRF1 as a potential biomarker and therapeutic target for MS, with implications for further modulation of myeloid cell function in neuroinflammatory diseases.

Fig. 7

Graphical model summarizing the protective role of ZNRF1 in EAE induction. Proposed schematic illustrating how ZNRF1 in myeloid cells regulates the surface expression of the key regulatory molecule, MHC-II, thereby controlling antigen-specific T cell proliferation and activation. This model highlights the suppressive role of ZNRF1 in myeloid cells during neuroinflammation