Diet modulates the therapeutic effects of dimethyl fumarate mediated by the immunometabolic neutrophil receptor HCAR2

Abstract

Monomethyl fumarate (MMF) and its prodrug dimethyl fumarate (DMF) are currently the most widely used agents for the treatment of multiple sclerosis (MS). However, not all patients benefit from DMF. We hypothesized that the variable response of patients may be due to their diet. In support of this hypothesis, mice subjected to experimental autoimmune encephalomyelitis (EAE), a model of MS, did not benefit from DMF treatment when fed a lauric acid (LA)-rich diet. Mice on normal chow (NC) diet, in contrast, and even more so mice on high-fiber (HFb) diet showed the expected protective DMF effect. DMF lacked efficacy in the LA diet-fed group despite similar resorption and preserved effects on plasma lipids. When mice were fed the permissive HFb diet, the protective effect of DMF treatment depended on hydroxycarboxylic receptor 2 (HCAR2), which is highly expressed in neutrophil granulocytes. Indeed, deletion of Hcar2 in neutrophils abrogated DMF protective effects in EAE. Diet had a profound effect on the transcriptional profile of neutrophils and modulated their response to MMF. In summary, DMF required HCAR2 on neutrophils as well as permissive dietary effects for its therapeutic action. Translating the dietary intervention into the clinic may improve MS therapy.

Keywords: experimental autoimmune encephalomyelitis; medicine; mouse; multiple sclerosis; neuroscience; neutrophils.

Figure 1.. Diet modulates dimethyl fumarate (DMF) efficacy in experimental autoimmune encephalomyelitis (EAE).

(A) Diet intervention started 14 days before immunization and was followed by DMF treatment (from dpi 3) and scoring (from dpi 7), feces collection 1 day before termination (dpi 28). (B) Body weight of mice on lauric acid diet (LAD), normal chow diet (NCD), and high-fiber diet (HFbD) during the course of EAE. DMF reduced weight loss in mice on NCD and HFbD, while the effect was lost in the LAD group. (C) Clinical scores in mice treated orally with vehicle or DMF (50 mg/kg body weight, twice daily). (D) Area under the curve of clinical scores, maximum scores, and the day of disease onset in mice on the three diets with or without DMF treatment. Dpi, day post-immunization. *p<0.05, **p<0.01, ***p<0.001. Means ± s.e.m. are shown. Points represent individual mice (D). Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 2.. Effects of diet and dimethyl fumarate (DMF) on metabolism.

(A) Monomethyl fumarate (MMF) plasma concentrations in mice that received lauric acid diet (LAD), normal chow diet (NCD), or high-fiber diet (HFbD) and oral DMF (50 mg/kg body weight) 20 min before blood sampling. (B) Sparse partial least-squares discriminant analysis (sPLS-DA) plot for the lipidomics dataset with corresponding loading plots for components 1, 2, and 3 (top 10). (C) Effects of LAD versus HFbD on plasma lipids in mice treated with vehicle are shown as fold change (FC). Each dot represents an individual lipid, sorted by the number of double bonds. Colored lipids were significantly changed by the diet (n=9–10 per group, Mann-Whitney U test, p≤0.05 (false discovery rate [FDR] adjusted), FC threshold = 1.5 [dotted line]). (D) Relative plasma concentration of fatty acids (FA) 18:2 in mice treated with vehicle. (E) Effects of DMF treatment on plasma lipids within diet groups are shown as FC. Each dot represents an individual lipid, sorted by the lipid class. Colored lipids were significantly changed by the DMF treatment (n=8–10 per group, Mann-Whitney U test, p≤0.05 [unadjusted], FC threshold = 1.5 [dotted line]). Major lipid classes are depicted. CAR, carnitines; CE, cholesterol ester; Cer, ceramides; FFA, free fatty acids; LNAPE, N-acyl lysophosphatidylethanolamine; PC, phosphatidylcholines; PC-O, ether-linked phosphatidylcholines; SM, sphingomyelins; TG, triglycerides; TG-O, ether-linked triglycerides. Samples in (B–E) were obtained on day post-immunization 28 in EAE in the experiment shown in Figure 1. **p<0.01, ***p<0.001. Means ± s.e.m. are shown. Points represent individual mice (A, B, D). Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 2—figure supplement 1.. Plasma concentration of short-chain fatty acids (SCFAs).

(A) Concentrations of acetic acid in plasma did not differ between diets or dimethyl fumarate (DMF) treatment. (B) Concentrations of propionic acid in plasma did not differ between diets or DMF treatment. Samples were obtained in the experiment shown in Figure 1. Means ± s.e.m. are shown. Points represent individual mice. Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 2—figure supplement 2.. Chromatograms and fragmentation spectra.

Representative chromatograms and fragmentation mass spectra of monomethyl fumarate (MMF) and the internal standard MMF-d3 (retention time = 3.34 min; m/z=129.019 and 132.040).

Figure 3.. Effects of diet and dimethyl fumarate (DMF) on metabolic parameters.

(A) Sparse partial least-squares discriminant analysis (sPLS-DA) plot for the metabolomics dataset with corresponding loading plots for components 1, 2, and 3 (top 10). Components 1 and 2 mainly reflected diet-induced effects, while component 3 reflected DMF-induced effects. (B) Relative plasma concentrations of the isoflavone glycitein and the isoflavone metabolite equol in mice treated with different diets and with or without DMF. (C) Relative plasma concentrations of N-acetyltyrosine and sphingosine 1-phosphate in mice treated with different diets and with or without DMF. (D) Relative plasma concentrations of glutamyl-glutamine and uracil in mice treated with different diets and with or without DMF. Samples were obtained in the experiment shown in Figure 1. *p<0.05, **p<0.01, ***p<0.001. Means ± s.e.m. are shown. Points represent individual mice. Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 4.. Effects of diet and dimethyl fumarate (DMF) on microbiota.

(A) Diet affected the microbiota of mice subjected to experimental autoimmune encephalomyelitis (EAE) and treated with vehicle or DMF as indicated. High-fiber (HFbD) diet increased short-chain fatty acid (SCFA) producers at the phylum and genus level. (B) α-Diversity was significantly reduced by the HFb diet compared to lauric acid (LAD) and normal chow (NCD) diets, but DMF treatment had no effect. Similarly, only diet, not treatment, affected β-diversity as shown by principal coordinates analysis and PERMANOVA. (C) HFb diet or DMF treatment significantly increased the relative abundance of the SCFA producers Prevotellamassilia or Parabacteroides and Acetatifactor, respectively. Fecal samples were obtained from the experiment shown in Figure 1 at day post-immunization (dpi) 28. *p<0.05, **p<0.01, ***p<0.001. Means ± s.e.m. are shown. Points represent individual mice. Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 5.. HCAR2 mediates the dimethyl fumarate (DMF) effect in experimental autoimmune encephalomyelitis (EAE) in mice on high-fiber (HFb) diet.

Mice were switched to HFb diet 2 weeks before immunization and treated orally with vehicle or DMF (50 mg/kg body weight, twice daily) from day post-immunization (dpi) 3. (A) Clinical scores of Hcar2^+/+ and Hcar2^-/- mice. DMF treatment significantly improved the neurological deficit in Hcar2^+/+ mice, while the effect was lost in Hcar2^-/- mice, indicating that HCAR2 mediates the protective effect of DMF. (B) Area under the curve of clinical scores in Hcar2^+/+ and Hcar2^-/- mice treated with vehicle or DMF. *p<0.05, **p<0.01. Means ± s.e.m. are shown. Points represent individual mice (B). Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 6.. Hcar2 reporter mice Hcar2^mRFP indicate that diet and dimethyl fumarate (DMF) treatment have no marked effects on Hcar2 expression in peripheral immune cells of mice subjected to experimental autoimmune encephalomyelitis (EAE).

After feeding Hcar2^mRFP mice with high-fiber diet (HFbD) or lauric acid diet (LAD) for 2 weeks, EAE was induced, and animals were treated orally with vehicle or DMF (50 mg/kg body weight, twice daily from day post-immunization [dpi] 3). On dpi 16/17, peripheral immune cells were analyzed by FACS. For the gating strategy, see Figure 6—figure supplement 1. The count of neutrophils or monocytes was not different between experimental groups (Figure 6—figure supplement 1B). (A) Histograms show that in both diets, regardless of treatment, more than 95% of neutrophils were RFP-positive. Hcar2 expression per neutrophil, represented as standardized unit (SU) of mRFP⁺, did not depend on DMF treatment or diet. (B–D) Among monocyte populations, Ly6C^int and Ly6C^low monocytes expressed mRFP at intermediate levels while almost all pro-inflammatory Ly6C^high monocytes were mRFP-negative. DMF did not affect their Hcar2-mRFP expression per cell in both diets. HFbD slightly increased mRFP expression in Ly6C^int monocytes. *p<0.05. Means ± s.e.m. are shown. Points represent individual mice (B). Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 6—figure supplement 1.. Gating strategy of FACS measurement of mouse blood.

(A) Blood cells were gated into viable and CD45-positive cells. Subsequently, the leukocyte population was gated into CD45⁺CD11b⁺Ly6G⁺Ly6C^- neutrophils, CD45⁺CD11b⁺Ly6G^-Ly6C^hi, CD45⁺CD11b⁺Ly6G^-Ly6C^int, and CD45⁺CD11b⁺Ly6G^-Ly6C^low monocytes. (B) Diet and dimethyl fumarate (DMF) treatment had no effects on CD45⁺CD11b⁺Ly6G⁺Ly6C^- neutrophils, CD45⁺CD11b⁺Ly6G^-Ly6C^int, and CD45⁺CD11b⁺Ly6G^-Ly6C^low monocytes counts. However, high-fiber diet (HFbD) led to a decrease in the number of CD45⁺CD11b⁺Ly6G^-Ly6C^hi monocytes compared to lauric acid diet (LAD). Means ± s.e.m. are shown. Points represent individual mice (B).

Figure 6—figure supplement 2.. Hcar2 reporter mice Hcar2^mRFP indicate that diet and dimethyl fumarate (DMF) treatment have no marked effects on Hcar2 expression in immune cells in the brain of mice subjected to experimental autoimmune encephalomyelitis (EAE).

After feeding Hcar2^mRFP mice with high-fiber diet (HFbD) or lauric acid diet (LAD) for 2 weeks, EAE was induced, and animals were treated orally with vehicle or DMF (50 mg/kg body weight, twice daily from day post-immunization [dpi] 3). On dpi 17, immune cells in the brain were analyzed by FACS. For the gating strategy, see Supplementary file 4. (A) Hcar2 expression in CD45^lowCD11⁺ microglia, represented as standardized unit (SU) of mRFP. (B) Hcar2 expression in Ly6G⁺Ly6C⁺ neutrophils, represented as SU of mRFP per cell. (C) Hcar2 expression in Ly6G⁺Ly6C^high (left panel) and Ly6G⁺Ly6C^low (right panel) monocyte-derived macrophages, represented as SU of mRFP. *p<0.05. Means ± s.e.m. are shown. Points represent individual mice. Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 7.. Conditional knockout of Hcar2 in neutrophils (Hcar2^nKO) decreases HCAR2-stimulated increase in intracellular Ca²⁺ concentrations ([Ca²⁺]_i).

Neutrophils were prepared from bone marrow of mice. (A) Stimulation with nicotinic acid (NA, 100 µM, 60 s) increased [Ca²⁺]_i concentrations in a majority of Hcar2^+/+ neutrophils, while it had no effect in Hcar2^-/- neutrophils. Hcar2^Fl/Fl neutrophils responded in a similar manner to NA stimulation as Hcar2^+/+ cells. The amplitude of [Ca²⁺]_i stimulated by nicotinic acid was lower in Hcar2^nKO neutrophils. Representative traces of individual cells are shown. (B) The rate of neutrophils responding to NA (100 µM) was significantly reduced in Hcar2^nKO mice compared to Hcar2^Fl/Fl animals. (C) Stimulation with monomethyl fumarate (MMF) (100 µM) increased [Ca²⁺]_i in neutrophils from Hcar2^Fl/Fl but not Hcar2^nKO mice. (D) The rate of neutrophils responding to MMF (100 µM) was significantly reduced in Hcar2^nKO mice compared to Hcar2^Fl/Fl animals. **p<0.01, ***p<0.001, ****p<0.0001. Means ± s.e.m. are shown. Points represent individual mice. Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 8.. The therapeutic effect of dimethyl fumarate (DMF) in experimental autoimmune encephalomyelitis (EAE) depends on HCAR2 on neutrophils.

(A) Clinical scores of Hcar2^FL/FL and Hcar2^nKO mice fed high-fiber (HFb) diet and treated orally with vehicle or DMF (50 mg/kg body weight, twice daily). DMF treatment significantly improved the neurological deficit in Hcar2^FL/FL mice, while the effect was lost in Hcar2^nKO mice, indicating that HCAR2 mediates the protective effect of DMF in neutrophils. (B) DMF treatment significantly reduced the area under the curve of clinical scores in Hcar2^FL/FL but not in Hcar2^nKO mice. **p<0.01. Means ± s.e.m. are shown. Points represent individual mice (B). Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 9.. Dietary effects on the transcriptional profile of neutrophils.

After mice were fed with lauric acid (LAD), normal chow (NCD), or high-fiber (HFbD) diets for 2 weeks, neutrophils were isolated from bone marrow and treated with vehicle or monomethyl fumarate (MMF, 100 µM, 3 hr) in vitro before bulk RNA sequencing. (A) Heatmap of the 1000 most variable genes is shown, indicating clear differences between diet groups. In contrast, MMF treatment of neutrophils had no marked effects. (B) Top ranking gene ontology (GO) terms identified by comparing diet or treatment groups. (C) Among five genes upregulated in neutrophils of MS patients (Shi et al., 2022), Btg2 was downregulated by MMF treatment and Il1b by diet. *p<0.05. Means ± s.e.m. are shown. Points represent individual mice (C). Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.

Figure 10.. High-fiber diet (HFbD) has a permissive effect on the neutrophil response to monomethyl fumarate (MMF).

Neutrophils were collected from the bone marrow of mice fed lauric acid diet (LAD), normal chow diet (NCD), or HFbD and stimulated with MMF (200 µM). (A) MMF significantly inhibited the adhesion of calcein-labeled neutrophils (green) to activated bEnd.3 brain endothelial cells when neutrophils were isolated from HFbD-fed mice, while no effects were observed in mice fed with LAD or NCD. In the left panels, representative images from the HFbD group are shown. (B) MMF treatment significantly reduced the phorbol 12-myristate 13-acetate (PMA)-triggered release of neutrophil extracellular traps (NETs, arrows) when mice were fed HFbD but not NCD or LAD. In the left panels, representative images of neutrophils treated with PMA + MMF of the three diets are shown. *p<0.05, **p<0.05. Means ± s.e.m. are shown. Points represent cultures from individual mice. Scale bars, 50 µm. Detailed information on the exact test statistics, sidedness, and values is provided in Supplementary file 6.