Dietary cholesterol promotes repair of demyelinated lesions in the adult brain

Abstract

Multiple Sclerosis (MS) is an inflammatory demyelinating disorder in which remyelination failure contributes to persistent disability. Cholesterol is rate-limiting for myelin biogenesis in the developing CNS; however, whether cholesterol insufficiency contributes to remyelination failure in MS, is unclear. Here, we show the relationship between cholesterol, myelination and neurological parameters in mouse models of demyelination and remyelination. In the cuprizone model, acute disease reduces serum cholesterol levels that can be restored by dietary cholesterol. Concomitant with blood-brain barrier impairment, supplemented cholesterol directly supports oligodendrocyte precursor proliferation and differentiation, and restores the balance of growth factors, creating a permissive environment for repair. This leads to attenuated axon damage, enhanced remyelination and improved motor learning. Remarkably, in experimental autoimmune encephalomyelitis, cholesterol supplementation does not exacerbate disease expression. These findings emphasize the safety of dietary cholesterol in inflammatory diseases and point to a previously unrecognized role of cholesterol in promoting repair after demyelinating episodes.

Figure 1. Dietary cholesterol does not aggravate EAE pathology.

(a) Clinical score of mice with MOG-EAE on normal chow or chow supplemented with 5% cholesterol (n=12–16 mice, 2 independent experiments). Start of cholesterol feeding was prophylactic, two weeks before immunization. Arrows illustrate the time points of analyses at the peak of clinical symptoms (16–18 dpi) and at remission (28 dpi). (b) Body weight of experimental animals as in (a) assessed from the day of induction of EAE to the end of monitoring clinical scores (28 days). Data is expressed as mean weight±s.e.m. of n=12–16 animals. Onset of clinical symptoms was paralleled by a drop in body weight, and mice gained weight only after the peak of disease. (c) Lesion characteristics were determined on sections of lumbar spinal cord from mice fed normal chow or cholesterol enriched chow (n=5 animals, representative images on the left, scales 200 μm). Luxol fast blue-periodic acid-Schiff-hematoxylin (LFB/PAS) staining was used to determine the lesion area and number of lesions per section (arrow). Immuno-labeling for myelin basic protein (MBP) was used to determine the per cent of myelinated area within a lesion (defined in the DAPI channel as clusters of >20 nuclei, marked by arrows). On sections immuno-labeled for APP, the number of axonal speroids (arrows) per square mm white matter area was counted, as a readout of axonal damage. In remission, unpaired Student's t-test revealed significantly less axonal damage in cholesterol fed animals. Sections triple stained for microglia/macrophages, T cells, and astrocytes (Iba1-CD3-GFAP triple immuno-labeling) were used to assess the cellular composition of lesions. Unpaired t-tests revealed significantly reduced densities of microglia/macrophages and T cells in cholesterol fed animals (*, P<0.05). Bars represent mean values with individual data points.

Figure 2. Increased BBB permeability in cuprizone treated mice.

(a) Extravasation of Evans blue on sections of the corpus callosum. In control animals, Evans blue fluorescence is restricted to blood vessels but extravasates in mice on cuprizone (arrows) (scale, 50 μm). (b) BBB permeability was measured by Evans blue (EB) extravasation in brains of animals fed cuprizone (cup) for 5 weeks on normal chow or cholesterol supplemented chow, or in brains of animals with EAE 2d after the peak of clinical symptoms (n=3 animals). All treatment groups were normalized to untreated control animals (n=5) and compared by one way ANOVA (P<0.0001). Nutritional cholesterol did not influence BBB permeability. Bars represent mean±s.e.m. (c) Extravasation of bodipy-cholesterol. Maximum intensity projection of bodipy-cholesterol fluorescence in the corpus callosum (delineated by dashed lines) of mice that were kept on cuprizone for 5 weeks in comparison to untreated mice (control) (scale, 50 μm). (d) Quantification of bodipy-cholesterol extravasation after extraction. Data are expressed as fold changes±s.e.m. in cuprizone treated mice compared with untreated control animals (n=6 mice per group, unpaired Student's t-test, P<0.0001).

Figure 3. Cholesterol does not affect cuprizone mediated demyelination.

(a) Scheme depicting the time course of demyelination/remyelination during 6 week cuprizone feeding (upper panel, based on own results and on other studies2528) to the treatment paradigm. To assess the influence of high-cholesterol feeding on demyelination, mice on normal chow or high-cholesterol chow additionally received cuprizone in the diet for between 2 and 5 weeks (black bars) after which mice were analysed histologically. (b) Representative pictures of the corpus callosum of untreated control mice and mice after 5 weeks on cuprizone with the corresponding quantification on the right. Assessed were myelination (Gallyas silver impregnation), the number of mature oligodendrocytes (CAII), the number of oligodendrocyte lineage cells (Olig2), activated microglia (MAC3) and astrocytes (GFAP). Each bar represents the mean value for 3–5 (week 2–4) or 9–10 (week 5; untreated controls, ctrl) animals per condition with individual data points (scale 100 μm). (c) APP positive spheroids per mm² in the corpus callosum at the end of 2–5 weeks of cuprizone with or without cholesterol supplementation (n=3–4 animals at 2 and 3 weeks, n=4–5 at week 4, n=6 untreated controls, n=9–10 at week 5).

Figure 4. Cholesterol facilitates remyelination after chronic cuprizone exposure.

(a) Scheme depicting the time course of demyelination/remyelination during cuprizone feeding to the treatment paradigm. To assess the influence of high-cholesterol feeding on spontaneous remyelination, mice received cuprizone in normal chow or chow supplemented with cholesterol for 5, 6 or 12 weeks (black bars) after which mice were analysed histologically. (b) Evaluation of disease in the corpus callosum of mice that were treated with cuprizone for 5, 6 or 12 weeks on normal chow or chow enriched with cholesterol. Corresponding representative pictures of the 12 weeks treatment cohort are on the left. Assessed were myelination (Gallyas silver impregnation), the number of mature oligodendrocytes (CAII), the number of oligodendrocyte lineage cells (Olig2), activated microglia (MAC3) and astrocytes (GFAP). Each bar represents the mean value of 4 (week 12) or 8–10 (week 5, 6) animals per condition with individual data points (scale 100 μm). (c) Myelinated axons per 10 μm² in the corpus callosum at the end of 6 and 12 weeks of cuprizone with or without cholesterol supplementation (n=4 animals, Two-way ANOVA and Sidak's post test). (d) APP positive spheroids per mm² in the corpus callosum at the end of 6 and 12 weeks of cuprizone with or without cholesterol supplementation (n=3–8 animals, Two-way ANOVA and Sidak's post test). Asterisks represent significant differences with *P<0.05; **P<0.01; ****P<0.0001.

Figure 5. Cholesterol facilitates remyelination after cuprizone withdrawal.

(a) Scheme depicting the time course of demyelination/remyelination during cuprizone feeding (remyelination after cuprizone withdrawal in purple). The influence of cholesterol on remyelination was assessed by feeding mice cuprizone in normal chow for 4 weeks (4, black bars) followed by ‘induced remyelination' after cuprizone withdrawal for 1 (4+1) or 2 (4+2) weeks on normal chow or cholesterol supplemented chow. (b) Representative pictures of the corpus callosum of mice after one week (4+1) remyelination. Corresponding quantification is on the right also including values for 2 weeks remyelination (4+2). Assessed were myelination (Gallyas silver impregnation), the number of mature oligodendrocytes (CAII), the number of oligodendrocyte lineage cells (Olig2), activated microglia (MAC3), and astrocytes (GFAP). Each bar represents the mean value from n=4–5 (4 and 4+2) or n=7 (4+1) animals (scale, 100 μm; Two-way ANOVA and Sidak's post test). (c) Quantification of proliferating OPCs (PCNA positive Olig2 positive) in the corpus callosum of mice after 4+1 treatment paradigm (4+1) or after 12 weeks (12) of cuprizone. Each bar represents the mean of n=6–7 (week 4+1), or n=4 (week 12) animals (Student's t-test). (d) Quantification of newly differentiated postmitotic oligodendrocytes (TCF4 positive, PCNA negative) in the corpus callosum treated as in c). Each bar represents the mean of n=6–7 (week 4+1), or n=4 (week 12) animals (Student's t-test). (e) Myelinated axons per 10 μm² in the corpus callosum at the end of the 4+1 (n=7) and 4+2 (n=4) treatment paradigm (two-way ANOVA and Sidak's post test). (f) APP positive spheroids per mm² in the corpus callosum (4+1 n=7; 4+2 n=3–4 animals, two-way ANOVA and Sidak's post test). (g) Motor learning as assessed by maximum velocity (Vmax) on a complex wheel (n=6 animals), expressed as per cent of the Vmax on a training wheel (mean of the last 7 days before changing to a complex wheel). Statistical evaluation of Vmax was done by Two-way ANOVA (cholesterol effect P<0.0001) and Sidak's post tests. Asterisks represent significant differences with *P<0.05; **P<0.01; ***P<0.001.

Figure 6. Cholesterol supports remyelination in the lysolecithin model.

(a) Representative images of spinal cord sections 14 days post lesion (dpl) with 1 μl 1% lysolecithin in the ventral spinal cord with quantification of n=5 (cholesterol chow) and n=6 (normal chow) animals. Student's t-tests revealed significantly more Olig2 positive oligodendroglial cells within the lesion area (P<0.0001), and significantly more MBP positive area (P=0.027; scales, 100 μm). (b,c) Representative electron micrographs (scale 1 μm) and quantification of myelin sheath thickness and the portion of remyelinated axons in control and cholesterol fed mice at 14 dpl by g-ratio analysis (n=3 animals per group). (d) Body weight of experimental animals assessed at the day of lesion (day 0), after 7d and after 14d at the end of the experiment. Shown are the means±s.e.m. of n=9 (chol chow) to 10 (normal chow) animals. Two-way ANOVA with Sidaks post tests revealed a significant influence of cholesterol feeding at both time points (7 dpi P<0.0003, 14 dpi P=0.0362).

Figure 7. Cholesterol alters the expression profile of growth factors.

(a) Differentiation time course of OPCs in oligodendroglial enriched cultures in the presence or absence of cholesterol supplementation (bars represent mean of n=4 cultures with individual data points). Drawings illustrate chosen categories of oligodendrocyte differentiation. In each category, significance was assessed by two-way ANOVA and Sidak's post tests. (b) Myelination at 20–28 days in vitro (DIV) in myelinating cocultures in the presence or absence of cholesterol (n=5–9 cultures). Myelin segments and axons were counted (see Supplementary Fig. 8b; two-way ANOVA and Sidak's post tests). (c–h) Quantitative RT-PCR analysis on dissected corpus callosi from mice after ‘induced remyelination' (4+1 weeks) and controls determining the expression of oligodendrocyte and myelin related genes (c; Car2, Plp1, Olig2), marker genes for microglia (Aif1) and astrocytes (Gfap) (d), genes involved in cholesterol synthesis (e; Hmgcr, Fdft1, Srebf2) and uptake (f; Ldlr, Lrp1), and growth factors downregulated (g; Pdgfa, Fgf2) and upregulated by cholesterol supplementation (h; Fgf1, Fgf9, Fgf12, Shh, Fgf17, Fgf22). Bars represent the means (n=4 animals) with individual data points (Student's t tests) normalized to untreated control mice (set to 1, grey line). (i) Differentiation of rat oligodendroglial cells in cultures supplemented with FGF1 and FGF2 (concentrations in ng per ml as indicated) in the presence or absence of cholesterol. Bars represent mean percentage of cells in each category of n=3 cultures (two-way ANOVA with Sidak's post test). (j) Proliferation of OPCs in response to growth factors and cholesterol. OPCs were cultured in the presence or absence of the growth factors (100 ng ml⁻¹) FGF1 or FGF2 with or without cholesterol for 24 h. Data are mean EdU positive cells of all oligodendroglial cells±s.e.m. (n=13 (no GF, FGF2) or n=7 (FGF1) cultures of individual rats; Student's t-tests). (k) Quantitative RT-PCR on primary astrocytes treated with cuprizone (cup) with or without cholesterol (chol) supplementation. Bars represent the mean of n=3 independent experiments with individual data points compared with untreated cultures (set to 1, grey line; one-way ANOVA with Sidak's post tests. Asterisks represent significant differences with *P<0.05; **P<0.01; ***P<0.001.

Figure 8. Working model of repair processes influenced by cholesterol.

Working model of nutritional cholesterol mediated repair processes. Cuprizone exposure causes oligodendrocyte loss and demyelination and slow repair (left panel) because of OPC depletion, imbalanced growth factors, and low local availability of cholesterol. In case of nutritional supplementation, cholesterol from the circulation enters the CNS because of increased BBB permeability (red arrows) increasing the local cholesterol availability (1). There, cholesterol rebalances the expression of growth factors and mitogens synthesized e.g. by astrocytes (2). This simultaneously enhances OPC proliferation (3) and opens a window for OPC differentiation. Cholesterol directly facilitates oligodendrocyte differentiation, presumably by relieving cells from time and energy intensive cholesterol synthesis (4). Altogether, these effects provide a ‘fast track' to remyelination and repair (5).