{"id":90,"date":"2019-07-04T02:39:51","date_gmt":"2019-07-04T02:39:51","guid":{"rendered":"http:\/\/wickedsister.evit.com.au\/?p=90"},"modified":"2019-07-04T02:39:51","modified_gmt":"2019-07-04T02:39:51","slug":"diet","status":"publish","type":"post","link":"https:\/\/wickedsister.evit.com.au\/index.php\/2019\/07\/04\/diet\/","title":{"rendered":"Diet"},"content":{"rendered":"\n

In demyelinating diseases such as multiple sclerosis (MS), the failure to remyelinate contributes to axonal damage1<\/a><\/sup>,\n a major factor in persistent disability. Remyelination failure can be \nattributed partially to an insufficient capacity of resident \noligodendrocyte precursor cells (OPC) to proliferate, migrate, \ndifferentiate and initiate myelin membrane growth2<\/a><\/sup>,3<\/a><\/sup>.\n There is now good evidence to implement therapies that combine the \nestablished immunosuppressive treatment of MS with compounds that \nstimulate remyelination and hence may secondarily limit axonal damage4<\/a><\/sup>,5<\/a><\/sup>.\n A number of factors that support differentiation of OPCs have been \nreported recently, some of which are linked to cholesterol metabolism in\n differentiating oligodendrocytes6<\/a><\/sup>,7<\/a><\/sup>,8<\/a><\/sup>,9<\/a><\/sup>.<\/p>\n\n\n\n

Patients with MS have disturbed brain lipid metabolism10<\/a><\/sup>, but serum lipid profiles are in the normal range11<\/a><\/sup>. During active disease and disease progression, total cholesterol levels can rise to the upper limit of the normal range12<\/a><\/sup>,13<\/a><\/sup>,14<\/a><\/sup>,15<\/a><\/sup>.\n Increased dietary intake of cholesterol is assumed to increase serum \ncholesterol and stimulate immunological responses in inflammatory \ndiseases16<\/a><\/sup>.\n However, it is unclear whether the elevated serum cholesterol in MS \npatients (i) contributes to disease progression, (ii) is a consequence \nof acute disease or (iii) reflects an attempt to counterbalance the \npathophysiological manifestation of the disease.<\/p>\n\n\n\n

We previously showed that cholesterol is rate limiting for CNS myelination17<\/a><\/sup> and that nutritional cholesterol supplementation can stimulate developmental myelination in a mouse model of leukodystrophy18<\/a><\/sup>.\n Here, we investigate the effects of increased dietary cholesterol on \ndisease parameters in three distinct mouse models of MS, that is, on (i)\n inflammation and demyelination in experimental autoimmune \nencephalomyelitis (EAE), (ii) remyelination in lysolecithin induced \nlesions and (iii) demyelination and remyelination in the cuprizone \nmodel. High-cholesterol chow does not aggravate clinical symptoms nor \ninflammatory parameters in EAE or alter demyelination in cuprizone \ntreated animals. Rather, we identify a novel function for cholesterol in\n myelin repair in adult mice. Dietary cholesterol modulates the profile \nof growth factors, simultaneously enhancing OPC proliferation and \noligodendrocyte differentiation, thereby facilitating remyelination and \nreducing axonal injury. These data have implications for the treatment \nof demyelinating diseases.Go to:<\/a><\/p>\n\n\n\n

Results<\/h2>\n\n\n\n

Cholesterol supplementation does not affect pathology in EAE<\/h3>\n\n\n\n

To\n test whether elevated serum cholesterol is a biomarker of acute \ninflammatory disease, we induced MOG-EAE and determined serum \ncholesterol at the peak of clinical symptoms, typically 16\u201320 days after\n induction. Surprisingly, in acute EAE, total serum cholesterol was \nreduced to about 75% of normal values (76\u00b12\u2009mg\u2009dl\u22121<\/sup>\u00b1s.e.m. cholesterol in EAE mice compared with 103\u00b12\u2009mg\u2009dl\u22121<\/sup> in untreated controls, n<\/em>=6\u20139, P<\/em><0.0001, Student’s t<\/em>-test). Similar reductions were observed during remission at 28 days after immunization (76\u00b11\u2009mg\u2009dl\u22121<\/sup>\u00b1s.e.m., n<\/em>=18, P<\/em><0.0001 Student’s t<\/em>-test).<\/p>\n\n\n\n

Next,\n we asked whether dietary cholesterol supplementation worsens acute \ninflammatory disease. Unexpectedly, mice on a high-cholesterol chow (5% \nw\/w cholesterol, fat content unchanged) either prophylactically, two \nweeks before inducing MOG-EAE, or therapeutically with onset of clinical\n symptoms, showed similar disease onset (normal chow 12.6\u00b10.3 days; \ncholesterol 12.6\u00b10.4d, n<\/em>=12\u201316), mean clinical scores and body weight, as controls, during the 28 days of monitoring (Fig. 1a,b<\/a>; Supplementary Fig. 1<\/a>). Moreover, high-cholesterol chow did not correct the reduced serum cholesterol (77\u00b18\u2009mg\u2009dl\u22121<\/sup>, n<\/em>=6).\n Correspondingly, at the peak of the clinical symptoms, dietary \ncholesterol did not influence the level of inflammation: \nhistopathological lesions in the lumbar spinal cord white matter as well\n as the immune cell infiltration and characteristics of the \npro-inflammatory milieu were comparable in extent and composition (Fig. 1c<\/a>; Supplementary Fig. 2<\/a>). These findings are in agreement with dietary cholesterol supplementation in the Theiler’s virus model of MS (ref. 19<\/a><\/sup>). Nonetheless, inflammation was slightly ameliorated in cholesterol fed animals in remission, 28d after immunization (Fig. 1<\/a>, Supplementary Fig. 2<\/a>).\n Reduced infiltration of T cells and microglia\/macrophages was \naccompanied by attenuated expression of several pro-inflammatory \nmarkers, such as interferon-\u03b3 (IFN\u03b3), interleukin 17 (IL-17), \ngranulocyte-macrophage colony-stimulating factor (GM-CSF), tumour \nnecrosis factor (TNF), and major histocompatibility complex II (MHCII). \nTaken together, EAE is associated with decreased serum cholesterol that \nis not restored by supplemented cholesterol. Importantly, cholesterol \ndoes not exacerbate disease but even slightly ameliorates inflammation \nduring remission, suggesting it is safe to administer in inflammatory \ndiseases. As cholesterol supplementation promotes developmental \nmyelination18<\/a><\/sup>, these data prompted us to examine cholesterol supplementation in a remyelination paradigm.<\/a>Open in a separate window<\/a>Figure 1<\/a>Dietary cholesterol does not aggravate EAE pathology.<\/strong><\/p>\n\n\n\n

(a<\/strong>) Clinical score of mice with MOG-EAE on normal chow or chow supplemented with 5% cholesterol (n<\/em>=12\u201316\n mice, 2 independent experiments). Start of cholesterol feeding was \nprophylactic, two weeks before immunization. Arrows illustrate the time \npoints of analyses at the peak of clinical symptoms (16\u201318\u2009dpi) and at \nremission (28\u2009dpi). (b<\/strong>) Body weight of experimental animals as in (a<\/strong>)\n assessed from the day of induction of EAE to the end of monitoring \nclinical scores (28 days). Data is expressed as mean weight\u00b1s.e.m. of n<\/em>=12\u201316\n animals. Onset of clinical symptoms was paralleled by a drop in body \nweight, and mice gained weight only after the peak of disease. (c<\/strong>)\n Lesion characteristics were determined on sections of lumbar spinal \ncord from mice fed normal chow or cholesterol enriched chow (n<\/em>=5\n animals, representative images on the left, scales 200\u2009\u03bcm). Luxol fast \nblue-periodic acid-Schiff-hematoxylin (LFB\/PAS) staining was used to \ndetermine the lesion area and number of lesions per section (arrow). \nImmuno-labeling for myelin basic protein (MBP) was used to determine the\n per cent of myelinated area within a lesion (defined in the DAPI \nchannel as clusters of >20 nuclei, marked by arrows). On sections \nimmuno-labeled for APP, the number of axonal speroids (arrows) per \nsquare mm white matter area was counted, as a readout of axonal damage. \nIn remission, unpaired Student’s t<\/em>-test revealed significantly \nless axonal damage in cholesterol fed animals. Sections triple stained \nfor microglia\/macrophages, T cells, and astrocytes (Iba1-CD3-GFAP triple\n immuno-labeling) were used to assess the cellular composition of \nlesions. Unpaired t<\/em>-tests revealed significantly reduced densities of microglia\/macrophages and T cells in cholesterol fed animals (*, P<\/em><0.05). Bars represent mean values with individual data points.<\/p>\n\n\n\n

Cuprizone lowers serum cholesterol and affects BBB integrity<\/h3>\n\n\n\n

Under\n physiological conditions, the blood-brain barrier (BBB) prevents the \npassage of cholesterol from the circulation into the CNS (refs 21<\/a><\/sup>, <\/sup>22<\/a><\/sup>).\n Therefore, we tested whether dietary cholesterol could penetrate the \nCNS in cuprizone fed mice. Surprisingly, BBB integrity was compromised \nin mice treated with cuprizone for 4 weeks, as indicated by \nextravasation of Evans blue dye into the CNS, following systemic \nadministration (Fig. 2a,b<\/a>).\n Systematic evaluation revealed increased BBB permeability during the \nentire treatment period of up to 12 weeks of cuprizone feeding (1.4\u00b10.1 \nfold, n<\/em>=4 P<0.05 Student’s t<\/em>-test).<\/a>Figure 2<\/a>Increased BBB permeability in cuprizone treated mice.<\/strong><\/p>\n\n\n\n

(a<\/strong>)\n Extravasation of Evans blue on sections of the corpus callosum. In \ncontrol animals, Evans blue fluorescence is restricted to blood vessels \nbut extravasates in mice on cuprizone (arrows) (scale, 50\u2009\u03bcm). (b<\/strong>)\n BBB permeability was measured by Evans blue (EB) extravasation in \nbrains of animals fed cuprizone (cup) for 5 weeks on normal chow or \ncholesterol supplemented chow, or in brains of animals with EAE 2d after\n the peak of clinical symptoms (n<\/em>=3 animals). All treatment groups were normalized to untreated control animals (n<\/em>=5) and compared by one way ANOVA (P<\/em><0.0001). Nutritional cholesterol did not influence BBB permeability. Bars represent mean\u00b1s.e.m. (c<\/strong>)\n Extravasation of bodipy-cholesterol. Maximum intensity projection of \nbodipy-cholesterol fluorescence in the corpus callosum (delineated by \ndashed lines) of mice that were kept on cuprizone for 5 weeks in \ncomparison to untreated mice (control) (scale, 50\u2009\u03bcm). (d<\/strong>)\n Quantification of bodipy-cholesterol extravasation after extraction. \nData are expressed as fold changes\u00b1s.e.m. in cuprizone treated mice \ncompared with untreated control animals (n<\/em>=6 mice per group, unpaired Student’s t<\/em>-test, P<\/em><0.0001).<\/p>\n\n\n\n

The extent of extravasation was much smaller than in EAE, likely explaining why previous studies have missed this BBB defect23<\/a><\/sup>,24<\/a><\/sup>,25<\/a><\/sup>,26<\/a><\/sup>.\n Notably, dietary cholesterol did not influence BBB permeability. When \ntested one week after a single injection of bodipy-cholesterol, the \nfluorescence from this cholesterol derivative (its biophysical \nproperties are very similar to unmodified cholesterol27<\/a><\/sup>)\n was readily detectable in the corpus callosum of cuprizone fed mice (in\n contrast to untreated controls) with a pattern typical for an \nintracellular localization, potentially in glial cells (Fig. 2c<\/a>). Quantification of extravasated bodipy-cholesterol revealed a \u223c3-fold increase in comparison to control mice (Fig. 2d<\/a>). Thus, in cuprizone fed mice, peripheral cholesterol can cross the BBB.<\/p>\n\n\n\n

Cuprizone mediated demyelination is unaltered by cholesterol<\/h3>\n\n\n\n

Next we tested whether nutritional cholesterol altered histopathology during the demyelination phase of cuprizone treatment (Fig. 3a<\/a>)25<\/a><\/sup>,28<\/a><\/sup>.\n In the corpus callosum, oligodendrocyte loss and demyelination evolved \nover the same time course in control and cholesterol supplemented mice (Fig. 3b<\/a>),\n leading to almost complete depletion of mature oligodendrocytes after \nfour weeks. In addition, oligodendroglial numbers (Olig2, \noligodendrocyte lineage transcription factor 2 marks OPCs and \noligodendrocytes), astrogliosis (GFAP, glial fibrillary acidic protein) \nand microgliosis (MAC3, macrophage-3 antigen) steadily increased in a \ncomparable manner in both groups, and axonal damage (APP positive \nspheroids, Fig. 3c<\/a>)\n was similar at all time points tested. Taken together, cholesterol \nsupplementation does not interfere with the cuprizone treatment, and \nmature oligodendrocytes do not escape the toxic insult.<\/a>Open in a separate window<\/a>Figure 3<\/a>Cholesterol does not affect cuprizone mediated demyelination.<\/strong><\/p>\n\n\n\n

(a<\/strong>)\n Scheme depicting the time course of demyelination\/remyelination during 6\n week cuprizone feeding (upper panel, based on own results and on other \nstudies25<\/a><\/sup>,28<\/a><\/sup>)\n to the treatment paradigm. To assess the influence of high-cholesterol \nfeeding on demyelination, mice on normal chow or high-cholesterol chow \nadditionally received cuprizone in the diet for between 2 and 5 weeks \n(black bars) after which mice were analysed histologically. (b<\/strong>)\n Representative pictures of the corpus callosum of untreated control \nmice and mice after 5 weeks on cuprizone with the corresponding \nquantification on the right. Assessed were myelination (Gallyas silver \nimpregnation), the number of mature oligodendrocytes (CAII), the number \nof oligodendrocyte lineage cells (Olig2), activated microglia (MAC3) and\n astrocytes (GFAP). Each bar represents the mean value for 3\u20135 (week \n2\u20134) or 9\u201310 (week 5; untreated controls, ctrl) animals per condition \nwith individual data points (scale 100\u2009\u03bcm). (c<\/strong>) APP positive spheroids per mm2<\/sup> in the corpus callosum at the end of 2\u20135 weeks of cuprizone with or without cholesterol supplementation (n<\/em>=3\u20134 animals at 2 and 3 weeks, n<\/em>=4\u20135 at week 4, n<\/em>=6 untreated controls, n<\/em>=9\u201310 at week 5).<\/p>\n\n\n\n

Cholesterol facilitates remyelination and motor learning<\/h3>\n\n\n\n

Next,\n we tested the hypothesis that dietary cholesterol supplementation \nenhances adult remyelination. When mice are continuously exposed to \ncuprizone, an episode of spontaneous repair occurs in the sixth week, \nresulting in marked remyelination (Fig. 4a<\/a>)25<\/a><\/sup>,29<\/a><\/sup>. At this time point, cholesterol neither influenced oligodendrocyte numbers, remyelination nor glial responses (Fig. 4b,c<\/a>).\n However, the density of APP positive axonal spheroids in cholesterol \nfed animals was reduced, suggesting attenuated axonal damage (Fig. 4d<\/a>).<\/a>Open in a separate window<\/a>Figure 4<\/a>Cholesterol facilitates remyelination after chronic cuprizone exposure.<\/strong><\/p>\n\n\n\n

(a<\/strong>)\n Scheme depicting the time course of demyelination\/remyelination during \ncuprizone feeding to the treatment paradigm. To assess the influence of \nhigh-cholesterol feeding on spontaneous remyelination, mice received \ncuprizone in normal chow or chow supplemented with cholesterol for 5, 6 \nor 12 weeks (black bars) after which mice were analysed histologically. (b<\/strong>)\n Evaluation of disease in the corpus callosum of mice that were treated \nwith cuprizone for 5, 6 or 12 weeks on normal chow or chow enriched with\n cholesterol. Corresponding representative pictures of the 12 weeks \ntreatment cohort are on the left. Assessed were myelination (Gallyas \nsilver impregnation), the number of mature oligodendrocytes (CAII), the \nnumber of oligodendrocyte lineage cells (Olig2), activated microglia \n(MAC3) and astrocytes (GFAP). Each bar represents the mean value of 4 \n(week 12) or 8\u201310 (week 5, 6) animals per condition with individual data\n points (scale 100\u2009\u03bcm). (c<\/strong>) Myelinated axons per 10\u2009\u03bcm2<\/sup> in the corpus callosum at the end of 6 and 12 weeks of cuprizone with or without cholesterol supplementation (n<\/em>=4 animals, Two-way ANOVA and Sidak’s post test). (d<\/strong>) APP positive spheroids per mm2<\/sup> in the corpus callosum at the end of 6 and 12 weeks of cuprizone with or without cholesterol supplementation (n<\/em>=3\u20138 animals, Two-way ANOVA and Sidak’s post test). Asterisks represent significant differences with *P<\/em><0.05; **P<\/em><0.01; ****P<\/em><0.0001.<\/p>\n\n\n\n

After\n chronic cuprizone exposure (12 weeks), a second episode of weak and \ntransient remyelination (up to 20% of full myelination) occurs (Fig. 4a<\/a>). However, even if cuprizone is withdrawn at this point, repair is very limited30<\/a><\/sup>.\n Thus, despite a considerable density of OPCs and mature \noligodendrocytes, remyelination is marginal and astrogliosis substantial\n (Fig. 4b<\/a>,\n blue bars at 12 weeks). Remarkably, cholesterol supplementation \nincreased remyelination \u223c1.6-fold as assessed in Gallyas silver \nimpregnated sections (Fig. 4b<\/a>) and in electron micrographs of the corpus callosum (Fig. 4c<\/a>, Supplementary Fig. 4<\/a>). Coupled to this, a similar increase in OPCs and in mature oligodendrocytes was observed (Fig. 4b<\/a>, 12 weeks). In addition, the positive influence of cholesterol was associated with increased body weight (Supplementary Fig. 5<\/a>). Thus, in the context of recurrent depletion of mature oligodendrocytes, cholesterol supplementation enhances tissue repair.<\/p>\n\n\n\n

To\n specifically determine the effect of cholesterol during remyelination, \nwe exposed mice to cuprizone for four weeks to achieve complete \ndemyelination, then withdrew cuprizone to induce remyelination (\u2018induced\n remyelination’) (Fig. 5a<\/a>).\n Mice fed normal chow during the first 7 days after cuprizone withdrawal\n demyelinated further and had only slightly increased oligodendrocyte \ndensities (Fig. 5b<\/a>,\n compare blue bars 4 and 4+1). In contrast, cholesterol supplementation \nfollowing cuprizone withdrawal dramatically increased OPC proliferation \nand augmented Olig2 positive cell density 1.5-fold (Fig. 5b,c<\/a>).\n Densities of newly differentiated TCF4+ PCNA\u2212 (TCF4, also called \nTCF7L2, transcription factor 7-like 2; PCNA, proliferating cell nuclear \nantigen) oligodendrocytes were also increased by cholesterol (Fig. 5d<\/a>, Supplementary Fig. 6<\/a>), similarly as found in actively repairing lesions from patients with MS31<\/a><\/sup>,32<\/a><\/sup>,33<\/a><\/sup>. The resulting 2.7-fold increase in mature oligodendrocytes (Fig. 5b<\/a>,\n time point 4+1) led to a 1.8-fold increase in myelin content on Gallyas\n silver impregnated sections and on electron micrographs (Fig. 5e<\/a>; Supplementary Fig. 4<\/a>).\n Cholesterol supplementation also altered the glial response, leading to\n a \u223c30% increase in astrocytes and \u223c50% reduction in microglial cells (Fig. 5b<\/a>, 4+1). Axonal damage was attenuated to \u223c70% in cholesterol fed animals of controls (Fig. 5f<\/a>).\n These histological signs of repair were associated with a net gain in \nbody weight, occurring within 7 days of cholesterol supplementation and \ncontrasting with weight maintenance in mice fed normal chow (Supplementary Fig. 5<\/a>).\n The beneficial effect of cholesterol persisted, leading to a robust \nincrease in mature oligodendrocytes and myelin content at 2 weeks after \ncuprizone withdrawal (Fig. 5b<\/a>, 4+2); a result that was confirmed on electron micrographs (Fig. 5e<\/a>).<\/a>Open in a separate window<\/a>Figure 5<\/a>Cholesterol facilitates remyelination after cuprizone withdrawal.<\/strong><\/p>\n\n\n\n

(a<\/strong>)\n Scheme depicting the time course of demyelination\/remyelination during \ncuprizone feeding (remyelination after cuprizone withdrawal in purple). \nThe influence of cholesterol on remyelination was assessed by feeding \nmice cuprizone in normal chow for 4 weeks (4, black bars) followed by \n\u2018induced remyelination’ after cuprizone withdrawal for 1 (4+1) or 2 \n(4+2) weeks on normal chow or cholesterol supplemented chow. (b<\/strong>)\n Representative pictures of the corpus callosum of mice after one week \n(4+1) remyelination. Corresponding quantification is on the right also \nincluding values for 2 weeks remyelination (4+2). Assessed were \nmyelination (Gallyas silver impregnation), the number of mature \noligodendrocytes (CAII), the number of oligodendrocyte lineage cells \n(Olig2), activated microglia (MAC3), and astrocytes (GFAP). Each bar \nrepresents the mean value from n<\/em>=4\u20135 (4 and 4+2) or n<\/em>=7 (4+1) animals (scale, 100\u2009\u03bcm; Two-way ANOVA and Sidak’s post test). (c<\/strong>)\n Quantification of proliferating OPCs (PCNA positive Olig2 positive) in \nthe corpus callosum of mice after 4+1 treatment paradigm (4+1) or after \n12 weeks (12) of cuprizone. Each bar represents the mean of n<\/em>=6\u20137 (week 4+1), or n<\/em>=4 (week 12) animals (Student’s t<\/em>-test). (d<\/strong>)\n Quantification of newly differentiated postmitotic oligodendrocytes \n(TCF4 positive, PCNA negative) in the corpus callosum treated as in c<\/strong>). Each bar represents the mean of n<\/em>=6\u20137 (week 4+1), or n<\/em>=4 (week 12) animals (Student’s t<\/em>-test). (e<\/strong>) Myelinated axons per 10\u2009\u03bcm2<\/sup> in the corpus callosum at the end of the 4+1 (n<\/em>=7) and 4+2 (n<\/em>=4) treatment paradigm (two-way ANOVA and Sidak’s post test). (f<\/strong>) APP positive spheroids per mm2<\/sup> in the corpus callosum (4+1 n<\/em>=7; 4+2 n<\/em>=3\u20134 animals, two-way ANOVA and Sidak’s post test). (g<\/strong>) Motor learning as assessed by maximum velocity (Vmax) on a complex wheel (n<\/em>=6\n animals), expressed as per cent of the Vmax on a training wheel (mean \nof the last 7 days before changing to a complex wheel). Statistical \nevaluation of Vmax was done by Two-way ANOVA (cholesterol effect P<\/em><0.0001) and Sidak’s post tests. Asterisks represent significant differences with *P<\/em><0.05; **P<\/em><0.01; ***P<\/em><0.001.<\/p>\n\n\n\n

To\n examine the generality of this response, we investigated whether \ndietary cholesterol enhanced remyelination in another, completely \ndistinct in vivo<\/em> model of remyelination that is accompanied by \nconfined BBB disruption. Localized injection of lysolecithin into the \nventral-lateral spinal cord of adult mice was used to produce focal \ndemyelination. As in the cuprizone model, demyelination was associated \nwith a reduction in serum cholesterol to about 70% of untreated \ncontrols. Further, dietary cholesterol (2% w\/w for 14 days) increased \nserum cholesterol slightly (79\u00b13\u2009mg\u2009dl\u22121<\/sup>\u00b1s.e.m. in cholesterol fed mice compared with 72\u00b16\u2009mg\u2009dl\u22121<\/sup> in chow fed controls, n<\/em>=3\u20135), enhanced remyelination and significantly increased the density of oligodendroglial cells within the lesion (Fig. 6a\u2013c<\/a>). The beneficial effect of cholesterol was also reflected in significantly increased body weight, relative to chow fed mice (Fig. 6d<\/a>).<\/a>Open in a separate window<\/a>Figure 6<\/a>Cholesterol supports remyelination in the lysolecithin model.<\/strong><\/p>\n\n\n\n

(a<\/strong>)\n Representative images of spinal cord sections 14 days post lesion (dpl)\n with 1\u2009\u03bcl 1% lysolecithin in the ventral spinal cord with \nquantification of n<\/em>=5 (cholesterol chow) and n<\/em>=6 (normal chow) animals. Student’s t<\/em>-tests revealed significantly more Olig2 positive oligodendroglial cells within the lesion area (P<\/em><0.0001), and significantly more MBP positive area (P<\/em>=0.027; scales, 100\u2009\u03bcm). (b<\/strong>,c<\/strong>)\n Representative electron micrographs (scale 1\u2009\u03bcm) and quantification of \nmyelin sheath thickness and the portion of remyelinated axons in control\n and cholesterol fed mice at 14 dpl by g-ratio analysis (n<\/em>=3 animals per group). (d<\/strong>)\n Body weight of experimental animals assessed at the day of lesion (day \n0), after 7d and after 14d at the end of the experiment. Shown are the \nmeans\u00b1s.e.m. of n<\/em>=9 (chol chow) to 10 (normal chow) animals. \nTwo-way ANOVA with Sidaks post tests revealed a significant influence of\n cholesterol feeding at both time points (7\u2009dpi P<\/em><0.0003, 14\u2009dpi P<\/em>=0.0362).<\/p>\n\n\n\n

To\n investigate whether the histopathological improvements in cholesterol \nfed animals was associated with improved clinical measures, we returned \nto the \u2018induced remyelination’ paradigm in the cuprizone model (for a \nscheme of experimental paradigm, see Supplementary Fig. 7<\/a>),\n measuring the maximum running velocity (Vmax) on a running wheel. \nFirst, a training wheel with regularly spaced rungs was placed into the \ncages to improve cardiopulmonary and musculoskeletal strength. One week \nafter cuprizone withdrawal, the training wheel was replaced by a complex\n wheel with irregularly spaced rungs to measure bilateral sensorimotor \ncoordination34<\/a><\/sup> (Supplementary Fig. 7<\/a>).\n The Vmax of mice remyelinated on normal chow dropped to about 40% of \nlevels on the training wheel, and did not improve above 75% (Fig. 5g<\/a>).\n In contrast, mice receiving cholesterol supplementation showed a less \nsevere drop in Vmax (to 63%), followed by a steady increase that reached\n the velocity achieved on the training wheel after two weeks. \nImportantly, in control mice (without cuprizone) cholesterol \nsupplementation, did neither influence performance on the training wheel\n nor motor learning (Vmax, run duration, number of runs and running \ndistance on the complex wheel) (Supplementary Fig. 7<\/a>\n and not shown). Hence, cholesterol supplementation enhances repair \nafter demyelination and improves neurological outcomes by supporting \noligodendrocyte proliferation and differentiation, promoting \nremyelination, decreasing microgliosis, and attenuating axonal damage in\n a permissive environment (\u2018induced remyelination’ after cuprizone \nwithdrawal).<\/p>\n\n\n\n

Cholesterol changes the expression profile of growth factors<\/h3>\n\n\n\n

To\n obtain insight into the mechanism by which cholesterol supports the \nsimultaneous expansion of OPC and oligodendrocyte densities, we \nmonitored differentiation of cultured primary oligodendrocytes in \ndefined Sato media, with or without cholesterol supplementation. \nOligodendrocytes differentiated significantly faster in the presence of \ncholesterol, as indicated by expression of differentiation markers and \nmorphological changes (Fig. 7a<\/a>, Supplementary Fig. 8a<\/a>). However, the final stage of maturation after 5d in culture was unchanged, as shown previously18<\/a><\/sup>.\n Similarly, the rate of myelination as measured by MBP (myelin basic \nprotein) positive area per axonal area (SMI31, phosphorylated axonal \nneurofilaments), was increased in spinal cord co-cultures differentiated\n in the presence of cholesterol (Fig. 7b<\/a>, Supplementary Fig. 8b<\/a>); neither the final degree of neurite outgrowth35<\/a><\/sup>\n nor myelination, were influenced by cholesterol. These findings suggest\n that external cholesterol directly facilitates oligodendrocyte \ndifferentiation and the synthesis of myelin membranes.<\/a>Figure 7<\/a>Cholesterol alters the expression profile of growth factors.<\/strong><\/p>\n\n\n\n

(a<\/strong>)\n Differentiation time course of OPCs in oligodendroglial enriched \ncultures in the presence or absence of cholesterol supplementation (bars\n represent mean of n<\/em>=4 cultures with individual data points). \nDrawings illustrate chosen categories of oligodendrocyte \ndifferentiation. In each category, significance was assessed by two-way \nANOVA and Sidak’s post tests. (b<\/strong>) Myelination at 20\u201328 days in vitro<\/em> (DIV) in myelinating cocultures in the presence or absence of cholesterol (n<\/em>=5\u20139 cultures). Myelin segments and axons were counted (see Supplementary Fig. 8b<\/a>; two-way ANOVA and Sidak’s post tests). (c<\/strong>\u2013h<\/strong>)\n Quantitative RT-PCR analysis on dissected corpus callosi from mice \nafter \u2018induced remyelination’ (4+1 weeks) and controls determining the \nexpression of oligodendrocyte and myelin related genes (c<\/strong>; Car2, Plp1, Olig2<\/em>), marker genes for microglia (Aif1<\/em>) and astrocytes (Gfap<\/em>) (d<\/strong>), genes involved in cholesterol synthesis (e<\/strong>; Hmgcr, Fdft1, Srebf2<\/em>) and uptake (f<\/strong>; Ldlr, Lrp1<\/em>), and growth factors downregulated (g<\/strong>; Pdgfa, Fgf2<\/em>) and upregulated by cholesterol supplementation (h<\/strong>; Fgf1, Fgf9, Fgf12, Shh, Fgf17, Fgf22<\/em>). Bars represent the means (n<\/em>=4 animals) with individual data points (Student’s t tests) normalized to untreated control mice (set to 1, grey line). (i<\/strong>)\n Differentiation of rat oligodendroglial cells in cultures supplemented \nwith FGF1 and FGF2 (concentrations in ng per ml as indicated) in the \npresence or absence of cholesterol. Bars represent mean percentage of \ncells in each category of n<\/em>=3 cultures (two-way ANOVA with Sidak’s post test). (j<\/strong>)\n Proliferation of OPCs in response to growth factors and cholesterol. \nOPCs were cultured in the presence or absence of the growth factors \n(100\u2009ng\u2009ml\u22121<\/sup>) FGF1 or FGF2 with or without cholesterol for 24\u2009h. Data are mean EdU positive cells of all oligodendroglial cells\u00b1s.e.m. (n<\/em>=13 (no GF, FGF2) or n<\/em>=7 (FGF1) cultures of individual rats; Student’s t<\/em>-tests). (k<\/strong>)\n Quantitative RT-PCR on primary astrocytes treated with cuprizone (cup) \nwith or without cholesterol (chol) supplementation. Bars represent the \nmean of n<\/em>=3 independent experiments with individual data points\n compared with untreated cultures (set to 1, grey line; one-way ANOVA \nwith Sidak’s post tests. Asterisks represent significant differences \nwith *P<\/em><0.05; **P<\/em><0.01; ***P<\/em><0.001.<\/p>\n\n\n\n

In\n principle, a substantial induction of OPC differentiation could be \nunfavourable, if it occurs at the expense of OPC numbers. Indeed, \ngradual depletion of OPCs was observed in cholesterol supplemented \noligodendroglial cultures (Fig. 7a<\/a>, left panel). Thus, the expansion of proliferative OPCs in vivo<\/em> (Fig. 5c<\/a>)\n is likely an indirect consequence of additional factors from the local \nenvironment. To identify factors that mediate cholesterol dependent OPC \nproliferation, we analysed another cohort of mice in the \u2018induced \nremyelination’ treatment paradigm (4+1 weeks), using quantitative RT-PCR\n on dissected corpus callosi. In agreement with our histological data, \noligodendrocyte related genes were (i) strongly downregulated in \ncuprizone fed mice in comparison to untreated controls (grey line) and \n(ii) significantly enhanced in cholesterol fed animals in comparison to \nchow fed animals (Fig. 7c<\/a>, compare Fig. 5<\/a>, Supplementary Table 1<\/a>). Similarly, the astrogliosis (Gfap<\/em>) and diminished microgliosis (Aif1<\/em>, allograft inflammatory factor 1) were also reflected in the expression levels of respective marker genes (Fig. 7d<\/a>).\n Surprisingly, cholesterol supplementation did not lead to feedback \ninhibition of cholesterol synthesis, but rather, increased the \nexpression of genes involved in cholesterol synthesis and uptake (Fig. 7e,f<\/a>), likely indicating enhanced remyelination. In contrast, expression of LXR family genes, which influence OPC differentiation36<\/a><\/sup>, was not affected by cholesterol (Supplementary Table 1<\/a>).<\/p>\n\n\n\n

The expression of growth factors involved in OPC survival, proliferation, migration or differentiation28<\/a><\/sup>, including Igf1<\/em> (insulin-like growth factor), Cntf<\/em> (ciliary neurotrophic factor), Inhba<\/em> (inhibin beta-A, also called activin beta-A) and Egf<\/em>\n (epidermal growth factor) was strongly increased (2\u201320 fold) by \ncuprizone, but was not further regulated by cholesterol supplementation (Supplementary Table 1<\/a>). A set of genes whose products are known to inhibit differentiation of OPCs, such as Fgf2<\/em> (fibroblast growth factor 2) and Pdgfa<\/em> (platelet derived growth factor alpha)37<\/a><\/sup>,38<\/a><\/sup>,39<\/a><\/sup>,\n was also strongly upregulated by cuprizone (8\u201312 fold higher than \nuntreated controls). Strikingly, the expression of these mitogens was \nattenuated in cholesterol fed animals to levels only 3\u20138 fold higher \nthan in untreated controls (Fig. 7g<\/a>).\n Moreover, in comparison to untreated controls, expression of another \nset of factors, some of which are known to facilitate differentiation of\n oligodendrocytes39<\/a><\/sup>,40<\/a><\/sup>, such as Fgf1<\/em> and Shh<\/em> (sonic hedgehog), was reduced by cuprizone, but strongly elevated by cholesterol supplementation (Fig. 7h<\/a>).\n Expression of FGF receptors (1\u20133) was not influenced by cholesterol \n(data not shown). Demonstrating the generality of these findings (Supplementary Table 2<\/a>),\n cholesterol influenced the profile of growth factor expression in a \nsimilar manner in mice treated chronically with cuprizone (12 weeks, see\n Fig. 4<\/a>).\n In contrast to the \u2018induced remyelination’ paradigm, expression of \nenzymes involved in cholesterol synthesis was reduced in this cohort, \nsuggesting feedback inhibition after remyelination is accomplished (Supplementary Table 2<\/a>).<\/p>\n\n\n\n

To\n determine whether the growth factor expression profile observed in \ncholesterol treated mice might be causally related to the enhanced \nrepair, we tested whether these growth factor combinations directly \nenhance OPC differentiation in vitro<\/em>, a surrogate for remyelination in vivo<\/em>. Indeed, differentiation was enhanced when OPCs were cultured for 3 days in media supplemented with 90\u2009ng\u2009ml\u22121<\/sup> FGF1, 35\u2009ng\u2009ml\u22121<\/sup> FGF2 and cholesterol (exemplifying cuprizone+cholesterol chow), in comparison to 45\u2009ng\u2009ml\u22121<\/sup> FGF1 and 80\u2009ng\u2009ml\u22121<\/sup> FGF2 (exemplifying cuprizone+normal chow) (Fig. 7i<\/a>). These data suggest the changes in growth factor expression are directly contributing to the improved repair.<\/p>\n\n\n\n

Next,\n we cultured OPCs for 24\u2009h in the presence of EdU \n(5-ethynyl-2\u2032-deoxyuridine), a marker of cells in S-phase of the cell \ncycle, to determine whether the proliferative effect of growth factors \nwas modified by cholesterol. Compared with vehicle treated controls, \nFGF2 doubled the number of EdU positive cells (as expected41<\/a><\/sup>), while FGF2 plus cholesterol elicited a threefold increase in this population (Fig. 7j<\/a>),\n suggesting that cholesterol potentiates the effects of FGF2. Indeed, \ncholesterol alone only slightly increased the proportion of EdU+ cells \nin these cultures (Fig. 7j<\/a>). We speculate that, despite attenuated Fgf2<\/em> expression (compare Fig. 7g<\/a>), potentiated FGF2 signalling contributes to the expansion of proliferating OPCs in cholesterol fed animals (compare Fig. 5c<\/a>).<\/p>\n\n\n\n

As\n only relatively few microglial cells are present in the corpus callosum\n of cholesterol fed mice in the \u2018induced remyelination’ paradigm (4+1 \nweeks) and in the \u2018chronic cuprizone’ paradigm (12 weeks), we \nhypothesized that astrocytes contributed principally to the altered \nprofile of growth factors. Indeed, while primary astrocytes \ndownregulated Fgf1<\/em> expression in response to cuprizone, its expression was upregulated in response to cholesterol, irrespective of cuprizone (Fig. 7k<\/a>), correlating with our in vivo<\/em>\n data. Taken together, in the cuprizone model, cholesterol \nsupplementation modulates the expression profile of growth factors, \nrebalancing proliferative and differentiation signals creating a \npermissive environment for repair.Go to:<\/a><\/p>\n\n\n\n

Discussion<\/h2>\n\n\n\n

Cholesterol availability is a prerequisite for myelination17<\/a><\/sup>,42<\/a><\/sup>,43<\/a><\/sup>\n and, as we show here, exogenous cholesterol directly increases the rate\n of OPC differentiation. In agreement with our findings, failure to \nupregulate expression of sterol synthesis enzymes leads to arrested \ndifferentiation of Tcf4<\/em> mutant OPCs, which can partially be rescued by cholesterol supplementation44<\/a><\/sup>. Further, cholesterol synthesis is enhanced during remyelination in mice29<\/a><\/sup> and statin administration (inhibitors of sterol and isoprenoid synthesis) interferes with remyelination in the cuprizone model45<\/a><\/sup>.\n Nonetheless, monotherapy with statins ameliorates clinical scores in \nEAE; an effect associated with decreased CNS infiltration and \ninflammatory activity of T cells, likely reducing demyelination46<\/a><\/sup>,47<\/a><\/sup>,48<\/a><\/sup>.\n The outcomes of studies using statins in MS patients are contradictory,\n probably because of the disparate effects of statins on inflammation \n(beneficial46<\/a><\/sup>,47<\/a><\/sup>,48<\/a><\/sup>) and on remyelination (detrimental45<\/a><\/sup>).\n Accordingly, a recent meta-analysis does not recommend statin treatment\n for relapsing-remitting MS or clinically isolated syndrome49<\/a><\/sup>.\n Hence, we hypothesize that remyelination failure in MS reflects, at \nleast partially, the inability to locally increase the cholesterol \ncontent in demyelinated lesions.<\/p>\n\n\n\n

This hypothesis is \nsupported by the current study. Exogenous cholesterol enters the CNS \nthrough an impaired blood-brain barrier, resulting in enhanced repair \nand an amelioration of the neurological phenotype in two distinct models\n of remyelination. Our data suggest that cholesterol directly \nfacilitates repair by modulating the profile of growth factor \nexpression, promoting OPC differentiation and, together with the mitogen\n FGF2, potentiating OPC proliferation. Importantly, cholesterol \nsupplementation does not exacerbate inflammation in EAE.<\/p>\n\n\n\n

What\n could prevent the increase of cholesterol in OPCs in demyelinated \nlesions? In patients with MS and in models of demyelination, CNS \ncholesterol homoeostasis is destabilized by a variety of mechanisms. \nFirst, expression of enzymes involved in cholesterol synthesis is \nreduced in demyelinated lesions (our study and refs 9<\/a><\/sup>, <\/sup>29<\/a><\/sup>, <\/sup>50<\/a><\/sup>).\n Second, intercellular cholesterol transport in the CNS is perturbed in \npatients with MS, because of reduced abundance of relevant proteins such\n as ApoE (Apolipoprotein E)10<\/a><\/sup>; and in mouse mutants with BBB disruption22<\/a><\/sup>,\n by uncontrolled flux of sterols in and out of the brain. BBB disruption\n has been shown by diffusion MRI in inflammatory diseases of the brain, \nsuch as MS (ref. 51<\/a><\/sup>)\n and, as we demonstrate here for the first time, is also a feature in \nthe cuprizone model. Third, the decrease in serum cholesterol in both \nEAE and cuprizone mouse models, probably contributes to the impairment \nof CNS cholesterol homoeostasis. Whether patients with MS experience a \ndrop in serum cholesterol during acute demyelinating episodes is \nunknown, and its analysis complicated by the fact that the standard \nfirst-line interferon beta treatment itself reduces total serum \ncholesterol52<\/a><\/sup>. Finally, the OPCs in chronically demyelinated lesions in MS50<\/a><\/sup> and mouse models29<\/a><\/sup>,\n fail to upregulate lipid synthesis and differentiate, potentially as a \nconsequence of an imbalance in signalling, as previously hypothesized4<\/a><\/sup>. Here, we demonstrate an imbalance in expression of growth factors in the cuprizone model, in accordance with previous studies28<\/a><\/sup>,53<\/a><\/sup>,54<\/a><\/sup>,55<\/a><\/sup>.\n The growth factor profile associated with cuprizone alone, such as high\n levels of FGF2 and PDGFa, is predicted to facilitate OPC proliferation \nbut impede efficient remyelination, particularly after chronic \ndemyelination.<\/p>\n\n\n\n

Specifically, FGF signalling could critically influence the fate of demyelinated lesions39<\/a><\/sup>,53<\/a><\/sup>,56<\/a><\/sup>,57<\/a><\/sup>.\n FGF signalling comprises a very complex network, including 24 FGF \nfamily members and four different receptors, whose signalling outcome \ndepends on various splice isoforms and on the multifaceted crosstalk \nbetween different pathways58<\/a><\/sup>.\n Here, we focused on the two major FGF members involved in myelination, \nFGF1 and FGF2. FGF2 has been implicated in OPC proliferation, migration \nand inhibition of oligodendrocyte differentiation38<\/a><\/sup>,39<\/a><\/sup>,41<\/a><\/sup>. In the cuprizone model and in patients with MS, FGF2 abundance correlates with the degree of OPC proliferation28<\/a><\/sup>,53<\/a><\/sup>,56<\/a><\/sup>,59<\/a><\/sup>.\n FGF2 is increased in regions of active OPC proliferation and ongoing \nremyelination, such as active lesions or the rim of demyelinated \nlesions, while it is downregulated in remyelinated shadow plaques, and \nit is low in abundance in normal appearing white matter and in the core \nof demyelinated silent lesions59<\/a><\/sup>. We show that cholesterol administration attenuates the overexpression of Fgf2<\/em>\n and other mitogens in the cuprizone model. Surprisingly, this did not \nrestrict proliferation but augmented OPC numbers, likely through synergy\n with cholesterol (see below).<\/p>\n\n\n\n

FGF1 is reduced in active MS lesions60<\/a><\/sup>, and increased expression is only found in remyelinated lesions56<\/a><\/sup>. In contrast to FGF2, FGF1 is not mitogenic for OPCs (our study and refs 39<\/a><\/sup>, <\/sup>56<\/a><\/sup>). Rather, FGF1 accelerates myelination in vitro<\/em>56<\/a><\/sup>,\n to an extent remarkably similar to what we observed in cholesterol \ntreated cultures. FGF1 might support CNS repair by inducing lipid \nsynthesis and secretion by astrocytes61<\/a><\/sup>.\n The crosstalk between FGF signalling and regulation of cholesterol \nmetabolism is supported by the presence of sterol responsive elements \n(consensus sequences for the SREBF2 transcription factor that increases \ncholesterol synthesis) in the Fgf1<\/em> and Fgf2<\/em> promoters (https:\/\/www.genomatix.de\/<\/a>).\n Altered cellular cholesterol levels can modulate signalling pathways, \nas shown for wnt signalling; the cholesterol content regulates the \nrecruitment of different sets of scaffolding\/adaptor proteins to the \nplasma membrane62<\/a><\/sup>. We demonstrate that cholesterol induces the expression of Fgf1<\/em>\n in astrocytes; however, the affected signalling remains enigmatic. In \naddition to astrocytes, other cells such as microglia, OPCs, neurons and\n vascular cells likely participate in growth factor synthesis. In \ncholesterol treated mice, the expression profile of growth factors was \naltered such that the mitogens FGF2 and PDGF\u03b1 were attenuated and \ndifferentiating cues such as FGF1 and Shh were enhanced. Other factors \nthat are unaffected by cholesterol probably also contributed to the \nrepair process.<\/p>\n\n\n\n

Our data reveal a previously unknown function of nutritional cholesterol in adult remyelination (Fig. 8<\/a>\n shows a working model). In response to cuprizone-mediated demyelination\n in mice, the secreted mitogens and growth factors favour the \nproliferation and oppose the differentiation of OPCs, which slows and \nultimately impairs remyelination. Dietary supplementation increases \ncholesterol availability within the demyelinated CNS and this is \nassociated with rebalancing of growth factor expression. The altered \nprofile of growth factors in cholesterol treated mice simultaneously \nfacilitates OPC proliferation and oligodendrocyte differentiation in vivo<\/em>.\n Thus arrested repair can be overcome by increasing the local \navailability of cholesterol which we achieved by nutritional \nsupplementation.<\/a>Open in a separate window<\/a>Figure 8<\/a>Working model of repair processes influenced by cholesterol.<\/strong><\/p>\n\n\n\n

Working\n model of nutritional cholesterol mediated repair processes. Cuprizone \nexposure causes oligodendrocyte loss and demyelination and slow repair \n(left panel) because of OPC depletion, imbalanced growth factors, and \nlow local availability of cholesterol. In case of nutritional \nsupplementation, cholesterol from the circulation enters the CNS because\n of increased BBB permeability (red arrows) increasing the local \ncholesterol availability (1). There, cholesterol rebalances the \nexpression of growth factors and mitogens synthesized e.g. by astrocytes\n (2). This simultaneously enhances OPC proliferation (3) and opens a \nwindow for OPC differentiation. Cholesterol directly facilitates \noligodendrocyte differentiation, presumably by relieving cells from time\n and energy intensive cholesterol synthesis (4). Altogether, these \neffects provide a \u2018fast track’ to remyelination and repair (5).<\/p>\n\n\n\n

We\n envision that moderate concentrations of the mitogen FGF2, when in \nsynergy with cholesterol, potentiate OPC proliferation, and at the same \ntime opens a window for OPC differentiation that is also enhanced by \nincreased levels of pro-differentiation factors. In addition, \ncholesterol might directly facilitate oligodendrocyte differentiation, \nby relieving cells of the burden of establishing the complex time- and \nenergy-intensive anabolic cholesterol pathway. Supplemented cholesterol \ncan directly support myelination by incorporation into myelin membranes,\n as shown previously in a leukodystrophy model18<\/a><\/sup>. The current study suggests that cholesterol provides a \u2018fast track’ to remyelination and repair.<\/p>\n\n\n\n

In\n contrast to the beneficial effect on remyelination, high-cholesterol \nchow (2% cholesterol) has no effect on demyelination and oligodendrocyte\n survival in the cuprizone model, likely because oligodendrocyte loss is\n induced by direct cuprizone-mediated damage, and is concomitant with a \nlow-grade inflammatory cascade involving T cells, astrocytes and \nmicroglia63<\/a><\/sup>.\n Consistent with cholesterol supplementation not exacerbating \ndemyelination in the cuprizone model, high-cholesterol chow (5% \ncholesterol) did not aggravate disease in EAE. Moreover, cholesterol \nsupplementation attenuated axonal damage in both models; during active \nremyelination in the cuprizone model and during remission in EAE. \nLikely, this is secondary to ameliorated disease states and balanced \nexpression of growth factors and pro-inflammatory factors. We found that\n dietary cholesterol slightly ameliorated inflammation in EAE, while, in\n contrast, a high-fat chow aggravates EAE symptoms64<\/a><\/sup>,65<\/a><\/sup>. Medium and long chain fatty acids of the high-fat chow probably modulate T cell differentiation66<\/a><\/sup>.\n Whether fatty acids contribute to the increased disease activity in \nsome MS patients with elevated serum cholesterol remains unclear12<\/a><\/sup>. Importantly, feeding cholesterol appears to be safe in mice with inflammatory disease.<\/p>\n\n\n\n

Taken\n together, our data show that demyelinating disease destabilizes \nperipheral and CNS cholesterol homoeostasis. Dietary cholesterol \nsupplementation supports cholesterol metabolism in the CNS and has the \nremarkable potential to ameliorate disease by facilitating several \nrepair mechanisms, leading to improved remyelination and neurological \noutcome. This study highlights the safety of dietary cholesterol and \nmight have implications for the management of demyelinating diseases, \nbut further studies, especially in combination with immune suppressive \ndrugs, are required to determine its feasibility for patients.Go to:<\/a><\/p>\n\n\n\n

Methods<\/h2>\n\n\n\n

Mice<\/h3>\n\n\n\n

All\n animal studies were performed in compliance with the animal policies of\n the Max Planck Institute of Experimental Medicine, and were approved by\n the German Federal State of Lower Saxony. Adult male C57BL\/6N mice \n(8\u201310 weeks of age) were taken for all analyses. Animals were randomly \nassigned to an experimental group. Mice were fed normal chow (V1124 \nssniff Spezialdi\u00e4ten GmbH, Germany) or chow supplemented with either 2% \nw\/w (cuprizone and lysolecithin experiments) or 5% w\/w cholesterol (EAE \nexperiments).<\/p>\n\n\n\n

For MOG-EAE, mice purchased from Charles \nRiver were immunized subcutaneously with 200\u2009\u03bcg myelin oligodendrocyte \nglycoprotein peptide 35\u201355 (MOG35\u201355) in complete Freund’s adjuvant (M. tuberculosis<\/em> at 3.75\u2009mg\u2009ml\u22121<\/sup>) and i.p. injected twice with 500\u2009ng pertussis toxin as described67<\/a><\/sup>.\n Animals were examined daily and scored for clinical signs of the \ndisease. If disease did not start within 15 days after induction or the \nclinical score rose above 4, animals were excluded from the analysis. \nThe clinical score was: 0 normal; 0.5 loss of tail tip tone; 1 loss of \ntail tone; 1.5 ataxia, mild walking deficits (slip off the grid); 2 mild\n hind limb weakness, severe gait ataxia, twist of the tail causes \nrotation of the whole body; 2.5 moderate hind limb weakness, cannot grip\n the grid with hind paw, but able to stay on a upright tilted grid; 3 \nmild paraparesis, falls down from a upright tiled grid; 3.5 paraparesis \nof hind limbs (legs strongly affected, but move clearly); 4 paralysis of\n hind limbs, weakness in forelimbs; 4.5 forelimbs paralyzed; 5 \nmoribund\/dead. Mice received 5% cholesterol chow commencing either two \nweeks before immunization defined as prophylactic regimen or at the \nfirst appearance of EAE symptoms defined as therapeutic regimen and \ncontinued until day 28.<\/p>\n\n\n\n

For cuprizone experiments, mice\n were fed 0.2% w\/w cuprizone (Sigma-Aldrich Inc., Germany) in powder \nchow with or without cholesterol for \u2018demyelination’ (2\u20135 weeks) and \n\u2018chronic cuprizone’ (6 and 12 weeks) paradigms. For \u2018induced \nremyelination’ experiments, mice were fed cuprizone in standard chow for\n 4 weeks, followed by cuprizone withdrawal and feeding mice standard \nchow with or without cholesterol supplementation. Mice were fed three \ntimes a week an exceeding amount of chow by dispenser. Food intake and \nanimal weight was monitored. Age-matched untreated controls were fed \nstandard powder chow.<\/p>\n\n\n\n

Focal spinal cord demyelinating \nlesions were induced under anaesthesia by stereotactic injection of 1\u2009\u03bcl\n lysolecithin (1%, from egg yolk, alpha-lysophosphatidylcholine, Sigma) \ninto the ventro-lateral funiculus at Th10 of 8-week old animals, as \npreviously described68<\/a><\/sup>. The injection was performed with a 10\u2009\u03bcl Hamilton syringe, fitted with a thin tapered glass tip, at a rate of \u223c1\u2009\u03bcl\u2009min\u22121<\/sup>.\n This procedure created fusiform demyelinating lesions, 5\u20136\u2009mm in \nlength. At the day of injection, mice were randomly assigned to normal \nor 2% cholesterol chow for 14 days, after which the animals were killed,\n and the spinal cord processed for histology.<\/p>\n\n\n\n

Bodipy-cholesterol injections were done as described18<\/a><\/sup>. Briefly, bodipy-cholesterol (Topflour, Avanti Polar Lipids) was injected i.p. (16\u2009\u03bcg\u2009g\u22121<\/sup>\n body weight). After one week, mice were perfused, and \nbodipy-cholesterol fluorescence was analysed on vibratome sections using\n a custom made two-photon laser scanning microscope equipped with a \ntitanium-sapphire laser and a \u00d7 20 water immersion objective (NA 1.0). \nZ-stacks of 100-\u03bcm depth were obtained and processed to maximum \nintensity projections. For tracer quantification of bodipy-cholesterol \n(5\u2009\u03bcg\u2009g\u22121<\/sup> body weight, i.p. injection 7d circulation time) or Evans blue (50\u2009\u03bcg\u2009g\u22121<\/sup>\n body weight, i.v. injection, 4\u2009h circulation time) animals were \nperfused with PBS to remove tracer from the circulatory system. Brains \nwere dissected and immediately frozen on dry ice, weighed and stored at \n\u221280\u2009\u00b0C for further processing. Tissue was lyophilized (Christ LMC-1 BETA\n 1-16) at \u201336\u2009\u00b0C for 24\u2009h under vacuum of 0.2\u2009mBar. For tracer \nextraction, hemispheres were incubated shaking in 10\u2009\u03bcl formamide per mg\n brain at 57\u2009\u00b0C for 24\u2009h. Integrated density of tracer fluorescence was \ndetermined in triplicates on a fluorescent microscope (Observer Z2, \nZeiss, Germany), equipped with an AxioCam MRc3, \u00d7 1 Camera Adaptor and \nthe ZEN 2012 blue edition software recorded at \u00d7 10 magnification \n(Plan-Apochromat \u00d7 10\/0.45 M27). Tracer concentration was calculated \nusing a standard curve and normalized to matched controls (set to 1).<\/p>\n\n\n\n

Motor skill performance was assessed essentially as described34<\/a><\/sup>. Mice were randomly divided into two treatment and two control groups (n<\/em>=6\u201313)\n and housed in individual cages that allow computer-controlled recording\n of wheel rotation as a function of time (MatLab-based custom software).\n The axis of each wheel was attached to a rotation sensor with a \nresolution of 16 per turn. One wheel revolution comes up to a running \ndistance of 35.5\u2009cm. The running wheel revolutions were recorded \ncontinuously at a sampling rate of 1\/0.48\u2009s by a customized recording \ndevice and software (Boenig & Kallenbach oHG, Dortmund, Germany). \nMice in treatment groups were treated as in the \u2018induced remyelination’ \nparadigm (feeding 4 weeks cuprizone in normal chow followed by \nwithdrawal of cuprizone and feeding normal chow or cholesterol \nsupplemented chow). Control animals received normal chow for the entire \nexperiment or were switched to cholesterol chow after 4 weeks. One week \nafter the start of the experiment training wheels with regularly spaced \nrungs were placed into the cages for adaptation of cardiopulmonary and \nmusculoskeletal strength. One week after the switch of diets (week 5 of \nexperiment), wheels were replaced by complex wheels with irregularly \nspaced rungs to assess the bilateral sensorimotor coordination that \nlikely involves the cerebellum and motor cortex and connecting white \nmatter such as the corpus callosum. Specifically, we measured maximum \nrunning velocity (Vmax<\/sub>), in addition to the total maximum run duration (Dmax<\/sub>), accumulative distance in metres (Distac<\/sub>) and the number of individual runs (Nrun<\/sub>). Parameters were logged once daily (12 am).<\/p>\n\n\n\n

Serum analyses<\/h3>\n\n\n\n

Animals\n were fasted for 4\u2009h, blood was collected from the retroorbital sinus, \nand serum was prepared after clotting by centrifugation. Cholesterol \nmeasurements were done with the architectII system (Abbott Diagnostics).<\/p>\n\n\n\n

Cell isolation and flow cytometry<\/h3>\n\n\n\n

Single-cell\n suspensions from spinal cords were obtained via mechanical dissociation\n on a cell strainer. Immune cells were separated over a two-phase \nPercoll-density gradient. Staining of \u03b1\u03b2TCR\/CD4+<\/sup> T cells, \u03b1\u03b2TCR\/CD8+<\/sup>\n T cells and CD45\/CD11b cells (macrophages\/microglia) was performed \nusing the following antibodies in a 1:200 dilution: Anti-CD3e (clone \n145-2C11), BioLegend; anti-CD4 (clone GK 1.5), BD; anti-CD8 (clone \n53-6.7), BD; anti-CD8 (clone 53\u20136.7), BD; anti-CD11b (clone M1\/70), \nBioLegend; anti-CD45.2 (clone 104), BioLegend. The addition of Calibrite\n APC beads (BD) allowed for cell quantification. Flow cytometry was \nperformed using a FACSCalibur operated by Cell Quest software (Becton \nDickinson).<\/p>\n\n\n\n

Histochemistry<\/h3>\n\n\n\n

Anesthetized\n mice were perfused with 4% formaldehyde (PFA). Brain samples of \ncuprizone treated animals were cut at Bregma 1.58 for comparable \npathology because the extent of cuprizone mediated demyelination \nstrongly depends on the rostral\/caudal position69<\/a><\/sup>.\n Tissue was postfixed overnight, embedded in paraffin and cut into 5\u2009\u03bcm \nsections (HMP 110, MICROM). Gallyas silver impregnation was done as \ndescribed18<\/a><\/sup>. For immunohistological analyses, sections were deparaffinized followed by antigen-retrieval in sodium citrate buffer (0.01\u2009M,\n pH 6.0). For immunofluorescence, sections were blocked with serum free \nprotein block (Dako). Primary antibodies were diluted in 2% bovine serum\n albumin (BSA)\/PBS and incubated for 48\u2009h followed by fluorophor coupled\n secondary antibodies. For immunohistochemistry, endogenous peroxidase \nactivity was blocked with 3% hydrogen peroxide. Sections were then \nblocked (20% goat serum in BSA\/PBS) and incubated with primary \nantibodies. Detection was done with the LSAB2 kit (Dako, Hamburg, \nGermany) or the Vector Elite ABC kit (Vector Labs). HRP substrate \n3,3\u2032-Diaminobenzidine (DAB) was applied by using the DAB Zytomed Kit \n(Zytomed Systems GmbH). Haematoxylin stain was done to label nuclei. \nSections were dehydrated before mounting (Eukitt). Specimens were \nanalysed on an Axio Imager.Z1 (Zeiss) equipped with an AxioCam MRc3, \u00d7 \n0.63 Camera Adaptor and the ZEN 2012 blue edition software using \u00d7 10 \nobjective (Plan Apochromat \u00d7 10\/0.45 M27) or \u00d7 20 objective \n(Plan-Apochromat \u00d7 20\/0.8) and evaluated with Image J software. \nQuantification of areas (Gallyas, GFAP, MAC3) were done by applying \nsemi-automated ImageJ software macro to threshold (variable threshold in\n case of Gallyas and fixed threshold for antibody stainings) and colour \ndeconvolute the images of the corpus callosum above the fornix (Bregma \n1.58). Three to five sections per animal were analysed. Quantification \nof EAE lumbar spinal cord lesions was done on two to four quadruple \nstained sections (Iba1 (induction of brown adipocytes 1), CD3, GFAP, \nDAPI) per animal recorded with tile region setup and shading correction.\n Lesion area was defined by focal accumulation of at least 20 DAPI \npositive cells, the presence of microglia and infiltration of CD3 \npositive cells. Lesion area, number of Iba1 and CD3 positive cells and \nGFAP positive area were evaluated. Quantification of cuprizone treated \nanimals: cell number (CAII (carbonic anhydrase 2), Olig-2, TCF4, PCNA), \nAPP positive spheroids and area (Gallyas, GFAP, MAC3) was done in the \ncorpus callosum above the fornix (Bregma 1.58). Three to five sections \nper animal were analysed. Microscope settings are listed in Supplementary Table 4<\/a>.<\/p>\n\n\n\n

Electron microscopic analysis was done as previously described17<\/a><\/sup>. Briefly, tissue was fixed in 4% PFA, 2.5% Glutaraldehyde, 0.1\u2009M\n Phosphate buffer and sagittal sections were cut on a vibratome (Leica \nVT1200, 300\u2009\u03bcm). The corpus callosum with adjacent tissue (\u22120.04\u2009mm \nlateral) was punched with a 2\u2009mm diameter punching tool and embedded in \nepon (EMTP, Leica). At least 15 digital pictures (\u00d7 12,000 \nmagnification, TRS, Moorenweis) of uranyl acetate contrasted ultrathin \nsections were taken with the Zeiss EM900.<\/p>\n\n\n\n

Antibodies<\/h3>\n\n\n\n

The\n following antibodies were used: APP (Chemicon MAB348), CAII (Said \nGhandour); CD3 (Serotec MCA1477);CD3e (Biolegend clone 145-2C11), CD4 \n(Becton Dickinson clone GK1.5), CD8 (Becton Dickinson clone 53-6.7), \nCD11b (Biolegend clone M1\/70), CD45.2 (Biolegend clone 104), CNP \n(2\u2032,3\u2032-Cyclic-nucleotide 3\u2032-phosphodiesterase, Sigma C5922), GFAP \n(Chemicon MAB3402), Iba1 (Wako 019-19741), MAC3 (Pharmigen 01781D); MBP \n(Serotec MCA409S), Olig2 (Prof Charles Stiles\/ Dr. John Alberta, DF308),\n PCNA (Abcam ab29), SMI31 (Covance SMI-31P), TCF4 (Millipore 04-1080).<\/p>\n\n\n\n

Expression analyses<\/h3>\n\n\n\n

For\n the characterization of the proinflammatory milieu in EAE mice, RNA \nfrom total spinal cord lysates was isolated using Trizol (Thermo \nFisher). complementary DNA (cDNA) was synthesized using RevertAid First \nStrand cDNA Synthesis Kit (Thermo Fisher) according to the \nmanufacturer’s protocol. Quantitative RT-PCR was performed using a \nStepOnePlus Real-Time PCR System operated by StepOnePlus Software v2.0. \nTarget-specific FAM- and TAMRA-labeled TaqMan probes were used in all \ncases. Measurements were performed in independent duplicates. Gene \nexpression was normalized to \u03b2-actin. Relative changes in gene \nexpression were analysed via the 2\u0394\u0394C<\/em>(T) method.<\/p>\n\n\n\n

For\n expression analyses on brain sections, mice were killed by cervical \ndislocation and brains were quickly cooled and sliced coronally using a \nbrain matrix (Asi-Instruments). The corpus callosum was dissected from \nBregma +1.10 to \u22122.46 and RNA was extracted using RNeasy Mini (Qiagen). \nThe concentration and quality of RNA was evaluated using a NanoDrop \nspectrophotometer and RNA Nano (Agilent). cDNA was synthesized with \nSuperscript III (Invitrogen) and quantitative PCRs were done in \ntriplicates with the GoTaq master mix (Promega) on a 7500 Fast Real-Time\n PCR System (Applied Biosystems). Expression values were normalized to \nthe geometric mean of two housekeeping genes, Hprt \n(Hypoxanthin-Phosphoribosyl-Transferase 1) and Rplp0 (60S acidic \nribosomal protein P) and analysed by the \u0394\u0394Ct<\/em> method.<\/p>\n\n\n\n

Expression of the following genes was measured: Abca1 (ATP-binding cassette transporter A1), Actb<\/em> (beta actin), Aif1<\/em> (allograft inflammatory factor 1), Apoe<\/em> (apolipoprotein E), Bdnf<\/em> (Brain-derived neurotropic factor), Bmp2 and Bmp4 (Bone morphogenic protein 2 and 4), Car2<\/em> (carbonic anhydrase 2), Ch25h (Cholesterol 25-Hydroxylase ), Cntf<\/em> (ciliary neurotrophic factor), Cyp27a1<\/em> (Sterol 27-hydroxylase ), Cyp46a1<\/em> (Cholesterol 24-hydroxylase), Cyp51a1<\/em> (Sterol 14 alpha-demethylase), Dhcr24<\/em> (24-Dehydrocholesterol reductase), Egf<\/em> (epidermal growth factor), Fdft1<\/em> (Farnesyl-Diphosphate Farnesyltransferase 1), members of the Fgf<\/em>\n (fibroblast growth factor) gene family, Gfap (glial fibrillary acidic \nprotein), Gmcsf (granulocyte-macrophage colony-stimulating factor, \nCSF2), H2-DMb2<\/em> (MHCII, major histocompatibility complex class II), Hmgcr<\/em> (3-Hydroxy-3-Methylglutaryl-CoA Reductase), Hmgcs1<\/em> (3-Hydroxy-3-methylgutaryl-CoA synthase 1), Ifng<\/em> (interferon gamma), Igf1<\/em> (insulin-like growth factor 1), Il2<\/em> (interleukin 2), Il17<\/em>, Inhba<\/em> (inhibin beta-A, also called activin beta-A), Ldlr<\/em> (Low density lipoprotein receptor), Lrp1<\/em> (Low density lipoprotein receptor-related protein 1), Mvk<\/em> (Mevalonate kinase), Ntf3<\/em> (Neurotrophin 3), Ngf<\/em> (Nerve growth factor ), Nr1h3<\/em> (Liver X receptor alpha, LXR alpha), Nr1h2<\/em> (Liver X receptor beta, LXR beta), Olig2<\/em> (oligodendrocyte lineage transcription factor 2), Pdgfa<\/em> (platelet derived growth factor alpha), Plp1<\/em> (proteolipid protein 1), Ptn (Pleiotrophin), Rxrg<\/em> (Retinoic X receptor gamma, RXR gamma), S100b<\/em> (S100 calcium-binding protein B), Shh<\/em> (sonic hedgehog), Srbf2<\/em> (sterol regulatory element binding transcription factor 2), Tnf<\/em> (tumour necrosis factor), Vldlr (Very low density lipoprotein receptor). All primer sequences are listed in Supplementary Table 3<\/a>.<\/p>\n\n\n\n

Cell cultures<\/h3>\n\n\n\n

For primary oligodendrocyte cultures, dissected cortices of newborn mice or rats were digested in 0.25\u2009mg\u2009ml\u22121<\/sup> Trypsin\/EDTA for 10\u2009min followed by triturating and plating in plating media (DMEM 4.5\u2009g\u2009l\u22121<\/sup> Glucose, 10% fetal calf serum containing about 300\u2009\u03bcg\u2009ml\u22121<\/sup>\n cholesterol, GlutaMAX, penicillin\/streptomycin). About 14 days after \nplating, OPCs were isolated by differential shaking and lectin panning, \nand plated in differentiating Sato media (DMEM 4.5\u2009g\u2009l\u22121<\/sup> glucose, 4\u2009mM glutamine, 5\u2009\u03bcg\u2009ml\u22121<\/sup> insulin, 16\u2009\u03bcg\u2009ml\u22121<\/sup> putrescine, 6.2\u2009ng\u2009ml\u22121<\/sup> progesterone, 5\u2009ng\u2009ml\u22121<\/sup> sodium selenite, 400\u2009ng\u2009ml\u22121<\/sup>\nL-thyroxine, 400\u2009ng\u2009ml\u22121<\/sup> triiodothyroxine, 50\u2009\u03bcg\u2009ml\u22121<\/sup>\n holo-transferrin, penicillin\/streptomycin, lacking any cholesterol \nsource). Cell purity was routinely determined by immune stainings and \nalways exceeded 95%. Cholesterol (10\u2009\u03bcg\u2009ml\u22121<\/sup>) was added from a 10\u2009mg\u2009ml\u22121<\/sup>\n stock solution in ethanol. Control cultures received 0.1% ethanol. In \ncase of growth factor supplementation assays, cells were allowed to \nadhere for 2\u2009h before treatment with EdU (5-ethynyl-2\u2032-deoxyuridine, \n10\u2009\u03bcM, Invitrogen) and growth factors (FGF1, FGF2; Peprotech) at \n100\u2009ng\u2009ml\u22121<\/sup> for proliferation experiments or 45\u2009ng\u2009ml\u22121<\/sup> FGF1 plus 80\u2009ng\u2009ml\u22121<\/sup> FGF2 or 90\u2009ng\u2009ml\u22121<\/sup> FGF1 plus 35\u2009ng\u2009ml\u22121<\/sup> FGF2 for differentiation experiments. Cultures were fixed with PFA and permeabilized with 0.5% Triton X100.<\/p>\n\n\n\n

Myelinating co-cultures were established as described70<\/a><\/sup>\n with minor modifications. Briefly, six E13 embryonic spinal cords per \nculture were digested in 0.125% Trypsin solution in HBSS (without Ca+2<\/sup> and Mg+2<\/sup>) at 37\u2009\u00b0C for 20\u2009min. After stopping the digestion with 1\u2009ml plating media (DMEM, 25% horse serum, 25% HBSS, 50\u2009\u03bcg\u2009ml\u22121<\/sup> DNAse) the tissue was homogenized by gentle trituration and centrifuged for 5\u2009min. 150,000 cells were plated per poly-L-lysine\n coated coverslip; 3 coverslips per 35\u2009mm Petri dish. After cell \nattachment in plating media, differentiation media was added (low \nglucose DMEM, 10\u2009\u03bcg\u2009ml\u22121<\/sup> insulin, 10\u2009ng\u2009ml\u22121<\/sup> biotin, 50\u2009nM hydrocortisone, 0.5% N1-mix). N1 mix was 1\u2009mg\u2009ml\u22121<\/sup>\n apo-transferrin, 20\u2009mM putrescine, 4\u2009\u03bcM progesterone, and 6\u2009\u03bcM sodium \nselenite. 50% media change was performed every 24\u201348\u2009h with \ndifferentiation media. After 12 days, insulin was removed from \ndifferentiation media. Coverslips were fixed with PFA after 20, 24 and \n28 days in culture and permeabilized with \u221220\u2009\u00b0C methanol for 10\u2009min.<\/p>\n\n\n\n

Fixed\n and permeabilized cells were blocked with 10% horse serum in PBS and \nincubated with primary antibodies in blocking solution followed by \nsecondary antibodies together with click-it kit for detection of EdU and\n DAPI for nuclear staining. Coverslips were mounted on slides with aqua \npolymount. On five randomly chosen visual fields of primary \noligodendrocyte cultures (\u00d7 10 magnification), stainings were evaluated.\n For differentiation of oligodendrocytes, CNP and MBP positive cells \nwere categorized according to morphological criteria (see Fig. 7a<\/a>).\n New oligodendrocytes (new OL) were CNP-positive MBP-negative cells with\n complex processes. MBP+OL cells contained few MBP-positive \nintracellular spots but did not form sheaths. Mature oligodendrocytes \nwere CNP- and MBP-positive with sheaths. In myelinating co-cultures, the\n axonal (SMI31) area and the area with myelin sheaths (MBP) of seven \nrandomly chosen visual fields of myelinating co-cultures (\u00d7 10 \nmagnification) was measured after binarization of thresholded images. \nSpecimens were analysed on an Axiophot observer.Z1 (Zeiss) equipped with\n an AxioCam MRm and the ZEN 2012 blue edition software and evaluated \nwith Image J software. Microscope settings are listed in Supplementary Table 4<\/a>.<\/p>\n\n\n\n

Statistical analyses<\/h3>\n\n\n\n

Statistical evaluation was done by unpaired Student’s t<\/em>-test\n for pairwise comparisons or by ANOVA for comparisons of more than two \ngroups as stated in the figure legends. Two-way ANOVA was combined with a\n post test to evaluate individual groups. For all statistical tests, \nsignificance was measured against an alpha value of 0.05. All error bars\n show s.e.m. P<\/em> values are shown as *P<\/em><0.05; **P<\/em><0.01; ***P<\/em><0.001.\n No statistical methods were used to predetermine sample sizes, but our \nsample sizes are similar to those reported in previous publications25<\/a><\/sup>,28<\/a><\/sup>. Data analysis was performed blind to the experimental groups.<\/p>\n\n\n\n

Data availability<\/h3>\n\n\n\n

All data generated or analysed during this study are included in this published article (and its supplementary information<\/a> files) or available from the authors on request.Go to:<\/a><\/p>\n\n\n\n

Additional information<\/h2>\n\n\n\n

How to cite this article:<\/strong> Berghoff, S. A. et al<\/em>. Dietary cholesterol promotes repair of demyelinated lesions in the adult brain. Nat. Commun.<\/em>\n8,<\/strong> 14241 doi: 10.1038\/ncomms14241 (2017).<\/p>\n\n\n\n

Publisher’s note:<\/strong> Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Go to:<\/a><\/p>\n\n\n\n

Supplementary Material<\/h2>\n\n\n\n

Supplementary Information: <\/strong><\/p>\n\n\n\n

Supplementary Figures, Supplementary Tables and Supplementary ReferencesClick here to view.<\/a>(39M, pdf)<\/sup>Go to:<\/a><\/p>\n\n\n\n

Acknowledgments<\/h2>\n\n\n\n

We\n are grateful to Klaus-Armin Nave for constant support. We cordially \nthank Annette Fahrenholz, Tanja Freerck, Boguslawa Sadowski and Lennart \nWiegand for technical support. We thank Prof Charles Stiles, Dr John \nAlberta and Prof Said Ghandour for generous gifts of antibodies. This \nwork was funded by the Deutsche Forschungsgemeinschaft (SA 2014\/2-1 to \nG.S.). J.E. and W.M. were funded by an ERC Advanced grant awarded to \nKlaus-Armin Nave.Go to:<\/a><\/p>\n\n\n\n

Footnotes<\/h2>\n\n\n\n

The authors declare no competing financial interests.<\/p>\n\n\n\n

Author contributions<\/strong>\n G.S., S.A.B., J.M.E., D.L. and W.M. participated in the planning and \ndesigning the experiments. S.A.B. was involved in performing all \nexperiments. N.G. and J.M.E. performed cell culture experiments, S.K.S.,\n P.D. and J.W., characterized the mouse pathologies, C.B. performed \nexpression analyses, B.B. was involved in EAE experiments, L.H. and F.O.\n did flow cytometry with subsequent expression analyses, T.R. performed \nelectron microscopy analysis, W.M. edited the manuscript, S.A.B., J.M.E.\n and G.S. wrote the manuscript.<\/p>\n","protected":false},"excerpt":{"rendered":"

In demyelinating diseases such as multiple sclerosis (MS), the failure to remyelinate contributes to axonal damage1, a major factor in persistent disability. Remyelination failure can be attributed partially to an insufficient capacity of resident oligodendrocyte precursor cells (OPC) to proliferate, migrate, differentiate and initiate myelin membrane growth2,3. There is now good evidence to implement therapies…<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-90","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/posts\/90","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/comments?post=90"}],"version-history":[{"count":1,"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/posts\/90\/revisions"}],"predecessor-version":[{"id":91,"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/posts\/90\/revisions\/91"}],"wp:attachment":[{"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/media?parent=90"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/categories?post=90"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wickedsister.evit.com.au\/index.php\/wp-json\/wp\/v2\/tags?post=90"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}

We\n first tested whether serum cholesterol was altered in the cuprizone \nmodel of demyelinating disease (see also below). Surprisingly, after 4 \nweeks on cuprizone, mice had markedly reduced total serum cholesterol \n(76\u00b13\u2009mg\u2009dl\u22121<\/sup>\u00b1s.e.m. in comparison to 103\u00b13\u2009mg\u2009dl\u22121<\/sup> in controls, n<\/em>=9\u201313, P<\/em><0.0001, Student’s t<\/em>-test). Although liver function values were normal (Supplementary Fig. 3<\/a>), we cannot exclude the possibility that this is due in part to altered liver metabolism20<\/a><\/sup>. In contrast to EAE, dietary supplementation with 2% w\/w cholesterol normalized total serum cholesterol (106\u00b15\u2009mg\u2009dl\u22121<\/sup>, n<\/em>=13).<\/p>\n\n\n\n