Multiple Sclerosis Resource Centre

Welcome to the Multiple Sclerosis Resource Centre. This website is intended for international healthcare professionals with an interest in Multiple Sclerosis. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Cholesterol and markers of cholesterol turnover in multiple sclerosis: relationship with disease outcomes

Multiple Sclerosis and Related Disorders, Volume 5, January 2016, Pages 53–65


Multiple sclerosis (MS) is a chronic central nervous system disease that is associated with progressive loss of myelin and subsequent axonal degeneration. Cholesterol is an essential component of mammalian cellular and myelin membranes. In this systematic review, we examined the relationship between levels of cholesterol and markers of cholesterol turnover in circulation and/or cerebrospinal fluid (CSF) and disease outcomes in adults with clinically isolated syndrome (CIS) or confirmed MS. Studies suggest that elevated levels of circulating low density lipoprotein cholesterol (LDL), total cholesterol, and particularly, apolipoprotein B and oxidized LDL are associated with adverse clinical and MRI outcomes in MS. These relationships were observed as early as CIS. The studies also suggest that oxysterols, cholesterol precursors, and apolipoprotein E may be markers of specific disease processes in MS, but more research is required to elucidate these processes and relationships. Taken together, the data indicate that cholesterol and markers of cholesterol turnover have potential to be used clinically as biomarkers of disease activity and may even be implicated in the pathogenesis of MS.


  • We examine 21 studies of cholesterol and cholesterol metabolites in MS progression.
  • Elevated total cholesterol, LDL, ox-LDL, ApoB are associated with adverse outcomes.
  • Further work is needed to address possible associations between ApoE or ApoD and MS outcome.

Keywords: Cholesterol, Biomarker, Multiple sclerosis, Clinically isolated syndrome, Apolipoprotein, ApoA, ApoB, ApoD, ApoE, Lipoprotein, LDL, HDL, Oxysterol, 24-hydroxycholesterol.

1. Introduction

Multiple sclerosis (MS) is a chronic central nervous system (CNS) disease characterized pathologically by inflammation, loss of myelin, and axonal loss. These pathological features translate into significant clinical disability and a highly variable and unpredictable disease course. There is therefore a need to develop readily accessible biomarkers that relate to disease evolution and outcomes and that enhance our understanding of specific disease processes in MS (Comabella and Montalban, 2014 and Teunissen et al, 2005).

Cholesterol is as an essential component of cellular and myelin membranes, a cofactor for signalling molecules, and a precursor of steroid hormones, bile acids and vitamin D (Berg et al., 2002). Cholesterol homeostasis is compartmentalized, with only limited interaction between brain/CSF and circulation (Di Paolo and Kim, 2011). The first investigation into the role of cholesterol in the pathophysiology of MS was performed in 1926 (Poynder and Russell, 1926) – at a time when its presence in CSF was believed to reflect the “wasting of nerve structures” (Mott, 1910; reviewed in Plum, 1960). Subsequently, post-mortem data showed that esterification of free cholesterol is a characteristic of demyelination in MS white matter and spinal cord (Cumings, 1953, Cumings, 1955, Shah and Johnson, 1980, and Yu et al, 1982). Early research also showed a positive association between esterified cholesterol in CSF and greater disability in MS (Pedersen, 1974). More recently, the discovery of new forms of cholesterol and of markers of cholesterol turnover has opened up new avenues for research. For instance, associations between MS disease outcomes and levels of cholesterol precursors, oxysterols, and apolipoprotein (Apo) E in circulation and CSF have been reported (van de Kraats et al, 2014 and Vuletic et al, 2014). Similarly, there are reports of associations between circulating lipoprotein-bound cholesterol and MS disease outcomes in MS, as well as correlation with risk factors for disease progression such as vitamin D deficiency and humoral responses to Epstein–Barr virus (Palavra et al, 2013, Weinstock-Guttman et al, 2011a, and Weinstock-Guttman et al, 2013a). This line of research is ever more pertinent with the recent publication of the MS-STAT study: a randomized, placebo-controlled trial (RCT) which showed that a high dose of the cholesterol-reducing medication simvastatin attenuated brain atrophy and disease progression among 140 patients with secondary progressive MS (SPMS) over 2 years (Chataway et al., 2014). Altogether, these data suggest that cholesterol and related molecules have the potential to be used clinically as biomarkers of treatment and disease outcomes in MS, and that they may represent potential new therapeutic targets. In order to further explore these putative associations, we conducted a systematic review of studies (dated 1983–2015) that examined cholesterol and markers of cholesterol turnover in circulation and/or CSF in relation to disease outcomes in MS.

2. Cholesterol synthesis and metabolism

The synthesis, transport and metabolism of cholesterol are complex, multi-step processes (outlined in Fig. 1) that give rise to a plethora of potential biomarkers. Cholesterol synthesis is initiated by the conversion of acetyl-coenzyme A (CoA) to mevalonate by HMG-CoA-reductase. Mevalonate is subsequently converted to lanosterol and then to cholesterol itself (Ginsberg, 1998). As it is largely water-insoluble, cholesterol is carried in the bloodstream bound to lipid transporters (lipoproteins), which require apolipoproteins (e.g. ApoA1, A2, B, D, and E) for their formation. ApoB is a major component of low density lipoprotein (LDL), intermediate density cholesterol (IDL) and very low density cholesterol (VLDL), and composes 95%, 60%, and 30% of their protein content, respectively (Ginsberg, 1998). LDL, IDL, and VLDL and ApoB are synthesized exclusively in the periphery and cannot pass through an intact blood-brain barrier (BBB; Carlsson et al., 1991; Di Paolo and Kim, 2011). ApoA1 and ApoA2 are major components of high-density lipoprotein (HDL), consisting of 70–80% and 10–15% of its protein mass, respectively (Ginsberg, 1998). Other minor components of HDL include ApoE and ApoD (Ferretti and Bachetti, 2011). HDL and its component apolipoproteins can be synthesized in the periphery, as well as in the brain, and small amounts of HDL can cross the BBB (Di Paolo and Kim, 2011, Ferretti and Bacchetti, 2011, Harr et al, 1996, and Poirier et al, 1991).

Fig. 1

Fig. 1 Pathway of cholesterol synthesis and its metabolites. Cholesterol is the most prevalent steroid in mammals. Acetyl coenzyme A (Acetyl CoA) is converted to 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) which is then converted to mevalonate. The latter reaction is inhibited by the statin class of drugs. Farnesyl pyrophosphate gives rise to several ubiquinone precursors, one of which is the cholesterol precursor squalene, which is cyclized to form lanosterol. The direct precursors of cholesterol are desmosterol and lathosterol. Oxidation of cholesterol leads to creation of the byproducts 24- and 27-hydroxycholesterol, amongst others. Multiple steps and enzymes are omitted for clarity.

Cholesterol metabolism follows primarily from oxidation and subsequent conversion to oxysterols, including 27-hydroxycholesterol (27-OHC) by the cytochrome P450 enzyme CYP27A1 outside the CNS and 24-hydroxycholesterol (24-OHC) by CYP46A1 in neuronal and glial cells (Lütjohann et al., 1996). Following conversion, ~99% of 24-OHC is passed into circulation and ~1% into CSF. Subsequently, both oxysterols are converted into bile acids in the liver (Lütjohann et al., 1996). Additionally, excess cholesterol is transported by HDL, ApoA1, ApoD, and ApoE (Ginsburg, 1998; Demeester et al., 2000; Rassart et al., 2000).

3. Methods

A systematic search was performed in the electronic databases PubMed and EMBASE, using the keywords “cholesterol” OR “apolipoprotein” OR “phospholipid transfer protein” OR “oxysterol” AND “multiple sclerosis.” Two reviewers (SZ, KM) independently screened all abstracts and titles of studies identified by the search strategy for inclusion and appropriateness based on the selection criteria: differences regarding study inclusion were resolved by consensus agreement with the third reviewer (MR). We included studies of adults with clinically isolated syndrome (CIS) or confirmed MS according to the Poser criteria (Poser et al., 1983) or more recent diagnostic criteria. Thus, the search included studies published in English from March 1st, 1983 to October 1st, 2015. Studies were also identified via cross-referencing of primary papers and review articles. In addition to cholesterol itself, we included molecules that would indicate cholesterol turnover including: transfer proteins (e.g. ApoA1, ApoA2, ApoB, ApoD, ApoE), oxysterols (e.g. 24-OHC, 27-OHC), and cholesterol precursors (e.g. lanosterol, lathosterol, and desmosterol) in circulation and/or CSF.

To be included, studies must have measured biological levels of cholesterol biomarkers and have analyzed their relationship with MS disease outcomes (e.g. magnetic resonance imaging [MRI], or optical coherence tomography [OCT] or clinical disability measures such as Expanded Disability Status Scale [EDSS]). The following data were excluded: (1) analyses of gene polymorphisms, (2) tissue analyses, (3) comparisons between MS subtypes, (4) estimation of cholesterol levels via patient questionnaire.

4. Results and discussion

Twenty-one publications met the criteria for inclusion. Findings are summarized in Table 1.

Table 1 Studies of cholesterol and markers of cholesterol turnover in MS.

Study (sample) N (course) [design, follow-up time] Marker Outcome (p for significance) <analysis type> Main results Notes
Mandoj 2015 (serum) 84 RR (58 relapse; 26 remission), 16 SPMS [cross-sectional] LDL-C HDL-C TG Lp(a) EDSS (p≤0.05) <ANOVA, correlation> LDL-C ↑


  • EDSS (p=0.007) ↑

TC ↑

  • EDSS (p=0.01) ↑


A positive correlation was found between age and TC (p=0.006) and LDL-C levels (p=0.0001) and between disease duration and LDL-C levels (p=0.02)
Palavra 2013 (serum, plasma) 30 (RR) [cross-sectional] LDL-C Ox-LDL HDL-C TC EDSS (p≤0.05) <Pearson correlation> LDL-C ↑


  • EDSS (p=0.050) ↑

Ox-LDL ↑

  • EDSS (p=0.011) ↑

TC ↑

  • EDSS (p=0.027) ↑


Kardys 2013 (serum) 121 (RR), 11 (SP), 4 (PP) [retrospective: within +6 months of OCT assessment] LDL-C HDL-C TC RNFL, PRVEP (p≤0.05) <linear mixed-model regression> LDL-C ↑


  • RNFL thickness (p=0.39) –
  • PRVEP latency (p=0.57) –
  • Visual acuity (p=0.69) –


  • RNFL thickness (p=0.008)↓
  • PRVEP latency (p=0.043) ↑
  • Visual acuity (p=0.066) –

TC ↑

  • RNFL thickness (p=0.082) –
  • PRVEP latency (p=0.30) –
  • Visual acuity (p=0.41) –

LDL-C >100 mg/dl status

  • RNFL thickness (p=0.022)↓
  • PRVEP latency (p=0.017) ↑
  • Visual acuity (p=0.83) –

HDL-C >60 mg/dl status

  • RNFL thickness (p=0.001)↓
  • PRVEP latency (p=0.12) –
  • Visual acuity (p=0.99) – TC >200 mg/dl status
  • RNFL thickness (p=0.001) ↓
  • PRVEP latency (p=0.10) –
  • Visual acuity (p=0.55) –


No significant association of cholesterol with RNFL thickness status in eyes that were not affected by ON
Weinstock-Guttman 2011 (serum) 395 (RR), 82, (SP), 15 (PP) [retrospective: 2.2 years#] LDL-C HDL-C TG TC CEL, EDSS, MSSS, T2L, T1L, BPF (p≤0.01) <multiple linear and logistic regression> LDL-C ↑


  • EDSS (p=0.006) ↑
  • MSSS (p=0.012) –
  • CEL presence (p=0.80) –
  • T1L volume (p=NR) –
  • T2L volume (p=NR) – HDL-C ↑
  • CEL volume (p<0.001) ↓
  • CEL presence (p=0.01) ↓
  • EDSS worsening (p=0.79) –
  • T1L volume (p=NR) –
  • T2L volume (p=NR) – TC ↑
  • EDSS (p=0.001) ↑
  • MSSS (p=0.008) ↑
  • BPF (p=0.033) –
  • CEL number (p=0.046) –
  • CEL volume decrease (p=0.20) –
  • CEL presence (p=0.44) –
  • T1L volume (p=NR) –
  • T2L volume (p=NR) –


  • CEL volume (p=0.025) –
  • EDSS (p=0.047) –
  • MSSS at baseline (p=0.080) –
  • EDSS≥4 at baseline (p=0.082) –
  • T1L volume (p=NR) –
  • T2L volume (p=NR) –

TG ↑

  • CEL volume (p=0.023) –
  • EDSS (p=0.025) –
  • MSSS (p=0.037) –
  • CEL presence (p=0.038)


Weinstock- Guttman 2013 (serum) Browne 2014 (serum) Fellows 2015 (serum, CSF) 135 (CIS) [prospective, 2 years] 181 (CIS) [prospective, 2 years] 154 (CIS) [prospective, 2 years] LDL-C HDL-C TC Lp(a) ApoB ApoA1 ApoA2 ApoE CEL, T2L, NBV, GM, relapse (p≤0.01) <regression> LDL-C ↑


  • New T2L (p=0.006)↑
  • New/enlarging T2L (p=0.008)↑
  • NBV at baseline (p=0.020) –
  • GM atrophy (p=0.047) –
  • New CEL (p=0.11) –
  • Number of relapses (p=NR) – HDL-C ↑
  • MRI outcomes (p=NR) –
  • Number of relapses (p=NR) – TC ↑
  • New T2L (p=0.001) ↑
  • GM atrophy (p=0.050) –
  • NBV at baseline (p=0.059) –
  • New CEL (p=0.075) –
  • Enlarging T2L (p=0.34) –
  • Number of relapses (p=NR) –

LDL-C >100 mg/dl status

  • GM atrophy (p=0.032) –
  • New T2L (p=0.047) –
  • Number of relapses (p=NR) – TC >200 mg/dl status
  • New T2L (p<0.001) ↑
  • New/enlarging T2L (p<0.001) ↑
  • Number of relapses (p=NR) – ApoB ↑
  • New T2L (p<0.001) ↑
  • New/enlarging T2L (p<0.001) ↑ ApoA1 ↑
  • New T2L (p=0.025) – ApoE ↑
  • Deep GM atrophy (p<0.001) ↑
  • Thalamic atrophy (p=0.047) –


Risk of developing clinically definite MS was not associated with any lipid profile variables.
Higher levels of serum HDL-C and ApoAI were associated with lower CSF total protein level, CSF albumin level, albumin quotient and CSF IgG level (all p<0.001 for HDL-C and all p<0.01 for ApoAI). HDL-C was also associated with CSF CD80+ (p<0.001) and with CSF CD80+CD19+ (p=0.007) cell frequencies.
Giubilei 2002 (plasma) 18 (CIS) [prospective, 6 months] LDL-C TC ApoE VLDL-C TG VLDL-TG CEL, EDSS (p≤0.05) <correlation, regression> LDL-C ↑


  • CEL number (p=0.02) ↑

TC ↑

  • CEL number (p=0.011) ↑

ApoE ↑

  • CEL number (NR) –


No relationship between CELs and VLDL-C, TG or VLDL-TG
Tettey 2014a,b (serum) 149 (RR), 20 (SP), 9 (PP) [prospective, 2.2 years#] LDL-C HDL-C TC Lp(a) ApoB ApoA1 ApoE EDSS, relapse (p≤0.05) <regression> LDL-C ↑


  • EDSS at baseline (p=0.26) –
  • EDSS over time (p=0.22) –
  • Hazard of relapse (p=0.95) –


  • EDSS at baseline (p=0.96) –
  • EDSS over time (p=0.29) –
  • Hazard of relapse (p=0.87) –

TC ↑

  • EDSS at baseline (p=0.037) ↑
  • EDSS over time (p=0.33) –
  • Hazard of relapse (p=0.94) –


  • EDSS at baseline (p=0.52) –
  • EDSS over time (p=0.029) ↑
  • Hazard of relapse (p=0.86) –

TG ↑

  • EDSS at baseline (p=0.19) –
  • Hazard of relapse (p=0.79) –

Lp(a) ↑

  • EDSS at baseline (p=0.20) –
  • Hazard of relapse (p=0.07) –

ApoB ↑

  • EDSS at baseline (p=0.003) ↑

EDSS over time (p=0.10) –

  • Hazard of relapse (p=0.76) –

ApoB:ApoA1 ↑

  • EDSS at baseline (p=0.018) ↑
  • EDSS over time (p=0.16) –
  • Hazard of relapse (p=0.78) –

ApoA1 ↑

  • EDSS at baseline (p=0.41) –

EDSS over time (p=0.96) –

  • Hazard of relapse (p=0.84) –

ApoE ↑EDSS at baseline (p=0.26) –

  • Hazard of relapse (p=0.63) –


Only 141 RR patients included in hazard of relapse analyses.
Rifai, 1987 (serum, CSF) 33 RR (22 relapse; 11 remission) [cross-sectional] ApoE Relapse (p<0.01) <t-test> ApoE↑(serum)


  • Relapse (p<0.001) ↑ ApoE↑(CSF)
  • Relapse (p=NR) –


Carlsson 1991 (serum, CSF) 35 RR (24 relapse; 11 remission) [cross-sectional] ApoE Relapse (NR) <Mann–Whitney Test> ApoE ↑(serum)


  • Relapse (p=NR) ↑

ApoE ↑(CSF)+

  • Relapse (p=NR) –


Serum lipoprotein profiles of MS patients in remission were shifted to lower values as compared to MS patients in relapse. However, ApoE levels showed the most significant difference between the two groups
Gaillard, 1998 (serum, CSF) 16 (RR), 18 (progressive) [cross-sectional] ApoE EDSS (p<0.05) <regression> ApoE ↑(serum and CSF)


  • EDSS (p=NR) –


Pirttila 2000 (serum, CSF) 55 (RR), 13 (SP), 25 (PP) [cross-sectional] ApoE Relapse (NR) <correlation> ApoE ↑(serum and CSF)


  • Time from last relapse (p=NR) –


Vuletic 2014 (CSF) 39 (RR), 37 (SP), 15 (PP) [cross-sectional] ApoE PLTP NBV, T2L, T1L, (p<0.05) <regression> ApoE ↑


  • NBV (p=0.066) – ApoE↑PLTP ↓
  • NBV (p=0.001)↑
  • T2L volume (p<0.036) ↓+
  • T1L count (p=0.238) –
  • T1L volume (p=0.686) –


No association between ApoE by itself and T2L or T1L volume/count
Reindl 2001 (serum, CSF) 12 (CIS), 17 (RR), 4 (SP) [cross-sectional] ApoD EDSS (p<0.05, NR) <correlation> ApoD ↑(serum and CSF)


  • EDSS (p=NR) –
  • Relapses (p=NR) –


van de Kraats 2014 (serum, CSF) 51 (RR), 39 (SP), 15 (PP) [cross-sectional] 24-OHC 27-OHC Lathosterol Lanosterol NBV, T2L, T1L, CEL (p<0.05) <correlation> 24-OHC ↑(serum)


  • NBV (p=0.004) [RR] ↓ Lanosterol↑(CSF)
  • T2L load (p<0.03) ↑


Teunissen, 2003 (serum) 20 (RR), 20 (SP), 20 (PP) [prospective, 1.4 years] 24-OHC 27-OHC Lathosterol Lanosterol EDSS (p<0.05) <correlation> 24-OHC ↑


  • EDSS at baseline (p=NR) –
  • EDSS change (p=NR) –


Age negatively associated with 24-OHC in PP
Leoni, 2002 (plasma, CSF) 77 (RR), 15 (SP), 3 (PP), 23 (undefined); 19 CEL positive (CSF only), 29 CEL negative (CSF only) [cross-sectional] 24-OHC EDSS, CEL (p<0.05) <correlation> 24-OHC ↑(plasma)


  • EDSS (p<0.05) ↓ 24-OHC ↑(CSF)
  • CEL positive (p<0.05)↑


Spikes in plasma 24-OHC were only found in young patients with positive cerebral MRI scans
Leoni, 2004 (CSF) 88 RR (32 CEL positive; 56 CEL negative) [retrospective] 24-OHC 27-OHC CEL (p<0.05) <ANOVA> 24-OHC ↑+


  • CEL positive (p<0.001) ↑
  • CEL negative (p=NR)

27-OHC ↑+

  • CEL positive (p<0.001) ↑
  • CEL negative (p<0.05) ↑


Karrenbauer, 2006 (plasma) 27 (RR), 19 (PP) [cross-sectional] 24-OHC EDSS, CEL, T2L, T1L (p<0.0083) <regression> 24-OHC ↑


  • T1L volume (p=0.052) [RR] –
  • T1L volume (p=0.854) [PP] –
  • T2L volume (p=0.078) [RR] –
  • T2L volume (p=0.247) [PP] –
  • CEL volume (p=0.937) [RR] –
  • CEL volume (p=0.158) [PP] –
  • EDSS (p=0.465) [RR] –
  • EDSS (p=0.792) [PP] –


Age negatively associated with 24-OHC in PP

↑=increased;↓= decreased; – no significant relationship and no trend; #=mean;+= versus healthy controls; NR=not reported; RR=relapsing-remitting; SP=secondary progressive; PP=primary progressive; CIS=clinically isolated syndrome; ON=optic neuritis; ApoE=apolipoprotein E; ApoB=apolipoprotein B; ApoA1=apolipoprotein A1; ApoA2=apolipoprotein A2; ApoD=apolipoprotein D; PLTP=phospholipid transfer protein; 24-OHC=24-hydroxycholesterol; 27-OHC=27-hydroxycholesterol; CSF=cerebrospinal fluid; LDL-C=low-density lipoprotein; HDL-C=high-density lipoprotein; Lp(a)=lipoprotein(a); TC=total cholesterol; TG=triglycerides; RNFL=retinal nerve fiber layer; PRVEP=pattern reversal visual-evoked potential; EDSS=Expanded Disability Status Scale; MSSS=MS Severity Scale; OCT=optical coherence tomography; BMI=Body Mass Index; CEL=contrast-enhancing lesion; CEL+=contrast-enhancing lesions present on MRI; CEL-=contrast-enhancing lesions absent on MRI; T1L=T1 lesion; T2L=T2 lesion; BPF=brain parenchymal fraction; NBV=normalized brain volume; GM=gray matter; ANOVA=analysis of variance.

4.1. Lipoprotein-bound cholesterol and apolipoproteins A1, A2 and B

All studies that measured circulating cholesterol found that elevated circulating LDL and/or total cholesterol are associated with adverse outcomes in MS (Table 1), including clinical progression as measured by Expanded Disability Status Scale (EDSS) score, contrast-enhancing lesions (CELs), total T2 lesion load on MRI, and retinal nerve fibre layer (RNFL) thinning on OCT (optical coherence tomography). In clinically isolated syndrome (CIS), there is evidence of a positive association between baseline total and LDL plasma cholesterol and mean number of CELs among 18 patients over 6 months (Giubilei et al., 2002). Similarly, higher LDL and total cholesterol levels were related to increased cumulative number of new T2 lesions among 135 patients with CIS over a 2-year period (Weinstock-Guttman et al., 2013b). In relapsing-remitting MS (RRMS), serum LDL and total cholesterol positively correlated with EDSS score at baseline (Palavra et al., 2013). Likewise, elevated total serum cholesterol was associated with a higher EDSS at baseline and elevated total cholesterol:HDL ratio was associated with subsequent worsening of EDSS after a mean of 2.2 years in a group of 178 MS patients of mixed subtypes of which 83% were RRMS (Tettey et al., 2014a). However, there was no relationship with hazard of relapse in the RRMS cohort (Tettey et al., 2014b). Another study found a relationship between worsening of the EDSS over 2.2 years and higher baseline serum LDL and total cholesterol among 492 MS patients of mixed subtypes of which 80% had RRMS (Weinstock-Guttman et al., 2011b). The same group later showed that serum LDL cholesterol was positively related to lower RNFL thickness and longer pattern reversal visual-evoked potential (PRVEP) latency (both reflecting greater degree of injury to the optic nerve) in eyes that were affected by optic neuritis among 136 MS patients of mixed subtypes of which 89% had RRMS (Kardys et al., 2013). Further, a recent study found a positive relationship between EDSS score and both LDL and total cholesterol level in a cross-sectional study of 100 MS patients (Mandoj et al., 2015).

While elevated LDL and total cholesterol appear to be associated with, and predictive of, both worse and worsening disease, potential causal mechanisms remain to be elucidated. Indeed, ‘non-classical’ markers of cardiovascular risk such as oxidized LDL (ox-LDL) and major apolipoprotein components of lipoprotein particles are more predictive of adverse outcomes than LDL itself (Browne et al, 2014, Palavra et al, 2013, and Tettey et al, 2014a). For instance, there is evidence that ox-LDL is a stronger predictor of elevated EDSS than either LDL or total cholesterol (Palavra et al., 2013). Tettey et al. (2014a) observed that lipoprotein-bound particles were not significantly related to outcomes when considered by themselves; however, elevated ApoB and ApoB:ApoA1 ratio were associated with higher EDSS at baseline (Tettey et al., 2014a). Likewise, Browne et al. (2014) reported baseline ApoB was more strongly associated with the development of new T2 lesions than lipoprotein cholesterol in CIS.

The apparent superiority of ApoB and ox-LDL over ‘classic’ markers of cardiovascular risk in predicting MS disease outcomes is perhaps not surprising given that they are also the strongest predictors of atherogenic risk (Huang et al, 2012 and Jacobson, 2011). Interestingly, like MS, atherosclerosis is a chronic inflammatory disease with components of both the innate and adaptive immune system contributing to disease pathogenesis (Hansson and Hermansson, 2011). LDL particles accumulate in the arterial wall, where they can be oxidatively modified (Tabas et al., 2007). Smaller and denser LDL particles are particularly prone to oxidative modification; the total LDL cholesterol level may not accurately reflect an increase of such particles because LDL particles can have variable cholesterol content. By contrast, the ratio between ApoB levels and the total number of atherogenic particles is 1:1 (Jacobson, 2011). Peroxidation of LDL generates highly reactive aldehydes such as malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE), which convert ApoB on the surface of LDL to a form recognized by the macrophage scavenger receptor, leading the macrophages to engulf the ox-LDL and become foam cells that create plaques on the arterial wall (Tabas et al, 2007 and Witztum and Steinberg, 1991).

Intriguingly, there is evidence that these deleterious processes can also contribute to disease pathogenesis in MS. Native LDL, MDA-LDL epitopes and myelin basic protein-derived peptides co-localize within foamy macrophages in early and actively demyelinating MS plaques (Newcombe et al., 1994). Importantly, these plaque macrophages also contained esterified cholesterol. This suggests that leakage of circulating LDL into parenchyma of MS plaques and subsequent oxidative modification by phagocytes may play an important role in demyelination (Newcombe et al., 1994). It also supports the theory that cholesterol esters in demyelinating MS plaques are derived from both circulating LDL and CNS myelin. The leakage of LDL into the MS brain is possible due to BBB damage inherent to MS and it can be exacerbated by elevated circulating cholesterol, which is known to independently increase BBB permeability in animals (Jiang et al., 2012).

In contrast to circulating cholesterol/ApoB, data regarding HDL and its major apolipoprotein components and MS outcomes have been equivocal. Weinstock-Guttman et al. (2011b) found that elevated HDL was related to fewer CELs and reduced CEL volume at baseline. Further, while Tettey et al. (2014a) did not find a relation between either total HDL or ApoA1 levels and positive outcomes, they did find that HDL:total cholesterol and ApoAI:ApoB ratios did relate to positive outcomes. Similarly, a recent study of CIS patients found elevated HDL to be associated with lower levels of BBB injury, as measured by a decreased ratio of CSF albumin to serum albumin, as well as decreased CD80+ and CD80+CD19+ cell extravasation into the CSF (Fellows et al., 2015). Other studies did not find a relationship between HDL, ApoA1, or ApoA2 and disease outcomes in CIS and MS (Browne et al, 2014, Giubilei et al, 2002, and Palavra et al, 2013). By contrast, Kardys et al. (2013) found that HDL was positively associated with indicators of optic nerve injury such as reduced RNFL thickness and prolonged PRVEP latency in MS.

These equivocal results may be in part explained by the fact that these studies did not differentiate between HDL subtypes and did not measure ox-HDL. Indeed, elevated HDL is typically protective in atherosclerosis via its ability to transport cholesterol away from the arterial wall and to neutralize the deleterious effects of ox-LDL and other mediators of inflammation and oxidative stress (Kontush and Chapman, 2006 and Navab et al, 2005). Nonetheless, different subtypes of HDL (i.e. small, intermediate, and large) vary in their effect on intravascular metabolism and in their anti-atherogenic properties (Kontush and Chapman, 2006). Moreover, conditions associated with chronic acute phase response and systemic inflammation, such as atherosclerosis, can change HDL into oxidized HDL (ox-HDL), making it a pro-inflammatory and pro-oxidant factor (Navab et al., 2005).

4.2. ApoE, ApoD, and phospholipid transfer protein

In addition to being minor components of HDL (Ferretti and Bacchetti, 2011), ApoE and ApoD can also act to remove cholesterol from injured nerves, to promote axonal regeneration and remyelination, and to reduce oxidation and inflammation (Boyles et al, 1990, Dassati et al, 2014, Miyata and Smith, 1996, Rassart et al, 2000, and Wei et al, 2013). However, current evidence for an association between these apolipoproteins and disease activity in MS has been equivocal (Table 1). Two early studies found increased serum ApoE levels among MS patients during a relapse, relative to those in remission, but there were no differences when examining CSF (Carlsson et al, 1991 and Rifai et al, 1987). Browne et al. (2014) found that serum ApoE was positively associated with brain atrophy in CIS, but that there was no relationship with relapses. There is also evidence that when taken together in a regression model, CSF ApoE and phospholipid transfer protein (PLTP; also involved in cholesterol transport and metabolism) could predict adverse MRI changes such as brain atrophy and development of new T2 lesions among 91 patients of mixed MS subtypes of which 43% had RRMS (Vuletic et al., 2014). This is in line with evidence that these two proteins are interdependent; for example, PLTP regulates ApoE expression and secretion in primary human astrocytes in vitro (Vuletic et al., 2005). By contrast, other studies found no relationship between serum or CSF levels of ApoE and clinical disease outcomes in MS patients of mixed subtypes (Gaillard et al, 1998, Pirttilä et al, 2000, and Tettey et al, 2014a), or presence of CELs on MRI in CIS (Giubilei et al., 2002). ApoD has not been widely investigated in MS; only one study exists and it failed to find an association with EDSS or relapses in serum or CSF among 33 MS patients of mixed subtypes (36% CIS, 52% RRMS; Reindl et al., 2001). Altogether, the available data do not point to a strong link between circulating or CSF ApoE and ApoD and MS disease outcomes. Given the conflicting evidence to date, more research is required to assess whether ApoE and/or ApoD can act as biomarkers of disease process in MS. Ideally, future studies would examine PLTP levels, enrol a large number of subjects, and include MRI outcomes.

4.3. Cholesterol precursors

Lanosterol is synthesized in the actively myelinating rat brain (Ramsey et al., 1972) and during remyelination following nerve injury (Yao, 1988). To date, only two studies measured cholesterol precursors in MS. Most notably, T2 hyperintense lesion load correlated with concentrations of lanosterol in CSF (but not serum) among 105 MS patients of mixed subtypes (49% RRMS) suggesting increased brain cholesterol synthesis in MS (van de Kraats et al., 2014; Table 1). Investigation into other cholesterol precursors (e.g. lathosterol, desmosterol) in serum and CSF have thus far failed to find a clear relationship with disease outcomes, but there are few studies and more research is required in order to elucidate a possible association (Teunissen et al, 2003 and van de Kraats et al, 2014).

4.4. Oxysterols

It is tempting to consider the cholesterol metabolites, 24-OHC and 27-OHC, as potential biomarkers for disease activity in MS. 24-OHC is primarily produced in the CNS, while 27-OHC is produced mostly in peripheral tissues. Because most brain cholesterol is contained within myelin (Di Paolo and Kim, 2011), elevated levels of circulating or CSF 24-OHC likely reflect changes in brain cholesterol turnover caused by demyelination. Moreover, the presence of 27-OHC in CSF is thought to indicate BBB dysfunction in neurological disorders, but the significance of circulating 27-OHC is unknown (Leoni et al., 2004). In one study of 88 RRMS patients, both 24-OHC and 27-OHC were elevated in the CSF of patients who had CELs (n=32; Leoni et al., 2004; Table 1), relative to healthy controls. CSF levels of 27-OHC were elevated to a considerably smaller degree in CEL negative-patients (n=56), relative to healthy controls, and no difference in CSF levels of 24-OHC was observed between CEL-negative patients and healthy controls. The presence of a CEL evidences concurrent BBB breakdown. In addition, van de Kraats et al. (2014) showed that serum (but not CSF) 24-OHC positively correlated with brain atrophy among their RRMS subgroup (n=51). However, no relationships were noted between T1 lesion or T2 lesion number and 24-OHC or 27-OHC levels in serum or CSF in any MS subtype. CEL counts were available from too few patients to perform a statistically valid analysis (van de Kraats et al., 2014). In a different study, the same investigators did not observe any relationship between serum 24-OHC or 27-OHC and EDSS at baseline or after 1.4 years of follow-up among RRMS (n=20), SPMS (n=20), or PPMS patients (n=20); however, disease progression is slow in MS and the follow-up time in this study was limited. In another study, there were no significant associations between plasma 24-OHC and T2 lesion or CEL volume among RRMS (n=27) or PPMS patients (n=19); however, a positive but non-significant trend was observed for T1 lesion volume in people with RRMS (Karrenbauer et al., 2006). Additionally, when examining 118 patients of mixed MS subtypes (65% RRMS), Leoni et al., (2002) found that spikes in plasma 24-OHC were only found among younger patients with positive cerebral MRI scans. In that study, sub-group analysis of RRMS patients for whom CSF was available showed that 24-OHC was significantly higher in patients with CELs (n=19), relative to those without (n=29; Leoni et al., 2002). By contrast, the authors found a significant inverse relationship between EDSS score and 24-OHC in plasma (Leoni et al., 2002). This observation could be explained by the fact that T1 and T2 lesion number and volume may actually decrease with disease progression because they tend to collapse and coalesce or disappear over time.

Turnover of brain cholesterol as measured by circulating 24-OHC likely reflects the degree of ongoing demyelination. It is thus perhaps not surprising that age and/or EDSS was negatively related to circulating 24-OHC in a number of studies (Leoni et al, 2002, Karrenbauer et al, 2006, and Teunissen et al, 2003) as older, progressive patients are known to experience fewer acute demyelinating events (Goldschmidt et al, 2009 and Ontaneda and Fox, 2015). In addition, the inverse relationship between circulating 24-OHC and age and/or EDSS may correspond to decreased turnover of brain cholesterol with loss of neuronal cells over time. Altogether, these data suggest that changes in levels of the oxysterols, 24-OHC and 27-OHC, may be associated with demyelination and BBB dysfunction. However, longitudinal studies that include MRI imaging are required to confirm these relationships.

In addition to their potential as biomarkers of disease processes, oxysterols may be involved in driving MS pathophysiology. For example, 24- and 27-OHC induce expression and synthesis of inflammatory cytokines, chemokines, and adhesion molecules (Poli et al., 2013). Serum 24- and 27-OHC levels increase prior to onset of clinical symptoms in experimental autoimmune encephalomyelitis (EAE), reaching a peak of 193% and 415% respectively (Teunissen et al., 2007). In the human brain, 24-OHC has been detected at concentrations of up to 30 μM (Lütjohann et al., 1996), whereas 24-OHC at concentrations higher than 10 μM induces cell death in SH-SY5Y cells and primary cortical neurons (Kölsch et al, 1999 and Noguchi et al, 2014). In a similar fashion, 27-OHC can be neurotoxic in vitro and cerebral accumulation of 27-OHC may be associated clinically aggressive disease progression in Alzheimer's disease (Poli et al, 2013 and Shafaati et al, 2011). Thus, accumulation of 24-OHC and 27-OHC may cause secondary damage to neurons via neurotoxicity and increased inflammation.

5. Implications for treatment of MS

Potential cholesterol-related mechanisms in MS disease progression are outlined in Fig. 2. Overall, this review suggests that the reduction of circulating cholesterol could be beneficial in MS, and epidemiologic studies support this conclusion. An early study showed that consumption of a low saturated fat diet was associated with significantly less disease progression and mortality among 140 MS patients who adhered to the diet (Swank and Dugan, 1990). More recently, several studies found that obesity in childhood and early adulthood is associated with an increased risk of developing MS (Hedström et al, 2012, Langer-Gould et al, 2013, and Munger et al, 2009). Similarly, there is evidence that vascular comorbidity (chiefly hypercholesterolemia and hypertension), whether present at symptom onset, diagnosis, or later in the disease course, was associated with a substantially increased risk of disability progression among a cohort of 8983 MS patients (Marrie et al., 2010). The deleterious effects found in these epidemiologic studies could be mediated directly by ApoB since it showed the strongest association with poor disease outcomes, when considered together with BMI and a host of other lipid markers (Tettey et al., 2014a).

Fig. 2

Fig. 2 Potential role of the cholesterol pathway in MS disease progression. Hypercholesterolemia (left side of figure) can lead to elevations in circulating LDL / ApoB and 27-hydroxycholesterol as well as blood–brain barrier (BBB) dysfunction. LDL is oxidized in the arterial wall by environmental stressors (e.g. smoking, stress), which can lead to enhanced immune recruitment. LDL can pass into the brain and be oxidized within MS lesions, thereby contributing to cell damage via its pro-inflammatory and neurotoxic effects. Reduced HDL may enhance injury to the BBB. 27-hydroxycholesterol can lead to enhanced immune recruitment and is a potential neurotoxin. Immune damage to myelin (right side of figure) can lead to generation of 24-hydroxycholesterol and lipid peroxidation by-products, which can also contribute to cell damage via their pro-inflammatory and neurotoxic effects.

Reduction of circulating cholesterol can be achieved through diet by lowering consumption of saturated fat and increasing consumption of niacin and plant sterols (Clarke et al, 1997, Jacobson, 2011, and Vanmierlo et al, 2015). Cholesterol reduction can also be achieved by treatment with the statin class of drugs that inhibit HMG-CoA reductase, the rate-limiting enzyme for cholesterol synthesis (Corsini et al., 1999). Recently, Pihl-Jensen et al., (2015) conducted a systematic review and meta-analysis that examined well-controlled RCTs of statins for the treatment of MS. The meta-analysis included trials in which patients had been treated with either simvastatin (40–80 mg) or atorvastatin (20–80 mg) for a duration of 6–25 weeks. No significant benefit of statins was observed in RRMS added to IFN-β therapy (5 studies), but there was a trend towards an increase in disease activity in the statin group as evidenced by an increase in new T2 lesions, the proportion of patients who relapsed, and whole brain atrophy. By contrast, statin monotherapy showed significant reduction in brain atrophy and disability progression in SPMS (MS-STAT study; Chataway et al., 2014; Pihl-Jensen et al., 2015). These findings are potentially exciting because no effective therapies currently exist for progressive MS (Ontaneda and Fox, 2015); however, further research is warranted.

The mechanisms responsible for the putative benefits of statins in SPMS remain to be elucidated. Statins have been shown to reduce LDL and ApoB by as much as 55% and 45%, respectively and increased HDL and ApoA1 by as much as 10% and 9%, respectively (Jones et al, 2003 and Jones et al, 2004). In addition to reduction of cholesterol, statins can promote other mechanisms that could be beneficial in MS, including inhibition of LDL oxidation, scavenger receptor expression, metalloproteinase secretion, superoxide generation and growth of macrophages induced by oxidized LDL (Corsini et al., 1999). These benefits could extend to the CNS because some statins are able to cross the BBB (simvastatin and atorvastatin). On the other hand, CNS penetrating statins could exert deleterious effects because cholesterol synthesis is necessary for remyelination. This could explain why there was a trend towards increased disease activity in the meta-analysis conducted by Pihl-Jensen et al. (2015). Indeed, there is evidence that prolonged simvastatin treatment can induce demyelination, reduce the number of mature oligodendrocytes, block differentiation of oligodendrocyte progenitor cells and inhibit remyelination in the cuprizone model (Miron et al., 2009). Remyelination in RRMS patients is more extensive than in progressive MS patients (Goldschmidt et al., 1914), suggesting that any potential benefits of CNS penetrating statins could be negated in the former individuals. It is possible that hydrophilic statins such as rosuvastatin and pravastatin that do not cross the BBB may be more beneficial among relapsing patients, however, well-controlled RCTs of these medications have not yet been conducted in MS (Corsini et al., 1999; Pihl-Jensen et al., 2015).

As statin treatment may be counterproductive in MS due to its potential for inhibiting remyelination, direct inhibition of ApoB and/or attenuation of cholesterol oxidation could be a promising strategy for the pharmacotherapy of MS. Other lipid regulating agents (some of which also possess anti-oxidant, anti-inflammatory and pro-myelinating activity) such as gemfibrozil, pioglitazone, and ezetimibe could be tried as alternatives to statins in MS (de Pablos-Velasco, 2010, Jacobson, 2011, Qin et al, 2014, and Roy and Pahan, 2009). Additionally, mimetic compounds such as the cholesterol derivative olesoxime (currently in phase 1b testing for MS: NCT01808885) and ApoE derived peptides could be beneficial. Indeed, there is evidence that olesoxime accelerates oligodendrocyte maturation, enhances myelination in vitro and attenuates demyelination / promotes remyelination in vivo (Magalon et al., 2012). Similarly, ApoE mimetics have been shown to ameliorate clinical disability and inflammatory infiltrates into the spinal cord in EAE (Li et al., 2006). Altogether, these data suggest that altering specific parts of the cholesterol metabolism via pharmacological means could be a promising strategy for the treatment of MS.

6. Conclusion

Currently, there is a need for biomarkers in MS to detect subclinical disease activity and repair and to predict future outcomes (Comabella and Montalban, 2014 and Teunissen et al, 2005). Cholesterol and markers of cholesterol turnover may contribute to the pathogenesis of MS and may prove to be useful biomarkers of disease activity and treatment efficacy. Elevated levels of circulating LDL cholesterol, oxidized LDL, total cholesterol and apolipoprotein B are associated with adverse clinical and MRI outcomes. Nonetheless, it remains unclear whether the relationship between cholesterol and MS disease outcomes is causal. Epidemiological research has shown that vascular comorbidities, including hypercholesterolemia, contribute to a more rapid MS disease progression (Marrie et al., 2010; Tettey et al., 2014c). Thus, it remains unclear whether high cholesterol itself leads to deleterious outcomes or if it is simply a marker of these outcomes. The cross-sectional design of many of the included studies further limits the ability to infer causation. Most studies did not analyze results by disease subtype; in light of the contrasting effects of statins on relapsing and progressive MS, it is possible that analyzing these subtypes together may have masked any true associations. Nevertheless, the research completed to date suggests there is great promise in these molecules as biomarkers and supports further study. Future studies should employ a longitudinal design, preferably in conjunction with MRI or OCT, in order to better understand the relationship between pathological mechanisms of specific disease processes in MS (e.g. demyelination, remyelination, axonal degeneration, BBB dysfunction) and levels of cholesterol and cholesterol-related biomarkers.

Declaration of conflicting interests

M.R. holds an EMD Serono, Canada and endMS Research and Training Network Transitional Career Development Award from the MS Society of Canada and the Multiple Sclerosis Scientific Research Foundation. Acceptance of this unrestricted educational grant by the MS Society of Canada does not constitute endorsement by researcher or the Society of any product(s) of EMD Serono, Canada. The MS Society of Canada is an independent, voluntary health agency and does not approve, endorse or recommend any specific product or therapy but provides information to assist individuals in making their own decisions. Funds from this award did not support the work contained here.


Simon Zhornitsky is recipient of a post-doctoral fellowship award from the MS Society of Canada. Kyla McKay is a recipient of the Alistair M. Fraser Master's Studentship Award from the MS Society of Canada. Manu Rangachari is supported by a Junior-1 «chercheurs-boursiers» salary support award from the Fonds de Recherche du Québec – Santé (FRQS). This work was completed as part of an endMS SPRINT interdisciplinary learning project while S.Z. and K.A.M. were enrolled in the Scholar Program for Researchers in Training (SPRINT). SPRINT is part of the endMS National Training Program funded by the Multiple Sclerosis Society of Canada through its related MS Scientific Research Foundation.


We thank V. Wee Yong for critical reading of the manuscript.


  • Berg et al., 2002 Berg, J.M., Tymoczko, J.L., Stryer, L., 2002. Important Derivatives of Cholesterol Include Bile Salts and Steroid Hormones. In: Biochemistry, 5th Ed., W H Freeman, New York.
  • Boyles et al., 1990 J.K. Boyles, L.M. Notterpek, L.J. Anderson. Accumulation of apolipoproteins in the regenerating and remyelinating mammalian peripheral nerve. Identification of apolipoprotein D, apolipoprotein A-IV, apolipoprotein E, and apolipoprotein A-I. J. Biol. Chem.. 1990;265(29):17805-17815
  • Browne et al., 2014 R.W. Browne, B. Weinstock-Guttman, D. Horakova, R. Zivadinov, M.L. Bodziak, M. Tamaño-Blanco, et al. Apolipoproteins are associated with new MRI lesions and deep grey matter atrophy in clinically isolated syndromes. J. Neurol. Neurosurg. Psychiatry. 2014;85(8):859-864 Crossref
  • Carlsson et al., 1991 J. Carlsson, V.W. Armstrong, H. Reiber, K. Felgenhauer, D. Seidel. Clinical relevance of the quantification of apolipoprotein E in cerebrospinal fluid. Clin. Chim. Acta. 1991;196(2–3):167-176 Crossref
  • Chataway et al., 2014 J. Chataway, N. Schuerer, A. Alsanousi, D. Chan, D. MacManus, K. Hunter, et al. Effect of high-dose simvastatin on brain atrophy and disability in secondary progressive multiple sclerosis (MS-STAT): a randomised, placebo-controlled, phase 2 trial. Lancet. 2014;383(9936):2213-2221 Crossref
  • Clarke et al., 1997 R. Clarke, C. Frost, R. Collins, P. Appleby, R. Peto. Dietary lipids and blood cholesterol: quantitative meta-analysis of metabolic ward studies. Br. Med. J.. 1997;314(7074):112-117
  • Comabella and Montalban, 2014 M. Comabella, X. Montalban. Body fluid biomarkers in multiple sclerosis. Lancet Neurol.. 2014;13(1):113-126 Crossref
  • Cumings, 1953 J.N. Cumings. The cerebral lipids in disseminated sclerosis and in amaurotic family idiocy. Brain 1953. 1953;76(4):551-562
  • Cumings, 1955 J.N. Cumings. Lipid chemistry of the brain in demyelinating diseases. Brain 1955. 1955;78(4):554-563
  • Dassati et al., 2014 S. Dassati, A. Waldner, R. Schweigreiter. Apolipoprotein D takes center stage in the stress response of the aging and degenerative brain. Neurobiol. Aging. 2014;35(7):1632-1642 Crossref
  • de Pablos-Velasco, 2010 P. de Pablos-Velasco. Pioglitazone: beyond glucose control. Expert. Rev. Cardiovasc. Ther.. 2010;8(8):1057-1067 Crossref
  • Demeester et al., 2000 N. Demeester, G. Castro, C. Desrumaux, C. De Geitere, J.C. Fruchart, P. Santens, et al. Characterization and functional studies of lipoproteins, lipid transfer proteins, and lecithin:cholesterol acyltransferase in CSF of normal individuals and patients with Alzheimer’s disease. J. Lipid Res.. 2000;41(6):963-974
  • Di Paolo and Kim, 2011 G. Di Paolo, T.W. Kim. Linking lipids to Alzheimer’s disease: cholesterol and beyond. Nat. Rev. Neurosci.. 2011;12(5):284-296 Crossref
  • Fellows et al., 2015 K. Fellows, T. Uher, R.W. Browne, B. Weinstock-Guttman, D. Horakova, H. Posova, et al. Protective associations of HDL with blood-brain barrier injury in multiple sclerosis patients. J. Lipid Res.. 2015;56(10):2010-2018
  • Ferretti and Bacchetti, 2011 G. Ferretti, T. Bacchetti. Peroxidation of lipoproteins in multiple sclerosis. J. Neurol. Sci.. 2011;311(1–2):92-97 Crossref
  • Gaillard et al., 1998 O. Gaillard, A. Gervais, D. Meillet, E. Plassart, B. Fontaine, O. Lyon-Caen, et al. Apolipoprotein E and multiple sclerosis: a biochemical and genetic investigation. J. Neurol. Sci.. 1998;158(2):180-186 Crossref
  • Giubilei et al., 2002 F. Giubilei, G. Antonini, S. Di Legge, M.P. Sormani, P. Pantano, R. Antonini, et al. Blood cholesterol and MRI activity in first clinical episode suggestive of multiple sclerosis. Acta Neurol. Scand.. 2002;106(2):109-112 Crossref
  • Ginsberg, 1998 H.N. Ginsberg. Lipoprotein physiology. Endocrinol. Metab. Clin. N. Am.. 1998;27(3):503-519 Crossref
  • Goldschmidt et al., 2009 T. Goldschmidt, J. Antel, F.B. Konig, W. Bruck, T. Kuhlmann. Remyelination capacity of the MS brain decreases with disease chronicity. Neurology. 2009;72(22):1914-1921 Crossref
  • Hansson and Hermansson, 2011 G.K. Hansson, A. Hermansson. The immune system in atherosclerosis. Nat. Immunol.. 2011;12(3):204-212 Crossref
  • Harr et al., 1996 S.D. Harr, L. Uint, R. Hollister, B.T. Hyman, A.J. Mendez. Brain expression of apolipoproteins E, J, and A–I in Alzheimer’s disease. J. Neurochem.. 1996;66(6):2429-2435
  • Hedström et al., 2012 A.K. Hedström, T. Olsson, L. Alfredsson. High body mass index before age 20 is associated with increased risk for multiple sclerosis in both men and women. Mult. Scler.. 2012;18(9):1334-1336
  • Huang et al., 2012 H. Huang, R. Ma, D. Liu, C. Liu, Y. Ma, W. Mai, et al. Oxidized low-density lipoprotein cholesterol and the ratio in the diagnosis and evaluation of therapeutic effect in patients with coronary artery disease. Dis. Markers. 2012;33(6):295-302 Crossref
  • Jacobson, 2011 T.A. Jacobson. Opening a new lipid “apo-thecary”: incorporating apolipoproteins as potential risk factors and treatment targets to reduce cardiovascular risk. Mayo Clin. Proc.. 2011;86(8):762-780 Crossref
  • Jiang et al., 2012 X. Jiang, M. Guo, J. Su, B. Lu, D. Ma, R. Zhang, et al. Simvastatin blocks blood–brain barrier disruptions induced by elevated cholesterol both in vivo and in vitro. Int. J. Alzheimers Dis.. 2012;2012:109324
  • Jones et al., 2003 P.H. Jones, M.H. Davidson, E.A. Stein, H.E. Bays, J.M. McKenney, E. Miller, et al. Comparison of the efficacy and safety of rosuvastatin versus atorvastatin, simvastatin, and pravastatin across doses (STELLAR* Trial). Am. J. Cardiol.. 2003;92(2):152-160 Crossref
  • Jones et al., 2004 P.H. Jones, D.B. Hunninghake, K.C. Ferdinand, E.A. Stein, A. Gold, R.J. Caplan, et al. Effects of rosuvastatin versus atorvastatin, simvastatin, and pravastatin on non-high-density lipoprotein cholesterol, apolipoproteins, and lipid ratios in patients with hypercholesterolemia: additional results from the STELLAR trial. Clin. Ther.. 2004;26(9):1388-1399 Crossref
  • Kardys et al., 2013 A. Kardys, B. Weinstock-Guttman, M. Dillon, M.W. Masud, N. Weinstock, N. Mahfooz, et al. Cholesterol affects retinal nerve fiber layer thickness in patients with multiple sclerosis with optic neuritis. Eur. J. Neurol.. 2013;20(9):1264-1271 Crossref
  • Karrenbauer et al., 2006 V.D. Karrenbauer, V. Leoni, E.T. Lim, G. Giovannoni, G.T. Ingle, J. Sastre-Garriga, et al. Plasma cerebrosterol and magnetic resonance imaging measures in multiple sclerosis. Clin. Neurol. Neurosurg.. 2006;108(5):456-460 Crossref
  • Kölsch et al., 1999 H. Kölsch, D. Lütjohann, A. Tulke, I. Björkhem, M.L. Rao. The neurotoxic effect of 24-hydroxycholesterol on SH-SY5Y human neuroblastoma cells. Brain Res.. 1999;818(1):171-175
  • Kontush and Chapman, 2006 A. Kontush, M.J. Chapman. Antiatherogenic small, dense HDL--guardian angel of the arterial wall?. Nat. Clin. Pract. Cardiovasc. Med.. 2006;3(3):144-153 Crossref
  • Langer-Gould et al., 2013 A. Langer-Gould, S.M. Brara, B.E. Beaber, C. Koebnick. Childhood obesity and risk of pediatric multiple sclerosis and clinically isolated syndrome. Neurology. 2013;80(6):548-552 Crossref
  • Leoni et al., 2002 V. Leoni, T. Masterman, U. Diczfalusy, G. De Luca, J. Hillert, I. Björkhem. Changes in human plasma levels of the brain specific oxysterol 24S-hydroxycholesterol during progression of multiple sclerosis. Neurosci. Lett.. 2002;331(3):163-166 Crossref
  • Leoni et al., 2004 V. Leoni, T. Masterman, F.S. Mousavi, B. Wretlind, L.O. Wahlund, U. Diczfalusy, et al. Diagnostic use of cerebral and extracerebral oxysterols. Clin. Chem. Lab. Med.. 2004;42(2):186-191
  • Li et al., 2006 F.Q. Li, G.D. Sempowski, S.E. McKenna, D.T. Laskowitz, C.A. Colton, M.P. Vitek. Apolipoprotein E-derived peptides ameliorate clinical disability and inflammatory infiltrates into the spinal cord in a murine model of multiple sclerosis. J. Pharmacol. Exp. Ther.. 2006;318(3):956-965
  • Lütjohann et al., 1996 D. Lütjohann, O. Breuer, G. Ahlborg, I. Nennesmo, A. Sidén, U. Diczfalusy, et al. Cholesterol homeostasis in human brain: evidence for an age-dependent flux of 24S-hydroxycholesterol from the brain into the circulation. Proc. Natl. Acad. Sci. USA. 1996;93(18):9799-9804
  • Magalon et al., 2012 K. Magalon, C. Zimmer, M. Cayre, J. Khaldi, C. Bourbon, I. Robles, et al. Olesoxime accelerates myelination and promotes repair in models of demyelination. Ann. Neurol.. 2012;71(2):213-226 Crossref
  • Mandoj et al., 2015 C. Mandoj, R. Renna, D. Plantone, I. Sperduti, G. Cigliana, L. Conti, et al. Anti-annexin antibodies, cholesterol levels and disability in multiple sclerosis. Neurosci. Lett.. 2015;606:156-160
  • Marrie et al., 2010 R.A. Marrie, R. Rudick, R. Horwitz, G. Cutter, T. Tyry, D. Campagnolo, et al. Vascular comorbidity is associated with more rapid disability progression in multiple sclerosis. Neurology. 2010;74(13):1041-1047 Crossref
  • Miron et al., 2009 V.E. Miron, S.P. Zehntner, T. Kuhlmann, S.K. Ludwin, T. Owens, T.E. Kennedy, et al. Statin therapy inhibits remyelination in the central nervous system. Am. J. Pathol.. 2009;174(5):1880-1890 Crossref
  • Miyata and Smith, 1996 M. Miyata, J.D. Smith. Apolipoprotein E allele-specific antioxidant activity and effects on cytotoxicity by oxidative insults and beta-amyloid peptides. Nat. Genet.. 1996;14(1):55-61 Crossref
  • Mott, 1910 F.M. Mott. The Oliver-Sharpey lectures on cerebrospinal fluid. Lancet. 1910;2(1):79
  • Munger et al., 2009 K.L. Munger, T. Chitnis, A. Ascherio. Body size and risk of MS in two cohorts of US women. Neurology. 2009;73(19):1543-1550 Crossref
  • Navab et al., 2005 M. Navab, G.M. Ananthramaiah, S.T. Reddy, B.J. Van Lenten, B.J. Ansell, S. Hama, et al. The double jeopardy of HDL. Ann. Med.. 2005;37(3):173-178 Crossref
  • Newcombe et al., 1994 J. Newcombe, H. Li, M.L. Cuzner. Low density lipoprotein uptake by macrophages in multiple sclerosis plaques: implications for pathogenesis. Neuropathol. Appl. Neurobiol.. 1994;20(2):152-162 Crossref
  • Noguchi et al., 2014 N. Noguchi, Y. Saito, Y. Urano. Diverse functions of 24(S)-hydroxycholesterol in the brain. Biochem. Biophys. Res. Commun.. 2014;446(3):692-696 Crossref
  • Ontaneda and Fox, 2015 D. Ontaneda, R.J. Fox. Progressive multiple sclerosis. Curr. Opin. Neurol.. 2015;28(3):237-243 Crossref
  • Palavra et al., 2013 F. Palavra, D. Marado, F. Mascarenhas-Melo, J. Sereno, E. Teixeira-Lemos, C.C. Nunes, et al. New markers of early cardiovascular risk in multiple sclerosis patients: oxidized-LDL correlates with clinical staging. Dis. Markers. 2013;34(5):341-348 Crossref
  • Pedersen, 1974 H.E. Pedersen. Cerebrospinal fluid cholesterols and phospholipids in multiple sclerosis. Acta Neurol. Scand.. 1974;50(2):171-182
  • Pihl-Jensen et al., 2015 G. Pihl-Jensen, A. Tsakiri, J.L. Frederiksen. Statin treatment in multiple sclerosis: a systematic review and meta-analysis. CNS Drugs. 2015;29(4):277-291 Crossref
  • Pirttilä et al., 2000 T. Pirttilä, M. Haanpää, P.D. Mehta, T. Lehtimäki. Apolipoprotein E (APOE) phenotype and APOE concentrations in multiple sclerosis and acute herpes zoster. Acta Neurol. Scand.. 2000;102(2):94-98
  • Plum, 1960 C.M. Plum. The cholesterol content of cerebrospinal fluid with special regard to the occurrence in cases with multiple sclerosis. Acta Psychiatr. Scand. Suppl.. 1960;35(148):S79-S100
  • Poirier et al., 1991 J. Poirier, M. Hess, P.C. May, C.E. Finch. Cloning of hippocampal poly(A) RNA sequences that increase after entorhinal cortex lesion in adult rat. Brain Res. Mol. Brain Res.. 1991;9(3):191-195 Crossref
  • Poli et al., 2013 G. Poli, F. Biasi, G. Leonarduzzi. Oxysterols in the pathogenesis of major chronic diseases. Redox Biol.. 2013;1:125-130 Crossref
  • Poser et al., 1983 C.M. Poser, D.W. Paty, L. Scheinberg, W.I. McDonald, F.A. Davis, G.C. Ebers, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann. Neurol. 1983;13(3):227-231 Crossref
  • Poynder and Russell, 1926 E.G.T. Poynder, J. Russell. The cholesterol content of CSF in certain mental diseases. J. Ment. Sci.. 1926;72:62 Crossref
  • Qin et al., 2014 L. Qin, Y.B. Yang, Y.X. Yang, N. Zhu, S.X. Li, D.F. Liao, et al. Anti-inflammatory activity of ezetimibe by regulating NF-κB/MAPK pathway in THP-1 macrophages. Pharmacology. 2014;93(1–2):69-75 Crossref
  • Ramsey et al., 1972 R.B. Ramsey, R.T. Aexel, J.P. Jones, H.J. Nicholas. Formation of methyl sterols in brain cholesterol biosynthesis. Sterol formation in vitro in actively myelinating rat brain. J. Biol. Chem.. 1972;247(11):3471-3475
  • Rassart et al., 2000 E. Rassart, A. Bedirian, S. Do Carmo, O. Guinard, J. Sirois, L. Terrisse, et al. Apolipoprotein D.. Biochim. Biophys. Acta. 2000;1482(1–2):185-198 Crossref
  • Reindl et al., 2001 M. Reindl, G. Knipping, I. Wicher, E. Dilitz, R. Egg, F. Deisenhammer, et al. Increased intrathecal production of apolipoprotein D in multiple sclerosis. J. Neuroimmunol.. 2001;119(2):327-332 Crossref
  • Rifai et al., 1987 N. Rifai, R.H. Christenson, B.B. Gelman, L.M. Silverman. Changes in cerebrospinal fluid IgG and apolipoprotein E indices in patients with multiple sclerosis during demyelination and remyelination. Clin. Chem.. 1987;33(7):1155-1157
  • Roy and Pahan, 2009 A. Roy, K. Pahan. Gemfibrozil, stretching arms beyond lipid lowering. Immunopharmacol. Immunotoxicol.. 2009;31:339-351 Crossref
  • Shafaati et al., 2011 M. Shafaati, A. Marutle, H. Pettersson, A. Lövgren-Sandblom, M. Olin, I. Pikuleva, et al. Marked accumulation of 27-hydroxycholesterol in the brains of Alzheimer’s patients with the Swedish APP 670/671 mutation. J. Lipid Res.. 2011;52(5):1004-1010 Crossref
  • Shah and Johnson, 1980 S.N. Shah, R.C. Johnson. Activity levels of cholesterol ester metabolizing enzymes in brain in multiple sclerosis: correlation with cholesterol ester concentrations. Exp. Neurol.. 1980;68(3):601-604 Crossref
  • Swank and Dugan, 1990 R.L. Swank, B.B. Dugan. Effect of low saturated fat diet in early and late cases of multiple sclerosis. Lancet. 1990;336(8706):37-39 Crossref
  • Tabas et al., 2007 I. Tabas, K.J. Williams, J. Borén. Subendothelial lipoprotein retention as the initiating process in atherosclerosis: update and therapeutic implications. Circulation. 2007;116(16):1832-1844 Crossref
  • Tettey et al., 2014a P. Tettey, S. Simpson Jr, B. Taylor, L. Blizzard, A.L. Ponsonby, T. Dwyer, et al. An adverse lipid profile is associated with disability and progression in disability, in people with MS. Mult. Scler.. 2014;20(13):1737-1744 Crossref
  • Tettey et al., 2014b P. Tettey, S. Simpson Jr, B. Taylor, L. Blizzard, A.L. Ponsonby, T. Dwyer, et al. Adverse lipid profile is not associated with relapse risk in MS: results from an observational cohort study. J. Neurol. Sci.. 2014;340(1–2):230-232 Crossref
  • Tettey et al., 2014c P. Tettey, S., Jr. Simpson, B.V. Taylor, I.A. van der Mei. Vascular comorbidities in the onset and progression of multiple sclerosis. J. Neurol. Sci. 2014;347(1-2):23-33 Crossref
  • Teunissen et al., 2005 C.E. Teunissen, C. Dijkstra, C. Polman. Biological markers in CSF and blood for axonal degeneration in multiple sclerosis. Lancet Neurol.. 2005;4(1):32-41 Crossref
  • Teunissen et al., 2003 C.E. Teunissen, C.D. Dijkstra, C.H. Polman, E.L. Hoogervorst, K. von Bergmann, D. Lütjohann. Decreased levels of the brain specific 24S-hydroxycholesterol and cholesterol precursors in serum of multiple sclerosis patients. Neurosci. Lett.. 2003;347(3):159-162 Crossref
  • Teunissen et al., 2007 C.E. Teunissen, S. Floris, M. Sonke, C.D. Dijkstra, H.E. De Vries, D. Lütjohann. 24S-hydroxycholesterol in relation to disease manifestations of acute experimental autoimmune encephalomyelitis. J. Neurosci. Res.. 2007;85(7) 1499-05
  • van de Kraats et al., 2014 C. van de Kraats, J. Killestein, V. Popescu, E. Rijkers, H. Vrenken, D. Lütjohann, et al. Oxysterols and cholesterol precursors correlate to magnetic resonance imaging measures of neurodegeneration in multiple sclerosis. Mult. Scler.. 2014;20(4):412-417 Crossref
  • Vanmierlo et al., 2015 T. Vanmierlo, J.F. Bogie, J. Mailleux, J. Vanmol, D. Lütjohann, M. Mulder, et al. Plant sterols: friend or foe in CNS disorders?. Prog. Lipid Res.. 2015;58:26-39 Crossref
  • Vuletic et al., 2005 S. Vuletic, E.R. Peskind, S.M. Marcovina, J.F. Quinn, M.C. Cheung, H. Kennedy, et al. Reduced CSF PLTP activity in Alzheimer’s disease and other neurologic diseases; PLTP induces ApoE secretion in primary human astrocytes in vitro. J. Neurosci. Res.. 2005;80(3):406-413 Crossref
  • Vuletic et al., 2014 S. Vuletic, H. Kennedy, J.J. Albers, J. Killestein, H. Vrenken, D. Lütjohann, et al. Cerebrospinal fluid apolipoprotein E and phospholipid transfer protein activity are reduced in multiple sclerosis; relationships with the brain MRI and CSF lipid variables. Mult. Scler. Relat. Disord.. 2014;3(4):533-541 Crossref
  • Wei et al., 2013 J. Wei, M. Zheng, P. Liang, Y. Wei, X. Yin, Y. Tang, et al. Apolipoprotein E and its mimetic peptide suppress Th1 and Th17 responses in experimental autoimmune encephalomyelitis. Neurobiol. Dis.. 2013;56:59-65 Crossref
  • Weinstock-Guttman et al., 2011a B. Weinstock-Guttman, R. Zivadinov, M. Ramanathan. Inter-dependence of vitamin D levels with serum lipid profiles in multiple sclerosis. J. Neurol. Sci.. 2011;311(1–2):86-91 Crossref
  • Weinstock-Guttman et al., 2011b B. Weinstock-Guttman, R. Zivadinov, N. Mahfooz, E. Carl, A. Drake, J. Schneider, et al. Serum lipid profiles are associated with disability and MRI outcomes in multiple sclerosis. J. Neuroinflammation. 2011;8:127 Crossref
  • Weinstock-Guttman et al., 2013a B. Weinstock-Guttman, D. Horakova, R. Zivadinov, M. Tamaño-Blanco, D. Badgett, M. Tyblova, et al. Interactions of serum cholesterol with anti-herpesvirus responses affect disease progression in clinically isolated syndromes. J. Neuroimmunol.. 2013;263(1–2):121-127 Crossref
  • Weinstock-Guttman et al., 2013b B. Weinstock-Guttman, R. Zivadinov, D. Horakova, E. Havrdova, J. Qu, G. Shyh, et al. Lipid profiles are associated with lesion formation over 24 months in interferon-β treated patients following the first demyelinating event. J. Neurol. Neurosurg. Psychiatry. 2013;84(11):1186-1191 Crossref
  • Witztum and Steinberg, 1991 J.L. Witztum, D. Steinberg. Role of oxidized low density lipoprotein in atherogenesis. J. Clin. Invest.. 1991;88(6):1785-1792 Crossref
  • Yao, 1988 J.K. Yao. Biosynthesis of membrane cholesterol during peripheral nerve development, degeneration and regeneration. Lipids. 1988;23(9):857-862 Crossref
  • Yu et al., 1982 R.K. Yu, K. Ueno, G.H. Glaser, W.W. Tourtellotte. Lipid and protein alterations of spinal cord and cord myelin of multiple sclerosis. J. Neurochem.. 1982;39(2):464-477 Crossref


a Department of Clinical Neurosciences, University of Calgary, Calgary, Canada

b Hotchkiss Brain Institute, University of Calgary, Calgary, Canada

c Division of Neurology, Faculty of Medicine, University of British Columbia, Vancouver, Canada

d Neurochemistry Laboratory and Biobank, Department of Clinical Chemistry, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, The Netherlands

e Department of Neuroscience, Centre de recherche du CHU de Québec – Université Laval, Québec, Canada

f Department of Molecular Medicine, Faculty of Medicine, Université Laval, Québec, Canada

Correspondence to: Centre de recherche du CHU de Québec – Université Laval, Pavillon CHUL, 2705 boul Laurier, Québec, Canada G1V 4G2.