Abstract
Purpose
To investigate levels of oxysterols in healthy control (HC) and multiple sclerosis (MS) patients and their inter-dependence with demographic, clinical characteristics and cholesterol biomarkers.
Methods
This study included 550 subjects (203 HC, 221 relapsing-remitting MS (RR-MS), 126 progressive MS (P-MS)). A complete lipid profile including total cholesterol (TC), high-density lipoprotein-cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C), apolipoproteins (Apo) A1, A2, B, E, C-reactive protein (CRP), 24-hydroxycholesterol (24HC), 25HC, 27HC, 7αHC and 7-ketocholesterol (7KC) was obtained. Lipoprotein particle sizing by H1 NMR was available for 432 subjects.
Results
The levels of 24HC, 27HC, and 7αHC (all p < 0.015) were lower in MS compared to HC, and 7KC was higher in P-MS compared to RR-MS (p < 0.001). TC, LDL-C and ApoB were associated with higher levels of all oxysterols (all p < 0.05) in HC. In MS, LDL-C was associated with higher levels of 24HC, 25HC, 7KC, and 7αHC (all p < 0.05), while TC and ApoB were associated with increased levels of all oxysterols (all p < 0.005).
Conclusions
The findings of lower 24HC, 27HC and 7αHC in MS compared to HC and higher 7KC in P-MS compared to RR-MS indicate that the oxysterol network is disrupted in MS.
INTRODUCTION
Multiple sclerosis (MS) is a chronic inflammatory disease characterized by demyelination in the central nervous system (CNS). Cholesterol is required for myelin structure and proper functioning of neuronal, vascular and immune cells in the CNS. About 25% of cholesterol in the human body is found in the brain.
The brain is dependent on de novo cholesterol synthesis from acetyl-coA because lipoprotein cholesterol from the peripheral circulation does not cross the blood-brain barrier (BBB). In our previous work, we have shown that higher levels of high-density lipoprotein cholesterol (HDL-C) are associated with lower measures of BBB breakdown, assessed using CSF measures of macromolecules, and CNS inflammation 1. Additionally, we found a significant association between low-density lipoprotein cholesterol (LDL-C) and total cholesterol (TC) levels and inflammatory MRI activity measures2, as well as worsening disability measured using the Expanded Disability Status Scale (EDSS) and the MS Severity Scale (MSSS) (10–13). However, the mechanisms underlying the associations of cholesterol biomarkers with disease progression in MS are unknown.
Oxysterols are oxygenated cholesterol metabolites that can cross the BBB, and are therefore thought to play an important role in brain cholesterol regulation3–5. Oxysterols such as 24HC, 27HC and 25HC are endogenous agonists for the liver X receptor (LXR), a nuclear receptor involved in regulating metabolic and immune processes6. 24HC is exclusively produced in the brain via cytochrome P450 (CYP) 46A1 in neurons, and is able to traverse the BBB; its plasma levels reflect neuronal loss and brain atrophy4,5. 27HC, produced by CYP27A1, also has the ability to cross the BBB, and is the most abundant oxysterol in the circulation5,7. Both 7αHC, formed via CYP7A, and 7KC, formed by auto-oxidation of cholesterol, stimulate an inflammatory phenotype in human endothelial cells. 25HC, which is produced by macrophages, has inflammatory and antiviral effects8.
Oxysterols thus regulate cholesterol homeostasis, enhance inflammatory reactions, contribute to endothelial cell dysfunction, and exhibit pro-oxidant effects7,9. There is now increasing evidence that oxysterols play a role in the pathophysiological processes mediating MS disease progression10. Based on these findings, our working hypothesis was that levels of oxysterols, particularly 24HC, 25HC, 27HC, 7KC and 7aHC, would be altered by MS disease course. The goals of this translational study are to identify the demographic, clinical and cholesterol biomarker covariates associated with oxysterols profiles in MS patients.
METHODS
Study Population
Study Setting
Single-center, prospective observational study.
Clinical Study Design
Samples and clinical data from an ongoing prospective study of clinical, genetic and environmental risk factors in MS at the MS Center of the State University of New York at Buffalo were used. The study included patients with MS, clinically isolated syndrome (CIS), healthy controls (HC) and controls with other neurological diseases (OND).
Patients and controls underwent neurological and MRI examinations, provided blood samples and responded to a structured questionnaire administered in-person 11. Body mass index (BMI) was derived from the patient’s height and weight. Smoking status was obtained from patient responses to the structured questionnaire.
Informed Consent
The University at Buffalo Human Subjects Institutional Review Board approved the study protocol, and all participants provided written informed consent.
Inclusion and Exclusion Criteria
Patients with the following characteristics: 18–65 years of age. Children younger than 18 years, adults over 65 years of age, CIS, neuromyelitis optica, or OND were excluded from this sub-study.
Oxysterol Assays
Oxysterols (7αHC, 24HC, 25HC, 27HC, 7KC) were determined in EDTA plasma samples by liquid chromatography-selected ion monitoring-stable isotope dilution-atmospheric pressure chemical ionization-mass spectroscopy (LC-SIM-SID-APCI-MS) as previously described 12.
Total sterols and oxysterols were released from steryl esters by cold saponification (0.35M KOH in ethanol containing 0.05% BHT and under argon atmosphere), neutralized and extracted on HyperSep C-18 solid phase extraction tubes before being concentrated and injected into a reversed phase, C-18 LC system using a gradient mobile phase of methanol and water. The Shimadzu 10ADVP LC system was interfaced with the 2010A MS by APCI interface in positive ion mode. Oxysterols were monitored by selected ion monitoring of the [M+H-H2O]+ molecular ion of the hydroxyl groups present on each (oxy)sterol. 7KC was monitored as the [M+H]+ ion of the 7-keto group. Deuterated (d7) 24-hydroxycholesterol, (d7) 7α-hydroxycholesterol and (d7) 7-ketocholesterol were used as stable isotope-labeled internal standards.
We, and others13, have found that oxysterol analytes are stable at −80°C storage. The mild room temperature saponification in the presence of butylated hydroxytoluene (BHT) under argon ameliorated autooxidation of cholesterol or degradation of pre-formed oxysterols, particularly 7KC. This method successfully resolves the most critical isomeric pairs including separation of 24HC from 25HC, 7αHC from 7βHC and 7KC from 7α-hydroxy cholesten-3-one12. The assay was systematically validated in accordance with the FDA guidance for bioanalytical methods. The coefficient of variation (CV) of each oxysterol analyte was ≤ 8.7%.
Lipids and Apolipoproteins
EDTA plasma samples for lipid and apolipoprotein analyses were obtained in the non-fasted state. Our previously published 14 methods for lipid and apolipoprotein analyses were used. Analysts were blinded to the clinical status of samples.
TC, HDL-C, phospholipids and triglycerides were measured with diagnostic reagent kits from Sekisui Diagnostics (Lexington, MA) an ABX Pentra 400 (Horiba Instruments, Irvine, CA) automated chemistry analyzer. LDL-C was obtained from the Friedewalde equation15. Immunoturbidometric diagnostic kits (Kamiya Biomedical, Thousand Oaks, CA) were used for the apolipoprotein (AI, AII, B and E) and high sensitivity C-reactive protein (CRP) assays. The coefficient of variation of these assays is < 5%.
Lipoprotein Particle Size Analysis
Lipoprotein particle number and size were measured using H1 NMR spectroscopy (LipoScience, Raleigh, NC). The measured amplitudes of the lipid methyl group NMR signals is used to calculate particle concentration, and the particles sizes are the sum of the diameter of each subclass multiplied by the relative mass percentages from the NMR signals.
Data Analysis
SPSS (IBM Inc., Armonk, NY, version 22.0) statistical program was used. In view of the multiple testing, the Benjamini-Hochberg method was used to assess significance with a target false discovery rate of q ≤ 0.0516. The Tables and Results summarize the raw, unadjusted p-values. Adjusted p-values (q-values) are shown only for variables with unadjusted p-values ≤ 0.05.
Secondary progressive (SP) and primary-progressive (PP) MS were categorized as progressive MS (P-MS)17.
The Mann-Whitney test was used to assess differences in oxysterol levels (24HC, 25HC, 27HC, 7αHC, 7KC) for dichotomous variables, e.g., between HC vs. MS patients RR-MS vs. PMS, and MS patients with borderline or high cholesterol (defined as TC > 200 mg/dl) vs. desirable cholesterol levels (TC ≤ 200 mg/dl),.
The associations of oxysterol variables (24HC, 25HC, 27HC, 7αHC, and 7KC) with lipid profile variables (HDL-C, LDL-C, TC, ApoA-I, ApoA-II, ApoB, ApoE and CRP) were investigated using linear regression analyses where the lipid variable of interest was the dependent variable and age, gender, BMI and quartiles of either 24HC, 25HC, 27HC, 7αHC, or 7KC were treated as covariates.
The Kruskal-Wallis test was used to assess whether oxysterol levels were significantly different between patients receiving interferon, glatiramer acetate or no current treatment. A Mann-Whitney test was used to assess whether oxysterols differed between patients who were presently receiving interferon and those who were not currently on an interferon regimen.
RESULTS
Demographic and Clinical Characteristics
The clinical, demographic and MRI characteristics of the study sample at baseline in healthy controls, RR-MS and P-MS are summarized in Table 1. The age, disability and MRI differences between RR-MS and P-MS groups are representative of the respective disease courses. The distribution of oxysterol levels and baseline lipid profiles (TC, HDL-C, and LDL-C) are shown in Table 2.
Table 1.
Demographic and clinical characteristics at baseline by disease course.
| Characteristic | All | HC | RR-MS | PMS |
|---|---|---|---|---|
| Sample size n | 550 | 203 | 221 | 126 |
| % Female | 377 (69%) | 120 (59%) | 167 (76%) | 90 (71%) |
| Age, years | 47.1 ± 12.0 | 45.6 ± 14 | 44.4 ± 10 | 54.3 ± 8.5 |
| Disease duration, years | 15.7 ± 10.6 | – | 12.2 ± 8.7 | 21.7 ± 11.1 |
| BMI | 25.8 ± 2.0 | 27.5 ± 5.8 | 27.2 ± 6.1 | 25.4 ± 5.2 |
| EDSSa | 3.0 (4.0) | – | 2.0 (2.0) | 6.0 (2.0) |
All continuous variables (age, disease duration) are mean ± standard deviation. For the ordinal EDSS, the median (inter-quartile range) is given.
Table 2.
Baseline oxysterol and lipid values by disease course.
| Characteristic | All Subjects | HC | RR-MS | PMS |
|---|---|---|---|---|
| n | 550 | 203 | 221 | 126 |
| 24HC, ng/ml | 40.7 ± 36 | 44.4 ± 44. | 40.4 ± 34 | 35.3 ± 20 |
| 25HC, ng/ml | 17.3 ± 22 | 17.8 ± 17 | 15.9 ± 11 | 18.6 ± 38 |
| 27HC, ng/ml | 235.5 ± 121 | 251.1 ± 117 | 233.2 ± 128 | 214.5 ± 114 |
| 7αHC, ng/ml | 88.7 ± 130 | 116.6 ± 188 | 73.4 ± 72 | 69.9 ± 74 |
| 7KC, ng/ml | 14.7 ± 11 | 14.0 ± 7 | 13.5 ± 8 | 18.1 ± 17 |
| LDL-C, mg/dL | 92.9 ± 27 | 95.7 ± 28 | 90.8 ± 28 | 91.9 ± 25 |
| HDLC, mg/dl | 68.0 ± 15 | 68.7 ± 16 | 66.8 ± 14 | 68.8 ± 15 |
| TC, mg/ml | 188.4 ± 35 | 193.0 ± 36 | 185.3 ± 35 | 186.2 ± 33 |
All variables are mean ± standard deviation. 24HC: 24 hydroxy cholesterol; 25HC: 25 hydroxy cholesterol; 27HC: 27 hydroxy cholesterol; 7aHC: 7a hydroxyl cholesterol; 7KC: 7 ketocholesterol; LDL-C: low-density lipoprotein cholesterol; HDL-C: high-density lipoprotein cholesterol.
Associations of Oxysterols with Clinical Factors in MS Patients
MS Disease Course
Figure 1 compares the baseline oxysterol levels between HC vs. MS and RR-MS vs. P-MS groups. The levels of 24HC (p = 0.015), 27HC (p = 0.004), and 7αHC (p < 0.001) were significantly lower in MS patients compared to HC. There were no differences in 25HC and 7KC levels between HC and RR-MS patients.
Figure 1.
Oxysterol levels in healthy controls (HC), relapsing remitting MS (RR-MS) and progressive MS (P-MS). Figure 1 shows the values of 24HC (Figure 1A), 25HC (Figure 1B), 27HC (Figure 1C), 7KC (Figure 1D), and 7αHC (Figure 1E) in ng/ml, by disease course (HC, RR-MS and P-MS). The bars compare mean oxysterol values shown on the y-axis by disease course shown on the x-axis. The error bars indicate the standard error of the mean. The p-values from a Mann-Whitney test are provided for the HC versus MS and RR-MS versus P-MS comparisons.
7KC levels were significantly higher in P-MS patients compared to RR-MS patients (p < 0.001). No differences were found between the two MS groups for 24HC, 25HC, 27HC, and 7αHC.
Disability Measures
The associations of oxysterols with disability measures (MSSS and EDSS) in both RR-MS and P-MS patients were investigated (data not shown). Increased levels of 24HC were significantly associated with MSSS (p = 0.045) in P-MS patients. There were no significant associations between any other oxysterol levels and MSSS and EDSS in the P-MS patient population. No associations were found between oxysterol levels and disability measures in RR-MS patients.
Disease-Modifying Therapies
We compared the levels of oxysterols between patients currently being treated with interferon and patients who are not currently being treated with interferon. The differences in 25HC and 7αHC levels between untreated, glatiramer acetate treated and interferon treated patients are summarized in Figure 2. We found that 25HC (p = 0.038) and 7αHC (p = 0.003) were significantly higher in interferon-treated patients. No significant differences were observed for 24HC, 27HC and 7KC.
Figure 2.
25HC and 7αHC levels (ng/ml) in MS patients on no treatment, glatiramer acetate or IFN treatment. The error bars indicate the standard error of the mean.
Associations of Oxysterols with Lipid Profile Variables
The associations between oxysterols and lipid variables (TC, HDLC, LDLC), apolipoproteins (ApoAI, ApoAII, ApoB, ApoE) and C-reactive protein (CRP) in both HC and MS patients are summarized in Table 3.
Table 3.
Associations between oxysterols and lipid profile variables in healthy control and MS patients. The partial correlation coefficient (p-value) are shown
| Lipid Variable | HC vs. MS | 24HC | 25HC | 27HC | 7KC | 7αHC |
|---|---|---|---|---|---|---|
| TC | HC | 0.37 (< 0.001) | 0.27 (< 0.001) | 0.35 (< 0.001) | 0.34 (< 0.001) | 0.52 (< 0.001) |
|
| ||||||
| MS | 0.26 (< 0.001) | 0.24 (< 0.001) | 0.16 (0.005) | 0.24 (< 0.001) | 0.35 (< 0.001) | |
|
| ||||||
| HDLC | HC | −0.10 (0.17) | 0.113 (0.13) | 0.05 (0.49) | 0.02 (0.79) | 0.10 (0.18) |
|
| ||||||
| MS | −0.056 (0.32) | 0.089 (0.11) | −0.06 (0.33) | −0.015 (0.79) | 0.06 (0.28) | |
|
| ||||||
| LDLC | HC | 0.39 (< 0.001) | 0.17 (0.024) | 0.25 (0.001) | 0.21 (0.005) | 0.36 (< 0.001) |
|
| ||||||
| MS | 0.22 (< 0.001) | 0.12 (0.028) | 0.10 (0.09) | 0.12 (0.03) | 0.21 (< 0.001) | |
|
| ||||||
| ApoAI | HC | 0.025 (0.73) | 0.13 (0.081) | 0.10 (0.17) | 0.013 (0.86) | 0.24 (0.001) |
|
| ||||||
| MS | −0.047 (0.40) | 0.06 (0.28) | −0.075 (0.18) | −0.07 (0.21) | 0.10 (0.077) | |
|
| ||||||
| ApoAII | HC | 0.19 (0.012) | 0.051 (0.49) | 0.26 (< 0.001) | 0.16 (0.03) | 0.35 (< 0.001) |
|
| ||||||
| MS | 0.091 (0.10) | 0.087 (0.12) | 0.12 (0.030) | −0.40 (0.48) | 0.19 (0.001) | |
|
| ||||||
| ApoB | HC | 0.32 (< 0.001) | 0.29 (< 0.001) | 0.39 (< 0.001) | 0.32 (< 0.001) | 0.36 (< 0.001) |
|
| ||||||
| MS | 0.29 (< 0.001) | 0.27 (< 0.001) | 0.31 (< 0.001) | 0.22 (< 0.001) | 0.21 (< 0.001) | |
|
| ||||||
| ApoE | HC | −0.14 (0.069) | −0.089 (0.23) | −0.093 (0.21) | −0.09 (0.26) | −0.10 (0.18) |
|
| ||||||
| MS | −0.07 (0.22) | −0.032 (0.57) | −0.024 (0.67) | 0.08 (0.16) | 0.05 (0.35) | |
|
| ||||||
| CRP | HC | 0.011 (0.88) | 0.042 (0.57) | 0.083 (0.26) | 0.08 (0.30) | 0.06 (0.42) |
|
| ||||||
| MS | 0.04 (0.47) | 0.075 (0.18) | −0.003 (0.96) | 0.03 (0.64) | −0.007 (0.90) | |
TC
Higher levels of all five of the oxysterols were associated with TC for both HC and MS populations (all p < 0.001 except 27HC in MS patients, p = 0.005).
In the HC group, 25HC (p = 0.007) and 7αHC (p = 0.028) were greater in subjects with borderline and high TC (TC > 200 mg/dl). MS patients with borderline and high TC had greater 24HC (p = 0.011), 25HC (p = 0.019) and 27HC (p = 0.002).
LDL-C
LDL-C was associated with higher levels of 24HC (p < 0.001), 25HC (p = 0.024), 27HC (p = 0.001), 7KC (p = 0.005) and 7αHC (p < 0.001) in HC. In MS patients, LDL-C was significantly associated with higher levels of 24HC (p < 0.001), 25HC (p = 0.028), 7KC (p = 0.030) and 7αHC (p < 0.001).
HDL-C
None of the oxysterols were significantly associated with HDL-C in either HC or MS patients. Thus, the associations of oxysterols with TC are principally driven by the associations of oxysterols with LDL-C. To further support this hypothesis, we conducted linear regression analyses that corrected for HDL-C values. The significant findings persisted.
Apolipoproteins
Apolipoproteins are functionally critical components of lipoproteins that enable physiological interactions. ApoB is the major apolipoprotein of LDL-C, whereas ApoAI and ApoAII are characteristic of HDL-C.
ApoB was significantly associated with increased levels of all oxysterols in both HC and MS patients (all p < 0.001), which is concordant with the LDL-C associations. ApoAII was associated with 27HC and 7αHC in both HC and MS. ApoAI was associated with 7αHC in HC. We did not obtain evidence for associations with ApoE and CRP.
Associations of Oxysterols with Lipoprotein Particle Sub-Classes
To further delineate the associations between oxysterols and lipoprotein particle sub-classes, we examined lipoprotein particle concentrations obtained by NMR particle sizing. The results are summarized in Table 4.
Table 4.
Associations between oxysterols and lipoprotein particle sub-class concentrations in HC and MS patients.
| Lipoprotein | HC vs. MS | 24HC | 25HC | 27HC | 7KC | 7αHC |
|---|---|---|---|---|---|---|
| Total VLDL | HC | 0.033 (0.69) | 0.22 (0.007) | 0.098 (0.23) | 0.13 (0.11) | 0.24 (0.003) |
|
| ||||||
| MS | 0.16 (0.009) | 0.22 (< 0.001) | 0.19 (0.002) | 0.13 (0.030) | 0.17 (0.005) | |
|
| ||||||
| Large VLDL | HC | 0.19 (0.019) | 0.29 (< 0.001) | 0.27 (0.001) | 0.29 (< 0.001) | 0.51 (< 0.001) |
|
| ||||||
| MS | 0.19 (0.002) | 0.20 (0.001) | 0.18 (0.003) | 0.18 (0.003) | 0.32 (< 0.001) | |
|
| ||||||
| Medium VLDL | HC | 0.064 (0.44) | 0.26 (0.001) | 0.17 (0.042) | 0.32 (< 0.001) | 0.42 (< 0.001) |
|
| ||||||
| MS | 0.088 (0.15) | 0.18 (0.002) | 0.14 (0.025) | 0.19 (0.002) | 0.29 (< 0.001) | |
|
| ||||||
| Small VLDL | HC | −0.045 (0.59) | 0.085 (0.30) | −0.017 (0.84) | −0.085 (0.30) | −0.051 (0.53) |
|
| ||||||
| MS | 0.11 (0.078) | 0.13 (0.039) | 0.11 (0.067) | 0.053 (0.39) | 0.011 (0.86) | |
|
| ||||||
| Total LDL | HC | 0.21 (0.011) | 0.30 (< 0.001) | 0.45 (< 0.001) | 0.36 (< 0.001) | 0.46 (< 0.001) |
|
| ||||||
| MS | 0.34 (< 0.001) | 0.32 (< 0.001) | 0.34 (< 0.001) | 0.23 (< 0.001) | 0.30 (< 0.001) | |
|
| ||||||
| Total IDL | HC | 0.056 (0.49) | 0.18 (0.029) | 0.10 (0.21) | 0.092 (0.26) | 0.20 (0.012) |
|
| ||||||
| MS | 0.041 (0.50) | 0.15 (0.012) | 0.088 (0.15) | 0.064 (0.29) | 0.008 (0.89) | |
|
| ||||||
| Large LDL | HC | 0.25 (0.002) | 0.11 (0.17) | 0.21 (0.010) | 0.068 (0.41) | 0.092 (0.26) |
|
| ||||||
| MS | 0.11 (0.073) | 0.068 (0.27) | 0.031 (0.61) | 0.063 (0.30) | −0.038 (0.53) | |
|
| ||||||
| Small LDL | HC | 0.032 (0.70) | 0.20 (0.015) | 0.30 (< 0.001) | 0.34 (< 0.001) | 0.41 (< 0.001) |
|
| ||||||
| MS | 0.22 (< 0.001) | 0.23 (< 0.001) | 0.27 (< 0.001) | 0.15 (0.011) | 0.33 (< 0.001) | |
|
| ||||||
| Total HDL | HC | −0.081 (0.32) | 0.009 (0.91) | −0.057 (0.49) | 0.050 (0.54) | 0.16 (0.047) |
|
| ||||||
| MS | 0.051 (0.40) | 0.056 (0.36) | 0.070 (0.25) | 0.015 (0.80) | 0.074 (0.23) | |
|
| ||||||
| Large HDL | HC | 0.071 (0.39) | −0.13 (0.13) | −0.12 (0.13) | −0.13 (0.12) | −0.090 (0.27) |
|
| ||||||
| MS | 0.003 (0.97) | −0.029 (0.64) | −0.14 (0.020) | 0.021 (0.74) | −0.087 (0.16) | |
|
| ||||||
| Medium HDL | HC | −0.024 (0.77) | 0.033 (0.68) | −0.065 (0.42) | 0.001 (0.99) | 0.11 (0.18) |
|
| ||||||
| MS | 0.058 (0.34) | 0.005 (0.94) | 0.081 (0.19) | −0.008 (0.90) | 0.060 (0.33) | |
|
| ||||||
| Small HDL | HC | −0.13 (0.11) | 0.14 (0.098) | 0.12 (0.15) | 0.208 (0.010) | 0.26 (0.001) |
|
| ||||||
| MS | 0.036 (0.56) | 0.10 (0.099) | 0.15 (0.013) | 0.039 (0.53) | 0.12 (0.045) | |
Values provided represent the Spearman correlation coefficient with p-values in parenthesis.
Overall, the oxysterols were associated with the VLDL and also, as expected from earlier analyses, with LDL lipoprotein particle concentrations in both MS and HC.
VLDL Particle Sub-Classes
Within the VLDL particle sub-classes, 25HC, 27HC, 7KC and 7αHC were associated with large and medium VLDL particle concentrations whereas levels of 24HC were only associated with large VLDL particle concentrations for both MS and HC.
LDL-C Particle Sub-classes
Within the LDL particle sub-classes, 25HC, 27HC, 7KC and 7αHC were associated with small LDL particle concentrations for both MS and HC. 24HC was associated with small LDL particle concentrations in MS but not for HC.
HDL-C Particle Sub-classes
By comparison, associations of oxysterols with HDL particle sub-class concentrations were weaker and sporadic. However, 7αHC was associated with small HDL particle concentrations for both MS and HC.
DISCUSSION
We investigated whether the levels of 24HC, 25HC, 27HC, 7αHC and 7KC differed between MS and HC and between RR-MS and P-MS, and whether the levels of these oxysterols were associated with lipid variables (HDL-C, LDL-C, TC, ApoAI, ApoAII, ApoB, ApoE and CRP) and demographic variables. We found that the levels of 24HC, 27HC and 7αHC were significantly lower in MS patients compared to HC, and that 7KC levels were significantly higher in P-MS patients compared to RR-MS patients, which supports findings from previous studies 18,19. Additionally, oxysterol levels in both MS and HC were pre dominantly associated with levels of LDL-C and its cognate apolipoprotein ApoB. The large and medium VLDL and small LDL particle sub-classes exhibited the strongest associations with the majority of oxysterols. 24HC is produced exclusively in the CNS and we hypothesize that the reduction in 24HC levels in MS is a marker of CNS cholesterol metabolism; the decrease therefore reflects changes to brain cholesterol homeostasis as a result of neurodegeneration. In contrast, the reductions in 27HC and 7αHC observed in MS are a result of changes to cholesterol metabolism pathways in the periphery and could facilitate in pro-inflammatory changes that enable MS disease progression. The observed increases in 7KC in PMS compared to RRMS reflect oxidative stress that promotes neurodegeneration in PMS.
Results from animal models offer insights into the effector mechanisms of oxysterols. Oxysterols exert immunomodulatory effects on dendritic cell, T cell, macrophage and B cell function via LXR and LXR-independent, EBI2-mediated signaling pathways. Oxysterols amplify and sustain inflammatory processes20 and regulate the expression of inflammatory mediators such as IL-6 and matrix metalloproteinase-9 (MMP-9)21,22. LXR signaling is critical for polarization of CD4 helper T cells into Th17 cells in humans23 and it also potently suppresses T cell proliferation22,24. LXR activation ameliorates experimental allergic encephalomyelitis (EAE) whereas LXR deficiency exacerbates it23. Oxysterols inhibit EAE by down-regulating IL-17 production and suppressing Th17 cells23. Dendritic cells produce cholesterol 27 hydroxylase during EAE 25. Thus the reductions in 24HC, 27HC and 7αHC levels and the increases in 7KC levels that we observed could potentially contribute to pathophysiological immune effects in MS.
There is increasing evidence for associations of cholesterol with disease progression in MS. We26, and others27–29, have shown that increased levels of LDL-C and TC are significantly associated with worsening disability on the EDSS scale. LDL-C and TC levels were associated with new T2 lesions and grey matter atrophy2. Increased levels of HDL-C and ApoAI were associated with less BBB breakdown as assessed with surrogate CSF measures1.
The pleiotropic effects of oxysterols on metabolic and neuroimmune processes, coupled with their ability to permeate the BBB, has generated interest in investigating the role of oxysterols in MS. 24HC, 25HC and 27HC are LXR ligands that coordinately down-regulate cholesterol synthesis as well as up-regulate the transporters such as the ABCA1 transporter involved in the removal of cellular cholesterol 9. There is evidence that interferon induces 25-hydroxylase, the enzyme responsible for the conversion of cholesterol to 25HC 30. This may explain our finding that 25HC was higher in interferon-treated MS patients.
7KC is an oxidative stress marker that stimulates phenotypic changes in macrophages that promote, invasion of the vascular endothelium, and leads to secretion of inflammatory factors that can lead to the formation of MS lesions 31. Demyelinating MS lesions have been found to contain foam cells containing ingested myelin-derived lipids 32. Boven et al. 32 used an in vitro model of myelin ingestion to assess anti-inflammatory and pro-inflammatory markers based on location within the lesion. They found that the lipid-filled macrophages located at the center of the lesions expressed anti-inflammatory cytokines, such as IL-1ra, CCL18, IL-10, TGF-β and IL-4, but the cells did not express the pro-inflammatory cytokines such as TNF-α, IL-1β, IL-12p40/70, regardless of location within the lesion 32. These authors proposed that lipid-ingested macrophages may have anti-inflammatory properties in the CNS and may also contribute to repair. They may contribute to the characteristic partial recovery that follows exacerbations in patients with RR-MS 32.
In a cross-sectional analysis, van de Kraats et al. compared the serum and CSF levels of 24HC, 27HC and lanosterol between MS patients and HC, and investigated the associations with MRI measures. They found that levels of serum 24HC were negatively associated with normalized gray matter volume (p = 0.004) in RR-MS patients 33. Additional studies have shown that serum levels of 24HC and 27HC were significantly lower in MS patients compared to HC, while lanosterol levels were increased in HC relative to MS patients 18,19. In our analysis we found supporting evidence that MS patients have lower levels of 24HC and 27HC compared to HC. Our findings indicating that age is negatively correlated with 24HC levels in HC and that 7αHC is lower in females compared to males is concordant with the results from a study by Stiles et al., who measured an extensive panel of sterols in a multi-ethnic, population-based sample 34.
To further delineate the significant associations between oxysterol levels and TC, LDL-C and ApoB, we investigated whether oxysterols were associated with particular sub-classes of lipoprotein particles. To the best of our knowledge, this was the first assessment of oxysterol associations with lipoprotein particle concentrations in MS patients. We found that the large and medium VLDL and small LDL particle sub class concentrations were most significantly associated with oxysterols. These results are consistent with the cardiovascular disease literature, where previous examinations of the oxysterols in lipoproteins revealed that small dense LDL are associated with the highest levels of oxidized lipids and have the strongest pro-atherogenic effect 35.
Although we were able to investigate numerous lipid profile and clinical covariates, our study is limited by its cross sectional design and the lack of data on statins. However, we expect the intriguing results of this study in MS will provide the rationale for well-controlled longitudinal studies.
In conclusion, the findings that lower levels of 24HC, 27HC and 7αHC are seen in MS patients compared to HC, and that higher 7KC levels are observed in P-MS compared to RR-MS indicate that the oxysterol network is disrupted in MS. Our findings also highlight the interdependence between oxysterols and LDL-C, which may explain the evidence that supports a role for LDL-C in the pathophysiology of MS.
Acknowledgments
The authors thank the patients who participated in this study.
FUNDING INFORMATION
Support from the National Multiple Sclerosis Society (RG4836-A-5) and the National Institute of Neurological Disorders and Stroke (1R21NS098169) to the Ramanathan laboratory is gratefully acknowledged.
Footnotes
DISCLOSURE
Kelly Fellows, Shreya Mukhopadhyay, Prachi Khare and Sandhya Krishnan Radhakrishnan have nothing to disclose.
Dr. Robert Zivadinov has received speaker honoraria and consultant fees from Teva Pharmaceuticals, Biogen Idec, Genzyme-Sanofi, Novartis, Claret Medical and EMD Serono. He has received research support from the National Multiple Sclerosis Society, Department of Defense, Biogen Idec, Teva Neuroscience, Teva Pharmaceuticals, EMD Serono, Genzyme-Sanofi, Claret Medical and Novartis.
Dr. Bianca Weinstock-Guttman received honoraria for serving in advisory boards and educational programs from Teva Pharmaceuticals, Biogen Idec, Novartis, Accorda EMD Serono, Pfizer, Novartis, Genzyme and Sanofi. She also received support for research activities from the National Institutes of Health, National Multiple Sclerosis Society, National Science Foundation, Department of Defense, EMD Serono, Biogen Idec, Teva Neuroscience, Cyberonics, Novartis, Acorda and the Jog for the Jake Foundation.
Dr. Murali Ramanathan received research funding from the National Multiple Sclerosis Society and the Department of Defense. He received compensation for serving as an Editor from the American Association of Pharmaceutical Scientists. These are unrelated to the research presented in this report.
Financial Conflicts: See disclosure statement.
Confidentiality: Use of the information in this manuscript for commercial, non-commercial, research or purposes other than peer review not permitted prior to publication without expressed written permission of the author.
AUTHOR CONTRIBUTIONS
Kelly Fellows – Data acquisition and analysis, manuscript drafting.
Shreya Mukhopadhyay – Data acquisition and analysis, manuscript drafting.
Prachi Khare – Data acquisition
Sandhya Krishnan Radhakrishnan – Data acquisition
Richard W. Browne – Data acquisition, manuscript preparation.
Bianca Weinstock-Guttman,– Data interpretation, manuscript preparation
Robert Zivadinov – MRI data analysis, manuscript preparation
Murali Ramanathan – Study concept and design, data analysis, manuscript preparation.
References
- 1.Fellows K, Uher T, Browne RW, et al. Protective Associations of HDL with Blood Brain Barrier Injury in Multiple Sclerosis Patients. Journal of lipid research. 2015 doi: 10.1194/jlr.M060970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Weinstock-Guttman B, Zivadinov R, Horakova D, et al. Lipid profiles are associated with lesion formation over 24 months in interferon-beta treated patients following the first demyelinating event. Journal of neurology, neurosurgery, and psychiatry. 2013;84(11):1186–91. doi: 10.1136/jnnp-2012-304740. [DOI] [PubMed] [Google Scholar]
- 3.Chen J, Costa LG, Guizzetti M. Assessment of cholesterol homeostasis in astrocytes and neurons. Methods in molecular biology. 2011;758:403–14. doi: 10.1007/978-1-61779-170-3_27. [DOI] [PubMed] [Google Scholar]
- 4.Leoni V, Caccia C. 24S-hydroxycholesterol in plasma: a marker of cholesterol turnover in neurodegenerative diseases. Biochimie. 2013;95(3):595–612. doi: 10.1016/j.biochi.2012.09.025. [DOI] [PubMed] [Google Scholar]
- 5.Vitali C, Wellington CL, Calabresi L. HDL and cholesterol handling in the brain. Cardiovascular research. 2014;103(3):405–13. doi: 10.1093/cvr/cvu148. [DOI] [PubMed] [Google Scholar]
- 6.Olkkonen VM, Beaslas O, Nissila E. Oxysterols and their cellular effectors. Biomolecules. 2012;2(1):76–103. doi: 10.3390/biom2010076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Poli G, Biasi F, Leonarduzzi G. Oxysterols in the pathogenesis of major chronic diseases. Redox biology. 2013;1:125–30. doi: 10.1016/j.redox.2012.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gold ES, Diercks AH, Podolsky I, et al. 25-Hydroxycholesterol acts as an amplifier of inflammatory signaling. Proceedings of the National Academy of Sciences of the United States of America. 2014;111(29):10666–71. doi: 10.1073/pnas.1404271111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Spann NJ, Glass CK. Sterols and oxysterols in immune cell function. Nat Immunol. 2013;14(9):893–900. doi: 10.1038/ni.2681. [DOI] [PubMed] [Google Scholar]
- 10.Zhornitsky S, McKay KA, Metz LM, et al. Cholesterol and markers of cholesterol turnover in multiple sclerosis: relationship with disease outcomes. Multiple sclerosis and related disorders. 2016;5:53–65. doi: 10.1016/j.msard.2015.10.005. [DOI] [PubMed] [Google Scholar]
- 11.Dolic K, Weinstock-Guttman B, Marr K, et al. Risk factors for chronic cerebrospinal venous insufficiency (CCSVI) in a large cohort of volunteers. PloS one. 2011;6(11):e28062. doi: 10.1371/journal.pone.0028062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Narayanaswamy R, Iyer V, Khare P, et al. Simultaneous determination of oxysterols, cholesterol and 25-hydroxy-vitamin D3 in human plasma by LC-UV-MS. PloS one. 2015;10(4):e0123771. doi: 10.1371/journal.pone.0123771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Helmschrodt C, Becker S, Thiery J, et al. Preanalytical standardization for reactive oxygen species derived oxysterol analysis in human plasma by liquid chromatography-tandem mass spectrometry. Biochemical and biophysical research communications. 2014;446(3):726–30. doi: 10.1016/j.bbrc.2013.12.087. [DOI] [PubMed] [Google Scholar]
- 14.Browne RW, Weinstock-Guttman B, Horakova D, et al. Apolipoproteins are associated with new MRI lesions and deep grey matter atrophy in clinically isolated syndromes. Journal of neurology, neurosurgery, and psychiatry. 2014;85(8):859–64. doi: 10.1136/jnnp-2013-307106. [DOI] [PubMed] [Google Scholar]
- 15.Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clinical chemistry. 1972;18(6):499–502. [PubMed] [Google Scholar]
- 16.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc B. 1995;57:289–300. [Google Scholar]
- 17.Lublin FD, Reingold SC. Defining the clinical course of multiple sclerosis: results of an international survey. National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis. Neurology. 1996;46(4):907–11. doi: 10.1212/wnl.46.4.907. [DOI] [PubMed] [Google Scholar]
- 18.Leoni V, Masterman T, Diczfalusy U, et al. Changes in human plasma levels of the brain specific oxysterol 24S-hydroxycholesterol during progression of multiple sclerosis. Neuroscience letters. 2002;331(3):163–6. doi: 10.1016/s0304-3940(02)00887-x. [DOI] [PubMed] [Google Scholar]
- 19.Teunissen CE, Dijkstra CD, Polman CH, et al. Decreased levels of the brain specific 24S-hydroxycholesterol and cholesterol precursors in serum of multiple sclerosis patients. Neuroscience letters. 2003;347(3):159–62. doi: 10.1016/s0304-3940(03)00667-0. [DOI] [PubMed] [Google Scholar]
- 20.Poli G, Biasi F, Leonarduzzi G. Oxysterols in the pathogenesis of major chronic diseases. Redox Biology. 2013;1(1):125–30. doi: 10.1016/j.redox.2012.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gargiulo S, Sottero B, Gamba P, et al. Plaque oxysterols induce unbalanced up-regulation of matrix metalloproteinase-9 in macrophagic cells through redox-sensitive signaling pathways: Implications regarding the vulnerability of atherosclerotic lesions. Free radical biology & medicine. 2011;51(4):844–55. doi: 10.1016/j.freeradbiomed.2011.05.030. [DOI] [PubMed] [Google Scholar]
- 22.Glass CK, Saijo K. Immunology: Oxysterols hold T cells in check. Nature. 2008;455(7209):40–1. doi: 10.1038/455040a. [DOI] [PubMed] [Google Scholar]
- 23.Cui G, Qin X, Wu L, et al. Liver X receptor (LXR) mediates negative regulation of mouse and human Th17 differentiation. J Clin Invest. 2011;121(2):658–70. doi: 10.1172/JCI42974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bensinger SJ, Bradley MN, Joseph SB, et al. LXR signaling couples sterol metabolism to proliferation in the acquired immune response. Cell. 2008;134(1):97–111. doi: 10.1016/j.cell.2008.04.052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Chalmin F, Rochemont V, Lippens C, et al. Oxysterols regulate encephalitogenic CD4(+) T cell trafficking during central nervous system autoimmunity. Journal of autoimmunity. 2015;56:45–55. doi: 10.1016/j.jaut.2014.10.001. [DOI] [PubMed] [Google Scholar]
- 26.Weinstock-Guttman B, Zivadinov R, Mahfooz N, et al. Serum lipid profiles are associated with disability and MRI outcomes in multiple sclerosis. Journal of neuroinflammation. 2011;8:127. doi: 10.1186/1742-2094-8-127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mandoj C, Renna R, Plantone D, et al. Anti-annexin antibodies, cholesterol levels and disability in multiple sclerosis. Neuroscience letters. 2015;606:156–60. doi: 10.1016/j.neulet.2015.08.054. [DOI] [PubMed] [Google Scholar]
- 28.Palavra F, Marado D, Mascarenhas-Melo F, et al. New markers of early cardiovascular risk in multiple sclerosis patients: oxidized-LDL correlates with clinical staging. Disease markers. 2013;34(5):341–8. doi: 10.3233/DMA-130979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tettey P, Simpson S, Jr, Taylor BV, et al. Vascular comorbidities in the onset and progression of multiple sclerosis. Journal of the neurological sciences. 2014;347(1–2):23–33. doi: 10.1016/j.jns.2014.10.020. [DOI] [PubMed] [Google Scholar]
- 30.Chen Y, Wang S, Yi Z, et al. Interferon-inducible cholesterol-25-hydroxylase inhibits hepatitis C virus replication via distinct mechanisms. Scientific reports. 2014;4:7242. doi: 10.1038/srep07242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Cui X, Kushiyama A, Yoneda M, et al. Macrophage foam cell formation is augmented in serum from patients with diabetic angiopathy. Diabetes research and clinical practice. 2010;87(1):57–63. doi: 10.1016/j.diabres.2009.10.011. [DOI] [PubMed] [Google Scholar]
- 32.Boven LA, Van Meurs M, Van Zwam M, et al. Myelin-laden macrophages are anti-inflammatory, consistent with foam cells in multiple sclerosis. Brain. 2006;129(Pt 2):517–26. doi: 10.1093/brain/awh707. [DOI] [PubMed] [Google Scholar]
- 33.van de Kraats C, Killestein J, Popescu V, et al. Oxysterols and cholesterol precursors correlate to magnetic resonance imaging measures of neurodegeneration in multiple sclerosis. Multiple sclerosis. 2014;20(4):412–7. doi: 10.1177/1352458513499421. [DOI] [PubMed] [Google Scholar]
- 34.Stiles AR, Kozlitina J, Thompson BM, et al. Genetic, anatomic, and clinical determinants of human serum sterol and vitamin D levels. Proceedings of the National Academy of Sciences of the United States of America. 2014;111(38):E4006–14. doi: 10.1073/pnas.1413561111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Chancharme L, Therond P, Nigon F, et al. LDL particle subclasses in hypercholesterolemia. Molecular determinants of reduced lipid hydroperoxide stability. Journal of lipid research. 2002;43(3):453–62. [PubMed] [Google Scholar]