27-Hydroxycholesterol Promotes Oligodendrocyte Maturation – Implications for Hypercholesterolemia Associated Brain White Matter Changes

Vilma Alanko Karolinska Institutet Tania Quintela-López University of the Basque Country: Universidad del Pais Vasco Adhara Gaminde-Blasco University of the Basque Country: Universidad del Pais Vasco Raúl Loera-Valencia Karolinska Institutet Alina Solomon University of Eastern Finland Kuopio Campus: Ita-Suomen yliopisto Kuopion kampus Ingemar Björkhem Karolinska Institutet Angel Cedazo-Minguez Karolinska Institutet Silvia Maioli Karolinska Institutet Graziella Tabacaru Karolinska Institutet María Latorre-Leal Karolinska Institutet Carlos Matute University of the Basque Country: Universidad del Pais Vasco Miia Kivipelto Karolinska Institutet Elena Alberdi University of the Basque Country: Universidad del Pais Vasco Anna Matton (  anna.matton@ki.se ) Karolinska Institutet https://orcid.org/0000-0002-7819-0495

Maturation of OPCs is essential for restoring some lost functions caused by demyelination (36). OPCs are vulnerable to several factors, including oxidative stress, amyloid-beta (Aβ), and age (37)(38)(39). Furthermore, along with ageing, properties and functionality of OPCs alternate both inter-and intraregionally in the brain resulting in a heterogeneous group of OPC subtypes (40), which subsequently may in uence the remyelinating capacity of these cells. Regarding impacts of Aβ on oligodendroglia, contradictory results exist; in several studies, Aβ has been observed to induce proliferation and maturation of oligodendrocytes, and even remyelination of myelin lesions in cell and animal models (41,42), while in others OPCs have been observed to lose their ability to proliferate in Aβ plaque environment (43,44). Exogenous agents may also affect OPCs. Statins induce the maturation of OPCs thus depleting the pool of progenitors (45). Hence, if OPCs go into senescence or initiate needless maturation processes, their capacity to remyelinate at later occasions would be insu cient. 27-OH reduces cholesterol synthesis in vitro and in vivo (46, 47), which raises concerns regarding proper myelin assembly and maintenance for which adequate cholesterol levels are essential. Still, to our knowledge, how 27-OH affects oligodendrocytes and myelination-related processes remain unknown. Here, we investigated whether 27-OH in uences oligodendrocyte maturation pro le and myelin integrity. First, the impacts of 27-OH treatment on cell viability in cultured oligodendrocytes were studied, whereafter maturation was assessed. Myelin RNA and protein levels were measured in primary 3D cocultures from mice. Learning and memory of female mice overexpressing sterol 27-hydroxylase (Cyp27Tg) and age-matched non-transgenic littermates (Tg−) were assessed with a battery of cognitive tests, and brain sections stained for markers of immature and mature oligodendrocytes, and myelin content. Lastly, associations between 27-OH and myelination proteins in CSF samples from memory clinic patients were analysed.

Cell viability assay
The Human Oligodendroglioma (HOG) cell line (Millipore, Burlington, MA, USA) is a clonal cancer cell line and originates from a surgically removed oligodendroglioma from a human patient. The cell line is described in more detail by Post and Dawson,and Buntinx et al. (48,49). The cells were cultured in DMEM media (Gibco, Thermo Fisher Scienti c, Waltham, MA, USA) with 10% heat-inactivated FBS (Gibco) and maintained at 37°C 5% CO 2 .
Cell viability assays for HOG cells and primary oligodendrocytes were performed with Calcein AM Cell Viability Assay (R&D Systems, Minneapolis, Minnesota, USA). HOG cells were seeded into 96-well plates at a density of 8000 cells per well. After 1 day in vitro, cells were incubated for 24 h with either 56.4 mM DMSO (Sigma-Aldrich, Saint Louis, Missouri, USA) (volume corresponding to that of 10 µM 27-OH treatment), 282 mM DMSO (volume corresponding to that of 50 µM 27-OH treatment), 1 µM, 10 µM, or 50 µM 27-OH (Avanti Polar Lipids, Alabaster, AL, USA). 27-OH was diluted in DMSO. Non-treated (data not shown) and 0.1 mM H 2 O 2 treated cells were included as negative and positive controls, respectively. The assay was performed according to the manufacturer's instructions using a 2 µM Calcein AM Working Solution. The uorescence of the plates was read with BMG Fluostar Galaxy (BMG Labtechnologies, Offenburg, Germany) with an excitation lter of 492 nm and an emission lter of 520 nm, and results were processed with its software (version 4.31-0).
For cell viability assay in 2D primary oligodendrocytes, tissue was obtained from optic nerves of 12-daysold Sprague Dawley rats, as described previously (50). Cells were seeded into 24-well plates bearing 12mm-diameter coverslips coated with poly-D-lysine (10 mg/ml) at a density of 1 × 10 4 cells per well. Cells were maintained at 37°C and 5% CO 2 in a chemically de ned medium (50). After 1 day in vitro, cultured oligodendrocytes were treated with increasing concentrations of DMSO or 27-OH (0.1-10 µM) for 24 h, as is indicated above. Cultured oligodendrocytes were incubated with Calcein-AM (Life Technologies, Carlsbad, CA, USA) at 1 µM and 37°C for 30 min in fresh culture medium and thereafter washed in prewarmed 0.1 M PBS three times. Emitted uorescence was measured by a Synergy HT (Biotek, Winooski, Vermont, USA) spectrophotometer using excitation wavelength at 485 nm and emission at 528 nm.

Primary 3D co-cultures
For 3D cell culturing C57/BLJ6 mice were used. Female mice were let to breed with males overnight to ensure accurate embryonic day (E) of the embryos used for primary cell cultures. At E16 -E17, the embryos were sacri ced by decapitation, heads of the embryos were collected and, kept in Hibernate™-E Medium (Gibco) until dissection of brains. Dissection was performed in EBSS solution (Life Technologies) and only dissected hippocampus and cortex were used for the co-culture set-up.
To extract the cells from the tissue, tissues were rst incubated at 37°C in a 1:1 mix of solution A (0.0625% trypsinization solution [Life Technologies] in EBSS) and solution B (0.08 mg/ml DNAse I solution [Roche, Basel, Schweiz] in EBSS) for approximately 15 min gently moving. Trypsinization was stopped with 10% heat-inactivated FBS or culture media #1 containing 2% B27 supplement (Gibco), 1% N2 supplement (Gibco), 100 units/ml penicillin and 100 µg/ml streptomycin (Thermo Fisher Scienti c), in DMEM/F-12 + GlutaMAX (Gibco). After supernatant removal, the tissue was mechanically disintegrated by pipetting up and down in culture media. The supernatant containing the cells was collected and centrifuged at 500 x g for 5 min.
For immunocytochemistry, pellet was thereafter resuspended in a new culture media #2 containing a 1:1 mix of Neurobasal media and DMEM/F12 + GlutaMAX, 0.5% GlutaMAX supplement (Gibco), 2% B27, 1% N2, and 50 ng/ml NGF (Gibco). For 3D culture droplets, diluted cell suspension and Matrigel (Corning Life Science, Corning, NY, USA) were mixed 1:1 to reach droplet sizes of 50 µl. Droplets were cultured in individual dishes precoated with poly-D-lysine (MatTek, Ashland, MA, USA). Approximately 15,000 cells were seeded in each droplet. Droplets were let to settle for 1 h before adding 2 ml of the culture media #2. Cultures were treated at DIV1 with 1 µM 27-OH or 5.6 mM DMSO. Non-treated cultures were included as negative controls (data not shown). Two-thirds of media were replaced every four days until xation and immuno uorescence (IF) analysis. Cultures were maintained at 37°C 5% CO 2 and xed at DIV20.
For the time-course study, the pellet was resuspended in culture media #1. 110,000-125,000 cells were seeded per 75 µl droplet with Matrigel in a ratio of 1:1. Droplets were let to settle for 1 h before adding 1 ml of the culture media #1. Cultures were treated at DIV1 with 1 µM 27-OH or 5.6 mM DMSO. Cultures were maintained at 37°C 5% CO 2 until harvesting at DIV3, 5 or 10, by removing media, washing with PBS, and lysing with 300-350 µl RLT buffer (Qiagen, Hilden, Germany).
A Nikon Eclipse Ti spinning disk confocal microscope using 20X objective coupled to a camera with a pixel size of 6.45 µm was used to image the cultures, and images were processed with NIS-Element software (version 5.11.02, Nikon, Tokyo, Japan). The images were further analysed using Imaris x64 software (version 5.9.0, Oxford Instruments, Zurich, Switzerland) by creating surfaces of desired channels: DAPI alone represented the total number of cells, Olig2 + DAPI colocalization the cells that are of oligodendrocyte lineage, MBP − Olig2 + DAPI colocalization immature oligodendrocytes, and MBP + Olig2 + DAPI colocalization represented the mature oligodendrocytes.
Quantitative PCR (qPCR) RNA was isolated from 3D cultures using the RNeasy® Mini Kit (Qiagen), following the manufacturer's protocol. The concentration and quality of the RNA were measured with a nanodrop spectrophotometer (NanoDrop 1000, Thermo Fisher Scienti c). Reverse transcription was performed in the S1000 Thermal cycler (Applied Biosystems, Thermo Fisher Scienti c) to yield cDNA using a standardised protocol for high-capacity cDNA reverse transcription (Thermo Fisher Scienti c).
Gene expression of Pdgfrα (Mm00440701_m1, Life Technologies), Cc1 (Mm00545872_m1, Life Technologies), Olig2 (Mm01210556_m1, Life Technologies), and Gapdh (4352339E, Life Technologies) in primary cells from the time-course experiment was analysed by qPCR according to a standardised protocol for TaqMan™ gene expression assay (Thermo Fisher Scienti c), using the 7500 Fast Real-Time PCR System (Applied Biosystems). Copy numbers of Pdgfrα (platelet-derived growth factor receptor α) and Cc1 (adenomatous polyposis coli clone 1) mRNAs were adjusted by mRNA copy numbers of Olig2.
As an internal control, mRNA copy numbers of Olig2 were adjusted by copy numbers of Gapdh. The relative values of mRNA copy numbers were compared to the numbers in DMSO-treated DIV3 cultures.
CYP27 transgenic mice 8-month-old female, human sterol 27-hydroxylase overexpressing, transgenic mice (Cyp27Tg) and their age-matched non-transgenic littermates (Tg−) underwent behavioural studies. We studied two separate cohorts: For Morris Water Maze we tested 10 transgenic and 10 non-transgenic mice and for Y-maze and Fear Conditioning 9 transgenic and 5 non-transgenic mice. Cyp27Tg mice overexpress human CYP27A1 and have higher amounts of 27-OH when compared to WT mice, described in more detail by Meir et al. and Ali et al. (46,47). 4 Tg − and 4 Cyp27Tg mouse brain samples were used for Western blotting of total brain homogenates. Myelin enriched fractions were retrieved from 4 Tg − and 7 Cyp27Tg mouse brains.

Behavioural tests
Morris Water Maze was performed as described by Maioli et al. (51). Each mouse was tested for four trials per day, for ve consecutive days. Reference memory was evaluated on the sixth day (probe test).
Y-maze was performed as described by Eroli et al. (52). The fear conditioning test was performed as described by Eroli et al. (52), with the supplementation of the conditional stimulus of sound on day one. Mice were exposed to a 55 dB sound at 5000 Hz that lasted for 30 seconds and was followed by a mild foot shock (0.3 mA for 2 sec). The sound-shock pairing was repeated three times in total with a 50-sec interval between each one. Cue fear conditioning was performed on day three, where mice were let to explore the new surrounding for 2 min, whereafter the sound (55 dB at 5000 Hz) lasted for 2 minutes continuously. The rectangular-shaped chamber was replaced for a round-shaped chamber (20 cm diameter x 35 cm high), and the stainless-steel grid oor was replaced for a black at surface. Instead of wiping the chamber with ethanol, the chambers were cleaned between each mouse with hypochlorous water (50% dil). Freezing behaviour was de ned as the complete absence of mobility within the same area for 2 seconds or longer and was measured through TSE Multi Conditioning software. The freezing % recorded during the habituation phase of day 1 (as a measure of baseline freezing) was compared to freezing % of day 2 to evaluate the context memory. To assess the cue memory, we measured the freezing % on day 3 before and during the sound stimulus.
Luxol blue staining of mouse brain sections For myelin staining, the Luxol Fast Blue Stain Kit (Abcam) was used. Brain sections were incubated in LFB for 2 h at 60°C. After incubation, slides were rinsed thoroughly with distilled H 2 O and thereafter dipped in lithium carbonate for differentiation. Differentiation was continued by dipping the slides into 70% alcohol reagent until the grey matter was colourless. Additional sections were incubated with cresyl echt violet for 1 min at RT to con rm that tissue morphology is maintained during LFB staining (images not shown). After incubation, slides were quickly rinsed with distilled H 2 O, and sections were dehydrated with three changes in 99.5% ethanol. Sections were mounted on coverslips with Vecta Mount mounting media. Images were acquired with Nikon Eclipse E800 light microscope, using a 10X objective, coupled to a Nikon DS-Ri2 camera and processed with NIS-Elements imaging software (version 4.30.00). Blinded analysis of LFB stained sections was performed using ImageJ 1.52p (53) as described previously by Underhill et al. 2011 andKhodanovich et al. 2017 (54,55). The mean intensity of each brain section was measured from corpus callosum and mbria. Mean intensities of the red channel (I R ) in ROIs (Fig. 4A) were measured from RGB images of LFB stained sections without cresyl echt violet stain. Similarly, the mean intensity of the background (I B ) was measured in each image. Optical densities (%) of each region were calculated as , resulting in higher myelin content with increasing value.
Immuno uorescence staining of mouse brain sections Immuno uorescence (IF) for MBP was performed on free-oating sections. Slices were permeabilized and blocked in 4% normal goat serum (NGS), 0.1% Triton X-100 in PBS (blocking buffer) for 1 h and incubated overnight at 4°C with primary antibody against MBP (SMI 99, 1:1000; Biolegend, San Diego, CA, USA). Slices were washed in PBS and incubated with uorophore-conjugated Alexa secondary antibody (1:500) in blocking buffer for 1 h at RT. Samples were mounted with Fluoromount-G. Images were acquired with Zeiss AxioVision microscope using a 10X objective and analysis was carried out in 2-3 sections per subject. Images were taken with the same setting for all experiments and mean intensity values were quanti ed from corpus callosum and cortex with Image J software.
For IF of CC1 and PDGFRα in depara nised tissue, a heat-induced treatment (95°C for 5 min) in R-Universal buffer (Aptum, Southampton, UK) was performed on all sections for epitope recovery. The sections used for CC1 staining were washed with cold PBS and treated with ice-cold 100% EtOH for 10 min at − 20°C. This step was not performed on the sections used for PDGFRα staining. The sections were (A11006, Invitrogen) or Goat anti-mouse Alexa Fluor 488 1:300 (A28175, Invitrogen) and anti-NeuN and DAPI 1:1000 secondary antibodies for 2 h at RT and protected from light. Finally, the sections were treated with an auto uorescence eliminator reagent (Millipore) for 5 minutes to reduce auto uorescence and mounted with SlowFade™ Gold Antifade (Thermo Fisher Scienti c). The stained tissue was analysed using a confocal microscope (Zeiss LSM-800 Airy system) with a 20X objective. Three pictures from each region, corpus callosum, cortex, and hippocampus, were captured from one section of each mouse using the Zen software (ZEISS Microscopy, Jena, Germany). Imaging settings were kept constant for each staining type. In a blinded experiment, the number of PDGFRα and CC1 positive cells were quanti ed for each region (total area of approximately 918 mm 2 /region) by one rater.

Preparation of myelin enriched fractions
Dissected brains were removed from meninges, choroid plexus, cerebellum, and the brainstem. Brains were minced in DMEM. Total brain homogenates were incubated with 2.5 mg/ml DNAse I and 2.5 mg/ml Trypsine/EDTA for 30 min-1 h at 37°C on an orbital shaker at 180 rpm for tissue digestion. Thereafter, tissue suspension was mixed with DMEM and centrifuged at 1000 x g for 10 min at 4°C. Pellet was then resuspended in 20% w/v BSA-DMEM and centrifuged 1000 x g for 20 min at 4°C. The upper layer (myelin) was collected using a pasteur pipette.
Immunoblotting of brain and cerebrospinal uid For myelin enriched fractions, 25 µl of each sample was mixed with 25 µl of 2X sample buffer. This process was performed on ice to enhance the lysis process and avoid protein degradation. Samples were boiled at 95°C for 8 min. Consequently, they were centrifuged for 1 minute at 14,000 x g and the supernatants were collected. Then, total protein content was quanti ed through RC DC Protein Assay (Bio-Rad) and all the samples were brought to the same concentration. Finally, they were boiled again at 95°C for 8 minutes.

27-OH treatment exhibits toxicity in cultured oligodendrocytes
First the impact of 27-OH on oligodendrocyte viability was investigated in primary oligodendrocytes derived from rat optic nerves and human oligodendroglioma (HOG) cells. In primary oligodendrocytes, cell viability decreased with increasing concentrations of 27-OH, with a signi cant decline at 0.5 µM (9.3% [3.5] of control, p = 0.020, Fig. 1A). However, the reduction in cell viability showed a trend already at 0.
Since the 3D co-cultures were treated at DIV1 and we know that 27-OH is rapidly metabolised (57) Fig. 3E). On the rst day of Fear Conditioning (FC) mice were exposed to a high-frequency sound and an electric shock after the sound. On the second day, the oor was kept as in day 1 (context), but mice were not exposed to sound nor electrical shock. In both strains the time freezing increased signi cantly for second day (Cyp27Tg 29.9% [4.5], Tg − 41.9% [6.0], for both p < 0.0001, Fig. 3F). However, the amount of freezing during the second day was less for Cyp27Tg than for Tg − mice, yet the difference demonstrates a trend towards signi cance (p = 0.072). On the third day, mice were exposed to the same sound (cue) as on the rst day on a new oor (change of context  Fig. 3G). There was no signi cant difference between strains in freezing time before sound (p = 0.100) nor during sound (p = 0.629).
Chronic exposure to 27-OH in vivo does not alter myelin overall structure but increases MBP levels To assess overall myelin structure and level in the Tg − and Cyp27Tg mice, brain sections were stained with Luxol fast blue (LFB). Myelination of the corpus callosum at the age of 8 months did not differ between strains (p = 0.427, Fig. 4C). Neither was there a difference in LFB staining of mbria between strains (9.7%, p = 0.108, Fig. 4D). In order to investigate if oligodendrocyte maturation is enhanced and if there is a decrease in OPCs in Cyp27Tg brains, brain sections were stained for MBP, PDGFRα, and CC1.

27-OH levels in CSF associate with CNPase levels in memory clinic patients
To study the associations between 27-OH levels and proteins re ecting myelin integrity we further studied CSF samples from GEDOC memory clinic patients. Sociodemographic and clinical characteristics of the cohort are presented in Table 1. The results from the linear regression analyses are presented in Table 2.
We did not observe any statistically signi cant associations between MBP and 27-OH in any of the models. Nonetheless, CNPase associated signi cantly with 27-OH in Model 1 (β = 0.38, p = 0.022), and the signi cance remained even after further adjustment for AD-related pathologies, which are common in memory clinic patients: Aβ 42 , t-Tau, and p-Tau levels. The results did not change even if models were further adjusted for cognition (data not shown). As expected, the CSF levels of myelin proteins MBP and CNPase were associated with each other (Model 1: β = 0.70, p < 0.0001). The myelin protein levels did not associate with the levels of AD biomarkers (data not shown), except for MBP which was associated with Aβ 42 (β = 0.32, p = 0.034). Neither did the myelin proteins associate with cognition as measured by RAVLT or MMSE when adjusted for age, sex, and education (data not shown).

Discussion
The oxysterol 27-OH is suggested to be the missing link between midlife hypercholesterolemia and increased risk of developing AD (2). Nonetheless, relatively little is known about the effects 27-OH poses on the brain -especially on oligodendrocytes and myelination. In our study, we observed that a physiologically relevant concentration of 27-OH induces cell death in primary oligodendrocytes which is in line with a previous study in an astrocytic cell model (58). In another study 27-OH did not cause cell death in a murine oligodendrocyte cell line (59) which is in compliance to our nding that immortalized cells (HOG) were resistant to 27-OH toxicity at relevant concentrations. The main result in this study was the noteworthy induction of OPC maturation by 27-OH. We postulate that the signi cant reduction in OPCs in the 3D-cell model is mainly a result of maturation rather than of an extensive OPC cell death as the total number of oligodendrocytes was not altered and there was additionally a modest increase in mature oligodendrocytes. Oligodendrocytes of any differentiation stage are vulnerable to cytotoxicity but depending on the agent there might be differences in the overall effect between OPCs and mature oligodendrocytes, given their different gene expression pro le (60). As the cells were treated only once, the long-term effect observed is quite outstanding as 27-OH is rather quickly metabolised in about two hours (57). In some studies, Aβ has been demonstrated to cause a similar effect in OPCs, implying that OPCs are vulnerable and an induction of their differentiation resulting in impaired proliferation capabilities may play a role in changes in myelin integrity observed in AD (39,(41)(42)(43). The remodelling capacity is essential, as myelination continues long into adulthood (25). Induction of OPC maturation into myelinating oligodendrocytes is desired in demyelinating diseases, but the timing of maturation must be adjusted to the need. This suggests that hypercholesterolemia and oxysterol related OPC maturation may in fact impair the ability of proper and timely remyelination when needed. Our results support the ndings of Sim and colleagues (45) who detected a PPARγ-induced OPC maturation after simvastatin or pravastatin treatment. Due to the comparable impact on cholesterol metabolism and synthesis between statins and excess 27-OH, it would be of future interest to elucidate whether the molecular mechanisms behind the oligodendroglial effects of these agents are corresponding. To summarise, we hypothesise that a loss of proliferating OPCs due to chronic exposure to 27-OH could impact the remyelinating capacity. As oxysterols and 27-OH in particular, are known to affect also other cell types in the brain (18,21), the possibility that 27-OH would indirectly in uence OPCs cannot be excluded. We did not see a decrease in the OPC marker PDGFRα in old Cyp27Tg mice suggesting that in vivo, there may be compensatory mechanisms maintaining the OPC population. OPCs are nonetheless not spared from impacts that come along with ageing, and old OPCs do not respond similarly to maturation signals as young OPCs do (61).  Male Cyp27Tg mice have previously been reported to exhibit de cits in memory at the age of 8-9 months and for example impaired neuronal morphology and function (18,19). A similar effect of 27-OH overload on memory has been reported in rats (69). In mice modelling cerebrovascular pathology, high-cholesteroldiet induced white matter changes and reduced numbers of immature oligodendrocytes, accompanied by cognitive de cits (7). In our study, female Cyp27Tg mice did demonstrate impairment in learning in MWM at the age of 8 months, indicating no sex differences in the effect of 27-OH in spatial learning. In Y-maze and FC we were unable to detect signi cant differences; however, the control group consisted of only ve Tg − mice reducing the power to detect small differences. As observed in previous studies, 27-OH has a varying impact on different functions of the brain raging from morphological changes of neurons and reduced spine density in the hippocampus (18), decreased glucose uptake of neurons (19), induced in ammatory responses (20), to induced maturation of oligodendrocytes as observed in the current study.
Plausible alterations in remodelling capacity of oligodendroglia and myelin structure may contribute to the impaired cognition of animals with excess levels of 27-OH. At the age of 8 months, there was no difference in oligodendrocyte maturation, as there were similar numbers of immature PDGFRα positive and mature CC1 positive cells. Nor was there any major difference in the extent of myelination as observed with LFB staining. Nevertheless, the signi cant increase of MBP production in corpus callosum and imbalance in other myelin proteins as detected in myelin enriched fractions may imply that small alterations in white matter composition may contribute to the attenuated cognitive performance. It is to be emphasised, however, that our results only represent a snapshot in the lifespan of female Cyp27Tg mice, and therefore we are unable to draw conclusions on how these alterations have been generated and what is to come along when the mice age. Moreover, some cognitive dysfunction can only be observed in paradigms testing energy-intensive plasticity, such as hippocampal long-term potentiation, which is aberrant in Cyp27Tg mice at 2 months of age (70).
Myelin proteins are released into CSF when the white matter is damaged (71). In post-mortem brains of AD patients, depletion of cholesterol, MBP, and CNPase have been detected (72). Moreover, substantially higher CSF concentrations of 27-OH are present in brain-damaging diseases, including demyelinating polyneuropathy and AD (15,16), emphasising a plausible connection of this oxysterol in CSF and demyelination and/or neuronal damage. Increasing CSF 27-OH levels have previously been associated with WMH burden in an AD-like cohort (73), and some cases of spastic paraplegia type 5 -a disease that results in elevated 27-OH levels due to mutations in CYP7B1 gene that is responsible for catabolising excess 27-OH -white matter abnormalities have been reported (74). In our study, we found a positive association between CNPase and 27-OH that was not dependent on AD pathology. However, already in the mid-nineties Braak and Braak reported that the Tau-related changes seen in AD follow the inverse pattern of myelinogenesis (31), and later it has been proven that even demyelination follows this same pattern (32,75). Hence, there could be a possible association with the presence of neuro brillary tangles and myelin lesions, which should be investigated in post-mortem brain in the future. Based on the available literature (76), we postulate that CNPase in CSF theoretically could be a marker for oligodendrocyte maturation. In this context, CNPase association to 27-OH may represent its role in oligodendrocyte maturation. However, our results should only be regarded as exploratory due to the small sample size and thus limited statistical power. These results should be veri ed in larger cohorts and preferably together with neuroimaging data.

Conclusions
Based on our ndings, 27-OH affects oligodendrocytes in at least three ways: It stimulates maturation of OPCs into mature oligodendrocytes, alters protein levels in myelin, and induces cell death of oligodendrocytes. We discuss the possibility that these mechanisms may contribute to white matter changes and dementia resulting from hypercholesterolemia.

Consent for publication
Not applicable.

Competing interests
The authors have no competing interests to declare.    strains altered between the arms in similar amounts. (E) However, the number of entries was slightly increased in the Cyp27Tg group. Fear conditioning was performed on three consecutive days, where on day one mice were exposed to sound, following an electric shock. (F) On the second day, mice were exposed to the same context as on day one and freezing increased signi cantly in both strains. (G) On day three, mice were exposed to a new environment and the cue (sound) from the rst day. Both strains demonstrated increased freezing during sound. (A, F, G) Repeated-measures ANOVA, (B) Mann-Whitney U test two-tailed (C, D, E) unpaired Student's two-tailed t-test, * p < 0.05, *** p < 0.0001. All data presented as mean ± SD, except (A) is presented as mean ± SEM.

Figure 4
Cyp27Tg mice have preserved overall myelin structure. (A) Brain sections were stained with Luxol fast blue (LFB), and areas of corpus callosum (average of the three red squares) and mbria (one green square) were analysed for intensity differences between Cyp27Tg and Tg− mice. (B) Representative images of LFB stained corpus callosa and mbriae. Scale bar represents 100 µm. Intensity difference in (C) corpus callosum and (D) mbria. (C, D) Unpaired Student's two-tailed t-test, all data presented as mean ± SD.  Hypercholesterolemia results in increased peripheral concentrations of 27-OH. The oxysterol further traverses the blood-brain barrier into the brain and cerebrospinal uid (CSF) in a concentration-dependent manner. Normally, oligodendrocyte progenitor cells (OPC) mature into myelinating oligodendrocytes during physiological myelin-turnover. However, it seems that 27-OH induces maturation reducing the OPC