Dysregulated Cholesterol Clearance via CYP46A1 Contributes to Cerebellar Sterol Imbalance in Mecp2-Null Mice

TL;DR

This study shows that disrupted cholesterol clearance in the cerebellum of Rett syndrome mice leads to a buildup of cholesterol, potentially contributing to the disease's neurological symptoms.

Contribution

The study identifies impaired CYP46A1 expression as a novel mechanism for cerebellar cholesterol imbalance in Rett syndrome.

Findings

Total cholesterol levels are significantly elevated in cerebellar tissue and synaptosomes of Mecp2-null mice.

CYP46A1 expression is markedly reduced, indicating defective cholesterol clearance rather than increased synthesis.

Cholesterol biosynthetic regulators show no significant changes in expression.

Abstract

Rett syndrome (RTT) is a neurodevelopmental disorder characterized by motor deficits, partly attributed to cerebellar dysfunction. RTT is primarily caused by mutations in the gene encoding the methyl-CpG-binding protein 2 (MECP2), which has been implicated in cholesterol homeostasis by mechanisms that remain poorly understood. Given that brain cholesterol is primarily synthesized de novo and that disrupted cholesterol homeostasis is linked to various neurological disorders, we aimed to investigate cholesterol regulation in the cerebellum of Mecp2-null mice, a well-established RTT model. We measured total cholesterol levels in cerebellar tissue and cerebellar synaptosomes and assessed the expression of genes involved in cholesterol biosynthesis and intracellular transport. Our results show significantly elevated total cholesterol in both cerebellar tissue and synaptosomes. Furthermore, we identified a marked reduction in CYP46A1 expression, which is essential for the elimination of encephalon sterols. In contrast, key cholesterol biosynthetic regulators (Srebp2, Hmgcs1, Sqle) showed no significant changes in expression, suggesting an impaired cerebellar cholesterol turnover—driven by defective clearance—rather than enhanced synthesis may underlie the metabolic imbalance observed in the cerebellum of the RTT mouse model. Altogether, these findings provide a mechanistic insight into how MeCP2 deficiency disrupts cerebellar cholesterol homeostasis and highlight cholesterol clearance pathways as potential contributors to RTT pathology and a factor to consider for further RTT therapeutic approaches.

Figures

mRNA expression of cholesterol transport-related genes in the cerebellum of Mecp2-null mice. mRNA expression was evaluated by RT-qPCR in the cerebellum of Mecp2-/Y and their wild-type littermate mice at 3 (a) and 7 (b) weeks of age. 3-week-old Mecp2-/Y (n = 9, triangles) and wild-type (n = 5, circles); 7-week-old Mecp2-/Y (n = 8, triangles) and wild-type (n = 7, circles). ND: not detected. Graph bars represent means ± SEM. Statistical analysis was performed using multiple Student’s t-tests followed by the Holm–Šídák correction to control the Family-Wise Error Rate (FWER) ∗ p < 0.05. Detailed statistical parameters, including adjusted p-values, are provided in Table 3 and Table 4.

Tables

Table 3. Statistical analysis of mRNA expression for intracellular cholesterol transport-related genes in the cerebellum of 3-week-old mice.

Gene	Mean Difference	SE of Difference	t Ratio	p-Value	Adjusted p-Value	Significance
Apoe	−0.244	0.189	1.297	0.219	0.892	ns
Abca1	−0.347	0.352	0.986	0.344	0.947	ns
Npc1	−0.137	0.169	0.810	0.434	0.947	ns
Npc2	−0.366	0.195	1.875	0.085	0.657	ns
Cav1	0.001	0.245	0.004	0.997	1.000	ns
Cav2	−0.538	0.459	1.171	0.264	0.914	ns
Osbpl2	0.077	0.131	0.586	0.568	0.965	ns
Osbpl9	0.046	0.137	0.335	0.743	0.983	ns
Cyp46a1	−0.001	0.127	0.009	0.993	1.000	ns
Stard1	−0.369	0.198	1.859	0.088	0.657	ns
Stard4	−0.313	0.321	0.976	0.348	0.947	ns
Stard5	−0.364	0.233	1.562	0.144	0.789	ns

Table 4. Statistical analysis of mRNA expression for intracellular cholesterol transport-related genes in the cerebellum of 7-week-old mice.

Gene	Mean Difference	SE of Difference	t Ratio	p-Value	Adjusted p-Value	Significance
Apoe	−0.080	0.131	0.610	0.552	0.915	ns
Abca1	0.121	0.136	0.892	0.389	0.915	ns
Npc1	0.168	0.090	1.852	0.087	0.597	ns
Npc2	0.162	0.118	1.379	0.191	0.794	ns
Cav1	0.156	0.115	1.363	0.196	0.794	ns
Cav2	0.236	0.092	2.563	0.025	0.261	ns
Osbpl2	0.281	0.132	2.126	0.053	0.452	ns
Osbpl9	0.055	0.113	0.485	0.636	0.915	ns
Cyp46a1	0.340	0.096	3.556	0.004	0.045	*
Stard1	−0.013	0.140	0.090	0.930	0.930	ns
Stard3	0.099	0.115	0.857	0.407	0.915	ns
Stard4	0.181	0.122	1.486	0.161	0.794	ns
Stard5	−0.183	0.123	1.482	0.162	0.794	ns

Full text

1. Introduction

Rett syndrome (RTT; OMIM #312750) is a neurological disorder first described by Dr. Andreas Rett in 1966 [1], affecting approximately 1 in 10,000 live girl births [2,3]. RTT is characterized by progressive neurodevelopmental regression leading to severe cognitive and motor impairments [2]. RTT typically presents after an initial period of an apparently normal development, followed by rapid deterioration, including loss of acquired speech and the hallmark stereotyped hand movements. Subsequently, RTT girls develop behavioral abnormalities and further motor decline, often including irregular breathing, impaired ambulation, and seizures [2,4]. Overall, RTT leads to widespread synaptic and structural alterations, including reduced neuronal size, simplified dendritic arbors, and decreased spine density across multiple brain regions [5,6]. The brains of RTT patients show significant volume loss and progressive cerebellar atrophy, including Purkinje cell loss, which contributes to deterioration of the motor system [3,7,8].

Most RTT girls carry mutations in the gene encoding the methyl-CpG binding protein-2 (MECP2) [9]. MECP2 was initially described as a protein that specifically binds to methylated DNA at CpG dinucleotides [10], functioning as a transcriptional repressor [11]. However, it was later demonstrated that MECP2 could also activate transcription, compact chromatin structure, regulate microRNA processing, and regulate alternative splicing [2,11,12]. MECP2 is especially abundant in postmitotic neurons [13,14], although its expression has also been reported in non-neuronal cells in vitro [15,16] and in other tissues throughout the body [17,18]. Several mouse models of RTT have been generated by deleting or mutating Mecp2 [19,20,21,22,23]; however, the male Mecp2-null mouse model (Mecp2^-/Y^) recapitulates most of the symptoms observed in RTT girls, including reduced mobility, dystonia/stiffness, ataxic gait, and tremors [24,25,26,27,28]. It has been widely used in preclinical trials [25].

The encephalon, including the brain and cerebellum, is a major reservoir of cholesterol, accounting for approximately 20–23% of the body’s total cholesterol. Most brain cholesterol exists in a non-esterified form (99%), with 10–20% localized to the plasma membranes of neurons and astrocytes, and 70% in the myelin sheaths of oligodendrocytes [29,30,31]. Because plasma cholesterol cannot cross the blood–brain barrier (BBB), all brain cholesterol must be synthesized locally and cleared through a tightly controlled mechanism [30,32]. In neurons, the predominant route of sterol elimination is mediated by CYP46A1, which converts excess cholesterol into 24S-hydroxycholesterol (24S-OHC). This soluble oxysterol can cross the BBB and reach the liver for final clearance [33]. This pathway accounts for 30–40% of total brain cholesterol clearance and is essential for maintaining brain cholesterol homeostasis. Given its critical role, even moderate changes in CYP46A1 expression or activity can profoundly affect neuronal cholesterol flux, a mechanism increasingly implicated in neurodegenerative diseases such as Huntington’s disease (HD), Parkinson’s disease (PD), Alzheimer’s disease (AD), and multiple sclerosis (MS) [29,34].

Cholesterol plays a critical role in the cerebellum of RTT girls. It has been demonstrated that disruptions in cholesterol metabolism occur in both mouse models and in RTT girls, potentially contributing to the neurological deficit characteristic of RTT [35,36]. For example, reducing cholesterol synthesis using statins—a pharmacological inhibitor of cholesterol production—has been shown to improve motor function, extend lifespan, and alleviate both systemic and brain-specific lipid metabolic disturbances in Mecp2-null mice [37]. Moreover, reduced levels of cholesterol precursors (lanosterol, lathosterol, desmosterol) and oxysterols (27-hydroxycholesterol and 24S-hydroxycholesterol) in the brains of Mecp2-null mice suggest significant alterations in cholesterol metabolism [38]. Importantly, recent work has identified CYP46A1 as a potential contributor to RTT pathology. Given its central role in cholesterol turnover, diminished CYP46A1 levels or activity could impair sterol elimination and exacerbate neuronal dysfunction. Supporting this concept, CYP46A1 overexpression ameliorates behavioral deficits and extends lifespan in Mecp2-null mice [39]. Because the cerebellum is both structurally vulnerable in RTT and functionally dependent on cholesterol-rich synaptic membranes, disturbances in sterol homeostasis may disproportionately compromise cerebellar circuits, contributing to the RTT characteristic phenotype.

In summary, disruptions in cholesterol homeostasis, particularly in cholesterol clearance via CYP46A1, appear to play a critical role in RTT pathology. However, whether such disturbances arise from defective synthesis, impaired turnover, or region-specific metabolic vulnerability -especially within the cerebellum- remains insufficiently resolved. Our study shows that Mecp2-null mice exhibit increased cholesterol levels in the cerebellum and cerebellar synaptosomes, which is accompanied by a significant reduction in CYP46A1 expression, the primary enzyme responsible for cholesterol elimination in the encephalon. Together, these findings suggest that defective cholesterol clearance may be an underrecognized mechanism contributing to cerebellar dysfunction in RTT.

2. Results

2.1. Cholesterol Levels Are Increased in the Cerebellum of Mecp2-Null Mice

Since cholesterol is essential for proper central nervous system function and the motor capabilities are regulated by the cerebellum [29], we measured total cholesterol levels in the cerebellum of Mecp2-null mice at 7 weeks of age, a time when the characteristic RTT phenotype is evident. Our results show a significant increase in cerebellar cholesterol levels in Mecp2-null mice compared to wild-type mice (Figure 1a). Because synaptic compartments are particularly sensitive to fluctuations in membrane sterol content, we also measured total cholesterol levels in cerebellar synaptosomes. Similar to cerebellar cholesterol levels, total cholesterol was increased in synaptosomes isolated from the cerebellum of 7-week-old Mecp2-null mice compared to wild-type (Figure 1b and Figure S1).

These convergent increases in bulk and synaptic cholesterol prompted us to examine whether this imbalance reflects compensatory transcriptional responses within the cholesterol biosynthetic pathway.

2.2. Cholesterol Biosynthesis Gene Expression Remains Unchanged in the Cerebellum of Mecp2-Null Mice

At three weeks of age, Mecp2-null mice are still asymptomatic and do not yet display overt RTT-related phenotypes. Therefore, we analyzed the expression of genes related to cholesterol synthesis to establish a baseline control. Our results indicate that RNA levels of enzymes involved in cholesterol synthesis in the cerebellum remain unchanged compared to wild-type at this RTT presymptomatic age (Figure 2a and Table 1). At seven weeks of age, when RTT symptoms are evident, we investigated the biosynthetic arm of the cholesterol pathway by focusing on key regulatory nodes—specifically Srebp2, Hmgcs1, and Sqle—which have been reported to be altered in other RTT models. However, after applying the Holm–Šídák correction for multiple comparisons (Table 2), we found no significant differences in the mRNA expression levels of these three genes compared to wild-type mice (Figure 2b). This lack of transcriptional downregulation in the cerebellum contrasts with findings in other brain regions of RTT models, particularly given the significantly elevated cerebellar cholesterol levels observed in our study.

To assess whether the unchanged mRNA levels were reflected at the protein level, we quantified the protein abundance of SREBP2 and HMGCS1. Given that cholesterol biosynthesis-related enzymes are membrane-associated, we specifically isolated the membrane fraction for the analysis and used Na^+^/K^+^ ATPase as a loading control. Consistent with the gene expression data, our results indicate no significant changes in the protein levels of SREBP2 (Figure 3a) nor HMGCS1 (Figure 3b).

Together, these results show that the expression of key cholesterol biosynthesis regulators remains stable at both the transcriptional and translational levels. The absence of expected downregulation in the presence of elevated sterol levels suggests a blunted homeostatic feedback response in the cerebellum of Mecp2-null mice, rather than a primary defect in the expression of genes and protein levels critical for determining cholesterol levels.

2.3. CYP46A1 Levels Are Decreased in the Cerebellum of Mecp2-Null Mice

Next, to determine whether elevated cholesterol levels are associated with defects in cholesterol transport in the cerebellum of Mecp2-null mice, we analyzed the expression of cholesterol-associated transporters.

As with cholesterol synthesis markers, we first examined cholesterol transport markers in 3-week-old Mecp2-null mice, before the onset of RTT phenotype. At this stage, RNA expression of the measured markers did not differ between Mecp2-null and wild-type mice (Figure 4a). However, by seven weeks of age, Cyp46a1 expression was significantly reduced in Mecp2-null mice compared to wild-type mice (Figure 4b). CYP46A1 hydroxylates cholesterol at position 24, converting it to 24S-hydroxycholesterol (24S-OHC), a more soluble derivative that can be transported across the blood–brain barrier to the liver for elimination [38]. Importantly, CYP46A1 is the principal neuronal enzyme responsible for sterol turnover in the brain, and even modest reductions in its activity can markedly impact cholesterol efflux.

mRNA expression of cholesterol transport-related genes in the cerebellum of Mecp2-null mice. mRNA expression was evaluated by RT-qPCR in the cerebellum of Mecp2-/Y and their wild-type littermate mice at 3 (a) and 7 (b) weeks of age. 3-week-old Mecp2-/Y (n = 9, triangles) and wild-type (n = 5, circles); 7-week-old Mecp2-/Y (n = 8, triangles) and wild-type (n = 7, circles). ND: not detected. Graph bars represent means ± SEM. Statistical analysis was performed using multiple Student’s t-tests followed by the Holm–Šídák correction to control the Family-Wise Error Rate (FWER) ∗ p < 0.05. Detailed statistical parameters, including adjusted p-values, are provided in Table 3 and Table 4.

Thus, the decrease in Cyp46a1 observed in the cerebellum of Mecp2-null mice represents a strong mechanistic candidate for the elevated cholesterol levels detected in cerebellar tissue and synaptosomes.

This pattern suggests that impaired cholesterol clearance, rather than enhanced synthesis, may be a key metabolic disturbance emerging during RTT progression. To further investigate this, we analyzed CYP46A1 protein levels. Consistent with the RNA data, our results confirm a significant decrease in CYP46A1 protein levels in the cerebellum of 7-week-old Mecp2-null mice (Figure 5). Together, the convergent reduction in Cyp46a1 mRNA and protein levels provides strong evidence that Mecp2 deficiency disrupts cerebellar cholesterol turnover at the level of neuronal sterol hydroxylation.

2.4. The Promoter Region of the Cyp46a1 Coding Gene Contains a CpG Island

MECP2 preferentially binds to methylated CpG dinucleotides, which are commonly found in DNA CpG islands, and interacts with transcriptional co-repressors [40,41]. Consequently, its deletion in the RTT mouse model alters gene expression [42] and delays motor learning [43], underscoring the critical role of MECP2 in this brain region. Given the transcriptional regulatory role of MECP2, and the emerging evidence linking it to metabolic gene control, we used bioinformatics tools to analyze the presence of CpG islands in the promoter region of the Cyp46a1 coding gene (Figure 6).

The results indicate that the promoter region of the CYP46A1 coding gene contains a CpG island enriched in CpG dinucleotides (Figure 6). The presence of this CpG island, along with the decreased expression of CYP46A1 in Mecp2-null mice, suggests that MECP2 might play a crucial role in epigenetic regulation of cholesterol metabolism associated with cerebellum postnatal neurodevelopment. Although bioinformatic prediction does not demonstrate direct binding, these findings raise the mechanistic possibility that MECP2 loss perturbs DNA methylation–sensitive transcriptional control of sterol-related genes, particularly Cyp46a1, potentially contributing to impaired cholesterol turnover in the cerebellum of RTT.

3. Discussion

In this study, we evaluated cholesterol levels in the cerebellum of Mecp2-null mice before and during the progression of RTT-related phenotype. At 7 weeks of age, Mecp2-null mice exhibited significantly elevated cholesterol levels in both the cerebellum and isolated cerebellar synaptosomes. This increase was accompanied by stable gene expression levels of Srebp2, Sqle, Hmgcs1, in contrast to a significant reduction in Cyp46a1, as well as a decrease in CYP46A1 protein levels, suggesting that Mecp2 plays a role in cholesterol metabolism. These findings collectively indicate that Mecp2 deficiency disrupts cerebellar cholesterol homeostasis by impairing clearance rather than altering synthesis. Specifically, the reduction in CYP46A1, coupled with sustained biosynthetic activity, suggests a clearance bottleneck rather than generalized metabolic downregulation.

Cholesterol plays a crucial role in synaptic function, and its dysregulation has been linked to neurodevelopmental and neurodegenerative disorders [29,44]. Given that cholesterol and its derivatives, such as 24-OHC, influence synaptic plasticity and neuronal signaling [39,45], disruptions in cholesterol homeostasis may contribute to the motor impairments observed in RTT. Our results reveal that while the expression of biosynthetic genes (Srebp2, Hmgcs1, Sqle) remains unchanged at both 3 and 7 weeks of age, Cyp46a1 levels significantly decline by the symptomatic stage. This suggests that the elevated cholesterol levels observed in the cerebellum of 7-week-old mice are driven by impaired clearance rather than enhanced synthesis. This temporal divergence—where homeostatic turnover fails during disease progression—aligns with the concept of MECP2-dependent epigenetic regulation, specifically affecting the maintenance of metabolic genes required for postnatal neurodevelopment.

Consistent with this mechanism, CYP46A1 acts as the key enzyme in central nervous system cholesterol clearance, converting cholesterol into the soluble metabolite 24-OHC for elimination from the encephalon [38,46]. The observed reduction in CYP46A1 at both RNA expression and protein levels in Mecp2-null provides a direct explanation for the impaired cholesterol turnover and the subsequent elevated cholesterol levels detected in the cerebellum of Mecp2-null mice and isolated cerebellar synaptosomes. Given CYP46A1′s role in modulating synaptic excitability and NMDAR signaling [39], its reduction potentially exerts functional implications beyond sterol accumulation, affecting cerebellar circuit performance and contributing to the RTT motor phenotype.

Interestingly, Cyp46a1 deletion in mice does not increase brain cholesterol levels [47]. This discrepancy with our findings may be attributed to several factors. First, in Cyp46a1-null mice, the enzyme is absent throughout development and growth, whereas in our model, Mecp2-null mice exhibit normal Cyp46a1 expression at 3 weeks, followed by a significant decrease at 7 weeks. Additionally, in Cyp46a1-null mice, cholesterol levels were measured in the whole encephalon. In contrast, our analysis focuses on one of the encephalic structures, the cerebellum, and also isolated cerebellar synaptosomes, providing a more anatomically region- and cell-specific perspective on cholesterol metabolism. Moreover, RTT models display broader transcriptional instability and compromised compensatory pathways, which may exacerbate sterol imbalance compared to Cyp46a1-null mice.

Our bioinformatic analysis of the Cyp46a1 promoter region identified a prominent CpG island, suggesting potential epigenetic regulation by MECP2. Although this analysis does not establish direct MECP2 binding, the presence of this CpG-rich regulatory region is consistent with a model in which MECP2 modulates metabolic gene expression via methylation-sensitive mechanisms [2,10,11,12,40,41]. This interpretation is supported by cerebellar transcriptomic analyses showing that altered Mecp2 levels lead to coordinated changes in hundreds of genes, including metabolic pathways, with most transcripts being downregulated in Mecp2-null and upregulated in Mecp2-Tg mice [42].

Altogether, our findings highlight impaired cholesterol clearance—driven by reduced Cyp46a1 mRNA expression and protein levels—as a previously underappreciated contributor to cerebellar metabolic dysfunction in RTT. This work provides a conceptual extension to existing RTT models by pointing to cholesterol turnover, rather than synthesis alone, as a key mechanistic node of vulnerability. Restoring cholesterol homeostasis, either by modulating CYP46A1 activity or by targeting upstream regulatory pathways, may thus represent a promising therapeutic strategy that warrants further preclinical investigation.

Limitations of the Study

Our findings identify impaired cholesterol clearance as a mechanistic node; however, certain experimental limitations should be noted. First, our biochemical and molecular analyses reflect steady-state levels rather than dynamic metabolic flux or turnover rates. Second, this study did not employ global lipidomic profiling or isotope labeling to trace precise cholesterol dynamics specifically within the cerebellum. Building on these limitations, our laboratory has uncovered significant remodeling of lipid composition in white adipose tissue [28] and in the encephalon of the same murine RTT model (unpublished data). We therefore hypothesize that cerebellar lipid homeostasis may likewise be disrupted, potentially contributing to region-specific neuropathophysiology. However, future investigations using mass spectrometry-based lipidomics and carbon-labeled tracking will be necessary to fully dissect the precise metabolic fluxes and specific lipid species altered in the cerebellum and their direct impact on the RTT motor phenotype.

4. Materials and Methods

4.1. Animals

Mecp2-null male mice (B6.129P2(C)-Mecp2^tm1^.^1Bird^/J, Strain #003890) and wild-type control mice were housed in a temperature-controlled room under a 12 h light/dark cycle, with food and water available ad libitum. All animal protocols were approved by the Bioethics and Biosafety Committee of Universidad San Sebastián, Chile (06-2021-10).

4.2. Isolation of Cerebellum and Cerebellar Synaptosomes

Mecp2-null male mice and their wild-type controls at 3 and 7 weeks of age were euthanized, and their cerebellums were collected. Synaptosomes were isolated from the cerebellum as previously described [48].

4.3. Measurement of Cholesterol in Cerebellum and Cerebellar Synaptosomes

Cholesterol extraction from cerebellum and cerebellar synaptosomes samples was performed using the Folch method [49,50]. Samples were first homogenized in PBS containing Halt™ Protease Inhibitor Cocktail, EDTA-Free (100X) (Thermo Fisher Scientific, #87785, Rockford, IL, USA). Methanol (3 mL) was then added, followed by sonication for 30 min. Chloroform (6 mL) was added, and the mixture was gently shaken for 1 min before incubation at 50 °C for 30 min in a thermoregulated bath. Subsequently, 2 mL of water was added, and the samples were left at 4 °C for at least 2 h. The mixture was centrifuged at 7000 rpm for 20 min at 4 °C, and the aqueous phase (upper layer) was carefully removed using a Pasteur pipette. The chloroform phase (lower layer) was transferred to a separate tube and evaporated at room temperature to isolate cholesterol. Cholesterol concentration was measured using the Amplex® Red Cholesterol Assay Kit (Thermo Fisher Scientific, #A12216, Waltham, MA, USA) according to the manufacturer’s instructions.

4.4. Real-Time PCR (qPCR)

Total RNA was extracted using the TRI Reagent® (Sigma-Aldrich, #T9424, St. Louis, MO, USA) and subsequently treated with the TURBO™ DNase kit (Ambion, #AM1907, Carlsbad, CA, USA) to remove genomic DNA contamination. Reverse transcription was carried out using the SuperScript IV Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. Quantitative PCR (qPCR) was performed on a StepOne Real-Time PCR System thermal cycler (Applied Biosystems). Gene expression levels were calculated using the 2^−ΔΔCt^ method. All mRNA levels were normalized to Ppia (Peptidylprolyl isomerase A) as a housekeeping gene [51]. All primer sequences utilized in this study are provided in Table 5.

4.5. Western Blot (WB)

Cerebellar pro5tein extraction was performed using the Mem-PER™ Plus Membrane Protein Extraction Kit (Thermo Scientific, #89842, Rockford, IL, USA). Protein concentration was determined with the BCA Protein Assay Kit (Thermo Scientific, #23225, Rockford, IL, USA) according to the manufacturer’s instructions. Protein extracts from the cerebellum and cerebellar synaptosomes were separated by 10–12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto PVDF membranes. Membranes were blocked with 5% non-fat dry milk in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) and incubated overnight at 4 °C with the following primary antibodies: anti-synapsin (1:1000, sc-376623, Santa Cruz Biotechnology, Dallas, TX, USA), anti-PSD95 (1:500, sc-32290, Santa Cruz Biotechnology), anti-actin (1:1000, sc-1616, Santa Cruz Biotechnology), anti-SREBP2 (1:500, PA1-338, Invitrogen, Carlsbad, CA, USA), anti-HMGCS1 (1:1000, 42201, Cell Signaling, Danvers, MA, USA), anti-CYP46 (1:500, sc-136148, Santa Cruz Biotechnology), and anti-ATPase (Na^+^/K^+^) alpha-1 subunit (1:500, AB_528090, DSHB). After washing with TBS-T, membranes were incubated with HRP-conjugated secondary antibodies, and protein bands were visualized using chemiluminescence with a G:Box imaging system (Syngene).

4.6. CpG Islands Prediction

The promoter and 5′ regulatory sequences of the Cyp46a1 gene (NM_010010.2) were retrieved from the NCBI database and analyzed using the UCSC Genome Browser (mouse assembly mm10). The identification of CpG islands was based on the default tracks provided by the browser, which define these regions as DNA segments of at least 200 bp with a guanine-plus-cytosine (G + C) content > 50% and an observed/expected CpG ratio > 0.6. The analysis was centered on the proximal promoter region, as annotated in the EPD-new database, relative to the transcription start site (TSS).