HepG2 (human hepatocellular carcinoma) and CCF-STTG1 (human astrocytoma) cells (ECACC, UK) were cultured in DMEM/F-12 with GlutaMAX™ supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S) (all from Thermo Fisher Scientific, Waltham, MA, USA) at 37 °C in a humidified 5% CO₂ incubator.

Primary human astrocytes (#1800, ScienCell, Carlsbad, CA, USA), kindly provided by Prof. B. Broux (Hasselt University, Belgium), were cultured in astrocyte medium (#1801, ScienCell) supplemented with 2% FBS, 1% astrocyte growth supplement, and 1% P/S in PLL-coated flasks, following the manufacturer’s protocol. Cells were seeded at 3.75 × 10⁶ cells per T-75 flask and cultured until 90% confluence.

THP-1 cells (human monocytic cell line; kindly provided by C. van Holten-Leelen, Erasmus MC, Rotterdam, The Netherlands) were cultured in RPMI 1640 with GlutaMAX™ and 25 mM HEPES, supplemented with 10% heat-inactivated FBS, 1% P/S, 1 mM sodium pyruvate, 0.05 mM 2-mercaptoethanol, and 11 mM glucose. For cytokine assays, cells were seeded in 24-well plates at 0.5 × 10⁶ cells/well and differentiated with 100 ng/mL phorbol 12-myristate 13-acetate (PMA, Merck) for 72 h, followed by 24 h in medium with 50 ng/mL PMA. For cholesterol efflux experiments, THP-1 cells (ECACC; Sigma-Aldrich, St. Louis, MO, USA) were cultured in RPMI 1640 medium (Euroclone, Milan, Italy) supplemented with 25 mM HEPES, 0.05 mM 2-mercaptoethanol, 0.5% v/v gentamicin, 11 mM glucose, 1 mM sodium pyruvate, and 10% FBS. Cells were seeded in 24-well plates at 0.5 × 10⁶ cells/well and differentiated with 100 ng/mL PMA for 72 h. Because these experiments were conducted at the Erasmus MC and at the University of Parma respectively, THP-1 cells were obtained from the locally available sources. Both cell lines were authenticated and maintained under comparable culture conditions to ensure experimental consistency.

HepG2 or CCF-STTG1 cells (0.55 × 10⁶) were plated in 4 mL DMEM/F-12 on T-25 dishes 24 h prior to transfection. Cells were transfected with pcDNA3.1/V5H6 vectors encoding LXRα or LXRβ, RXRα, LXRE-firefly luciferase, and renilla luciferase using FuGENE^® 6 (E2692, Promega, USA) (Zhan et al., 2023). After overnight incubation at 37°C/5% CO₂, cells were trypsinized, seeded in 96-well plates, and cultured in DMEM/F-12 with 10% FBS and 1% P/S. Cells were then incubated for 24 h with 2.5 μM 24(S/R)-saringosterol or fucosterol (isolated from Sargassum fusiforme, Prof. Liu, Ocean University of China) in phenol red-free medium containing 10% heat-inactivated FBS. Compounds were dissolved in ethanol, and vehicle controls containing the same final ethanol concentration (≤ 0.1% v/v) were included in all experiments to account for potential solvent effects. To confirm that ethanol at the applied concentrations does not affect cell viability, MTT assays were performed using 0.1%, 0.2%, and 0.3% ethanol; no significant effect on cell metabolism or viability was observed. Firefly and renilla luciferase activities were measured using the Dual-Luciferase^® Reporter (DLR™) Assay System (E1980, Promega, USA) on a Perkin Elmer Victor X4 luminometer.

HepG2 and CCF-STTG1 cells plated in 12-well plates were incubated with 24(S)-saringosterol, 24(R)-saringosterol, fucosterol (isolated from Sargassum fusiforme, provided by Prof. Liu, Ocean University of China), or SH42 (Item No. 34677, Cayman, USA), a selective DHCR24 inhibitor, in standard DMEM/F-12 culture medium for 24 h. Then, the extracellular medium was collected. The cells were washed with non-supplemented DMEM/F-12 (4 °C), trypsinized, and collected by centrifugation (4 °C, 1500 rpm, 5 min). Medium and cell samples were stored at -80 °C for subsequent analysis.

HepG2 cells were seeded at 1 × 10⁶ cells/well in 12-well plates and cultured for 24 h in DMEM/F-12 with 10% FBS and 1% P/S. Medium was replaced with DMEM/F-12 containing 10 mM 1-¹³C-acetate and 2.5 µM of 24(S)-saringosterol, 24(R)-saringosterol, or fucosterol. Cells were incubated for 48 or 72 h, with medium refreshed every 24 h. After incubation, cells were washed, trypsinized, centrifuged (4 °C, 1500 rpm, 5 min), dried, and weighed. Cells and media were analyzed for sterol content by GC-MS. Incorporation of ¹³C-acetate into cholesterol was quantified by analyzing m/z 372–379 and 462–469, correcting for natural ¹³C abundance (1.1%), and compared with unlabeled controls (Ahmed et al., 2014; Gkiouli et al., 2019).

THP-1-derived macrophages were labeled for 24 h with [1,2-³H]-cholesterol (2 μCi/mL, #NET13900, PerkinElmer, MA, USA) in RPMI with 1% FBS and 2 μg/mL ACAT inhibitor (#S9318, Sandoz 58-035, Sigma-Aldrich, St. Louis, MO, USA), according to a standardized radioisotopic technique (Turri et al., 2023). Cells were equilibrated for 20 h in 0.2% BSA medium (#A8806, Sigma-Aldrich) supplemented with 22-hydroxycholesterol (22-OHC, 12.4 µM, #H9384) and 9-cis-retinoic acid (9cRA, 10 µM, #R4643, Sigma-Aldrich) and treated with saringosterol, fucosterol, and desmosterol (1.25, 2.5, 5 µM) or vehicle (ethanol). Cholesterol efflux was induced for 6 h using 2% pooled human serum, 10 µg/mL ApoA-I, or 12.5 µg/mL HDL (#A0722 and #LP3 all from Sigma-Aldrich). Cholesterol efflux capacity (CEC) was calculated as the percentage of radiolabeled cholesterol released in culture medium relative to total cellular radioactivity and normalized to ethanol-treated controls. Intra-assay coefficient of variation for ApoA-I- and HDL-mediated CEC was <10%.

THP-1-derived macrophages were stimulated with 10 ng/mL LPS (Merck, St. Louis, MO, USA) and incubated with saringosterol, fucosterol, or desmosterol at 5 or 10 µM for 24 h at 37 °C and 5% CO₂. Culture media were collected and stored at −80 °C. Pro- and anti-inflammatory cytokines (IL-1β, IL-8, TNF-α, IL-6, IL-10) were quantified using a Human Magnetic Luminex Discovery Assay Kit (Kit Lot: L140180, Cat #: LXSAHM, R&D Systems, Abdingdon, UK) with data acquisition on BioPlex MAGPIX™ and analysis in Bio-Plex Manager MP software (Bio-Rad, Hercules, CA, USA).

Wild-type male C57BL6/J mice (8–10 weeks old) were obtained through in-house breeding (fucosterol trial: breeding protocol ID202132B, Hasselt University, saringosterol trial: breeding protocol 1005EM, University of Groningen). The mice were housed in a conventional animal facility, with ad libitum access to food and water, and maintained on an inversed 12 h light/dark cycle. The animal procedures were approved by the ethical committees for animal experiments of Hasselt University (fucosterol trial: protocol ID202249) or the University of Groningen (saringosterol trial: protocol ID AVD10500202115290) in accordance with institutional guidelines. Mice received either non-supplemented standard chow (vehicle, n = 5) or fucosterol (Lemeitian Medicine, Chengdu, China)-supplemented chow (0.2% w/w, n = 5) or saringosterol (COMFiON BV, Leimuiden, The Netherlands)-supplemented chow (0.02% w/w, n = 5) for 7 consecutive days. The fucosterol and saringosterol dosages were based on their average concentrations in seaweed extracts used in previous experiments (Bogie et al., 2019; Martens et al., 2024).

Mice were euthanized via intraperitoneal injection of Dolethal (200 mg/kg, Vetoquinol, Aartselaar, Belgium), followed by cardiac puncture and transcardiac perfusion with heparin-PBS. Blood was centrifuged at 4000 x g for 5 minutes to obtain serum, which was stored at -80°C. Brains were hemisected; the right cerebellum and hippocampus were snap-frozen for sterol analyses, and the left hippocampus for RNA sequencing. Livers were removed, snap-frozen, and stored at -80°C for sterol and triglyceride measurements.

Cells or tissue samples were spun in a speed vacuum dryer (Thermo Fisher Scientific) and weighed to determine the dry weight. (Oxy)sterols and stanols were quantified as previously described (Lütjohann et al., 2002; Mackay et al., 2014; Vanbrabant et al., 2021), using 50 µg 5α-cholestane (1 mg/mL, Serva, Heidelberg, Germany) and 1 µg epicoprostanol (100 µg/mL, Sigma, Deisenhofen, Germany) as internal standards. Steroids were extracted with cyclohexane after saponification and neutralization, evaporated, and derivatized to (di-)trimethylsilyl (TMSi)-ethers by adding 300 µL TMSi-reagent (pyridine-hexamethyldisilazane-chlorotrimethylsilane, 9:3:1, v/v/v; Merck, Darmstadt, Germany) and incubating for 2 h at 90°C. Derivatized samples were evaporated under nitrogen at 65°C, dissolved in 80 µL n-decane, and analyzed by GC-MS-SIM. Concentrations of sterols and stanols, including 24(S)- and 24(R)-saringosterol, fucosterol, cholesterol precursors, oxycholesterols, and plant sterols/stanols, were calculated from standard curves using epicoprostanol, while cholesterol concentrations were determined by one-point calibration with 5α-cholestane using GC-FID (Šošić-Jurjević et al., 2019).

Liver samples were homogenized using the BioSpec Mini-Beadbeater (Biospec Products, Bartlesville, OK, USA). Lipid extraction was performed as described by Bligh and Dyer (1959). Serum samples were collected as described in the tissue preparation section. Triglyceride concentrations in hepatic lipid extracts and serum were determined using a colorimetric enzymatic assay (DiaSys Diagnostic Systems, Holzheim, Germany) based on glycerol measurement. In brief, triglycerides were enzymatically hydrolyzed by lipoprotein lipase (LPL) to release glycerol and free fatty acids. The released glycerol was phosphorylated by glycerol kinase (GK) to glycerol-3-phosphate, which was then oxidized by glycerol-3-phosphate oxidase (GPO) to produce hydrogen peroxide. Hydrogen peroxide reacted with 4-aminoantipyrine and 4-chlorophenol in the presence of peroxidase (POD) to form quinoneimine, a chromophore detected colorimetrically. Baseline free glycerol in samples was corrected according to the manufacturer’s instructions.

Hippocampus samples of wild-type mice were homogenized (BioSpec Mini-Beadbeater, Biospec Products, Bartlesville, OK, USA), and total RNA was isolated as described above. RNA integrity was assessed using the RNA 6000 Nano Lab-on-a-Chip kit and a Bioanalyzer 2100 (Agilent Technologies, Amstelveen, The Netherlands). Libraries were prepared using the NEBNext Ultra II Directional RNA Library Prep Kit (NEB #E7760S/L, New England Biolabs, Ipswich, MA, USA), including mRNA isolation, fragmentation, cDNA synthesis, adapter ligation, and PCR amplification. Library quality and size distribution (300–500 bp) were confirmed with a Fragment Analyzer (Agilent Technologies). Sequencing was performed on an Illumina NovaSeq6000 (GenomeScan B.V., Leiden, the Netherlands) at 1.1 nM library concentration, yielding ≥15 million 150 nt paired-end reads per sample. Reads were aligned to Mus musculus GRCm38.gencode.vM19 using STAR 2.5, counted with HTSeq-count v0.6.1p1, and differential expression was analyzed using DESeq2. Functional enrichment of genes with p < 0.01 was performed using Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) pathways in R (version 4.4.1) using the clusterProfiler and enrichplot packages. Pathways with adjusted p < 0.01 were considered significantly enriched.

The chemical structures of saringosterol, fucosterol, and desmosterol were uploaded to the PubChem compound website (https://pubchem.ncbi.nlm.nih.gov/) and exported as an SDF file. Next, the SDF file was uploaded to the PharmMapper database for target prediction. Targets were identified with the fit score threshold set at 3.5 and the Z-score set at 1 (Wang et al., 2016; Wang et al., 2017). The human target proteins (25 predicted targets of saringosterol, 29 predicted targets of fucosterol, and 30 predicted targets of desmosterol) were imported into and analyzed by String (https://string-db.org/). KEGG pathways and Reactome pathways were analyzed to explore the effects of saringosterol, fucosterol, and desmosterol on metabolic pathways. The Venn diagram was drawn through venny 2.1.0 (https://bioinfogp.cnb.csic.es/tools/venny/).

Statistical analyses were performed using GraphPad Prism 9.5. Data are presented as mean ± SD. Differences between two groups were analyzed using the Mann–Whitney U test, and multiple-group comparisons were performed using the Kruskal–Wallis test followed by Dunn’s post hoc test. When comparing a single experimental group against a reference value of 1, a one-sample Wilcoxon signed-rank test was used. RNA sequencing data were analyzed using the Differential Gene Expression–Gene Set Analysis (DGE–GSA) algorithm, with statistical significance defined as p < 0.05.

Both fucosterol and saringosterol activated LXRα and LXRβ in HepG2 and in CCF-STTG1 cells (Figure 1). However, the activation potency of fucosterol was consistently lower than that of saringosterol. At the two tested concentrations (2.5 and 5.0 µM), both saringosterol epimers [24(S) and 24(R)] and fucosterol were internalized dose-dependently, with a similar fraction of sterols internalized (Figures 2A, B). Time-course experiments revealed comparable uptake kinetics across all sterols, but the intracellular levels of fucosterol reached only 40–60% of those of either saringosterol epimer within 24 h (Figure 2C). Despite increased cellular content over time, the apparent equilibration rate declined within 24 h for all sterols, with fucosterol exhibiting approximately half the rate observed for both saringosterol epimers (Figure 2D). All sterols underwent substantial re-release into the culture medium, indicating rapid equilibration with the extracellular environment. Notably, fucosterol was re-released to a higher extend than both saringosterol epimers, resulting in lower net cellular accumulation (Figures 2E, F). The weaker LXR activation observed for fucosterol may correlate with its reduced cellular retention, which results from both slower uptake and enhanced re-release.

We investigated whether the distinct LXR activation profiles of fucosterol and saringosterol differentially affect cholesterol biosynthesis in vitro. We first measured desmosterol, the immediate precursor of cholesterol and an endogenous LXR ligand. In CCF-STTG1 cells, fucosterol caused a trend toward increased intracellular desmosterol, accompanied by a non-significant increase in extracellular desmosterol (p > 0.05; Supplementary Figure S1), comparable to the Δ24-dehydrocholesterol reductase (DHCR24) inhibitor SH42 (Müller et al., 2017). In primary human astrocytes, fucosterol and saringosterol exerted opposing trends on desmosterol (increase of 31.9 ± 48.9% vs. decrease of 46.3 ± 27.4%), though these changes did not reach statistical significance (Supplementary Figure S2).

In HepG2 cells, as illustrated in the schematic representation of the cholesterol metabolic pathway (Figure 3A), fucosterol increased intracellular desmosterol while reducing lathosterol, without significantly affecting cholesterol or lanosterol levels (Figures 3B–E). While fucosterol caused a ~2-fold increase in desmosterol, SH42 induced a ~15-fold increase (Supplementary Figure S3). In contrast, both 24(S)- and 24(R)-saringosterol dose-dependently decreased desmosterol, cholesterol, and lathosterol, while lanosterol remained unchanged. (Figures 3B–E). Downstream cholesterol metabolites were also differentially affected. At 2.5 µM, 5α-cholestanol showed a decreasing trend with all sterols. At 5.0 µM, a significant reduction was observed for both 24(S)- and 24(R)-saringosterol, whereas fucosterol did not significantly alter 5α-cholestanol levels (Figure 3F). Both saringosterol epimers, but not fucosterol, dose-dependently increased 27-hydroxycholesterol (27-OHC) up to 3–5-fold (Figure 3G).

To quantify the effects of fucosterol and saringosterol on cholesterol biosynthesis, we performed stable isotope tracing in HepG2 cells using [1-¹³C]-acetate (Figure 4A). Both 24(S)- and 24(R)-saringosterol significantly suppressed de novo cholesterol synthesis, reducing intracellular ¹³C-cholesterol by 35–37% at 48 h and 27–35% at 72 h (p < 0.05, Figures 4B, C). Extracellular ¹³C-cholesterol was also decreased across all time points (24–72 h; Figures 4B, C). Conversely, fucosterol increased both intracellular (17.38 ± 9.07%) and extracellular (15.61 ± 7.36%) ¹³C-cholesterol levels, but only at the 72 h (p < 0.05), with no significant effects observed at earlier time points (Figures 4C, D).

Consistent with the results obtained in CCF-STTG1 and HepG2 cells (Figures 3, S1), fucosterol, but not saringosterol, increased intracellular desmosterol concentrations in THP-1-derived macrophages at 5.0 µM, as well as in the culture medium (Figures 5A, B). An increase in desmosterol concentrations in macrophages has been shown to dampen the inflammatory response (Spann et al., 2012; Zhang et al., 2021). Following LPS stimulation, saringosterol, desmosterol, and the DHCR24 inhibitor SH42, which promotes intracellular desmosterol accumulation, all reduced the production of TNF-α and IL-6 (Figures 5C, D). The production of IL-1β and IL-10 remained unaffected across all treatments (Figures 5E, F).

Cholesterol accumulation in macrophages is known to promote inflammatory responses, whereas enhanced cholesterol efflux can counteract this effect (Tall and Yvan-Charvet, 2015). We assessed the ability of saringosterol, fucosterol, and desmosterol to induce cholesterol efflux from THP-1-derived macrophages (Figure 5G). Cholesterol efflux to ApoA-I was increased by saringosterol (5.0 μM) and fucosterol (2.5 µM), and by desmosterol (2.5 and 5.0 µM) (Figure 5H). Fucosterol and desmosterol at 2.5 µM also significantly increased cholesterol efflux to HDL, while saringosterol did not (Figure 5I). Cholesterol efflux to serum was significantly induced exclusively by desmosterol (2.5 and 5 µM, Figure 5J).

Wild-type mice received diet supplementation with fucosterol for one week to determine the effect on sterol metabolism and the early phase modulatory effect on the transcriptome. Fucosterol supplementation led to an accumulation of fucosterol in liver, serum, cerebellum, and hippocampus (Figures 6A–D). Desmosterol concentrations were modestly increased in the liver (p < 0.01), but not in serum, hippocampus, or cerebellum. Alongside the rise in fucosterol, concentrations of other phytosterols were significantly reduced in liver and serum (Figures 6C, D). Fucosterol supplementation also lowered hepatic 5α-cholestanol, as well as 7α-OHC, and 27-OHC in serum (Figures 6C, D).

Saringosterol supplementation for one week increased saringosterol concentrations in the liver, serum, and cerebellum (Supplementary Figure S4). It also decreased the hepatic fucosterol and campesterol concentrations, whereas other plant sterols and total plant sterols remained unaffected. Unlike the reduction in desmosterol upon saringosterol administration to HepG2 cells, diet supplementation with saringosterol to wild-type mice slightly increased hepatic desmosterol concentrations, similar to fucosterol (Supplementary Figure S4B). Saringosterol supplementation decreased the lathosterol concentration and increased the 7α-OHC concentration in cerebellum (Supplementary Figure S4A), as well as 7α-OHC in liver and 27-OHC in serum (Supplementary Figures S4B, C).

To evaluate whether fucosterol affects hepatic lipogenesis, a common side effect of most synthetic LXR agonists (Cha and Repa, 2007), we assessed serum and hepatic triglyceride (TG) concentrations in mice after one week of fucosterol treatment: TG in serum and liver remained unaltered (Figure 7). Consistently, fucosterol did not affect the expression of SREBF1, the master regulator of lipogenesis, or SCD1, a key enzyme in fatty acid synthesis, in HepG2 cells (Supplementary Figure S5). These findings demonstrate that fucosterol administration does not induce lipogenesis or hypertriglyceridemia under these experimental conditions.

Although fucosterol accumulated in the brain following oral administration, it did not significantly affect levels of desmosterol or of other sterols in cerebellum or hippocampus, suggesting that its neurological effects are unlikely to be mediated via local changes in sterol composition, particularly cholesterol intermediates such as desmosterol. To explore the potential neural impact of fucosterol beyond sterol alterations, we performed RNA-sequencing of hippocampal tissue from mice after one week of its diet supplementation (Figure 8).

Transcriptome profiling revealed that fucosterol primarily modulated genes involved in neuronal structure and signaling. Gene Ontology (GO) enrichment highlighted cellular components such as axons and microtubules, and molecular functions related to myosin II binding, small GTPase binding, and histone modification. Biological processes included RNA polymerase III-mediated transcription, oligodendrocyte development, dendritic spine formation, and microtubule polymerization. KEGG analysis showed enrichment in apelin signaling, cholinergic synapses and insulin signaling pathways, suggesting enhanced neuronal plasticity and intracellular signaling (Figures 8A, B).

In contrast, saringosterol enriched pathways related to proteostasis and RNA metabolism, including protein refolding, RNA splicing and RNA catabolic process. Enriched cellular components included the proteasome, nuclear pores and ER-Golgi compartments. For molecular functions, enriched terms were mainly associated with chromatin modification, RNA regulation, and ubiquitin-mediated processes, such as histone methyltransferase activity, RNA polymerase II complex binding, NF-κB binding, and ubiquitin recognition. KEGG pathways implicated in neurodegenerative diseases—such as AD, Huntington’s, prion, and Parkinson’s diseases—were also among the top 10 enriched terms (Figures 8C, D), suggesting an effect on neurodegeneration-related processes.

Direct comparison of fucosterol versus saringosterol further underscored their differential effects. Fucosterol treatment was associated with enriched GO terms in chromatin remodeling, proteasome complex, microtubule anchoring, and positive regulation of transcription by RNA polymerase III (Figure 8E). KEGG pathways significantly enriched in this comparison included multiple neurodegenerative diseases (e.g., Parkinson’s disease, Huntington’s disease, prion disease, AD) (Figure 8F). Notably, these neurodegeneration-related pathways were absent in the fucosterol versus control group but present in both the saringosterol versus control and fucosterol versus saringosterol comparisons.

Importantly, hippocampal transcriptomic analyses did not reveal strong enrichment of cholesterol metabolism pathways, suggesting that the neural effects of fucosterol and saringosterol are mediated primarily through transcriptional programs related to neuronal structure, signaling, and protein homeostasis rather than through direct modulation of sterol metabolism.

To capture potential systemic mechanisms that might not be evident from brain-restricted transcriptome data, we performed in silico target prediction for fucosterol, saringosterol, and desmosterol using PharmMapper (Wang et al., 2016; Wang et al., 2017), with results for Homo sapiens summarized in Supplementary Table S1. Pathway enrichment analysis using Reactome and KEGG via STRING revealed several putative mechanisms through which these sterols may influence both brain and systemic cholesterol homeostasis and broader metabolic functions.

Shared Reactome pathways enriched by all three sterols included SUMOylation of nuclear receptors, bile acid synthesis via 27-OHC, nuclear receptor-mediated transcription, and general steroid metabolism (Figure 9A). These common pathways point to overlapping roles in lipid homeostasis and nuclear receptor signaling, consistent with hippocampal transcriptome findings implicating lipid-related signaling and cytoskeletal organization.

Fucosterol and saringosterol jointly enriched pathways related to LXR-mediated cholesterol uptake, bile acid homeostasis, lipogenesis, and PPARα-activated gene expression, further reinforcing their capacity to regulate lipid metabolism at the transcriptional level. Uniquely, fucosterol and desmosterol were associated with pathways related to retinoid metabolism, including the canonical retinoid cycle in rods, retinoid transport, and visual phototransduction, indicating a potential link between these sterols and retinoid-sensitive transcriptional networks. Fucosterol-specific targets were enriched in heme degradation and platelet sensitization by LDL.

KEGG enrichment also revealed a systemic role in detoxification and metabolic adaptation. Targets of all three sterols were significantly enriched in xenobiotic metabolism pathways, including cytochrome P450-related drug metabolism, steroid hormone biosynthesis, and insulin resistance. Additionally, desmosterol was significantly enriched in pathways such as chemical carcinogenesis, glutathione metabolism, Th17 cell differentiation, and adipocytokine signaling, suggesting a potential immunometabolic crosstalk mediated by this sterol (Figure 9B).

Overall, whereas hippocampal transcriptome primarily reflected neural-specific effects, these target predictions highlight broader systemic roles, particularly in cholesterol metabolism and lipid homeostasis, that are not directly apparent in the brain.