Cedrol mitigates hepatic lipid accumulation and adipocyte hypertrophy induced by corticosteroids through the inhibition of glucocorticoid receptor activity

Abstract

Cedrol is a sesquiterpene alcohol derived from ginger and cedar oil. Cedrol has multiple pharmacological effects such as sedation, promoting hair growth, decreasing blood pressure and reducing obesity. But its pharmacological mechanisms are not fully understood and its direct targets remain unknown. Glucocorticoids, particularly dexamethasone, are stress hormones in the body and interact with glucocorticoid receptors (GRs) to regulate physiological functions such as metabolism. In this study, we find that cedrol effectively mitigates the lipid accumulation in liver and adipose tissues induced by dexamethasone in adult male mice. Cedrol treatment also inhibits the dexamethasone-induced expression of genes involved in lipid metabolism, including Cd36, C/ebpβ, Srebp1, Fas and Scd1 in the liver. In addition, cedrol binds to GR in the cellular thermal shift assay and shows antagonistic activity in luciferase reporter assays. These results indicate that cedrol has GR antagonist activity, which may be responsible for its effect on lipid metabolism, and suggest that cedrol could also be potentially used in the treatment of lipid metabolism disorders induced by high glucocorticoids.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-32917-8.

Introduction

Cedrol is an important sesquiterpene alcohol found in cedar oil, which is extracted from various Cupressus and Juniperus species of the family Cupressaceae and can also be isolated from ginger^1,2. It is approved by the U.S. Food and Drug Administration (FDA) as an adjuvant or a flavoring in cosmetics, foods, and drugs³. Studies have shown that exposure to cedrol may induce anxiolytic effects⁴. Topical application of cedrol promotes hair growth in mice⁵. Cedrol can also decrease the blood pressure in humans by inhalation and ameliorate high-fat diet-induced obesity in mice^6,7. However, the detailed mechanism for cedrol’s pharmacological effects is not fully understood, and the direct target of cedrol has not yet been identified.

Glucocorticoids (GCs), one type of steroid hormones produced by the adrenal glands, bind with glucocorticoid receptors (GRs) to regulate specific gene expression, thereby playing a pivotal role in various physiological processes, including metabolism, stress response, etc.⁸. Elevated glucocorticoid levels could lead to anxiety⁹, hair loss¹⁰ and elevate blood pressure¹¹. Moreover, elevated glucocorticoid levels can disrupt lipid homeostasis, resulting in conditions such as central obesity and Cushing’s syndrome^12,13. Disrupted lipid homeostasis could lead to increased concentrations of triglycerides (TG) and total cholesterol (T-CHO) in the blood and liver, which contribute to the pathogenesis of atherosclerosis¹⁴ and non-alcoholic fatty liver disease (NAFLD)¹⁵, respectively. Therefore, cedrol’s pharmacological actions of anxiolytic effects, promoting hair growth, decreasing blood pressure and mitigating obesity may be related to its action on glucocorticoids. We hypothesize that cedrol could mitigate corticosteroid-induced central obesity and lipid metabolism dysfunction in vivo.

In this study, we reveal that cedrol could indeed mitigate corticosteroid (dexamethasone)-induced hepatic lipid accumulation and adipocyte hypertrophy in vivo, probably by acting as a GR antagonist. These results reveal novel pharmacological mechanisms and a potential direct target of cedrol, which may shed light on the understanding of cedrol’s multiple beneficial functions in humans.

Results

Cedrol inhibits dexamethasone (dex)’s effect on lipid metabolism in vivo

We found that in mice, dexamethasone (dex) treatment reduced body weight, induced leg and gonadal white adipose tissue (WAT) mass as well as interscapular brown adipose tissue (BAT) mass (Figs. 1A–C, S1A,B). Cedrol alone had no significant effect on body weight (BW), WAT and BAT mass (Figs. 1A–C, S1A,B). Cedrol treatment with dex attenuated the dex-induced increase in leg WAT mass (Fig. 1A). For gonadal WAT mass, there was no difference between cedrol + dex group with control group, indicating that cedrol could inhibit dex-induced gonadal WAT mass, although there was no statistical difference between cedrol+dex group and dex alone group (Fig. 1B). Cedrol treatment did not inhibit dex-induced BAT mass increase (Fig. 1C). The food intake was not altered by either dex or cedrol treatment (Fig. S1C).

Fig. 1.

Effect of cedrol (20 mg/kg/every two days) and dex (20 mg/kg, every two days) treatment for 16 days in adult male C57BL/6 J mice. (A) Leg WAT mass/body weight (BW). (B) Gonadal WAT mass/BW. (C) Interscapular BAT mass/BW. (D) Serum TG. (E) Liver TG. (F) Serum T-CHO. (G) Liver weight/BW. (H) Serum AST. (I) Serum ALT. The data are shown as mean ± SD (n = 5). *P < 0.05, **P < 0.01, ***P < 0.001 compared to control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to dex group.

Dex treatment induced serum and liver TG levels and serum T-CHO level, whereas cedrol treatment reduced dex-induced serum and liver TG levels and serum T-CHO level (Fig. 1D–F). Liver weight was induced by dex treatment, and cedrol treatment reduced liver weight, although there was no statistical significance between cedrol+dex group and dex alone group (Fig. 1G). Cedrol treatment reduced dex-induced serum levels of ALT and AST (Fig. 1H,I). Consistently, in another set of experiments, we found that cedrol inhibited dex-induced serum and liver TG in both male and female C57BL/6 J mice, similarly to a known GR antagonist, mifepristone (Fig. S2A–D). Taken together, these data reveal that cedrol mitigated the effect of dex on lipid accumulation and metabolism in adipose tissue and liver in vivo.

Histological analysis of liver tissue revealed that dex increased lipid droplets in the liver, and cedrol treatment drastically reduced dex-induced lipid accumulation in liver (Fig. 2A–D). After dex treatment, large vacuoles of lipid droplets were found in the cytoplasm of liver cells with the nucleus pushed to the side (Fig. 2A,B). After cedrol treatment, only small lipid droplets could be seen in the liver (Fig. 2D,E). Oil Red O staining in liver showed that cedrol treatment resulted in a significant reduction in lipid droplet size (Fig. 2F–J). Treatment with cedrol alone did not increase lipid droplets or alter lipid droplet size in liver cells compared to control (Fig. 2C,H).

Fig. 2.

Histological analysis of the liver. (A–D) H&E staining of liver in (A) control group, (B) dex group, (C) cedrol group, (D) dex + cedrol group. The images in the left and right columns were taken with 100 × and 400 × magnifications, respectively. Scale bar: 100 µm, and 50 µm for 100 × and 400 × magnification images, respectively. The black rectangle represents the zoomed-in area. (E) Quantification of the area of lipid droplets in H&E staining of liver. (F–I) Oil Red O staining of liver in (E) control group, (F) dex group, (G) cedrol group, and (H) dex + cedrol group. The images in the left, middle and right columns were taken with 100×, 400 × and 1000 × magnifications, respectively. Scale bar: 100 µm, 50 µm and 10 µm for 100×, 400 × and 1000 × magnification images, respectively. The black rectangle represents the zoomed-in area. (J) Quantification of the area of lipid droplets from Oil Red O staining of the liver. The data are shown as mean ± SD (n = 5). *P < 0.05, **P < 0.01, ***P < 0.001 compared to control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to dex group.

Histological analysis of leg WAT showed that the size of adipocytes in WAT increased significantly after dex treatment (Fig. 3A,B,E). Treatment with cedrol restored the size of adipocytes to the control level (Fig. 3D,E). Treatment with cedrol alone had no effect on adipocyte size (Fig. 3C). These results indicate that cedrol has an inhibitory effect on dex-induced adipocyte hypertrophy. In interscapular BAT, we found that supplementation of cedrol significantly inhibited the accumulation of lipid in BAT cells induced by dex (Fig. 3F–J). In sum, these results indicate that cedrol not only protected the liver against dex-induced lipid accumulation and liver injury, but also reduced dex-induced lipid accumulation in adipose tissues.

Fig. 3.

Histological analysis of adipose tissues. (A–D) H&E staining of leg WAT in (A) control group, (B) dex group, (C) cedrol group, (D) dex + cedrol group. Scale bar: 50 µm. (E) Quantification of average WAT adipocyte size. (F–I) H&E stained interscapular BAT in (F) control group, (G) dex group, (H) cedrol group, (I) dex + cedrol group. The images in the left and right columns were taken with 100 × and 400 × magnification, respectively. Scale bar: 100 µm, and 50 µm for 100 × and 400 × magnification images, respectively. The black rectangle represents the zoomed-in area. (J) Quantification of the area of BAT lipid droplets. The data are shown as mean ± SD (n = 5). *P < 0.05, **P < 0.01, ***P < 0.001 compared to control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to dex group.

Cedrol regulates the expression of lipid metabolism genes in the liver

To explore the mechanism of action of cedrol in reducing dex-induced hepatic lipid accumulation and adipocyte hypertrophy, we analyzed the expression levels of genes involved in lipid metabolism and lipogenesis in the liver by qPCR (Fig. 4). The result showed that dex significantly induced the expression of fatty acid transporter (FAT/Cd36), a gene that plays an important role in promoting fatty acid uptake, and liver lipid accumulation¹⁶, and this induction was significantly reduced by cedrol treatment (Fig. 4A). CD36 plays an important role in the pathogenesis of NAFLD as it promotes fatty acid uptake, which leads to the accumulation of TG in hepatocytes¹⁷. While there are currently no clinically approved CD36 inhibitors, CD36 remains a promising target for NAFLD. In recent years, there has been growing interest in the potential role of natural products that act through CD36 in treating NAFLD¹⁸. Therefore, cedrol may potentially serve as a drug candidate for the development of CD36-targeted therapy for NAFLD.

Fig. 4.

qPCR analysis of gene expression in mouse liver. (A–H). Expression of (A) Cd36, (B) C/ebpβ, (C) Srebp1, (D) Fas, (E) Scd1, (F) Fkbp51, (G) Fkbp52, (H) Grα. The relative gene expression levels were normalized with GAPDH. FAT/CD36, fatty acid translocase; C/EBPβ, CCAAT/enhancer binding protein beta; SREBP1, sterol-regulatory-element-binding protein 1; SCD1, stearoyl-CoA desaturas-1; FAS, fatty acid synthase; GR, glucocorticoid receptor; FKBP5, FK506-binding protein 5; GAPDH, Glyceraldehyde 3-phosphate dehydrogenase. Results are shown as means ± SD (n = 5). *P < 0.05, **P < 0.01, ***P < 0.001 compared to control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to dex group.

CCAAT/enhancer binding protein beta (C/ebpβ), a downstream gene of GR, plays an important role in the transcriptional regulation of FAT/Cd36. We found that C/ebpβ expression in liver was significantly induced by dex and inhibited after cedrol treatment (Fig. 4B). Sterol-regulatory-element-binding protein 1 (SREBP1) is involved in the transcriptional regulation of fatty acid synthesis-related genes, such as stearoyl-CoA desaturase-1 (SCD1) and fatty acid synthase (FAS)¹⁹. The expressions of Srebp1, Fas, and Scd1 were significantly increased by dex and inhibited after treatment with cedrol (Fig. 4C–E). In sum, these results indicate that cedrol could inhibit the expression of GR downstream genes, which are related to lipid metabolism.

qPCR analysis revealed that cedrol could inhibit dex-induced Fkbp51 and Fkbp52 mRNA levels (Fig. 4F,G), which is consistent with an early study showing that Fkbp5 mRNA expression was inhibited by a selective GR antagonist²⁰. Research has shown that FKBP51 and FKBP52 have different effects on GR. FKBP51 inhibits nuclear transactivation of GR, whereas FKBP52 promotes GR translocation²¹. We find that dex induced Fkbp52 to about 4 times of the control level and promoted GR activation, whereas cedrol cotreatment inhibited Fkbp52 expression level to the control level and inhibited GR activation (Fig. 4G). The changes in Fkbp51 expression level by dex and cedrol were much weaker than those of Fkbp52 (Fig. 4F). Therefore, the downregulation of Fkbp52 by cedrol may potentially contribute to the reduced GR transactivation by cedrol.

It is of note that in the liver, the mRNA level of Grα and Grβ was inhibited by dex (Fig. 4H,I), which may be related to the negative feedback regulation. Cedrol treatment restored Grβ but not Grα mRNA level (Fig. 4H,I).It has been reported that GRβ could inhibit GRα activation, although GRβ does not directly bind to glucocorticoids²². Therefore, the increase of Grβ mRNA level by cedrol treatment may also potentially contribute to the reduced activation of GRα by cedrol to some extent.

Cedrol can directly bind to GR and is a GR antagonist in vitro

To investigate whether cedrol directly binds and antagonizes GR, we carried out a series of experiments. First, we used molecular docking analysis, which revealed that hydrogen bonds were found to form between cedrol and GR, with a binding energy of ﹣31.0 kcal/mol (Fig. 5A). As comparison, the binding energy between GR and its ligand mifepristone is ﹣37 kcal/mol. Molecular docking suggests cedrol binds to the same binding pocket of dex with two hydrogen bonds formed between cedrol and N564 and Q570 (Fig. 5A). Because N564 can also form hydrogen bonds with dex²³, cedrol may be able to compete with the binding of dex with GR. Then we performed the Cellular Thermal Shift Assay (CETSA) to further evaluate the binding between cedrol and GR (Fig. 5B–E). We found that treatment of HeLa cell lysates with cedrol increased GR stability at higher temperatures (55 °C, 57 °C) in a similar way as the known agonist dex and antagonist mifepristone (Fig. 5B,C). In addition, isothermal dose–response CETSA (ITDRCETSA) was performed with different concentrations of cedrol at 57 °C (Fig. 5D,E). The result further demonstrated that GR was stabilized by the presence of cedrol in a concentration-dependent manner (Fig. 5D,E). Taken together, these results indicate that cedrol could probably bind to GR directly.

Fig. 5.

Molecular docking analysis of the cedrol-GR interaction and GCs antagonist activity of cedrol in vitro. (A) The interaction of cedrol and GR (PDB ID: 3H52) was revealed by molecular docking. The protein structure is shown in ribbons (colored blue). Cedrol and key amino acids in the binding site are shown in sticks. Pink dashed line: hydrogen bond. (B) Thermal stability of GR (determined by Cellular Thermal Shift Assay, CETSA) treated with dex (15 μM), cedrol (15 μM) or mifepristone (15 μM) at different temperatures. (C) Relative intensity of the protein bands in (B) was quantified using the ImageJ software. For CETSA curves, the band intensity was calculated relative to the respective band intensity at 37 °C for each group. Results are shown as means ± SD (n = 3). (D) ITDRCETSA result of GR at 57 °C in response to increasing concentrations of cedrol. (E) Relative intensity of the protein bands in (D) normalized to GAPDH. Results are shown as means ± SD (n = 3). (F) Reporter assay in HeLa cells for cedrol and dex (dexamethasone, a known GR agonist). (G) Reporter assay in HEK-293 T cells with wild-type and N564A GR plasmids. (H) Reporter assay in HEK-293 T cells with wild-type and Q570A GR plasmids. Results are shown as means ± SD (n = 4). *P < 0.05, **P < 0.01, ***P < 0.001 compared to control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to dex group.

The luciferase reporter gene assay was used to evaluate the effect of cedrol on the transcriptional activity of GR in HeLa cells. If GR is activated by ligands, it can bind to the glucocorticoid response element (GRE) and induce downstream luciferase gene expression. We found that at the concentration of 80 nM, hydrocortisone and dex, two known GR agonists, induced luciferase activity to the level of 4–5 times that of the control group (Figs. 5F, S3B). The treatment of cedrol inhibited hydrocortisone and dex-induced luciferase activity dose-dependently to the control level (Figs. 5F, S3B). Cedrol treatment itself did not induce luciferase activity (Fig. S3A). In addition, cedrol treatment did not significantly alter the expression level of GR in HeLa cells (Fig. S3C).

Single-point mutations of two residues in the GR-LBD (N564 and Q570) were constructed, which were identified by docking as potential residues that interact with cedrol. The full-length wild-type or mutant GR plasmids were co-transfected with the GRE-Luc reporter plasmid into HEK-293 T cells, followed by treatment with dex or cedrol. The results show that the N564A mutation completely abolished the agonistic activity of dex (Fig. 5G), indicating that N564 is indispensable for dex binding. In contrast, the Q570A mutation did not affect GR activation by dex (Fig. 5H), suggesting that Q570 is not important for dex binding. Therefore, dex mainly interacts with N564, but not Q570, in GR. It is of note that antagonism of cedrol on GR activation by dex was completely abolished with Q570A mutation, which indicates that Q570 is essential for cedrol binding. Although cedrol and dex bind GR-LBD via interacting with Q570 and N564, respectively, these two amino acids are very close to each other and located in the same binding pocket (Fig. 5A). Thus, cedrol appears to bind GR-LBD in a site overlapping with that of dex, suggesting it may act as a competitive antagonist. In sum, these studies indicate that cedrol directly binds to GR and serves as a GR antagonist to inhibit GR transcriptional activity. This result supports our in vivo findings that cedrol inhibits GR downstream gene expression in the liver (Fig. 4). Therefore, these in vitro studies indicate that cedrol directly binds to GR and acts as an antagonist, consistent with the observed inhibition of GR downstream genes in vivo.

Then, we investigated the effect of cedrol on the dex-induced subcellular translocation of GR. As demonstrated by immunofluorescence microscopy, treatment with dex alone induced prominent translocation of GR from the cytoplasm to the nucleus in HeLa cells (Fig. 6A), indicating GR activation with dex treatment. In contrast, cedrol (160 nM) alone showed no significant effect on GR localization, with fluorescence distributed in both the cytoplasm and nucleus, which was similar to the vehicle-treated control. However, treatment with cedrol almost completely abolished dex-induced GR nuclear translocation (Fig. 6). Quantitative analysis revealed that the ratio of nuclear-to-cytoplasmic (N/C) fluorescence intensity of GR was reduced to near the basal level upon treatment with cedrol and dex at the same time (Fig. 6B). These results suggest that cedrol functions as a GR antagonist that attenuates dex-induced GR activation and nuclear translocation.

Fig. 6.

Cedrol inhibits dex-induced nuclear translocation of GR in HeLa cells. (A) Representative immunofluorescence images of HeLa cells treated with vehicle, dex (80 nM), cedrol (160 nM) or dex (80 nM) + cedrol (160 nM) for 8 h. Cells were fixed and then stained for endogenous GR (using mouse anti-GR antibody) and DNA (DAPI). Scale bar = 10 µm. (B) Quantitative analysis of GR subcellular localization. The nuclear-to-cytoplasmic (N/C) ratio of GR fluorescence intensity was calculated for individual cells. Data represent the mean N/C ratio of GR fluorescence intensity ± SD from three independent experiments (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 compared to control, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to Dex group.

Discussion

In this study, we provide evidence showing that the bioactive sesquiterpene cedrol mitigates dex-induced hepatic steatosis, liver damage and adipocyte hypertrophy in mice (Figs. 1, 2, 3). Our data further suggest that these effects are associated with the inhibition of GR transcriptional activity and downstream lipid metabolism genes, such as Cd36, C/ebpβ, and Srebp1, in the liver (Fig. 4). We demonstrated that cedrol is a novel natural GR antagonist in vitro and in vivo (Figs. 5 and 6).

Non-alcoholic fatty liver disease (NAFLD) is a liver condition that occurs when there is an accumulation of excess lipid, especially TG, in the liver¹⁵. NAFLD can progress to serious liver damage, including cirrhosis and liver cancer. Similar to our results, previous studies have found that chronic administration of dex can lead to hepatic lipid accumulation, a characteristic symptom of NAFLD²⁴. In addition, lipidomic analysis of an acute hepatocyte-specific GR knockout model showed that GR knockout resulted in impaired regulation of TG, non-esterified fatty acids and sphingolipids, thereby affecting lipid metabolism²⁵. The observed reduction in serum ALT and AST levels following cedrol treatment indicates its hepatoprotective potential against dex-induced liver injury (Fig. 2H,I). In addition, cedrol could reduce the accumulation of TG and lipid droplets in the liver (Figs. 1E and 2). Therefore, cedrol could potentially be used for treating dex-induced NAFLD.

The results of this study can also shed light on the pharmacological mechanism of cedrol in other scenarios. For example, cedrol has a significant sedative effect and can relieve anxiety⁴. We hypothesize that the reported sedative and anxiolytic effects of cedrol might be partially mediated through its GR antagonist activity, given the central role of glucocorticoids in stress response. This hypothesis warrants direct testing in relevant models. Previous research findings suggested that chronic restraint stress may inhibit hair follicle growth via regulating the key elements of the central hypothalamic–pituitary–adrenal (HPA) axis, like GR²⁶. Therefore, the finding that topical application of cedrol promotes hair growth in mice⁵ might also be related to its GR antagonist activity. Further studies are merited to explore the mechanism of action of cedrol as a GR antagonist in these disease models.

Lastly, it is of note that cedrol has weak androgen receptor (AR) agonist activity in reporter assay in vitro (Fig. S4A). However, in our animal experiment, no effects of cedrol on male reproductive system organs such as seminal vesicles were found in male C57BL/6 J mice (Fig. S4B), indicating that at the dose used, cedrol’s effect on lipid metabolism in vivo is likely mediated primarily through GR rather than AR activation. For estrogen receptor, no significant ERα/ERβ activation was observed in the reporter assays by cedrol treatment (Fig. S4C,D). It merits further investigation to reveal other potential targets of cedrol, such as LXR, which also modulates lipid metabolism.

This study has some limitations. First, our in vivo conclusions are primarily based on a mouse model of dex-induced metabolic disturbance. Further studies are merited to study whether these findings are relevant to human pathophysiology of endogenous glucocorticoid excess. Second, while we identified GR antagonism as a key mechanism, the contributions of other potential targets of cedrol such as other nuclear receptors to the observed metabolic effects cannot be fully ruled out and merit further investigation. Third, the pharmacokinetics, optimal dosing regimen, and long-term safety profile of cedrol for potential therapeutic applications remain to be determined. Finally, the precise molecular mechanisms by which cedrol inhibits GR function warrant further elucidation. Our data have shown that cedrol could decrease dex-induced gene transcription. However, whether cedrol disrupts GR’s interaction with transcriptional coactivators (e.g., p300/CBP) or alters the binding of GR to promoters of genes merits further investigation using chromatin immunoprecipitation (ChIP) sequencing and co-immunoprecipitation (Co-IP) experiments for p300/CBP. In addition, cedrol may have potential indirect interaction with GR, such as interfering with GR-cochaperone HSP90/PPID interaction or GR dimerization and activation, which merits further investigation.

In conclusion, our study identifies cedrol as a natural GR antagonist and provides evidence for its potential use in ameliorating glucocorticoid-induced disturbances in lipid metabolism.

Materials and methods

Chemicals and reagents

Dex and mifepristone were purchased from Macklin Biochemical Co., Ltd. (Shanghai, China). Cedrol was purchased from Aladdin Biochemical Co., Ltd. (Shanghai, China). Glutaraldehyde, ethanol anhydrous, corn oil, isopropyl alcohol, formaldehyde solution, phosphate buffer saline (PBS) and Dimethyl sulfoxide (DMSO) were purchased from Macklin Biochemical Co., Ltd. (Shanghai, China). Dulbecco Modified Eagle Medium (DMEM), fetal bovine serum (FBS) and penicillin–streptomycin (P/S) were purchased from Gibco Co., Ltd. (Grand Island, NY, USA). Lipofectamine 3000 transfection reagent was purchased from Thermo Fisher Scientific (San Jose, CA, USA). FastPure tissue total RNA isolation kit V1, HiScript III 1st strand cDNA synthesis kit, and Tan Pro Universal SYBR qPCR Master Mix were purchased from Vazyme Biotech Co., Ltd. (Nanjing, China). Firefly luciferase reporter gene assay kit, protease inhibitor cocktail and Bicinchoninic Acid Assay (BCA) protein assay kit were purchased from UElandy Co., Ltd. (Suzhou, China). The plasmids pGL3- glucocorticoid response element (GRE)-Luciferase, pGL3-androgen response element (ARE)-Luciferase, pGL3-estrogen response element (ERE)-Luciferase, pcDNA3.1-GR, pcDNA3.1-AR, pcDNA3.1-ERα and pcDNA3.1-ERβ were constructed by GENEWIZ, Inc. (Suzhou, China). The triglyceride (TG), total cholesterol (T-CHO), alanine transaminase (ALT) and aspartate aminotransferase (AST) kits were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China).

Animals and treatment

All experimental procedures involving animals were approved by the Institutional Animal Care and Use Committee (IACUC) of The Chinese University of Hong Kong (Shenzhen) (Protocol Number: CUHKSZ-AE2021005). This study was conducted in compliance with the ARRIVE guidelines. All methods were performed in accordance with the relevant guidelines and regulations. C57BL/6 J mice (9–10 weeks old, weighing 20–25 g) were purchased from Guangdong Laboratory Animal Center (Guangdong, China). All animals were acclimatized for 7 days before the experiments. Mice were housed in specific-pathogen-free (SPF) conditions with 22 ± 2 ℃, 50–60% humidity and an 12h light / dark cycle. Mice had free access to water and a commercially prepared laboratory animal diet. A total of 20 male mice were weighed and then randomly divided into 4 groups (n = 5 per group): vehicle control, dex (20 mg/kg), cedrol (20 mg/kg), and dex (20 mg/kg) + cedrol (20 mg/kg). In a second batch of animal experiments with both male and female C57BL/6 J mice, animals were randomly divided into 4 groups (n = 5 per group): vehicle control, dex (20 mg/kg), dex (20 mg/kg) + cedrol (20 mg/kg), and dex (20 mg/kg) + mifepristone (20 mg/kg).

Drugs were dissolved in DMSO, diluted with corn oil and administered by intraperitoneal injection to all animals every two days for 16 days. The doses were selected based on the use of dex in literature²⁷. After treatment, mice were fasted overnight and euthanized. Mice were anesthetized using isoflurane (3% for induction, 2% for maintenance in medical-grade oxygen) using a precision vaporizer. Following confirmation of deep anesthesia (assessed by absence of pedal reflex), euthanasia was performed by exsanguination via hepatic portal vein puncture. Blood samples were collected for subsequent analysis. Blood was collected into ethylenediaminetetraacetic acid (EDTA) tubes, and plasma samples were prepared by centrifugation at 1000 × g for 10 min and stored in aliquots at -80℃. Liver was quickly removed, gonadal white adipose tissue (WAT) and leg WAT were taken and weighed from all animals. Interscapular brown adipose tissue (BAT) was also collected and weighed. Half of these tissues were stored in formaldehyde for histological studies, and the other half were immediately frozen in liquid nitrogen and then stored at -80℃. The biochemical parameters in the plasma were measured with the commercial kits according to the manufacturer’s instructions.

Histological analysis

The hematoxylin and eosin (H&E) and Oil Red O staining of liver and adipose tissues were performed by the Best Medical Diagnostic Technology Co., Ltd. (Beijing, China). Part of these tissues were fixed in paraformaldehyde, dehydrated, embedded in paraffin, sectioned with 10 µm thickness and stained with H&E using standard protocols. A separate part of these tissues were used for frozen sections and Oil Red O staining, with hematoxylin counterstaining. Images were captured using an optical microscope (ML31, Mshot, Guangzhou, China). For the quantitative analysis of the total area of lipid droplets and the average size of adipocytes, we randomly selected 5 sections from each animal and 5 fields from each section. ImageJ software was used for the quantitative analysis.

Real-time PCR (qPCR)

RNA was isolated from animal tissues or HeLa cells with the FastPure tissue total RNA isolation kit V1. Total purified RNA was quantified using NanoDrop (Thermo Fisher Scientific, USA). Then, cDNA was synthesized by reverse transcription using a high-capacity cDNA reverse transcription kit. qPCR was performed with ChamQ Universal SYBR qPCR Master Mix in a QuantStudio 6 Flex Real-Time PCR System (Thermo Fisher Scientific, USA). All primers were synthesized by Sangon Biotech (Shanghai, China) and the primer sequences for the genes analyzed are listed in Tables S1 and S2. All samples were run in duplicate and repeated three times. The relative expression values of the target genes were normalized to Glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The primers for mice qPCR were chosen based on a previously published study²⁸. The expression was evaluated using ΔΔCT method by QuantStudio™ Real-Time PCR Software v1.7.1 (Thermo Fisher Scientific, USA).

Computational modeling

The computational modelling analysis was carried out according to our previous report²⁹. Briefly, energy minimizations, molecular dynamics simulations, and molecular docking were performed with the DiscoveryStudio (DS) modelling software (Version 16.1.0.15350, Dassault Systemes BIOVIA (San Diego, CA, USA) installed on a Windows Server R1 operating system on a Dell PowerEdge R740 workstation. The 1D structure of cedrol was obtained from the PubChem database, and then the crystal structure of the human GR-mifepristone complex (PDB ID: 3H52) was obtained from the RCSB PDB database. Since our results showed that cedrol could potentially act as a GR antagonist, this GR structure (PDB ID: 3H52) in the antagonist-bound conformation was used as the protein structure for docking.

Prior to docking, the ligand and protein structures underwent preprocessing using the DS modelling software. Water molecules have been removed, and hydrogen added back. Bonds and bond order have been corrected. Then the structures underwent energy minimization and molecular docking analysis. The binding energy (kcal/mol) value indicates the strength of interaction between the ligand and protein. A lower binding energy suggests a more stable binding between the receptor and ligand. The three-dimensional and two-dimensional images were prepared using PyMOL and LigPlus software, respectively.

Cell culture

The HeLa cells and HEK-293 T were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The HeLa cells and HEK-293 T were grown in complete DMEM supplemented with 10% FBS and 1% P/S. All cell culture dishes and plates were incubated at a 37 ℃ atmosphere containing 5% CO₂.

Site-directed mutagenesis of plasmids

Fast Site-Directed Mutagenesis Kit (KM101, TIANGEN Biochemical Technology Co., Ltd., Beijing, China) was used to construct the point mutation in the human full-length GR. The primers (Table S3) were designed and synthesized by Shenzhen Hechengyuan Biotechnology Co., Ltd (Shenzhen, China). A PCR reaction system (total volume 50 μl) was prepared with 10–100 ng DNA Template pcDNA3.1-GR, 400 nM forward and reverse mutation primers, 10 μl 5 × FastAlteration Buffer, and FastAlteration DNA Polymerase 0.02U/μl. The PCR product was digested with Dpn I restriction enzyme at 37℃ for 1 h. Then, the reaction product was transformed into bacteria and single colonies were picked for sequencing by Sangon Biotech Co., Ltd (Shanghai, China).

Luciferase reporter assay

HeLa cells with endogenous GR expression were seeded in 24-well plates (1 × 10⁴ cells/well) and cultured for two days in complete DMEM medium supplemented with 10% FBS and 1% P/S. The pGL3-GRE-Luciferase was transfected into the HeLa cells using Lipofectamine 3000 transfection reagent with 0.25 μg/well plasmid. Eight hours after transfection, cells in 24-well plates were treated with the compounds in different concentrations for 48 h. Then, cells were lyzed, and luciferase activity was measured with a firefly luciferase reporter gene assay kit by PerkinElmer Envision multimode plate reader (PerkinElmer, Waltham, MA, USA). The luciferase activity in each well was normalized by the protein concentration.

HEK-293 T (2 × 10⁴ cells/well) cells were seeded in 24-well plates (CORNING, NY, USA) and cultured for 72 h in complete RPMI-1640 medium with 10% DCC-treated-FBS and 1% P/S. The ARE-Luciferase reporter plasmid and the expression plasmid AR were co-transfected into the cells using Lipofectamine 3000 transfection reagent at 0.25 and 0.1 μg/well, respectively. Eight hours after transfection, cells in 24-well plates were treated with the compounds in different concentrations for 48 h. Then, cells were lyzed, and luciferase activity was measured with a Firefly luciferase reporter gene assay kit (UElandy, Suzhou, China) by PerkinElmer EnVision multimode plate reader (PerkinElmer, Waltham, MA, USA). The luciferase activity in each well was normalized by the protein concentration.

The ER reporter assay using ERα and ERβ expression plasmids and pGL3-ERE-Luciferase was carried out in a similar way as the above AR reporter assay. The GR reporter assay using WT and mutant (N564A, Q570A) GR expression plasmids and pGL3-GRE-Luciferase was carried out in a similar way as the above AR reporter assay.

Cellular thermal shift assay (CETSA)

CETSA was performed according to a previously described method³⁰. Briefly, HeLa cells were harvested and lyzed in RIPA buffer supplemented with the protease inhibitor cocktail for 30 min on ice. Then, cell lysates were sonicated for 3 min. The cell homogenate was centrifuged at 10,000 × g for 5 min at 4 °C, and the protein concentration of the supernatant was measured by BCA. The supernatant was diluted to a protein concentration 1 mg/mL, and then 5 µL was added to an equal volume of the diluted drugs in PBS with a final concentration of 15 μM cedrol, 15 μM dex and 15 μM mifepristone. The reaction mixture was heated at different temperatures for 5 min and then cooled on ice for 10 min. Meanwhile, 5 µL of protein was mixed with an equal volume of the diluted different concentrations of cedrol at 57 °C for 5 min and then cooled on ice for 10 min. The protein samples were mixed with 5 × SDS sample buffer and then separated by electrophoresis using SDS-PAGE. The following primary antibodies were used for immunoblotting: GR (#12041S, Cell Signalling Technology, Massachusetts, USA), GAPDH (#A19056, ABclonal, Wuhan, China). The second antibody is Goat anti-rabbit HRP conjugated antibodies (#A0108, Beyotime, Shanghai, China). For quantification of Western blots, the signal for each band was quantified using ImageJ software (National Institutes of Health, Bethesda, Maryland, USA).

Immunofluorescent staining

Serum-starved HeLa cells seeded on coverslips were subjected to an 8-h treatment with PBS, dex (80 nM), cedrol (160 nM) or dex (80 nM) + cedrol (160 nM). Following treatment, cells were fixed with 4% paraformaldehyde and permeabilized using 0.2% Triton X-100. Immunofluorescence staining was performed with the primary antibody rabbit anti-GR (CST) and secondary antibody goat anti-rabbit IgG (H + L) Cross-Adsorbed Alexa Fluor™ 488 (A11008, Thermo Fisher Scientific, USA). Nuclei were counterstained with DAPI. Imaging was carried out using a Zeiss LSM900 laser scanning confocal microscope (Oberkochen, Germany). The subcellular distribution of GR protein was quantified using the nuclear-to-cytoplasmic (N/C) fluorescence intensity ratio using the built-in analysis software of the confocal imaging system. Data are presented as the average of nuclear-to-cytoplasmic (N/C) fluorescence intensity ratio ± SD from three independent repeats (n = 3), with more than 100 cells quantified in each treatment.

Statistical analysis

All experimental data were analyzed using GraphPad Prism 9 software program (Inc., San Diego, CA, USA). Values are expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was performed to determine whether differences existed between all groups, followed by Tukey’s multiple comprisons test, and then the Student’s unpaired t-test was employed to determine the significance of the differences. Differences with a value of p < 0.05 were considered statistically significant.

Author contributions

Minghui Shu: Writing—original draft & editing, Methodology, Investigation, Formal analysis. Pan Wang: Writing—review & editing, Supervision, Methodology, Conceptualization.

Funding

This study is supported by Ganghong Young Scholars Development Fund (No.1011E0009, 1011E0031), Shenzhen Key Laboratory of Steroid Drug Discovery and Development (No. ZDSYS10190901093417963), Shenzhen-Hong Kong Cooperation Zone for Technology and Innovation (No. HZQB-KCZYB-1010056), Shenzhen Science and Technology Program (No. JCYJ20220818103008017) and Longgang District Science and Technology Bureau’s Key Laboratory Program.

Data availability

The data in this study will be made available upon request by contacting the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.

Ethics statement

This study was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) at the Chinese University of Hong Kong (Shenzhen) (Protocol Number: CUHKSZ-AE2021005) dated Nov 28th, 2021 in compliance with regulations and Animal Research Reporting of In Vivo Experiments (ARRIVE) guidelines.