Hepatic-specific vitamin D receptor downregulation alleviates aging-related metabolic dysfunction-associated steatotic liver disease

Abstract

BACKGROUND

Metabolic dysfunction-associated steatotic liver disease (MASLD) is defined by the abnormal lipid deposition in hepatocytes. The prevalence of MASLD is significantly increased in the elderly population, suggesting that aging may be related to the occurrence of MASLD. Emerging evidences suggest that vitamin D receptor (VDR) may be implicated in the progression of MASLD. Therefore, additional researches are warranted to elucidate whether VDR plays a role in aging-related MASLD.

AIM

To investigate the relationship between aging and MASLD and explore the role and related mechanisms of VDR in aging-related MASLD.

METHODS

Cellular senescence models were established, and the senescence phenotype of telomerase RNA component knockout mice was validated. These mice were then used as a senescence model for subsequent studies. Changes in VDR expression in the livers of aging mice were examined. VDR knockdown models, including cell knockdown models and hepatic-specific VDR knockout mice, were constructed, and MASLD was established in these models. Additionally, vitamin D (VD)-supplemented models, including senescent liver cell lines and senescent mice, were constructed.

RESULTS

The steatosis in senescent liver cells was more severe than in normal cells (P < 0.05). Moreover, hepatic steatosis was significantly more pronounced in senescence model mice compared to control group when the MASLD model was successfully induced (P < 0.05). Therefore, we concluded that aging aggravated hepatic steatosis. The hepatic expression of VDR increased after aging. VDR knockdown in senescent liver cells and senescent mice alleviated hepatic steatosis (P < 0.05). When senescent liver cells were stimulated with VD, cellular steatosis was aggravated (P < 0.05). However, VD supplementation had no effect on aging mice.

CONCLUSION

Aging can lead to increased hepatic steatosis, and the hepatic-specific knockdown of VDR alleviated aging-related MASLD. VDR could serve as a potential molecular target for aging-related MASLD.

Key Words: Aging; Metabolic dysfunction-associated steatotic liver disease; Vitamin D receptor; Steatosis; Hepatocytes

Core Tip: We showed that aging exacerbated the pathogenesis of metabolic dysfunction-associated steatotic liver disease. Hepatic-specific knockdown of vitamin D receptor alleviated the progression of aging-related metabolic dysfunction-associated steatotic liver disease. Vitamin D receptor is a crucial factor implicated in steatosis.

INTRODUCTION

Metabolic dysfunction-associated steatotic liver disease (MASLD) is defined by the excessive intracellular lipid deposition in liver cells, primarily presenting as diffuse bullae-like steatosis[1,2]. The pathological progression of MASLD includes steatosis without substantial inflammation, metabolic dysfunction-associated steatohepatitis (MASH), cirrhosis, and ultimately hepatocellular carcinoma[3-6]. The excessive accumulation of lipids in the liver induces hepatocyte injury, triggers inflammation and leads to insulin resistance (IR). These factors, in turn, exacerbate hepatic steatosis, creating a vicious cycle that drives the progression of MASLD[7,8].

Aging can induce cellular senescence and is associated with the development of MASLD[9-11]. With aging, the structure and function of the liver change, thereby interfering with glucose regulation and lipid processing pathways. Studies have shown that an increase in senescent cells in the liver contributes to hepatic fat deposition and steatosis[12]. Additionally, aging exacerbates MASLD progression through the induction of ferroptosis in hepatocytes[13].

Regarding the relationship between human aging and MASLD, age is the main risk factor for MASLD[14,15]. Global population-based epidemiological studies have revealed that MASLD impacts around 25.2% of individuals worldwide, while its prevalence in China is higher at 29.2%[16]. Significant differences have been observed across the age distributions of the population. A cross-sectional study suggested that the prevalence of MASLD in individuals over 50 years old is 30.7%, whereas the prevalence of MASLD in those under 50 years old is only 19.8%[17]. Moreover, some basic studies have shown that the levels of many senescence markers, such as P16, P21, and P53, are evidently increased in the hepatocytes and adipose tissue of patients with MASLD[18,19] and that the degree of MASLD correlates positively with the levels of hepatocyte senescence markers[12,20,21]. Liver biopsies obtained from patients with MASLD exhibited elevated expression of p53 compared to those from healthy individuals[22]. However, the link and associated mechanisms between aging and MASLD have not yet been fully elucidated. Further studies are needed.

As a lipophilic secosteroid, vitamin D (VD) is primarily synthesized mainly by the exposure of the skin to the ultraviolet radiation in sunlight[23,24]. Once produced, VD is transported to the liver through the bloodstream, bound to albumin and VD-binding protein. With the participation of 25-hydroxylase (OH) secreted by the liver, VD is hydroxylated and forms 25-(OH)-VD3, after which is transported to the kidney through the blood and further converted into biologically active 1,25-(OH) 2-VD3 by 1α-hydroxylase[25-27]. Ultimately, following systemic circulation, VD-binding protein acts as a carrier to deliver VD to target organs like the small intestine and bone. Here, it binds to the VD receptor (VDR) and exerts the associated biological effects[28-30].

VDR belongs to the nuclear receptor superfamily. Its protein structure is shared by the nuclear receptor, i.e., an N-terminal DNA binding domain that binds to the VDR enhancer region of the downstream gene promoter and plays a transcriptional regulatory role[31-33]. VDR also contains a C-terminal ligand-binding domain[31].

VDR has been reported to be involved in MASLD in several studies in recent years. The expression of VDR in the livers of mice fed a high-fat diet (HFD) was obviously increased. Furthermore, compared to wild-type (WT) mice, VDR-knockout mice presented significantly less hepatic steatosis after HFD consumption[34]. Other studies have shown that in hepatic-specific VDR-knockout mice, a high-fat model of hepatic steatosis was significantly aggravated[35]. Thus, it appears that VDR plays a role in the progression of MASLD. Additionally, aging has an impact on the metabolism of VD. Older people may be exposed to sunlight for decreased periods because of decreased mobility or spending more time indoors, and the capacity of the skin to synthesize VD diminishes with age[36,37]. In addition, the physiological response to VD is weakened, resulting in relative VD resistance[38]. After aging, renal function and the ability to activate VD decrease[39,40].

Therefore, aging and VDR may be associated with the development of MASLD. We speculated that VDR is involved in aging-related MASLD and performed relevant research.

MATERIALS AND METHODS

Animal study design

Animals: Male C57BL/6 mice (WT mice) and telomerase RNA component knockout mice (Terc^-/- mice) were used. Using the clustered regularly interspaced short palindromic repeats/CAS9 system, VDR^flox/flox mice (Biocytogen, Beijing, China) on a C57BL/6J background were generated. Hepatocyte-specific VDR knockout mice (Vdr^Cre/+ mice) were subsequently generated by crossing VDR^flox/flox mice with albumin promoter-driven Cre recombinase transgenic mice (Biocytogen). Terc^-/- mice were injected with adeno-associated virus 8 (AAV8)-short hairpin (sh)-VDR (AAV8-shVDR) or AAV8-sh-negative control (AAV8-shNC) (Vigene Bio, Jinan, Shandong Province, China) via the tail vein to obtain hepatic-specific VDR knockdown mice and the corresponding control mice. Moreover, Terc^-/- mice were injected with AAV8-VDR (AAV8-VDR) or AAV8-negative control (AAV8-NC) (Hanbio, Shanghai, China) via the tail vein to obtain hepatic-specific VDR-overexpressing mice and the corresponding control mice. All experimental mice were maintained in standardized conditions with controlled ambient temperature (23 ± 2 °C) and a 12-hour light/dark cycle, provided with unrestricted access to food and water throughout the study period.

Diets and experimental procedure: Mice were randomly assigned to experimental groups and control groups. In the experimental groups, an HFD (Research Diet, New Brunswick, NJ, United States) was administered for 8 to 14 weeks to establish the MASLD model. A methionine-choline-deficient diet (MCD) (Research Diet) was administered for 6 weeks to establish the MASH model. For the VD supplementation experiment, a VD-supplemented diet [standard chow diet (SCD) + VD)] (Shuyishuer Bio, Changzhou, Jiangsu Province, China) was administered for 14 weeks. This diet contained 2000 IU of VD per 4057 kcal. The corresponding control groups were fed a SCD (Shuyishuer Bio). At the conclusion of the animal experiments, the mice were fasted overnight for 16 hour prior to being sacrificed. All animal procedures adhered to the ethical guidelines for animal use and were approved by the Ethics Committee of The First Affiliated Hospital, Zhejiang University School of Medicine (No. 2024-1652).

Laboratory investigations

Biochemical and metabolic indicators: Serum triglyceride (TG), total cholesterol (TC), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels, and hepatic TG and TC levels were detected with the corresponding kits (Applygen, Beijing, China). All the measurement procedures were conducted in accordance with the guidelines provided by the manufacturers.

Glucose tolerance test and insulin tolerance test: For the glucose tolerance test (GTT), the mice were fasted for 16 hours and then received an intraperitoneal injection of 1 g/kg glucose (Sigma, St. Louis, MO, United States). For the insulin tolerance test (ITT), the mice were fasted for 6 h and subsequently administered an intraperitoneal injection of 0.75 U/kg of regular human insulin (Wanbang, Xuzhou, Jiangsu Province, China). Blood glucose levels were monitored using a glucometer (Onetouch, Milpitas, CA, United States) at baseline (0 minute) and at 15, 30, 60, 90 and 120 minutes post-injection.

Histological analyses

The livers from the mice were first fixed with 10% neutral formalin, followed by paraffin embedding and sectioning for hematoxylin-eosin (HE) staining. For Oil Red O staining and senescence-associated β-galactosidase staining, fresh liver tissues were embedded in optimal cutting temperature compound at -20 °C and then sectioned. An Oil Red O staining kit (Jiancheng, Nanjing, Jiangsu Province, China) and a β-galactosidase staining kit (Solarbio, Beijing, China) were used.

Cell cultures and treatments

The normal mouse hepatocyte cell line (alpha mouse liver 12, AML12) and human hepatoma cell line (human hepatocellular carcinoma, HepG2) were purchased from the Chinese Academy of Science (Shanghai, China). AML12 cells were cultured in dulbecco’s modified eagle medium (DMEM)/F12 (HyClone, Logan, UT, United States) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA, United States). HepG2 cells were cultured in DMEM (HyClone) supplemented with 10% fetal bovine serum (Invitrogen). All the cell lines were cultured at 37 °C under an atmosphere containing 5% carbon dioxide. Oleic acid (OA) (Sigma) and palmitic acid (PA) (Sigma) were dissolved in a 25% bovine serum albumin (BSA) (SangonBio, Shanghai, China) solution to obtain PA (20 mmol/L) and OA (20 mmol/L) solutions. Free fatty acids (FFAs) were prepared by mixing OA and PA at a ratio of 2:1. Cells were treated with PA (300 μM) or FFA (1 mmol/L) and incubated for 24 hours to establish a cellular model of MASLD. Nutlin-3a (10 μM; APExBIO, Houston, TX, United States) dissolved in dimethyl sulfoxide was applied to the cells and incubated for 48 hours to modulate the expression of P53. We also dissolved 1,25(OH)2VD3 (calcitriol) (100 nM; Sigma) in ethanol, which was incubated with the cells for 12 hours, followed by costimulation with PA for another 24 hours. The AML12 cells and HepG2 cells were transfected with small interfering RNAs (siRNAs), including the small interfering-VDR (siVDR) or the small interfering-negative control (siNC) (RuiboBio, Guangzhou, Guangdong Province, China), using Lipofectamine 3000 (Invitrogen) to knock down VDR expression in the cell lines.

Western blotting

Proteins were extracted from treated cells or liver samples using radio immunoprecipitation assay buffer (FudeBio, Hangzhou, Zhejiang Province, China) supplemented with protease and phosphatase inhibitors (FudeBio). The protein concentrations were subsequently measured using a bicinchoninic acid protein assay kit (Beyotime, Shanghai, China). The samples were adjusted to the same concentration and heated for denaturation. Protein samples and molecular weight markers were loaded onto sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels and run at 80-130 V until the proteins were fully separated. The proteins were subsequently transferred to a polyvinylidene difluoride membrane (Millipore, Danvers, MA, United States). After blocking with 5% nonfat milk, the corresponding antibodies were used to detect specific proteins. The following antibodies were used: Those from Proteintech (Rosemont, IL, United States) included P16 (1:1000, 10883-1-AP), P21 (1: 1000, 10355-1-AP), and LIPIN1 (1:1000, 27026-1-AP) antibodies; those from Cell Signaling Technology (Danvers, MA, United States) included glyceraldehyde-3-phosphate dehydrogenase (1:1000, 2118) and P53 (1:1000, 2524) antibodies; those from Santa Cruz Biotechnology (Dallas, TX, United States) included VDR antibodies (1:300, sc-13133); and those from Abcam (Cambridge, United Kingdom) included FASN antibodies (1:1000, ab128856).

Quantitative real-time polymerase chain reaction

RNA extraction was conducted using the SteadyPure Quick RNA extraction kit (Accurate, Changsha, Hunan Province, China) and was subsequently reverse transcribed into cDNA. Quantitative real-time polymerase chain reaction (qPCR) was carried out using SYBR Green (Accurate) on a QuantStudio5dx system. The qPCR primers are detailed in Supplementary Table 1. Glyceraldehyde-3-phosphate dehydrogenase served as an internal control to normalize the relative expression levels of the target RNA, and the expression levels were calculated using the ΔΔ^Ct method.

Statistical analysis

All the data are expressed as the means ± SDs. Statistical comparisons between groups were conducted using either unpaired two-tailed student’s t tests or one-way analysis of variance with Tukey’s correction, as appropriate. All analyses were conducted using GraphPad Prism 9.0 (La Jolla, CA, United States), and a P < 0.05 was deemed statistically significant.

RESULTS

Cellular senescence aggravated steatosis

We used two common hepatocyte lines, AML12 and HepG2, and two classical cell senescence modeling approaches to establish models of hepatocyte senescence. AML12 and HepG2 cells were treated with hydrogen peroxide (H₂O₂) (400 μM) for 6 hours or irradiated with 8 Gy of X-rays and then stained with β-galactosidases. Both cell lines presented greater degrees of staining after senescence was established (Figure 1A and B). We then examined the changes in the levels of several senescence markers after cellular senescence modeling. When AML12 and HepG2 cells were treated with H₂O₂, the mRNA levels of P53 were increased, but the protein levels of P16, P21, and P53 were not different (Figure 1C and D). When AML12 and HepG2 cells were subjected to irradiation, the mRNA levels of P21 and P53 and the protein levels of P21 and P53 were higher (Figure 1E and F). Therefore, the hepatocyte senescence models were established successfully. We used different concentrations of PA or FFAs to stimulate hepatocyte steatosis. In AML12 cells, cellular senescence and stimulation with different concentrations of PA or FFAs resulted in increased levels of TGs and lipids, as shown by Oil Red O staining (Figure 1G-I). Therefore, cellular senescence can aggravate steatosis.

Figure 1 Establishment of a cellular senescence model and aggravation of steatosis in senescent cells. A: HepG2 and AML12 cells were treated with hydrogen peroxide (H₂O₂) (400 μM) for 6 hours and stained for β-galactosidase (× 400); B: HepG2 and AML12 cells were irradiated with 8 Gy of X-rays and stained for β-galactosidase (× 400); C-F: Protein expression levels of P16, P21, and P53 and corresponding quantitative results, and mRNA levels of P16, P21, and P53 in HepG2 and AML12 cells after cellular senescence was established; G: Cellular levels of triglycerides (TGs) and Oil Red O staining (× 400) of AML12 cells treated with H₂O₂ (400 μM) for 6 hours and stimulated with different concentrations of palmitic acid (PA) (100 μM, 200 μM, or 300 μM) for 24 hours; H: Cellular levels of TGs and Oil Red O staining (× 400) of AML12 cells irradiated with 8 Gy of X-rays and stimulated with different concentrations of PA (100 μM, 200 μM, or 300 μM) for 24 hours; I: Cellular levels of TGs and Oil Red O staining (× 400) of AML12 cells irradiated with 8 Gy of X-rays and stimulated with different concentrations of free fatty acids (0.6 mmol/L, 0.8 mmol/L, or 1 mmol/L) for 24 hours. Protein levels and mRNA levels were normalized to glyceraldehyde-3-phosphate dehydrogenase. The data are presented as the means ± SDs. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P calculated according to student’s t tests. H₂O₂: Hydrogen peroxide; NC: Negative control; BSA: Bovine serum albumin; SA-β-Gal: Senescence-associated β-galactosidase; PA: Palmitic acid; FFA: Free fatty acids; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

Senescence aggravated hepatic steatosis in mice

We examined the liver senescence phenotype of Terc^-/- mice at 3 months of age. Liver sections were stained for β-galactosidase, and Terc^-/- mice presented greater degrees of staining than WT mice (Figure 2A). The mRNA levels of P16 and P53 and the protein expression level of P53 were obliviously higher in Terc^-/- mice than in WT mice (Figure 2B). No significant differences in body weight; liver weight; the serum levels of ALT, AST, and TGs; or the hepatic levels of TGs and TC. The only difference was in the serum TC level (Figure 2C). HE and Oil Red O staining of the liver sections revealed that the two groups of mice had no significant hepatic steatosis (Figure 2D). Therefore, we concluded that the aging phenotype of Terc^-/- mice was more obvious than that of WT mice and that no spontaneous hepatic steatosis was observed at 3 months of age. Thus, we used Terc^-/- mice as an aging model for subsequent studies.

Figure 2 Establishment of a mouse aging model and aggravation of hepatic steatosis after senescence. A: Β-Galactosidase staining (× 200) of liver sections from 3-month-old telomerase RNA component knockout mice (Terc^-/-) or wild-type (WT) mice; B: Protein expression levels of P16, P21, and P53 and corresponding quantitative results, and mRNA levels of P16, P21, and P53 in 3-month-old Terc^-/- or WT mice; C: Weights, liver weights, serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), triglycerides (TGs), and total cholesterol (TC) and hepatic levels of TGs and TC in 3-month-old Terc^-/- or WT mice; D: Hematoxylin-eosin (HE) and Oil Red O staining (× 200) of the liver sections; E-H: The 3-month-old Terc^-/- or WT mice were fed a high-fat diet or standard chow diet (SCD) for 8 weeks to establish a model of metabolic dysfunction-associated steatotic liver disease; Weights and changes in weight (E); Glucose tolerance test and associated area under the curve. Insulin tolerance test and associated area under the curve (F); Liver weights, liver-to-body weight ratios, serum levels of ALT, AST, TGs, and TC, and hepatic levels of TGs and TC (G); HE and Oil Red O staining (× 200) of the liver sections (H); I-K: Three-month-old Terc^-/- or WT mice were fed a methionine/choline-deficient diet or SCD for 6 weeks to establish a model of metabolic dysfunction-associated steatohepatitis; Weights and changes in weight (I); Liver weights, liver-to-body weight ratio, serum levels of ALT, AST, TGs, and TC, and hepatic levels of TGs (J); HE and Oil Red O staining (× 200) of the liver sections. Protein levels and mRNA levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase (K). The data are presented as the means ± SDs. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P as determined via student’s t tests and one-way analysis of variance with Tukey’s correction. WT: Wild-type; HE: Hematoxylin-eosin; MCD: Methionine-choline-deficient diet; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TG: Triglycerides; TC: Total cholesterol; SCD: Standard chow diet; HFD: High-fat diet; GTT: Glucose tolerance test; ITT: Insulin tolerance test; AUC: Area under the curve; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

First, 3-month-old Terc^-/- or WT mice were fed an HFD or SCD for 8 weeks to establish a model of MASLD. After HFD feeding, Terc^-/- mice had greater weights than WT mice (Figure 2E), and the GTT and ITT revealed more severe abnormalities in glucose tolerance and IR in Terc^-/- mice (Figure 2F). The hepatic TG and TC levels and HE and Oil Red O staining of the liver sections also showed that Terc^-/- mice presented more severe hepatic steatosis (Figure 2G and H).

Second, 3-month-old Terc^-/- or WT mice were fed an MCD or SCD for 6 weeks to establish a model of MASH. After consuming the MCD diet, weight loss did not differ between Terc^-/- and WT mice. Liver weights, liver-to-body weight ratios, serum levels of ALT and AST, and hepatic TG levels were higher in Terc^-/- mice than in WT mice (Figure 2I and J). HE and Oil Red O staining of liver sections showed that Terc^-/- mice presented more severe hepatic steatosis (Figure 2K). The use of Terc^-/- mice as an aging model for establishing MASLD and MASH revealed that hepatic steatosis was more pronounced in senescent mice, indicating that senescence could aggravate hepatic steatosis in these mice.

VDR expression was upregulated in the liver when senescence was induced

After Terc^-/- and WT mice were fed an MCD diet for 6 weeks, the protein and mRNA levels of VDR and P53 in Terc^-/- mice were higher than those in WT mice (Figure 3A). Therefore, we hypothesized that VDR expression would be increased in the liver after senescence. For further examination, we selected naturally aged WT mice (18 months old) and compared them with young WT mice (2 months old). Both the protein and mRNA levels of VDR and P53 were higher in the livers of aged WT mice (Figure 3B). We then examined changes in VDR expression in senescent AML12 and HepG2 cells. The mRNA level of VDR was higher in the senescent cells than in the control cells, except when HepG2 cells were treated with H₂O₂ (Figure 3C-F).

Figure 3 Vitamin D receptor expression in the liver increased after senescence. A: Three-month-old telomerase RNA component knockout mice (Terc^-/-) or wild-type (WT) mice were fed a methionine/choline-deficient diet for 6 weeks. The protein expression levels of P53 and vitamin D receptor (VDR) and the corresponding quantitative results, and the mRNA levels of P53 and VDR are shown; B: Two-month-old WT mice and 18-month-old WT mice were fed a standard chow diet. Protein expression levels of P53 and VDR and corresponding quantitative results, and mRNA levels of P53 and VDR; C-F: MRNA levels of VDR in AML12 or HepG2 cells after senescence was established; AML12 cells were treated with hydrogen peroxide (H₂O₂) (400 μM) for 6 hours (C); AML12 cells were irradiated with 8 Gy of X-rays (D); HepG2 cells were treated with H₂O₂ (400 μM) for 6 hours (E); HepG2 cells were irradiated with 8 Gy of X-rays (F); G: MRNA levels of P53 and VDR in AML12 or HepG2 cells treated with Nutlin-3a (10 μM) for 48 hours; H: MRNA levels of P53 and VDR in senescent AML12 or HepG2 cells treated with Nutlin-3a (10 μM) for 48 hours. Protein and mRNA levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase. The data are presented as the means ± SDs. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P calculated according to student’s t tests. WT: Wild-type; MCD: Methionine-choline-deficient diet; H₂O₂: Hydrogen peroxide; VDR: Vitamin D receptor; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; NC: Negative control.

Therefore, the results of the animal and cell-based experiments were consistent. These results showed similar expression patterns of P53 and VDR. P53 is a classical marker of senescence. Therefore, we considered whether the increased expression of VDR in the liver after aging was related to P53. We used a P53 agonist, Nutlin-3a, to induce P53 expression. The results revealed that in both normal and senescent liver cell lines, Nutlin-3a elevated the mRNA levels of both P53 and VDR (Figure 3G and H). Given these findings, we speculated that the upregulation of VDR observed in the liver after aging may be associated with the increased expression of P53.

The downregulation of VDR alleviated PA-induced steatosis when cellular senescence was induced

Because senescence can lead to the aggravation of liver steatosis and an increase in VDR expression, we investigated whether VDR could affect aging-related MASLD. Therefore, we constructed siVDR to knock down VDR expression in liver cell lines. VDR expression was significantly decreased at both the mRNA and protein levels after siVDR treatment (Figure 4A and B). Lower TG levels were observed in AML12 cells treated with H₂O₂ to induce senescence, transfected with siVDR to knock down the expression of VDR and then stimulated with PA. Oil Red O staining revealed a lower amount of lipids (Figure 4C and D). Similar results were obtained after the same treatments were administered to HepG2 cells (Figure 4E and F). We therefore concluded that knocking down VDR in senescent liver cells could alleviate PA-induced steatosis.

Figure 4 Downregulation of vitamin D receptor alleviated palmitic acid-induced steatosis when cellular senescence was induced. A: Protein expression level of the vitamin D receptor (VDR) and corresponding quantitative results, and mRNA level of VDR in AML12 cells transfected with the small interfering-negative control (siNC) or the VDR knockdown construct small interfering-VDR (siVDR); B: Protein expression level of VDR and corresponding quantitative results, and mRNA level of VDR in HepG2 cells treated with siNC or siVDR; C and D: Cellular levels of triglycerides and Oil Red O staining (× 400) of AML12 cells treated with hydrogen peroxide (H₂O₂) (400 μM) for 6 hours, transfected with siNC or siVDR, and stimulated with palmitic acid (PA) (300 μM) for 24 hours; E and F: Cellular levels of triglycerides and Oil Red O staining (× 400) of HepG2 cells treated with H₂O₂ (400 μM) for 6 hours, transfected with siNC or siVDR, and stimulated with PA (300 μM) for 24 hours. Protein levels and mRNA levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase. The data are presented as the means ± SDs. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P calculated as determined via student’s t tests and one-way analysis of variance with Tukey’s correction. VDR: Vitamin D receptor; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; si-NC: Small interfering-negative control; si-VDR: Small interfering-vitamin D receptor; H₂O₂: Hydrogen peroxide; BSA: Bovine serum albumin; PA: Palmitic acid.

The hepatic-specific knockout of VDR ameliorated hepatic steatosis associated with senescence

We subsequently investigated the impact of VDR on hepatic steatosis in aging mice. We selected naturally aged 18-month-old VDR^flox/flox mice and VDR^Cre/+ mice. Their body weights were recorded each month. No differences in body weight were observed between the two groups (Figure 5A). The GTT results indicated that the abnormal glucose tolerance was ameliorated in VDR^Cre/+ mice. Moreover, the ITT results were not different between the two groups (Figure 5B). We found that hepatic TG levels were higher in VDR^flox/flox mice than in VDR^Cre/+ mice (Figure 5C). HE and Oil Red O staining of liver sections also showed that VDR^flox/flox mice presented more severe hepatic steatosis (Figure 5D), which indicated that the hepatic-specific knockout of VDR could ameliorate aging-related MASLD in aging mice. We further examined the hepatic levels of malondialdehyde and performed immunohistochemistry for 4HNE and F480 (Figure 5E and F). No obvious differences were detected between the two groups, suggesting that the ameliorative effects may not be related to lipid peroxidation and inflammation.

Figure 5 The hepatic-specific knockout of vitamin D receptor ameliorated hepatic steatosis associated with senescence. Eighteen-month-old vitamin D receptor (VDR)^flox/flox mice and 18-month-old VDR^Cre/+ mice were fed a standard chow diet. A: Weights and changes in weight; B: Glucose tolerance test and associated area under the curve (AUC). Insulin tolerance test and associated AUC; C: Liver weights, liver-to-body weight ratio; serum levels of alanine aminotransferase, aspartate aminotransferase, triglycerides (TGs), and total cholesterol (TC), and hepatic levels of TGs and TC; D: Hematoxylin-eosin and Oil Red O staining (× 200) of the liver sections; E: Immunohistochemical (IEC) staining (× 200) for VDR in liver sections and hepatic levels of malondialdehyde; F: IHC staining (× 200) for F480 in liver sections. The data are presented as the means ± SDs. ^aP < 0.05. P calculated according to student’s t tests. SCD: Standard chow diet; GTT: Glucose tolerance test; ITT: Insulin tolerance test; AUC: Area under the curve; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TG: Triglycerides; TC: Total cholesterol; HE: Hematoxylin-eosin; MDA: Malondialdehyde; IEC: Immunohistochemical.

The hepatic-specific downregulation of VDR ameliorated hepatic steatosis in Terc^-/- mice

We used Terc^-/- mice as an aging model. They were injected with AAV8-shNC or AAV8-shVDR at 3 months of age. Two weeks post-injection, the mice were fed an HFD or SCD for 14 weeks to establish a model of MASLD. Mouse body weights were measured weekly throughout the study period, and the weights of the mice injected with AAV8-shVDR were lower than those of the mice injected with AAV8-shNC (Figure 6A). Next, we assessed the effect of AAV8-shVDR-mediated knockdown and found that the protein and mRNA levels of VDR and immunohistochemical staining for VDR in liver sections were decreased (Figure 6B). The ITT results indicated that IR was ameliorated in Terc^-/- mice injected with AAV8-shVDR. However, the GTT results were not different between the two groups (Figure 6C). Serum ALT and hepatic TG levels were higher in Terc^-/- mice injected with AAV8-shNC (Figure 6D), and HE and Oil Red O staining of liver sections showed that Terc^-/- mice injected with AAV8-shNC presented more severe hepatic steatosis (Figure 6E). Furthermore, the downregulation of VDR also tended to decrease the protein level of FASN, which can promote lipid synthesis, and increased the protein level of LIPIN1, which can induce lipid oxidation (Figure 6F). We concluded that the hepatic-specific downregulation of VDR in aging mice could alleviate HFD-induced hepatic steatosis by decreasing lipid synthesis and promoting lipid oxidation.

Figure 6 The hepatic-specific downregulation of vitamin D receptor ameliorated hepatic steatosis in telomerase RNA component knockout mice mice. Telomerase RNA component knockout mice were injected with adeno-associated virus 8 (AAV8)-short hairpin (sh)-negative control (NC) or AAV8-sh-vitamin D receptor (VDR) at 3 months of age. Two weeks after the injection, these mice were fed a high-fat diet for 14 weeks. A: Weights and changes in weight; B: The protein expression level of VDR and corresponding quantitative results, the mRNA level of VDR, and immunohistochemical staining (× 200) for VDR in liver sections; C: Glucose tolerance test and associated area under the curve (AUC). Insulin tolerance test and associated AUC; D: Liver weights, liver-to-body weight ratio, serum levels of alanine aminotransferase, aspartate aminotransferase, triglycerides (TGs), and total cholesterol, and hepatic levels of TGs; E: Hematoxylin-eosin and Oil Red O staining (× 200) of the liver sections; F: The protein expression levels of FASN and LIPIN1 and corresponding quantitative results. Protein levels and mRNA levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase. The data are presented as the means ± SDs. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P calculated according to student’s t tests. AAV8: Adeno-associated virus 8; shNC: Short hairpin-negative control; shVDR: Short hairpin vitamin D receptor; HE: Hematoxylin-eosin; VDR: Vitamin D receptor; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; GTT: Glucose tolerance test; ITT: Insulin tolerance test; AUC: Area under the curve; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TG: Triglycerides; TC: Total cholesterol.

The upregulation of VDR aggravated PA-induced steatosis when cellular senescence was induced

The knockdown of VDR in senescent liver cells alleviated PA-induced steatosis. We investigated whether elevated VDR expression affects cellular steatosis. We used VD to stimulate VDR expression and examined the effects of VD on AML12 and HepG2 cells separately. VDR expression was markedly elevated at both the mRNA and protein levels after VD treatment (Figure 7A and B). We used H₂O₂ or irradiation to establish senescence models in AML12 and HepG2 cells. The aging cells were then stimulated with VD for 12 hours and costimulated with VD and PA for 24 hours. The levels of TG and the degree of Oil Red O staining were increased in the group treated with VD. Moreover, the protein levels of VDR increased, and the protein levels of LIPIN1 decreased (Figure 7C-E). Interestingly, the expression of VDR decreased after PA stimulation compared with that after BSA stimulation in AML12 cells, unlike in HepG2 cells. However, similar results were not obtained with irradiated AML12 cells. These findings indicated that the upregulation of VDR in senescent liver cells could aggravate PA-induced steatosis via the downregulation of LIPIN1 to inhibit lipid oxidation.

Figure 7 The upregulation of vitamin D receptor aggravated palmitic acid-induced steatosis when cellular senescence was induced. A: Protein expression level of vitamin D receptor (VDR) and corresponding quantitative results, and mRNA level of VDR in AML12 cells treated with vitamin D (VD) (100 nM) for 12 hours; B: Protein expression level of VDR and corresponding quantitative results, and mRNA level of VDR in HepG2 cells treated with VD (100 nM) for 12 hours; C-E: Cellular levels of triglycerides, Oil Red O staining (× 400), and protein expression levels of VDR and LIPIN1 and corresponding quantitative results after cellular senescence was induced and cells were treated with VD and palmitic acid (PA). AML12 cells were treated with hydrogen peroxide (H₂O₂) (400 μM) for 6 hours, treated with VD (100 nM) for 12 hours, and then costimulated with VD (100 nM) and PA (300 μM) for 24 hours (C); HepG2 cells were treated with H₂O₂ (400 μM) for 6 hours, treated with VD (100 nM) for 12 hours, and then costimulated with VD (100 nM) and PA (300 μM) for 24 hours (D); HepG2 cells were irradiated with 8 Gy of X-rays, treated with VD (100 nM) for 12 hours, and then costimulated with VD (100 nM) and PA (300 μM) for 24 hours (E). Protein and mRNA levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase. The data are presented as the means ± SDs. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P calculated as determined via student’s t tests and one-way analysis of variance with Tukey’s correction. VDR: Vitamin D receptor; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; VD: Vitamin D; H₂O₂: Hydrogen peroxide; BSA: Bovine serum albumin; PA: Palmitic acid.

The hepatic-specific overexpression of VDR had no significant effect on hepatic steatosis in Terc^-/- mice

We used Terc^-/- mice as an aging model and injected them with AAV8-NC or AAV8-VDR at 3 months of age to determine whether VDR overexpression affected hepatic steatosis in aging mice. Two weeks after the injection, these mice were fed an HFD for 14 weeks to establish a model of MASLD. The body weights did not differ between the groups (Supplementary Figure 1A). We subsequently examined the effect of AAV8-VDR overexpression and found that the protein and mRNA levels of VDR and immunohistochemical staining for VDR in liver sections were increased in Terc^-/- mice injected with AAV8-VDR (Supplementary Figure 1B). The GTT; ITT; liver weight; liver-to-body weight ratio; serum levels of ALT, AST, TGs, and TC; and hepatic levels of TGs did not differ between the groups (Supplementary Figure 1C and 1D). HE and Oil Red O staining of the liver sections showed no differences (Supplementary Figure 1E). Therefore, we hypothesized that the hepatic-specific overexpression of VDR in an aging mouse model had no significant impact on hepatic steatosis.

VD supplementation had no significant effect on hepatic steatosis in old mice or Terc^-/- mice

Hepatic-specific overexpression of VDR had no obvious effect on hepatic steatosis in aging mice. Therefore, we fed aged mice an SCD supplemented with VD (SCD + VD). Fifteen-month-old WT mice were fed the SCD or SCD + VD for 14 weeks. Their body weights were recorded, and no differences were observed between the groups (Supplementary Figure 2A). We measured the VDR protein level, and it was obviously increased in the group fed with SCD + VD (Supplementary Figure 2B), indicating that dietary supplementation with VD could increase the expression of VDR in the liver. However, the liver weights; liver-to-body weight ratios; serum levels of ALT, AST, TGs, and TC; and hepatic levels of TG did not differ between the groups (Supplementary Figure 2C). HE and Oil Red O staining of the liver sections revealed no differences (Supplementary Figure 2D). Second, 3-month-old Terc^-/- mice were fed the SCD or SCD + VD for 14 weeks. The results were consistent with those from the old WT mice described above (Supplementary Figure 2E-H). We therefore postulated that increased hepatic VDR expression caused by dietary VD supplementation had no effect on hepatic steatosis.

DISCUSSION

This study investigated the role of VDR in the pathogenesis of aging-related MASLD. We found that aging can increase hepatic steatosis at the cellular and animal levels. VDR expression in the livers of aged mice was elevated compared to that in young mice, which might be related to the increased expression of P53, a classic marker of senescence. Knockdown of VDR expression in senescent model liver cell lines alleviated PA-induced hepatocyte steatosis. Hepatic-specific knockout of VDR alleviated hepatic steatosis in aged mice. The increase in VDR expression induced by VD aggravated PA-induced hepatocyte steatosis in senescent liver cells. However, Dietary VD supplementation in aged mice did not significantly impact hepatic steatosis. The impact of aging-related MASLD via the VDR may be associated with LIPIN1, which can affect the lipid oxidation process[41-43].

MASLD is a prevalent condition globally. The incidence of MASLD in the elderly population has significantly increased to more than 30%[17]. Aging is closely linked to the onset and progression of MASLD, and the accumulation of senescent cells in the liver drives hepatic fat accumulation and steatosis[12]. Aging and MASLD can affect each other. Aging may lead to hepatocyte metabolic disorders, a local inflammatory response, and a decreased regenerative capacity[9,12,44]. MASLD has also been shown to promote cellular senescence[12,20]. However, the specific mechanism of mutual regulation between senescence and MASLD awaits further investigation.

In this study, we found that aging can promote the development of MASLD. We established cellular senescence models and observed that more severe lipid deposition occurred in senescent cells under the same MASLD modeling conditions. We also examined the aging phenotype of Terc^-/- mice and used it as an aging model. We found that hepatic steatosis, abnormal glucose tolerance, and IR were more severe after MASLD modeling, which aligned with the findings form the cell-based experiments. Therefore, the factors that can affect aging-related MASLD were further explored.

Upon investigating the connection between aging and MASLD, we observed that the expression of VDR in the liver was markedly elevated in old mice compared to young mice. Moreover, the expression of P53 and VDR exhibited the same trend. P53 is a classic aging marker, and some studies have described a possible interaction between P53 and VDR[45]. We further examined the relationship between aging and VDR by administering Nutlin-3a, a P53 agonist, to both normal and senescent liver cell lines. Nutlin-3a treatment resulted in elevated mRNA levels of both P53 and VDR. Based on these findings, we assumed that aging could induce increased VDR expression in the liver, potentially through mechanisms involving increased P53 expression.

Under normal physiological conditions, VDR expression in hepatocytes is minimal, but is still detectable in rodent and human hepatocytes[46]. As a nuclear receptor, VDR transcriptionally regulates the expression of various genes, thereby exerting significant influence on a variety of physiological processes other than its physiological effects on the routine regulation of calcium and phosphorus metabolism[47]. The VDR is involved in regulating several hepatic functions, such as bile acid synthesis and lipid metabolism[48]. Previous studies have shown that VDR expression is significantly elevated in the livers of individuals with MASLD and in mice with diet-induced obesity[34]. In addition, a research has demonstrated that compared with WT mice, VDR knockout mice exhibit markedly reduced hepatic steatosis[34]. Other studies have shown that in hepatic-specific VDR-knockout mice, high-fat models of hepatic steatosis are significantly aggravated[35], suggesting that VDR is associated with the development and progression of MASLD. Moreover, aging has been shown to impact VD metabolism[40]. Whether VDR affects aging-related MASLD has not been sufficiently studied.

Considering that both aging and VDR may be associated with the occurrence of MASLD, and aging could induce increased VDR expression in the liver, we hypothesized that VDR might be involved in aging-related MASLD. Therefore, we further investigated the role of VDR in the pathogenesis of aging-related MASLD.

At the cellular level, we knocked down VDR expression in the AML12 and HepG2 cell lines, followed by MASLD modeling. These results suggested that VDR knockdown in senescent hepatocytes attenuated cellular steatosis. For the animal experiments, we selected naturally aged hepatic-specific VDR-knockout mice (VDR^Cre/+) and control mice (VDR^flox/flox) without any model, and found that hepatic steatosis and impaired glucose tolerance were decreased. For further testing, Terc^-/- mice were used as an aging model, were injected with AAV8-shVDR, and were fed an HFD. The results showed that hepatic steatosis and IR were decreased in Terc^-/- mice with hepatic-specific VDR knockdown. The expression of FASN, a protein involved in lipid synthesis, was reduced, while the expression of LIPIN1, a protein associated with lipid oxidation, was increased. The results of the cell-based experiments and two animal experiments were consistent, revealing that VDR downregulation in hepatocytes had a protective effect on aging-related MASLD and ameliorated abnormal glucose metabolism by increasing lipid oxidation. This finding contrasts with that of Zhang et al[35], who reported that the hepatic-specific knockout of VDR exacerbated MASLD at a young age, whereas it was alleviated at an advanced age, indicating that the effect of hepatic VDR on MASLD varies with age. This finding provides a new perspective for future research on VDR.

In recent years, numerous studies have highlighted the connection between VD and MASLD. Clinical investigations have demonstrated that VD deficiency is associated with MASLD[49-51]. Moreover, patients with MASLD confirmed by liver biopsy exhibit lower VD levels compared to healthy controls[52]. Although many relevant epidemiological studies exist, the specific mechanism remains unclear.

Because our results suggested that VDR knockdown in aging mice and aging liver cell lines may attenuate steatosis in MASLD models, we attempted to determine whether the overexpression of VDR in the liver and dietary supplementation with VD affected aging-related MASLD. At the cellular level, we used VD to stimulate the expression of VDR in the AML12 and HepG2 cell lines, which were used to model senescence, followed by MASLD modeling. The results suggested that VD treatment of aging hepatocytes alleviated cellular steatosis and that the expression of the lipid oxidation-associated protein LIPIN1 was increased.

When Terc^-/- mice were used as an aging model and injected with AAV8-VDR, no significant difference in hepatic steatosis was observed between Terc^-/- mice with hepatic-specific VDR overexpression and control mice. A VD-supplemented diet was subsequently administered to 15-month-old WT mice and Terc^-/- mice. No significant difference in hepatic steatosis was observed between the experimental and control groups. Integrating the findings from both cellular and animal studies, we proposed that the increased VDR expression stimulated by VD aggravated aging-related MASLD at the cellular level but not at the animal level. This difference may be related to the fact that the WT mice we selected were not old enough and that the aging phenotype of Terc^-/- mice may not be able to replicate the true aging environment. Therefore, more studies are warranted to explore the effect of VD supplementation on elderly patients with MASLD in the future.

CONCLUSION

Our findings showed that aging can lead to increased hepatic steatosis and upregulate VDR expression in the liver. VDR downregulation in hepatocytes has an important effect on aging-related MASLD, which could alleviate hepatic steatosis and ameliorate abnormal glucose metabolism by increasing lipid oxidation. VDR may be a therapeutic target for aging-related MASLD.