DNA AND CELL BIOLOGY Volume 33, Number 8, 2014 ª Mary Ann Liebert, Inc. Pp. 514–521 DOI: 10.1089/dna.2013.2290
FGF21 Increases Cholesterol Efflux by Upregulating ABCA1 Through the ERK1/2–PPARg–LXRa Pathway in THP1 Macrophage-Derived Foam Cells Xiao-Long Lin,1,2,* Xing-Lan He,1,* Jun-Fa Zeng,3 Hai Zhang,1 Yue Zhao,4 Jian-Kai Tan,4 and Zuo Wang1
FGF21, a member of the fibroblast growth factor superfamily, is an important endogenous regulator of systemic glucose and lipid metabolism. Elevated serum FGF21 levels have been reported in subjects with coronary heart disease and carotid artery plaques. However, whether FGF21 is associated with atherosclerotic diseases remains unclear. In this study, the effects of FGF21 on cholesterol efflux in THP1 macrophage-derived foam cells and the underlying mechanisms were investigated. THP1 macrophage-derived foam cells were incubated with 0, 25, 50, 100, 200, and 400 ng/mL of FGF21 for varying time periods (0, 6, 12, and 24 h). Cholesterol efflux onto apoA-1 was assessed by high-performance liquid chromatography assays, while change in ABCA1 expression was analyzed by western blot and real-time quantitative PCR. Incubation was performed with the ERK1/2specific inhibitor PD98059, PPARg-specific inhibitor GW9662, and LXRa siRNA. Our results show that FGF21 promotes cholesterol efflux and ABCA1 expression in THP1 macrophage-derived foam cells in a doseand time-dependent manner. In addition, inhibition of ERK1/2 or PPARg, or knockdown of LXRa attenuated FGF21-mediated promotion of ABCA1 expression and cholesterol efflux. These results demonstrate that FGF21 can promote cholesterol efflux by upregulating ABCA1 through the ERK1/2–PPARg–LXRa pathway in THP1 macrophage-derived foam cells.
Introduction
T
he fibroblast growth factor (FGF) superfamily has a wide range of biological functions. Among members of the endocrine subfamily, FGF19 and FGF21 possess distinct regulatory roles in glucose and lipid homeostasis, while FGF23 is an important regulator of phosphate and vitamin D metabolism (Cuevas-Ramos et al., 2012; Ding et al., 2012; Woo et al., 2013). Several studies have analyzed the role of FGF21 in glucose and lipid metabolism (Habegger et al., 2013; Li et al., 2013). Human recombinant FGF21 has been demonstrated to stimulate glucose incorporation in mouse and human adipocytes, and to lower blood glucose and triglyceride levels when administered to diabetic and obese mice as well as diabetic monkeys (Kharitonenkov et al., 2007). By contrast, FGF21-deficient mice showed mild weight gain, slightly impaired glucose homeostasis, and also developed hepatosteatosis and marked impairments in ketogenesis and glucose control when raised on a ketogenic diet (Badman
et al., 2009). These findings suggest that FGF21 is an important metabolic hormone in glucose and lipid homeostasis. A recent study has reported that serum FGF21 levels are increased in coronary heart disease (CHD) and that FGF21 is found in carotid artery plaques (Lin et al., 2010; An et al., 2012). Based on these results, FGF21 has been proposed to be associated with arteriosclerosis. However, the role of FGF21 in arteriosclerosis remains unclear. Atherosclerosis and cardiovascular diseases have emerged as serious health problems in the past few decades. A key part of atherosclerosis is failure of macrophages to restore their cellular cholesterol homeostasis and formation of foam cells (Ross, 1999). Reverse cholesterol transport (RCT) is important in decreasing the accumulation of lipids in foam cells along the arterial wall and preventing development of atherosclerosis. Cholesterol efflux from foam cells occurs through several pathways such as unmediated aqueous diffusion and specific receptor-mediated processes (Li et al., 2013; Voloshyna et al., 2013). Several studies have identified ABCA1 as a key player in the regulation of cholesterol
1
Key Laboratory for Arteriosclerology of Hunan Province, Institute of Cardiovascular Disease, University of South China, Hengyang, China. Pathology Department, The Third People’s Hospital of Huizhou, Guangdong Huizhou, China. 3 The Second Affiliated Hospital of the University of South China, Hengyang, China. 4 The First Affiliated Hospital of the University of South China, Hengyang, China. *Co-first authors. 2
514
FGF21 INCREASES CHOLESTEROL EFFLUX
efflux (Smith et al., 2004; Yokoyama, 2006; Liu et al., 2012). In addition, ABCA1 plays a critical role in cholesterol and HDL metabolism as well as macrophage RCT. Thus, the identification of molecules capable of increasing ABCA1 expression may be beneficial for the prevention of atherosclerosis. In the present study, we sought to explore the role of FGF21 in cholesterol efflux from THP1 (human monocytic THP1 cells) macrophage-derived foam cells to apoA-1 and the underlying mechanisms. FGF21 has been demonstrated to regulate lipid metabolism in adipocytes and hepatocytes (Kosola et al., 2012; Goetz, 2013). Several studies have shown that ERK1/2 signaling is the key pathway by which FGF21 regulates glycolipid metabolism through FGFRs. Unlike other receptors, the FGF receptor FGFR1 preferentially binds to FGF21 (Fisher et al., 2011; Iglesias et al., 2012). Several studies have shown that FGFR1 is highly expressed in intima foam cells (Hughes, 1997), suggesting that FGF21 activates ERK1/2 in macrophage-derived foam cells. PPAR–LXRa signaling is the main pathway regulating ABCA1, and evidence exists for PPARg agonist-mediated ABCA1 induction as well (Nakaya et al., 2011; Ozasa et al., 2011). ERK1/2 also activates PPARg, which is involved in cholesterol metabolism (Mulay et al., 2013; Tian et al., 2013). Based on these facts, we hypothesized that ERK1/2– PPARg–LXRa is activated by FGF21, which results in upregulation of ABCA1 expression and promotion of cholesterol efflux in macrophages. Since FGF21 is a metabolic hormone that regulates lipid homeostasis, we further speculated that FGF21 can act as a protective factor regulating lipid homeostasis in plaques. In the present study, we attempted to explore the FGF21 cholesterol efflux of the THP1 macrophage-derived foam cell to apoA-1 and its mechanism. Materials and Methods Materials
FGF21 was purchased from Sigma (St. Louis, MO). The TRIzol reagent (Invitrogen, Carlsbad, CA), BCA protein assay reagent (Pierce Chemical, Rockford, IL), ReverAidTM first strand cDNA synthesis kit (Fermentas, Burlington, Canada), rabbit monoclonal anti-ABCA1, anti-b-actin and rabbit polyclonal anti-PPARg, anti-phosphoPPAPg (anti-p-PPARg), anti-ERK1/2, anti-phospho ERK1/2 (anti-p-ERK1/2), and anti-LXRa were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The horseradish peroxidase-conjugated goat anti-rabbit polyclonal antibody was purchased from CWBIO (Beijing, China). All other reagents were obtained in the best grade available from commercial sources. Cell culture
Human THP1 cells were seeded on six-well plates at 1.0 · 106 cells/well in the RPMI1640 medium (Sigma) [with 10% fetal bovine serum (Invitrogen), 20 mg/mL penicillin, and 20 mg/mL streptomycin] and maintained at 37C in a humidified atmosphere of 5% CO2. The cells were differentiated into macrophages by adding 100 ng/mL phorbol 12-myristate-13acetate for 24 h, and the medium was then replaced with that containing ox-LDL (50 mg/mL) for 2 days to obtain fully differentiated foam cells before use in experiments.
515 Preparation of ox-LDL
The native low-density lipoprotein (LDL) was obtained from Sigma. LDL was oxidized with CuSO4 at 37C for 18 h and transferred into ethylene diamine tetraacetic acid (EDTA; 200 mM) in phosphate-buffered saline (PBS) for 24 h at 4C to remove Cu2 + . Subsequently, the product was dialyzed in PBS for 24 h at 4C to remove EDTA. LDL oxidation was confirmed by thiobarbituric acid reaction substances with malondialdehyde as the standard. The content of ox-LDL was 1.12 – 0.056 compared with 0.30 – 0.067 nmol/100 mg protein in the native LDL preparation ( p < 0.01). The ox-LDL was then sterilized by filtration and stored at 4C as previously described (Hu et al., 2010). RNA isolation and real-time quantitative PCR
Total RNA was extracted from macrophages using the Trizol reagent following the manufacturer’s instructions (Invitrogen). First-strand cDNA synthesis was done with 1 mg total RNA using a TaqMan Reverse Transcription Reagent Kit (Applied Biosystems, Grand Island, NY). Real-time quantitative PCR was performed with 2 mg cDNA using SYBR Green detection chemistry on a Roche LightCycler real-time PCR system (Roche Diagnostics, Mannheim, Germany). The primers used to amplify specific gene products were as follows: b-actin sense 5¢-GTGGGGCGCCCCAGGCACCA-3¢, antisense 5¢-CTCCTTAATGTCACG CACGATTC-3¢; human ABCA1 sense 5¢-TCC AGG CCA GTA CGG AAT TC3¢, antisense 5¢-ACT TTC CTC GCC AAA CCA GTA G-3¢. All data were evaluated with the Roche Light Cycler Run 5.32 software (Roche Diagnostics). Melt curve analyses of all realtime PCR products were performed as shown to ensure production of a single DNA duplex. Relative expression levels were determined by normalizing the expression of each gene to that of the b-actin gene by the 2 - DDCt method. High-performance liquid chromatography assays
High-performance liquid chromatography (HPLC) analysis was conducted as described previously (Hu et al., 2010). In short, cells were washed thrice with PBS. The appropriate volume (usually 1 mL) of 0.5% NaCl was added to *50–200 mg cellular proteins per mL. The cells were sonicated using an ultrasonic processor for 2 min. The protein concentration in the cell solution was measured using a BCA kit. A 0.1-mL aliquot of the cell solution (containing 5–20 mg protein) was used to measure free cholesterol, and a different aliquot was used to measure total cholesterol. Free cholesterol was dissolved in isopropanol (1 mg cholesterol/ mL) and stored at - 20C as a stock solution. A cholesterol standard calibration solution ranging from 0 to 40 mg cholesterol/mL was obtained by diluting the cholesterol stock solution in the same cell lysed buffer. Then, 0.1 mL of each sample (cholesterol standard calibration solutions or cell solutions) was supplemented with 10 mL of reaction mixture, including 500 mM MgCl2, 500 mM Trise HCl (pH 7.4), 10 mM dithiothreitol, and 5% NaCl. A total of 0.4 U cholesterol oxidases in 10 mL 0.5% NaCl were added to each tube for free cholesterol determination, while 0.4 U cholesterol oxidase and 0.4 U of cholesterol esterase were added to each tube for total cholesterol measurement. The total
516
reaction solution in each tube was incubated at 37C for 30 min, and 100 mL of methanol: ethanol mixture (1:1) was then added to stop the reaction. Each solution was kept cold for 30 min to allow protein precipitation and then centrifuged at 1500 rpm for 10 min at 15C. Then, 10 mL of supernatant was applied onto a chromatograph system (PerkinElmer, Inc., Waltham, MA), which consisted of a PerkinElmer series 200 vacuum degasser, pump, PerkinElmer series 600 LINK, PerkinElmer series 200 UV/vis detector, and a Discovery C-18 HLPC column (Supelco, Inc., Pennsylvania, PA). The column was eluted using an isopropanol:n-heptane:acetonitrile mixture (35:13:52) at a flow rate of 1 mL/min for 8 min. Absorbance at 216 nm was monitored. Data were analyzed using Total Chrom software (PerkinElmer, Inc.). Western blot analysis
Cells were harvested in 1 mM phenyl methyl sulfonyl fluoride (94:6), and then placed on ice for 20 min. After centrifugation at 10,000 rpm for 10 min at 4C, the supernatant was collected. The protein concentration was determined with the Hyclone–Pierce’s protein assay kit. Proteins were separated in sodium dodecyl sulfate–polyacrylamide electrophoresis gels (8%) and transferred to a polyvinylidene difluoride membrane. The membrane was immunoblotted first with antibodies to ABCA1 (1:300 dilution), PPARg (1:300 dilution), LXRa (1:400 dilution), and b-actin (1:500 dilution) at 4C overnight, and then with HRP-conjugated secondary antibodies (goat anti-rabbit). Visualization was done using the Enhanced Chemiluminescence detection system from Amersham Biosciences (Piscataway, NJ). Cellular cholesterol efflux experiments
The cells were cultured as described above and then labeled with 0.2 mCi/mL of [3H] cholesterol for 24 h. After 24 h, the cells were washed with fresh media, and FGF21 was added. The cells were washed thrice with PBS and incubated overnight in the RPMI 1640 medium containing 0.1% (w/v) BSA to allow equilibration of [3H] cholesterol in all cellular pools. Equilibrated [3H] cholesterol-labeled cells were washed with PBS and incubated in 2 mL of the efflux medium containing the RPMI 1640 medium and 0.1% BSA with 25 mg/mL human plasma apoA-1. A 150-mL sample of the efflux medium was obtained at designated times and passed through a 0.45-mm filter to remove any floating cells. The cell monolayer was washed with PBS and extracted with 0.15 M NaOH to determine cellular total [3H] radioactivity. Medium and cell-associated [3H] cholesterol were then measured by liquid scintillation counting. Percent efflux was calculated using the following equation: [total media counts/(total cellular counts + total media counts)] · 100% (Liu et al., 2013). Transfection of small interfering RNA
Small interfering RNA (siRNA) targeting LXRa was purchased from Santa Cruz Biotechnology. A control siRNA specific for the red fluorescent protein (CCACTAC CTGAGCACCCAG) was used as negative control. The cells were cultured without antibiotics and serum for 6 h at 50% confluence. Different concentrations of RNA interfer-
LIN ET AL.
ence reagent (A) and RNA transfection reagent (B) (i.e., Lipofectamine 2000 from Invitrogen) were diluted. Before transfection, A was mixed with B and incubated for 30 min. The cells were washed thrice with PBS after 6 h and then cultured in 30% serum DMEM. Statistical analysis
All results are expressed as the mean – SD from three independent experiments, and data were analyzed by the one-way ANOVA and Student’s t-test, using the SPSS 13.0 software. p < 0.05 was considered statistically significant. All data are expressed as mean – SD. Results FGF21 increased the expression of ABCA1 and cholesterol efflux, and activated ERK1/2 in THP1 macrophage-derived foam cell
As a key player in RCT, ABCA1 is critical for regulating cellular cholesterol homeostasis. In this study, we first examined the effect of FGF21 on ABCA1 expression in THP1 macrophage-derived foam cells by real-time quantitative PCR and western blot assays. As shown in Figure 1, FGF21 obviously upregulated ABCA1 expression at both mRNA and protein levels in a dose- and timedependent manner. Having confirmed that ABCA1 was upregulated by FGF21, we next examined the effect of FGF21 on apoA-1-specific cholesterol efflux in the THP1 macrophage-derived foam cell by HPLC and liquid scintillation counting assays. The cell cholesterol content was decreased (Tables 1 and 2), and the cholesterol to apoA-1 ratio was increased by FGF21 (Fig. 2C, D). These results indicate that FGF21 increases apoA-1-specific cholesterol efflux and upregulates ABCA1 expression in THP1 macrophage-derived foam cells. We then analyzed ERK1/2 and p-ERK1/2 levels in THP1 macrophage-derived foam cells under different FGF21 concentrations and incubation times by western blot analysis to investigate the effect of FGF21 on ERK1/2 activation. Our results show that ERK1/2 did not change with changes in time and concentration of FGF21 incubation, whereas p-ERK1/2 increased at 6 h and 100 ng/mL, and this increase was maintained until 24 h and 400 ng/mL (Fig. 2A, B). These data suggest that FGF21 can activate ERK1/2 in THP1 macrophage-derived foam cells in a dose- and timedependent manner. Effects of FGF21 on PPAR c and LXR a expression in THP1 macrophage-derived foam cell
The transcriptional cascade in the PPARg and LXRa pathways plays a critical role in maintaining cellular cholesterol homeostasis in macrophages. Previous studies have shown that PPARg activation upregulates LXRa expression, while PPARg and/or LXRa activation upregulates ABCA1 expression and facilitates cellular cholesterol efflux. To investigate whether LXRa and PPARg expression may be affected by FGF21, we used real-time quantitative PCR and western blot analyses. As shown in Figure 3A, LXRa mRNA and protein expression were both upregulated when the cells were treated with FGF21 in a dose-dependent (0, 25, 50, 100, 200, and 400 ng/mL) and time-dependent (BSA,
FGF21 INCREASES CHOLESTEROL EFFLUX
517
FIG. 1. Effect of fibroblast growth factor (FGF)21 on ABCA1 expression in macrophage-derived foam cells under different FGF21 concentrations and incubation times. (A, B) Macrophagederived foam cells were treated with 10 mg/mL apoA-1 and FGF21 at 0, 25, 50, 100, 200, and 400 ng/mL. ABCA1 mRNA and protein were measured by real-time quantitative PCR and western blot. (C, D) Macrophage-derived foam cells were treated with 10 mg/mL apoA-1 and 5 mg/ mL BSA for 24 h or with 10 mg/mL apoA-1 and 200 ng/mL FGF21 at 0, 6, 12, and 24 h. (A, B) ABCA1 mRNA and protein were measured by real-time quantitative PCR and western blot. ERK1/2 and p-ERK1/2 were measured by western blot. (D) Cellular cholesterol efflux was analyzed by liquid scintillation counting as shown above. *p < 0.05 versus 0 ng/mL.
0, 6, 12, and 24 h) manner, while PPARg protein and mRNA levels did not change. However, phosphorylation of PPARg increased significantly (Fig. 3B). Furthermore, GW9662 (10 mM), a PPARg antagonist, was able to reverse the upregulation of LXRa mRNA and protein expression (Fig. 3C). These results indicate that FGF21 activates PPARg and promotes LXRa expression in THP1 macrophage-derived foam cells. We then examined the effect of LXRa siRNA and/or GW9662 (PPARg antagonist) (10 mM) (Liu et al., 2013) on FGF21-induced change in ABCA1 expression. Upregulation of ABCA1 expression by FGF21 was reversed by both LXRa siRNA and GW9662. Moreover, the LXRa protein and mRNA levels decreased when treated with GW9662 (Fig. 3D). This indicates that FGF21 promotes ABCA1 upregulation through the PPARg–LXRa pathway. We also examined cholesterol efflux in cells using the liquid scintillation counting assay. Our results showed that treatment
with both LXRa siRNA and FGF21 (GW9662 and FGF21) blocked FGF21-mediated upregulation of cholesterol efflux compared with the controls (Fig. 3E). ERK1/2 is involved in the FGF21-activated PPAR c–LXR a pathway
To further confirm whether ERK1/2 is involved in the activation of PPARg and promotion of LXRa expression by FGF21, we treated THP1 macrophage-derived foam cells with PD98059, a p-ERK1/2-specific inhibitor. As shown in Figure 4A, treatment with PD98059 led to the suppression of p-ERK1/2 expression by 77%. Compared with FGF21 + PD98059, PD98059 significantly suppressed FGF21-induced upregulation of p-PPARg expression as well as LXRa mRNA and protein expression (Fig. 4B). These results suggest that activation of the PPARg–LXRa pathway by FGF21 is mediated through ERK1/2.
Table 1. Effect of FGF21 on Total Cholesterol, Free Cholesterol, and Cholesterol Ester at Different Concentrations in THP1 Macrophage-Derived Foam Cells FGF21
Control
25 ng/mL
50 ng/mL
100 ng/mL
200 ng/mL
400 ng/mL
Total cholesterol (mg/dL) Free cholesterol (mg/dL) Cholesterol ester (mg/dL) Cholesterol ester/total cholesterol (%)
492 – 36 196 – 22 296 – 28 60.2
487 – 34 197 – 22 290 – 30 59.5
479 – 2 196 – 24 283 – 27 59.1
398 – 25 167 – 21 231 – 26 58
298 – 29* 126 – 25* 172 – 32* 57.7
295 – 20* 125 – 25* 170 – 25* 57.6
The results are expressed as the mean – SD from three independent experiments, each performed in triplicate. *p < 0.05 versus control. FGF, fibroblast growth factor.
518
LIN ET AL.
Table 2. Effect of FGF21 on Total Cholesterol, Free Cholesterol, and Cholesterol Ester at Different Times in THP1 Macrophage-Derived Foam Cells Time Total cholesterol (mg/dL) Free cholesterol (mg/dL) Cholesterol ester (mg/dL) Cholesterol ester/total cholesterol (%)
BSA (24 h)
0h
6h
12 h
24 h
48 h
506 – 36 196 – 22 310 – 28 61.3
503 – 34 197 – 22 306 – 30 60.8
490 – 26 196 – 24 294 – 27 60
408 – 25 166 – 21 242 – 26 59.1
304 – 29* 125 – 25* 179 – 32* 58.5
299 – 20* 126 – 25* 173 – 25* 58.1
The results are expressed as the mean – SD from three independent experiments, each performed in triplicate. *p < 0.05 versus 0 h.
Blockade of ERK1/2 signaling inhibited FGF21-mediated upregulation of ABCA1 and reduced cholesterol efflux
PPARg–LXRa is the main pathway that controls ABCA1 expression and cholesterol efflux. As described above, our results showed that activation of the PPARg– LXRa pathway by FGF21 is mediated through ERK1/2. Given this finding, we speculated that preventing phosphorylation of ERK1/2 may inhibit ABCA1 expression and reduce cholesterol efflux. Cells treated with 200 ng/mL of FGF21 for 24 h exhibited strong upregulation of ABCA1 expression and cellular cholesterol efflux. By contrast, FGF21-induced upregulation of ABCA1 was blocked by PD98059, an inhibitor of ERK1/2 phosphorylation (Fig. 4C, D). These results indicate that FGF21 increases cholesterol efflux and upregulates ABCA1 expression in an ERK1/2-dependent manner.
FIG. 2. Effect of FGF21 on ERK1/2 activation and cholesterol efflux in macrophagederived foam cells under different concentrations and incubation times. (A, B) ERK1/2 and p-ERK1/2 were measured by western blot. (C, D) Cellular cholesterol efflux was analyzed by liquid scintillation counting assay as shown above. *p < 0.05 versus 0 ng/mL.
Discussion
FGF21, a member of a hormone-like subgroup within the FGF superfamily, is emerging as a key regulator of energy homeostasis and a novel target for the development of therapies for treatment of diabetes, cardiovascular disease, and obesity (An et al., 2012; Jung et al., 2012; Habegger et al., 2013). A previous study showed that incubation of rodent cardiac microvascular endothelial cells (CMECs) incubated with oxidized LDL led to upregulation of FGF21 mRNA and protein expression and inhibited CMEC apoptosis (Kosinski et al., 2012). This supports the hypothesis that FGF21 functions as an endogenous protective factor in the cardiovascular system that can improve endothelial function during early stages of atherosclerosis. Moreover, elevated serum FGF21 levels have recently been reported in patients with CHD and are associated with the presence of carotid artery plaques (Lin et al., 2010; An et al., 2012).
FGF21 INCREASES CHOLESTEROL EFFLUX
519
FIG. 3. Effects of FGF21 on PPARg and LXRa expression in THP1 macrophage-derived foam cells and effects of FGF21 on ABCA1 and cholesterol efflux in THP1 macrophage-derived foam cells during inhibition of PPARg activation and LXRa expression. (A, B) Macrophage-derived foam cells were treated with 10 mg/mL apoA-1 and with FGF21 at 0, 25, 50, 100, 200, and 400 ng/mL; macrophage-derived foam cells were treated with 25 mg/mL apoA-1 and 5 mg/mL BSA for 24 h or with 10 mg/mL apoA-1 and 200 ng/mL FGF21 at 0, 6, 12, and 24 h. *p < 0.05 versus vehicle. (C) Effects of FGF21 on LXRa expression in THP1 macrophage-derived foam cell when PPARg activation was inhibited. Macrophage-derived foam cells were divided into three groups [control, FGF21, FGF21 + GW9662 (10 mM GW9662)] and cultured in a medium at 37C for 24 h. Real-time quantitative PCR and western blot analysis were conducted to measure LXRa mRNA and protein levels, *p < 0.05 versus FGF21 + GW9662. (D, E) Effects of FGF21 on ABCA1 and cholesterol efflux in THP1 macrophage-derived foam cells when PPARg activation was inhibited. *p < 0.05 versus FGF21 + GW9662, #p < 0.05 versus FGF21 + LXRa siRNA. However, the role of FGF21 in atherosclerosis plaques remains unclear. In this study, we present the first evidence that FGF21 can modulate cholesterol efflux from macrophages to apoA-1. In addition, we show that this effect is mediated through the ERK1/2–PPARg–LXRa pathway. ABCA1 is the main protein that regulates cholesterol efflux from cells (Smith et al., 2004; Yokoyama, 2006; Liu et al., 2012). ABCA1 plays a critical role in cholesterol and HDL metabolism as well as macrophage RCT. In this study, we found that FGF21 promotes ABCA1 expression and mediates apoA-1-dependent cholesterol efflux. PPARg and LXRa are important transcription factors that regulate ABCA1 expression. Several studies have shown that PPARg activation upregulates ABCA1 expression in macrophages; however, this upregulation is inhibited in LXRa knockout macrophages (Lu¨ et al., 2010). This indicates that ABCA1 expression is regulated in an LXRa-dependent manner. Consistent with this, some studies have also shown that an LXRa response element exists in the ABCA1 promoter (Li et al., 2004). Likewise, a PPARg response element exists in
the LXRa promoter (Costet et al., 2000), indicating that PPARg activation regulates LXRa expression. In cultured macrophages, activation of PPARg promotes LXRa and ABCA1 expression and cholesterol efflux. The PPARg– LXRa–ABCA1 axis is a major pathway that induces cholesterol efflux in macrophages. In the present study, we found that FGF21 promoted LXRa expression in a dose- and timedependent manner, and the increase in the LXRa level was equal to that of ABCA1. Moreover, FGF21 did not affect PPARg expression, but activated PPARg in a dose- and timedependent manner. Furthermore, GW9662 (10 mM), a PPARg antagonist, inhibited LXRa and ABCA1 expression in THP1 cells. Knockdown of LXRa by siRNA produced the same result (Fig. 3C, D). These results suggest that PPARg is the main factor linking FGF21 and ABCA1. Taken together, our results support a model in which FGF21 upregulates ABCA1 expression by promoting the PPARg–LXRa pathway. ERK1/2 signaling is involved in regulating PPARg expression in various tissues (Costet et al., 2000; Wagner et al., 2003; Kaplan et al., 2010). PPARg–LXRa signaling is
520
LIN ET AL.
FIG. 4. Effects of FGF21 on PPARg, LXRa, and ABCA1 expression and cholesterol efflux in THP1 macrophage-derived foam cells under blockade of ERK1/2 activation. (A) Effect of FGF21 on phosphorylation of ERK1/2 under blockade of ERK1/2 activation. ERK1/2 and p-ERK1/2 protein levels were measured by western blot. (B) Effect of FGF21 on PPARg and p-PPARg expression under blockade of ERK1/2 activation. PPARg and p-PPARg proteins were measured by western blot. (C) Effect of FGF21 on ABCA1 expression under blockade of ERK1/2 activation. The ABCA1 protein was measured by western blot. (D) Effect of FGF21 on cholesterol efflux under blockade of ERK1/2 activation. Cell cholesterol efflux was analyzed by liquid scintillation counting assay as shown above. *p < 0.05 versus FGF21 + PD98059. the main pathway mediating regulation of ABCA1. Moreover, PPARg agonist-mediated induction of ABCA1 has been shown in a previous study (Nakaya et al., 2011; Ozasa et al., 2011). Furthermore, ERK1/2 signaling-mediated phosphorylation of PPARg regulates cholesterol metabolism (Mulay et al., 2013; Tian et al., 2013). Therefore, ERK1/2– PPARg may be activated by FGF21 and may upregulate ABCA1 expression and cholesterol efflux. In the current study, we found that FGF21 can activate ERK1/2 in a doseand time-dependent manner. This suggests that FGF21 increases cholesterol efflux by upregulating ABCA1 through the ERK1/2-PPARg–LXRa pathway. When the THP1 macrophage-derived foam cell was treated with an ERK1/2 inhibitor, phosphorylation of PPARg was reduced, FGF21mediated upregulation of LXRa expression was reversed, and ABCA1 expression and cholesterol efflux were both reduced. These results indicate that FGF21 can promote cholesterol efflux by upregulating ABCA1 through the ERK1/2–PPARg–LXRa pathway in THP1 macrophagederived foam cells. In summary, we have demonstrated that FGF21 induces upregulation of ABCA1 and increases cellular cholesterol efflux in an ERK1/2–PPARg–LXRa-dependent manner in
THP1 macrophage-derived foam cells. These findings provide evidence for the role of FGF21 in arteriosclerosis and a new direction for atherosclerosis prevention. Acknowledgments
This study was supported by the Innovative Research Team for Science and Technology in Higher Educational Institutions of Hunan Province and Natural Science Foundation of China (No. 81070221) and the Visiting Scholar Foundation of Key Laboratory for Biorheological Science and Technology (Chongqing University) of Ministry of Education (2010) and the construct program of the key discipline in Hunan province. Disclosure Statement
No competing financial interests exist. References
An, S.Y., Lee, M.S., Yi, S.A., et al. (2012). Serum fibroblast growth factor 21 was elevated in subjects with type 2 diabetes mellitus and was associated with the presence of carotid artery plaques. Diabetes Res Clin Pract 96, 196–203.
FGF21 INCREASES CHOLESTEROL EFFLUX
Badman, M.K., Koester, A., Flier, J.S., Kharitonenkov, A., and Maratos-Flier, E. (2009). Fibroblast growth factor 21deficient mice demonstrate impaired adaptation to ketosis. Endocrinology 150, 4931–4940. Costet, P., Luo, Y., Wang, N., Tall, A.R. (2000). Steroldependent transactivation of the abc1 promoter by the liver x receptor/retinoid x receptor. J Biol Chem 275, 28240–28245. Cuevas-Ramos, D., Aguilar-Salinas, C.A., and Go´mez-Pe´rez, F.J. (2012). Metabolic actions of fibroblast growth factor. Curr Opin Pediatr 2124, 523–529. Ding, X., Boney-Montoya, J., Owen, B.M., Bookout, A.L., Coate, K.C., Mangelsdorf, D.J., and Kliewer, S.A. (2012). Cell Metab 16, 387–393. Fisher, F.M., Estall, J.L., Adams, A.C., et al. (2011). Integrated regulation of hepatic metabolism by fibroblast growth factor 21 (FGF21) in vivo. Endocrinology 152, 2996–3004. Goetz, R. (2013). Metabolism: adiponectin-a mediator of specific metabolic actions of FGF21. Nat Rev Endocrinol 9, 506– 508. Habegger, K.M., Stemmer, K., Cheng, C., et al. (2013). Fibroblast growth factor 21 mediates specific glucagon actions. Diabetes 62, 1453–1463. Hu, Y.W., Wang, Q., Ma, X., et al. (2010). TGF-beta1 upregulates expression of ABCA1, ABCG1 and SR-BI through liver X receptor alpha signaling pathway in THP-1 macrophagederived foam cells. J Atheroscler Thromb 17, 493–502. Hughes, S.E. (1997). Differential expression of the fibroblast growth factor receptor (FGFR) multigene family in normal human adult tissues. J Histochem Cytochem 45, 1005–1019. Iglesias, P., Selgas, R., Romero, S., and Dı´ez, J.J. (2012). Biological role, clinical significance, and therapeutic possibilities of the recently discovered metabolic hormone fibroblastic growth factor 21. Eur J Endocrinol 167, 301–309. Jung, U.J., Torrejon, C., Chang, C.L., et al. (2012). Fatty acids regulate endothelial lipase and inflammatory markers in macrophages and in mouse aorta: a role for PPARg. Arterioscler Thromb Vasc Biol 32, 2929–2937. Kaplan, J.M., Hake, P.W., Denenberg, A., Nowell, M., Piraino, G., et al. (2010). Phosphorylation of extracellular signalregulated kinase (ERK)-1/2 is associated with the downregulation of peroxisome proliferator–activated receptor (PPAR)-g during polymicrobial sepsis. Mol Med 16, 491–497. Kharitonenkov, A., Wroblewski, V.J., Koester, A., et al. (2007). The metabolic state of diabetic monkeys is regulated by fibroblast growth factor-21. Endocrinology 148, 774–781. Kosinski, J.R., Hubert, J., and Carrington, P.E. (2012). The glucagon receptor is involved in mediating the body weightlowering effects of oxyntomodulin. Obesity (Silver Spring) 20, 1566–1571. Kosola, S., Lampela, H., Gylling, H., et al. (2012). Cholesterol metabolism altered and FGF21 levels high after pediatric liver transplantation despite normal serum lipids. Am J Transplant 12, 2815–2824. Li, A.C., Binder, C.J., Gutierrez, A., et al. (2004). Differential inhibition of macrophage foam-cell formation and atherosclerosis in mice by pparalpha, beta/delta, and gamma. J Clin Invest 114, 1564–1576. Li, G., Gu, H.M., Zhang, D.W., et al. (2013). ATP-binding cassette transporters and cholesterol translocation. IUBMB Life 65, 505–512. Li, H., Zhang, J., and Jia, W. (2013). Fibroblast growth factor 21: a novel metabolic regulator from pharmacology to physiology. Front Med 7, 25–30.
521
Lin, Z., Wu, Z., Yin, X., et al. (2010). Serum levels of FGF-21 are increased in coronary heart disease patientsand are independently associated with adverse lipid profile. PLoS One 5, e15534. Liu, H.F., Cui, K.F., Wang, J.P., et al. (2012). Significance of ABCA1 in human carotid atherosclerotic plaques. Exp Ther Med 4, 297–302. Liu, X.Y., Lu, Q., Ouyang, X.P., et al. (2013). Apelin-13 increases expression of ATP-binding cassette transporter A1 via activating protein kinase C a signaling in THP-1 macrophagederived foam cells. Atherosclerosis 226, 398–407. Lu¨, Y., Liu, J.H., Zhang, L.K., et al. (2010). Fibroblast growth factor 21 as a possible endogenous factor inhibits apoptosis in cardiac endothelial cells. Chin Med J (Engl) 123, 3417–3421. Mulay, V., Wood, P., Manetsch, M., et al. (2013). Inhibition of mitogen-activated protein kinase Erk1/2 promotes protein degradation of ATP binding cassette transporters A1 and G1 in CHO and HuH7 cells. PLoS One 8, e62667. Nakaya, K., Tohyama, J., and Naik, S.U. (2011). Peroxisome proliferator-activated receptor-a activation promotes macrophage reverse cholesterol transport through a liver X receptor-dependent pathway. Arterioscler Thromb Vasc Biol 31, 1276–1282. Ozasa, H., Ayaori, M., Iizuka, M., et al. (2011). Pioglitazone enhances cholesterol efflux from macrophages by increasing ABCA1/ABCG1 expressions via PPARg/LXRa pathway: findings from in vitro and ex vivo studies. Atherosclerosis 219, 141–150. Ross, R. (1999). Atherosclerosis—an inflammatory disease. N Engl J Med 340, 115–126. Smith, J.D., Le Goff, W., Settle, M., Brubaker, G., Waelde, C., Horwitz, A., and Oda, M.N. (2004). ABCA1 mediates concurrent cholesterol and phospholipid efflux to apolipoprotein A-I. J Biol Chem 279, 9963–9969. Tian, C., Ye, X., Zhang, R., et al. (2013). Green tea polyphenols reduced fat deposits in high fat-fed rats via erk1/2-PPARgadiponectin pathway. PLoS One 8, e53796. Voloshyna, I., Hai, O., Littlefield, M.J., Carsons, S., and Reiss, A.B. (2013). Resveratrol mediates anti-atherogenic effects on cholesterol flux in human macrophages and endothelium via PPARg and adenosine. Eur J Pharmacol 698, 299–309. Wagner, B.L., Valledor, A.F., Shao, G., et al. (2003). Promoterspecificroles for liver x receptor/corepressor complexes in the regulationof abca1 and srebp1 gene expression. Mol Cell Biol 23, 5780–5789. Woo, Y.C., Xu, A., Wang, Y., and Lam, K.S. (2013). Fibroblast growth factor 21 as an emerging metabolic regulator: clinical perspectives. Clin Endocrinol (Oxf) 78, 489–496. Yokoyama, S. (2006). ABCA1 and biogenesis of HDL. J Atheroscler Thromb 13, 1–15.
Address correspondence to: Zuo Wang, PhD Key Laboratory for Arteriosclerology of Hunan Province Institute of Cardiovascular Disease University of South China Hengyang 421001 China E-mail: [email protected] Received for publication November 26, 2013; received in revised form March 15, 2014; accepted March 15, 2014.
FGF21 Increases Cholesterol Efflux by Upregulating ABCA1 Through the ERK1/2–PPARg–LXRa Pathway in THP1 Macrophage-Derived Foam Cells Xiao-Long Lin,1,2,* Xing-Lan He,1,* Jun-Fa Zeng,3 Hai Zhang,1 Yue Zhao,4 Jian-Kai Tan,4 and Zuo Wang1
FGF21, a member of the fibroblast growth factor superfamily, is an important endogenous regulator of systemic glucose and lipid metabolism. Elevated serum FGF21 levels have been reported in subjects with coronary heart disease and carotid artery plaques. However, whether FGF21 is associated with atherosclerotic diseases remains unclear. In this study, the effects of FGF21 on cholesterol efflux in THP1 macrophage-derived foam cells and the underlying mechanisms were investigated. THP1 macrophage-derived foam cells were incubated with 0, 25, 50, 100, 200, and 400 ng/mL of FGF21 for varying time periods (0, 6, 12, and 24 h). Cholesterol efflux onto apoA-1 was assessed by high-performance liquid chromatography assays, while change in ABCA1 expression was analyzed by western blot and real-time quantitative PCR. Incubation was performed with the ERK1/2specific inhibitor PD98059, PPARg-specific inhibitor GW9662, and LXRa siRNA. Our results show that FGF21 promotes cholesterol efflux and ABCA1 expression in THP1 macrophage-derived foam cells in a doseand time-dependent manner. In addition, inhibition of ERK1/2 or PPARg, or knockdown of LXRa attenuated FGF21-mediated promotion of ABCA1 expression and cholesterol efflux. These results demonstrate that FGF21 can promote cholesterol efflux by upregulating ABCA1 through the ERK1/2–PPARg–LXRa pathway in THP1 macrophage-derived foam cells.
Introduction
T
he fibroblast growth factor (FGF) superfamily has a wide range of biological functions. Among members of the endocrine subfamily, FGF19 and FGF21 possess distinct regulatory roles in glucose and lipid homeostasis, while FGF23 is an important regulator of phosphate and vitamin D metabolism (Cuevas-Ramos et al., 2012; Ding et al., 2012; Woo et al., 2013). Several studies have analyzed the role of FGF21 in glucose and lipid metabolism (Habegger et al., 2013; Li et al., 2013). Human recombinant FGF21 has been demonstrated to stimulate glucose incorporation in mouse and human adipocytes, and to lower blood glucose and triglyceride levels when administered to diabetic and obese mice as well as diabetic monkeys (Kharitonenkov et al., 2007). By contrast, FGF21-deficient mice showed mild weight gain, slightly impaired glucose homeostasis, and also developed hepatosteatosis and marked impairments in ketogenesis and glucose control when raised on a ketogenic diet (Badman
et al., 2009). These findings suggest that FGF21 is an important metabolic hormone in glucose and lipid homeostasis. A recent study has reported that serum FGF21 levels are increased in coronary heart disease (CHD) and that FGF21 is found in carotid artery plaques (Lin et al., 2010; An et al., 2012). Based on these results, FGF21 has been proposed to be associated with arteriosclerosis. However, the role of FGF21 in arteriosclerosis remains unclear. Atherosclerosis and cardiovascular diseases have emerged as serious health problems in the past few decades. A key part of atherosclerosis is failure of macrophages to restore their cellular cholesterol homeostasis and formation of foam cells (Ross, 1999). Reverse cholesterol transport (RCT) is important in decreasing the accumulation of lipids in foam cells along the arterial wall and preventing development of atherosclerosis. Cholesterol efflux from foam cells occurs through several pathways such as unmediated aqueous diffusion and specific receptor-mediated processes (Li et al., 2013; Voloshyna et al., 2013). Several studies have identified ABCA1 as a key player in the regulation of cholesterol
1
Key Laboratory for Arteriosclerology of Hunan Province, Institute of Cardiovascular Disease, University of South China, Hengyang, China. Pathology Department, The Third People’s Hospital of Huizhou, Guangdong Huizhou, China. 3 The Second Affiliated Hospital of the University of South China, Hengyang, China. 4 The First Affiliated Hospital of the University of South China, Hengyang, China. *Co-first authors. 2
514
FGF21 INCREASES CHOLESTEROL EFFLUX
efflux (Smith et al., 2004; Yokoyama, 2006; Liu et al., 2012). In addition, ABCA1 plays a critical role in cholesterol and HDL metabolism as well as macrophage RCT. Thus, the identification of molecules capable of increasing ABCA1 expression may be beneficial for the prevention of atherosclerosis. In the present study, we sought to explore the role of FGF21 in cholesterol efflux from THP1 (human monocytic THP1 cells) macrophage-derived foam cells to apoA-1 and the underlying mechanisms. FGF21 has been demonstrated to regulate lipid metabolism in adipocytes and hepatocytes (Kosola et al., 2012; Goetz, 2013). Several studies have shown that ERK1/2 signaling is the key pathway by which FGF21 regulates glycolipid metabolism through FGFRs. Unlike other receptors, the FGF receptor FGFR1 preferentially binds to FGF21 (Fisher et al., 2011; Iglesias et al., 2012). Several studies have shown that FGFR1 is highly expressed in intima foam cells (Hughes, 1997), suggesting that FGF21 activates ERK1/2 in macrophage-derived foam cells. PPAR–LXRa signaling is the main pathway regulating ABCA1, and evidence exists for PPARg agonist-mediated ABCA1 induction as well (Nakaya et al., 2011; Ozasa et al., 2011). ERK1/2 also activates PPARg, which is involved in cholesterol metabolism (Mulay et al., 2013; Tian et al., 2013). Based on these facts, we hypothesized that ERK1/2– PPARg–LXRa is activated by FGF21, which results in upregulation of ABCA1 expression and promotion of cholesterol efflux in macrophages. Since FGF21 is a metabolic hormone that regulates lipid homeostasis, we further speculated that FGF21 can act as a protective factor regulating lipid homeostasis in plaques. In the present study, we attempted to explore the FGF21 cholesterol efflux of the THP1 macrophage-derived foam cell to apoA-1 and its mechanism. Materials and Methods Materials
FGF21 was purchased from Sigma (St. Louis, MO). The TRIzol reagent (Invitrogen, Carlsbad, CA), BCA protein assay reagent (Pierce Chemical, Rockford, IL), ReverAidTM first strand cDNA synthesis kit (Fermentas, Burlington, Canada), rabbit monoclonal anti-ABCA1, anti-b-actin and rabbit polyclonal anti-PPARg, anti-phosphoPPAPg (anti-p-PPARg), anti-ERK1/2, anti-phospho ERK1/2 (anti-p-ERK1/2), and anti-LXRa were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The horseradish peroxidase-conjugated goat anti-rabbit polyclonal antibody was purchased from CWBIO (Beijing, China). All other reagents were obtained in the best grade available from commercial sources. Cell culture
Human THP1 cells were seeded on six-well plates at 1.0 · 106 cells/well in the RPMI1640 medium (Sigma) [with 10% fetal bovine serum (Invitrogen), 20 mg/mL penicillin, and 20 mg/mL streptomycin] and maintained at 37C in a humidified atmosphere of 5% CO2. The cells were differentiated into macrophages by adding 100 ng/mL phorbol 12-myristate-13acetate for 24 h, and the medium was then replaced with that containing ox-LDL (50 mg/mL) for 2 days to obtain fully differentiated foam cells before use in experiments.
515 Preparation of ox-LDL
The native low-density lipoprotein (LDL) was obtained from Sigma. LDL was oxidized with CuSO4 at 37C for 18 h and transferred into ethylene diamine tetraacetic acid (EDTA; 200 mM) in phosphate-buffered saline (PBS) for 24 h at 4C to remove Cu2 + . Subsequently, the product was dialyzed in PBS for 24 h at 4C to remove EDTA. LDL oxidation was confirmed by thiobarbituric acid reaction substances with malondialdehyde as the standard. The content of ox-LDL was 1.12 – 0.056 compared with 0.30 – 0.067 nmol/100 mg protein in the native LDL preparation ( p < 0.01). The ox-LDL was then sterilized by filtration and stored at 4C as previously described (Hu et al., 2010). RNA isolation and real-time quantitative PCR
Total RNA was extracted from macrophages using the Trizol reagent following the manufacturer’s instructions (Invitrogen). First-strand cDNA synthesis was done with 1 mg total RNA using a TaqMan Reverse Transcription Reagent Kit (Applied Biosystems, Grand Island, NY). Real-time quantitative PCR was performed with 2 mg cDNA using SYBR Green detection chemistry on a Roche LightCycler real-time PCR system (Roche Diagnostics, Mannheim, Germany). The primers used to amplify specific gene products were as follows: b-actin sense 5¢-GTGGGGCGCCCCAGGCACCA-3¢, antisense 5¢-CTCCTTAATGTCACG CACGATTC-3¢; human ABCA1 sense 5¢-TCC AGG CCA GTA CGG AAT TC3¢, antisense 5¢-ACT TTC CTC GCC AAA CCA GTA G-3¢. All data were evaluated with the Roche Light Cycler Run 5.32 software (Roche Diagnostics). Melt curve analyses of all realtime PCR products were performed as shown to ensure production of a single DNA duplex. Relative expression levels were determined by normalizing the expression of each gene to that of the b-actin gene by the 2 - DDCt method. High-performance liquid chromatography assays
High-performance liquid chromatography (HPLC) analysis was conducted as described previously (Hu et al., 2010). In short, cells were washed thrice with PBS. The appropriate volume (usually 1 mL) of 0.5% NaCl was added to *50–200 mg cellular proteins per mL. The cells were sonicated using an ultrasonic processor for 2 min. The protein concentration in the cell solution was measured using a BCA kit. A 0.1-mL aliquot of the cell solution (containing 5–20 mg protein) was used to measure free cholesterol, and a different aliquot was used to measure total cholesterol. Free cholesterol was dissolved in isopropanol (1 mg cholesterol/ mL) and stored at - 20C as a stock solution. A cholesterol standard calibration solution ranging from 0 to 40 mg cholesterol/mL was obtained by diluting the cholesterol stock solution in the same cell lysed buffer. Then, 0.1 mL of each sample (cholesterol standard calibration solutions or cell solutions) was supplemented with 10 mL of reaction mixture, including 500 mM MgCl2, 500 mM Trise HCl (pH 7.4), 10 mM dithiothreitol, and 5% NaCl. A total of 0.4 U cholesterol oxidases in 10 mL 0.5% NaCl were added to each tube for free cholesterol determination, while 0.4 U cholesterol oxidase and 0.4 U of cholesterol esterase were added to each tube for total cholesterol measurement. The total
516
reaction solution in each tube was incubated at 37C for 30 min, and 100 mL of methanol: ethanol mixture (1:1) was then added to stop the reaction. Each solution was kept cold for 30 min to allow protein precipitation and then centrifuged at 1500 rpm for 10 min at 15C. Then, 10 mL of supernatant was applied onto a chromatograph system (PerkinElmer, Inc., Waltham, MA), which consisted of a PerkinElmer series 200 vacuum degasser, pump, PerkinElmer series 600 LINK, PerkinElmer series 200 UV/vis detector, and a Discovery C-18 HLPC column (Supelco, Inc., Pennsylvania, PA). The column was eluted using an isopropanol:n-heptane:acetonitrile mixture (35:13:52) at a flow rate of 1 mL/min for 8 min. Absorbance at 216 nm was monitored. Data were analyzed using Total Chrom software (PerkinElmer, Inc.). Western blot analysis
Cells were harvested in 1 mM phenyl methyl sulfonyl fluoride (94:6), and then placed on ice for 20 min. After centrifugation at 10,000 rpm for 10 min at 4C, the supernatant was collected. The protein concentration was determined with the Hyclone–Pierce’s protein assay kit. Proteins were separated in sodium dodecyl sulfate–polyacrylamide electrophoresis gels (8%) and transferred to a polyvinylidene difluoride membrane. The membrane was immunoblotted first with antibodies to ABCA1 (1:300 dilution), PPARg (1:300 dilution), LXRa (1:400 dilution), and b-actin (1:500 dilution) at 4C overnight, and then with HRP-conjugated secondary antibodies (goat anti-rabbit). Visualization was done using the Enhanced Chemiluminescence detection system from Amersham Biosciences (Piscataway, NJ). Cellular cholesterol efflux experiments
The cells were cultured as described above and then labeled with 0.2 mCi/mL of [3H] cholesterol for 24 h. After 24 h, the cells were washed with fresh media, and FGF21 was added. The cells were washed thrice with PBS and incubated overnight in the RPMI 1640 medium containing 0.1% (w/v) BSA to allow equilibration of [3H] cholesterol in all cellular pools. Equilibrated [3H] cholesterol-labeled cells were washed with PBS and incubated in 2 mL of the efflux medium containing the RPMI 1640 medium and 0.1% BSA with 25 mg/mL human plasma apoA-1. A 150-mL sample of the efflux medium was obtained at designated times and passed through a 0.45-mm filter to remove any floating cells. The cell monolayer was washed with PBS and extracted with 0.15 M NaOH to determine cellular total [3H] radioactivity. Medium and cell-associated [3H] cholesterol were then measured by liquid scintillation counting. Percent efflux was calculated using the following equation: [total media counts/(total cellular counts + total media counts)] · 100% (Liu et al., 2013). Transfection of small interfering RNA
Small interfering RNA (siRNA) targeting LXRa was purchased from Santa Cruz Biotechnology. A control siRNA specific for the red fluorescent protein (CCACTAC CTGAGCACCCAG) was used as negative control. The cells were cultured without antibiotics and serum for 6 h at 50% confluence. Different concentrations of RNA interfer-
LIN ET AL.
ence reagent (A) and RNA transfection reagent (B) (i.e., Lipofectamine 2000 from Invitrogen) were diluted. Before transfection, A was mixed with B and incubated for 30 min. The cells were washed thrice with PBS after 6 h and then cultured in 30% serum DMEM. Statistical analysis
All results are expressed as the mean – SD from three independent experiments, and data were analyzed by the one-way ANOVA and Student’s t-test, using the SPSS 13.0 software. p < 0.05 was considered statistically significant. All data are expressed as mean – SD. Results FGF21 increased the expression of ABCA1 and cholesterol efflux, and activated ERK1/2 in THP1 macrophage-derived foam cell
As a key player in RCT, ABCA1 is critical for regulating cellular cholesterol homeostasis. In this study, we first examined the effect of FGF21 on ABCA1 expression in THP1 macrophage-derived foam cells by real-time quantitative PCR and western blot assays. As shown in Figure 1, FGF21 obviously upregulated ABCA1 expression at both mRNA and protein levels in a dose- and timedependent manner. Having confirmed that ABCA1 was upregulated by FGF21, we next examined the effect of FGF21 on apoA-1-specific cholesterol efflux in the THP1 macrophage-derived foam cell by HPLC and liquid scintillation counting assays. The cell cholesterol content was decreased (Tables 1 and 2), and the cholesterol to apoA-1 ratio was increased by FGF21 (Fig. 2C, D). These results indicate that FGF21 increases apoA-1-specific cholesterol efflux and upregulates ABCA1 expression in THP1 macrophage-derived foam cells. We then analyzed ERK1/2 and p-ERK1/2 levels in THP1 macrophage-derived foam cells under different FGF21 concentrations and incubation times by western blot analysis to investigate the effect of FGF21 on ERK1/2 activation. Our results show that ERK1/2 did not change with changes in time and concentration of FGF21 incubation, whereas p-ERK1/2 increased at 6 h and 100 ng/mL, and this increase was maintained until 24 h and 400 ng/mL (Fig. 2A, B). These data suggest that FGF21 can activate ERK1/2 in THP1 macrophage-derived foam cells in a dose- and timedependent manner. Effects of FGF21 on PPAR c and LXR a expression in THP1 macrophage-derived foam cell
The transcriptional cascade in the PPARg and LXRa pathways plays a critical role in maintaining cellular cholesterol homeostasis in macrophages. Previous studies have shown that PPARg activation upregulates LXRa expression, while PPARg and/or LXRa activation upregulates ABCA1 expression and facilitates cellular cholesterol efflux. To investigate whether LXRa and PPARg expression may be affected by FGF21, we used real-time quantitative PCR and western blot analyses. As shown in Figure 3A, LXRa mRNA and protein expression were both upregulated when the cells were treated with FGF21 in a dose-dependent (0, 25, 50, 100, 200, and 400 ng/mL) and time-dependent (BSA,
FGF21 INCREASES CHOLESTEROL EFFLUX
517
FIG. 1. Effect of fibroblast growth factor (FGF)21 on ABCA1 expression in macrophage-derived foam cells under different FGF21 concentrations and incubation times. (A, B) Macrophagederived foam cells were treated with 10 mg/mL apoA-1 and FGF21 at 0, 25, 50, 100, 200, and 400 ng/mL. ABCA1 mRNA and protein were measured by real-time quantitative PCR and western blot. (C, D) Macrophage-derived foam cells were treated with 10 mg/mL apoA-1 and 5 mg/ mL BSA for 24 h or with 10 mg/mL apoA-1 and 200 ng/mL FGF21 at 0, 6, 12, and 24 h. (A, B) ABCA1 mRNA and protein were measured by real-time quantitative PCR and western blot. ERK1/2 and p-ERK1/2 were measured by western blot. (D) Cellular cholesterol efflux was analyzed by liquid scintillation counting as shown above. *p < 0.05 versus 0 ng/mL.
0, 6, 12, and 24 h) manner, while PPARg protein and mRNA levels did not change. However, phosphorylation of PPARg increased significantly (Fig. 3B). Furthermore, GW9662 (10 mM), a PPARg antagonist, was able to reverse the upregulation of LXRa mRNA and protein expression (Fig. 3C). These results indicate that FGF21 activates PPARg and promotes LXRa expression in THP1 macrophage-derived foam cells. We then examined the effect of LXRa siRNA and/or GW9662 (PPARg antagonist) (10 mM) (Liu et al., 2013) on FGF21-induced change in ABCA1 expression. Upregulation of ABCA1 expression by FGF21 was reversed by both LXRa siRNA and GW9662. Moreover, the LXRa protein and mRNA levels decreased when treated with GW9662 (Fig. 3D). This indicates that FGF21 promotes ABCA1 upregulation through the PPARg–LXRa pathway. We also examined cholesterol efflux in cells using the liquid scintillation counting assay. Our results showed that treatment
with both LXRa siRNA and FGF21 (GW9662 and FGF21) blocked FGF21-mediated upregulation of cholesterol efflux compared with the controls (Fig. 3E). ERK1/2 is involved in the FGF21-activated PPAR c–LXR a pathway
To further confirm whether ERK1/2 is involved in the activation of PPARg and promotion of LXRa expression by FGF21, we treated THP1 macrophage-derived foam cells with PD98059, a p-ERK1/2-specific inhibitor. As shown in Figure 4A, treatment with PD98059 led to the suppression of p-ERK1/2 expression by 77%. Compared with FGF21 + PD98059, PD98059 significantly suppressed FGF21-induced upregulation of p-PPARg expression as well as LXRa mRNA and protein expression (Fig. 4B). These results suggest that activation of the PPARg–LXRa pathway by FGF21 is mediated through ERK1/2.
Table 1. Effect of FGF21 on Total Cholesterol, Free Cholesterol, and Cholesterol Ester at Different Concentrations in THP1 Macrophage-Derived Foam Cells FGF21
Control
25 ng/mL
50 ng/mL
100 ng/mL
200 ng/mL
400 ng/mL
Total cholesterol (mg/dL) Free cholesterol (mg/dL) Cholesterol ester (mg/dL) Cholesterol ester/total cholesterol (%)
492 – 36 196 – 22 296 – 28 60.2
487 – 34 197 – 22 290 – 30 59.5
479 – 2 196 – 24 283 – 27 59.1
398 – 25 167 – 21 231 – 26 58
298 – 29* 126 – 25* 172 – 32* 57.7
295 – 20* 125 – 25* 170 – 25* 57.6
The results are expressed as the mean – SD from three independent experiments, each performed in triplicate. *p < 0.05 versus control. FGF, fibroblast growth factor.
518
LIN ET AL.
Table 2. Effect of FGF21 on Total Cholesterol, Free Cholesterol, and Cholesterol Ester at Different Times in THP1 Macrophage-Derived Foam Cells Time Total cholesterol (mg/dL) Free cholesterol (mg/dL) Cholesterol ester (mg/dL) Cholesterol ester/total cholesterol (%)
BSA (24 h)
0h
6h
12 h
24 h
48 h
506 – 36 196 – 22 310 – 28 61.3
503 – 34 197 – 22 306 – 30 60.8
490 – 26 196 – 24 294 – 27 60
408 – 25 166 – 21 242 – 26 59.1
304 – 29* 125 – 25* 179 – 32* 58.5
299 – 20* 126 – 25* 173 – 25* 58.1
The results are expressed as the mean – SD from three independent experiments, each performed in triplicate. *p < 0.05 versus 0 h.
Blockade of ERK1/2 signaling inhibited FGF21-mediated upregulation of ABCA1 and reduced cholesterol efflux
PPARg–LXRa is the main pathway that controls ABCA1 expression and cholesterol efflux. As described above, our results showed that activation of the PPARg– LXRa pathway by FGF21 is mediated through ERK1/2. Given this finding, we speculated that preventing phosphorylation of ERK1/2 may inhibit ABCA1 expression and reduce cholesterol efflux. Cells treated with 200 ng/mL of FGF21 for 24 h exhibited strong upregulation of ABCA1 expression and cellular cholesterol efflux. By contrast, FGF21-induced upregulation of ABCA1 was blocked by PD98059, an inhibitor of ERK1/2 phosphorylation (Fig. 4C, D). These results indicate that FGF21 increases cholesterol efflux and upregulates ABCA1 expression in an ERK1/2-dependent manner.
FIG. 2. Effect of FGF21 on ERK1/2 activation and cholesterol efflux in macrophagederived foam cells under different concentrations and incubation times. (A, B) ERK1/2 and p-ERK1/2 were measured by western blot. (C, D) Cellular cholesterol efflux was analyzed by liquid scintillation counting assay as shown above. *p < 0.05 versus 0 ng/mL.
Discussion
FGF21, a member of a hormone-like subgroup within the FGF superfamily, is emerging as a key regulator of energy homeostasis and a novel target for the development of therapies for treatment of diabetes, cardiovascular disease, and obesity (An et al., 2012; Jung et al., 2012; Habegger et al., 2013). A previous study showed that incubation of rodent cardiac microvascular endothelial cells (CMECs) incubated with oxidized LDL led to upregulation of FGF21 mRNA and protein expression and inhibited CMEC apoptosis (Kosinski et al., 2012). This supports the hypothesis that FGF21 functions as an endogenous protective factor in the cardiovascular system that can improve endothelial function during early stages of atherosclerosis. Moreover, elevated serum FGF21 levels have recently been reported in patients with CHD and are associated with the presence of carotid artery plaques (Lin et al., 2010; An et al., 2012).
FGF21 INCREASES CHOLESTEROL EFFLUX
519
FIG. 3. Effects of FGF21 on PPARg and LXRa expression in THP1 macrophage-derived foam cells and effects of FGF21 on ABCA1 and cholesterol efflux in THP1 macrophage-derived foam cells during inhibition of PPARg activation and LXRa expression. (A, B) Macrophage-derived foam cells were treated with 10 mg/mL apoA-1 and with FGF21 at 0, 25, 50, 100, 200, and 400 ng/mL; macrophage-derived foam cells were treated with 25 mg/mL apoA-1 and 5 mg/mL BSA for 24 h or with 10 mg/mL apoA-1 and 200 ng/mL FGF21 at 0, 6, 12, and 24 h. *p < 0.05 versus vehicle. (C) Effects of FGF21 on LXRa expression in THP1 macrophage-derived foam cell when PPARg activation was inhibited. Macrophage-derived foam cells were divided into three groups [control, FGF21, FGF21 + GW9662 (10 mM GW9662)] and cultured in a medium at 37C for 24 h. Real-time quantitative PCR and western blot analysis were conducted to measure LXRa mRNA and protein levels, *p < 0.05 versus FGF21 + GW9662. (D, E) Effects of FGF21 on ABCA1 and cholesterol efflux in THP1 macrophage-derived foam cells when PPARg activation was inhibited. *p < 0.05 versus FGF21 + GW9662, #p < 0.05 versus FGF21 + LXRa siRNA. However, the role of FGF21 in atherosclerosis plaques remains unclear. In this study, we present the first evidence that FGF21 can modulate cholesterol efflux from macrophages to apoA-1. In addition, we show that this effect is mediated through the ERK1/2–PPARg–LXRa pathway. ABCA1 is the main protein that regulates cholesterol efflux from cells (Smith et al., 2004; Yokoyama, 2006; Liu et al., 2012). ABCA1 plays a critical role in cholesterol and HDL metabolism as well as macrophage RCT. In this study, we found that FGF21 promotes ABCA1 expression and mediates apoA-1-dependent cholesterol efflux. PPARg and LXRa are important transcription factors that regulate ABCA1 expression. Several studies have shown that PPARg activation upregulates ABCA1 expression in macrophages; however, this upregulation is inhibited in LXRa knockout macrophages (Lu¨ et al., 2010). This indicates that ABCA1 expression is regulated in an LXRa-dependent manner. Consistent with this, some studies have also shown that an LXRa response element exists in the ABCA1 promoter (Li et al., 2004). Likewise, a PPARg response element exists in
the LXRa promoter (Costet et al., 2000), indicating that PPARg activation regulates LXRa expression. In cultured macrophages, activation of PPARg promotes LXRa and ABCA1 expression and cholesterol efflux. The PPARg– LXRa–ABCA1 axis is a major pathway that induces cholesterol efflux in macrophages. In the present study, we found that FGF21 promoted LXRa expression in a dose- and timedependent manner, and the increase in the LXRa level was equal to that of ABCA1. Moreover, FGF21 did not affect PPARg expression, but activated PPARg in a dose- and timedependent manner. Furthermore, GW9662 (10 mM), a PPARg antagonist, inhibited LXRa and ABCA1 expression in THP1 cells. Knockdown of LXRa by siRNA produced the same result (Fig. 3C, D). These results suggest that PPARg is the main factor linking FGF21 and ABCA1. Taken together, our results support a model in which FGF21 upregulates ABCA1 expression by promoting the PPARg–LXRa pathway. ERK1/2 signaling is involved in regulating PPARg expression in various tissues (Costet et al., 2000; Wagner et al., 2003; Kaplan et al., 2010). PPARg–LXRa signaling is
520
LIN ET AL.
FIG. 4. Effects of FGF21 on PPARg, LXRa, and ABCA1 expression and cholesterol efflux in THP1 macrophage-derived foam cells under blockade of ERK1/2 activation. (A) Effect of FGF21 on phosphorylation of ERK1/2 under blockade of ERK1/2 activation. ERK1/2 and p-ERK1/2 protein levels were measured by western blot. (B) Effect of FGF21 on PPARg and p-PPARg expression under blockade of ERK1/2 activation. PPARg and p-PPARg proteins were measured by western blot. (C) Effect of FGF21 on ABCA1 expression under blockade of ERK1/2 activation. The ABCA1 protein was measured by western blot. (D) Effect of FGF21 on cholesterol efflux under blockade of ERK1/2 activation. Cell cholesterol efflux was analyzed by liquid scintillation counting assay as shown above. *p < 0.05 versus FGF21 + PD98059. the main pathway mediating regulation of ABCA1. Moreover, PPARg agonist-mediated induction of ABCA1 has been shown in a previous study (Nakaya et al., 2011; Ozasa et al., 2011). Furthermore, ERK1/2 signaling-mediated phosphorylation of PPARg regulates cholesterol metabolism (Mulay et al., 2013; Tian et al., 2013). Therefore, ERK1/2– PPARg may be activated by FGF21 and may upregulate ABCA1 expression and cholesterol efflux. In the current study, we found that FGF21 can activate ERK1/2 in a doseand time-dependent manner. This suggests that FGF21 increases cholesterol efflux by upregulating ABCA1 through the ERK1/2-PPARg–LXRa pathway. When the THP1 macrophage-derived foam cell was treated with an ERK1/2 inhibitor, phosphorylation of PPARg was reduced, FGF21mediated upregulation of LXRa expression was reversed, and ABCA1 expression and cholesterol efflux were both reduced. These results indicate that FGF21 can promote cholesterol efflux by upregulating ABCA1 through the ERK1/2–PPARg–LXRa pathway in THP1 macrophagederived foam cells. In summary, we have demonstrated that FGF21 induces upregulation of ABCA1 and increases cellular cholesterol efflux in an ERK1/2–PPARg–LXRa-dependent manner in
THP1 macrophage-derived foam cells. These findings provide evidence for the role of FGF21 in arteriosclerosis and a new direction for atherosclerosis prevention. Acknowledgments
This study was supported by the Innovative Research Team for Science and Technology in Higher Educational Institutions of Hunan Province and Natural Science Foundation of China (No. 81070221) and the Visiting Scholar Foundation of Key Laboratory for Biorheological Science and Technology (Chongqing University) of Ministry of Education (2010) and the construct program of the key discipline in Hunan province. Disclosure Statement
No competing financial interests exist. References
An, S.Y., Lee, M.S., Yi, S.A., et al. (2012). Serum fibroblast growth factor 21 was elevated in subjects with type 2 diabetes mellitus and was associated with the presence of carotid artery plaques. Diabetes Res Clin Pract 96, 196–203.
FGF21 INCREASES CHOLESTEROL EFFLUX
Badman, M.K., Koester, A., Flier, J.S., Kharitonenkov, A., and Maratos-Flier, E. (2009). Fibroblast growth factor 21deficient mice demonstrate impaired adaptation to ketosis. Endocrinology 150, 4931–4940. Costet, P., Luo, Y., Wang, N., Tall, A.R. (2000). Steroldependent transactivation of the abc1 promoter by the liver x receptor/retinoid x receptor. J Biol Chem 275, 28240–28245. Cuevas-Ramos, D., Aguilar-Salinas, C.A., and Go´mez-Pe´rez, F.J. (2012). Metabolic actions of fibroblast growth factor. Curr Opin Pediatr 2124, 523–529. Ding, X., Boney-Montoya, J., Owen, B.M., Bookout, A.L., Coate, K.C., Mangelsdorf, D.J., and Kliewer, S.A. (2012). Cell Metab 16, 387–393. Fisher, F.M., Estall, J.L., Adams, A.C., et al. (2011). Integrated regulation of hepatic metabolism by fibroblast growth factor 21 (FGF21) in vivo. Endocrinology 152, 2996–3004. Goetz, R. (2013). Metabolism: adiponectin-a mediator of specific metabolic actions of FGF21. Nat Rev Endocrinol 9, 506– 508. Habegger, K.M., Stemmer, K., Cheng, C., et al. (2013). Fibroblast growth factor 21 mediates specific glucagon actions. Diabetes 62, 1453–1463. Hu, Y.W., Wang, Q., Ma, X., et al. (2010). TGF-beta1 upregulates expression of ABCA1, ABCG1 and SR-BI through liver X receptor alpha signaling pathway in THP-1 macrophagederived foam cells. J Atheroscler Thromb 17, 493–502. Hughes, S.E. (1997). Differential expression of the fibroblast growth factor receptor (FGFR) multigene family in normal human adult tissues. J Histochem Cytochem 45, 1005–1019. Iglesias, P., Selgas, R., Romero, S., and Dı´ez, J.J. (2012). Biological role, clinical significance, and therapeutic possibilities of the recently discovered metabolic hormone fibroblastic growth factor 21. Eur J Endocrinol 167, 301–309. Jung, U.J., Torrejon, C., Chang, C.L., et al. (2012). Fatty acids regulate endothelial lipase and inflammatory markers in macrophages and in mouse aorta: a role for PPARg. Arterioscler Thromb Vasc Biol 32, 2929–2937. Kaplan, J.M., Hake, P.W., Denenberg, A., Nowell, M., Piraino, G., et al. (2010). Phosphorylation of extracellular signalregulated kinase (ERK)-1/2 is associated with the downregulation of peroxisome proliferator–activated receptor (PPAR)-g during polymicrobial sepsis. Mol Med 16, 491–497. Kharitonenkov, A., Wroblewski, V.J., Koester, A., et al. (2007). The metabolic state of diabetic monkeys is regulated by fibroblast growth factor-21. Endocrinology 148, 774–781. Kosinski, J.R., Hubert, J., and Carrington, P.E. (2012). The glucagon receptor is involved in mediating the body weightlowering effects of oxyntomodulin. Obesity (Silver Spring) 20, 1566–1571. Kosola, S., Lampela, H., Gylling, H., et al. (2012). Cholesterol metabolism altered and FGF21 levels high after pediatric liver transplantation despite normal serum lipids. Am J Transplant 12, 2815–2824. Li, A.C., Binder, C.J., Gutierrez, A., et al. (2004). Differential inhibition of macrophage foam-cell formation and atherosclerosis in mice by pparalpha, beta/delta, and gamma. J Clin Invest 114, 1564–1576. Li, G., Gu, H.M., Zhang, D.W., et al. (2013). ATP-binding cassette transporters and cholesterol translocation. IUBMB Life 65, 505–512. Li, H., Zhang, J., and Jia, W. (2013). Fibroblast growth factor 21: a novel metabolic regulator from pharmacology to physiology. Front Med 7, 25–30.
521
Lin, Z., Wu, Z., Yin, X., et al. (2010). Serum levels of FGF-21 are increased in coronary heart disease patientsand are independently associated with adverse lipid profile. PLoS One 5, e15534. Liu, H.F., Cui, K.F., Wang, J.P., et al. (2012). Significance of ABCA1 in human carotid atherosclerotic plaques. Exp Ther Med 4, 297–302. Liu, X.Y., Lu, Q., Ouyang, X.P., et al. (2013). Apelin-13 increases expression of ATP-binding cassette transporter A1 via activating protein kinase C a signaling in THP-1 macrophagederived foam cells. Atherosclerosis 226, 398–407. Lu¨, Y., Liu, J.H., Zhang, L.K., et al. (2010). Fibroblast growth factor 21 as a possible endogenous factor inhibits apoptosis in cardiac endothelial cells. Chin Med J (Engl) 123, 3417–3421. Mulay, V., Wood, P., Manetsch, M., et al. (2013). Inhibition of mitogen-activated protein kinase Erk1/2 promotes protein degradation of ATP binding cassette transporters A1 and G1 in CHO and HuH7 cells. PLoS One 8, e62667. Nakaya, K., Tohyama, J., and Naik, S.U. (2011). Peroxisome proliferator-activated receptor-a activation promotes macrophage reverse cholesterol transport through a liver X receptor-dependent pathway. Arterioscler Thromb Vasc Biol 31, 1276–1282. Ozasa, H., Ayaori, M., Iizuka, M., et al. (2011). Pioglitazone enhances cholesterol efflux from macrophages by increasing ABCA1/ABCG1 expressions via PPARg/LXRa pathway: findings from in vitro and ex vivo studies. Atherosclerosis 219, 141–150. Ross, R. (1999). Atherosclerosis—an inflammatory disease. N Engl J Med 340, 115–126. Smith, J.D., Le Goff, W., Settle, M., Brubaker, G., Waelde, C., Horwitz, A., and Oda, M.N. (2004). ABCA1 mediates concurrent cholesterol and phospholipid efflux to apolipoprotein A-I. J Biol Chem 279, 9963–9969. Tian, C., Ye, X., Zhang, R., et al. (2013). Green tea polyphenols reduced fat deposits in high fat-fed rats via erk1/2-PPARgadiponectin pathway. PLoS One 8, e53796. Voloshyna, I., Hai, O., Littlefield, M.J., Carsons, S., and Reiss, A.B. (2013). Resveratrol mediates anti-atherogenic effects on cholesterol flux in human macrophages and endothelium via PPARg and adenosine. Eur J Pharmacol 698, 299–309. Wagner, B.L., Valledor, A.F., Shao, G., et al. (2003). Promoterspecificroles for liver x receptor/corepressor complexes in the regulationof abca1 and srebp1 gene expression. Mol Cell Biol 23, 5780–5789. Woo, Y.C., Xu, A., Wang, Y., and Lam, K.S. (2013). Fibroblast growth factor 21 as an emerging metabolic regulator: clinical perspectives. Clin Endocrinol (Oxf) 78, 489–496. Yokoyama, S. (2006). ABCA1 and biogenesis of HDL. J Atheroscler Thromb 13, 1–15.
Address correspondence to: Zuo Wang, PhD Key Laboratory for Arteriosclerology of Hunan Province Institute of Cardiovascular Disease University of South China Hengyang 421001 China E-mail: [email protected] Received for publication November 26, 2013; received in revised form March 15, 2014; accepted March 15, 2014.
