Suppression of Adiponectin Gene Expression by Histone Deacetylase Inhibitor Valproic Acid
http://www.100md.com
《内分泌学杂志》
Graduate Center for Nutritional Sciences (L.Q., J.Sh.), University of Kentucky, Lexington, Kentucky 40536
Departments of Microbiology (J.Sc.), University of Colorado Health Sciences Center, Denver, Colorado 80262
Abstract
Valproic acid (VPA) has been used for the treatment of epilepsy and bipolar disorders for more than 30 yr. Obesity and insulin resistance are common side effects of VPA treatment. Adiponectin is an adipocyte-derived protein that plays an important role in controlling insulin sensitivity and glucose homeostasis. In this report, we examined the effects of VPA on adiponectin gene expression in C57BL/6J mice and in differentiated 3T3-L1 adipocytes. VPA treatment significantly decreased adiponectin protein and mRNA levels in both mice and 3T3-L1 adipocytes. The adipocyte study showed that VPA inhibited adiponectin gene expression in a dose- and time-dependent manner. Repression of adiponectin expression by VPA occurred at the transcription level and correlated with inhibition of histone deacetylase activity. Therapeutic concentrations of VPA increased overall histone acetylation and increased adiponectin promoter-driven luciferase expression in fibroblasts, but decreased adiponectin promoter activity in differentiated 3T3-L1 adipocytes. VPA treatment decreased adipogenic transcription factor CCAAT/enhancer binding protein- (C/EBP) levels and binding of C/EBP to the adiponectin promoter without altering the levels of peroxisome proliferator-activated receptor- and steroid regulatory element binding protein-1. Furthermore, VPA did not suppress adiponectin gene expression in C/EBP gene-deficient adipocytes that stably expressed exogenous peroxisome proliferator-activated receptor-2. Together, these results demonstrate that histone deacetylase inhibitor VPA suppresses adiponectin gene expression in mature adipocytes. The study also provides evidence that diminished C/EBP protein level and decreased binding at the adiponectin promoter mediate the inhibitory effects of VPA on adiponectin gene transcription.
Introduction
VALPROIC ACID (VPA, 2-propulpentanoic acid) is a short-chain, branched fatty acid that has been widely used as an anticonvulsant agent for the treatment of epilepsy and as a mood stabilizer for the treatment of bipolar disorder (1). Although VPA is well tolerated by patients, it can induce birth defects, polycystic ovary syndrome, obesity, and insulin resistance (1, 2, 3, 4, 5). All available data from prospective, retrospective, and cross-sectional studies demonstrate that VPA treatment is associated with a significant increase in body weight (1, 4, 5). VPA-induced weight gain strongly correlates with basal hyperinsulinemia and insulin resistance (5). However, the mechanisms underlying VPA-induced weight gain and insulin resistance are poorly understood.
Adiponectin is a recently identified adipose-derived hormone (6, 7, 8, 9). Adiponectin has diverse roles that include regulation of insulin sensitivity and energy homeostasis, immunological responses, and the development of vascular disease (10, 11, 12, 13). Administration of adiponectin improves fatty acid oxidation in the liver and skeletal muscle in vivo and reduces hepatic glucose production in vitro (14, 15, 16). These metabolic effects appear to be mediated through the adiponectin receptor and AMP kinase signaling pathway (16, 17, 18). Expression and serum levels of adiponectin are significantly reduced in humans and animals with obesity and type 2 diabetes (19, 20, 21, 22). VPA-induced obesity is associated with elevated serum insulin and leptin (5, 23). There are no available data regarding whether VPA treatment changes serum adiponectin levels. Because VPA treatment can induce significant weight gain and obesity that inversely correlate with adiponectin expression, it is reasonable to speculate that VPA treatment might alter adiponectin expression and contribute to obesity-associated insulin resistance.
Transcription in mammalian cells does not occur on naked DNA and is critically influenced by the manner in which DNA is packaged. DNA is packaged into chromatin, a highly organized and dynamic protein-DNA complex. The fundamental unit of chromatin is the nucleosome, which is composed of an octamer of four core histones, i.e. an H3/H4 tetramer and two H2A/H2B dimmers, wrapping 146 bp of DNA. Histone modifications, include site-specific acetylation, methylation, phosphorylation, and ubiquitination, affect chromatin structure and gene expression through nucleosomal core structure remodeling. Acetylation of histones, particularly H3 and H4, near the N terminus, is a critical regulatory mechanism by which gene expression is regulated. In general, increased levels of histone acetylation loosens chromatin packaging and have been correlated with transcriptional activation, whereas decreased levels of histone acetylation are associated with transcriptional repression (25). Steady-state levels of acetylation of the core histones result from the balance between the opposing activities of histone acetyltransferases and histone deacetylases (HDACs). Recently, VPA has been shown to inhibit HDACs (26, 27). Thus, inhibition of HDAC activity provides a new avenue for studying how VPA affects genes expression.
In this study, we show that 1 wk of VPA treatment significantly suppresses adiponectin gene expression in the epididymal adipose tissues from C57BL/6J mice. Our in vitro studies reveal that VPA at therapeutic concentrations inhibits adiponectin gene expression in mature 3T3-L1 adipocytes. The inhibitory effects were dose and time dependent and occurred through the inhibition of HDACs. Furthermore, our study provides evidence that VPA treatment reduces CCAAT/enhancer-binding protein (C/EBP) protein level and binding at the adiponectin promoter, the likely mechanism by which VPA suppresses adiponectin gene expression in differentiated adipocytes.
Materials and Methods
Materials
VPA, trichostatin A (TSA), sodium butyrate, bovine insulin, thiazolidinedione (TZD), dexamethasone, and isobutylmethylxanthine were purchased from Sigma (St. Louis, MO). Valpromide (VPM) was a gift from Katwijk Chemie B.B. (Katwijk, The Netherlands). 2-Methyl-2-pentenoic acid (2M2P) was purchased from Aesar (Ward Hill, MA). Penicillin-streptomycin and DMEM were purchased from Invitrogen (Carlsbad, CA). Anti-PPAR (peroxisome proliferator-activated receptor) , C/EBP, and steroid regulatory element binding protein-1 (SREBP1) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antiadiponectin and anti-aP2 (adipocyte-specific fatty acid-binding protein) monoclonal antibodies were purchased from R&D Systems (Minneapolis, MN). Anti--actin antibody was purchased from Cell Signaling Technology (Beverly, MA). Anti-acetyl-histone H3 and -acetyl-histone H4 were purchased from Upstate Biotechnologies (Lake Placid, NY).
Experimental animals
Male C57B/6J mice (10 wk old) were purchased from The Jackson Laboratory (Bar Harbor, ME). All mice were maintained under standardized conditions with 12-h light, 12-h dark cycle. The mice were randomly divided into three groups (n = 5). VPA was dissolved in saline. VPA (200 mg/kg) or saline were injected ip in a volume of 4 ml/kg three times daily. At this dosage and administration, serum VPA can reach therapeutic concentrations in humans (28, 29). For the pair-feeding mice, food supply was calculated based on the intake of VPA-treated mice and provided daily.
Cell culture
3T3-L1 fibroblasts were purchased from ATCC. 3T3-L1CAR1 cells stably express the coxsackie-adenovirus receptor, which improves the efficiency of adenoviral transduction for gain or loss of function studies (30, 31). Mouse C/EBP gene-deficient (C/EBP–/–) fibroblasts were a gift from Dr. Gretchen J. Darlington (Baylor College of Medicine, Houston, TX). Stable cell lines were created by transfection with pcDNA-PPAR2 plasmid or retroviral transduction of the human C/EBP cDNA. The retroviral vector was packaged in GP2–293 cells (CLONTECH, Palo Alto, CA). The cells were maintained in DMEM with 10% FBS (Gemini Bio-Products, Woodland, CA). Adipocyte conversion was induced by treating 2-d postconfluent cultures with DMEM containing 0.25 μM dexamethasone, 0.5 mM isobutylmethylxanthine, and 10–7 M insulin for 48 h, then with DMEM containing 10–7 M insulin for 2 d, then insulin was withdrawn from medium for the next 4 d (30, 32). For the induction of adipocyte differentiation of C/EBP- or/and PPAR2-expressing C/EBP–/– fibroblasts, 1 μM TZD was added with the above cocktail. For adenovirus transduction studies, purified adenoviruses were used at 100 pfu/cell. Twenty-four hours after transduction, protein was isolated as described below.
Oil Red O staining
After VPA treatment, cells were washed twice with PBS and fixed with 10% formalin in PBS for 1 h. Cells were stained with Oil Red O. Excess stain was washed out with water. The stained cells were dried in air. Stained oil droplets were dissolved and extracted in isopropanol containing 4% vol/vol Nonidet P-40. The samples were quantified by measuring the absorbance at 520 nM.
Apoptosis assay
Cysteine aspartic acid-specific protease (caspase) family members play key effector roles in apoptosis in mammalian cells. Caspase-3/7 activity was measured as an indicator of apoptosis using Apo-One homogenous caspase-3/7 assay kit (Promega, Madison, WI). Eight days after the initiation of adipocyte differentiation, 3T3-L1 adipocytes were treated with or without 1 mM VPA or 30 ng/ml TNF for 24 h. Cells were lysed, and caspase-3/7 activity was measured by cleavage of the caspase-3/7 substrate provided by the manufacturer using a Gemini XPS spectrofluorometer (Molecular Devices, Sunnyvale, CA) with 485 nm excitation and 530 nm emission.
Plasmid constructs and generation of recombinant adenovirus
A 1.3-kb fragment of the 5'-flanking region of the mouse adiponectin gene, extending from –1300 bp to + 18 bp relative to the cap site, was generated by PCR using mouse genomic DNA as template. The sequences of the primers are the following: forward primer, 5'-TGCTCCCGAGAATCAGCTCT-3'; reverse primer, 5'-GACAATCGTACAGACAGAAA-3'. The PCR product was ligated into pGl3-basic (mPro-Luc). The full-length human PPAR2 cDNA was inserted into pcDNA3.1 to generate pcDNA-hPPAR2, as described previously (32). To create a retrovirus vector encoding C/EBP, human C/EBP cDNA was inserted into pBabe-puro (gift from Dr. Sheila Stewart, Washington University School of Medicine, St. Louis, MO). The inserted DNA fragments for these constructs were verified by restriction analysis and sequencing. An adenovirus vector encoding small interfering RNA (siRNA) directed against C/EBP was described previously (32). Transduction efficiency was monitored by fluorescence microscopy for green fluorescent protein (GFP) that is coexpressed in the AdTrack-AdEasy recombinants.
Immunoblotting
Eight days after the induction of adipocyte differentiation, 3T3-L1 or adenovirus transduced 3T3-L1CAR1 adipocytes were treated with VPA or analogs for the times indicated in the figure legends. Nuclear or total protein were extracted and analyzed as described previously (32), using antibodies raised against mouse adiponectin, C/EBP, PPAR, SREBP1, aP2, or -actin. The bands from Western blots were quantified using Kodak Image Station 2000R and 1D software (Kodak, New Haven, CT).
Quantitative RT-PCR analysis
Total RNA was prepared from 3T3-L1 adipocytes with Trizol following the manufacturer’s protocol (Invitrogen). cDNA was synthesized using SuperScript III Reverse Transcriptase and oligo(deoxythymidine)12–18 primer (Invitrogen). Real-time PCR was performed using the Mx3000P Real-Time PCR system (Stratagene, La Jolla, CA) using SYBR Green dye (Molecular Probes, Eugene, OR). The sequences for the primers are: adiponectin 5'AAAGGAGAGCCTGGAGAAGC-3' and 5'-AAAGGAGAGCCTGGAGAAGC-3'; C/EBP 5'-TGGACAAGAACAGCAACGAG-3' and 5'-TCACTGGTCAACTCCAGCAC-3'; and -actin 5'-TACAGCTTCACCACCACAGC-3' and 5'-TACAGCTTCACCACCACAGC-3'. The levels of PCR product were calculated from standard curves established from each primer pair. Expression data were normalized to the amount of -actin PCR product.
Quantitative chromatin immunoprecipitation (ChIP) assays
ChIP was described in detail previously (32, 33). Briefly, adipocytes were fixed and chromatin was sheared by sonication. Chromatin complexes were immunoprecipitated for 12–18 h at 4 C while rotating with 5 μg of anti-C/EBP, Ac-H3, or Ac-H4 antibody or with normal rabbit serum to provide controls. Immune complexes were collected with 40 μl of protein A-agarose. Coimmunoprecipitated promoter fragments were quantitated by real-time PCR using continuous SYBR Green I monitoring as detailed previously (34). The primer sequences are: forward 5'-GTATGGGATCCGGTCTAGCA; reverse, 5'-ATTCCCAGCACCCACAGTAA. Data were expressed as fold-differences relative to control conditions, in which normal rabbit serum was used instead of specific antibody in the ChIP. The formula and calculation were described in detail previously (34). ChIP assays were independently performed at least four times. Each ChIP sample was quantitated in triplicate on at least two separate occasions. A melting curve analysis was performed for each sample after PCR amplification to ensure that a single product of expected melting curve characteristics was obtained.
Statistical analysis
Data are expressed as mean ± SD. Statistical analysis was performed using the Student’s t test (group comparison) or ANOVA analyses (dosage and time course), followed by contrast test with Tukey or Dunnett error protection. Differences were considered significant at P < 0.05.
Results
VPA suppresses adiponectin gene expression in mice
To determine whether VPA alters adiponectin expression, C57BL/6J mice were treated with VPA for 7 d. Surprisingly, instead of increasing body weight, VPA treatment led to significantly decreased body weight and epididymal fat (Table 1). Food intake of VPA-treated mice was markedly reduced compared with saline-treated controls (Table 1). Because food intake and body weight closely associate with adiponectin expression, a group of pair-feeding mice were included as further control. The results showed that, although pair-feeding reduced body weight slightly, there was no significant reduction of epididymal fat tissues (Table 1). To determine insulin sensitivity, insulin challenge tests were carried out and showed no significant difference in insulin-stimulated glucose reduction among the three groups (data not shown). These data suggest that VPA treatment decreases body weight and white adipose tissue in mice in a manner not solely dependent on decreased food intake. The mechanisms underlying the changes in weight and epididymal fat are currently under investigation.
Serum adiponectin levels in the mice were measured by Western blot. As shown in Fig. 1A, plasma adiponectin was significantly reduced in VPA-treated mice, compared with control or pair-feeding mice (P < 0.05). Interestingly, pair-feeding increased plasma adiponectin markedly (P < 0.01 vs. VPA-treated or saline control mice). A recent study reported that fasting and refeeding alters adiponectin expression in mice (35). Comparing the adiponectin protein levels between control and pair-feeding mice, our study indicates food restriction increases plasma adiponectin in mice. Adiponectin and adipogenic transcription factor C/EBP mRNA levels of in epididymal fat were significantly lower in VPA-treated mice than that in control or pair-feeding mice (P < 0.01, Fig. 1B). Together, these results demonstrated that VPA treatment inhibits adiponectin gene expression in mice. Although this in vivo study does not exclude other mechanisms that lead to decreased plasma adiponectin concentration, the evidence suggests that VPA-dependent effects on transcription contribute to decreased plasma adiponectin.
Apparently, insulin sensitivity of VPA-treated mice does not correlate with decreased adiponectin in circulation. We speculate that the relative short-term of treatment may be responsible for this. These results also suggest that VPA may alter carbohydrate metabolism via other mechanisms in vivo.
VPA inhibits adiponectin gene expression in mature adipocytes
To study whether VPA directly affects adiponectin expression, fully differentiated 3T3-L1 adipocytes (8 d after the initiation of differentiation) were used in this study. Intracellular adiponectin protein levels were measured by Western blot. As shown in Fig. 2A, adiponectin protein level was significantly reduced in adipocytes treated for 8 h with VPA (P < 0.05). Intracellular adiponectin protein levels were suppressed by 70–74% after 16 or 24 h treatment and reached their lowest levels (P < 0.001) (Fig. 2A). Interestingly, acute VPA treatment (1–4 h) tended to slightly increase adiponectin protein levels (P > 0.05). Adiponectin mRNA levels exhibited changes that paralleled changes in protein levels after treatment with 1 mM VPA for 8 h (Fig. 2B). Similar results were obtained using Chub-S7 human adipocytes (data not shown).
To test the dose-dependent effects of VPA, differentiated 3T3-L1 adipocytes were treated with different amounts of VPA for 16 h. Intracellular adiponectin protein levels were suppressed with increasing VPA concentration (Fig. 2C). These results indicate that at therapeutic concentrations VPA inhibits adiponectin gene expression in a time- and dose-dependent manner.
To determine whether VPA treatment alters adiponectin secretion, adiponectin levels in the medium were measured. There were significant reductions of adiponectin in the medium after 8 h VPA treatment. Adiponectin protein levels in the medium correlated well with intracellular levels during treatment with VPA (data not shown). Thus, VPA treatment does not appear to affect adiponectin secretion.
VPA inhibits adiponectin gene expression without inducing adipocyte dedifferentiation or apoptosis
The presence of VPA during adipocyte differentiation blocks adipogensis (36), raising the possibility that the VPA-reduced adiponectin expression is caused by adipocyte dedifferentiation or change of cellular phenotype. To address this possibility, triacylglycerol droplets were stained with Oil Red O and quantitatively analyzed in differentiated mature 3T3-L1 adipocytes that were treated with or without VPA (1 mM) for 12 or 24 h. TZD (1 μM) was used as positive control. As shown in Fig. 3A, there was no difference in the triacylglycerol droplet amounts between control and 12-h VPA-treated cells (P > 0.05). There was a small but statistically insignificant decrease after 24 h of VPA treatment. The study also showed that VPA treatment (1 mM, 24 h) slightly reduced the mRNA levels of perilipin in 3T3-L1 adipocytes, but without statistical difference (data not shown).
Although well tolerated by patients, VPA and its analogs inhibit cell proliferation and induce apoptosis in several cancer cell lines (1, 37, 38). The viability of VPA-treated adipocytes was examined using trypan blue exclusion. Surprisingly, exposure of mature 3T3-L1 adipocytes to 1 mM VPA did not reduce cell viability (Fig. 3B). Caspase-3/7 are cysteine proteases and are activated by proteolytic cleavage upon activation of the apoptotic cascade. Measuring casepase-3/7 activity has been used for studying apoptosis. After 24 h VPA treatment, there was no significant increase of caspase-3/7 activity compared with control cells (Fig. 3C). However, TNF treatment increased caspase-3/7 activity markedly. Taken together, these data suggest that VPA-inhibited adiponectin expression is not caused by dedifferentiation or induction of apoptosis.
Inhibition of HDACs by VPA appears to be required for suppressing adiponectin gene expression
VPA has recently been shown to inhibit a subset of HDACs (26, 27). To determine whether VPA-suppressed adiponectin expression in 3T3-L1 adipocytes was mediated by HDAC inhibition, a series of VPA analogs was used. Similar to VPA, treatment of 3T3-L1 adipocytes for 16 h with 3 nM TSA or 1 mM sodium butyrate decreased the intracellular levels of adiponectin protein (Fig. 4) and mRNA (data not shown). The concentrations of VPA, TSA, and sodium butyrate used in this study are within the HDAC IC50 ranges (39). Two VPA analogs, VPM and 2M2P, with minimal HDAC inhibitory effect (HDAC IC50 > 20 mM for VPM and HDAC IC50 = 15 mM for 2M2P) (39), were used to control for effects other than inhibition of HDACs. The concentrations of VPM (1 mM) and 2M2P (1 mM) used in this study have almost no HDAC inhibitory effect (39). As shown in Fig. 4, 16-h exposure to VPM or 2M2P did not reduce intracellular adiponectin protein levels in 3T3-L1 adipocytes. Total acetylated histone H4 was also measured. As expected, H4 acetylation was significantly increased in VPA, TSA, and sodium butyrate-treated cells but was not changed in VPM- or 2M2P-treated cells. Thus, these results suggest that inhibition of HDACs activity is correlated in VPA-induced down-regulation of adiponectin expression in mature adipocytes.
VPA inhibits adiponectin expression at the transcriptional level
As stated above, VPA reduced adiponectin mRNA level in adipocytes, and the inhibition was time and dose dependent. We also studied the effects of VPA on the stability of adiponectin mRNA. Treatment with 1 mM VPA did not alter the half-life of adiponectin mRNA (data not shown), suggesting that VPA inhibits adiponectin expression at the level of transcription. To further investigate whether the suppression of adiponectin gene expression by VPA occurs at the transcriptional level, a mouse adiponectin promoter-luciferase gene construct was created using pGl3-basic as vector. The gene reporter construct and transfection efficiency control pCMV--galactosidase were transiently transfected into 3T3-L1 fibroblasts or fully differentiated adipocytes (32). Twenty-four hours after transfection, the cells were treated with 1 mM VPA for 16 h. Luciferase activity was determined and normalized relative to -galactosidase activity. Similar to previous studies (32, 40), adiponectin promoter activity was 2-fold higher in differentiated adipocytes than in fibroblasts (Fig. 5, A and B). As shown in Fig. 5A, VPA increased adiponectin promoter activity in undifferentiated 3T3-L1 preadiocytes (P < 0.05). However, adiponectin promoter activity was significantly lower in VPA-treated differentiated 3T3-L1 adipocytes compared with vehicle control (P < 0.05, Fig. 5B). These results indicate that VPA decreases adiponectin gene expression at the transcriptional level and that the inhibition is adipocyte specific.
VPA suppresses adiponectin gene expression through down-regulating C/EBP
Because VPA inhibits adiponectin promoter activity in an adipocyte-specific manner, we conducted studies to determine whether the VPA-induced reduction of adiponectin in mature adipocytes results from changes in key adipogenic transcription factors. Nuclear protein was extracted from 3T3-L1 adipocytes and protein levels of C/EBP, PPAR, and SREBP1 were measured by Western blot. The results showed that, after 8 h of treatment with 1 mM VPA, both the 42- and 30-kDa C/EBP isoforms were reduced (Fig. 6A). Total C/EBP protein was reduced greater than 78% after 16 h treatment (Fig. 6A). PPAR and SREBP1 protein levels were detected using the same membranes. Interestingly, VPA treatment did not significantly change the protein levels of PPAR2, PPAR1, or cleaved SREBP1 in 3T3-L1 adipocytes (Fig. 6A). Because the SREBP1 antibody reacts with both SREBP1a and 1c, the Western results did not distinguish the isoforms.
The dependence of the reduction of C/EBP on VPA concentration was then examined. Exposure of 3T3-L1 adipocytes to 0.5 mM VPA for 16 h markedly reduced the levels of both the 42- and 30-kDa forms of C/EBP (Fig. 6B). Increasing reductions in the C/EBP isoforms were apparent with increasing VPA concentrations through 4 mM, the highest dose tested. PPAR1 and PPAR2 protein levels did not exhibit statistically significant change with VPA treatment at any tested concentration (Fig. 6B). These studies indicate that, similar to the effects on adiponectin expression, VPA treatment reduced C/EBP protein levels in 3T3-L1 adipocytes in a dose- and time-dependent manner.
Our previous study demonstrated that adiponectin gene transcription is well correlated with C/EBP protein levels in mature adipocytes (32). To determine whether the VPA-induced reduction in adiponectin expression is dependent on reductions in C/EBP, three experiments were conducted. In the first study, C/EBP protein was knocked down approximately 80% in 3T3-L1CAR1 adipocytes transduced with an adenovirus vector encoding an siRNA specific for C/EBP (32). The reduction in C/EBP protein was accompanied by a significant reduction in adiponectin protein that was similar to the reduction induced by VPA treatment (Fig. 7A).
In the second study, C/EBP gene-deficient (C/EBP–/–) mouse fibroblasts were stably transfected with C/EBP and/or PPAR2. Differentiation was induced following the protocol published previously (41). Consistent with our previous study, low levels of adiponectin protein were observed in cells expressing exogenous PPAR2 in the absence of C/EBP compared with adipocytes expressing both C/EBP and PPAR2 (Fig. 7B). Interestingly, treatment with 1 mM VPA did not suppress adiponectin protein levels in the adipocytes expressing PPAR2 alone (Fig. 7B). In contrast to the substantial reduction of adiponectin in 3T3-L1 adipocytes, VPA treatment led to only slightly reduced adiponectin in adipocytes that stably express C/EBP and PPAR2 (Fig. 7B). Furthermore, VPA treatment did not significantly reduce C/EBP protein in adipocytes that stably express C/EBP and PPAR2 (data not shown), which may due to the use of the exogenous retroviral long terminal repeat promoter to direct C/EBP expression. This result suggests that VPA treatment has no impact on exogenous promoter-driven C/EBP gene expression. The study also revealed that, consistent with the results from 3T3-L1 adipocytes, adiponectin protein levels correlated with C/EBP level in the adipocytes differentiated from mouse embryo fibroblasts.
A C/EBP binding site in the mouse promoter has recently been reported (42). Our group has recently identified a new C/EBP responding region in the mouse adiponectin promoter through which C/EBP acts as a coactivator to up-regulate adiponectin transcription (Qiao, L., and J. Shao, unpublished data). We next examined effects of VPA treatment on C/EBP binding to the adiponectin promoter region. Quantitative ChIP assay revealed that the binding of C/EBP at the proximal adiponectin promoter was significantly reduced by VPA treatment (P < 0.05), but bound acetyl-histone H3 and H4 were significantly increased (Fig. 8). Taken together, these data support the hypothesis that VPA-suppressed adiponectin expression is mediated by the reduction of C/EBP protein and impaired C/EBP binding to the adiponectin promoter in mature adipocytes.
Discussion
VPA has been used in the treatment of epilepsy and bipolar disorders for more than 30 yr. Clinical studies have demonstrated that VPA treatment usually induces obesity and insulin resistance. Adiponectin is an adipocyte-derived hormone that enhances insulin sensitivity and plays an important role in maintaining energy homeostasis. Numerous studies demonstrate that adiponectin gene expression inversely correlates with body weight. Therefore, it is reasonable to speculate that VPA-induced insulin resistance might occur through the down-regulation of adiponectin. Interestingly, a recent study has shown that VPA does not increase adipocyte differentiation in vitro (36). Instead, VPA blocks adipogenesis (36). VPA induces significant weight gain in rhesus monkeys but not rodents (43, 44, 45, 46). Furthermore, our study demonstrates that VPA actually decreases body weight and epididymal fat tissue, as well as adiponetin expression, in mice. Therefore, multiple mechanisms likely underlie VPA-induced weight gain and insulin resistance in humans. These studies suggest that VPA may induce human obesity through an indirect mechanism, rather than through direct effects on adipocyte differentiation.
Using an in vitro cell culture system, we show here that VPA directly inhibits adiponectin gene expression in differentiated adipocytes. Because the current study does not rule out the possibility of obesity-driven low adiponectin expression, we postulate that VPA-suppressed adiponectin gene expression is one of the mechanisms that contribute to VPA-induced insulin resistance. Unfortunately, there are no clinical data regarding whether VPA alters adiponectin expression in human subjects.
Adiponectin is exclusively expressed in adipocytes (6, 7). Adiponectin gene expression is turned on 2 d after the initiation of adipocyte differentiation and maintained at a relatively high level in mature adipocytes. Blockage of adipocyte differentiation reduces adiponectin expression. To avoid effects of VPA on differentiation (36), all of the cells used in this study were fully differentiated mature adipocytes and were treated beginning 8 d after initiation of adipocyte differentiation. The fact that treatment with VPA for 24 h did not alter triacylglycerol droplet content indicates that VPA treatment did not induce dedifferentiation of the adipocytes. Therefore, we conclude that VPA-suppressed adiponectin gene expression in mature adipocytes occurs through a mechanism distinct from that of the antiadipocyte differentiation effects of VPA.
VPA and certain of its analogs arrest growth and apoptosis in several types of cancer cells (1). Our study showed that the presence of 1 mM VPA in the cell culture medium, a level reached in patient serum during therapy of epilepsy with a daily dose of 20–30 mg/kg, did not induce significant loss of cell viability. In addition, Western blot studies showed that -actin levels were fairly stable and were not altered by VPA treatment of adipocytes. The mechanisms involved in VPA- or VPA analog-induced apoptosis are intricate and differ among cell types. Therapeutic concentrations of VPA do not induce apoptosis in most cell types (37, 38). Therefore, these results demonstrate that the inhibition of adiponectin gene expression by VPA in mature adipocytes is gene specific and not due to cellular toxicity or apoptosis.
Histone acetylation plays an important role in regulating transcription (25). Histone acetylation is dynamic and the level of acetylation is determined by the balance of the activities of acetyltransferases and HDACs. Recently, HDACs have been identified as direct targets of VPA (26, 27). VPA inhibits both class I and II HDACs, with a high potency for class I (26, 27, 39). Our study showed that VPA and its analogs with HDAC inhibitory effects suppress adiponectin expression. In contrast, the VPA analogs VPM and 2M2P, which do not inhibit HDAC activity, did not affect adiponectin protein levels in adipocytes (Fig. 4). Therefore, the inhibition of adiponectin gene expression by VPA is most likely mediated through HDAC inhibition. However, further studies are required to determine whether HDAC inhibition directly suppresses adiponectin gene expression.
In general, increased levels of histone acetylation by histone acetyltransferases loosens chromatin packaging and are correlated with transcriptional activation, whereas histone deacetylation by HDACs is associated with transcriptional repression (25). However, recent studies have demonstrated that histone deacetylation also associate with transcription activation of certain genes (47). VPA and other HDAC inhibitors increase histone acetylation universally by suppressing HDAC activity. VPA or TSA treatment increase transcription of genes including those encoding gelsolin, cytokeratin A, and c-fos (1). However, inhibition of HDAC activities also suppresses expression of certain genes including those encoding c-myc, IL-2, IL-8, and LC-pTP (1, 48, 49, 50). Inhibition of HDACs by TSA or trapoxin alters transcription, either up or down, of approximately 2% of genes (48). Therefore, HDAC inhibitors regulate gene expression either positively or negatively in a gene-specific manner. Our study showed that VPA treatment represses adiponectin gene transcription in mature adipocytes. Interestingly, our study also revealed that VPA increased mouse adiponectin promoter-driven luciferase expression in 3T3-L1 fibroblasts but decreased it in differentiated 3T3-L1 adipocytes. This finding suggests that, similar to most promoters tested, hyperacetylation of histones up-regulates adiponectin promoter activity. However, the adipocyte-specific inhibition of adiponectin promoter activity by VPA suggests that VPA treatment inhibits adiponectin gene expression through an indirect mechanism. Therefore, we hypothesized that VPA treatment might alter signal pathway(s) or transcription factor(s) and thereby inhibit adiponectin gene transcription in mature adipocytes.
C/EBP and PPAR are major adipogenic transcription factors that transactivate many adipocyte-specific genes. Our previous study and other studies have demonstrated that C/EBP is required to fully activate adiponectin gene expression (32, 41, 51). SREBP1c also plays a role in regulating adiponectin gene transcription. Our current study showed that VPA treatment significantly decreased C/EBP protein levels in differentiated adipocytes. However, VPA treatment did not alter protein levels of PPAR1, PPAR2, or SREBP1. Furthermore, VPA treatment did not alter adiponectin expression in C/EBP-deficient adipocytes that ectopicly express PPAR2 (Fig. 7B). These studies strongly support our hypothesis that by reducing the level of the adipogenic transcription factor C/EBP, VPA inhibits adiponectin gene transcription in adipocytes.
C/EBP-responsive elements that are associated with C/EBP and C/EBP have been identified in the mouse adiponectin promoter (42) (our unpublished data). While we were preparing this manuscript, a new study demonstrated that inhibition of HDACs increases C/EBP acetylation as well as histone acetylation (37). The study demonstrated that elevated acetylation of C/EBP inhibits its binding to target DNA sequences (37). Although we did not test the acetylation levels of C/EBP, our quantitative ChIP assay showed the binding of C/EBP to the adiponectin promoter region is significantly reduced in VPA-treated adipocytes. Therefore, we conclude that VPA inhibits adiponectin gene transcription in adipocytes by reducing the amount of C/EBP protein and its binding to the adiponectin promoter. In contrast to our study, Lagace et al. (24) reported that 3 h of VPA treatment transiently reduced PPAR and SREBP1a mRNA in adipocytes, but prolonged exposure (6 or 12 h) had no significant effect on protein or mRNA levels of C/EBP, PPAR, and SREBP1a. We do not have any evidence to explain the discrepancy but speculate that differences in the adipocyte differentiation procedure and differentiation stages of the adipocytes might contribute. The protein samples used in our study were nuclear extract while total cell lysis was used in the study by Lagace et al. Thus, differences in sample preparation may also contribute to the discrepant results. Strikingly, VPA treatment (1 mM, 12 h) significantly increases C/EBP protein levels in FAO hepatoma cells (Qiao, L., and J. Shao, unpublished data). These data indicate that VPA regulates C/EBP gene expression in a tissue-specific manner. Nevertheless, further studies are warranted to investigate the mechanism(s) whereby VPA represses C/EBP expression in mature adipocytes.
In summary, our results demonstrate that VPA suppresses adiponectin gene expression in mice and differentiated adipocytes, and that the suppression is correlated with the inhibition of HDAC activity of VPA. VPA-suppressed adiponectin gene expression occurs at the transcriptional level. We provide evidence that VPA suppresses adiponectin gene expression via down-regulation of C/EBP and its binding at the adiponectin promoter.
Acknowledgments
We thank Dr. Lisa A. Cassis (Graduate Center for Nutritional Sciences, University of Kentucky, Lexington, KY) for comments on the manuscript and Dr. Paul MacLean (Department of Medicine, University of Colorado Health Sciences Center, Aurora, CO) for helpful discussion.
Footnotes
This work was supported by grants from the American Diabetes Association (1-04-JF-44; to J.Sh.).
First Published Online November 10, 2005
Abbreviations: aP2, Adipocyte-specific fatty acid-binding protein; C/EBP, CCAAT/enhancer-binding protein; ChIP, chromatin immunoprecipitation; GFP, green fluorescent protein; HDAC, histone deacetylase; 2M2P, 2-methyl-2-pentenoic acid; PPAR, peroxisome proliferator-activated receptor ; siRNA, small interfering RNA; SREBP1, steroid regulatory element binding protein-1; TSA, trichostatin A; TZD, thiazolidinedione; VPA, valproic acid; VPM, valpromide.
Accepted for publication October 31, 2005.
References
Gurvich N, Klein PS 2002 Lithium and valproic acid: parallels and contrasts in diverse signaling contexts. Pharmacol Ther 96:45–66
Genton P, Bauer J, Duncan S, Taylor AE, Balen AH, Eberle A, Pedersen B, Salas-Puig X, Sauer MV 2001 On the association between Valproate and polycystic ovary syndrome. Epilepsia 42:295–304
Devlin MJ, Yanovski SZ, Wilson GT 2000 Obesity: what mental health professionals need to know. Am J Psychiatry 157:854–866
Corman C, Leung N, Guberman A 1997 Weight gain in epileptic patients during treatment with valproic acid: a retrospective study. Can J Neurol Sci 24:240–244
Verrotti A, Basciani F, De Simone M, Trotta D, Morgese G, Chiarelli F 2002 Insulin resistance in epileptic girls who gain weight after therapy with valproic acid. J Child Neurol 17:265–268
Scherer PE, Williams S, Fogliano M, Baldini G, Lodish HF 1995 A novel serum protein similar to C1q, produced exclusively in adipocytes. J Biol Chem 270:26746–26749
Hu E, Liang P, Spiegelman BM 1996 AdipoQ is a novel adipose-specific gene dysregulated in obesity. J Biol Chem 271:10697–10703
Maeda K, Okubo K, Shimomura I, Funahashi T, Matsuzawa Y, Matsubara K 1996 cDNA cloning and expression of a novel adipose specific collagen-like factor, apM1 (AdiPose most abundant gene transcript 1). Biochem Biophys Res Commun 221:286–289
Nakano Y, Tobe T, Choi-Miura NH, Mazda T, Tomita M 1996 Isolation and characterization of GBP28, a novel gelatin-binding protein purified from human plasma. J Biochem (Tokyo) 120:803–812
Kim S, Moustaid-Moussa N 2000 Secretory, endocrine and autocrine/paracrine function of the adipocyte. J Nutr 130:3110S–3115S
Kahn BB, Flier JS 2000 Obesity and insulin resistance. J Clin Invest 106:473–481
Diez JJ, Iglesias P 2003 The role of the novel adipocyte-derived hormone adiponectin in human disease. Eur J Endocrinol 148:293–300
Stefan N, Stumvoll M 2002 Adiponectin—its role in metabolism and beyond. Horm Metab Res 34:469–474
Berg AH, Combs TP, Du X, Brownlee M, Scherer PE 2001 The adipocyte-secreted protein Acrp30 enhances hepatic insulin action. Nat Med 7:947–953
Fruebis J, Tsao TS, Javorschi S, Ebbets-Reed D, Erickson MR, Yen FT, Bihain BE, Lodish HF 2001 Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci USA 98:2005–2010
Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, Yamashita S, Noda M, Kita S, Ueki K, Eto K, Akanuma Y, Froguel P, Foufelle F, Ferre P, Carling D, Kimura S, Nagai R, Kahn BB, Kadowaki T 2002 Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 8:1288–1295
Yamauchi T, Kamon J, Ito Y, Tsuchida A, Yokomizo T, Kita S, Sugiyama T, Miyagishi M, Hara K, Tsunoda M, Murakami K, Ohteki T, Uchida S, Takekawa S, Waki H, Tsuno NH, Shibata Y, Terauchi Y, Froguel P, Tobe K, Koyasu S, Taira K, Kitamura T, Shimizu T, Nagai R, Kadowaki T 2003 Cloning of adiponectin receptors that mediate antidiabetic metabolic effects. Nature 423:762–769
Tomas E, Tsao TS, Saha AK, Murrey HE, Zhang Cc C, Itani SI, Lodish HF, Ruderman NB 2002 Enhanced muscle fat oxidation and glucose transport by ACRP30 globular domain: acetyl-CoA carboxylase inhibition and AMP-activated protein kinase activation. Proc Natl Acad Sci USA 99:16309–16313
Arita Y, Kihara S, Ouchi N, Takahashi M, Maeda K, Miyagawa J, Hotta K, Shimomura I, Nakamura T, Miyaoka K, Kuriyama H, Nishida M, Yamashita S, Okubo K, Matsubara K, Muraguchi M, Ohmoto Y, Funahashi T, Matsuzawa Y 1999 Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 257:79–83
Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M, Okamoto Y, Iwahashi H, Kuriyama H, Ouchi N, Maeda K, Nishida M, Kihara S, Sakai N, Nakajima T, Hasegawa K, Muraguchi M, Ohmoto Y, Nakamura T, Yamashita S, Hanafusa T, Matsuzawa Y 2000 Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol 20:1595–1599
Statnick MA, Beavers LS, Conner LJ, Corominola H, Johnson D, Hammond CD, Rafaeloff-Phail R, Seng T, Suter TM, Sluka JP, Ravussin E, Gadski RA, Caro JF 2000 Decreased expression of apM1 in omental and subcutaneous adipose tissue of humans with type 2 diabetes. Int J Exp Diabetes Res 1:81–88
Tsao TS, Lodish HF, Fruebis J 2002 ACRP30, a new hormone controlling fat and glucose metabolism. Eur J Pharmacol 440:213–221
Verrotti A, Basciani F, Morresi S, de Martino M, Morgese G, Chiarelli F 1999 Serum leptin changes in epileptic patients who gain weight after therapy with valproic acid. Neurology 53:230–232
Lagace DC, McLeod RS, Nachtigal MW 2004 Valproic acid inhibits leptin secretion and reduces leptin messenger ribonucleic acid levels in adipocytes. Endocrinology 145:5493–5503
Struhl K 1998 Histone acetylation and transcriptional regulatory mechanisms. Genes Dev 12:599–606
Gottlicher M, Minucci S, Zhu P, Kramer OH, Schimpf A, Giavara S, Sleeman JP, Lo Coco F, Nervi C, Pelicci PG, Heinzel T 2001 Valproic acid defines a novel class of HDAC inhibitors inducing differentiation of transformed cells. EMBO J 20:6969–6978
Phiel CJ, Zhang F, Huang EY, Guenther MG, Lazar MA, Klein PS 2001 Histone deacetylase is a direct target of valproic acid, a potent anticonvulsant, mood stabilizer, and teratogen. J Biol Chem 276:36734–36741
Nau H 1985 Teratogenic valproic acid concentrations: infusion by implanted minipumps vs conventional injection regimen in the mouse. Toxicol Appl Pharmacol 80:243–250
Schank JR, Liles LC, Weinshenker D 2005 Reduced anticonvulsant efficacy of valproic acid in dopamine -hydroxylase knockout mice. Epilepsy Res 65:23–31
Orlicky DJ, DeGregori J, Schaack J 2001 Construction of stable coxsackievirus and adenovirus receptor-expressing 3T3-L1 cells. J Lipid Res 42:910–915
Ross SA, Song X, Burney MW, Kasai Y, Orlicky DJ 2003 Efficient adenovirus transduction of 3T3-L1 adipocytes stably expressing coxsackie-adenovirus receptor. Biochem Biophys Res Commun 302:354–358
Qiao L, MacLean PS, Schaack J, Orlicky DJ, Darimont C, Pagliassotti M, Friedman JE, Shao J 2005 C/EBP regulates human adiponectin gene transcription through an intronic enhancer. Diabetes 54:1744–1754
Shao J, Qiao L, Janssen RC, Pagliassotti M, Friedman JE 2005 Chronic hyperglycemia enhances PEPCK gene expression and hepatocellular glucose production via elevated liver activating protein/liver inhibitory protein ratio. Diabetes 54:976–984
Chakrabarti SK, James JC, Mirmira RG 2002 Quantitative assessment of gene targeting in vitro and in vivo by the pancreatic transcription factor, Pdx1. Importance of chromatin structure in directing promoter binding. J Biol Chem 277:13286–13293
Hazel M, Cooksey RC, Jones D, Parker G, Neidigh JL, Witherbee B, Gulve EA, McClain DA 2004 Activation of the hexosamine signaling pathway in adipose tissue results in decreased serum adiponectin and skeletal muscle insulin resistance. Endocrinology 145:2118–2128
Lagace DC, Nachtigal MW 2004 Inhibition of histone deacetylase activity by valproic acid blocks adipogenesis. J Biol Chem 279:18851–18860
Duan H, Heckman CA, Boxer LM 2005 Histone deacetylase inhibitors down-regulate bcl-2 expression and induce apoptosis in t(14;18) lymphomas. Mol Cell Biol 25:1608–1619
Sanjo Y, Yokoyama T, Sato S, Ikeda K, Nakajima R 1999 Ergonomic automated anesthesia recordkeeper using a mobile touch screen with voice navigtion. J Clin Monit Comput 15:347–356
Gurvich N, Tsygankova OM, Meinkoth JL, Klein PS 2004 Histone deacetylase is a target of valproic acid-mediated cellular differentiation. Cancer Res 64:1079–1086
Das K, Lin Y, Widen E, Zhang Y, Scherer PE 2001 Chromosomal localization, expression pattern, and promoter analysis of the mouse gene encoding adipocyte-specific secretory protein Acrp30. Biochem Biophys Res Commun 280:1120–1129
Gustafson B, Jack MM, Cushman SW, Smith U 2003 Adiponectin gene activation by thiazolidinediones requires PPAR2, but not C/EBP-evidence for differential regulation of the aP2 and adiponectin genes. Biochem Biophys Res Commun 308:933–939
Park S-k, Oh S-Y, Lee M-Y, Yoon S, Kim K-S, Kim J-w2004 CCAAT/enhancer binding protein and nuclear factor-Y regulate adiponectin gene expression in adipose tissue. Diabetes 53:2757–2766
Ferin M, Morrell M, Xiao E, Kochan L, Qian F, Wright T, Sauer M 2003 Endocrine and metabolic responses to long-term monotherapy with the antiepileptic drug valproate in the normally cycling rhesus monkey. J Clin Endocrinol Metab 88:2908–2915
Lagace DC, Nachtigal MW 2003 Valproic acid fails to induce polycystic ovary syndrome in female rats. Prog Neuropsychopharmacol Biol Psychiatry 27:587–594
Wolden-Hanson T, Gidal BE, Atkinson RL 1998 Evaluation of a rat model of valproate-induced obesity. Pharmacotherapy 18:1075–1081
Chapman JB, Cutler MG 1984 Sodium valproate: effects on social behaviour and physical development in the mouse. Psychopharmacology (Berl) 83:390–396
Nusinzon I, Horvath CM 2005 Histone deacetylases as transcriptional activators Role reversal in inducible gene regulation. Sci STKE 2005:re11
Van Lint C, Emiliani S, Verdin E 1996 The expression of a small fraction of cellular genes is changed in response to histone hyperacetylation. Gene Expr 5:245–253
Takahashi I, Miyaji H, Yoshida T, Sato S, Mizukami T 1996 Selective inhibition of IL-2 gene expression by trichostatin A, a potent inhibitor of mammalian histone deacetylase. J Antibiot (Tokyo) 49:453–457
Huang N, Katz JP, Martin DR, Wu GD 1997 Inhibition of IL-8 gene expression in Caco-2 cells by compounds which induce histone hyperacetylation. Cytokine 9:27–36
Park B-H, Qiang L, Farmer SR 2004 Phosphorylation of C/EBP at a consensus extracellular signal-regulated kinase/glycogen synthase kinase 3 site is required for the induction of adiponectin gene expression during the differentiation of mouse fibroblasts into adipocytes. Mol Cell Biol 24:8671–8680(Liping Qiao, Jerome Schaack and Jianhua )
Departments of Microbiology (J.Sc.), University of Colorado Health Sciences Center, Denver, Colorado 80262
Abstract
Valproic acid (VPA) has been used for the treatment of epilepsy and bipolar disorders for more than 30 yr. Obesity and insulin resistance are common side effects of VPA treatment. Adiponectin is an adipocyte-derived protein that plays an important role in controlling insulin sensitivity and glucose homeostasis. In this report, we examined the effects of VPA on adiponectin gene expression in C57BL/6J mice and in differentiated 3T3-L1 adipocytes. VPA treatment significantly decreased adiponectin protein and mRNA levels in both mice and 3T3-L1 adipocytes. The adipocyte study showed that VPA inhibited adiponectin gene expression in a dose- and time-dependent manner. Repression of adiponectin expression by VPA occurred at the transcription level and correlated with inhibition of histone deacetylase activity. Therapeutic concentrations of VPA increased overall histone acetylation and increased adiponectin promoter-driven luciferase expression in fibroblasts, but decreased adiponectin promoter activity in differentiated 3T3-L1 adipocytes. VPA treatment decreased adipogenic transcription factor CCAAT/enhancer binding protein- (C/EBP) levels and binding of C/EBP to the adiponectin promoter without altering the levels of peroxisome proliferator-activated receptor- and steroid regulatory element binding protein-1. Furthermore, VPA did not suppress adiponectin gene expression in C/EBP gene-deficient adipocytes that stably expressed exogenous peroxisome proliferator-activated receptor-2. Together, these results demonstrate that histone deacetylase inhibitor VPA suppresses adiponectin gene expression in mature adipocytes. The study also provides evidence that diminished C/EBP protein level and decreased binding at the adiponectin promoter mediate the inhibitory effects of VPA on adiponectin gene transcription.
Introduction
VALPROIC ACID (VPA, 2-propulpentanoic acid) is a short-chain, branched fatty acid that has been widely used as an anticonvulsant agent for the treatment of epilepsy and as a mood stabilizer for the treatment of bipolar disorder (1). Although VPA is well tolerated by patients, it can induce birth defects, polycystic ovary syndrome, obesity, and insulin resistance (1, 2, 3, 4, 5). All available data from prospective, retrospective, and cross-sectional studies demonstrate that VPA treatment is associated with a significant increase in body weight (1, 4, 5). VPA-induced weight gain strongly correlates with basal hyperinsulinemia and insulin resistance (5). However, the mechanisms underlying VPA-induced weight gain and insulin resistance are poorly understood.
Adiponectin is a recently identified adipose-derived hormone (6, 7, 8, 9). Adiponectin has diverse roles that include regulation of insulin sensitivity and energy homeostasis, immunological responses, and the development of vascular disease (10, 11, 12, 13). Administration of adiponectin improves fatty acid oxidation in the liver and skeletal muscle in vivo and reduces hepatic glucose production in vitro (14, 15, 16). These metabolic effects appear to be mediated through the adiponectin receptor and AMP kinase signaling pathway (16, 17, 18). Expression and serum levels of adiponectin are significantly reduced in humans and animals with obesity and type 2 diabetes (19, 20, 21, 22). VPA-induced obesity is associated with elevated serum insulin and leptin (5, 23). There are no available data regarding whether VPA treatment changes serum adiponectin levels. Because VPA treatment can induce significant weight gain and obesity that inversely correlate with adiponectin expression, it is reasonable to speculate that VPA treatment might alter adiponectin expression and contribute to obesity-associated insulin resistance.
Transcription in mammalian cells does not occur on naked DNA and is critically influenced by the manner in which DNA is packaged. DNA is packaged into chromatin, a highly organized and dynamic protein-DNA complex. The fundamental unit of chromatin is the nucleosome, which is composed of an octamer of four core histones, i.e. an H3/H4 tetramer and two H2A/H2B dimmers, wrapping 146 bp of DNA. Histone modifications, include site-specific acetylation, methylation, phosphorylation, and ubiquitination, affect chromatin structure and gene expression through nucleosomal core structure remodeling. Acetylation of histones, particularly H3 and H4, near the N terminus, is a critical regulatory mechanism by which gene expression is regulated. In general, increased levels of histone acetylation loosens chromatin packaging and have been correlated with transcriptional activation, whereas decreased levels of histone acetylation are associated with transcriptional repression (25). Steady-state levels of acetylation of the core histones result from the balance between the opposing activities of histone acetyltransferases and histone deacetylases (HDACs). Recently, VPA has been shown to inhibit HDACs (26, 27). Thus, inhibition of HDAC activity provides a new avenue for studying how VPA affects genes expression.
In this study, we show that 1 wk of VPA treatment significantly suppresses adiponectin gene expression in the epididymal adipose tissues from C57BL/6J mice. Our in vitro studies reveal that VPA at therapeutic concentrations inhibits adiponectin gene expression in mature 3T3-L1 adipocytes. The inhibitory effects were dose and time dependent and occurred through the inhibition of HDACs. Furthermore, our study provides evidence that VPA treatment reduces CCAAT/enhancer-binding protein (C/EBP) protein level and binding at the adiponectin promoter, the likely mechanism by which VPA suppresses adiponectin gene expression in differentiated adipocytes.
Materials and Methods
Materials
VPA, trichostatin A (TSA), sodium butyrate, bovine insulin, thiazolidinedione (TZD), dexamethasone, and isobutylmethylxanthine were purchased from Sigma (St. Louis, MO). Valpromide (VPM) was a gift from Katwijk Chemie B.B. (Katwijk, The Netherlands). 2-Methyl-2-pentenoic acid (2M2P) was purchased from Aesar (Ward Hill, MA). Penicillin-streptomycin and DMEM were purchased from Invitrogen (Carlsbad, CA). Anti-PPAR (peroxisome proliferator-activated receptor) , C/EBP, and steroid regulatory element binding protein-1 (SREBP1) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antiadiponectin and anti-aP2 (adipocyte-specific fatty acid-binding protein) monoclonal antibodies were purchased from R&D Systems (Minneapolis, MN). Anti--actin antibody was purchased from Cell Signaling Technology (Beverly, MA). Anti-acetyl-histone H3 and -acetyl-histone H4 were purchased from Upstate Biotechnologies (Lake Placid, NY).
Experimental animals
Male C57B/6J mice (10 wk old) were purchased from The Jackson Laboratory (Bar Harbor, ME). All mice were maintained under standardized conditions with 12-h light, 12-h dark cycle. The mice were randomly divided into three groups (n = 5). VPA was dissolved in saline. VPA (200 mg/kg) or saline were injected ip in a volume of 4 ml/kg three times daily. At this dosage and administration, serum VPA can reach therapeutic concentrations in humans (28, 29). For the pair-feeding mice, food supply was calculated based on the intake of VPA-treated mice and provided daily.
Cell culture
3T3-L1 fibroblasts were purchased from ATCC. 3T3-L1CAR1 cells stably express the coxsackie-adenovirus receptor, which improves the efficiency of adenoviral transduction for gain or loss of function studies (30, 31). Mouse C/EBP gene-deficient (C/EBP–/–) fibroblasts were a gift from Dr. Gretchen J. Darlington (Baylor College of Medicine, Houston, TX). Stable cell lines were created by transfection with pcDNA-PPAR2 plasmid or retroviral transduction of the human C/EBP cDNA. The retroviral vector was packaged in GP2–293 cells (CLONTECH, Palo Alto, CA). The cells were maintained in DMEM with 10% FBS (Gemini Bio-Products, Woodland, CA). Adipocyte conversion was induced by treating 2-d postconfluent cultures with DMEM containing 0.25 μM dexamethasone, 0.5 mM isobutylmethylxanthine, and 10–7 M insulin for 48 h, then with DMEM containing 10–7 M insulin for 2 d, then insulin was withdrawn from medium for the next 4 d (30, 32). For the induction of adipocyte differentiation of C/EBP- or/and PPAR2-expressing C/EBP–/– fibroblasts, 1 μM TZD was added with the above cocktail. For adenovirus transduction studies, purified adenoviruses were used at 100 pfu/cell. Twenty-four hours after transduction, protein was isolated as described below.
Oil Red O staining
After VPA treatment, cells were washed twice with PBS and fixed with 10% formalin in PBS for 1 h. Cells were stained with Oil Red O. Excess stain was washed out with water. The stained cells were dried in air. Stained oil droplets were dissolved and extracted in isopropanol containing 4% vol/vol Nonidet P-40. The samples were quantified by measuring the absorbance at 520 nM.
Apoptosis assay
Cysteine aspartic acid-specific protease (caspase) family members play key effector roles in apoptosis in mammalian cells. Caspase-3/7 activity was measured as an indicator of apoptosis using Apo-One homogenous caspase-3/7 assay kit (Promega, Madison, WI). Eight days after the initiation of adipocyte differentiation, 3T3-L1 adipocytes were treated with or without 1 mM VPA or 30 ng/ml TNF for 24 h. Cells were lysed, and caspase-3/7 activity was measured by cleavage of the caspase-3/7 substrate provided by the manufacturer using a Gemini XPS spectrofluorometer (Molecular Devices, Sunnyvale, CA) with 485 nm excitation and 530 nm emission.
Plasmid constructs and generation of recombinant adenovirus
A 1.3-kb fragment of the 5'-flanking region of the mouse adiponectin gene, extending from –1300 bp to + 18 bp relative to the cap site, was generated by PCR using mouse genomic DNA as template. The sequences of the primers are the following: forward primer, 5'-TGCTCCCGAGAATCAGCTCT-3'; reverse primer, 5'-GACAATCGTACAGACAGAAA-3'. The PCR product was ligated into pGl3-basic (mPro-Luc). The full-length human PPAR2 cDNA was inserted into pcDNA3.1 to generate pcDNA-hPPAR2, as described previously (32). To create a retrovirus vector encoding C/EBP, human C/EBP cDNA was inserted into pBabe-puro (gift from Dr. Sheila Stewart, Washington University School of Medicine, St. Louis, MO). The inserted DNA fragments for these constructs were verified by restriction analysis and sequencing. An adenovirus vector encoding small interfering RNA (siRNA) directed against C/EBP was described previously (32). Transduction efficiency was monitored by fluorescence microscopy for green fluorescent protein (GFP) that is coexpressed in the AdTrack-AdEasy recombinants.
Immunoblotting
Eight days after the induction of adipocyte differentiation, 3T3-L1 or adenovirus transduced 3T3-L1CAR1 adipocytes were treated with VPA or analogs for the times indicated in the figure legends. Nuclear or total protein were extracted and analyzed as described previously (32), using antibodies raised against mouse adiponectin, C/EBP, PPAR, SREBP1, aP2, or -actin. The bands from Western blots were quantified using Kodak Image Station 2000R and 1D software (Kodak, New Haven, CT).
Quantitative RT-PCR analysis
Total RNA was prepared from 3T3-L1 adipocytes with Trizol following the manufacturer’s protocol (Invitrogen). cDNA was synthesized using SuperScript III Reverse Transcriptase and oligo(deoxythymidine)12–18 primer (Invitrogen). Real-time PCR was performed using the Mx3000P Real-Time PCR system (Stratagene, La Jolla, CA) using SYBR Green dye (Molecular Probes, Eugene, OR). The sequences for the primers are: adiponectin 5'AAAGGAGAGCCTGGAGAAGC-3' and 5'-AAAGGAGAGCCTGGAGAAGC-3'; C/EBP 5'-TGGACAAGAACAGCAACGAG-3' and 5'-TCACTGGTCAACTCCAGCAC-3'; and -actin 5'-TACAGCTTCACCACCACAGC-3' and 5'-TACAGCTTCACCACCACAGC-3'. The levels of PCR product were calculated from standard curves established from each primer pair. Expression data were normalized to the amount of -actin PCR product.
Quantitative chromatin immunoprecipitation (ChIP) assays
ChIP was described in detail previously (32, 33). Briefly, adipocytes were fixed and chromatin was sheared by sonication. Chromatin complexes were immunoprecipitated for 12–18 h at 4 C while rotating with 5 μg of anti-C/EBP, Ac-H3, or Ac-H4 antibody or with normal rabbit serum to provide controls. Immune complexes were collected with 40 μl of protein A-agarose. Coimmunoprecipitated promoter fragments were quantitated by real-time PCR using continuous SYBR Green I monitoring as detailed previously (34). The primer sequences are: forward 5'-GTATGGGATCCGGTCTAGCA; reverse, 5'-ATTCCCAGCACCCACAGTAA. Data were expressed as fold-differences relative to control conditions, in which normal rabbit serum was used instead of specific antibody in the ChIP. The formula and calculation were described in detail previously (34). ChIP assays were independently performed at least four times. Each ChIP sample was quantitated in triplicate on at least two separate occasions. A melting curve analysis was performed for each sample after PCR amplification to ensure that a single product of expected melting curve characteristics was obtained.
Statistical analysis
Data are expressed as mean ± SD. Statistical analysis was performed using the Student’s t test (group comparison) or ANOVA analyses (dosage and time course), followed by contrast test with Tukey or Dunnett error protection. Differences were considered significant at P < 0.05.
Results
VPA suppresses adiponectin gene expression in mice
To determine whether VPA alters adiponectin expression, C57BL/6J mice were treated with VPA for 7 d. Surprisingly, instead of increasing body weight, VPA treatment led to significantly decreased body weight and epididymal fat (Table 1). Food intake of VPA-treated mice was markedly reduced compared with saline-treated controls (Table 1). Because food intake and body weight closely associate with adiponectin expression, a group of pair-feeding mice were included as further control. The results showed that, although pair-feeding reduced body weight slightly, there was no significant reduction of epididymal fat tissues (Table 1). To determine insulin sensitivity, insulin challenge tests were carried out and showed no significant difference in insulin-stimulated glucose reduction among the three groups (data not shown). These data suggest that VPA treatment decreases body weight and white adipose tissue in mice in a manner not solely dependent on decreased food intake. The mechanisms underlying the changes in weight and epididymal fat are currently under investigation.
Serum adiponectin levels in the mice were measured by Western blot. As shown in Fig. 1A, plasma adiponectin was significantly reduced in VPA-treated mice, compared with control or pair-feeding mice (P < 0.05). Interestingly, pair-feeding increased plasma adiponectin markedly (P < 0.01 vs. VPA-treated or saline control mice). A recent study reported that fasting and refeeding alters adiponectin expression in mice (35). Comparing the adiponectin protein levels between control and pair-feeding mice, our study indicates food restriction increases plasma adiponectin in mice. Adiponectin and adipogenic transcription factor C/EBP mRNA levels of in epididymal fat were significantly lower in VPA-treated mice than that in control or pair-feeding mice (P < 0.01, Fig. 1B). Together, these results demonstrated that VPA treatment inhibits adiponectin gene expression in mice. Although this in vivo study does not exclude other mechanisms that lead to decreased plasma adiponectin concentration, the evidence suggests that VPA-dependent effects on transcription contribute to decreased plasma adiponectin.
Apparently, insulin sensitivity of VPA-treated mice does not correlate with decreased adiponectin in circulation. We speculate that the relative short-term of treatment may be responsible for this. These results also suggest that VPA may alter carbohydrate metabolism via other mechanisms in vivo.
VPA inhibits adiponectin gene expression in mature adipocytes
To study whether VPA directly affects adiponectin expression, fully differentiated 3T3-L1 adipocytes (8 d after the initiation of differentiation) were used in this study. Intracellular adiponectin protein levels were measured by Western blot. As shown in Fig. 2A, adiponectin protein level was significantly reduced in adipocytes treated for 8 h with VPA (P < 0.05). Intracellular adiponectin protein levels were suppressed by 70–74% after 16 or 24 h treatment and reached their lowest levels (P < 0.001) (Fig. 2A). Interestingly, acute VPA treatment (1–4 h) tended to slightly increase adiponectin protein levels (P > 0.05). Adiponectin mRNA levels exhibited changes that paralleled changes in protein levels after treatment with 1 mM VPA for 8 h (Fig. 2B). Similar results were obtained using Chub-S7 human adipocytes (data not shown).
To test the dose-dependent effects of VPA, differentiated 3T3-L1 adipocytes were treated with different amounts of VPA for 16 h. Intracellular adiponectin protein levels were suppressed with increasing VPA concentration (Fig. 2C). These results indicate that at therapeutic concentrations VPA inhibits adiponectin gene expression in a time- and dose-dependent manner.
To determine whether VPA treatment alters adiponectin secretion, adiponectin levels in the medium were measured. There were significant reductions of adiponectin in the medium after 8 h VPA treatment. Adiponectin protein levels in the medium correlated well with intracellular levels during treatment with VPA (data not shown). Thus, VPA treatment does not appear to affect adiponectin secretion.
VPA inhibits adiponectin gene expression without inducing adipocyte dedifferentiation or apoptosis
The presence of VPA during adipocyte differentiation blocks adipogensis (36), raising the possibility that the VPA-reduced adiponectin expression is caused by adipocyte dedifferentiation or change of cellular phenotype. To address this possibility, triacylglycerol droplets were stained with Oil Red O and quantitatively analyzed in differentiated mature 3T3-L1 adipocytes that were treated with or without VPA (1 mM) for 12 or 24 h. TZD (1 μM) was used as positive control. As shown in Fig. 3A, there was no difference in the triacylglycerol droplet amounts between control and 12-h VPA-treated cells (P > 0.05). There was a small but statistically insignificant decrease after 24 h of VPA treatment. The study also showed that VPA treatment (1 mM, 24 h) slightly reduced the mRNA levels of perilipin in 3T3-L1 adipocytes, but without statistical difference (data not shown).
Although well tolerated by patients, VPA and its analogs inhibit cell proliferation and induce apoptosis in several cancer cell lines (1, 37, 38). The viability of VPA-treated adipocytes was examined using trypan blue exclusion. Surprisingly, exposure of mature 3T3-L1 adipocytes to 1 mM VPA did not reduce cell viability (Fig. 3B). Caspase-3/7 are cysteine proteases and are activated by proteolytic cleavage upon activation of the apoptotic cascade. Measuring casepase-3/7 activity has been used for studying apoptosis. After 24 h VPA treatment, there was no significant increase of caspase-3/7 activity compared with control cells (Fig. 3C). However, TNF treatment increased caspase-3/7 activity markedly. Taken together, these data suggest that VPA-inhibited adiponectin expression is not caused by dedifferentiation or induction of apoptosis.
Inhibition of HDACs by VPA appears to be required for suppressing adiponectin gene expression
VPA has recently been shown to inhibit a subset of HDACs (26, 27). To determine whether VPA-suppressed adiponectin expression in 3T3-L1 adipocytes was mediated by HDAC inhibition, a series of VPA analogs was used. Similar to VPA, treatment of 3T3-L1 adipocytes for 16 h with 3 nM TSA or 1 mM sodium butyrate decreased the intracellular levels of adiponectin protein (Fig. 4) and mRNA (data not shown). The concentrations of VPA, TSA, and sodium butyrate used in this study are within the HDAC IC50 ranges (39). Two VPA analogs, VPM and 2M2P, with minimal HDAC inhibitory effect (HDAC IC50 > 20 mM for VPM and HDAC IC50 = 15 mM for 2M2P) (39), were used to control for effects other than inhibition of HDACs. The concentrations of VPM (1 mM) and 2M2P (1 mM) used in this study have almost no HDAC inhibitory effect (39). As shown in Fig. 4, 16-h exposure to VPM or 2M2P did not reduce intracellular adiponectin protein levels in 3T3-L1 adipocytes. Total acetylated histone H4 was also measured. As expected, H4 acetylation was significantly increased in VPA, TSA, and sodium butyrate-treated cells but was not changed in VPM- or 2M2P-treated cells. Thus, these results suggest that inhibition of HDACs activity is correlated in VPA-induced down-regulation of adiponectin expression in mature adipocytes.
VPA inhibits adiponectin expression at the transcriptional level
As stated above, VPA reduced adiponectin mRNA level in adipocytes, and the inhibition was time and dose dependent. We also studied the effects of VPA on the stability of adiponectin mRNA. Treatment with 1 mM VPA did not alter the half-life of adiponectin mRNA (data not shown), suggesting that VPA inhibits adiponectin expression at the level of transcription. To further investigate whether the suppression of adiponectin gene expression by VPA occurs at the transcriptional level, a mouse adiponectin promoter-luciferase gene construct was created using pGl3-basic as vector. The gene reporter construct and transfection efficiency control pCMV--galactosidase were transiently transfected into 3T3-L1 fibroblasts or fully differentiated adipocytes (32). Twenty-four hours after transfection, the cells were treated with 1 mM VPA for 16 h. Luciferase activity was determined and normalized relative to -galactosidase activity. Similar to previous studies (32, 40), adiponectin promoter activity was 2-fold higher in differentiated adipocytes than in fibroblasts (Fig. 5, A and B). As shown in Fig. 5A, VPA increased adiponectin promoter activity in undifferentiated 3T3-L1 preadiocytes (P < 0.05). However, adiponectin promoter activity was significantly lower in VPA-treated differentiated 3T3-L1 adipocytes compared with vehicle control (P < 0.05, Fig. 5B). These results indicate that VPA decreases adiponectin gene expression at the transcriptional level and that the inhibition is adipocyte specific.
VPA suppresses adiponectin gene expression through down-regulating C/EBP
Because VPA inhibits adiponectin promoter activity in an adipocyte-specific manner, we conducted studies to determine whether the VPA-induced reduction of adiponectin in mature adipocytes results from changes in key adipogenic transcription factors. Nuclear protein was extracted from 3T3-L1 adipocytes and protein levels of C/EBP, PPAR, and SREBP1 were measured by Western blot. The results showed that, after 8 h of treatment with 1 mM VPA, both the 42- and 30-kDa C/EBP isoforms were reduced (Fig. 6A). Total C/EBP protein was reduced greater than 78% after 16 h treatment (Fig. 6A). PPAR and SREBP1 protein levels were detected using the same membranes. Interestingly, VPA treatment did not significantly change the protein levels of PPAR2, PPAR1, or cleaved SREBP1 in 3T3-L1 adipocytes (Fig. 6A). Because the SREBP1 antibody reacts with both SREBP1a and 1c, the Western results did not distinguish the isoforms.
The dependence of the reduction of C/EBP on VPA concentration was then examined. Exposure of 3T3-L1 adipocytes to 0.5 mM VPA for 16 h markedly reduced the levels of both the 42- and 30-kDa forms of C/EBP (Fig. 6B). Increasing reductions in the C/EBP isoforms were apparent with increasing VPA concentrations through 4 mM, the highest dose tested. PPAR1 and PPAR2 protein levels did not exhibit statistically significant change with VPA treatment at any tested concentration (Fig. 6B). These studies indicate that, similar to the effects on adiponectin expression, VPA treatment reduced C/EBP protein levels in 3T3-L1 adipocytes in a dose- and time-dependent manner.
Our previous study demonstrated that adiponectin gene transcription is well correlated with C/EBP protein levels in mature adipocytes (32). To determine whether the VPA-induced reduction in adiponectin expression is dependent on reductions in C/EBP, three experiments were conducted. In the first study, C/EBP protein was knocked down approximately 80% in 3T3-L1CAR1 adipocytes transduced with an adenovirus vector encoding an siRNA specific for C/EBP (32). The reduction in C/EBP protein was accompanied by a significant reduction in adiponectin protein that was similar to the reduction induced by VPA treatment (Fig. 7A).
In the second study, C/EBP gene-deficient (C/EBP–/–) mouse fibroblasts were stably transfected with C/EBP and/or PPAR2. Differentiation was induced following the protocol published previously (41). Consistent with our previous study, low levels of adiponectin protein were observed in cells expressing exogenous PPAR2 in the absence of C/EBP compared with adipocytes expressing both C/EBP and PPAR2 (Fig. 7B). Interestingly, treatment with 1 mM VPA did not suppress adiponectin protein levels in the adipocytes expressing PPAR2 alone (Fig. 7B). In contrast to the substantial reduction of adiponectin in 3T3-L1 adipocytes, VPA treatment led to only slightly reduced adiponectin in adipocytes that stably express C/EBP and PPAR2 (Fig. 7B). Furthermore, VPA treatment did not significantly reduce C/EBP protein in adipocytes that stably express C/EBP and PPAR2 (data not shown), which may due to the use of the exogenous retroviral long terminal repeat promoter to direct C/EBP expression. This result suggests that VPA treatment has no impact on exogenous promoter-driven C/EBP gene expression. The study also revealed that, consistent with the results from 3T3-L1 adipocytes, adiponectin protein levels correlated with C/EBP level in the adipocytes differentiated from mouse embryo fibroblasts.
A C/EBP binding site in the mouse promoter has recently been reported (42). Our group has recently identified a new C/EBP responding region in the mouse adiponectin promoter through which C/EBP acts as a coactivator to up-regulate adiponectin transcription (Qiao, L., and J. Shao, unpublished data). We next examined effects of VPA treatment on C/EBP binding to the adiponectin promoter region. Quantitative ChIP assay revealed that the binding of C/EBP at the proximal adiponectin promoter was significantly reduced by VPA treatment (P < 0.05), but bound acetyl-histone H3 and H4 were significantly increased (Fig. 8). Taken together, these data support the hypothesis that VPA-suppressed adiponectin expression is mediated by the reduction of C/EBP protein and impaired C/EBP binding to the adiponectin promoter in mature adipocytes.
Discussion
VPA has been used in the treatment of epilepsy and bipolar disorders for more than 30 yr. Clinical studies have demonstrated that VPA treatment usually induces obesity and insulin resistance. Adiponectin is an adipocyte-derived hormone that enhances insulin sensitivity and plays an important role in maintaining energy homeostasis. Numerous studies demonstrate that adiponectin gene expression inversely correlates with body weight. Therefore, it is reasonable to speculate that VPA-induced insulin resistance might occur through the down-regulation of adiponectin. Interestingly, a recent study has shown that VPA does not increase adipocyte differentiation in vitro (36). Instead, VPA blocks adipogenesis (36). VPA induces significant weight gain in rhesus monkeys but not rodents (43, 44, 45, 46). Furthermore, our study demonstrates that VPA actually decreases body weight and epididymal fat tissue, as well as adiponetin expression, in mice. Therefore, multiple mechanisms likely underlie VPA-induced weight gain and insulin resistance in humans. These studies suggest that VPA may induce human obesity through an indirect mechanism, rather than through direct effects on adipocyte differentiation.
Using an in vitro cell culture system, we show here that VPA directly inhibits adiponectin gene expression in differentiated adipocytes. Because the current study does not rule out the possibility of obesity-driven low adiponectin expression, we postulate that VPA-suppressed adiponectin gene expression is one of the mechanisms that contribute to VPA-induced insulin resistance. Unfortunately, there are no clinical data regarding whether VPA alters adiponectin expression in human subjects.
Adiponectin is exclusively expressed in adipocytes (6, 7). Adiponectin gene expression is turned on 2 d after the initiation of adipocyte differentiation and maintained at a relatively high level in mature adipocytes. Blockage of adipocyte differentiation reduces adiponectin expression. To avoid effects of VPA on differentiation (36), all of the cells used in this study were fully differentiated mature adipocytes and were treated beginning 8 d after initiation of adipocyte differentiation. The fact that treatment with VPA for 24 h did not alter triacylglycerol droplet content indicates that VPA treatment did not induce dedifferentiation of the adipocytes. Therefore, we conclude that VPA-suppressed adiponectin gene expression in mature adipocytes occurs through a mechanism distinct from that of the antiadipocyte differentiation effects of VPA.
VPA and certain of its analogs arrest growth and apoptosis in several types of cancer cells (1). Our study showed that the presence of 1 mM VPA in the cell culture medium, a level reached in patient serum during therapy of epilepsy with a daily dose of 20–30 mg/kg, did not induce significant loss of cell viability. In addition, Western blot studies showed that -actin levels were fairly stable and were not altered by VPA treatment of adipocytes. The mechanisms involved in VPA- or VPA analog-induced apoptosis are intricate and differ among cell types. Therapeutic concentrations of VPA do not induce apoptosis in most cell types (37, 38). Therefore, these results demonstrate that the inhibition of adiponectin gene expression by VPA in mature adipocytes is gene specific and not due to cellular toxicity or apoptosis.
Histone acetylation plays an important role in regulating transcription (25). Histone acetylation is dynamic and the level of acetylation is determined by the balance of the activities of acetyltransferases and HDACs. Recently, HDACs have been identified as direct targets of VPA (26, 27). VPA inhibits both class I and II HDACs, with a high potency for class I (26, 27, 39). Our study showed that VPA and its analogs with HDAC inhibitory effects suppress adiponectin expression. In contrast, the VPA analogs VPM and 2M2P, which do not inhibit HDAC activity, did not affect adiponectin protein levels in adipocytes (Fig. 4). Therefore, the inhibition of adiponectin gene expression by VPA is most likely mediated through HDAC inhibition. However, further studies are required to determine whether HDAC inhibition directly suppresses adiponectin gene expression.
In general, increased levels of histone acetylation by histone acetyltransferases loosens chromatin packaging and are correlated with transcriptional activation, whereas histone deacetylation by HDACs is associated with transcriptional repression (25). However, recent studies have demonstrated that histone deacetylation also associate with transcription activation of certain genes (47). VPA and other HDAC inhibitors increase histone acetylation universally by suppressing HDAC activity. VPA or TSA treatment increase transcription of genes including those encoding gelsolin, cytokeratin A, and c-fos (1). However, inhibition of HDAC activities also suppresses expression of certain genes including those encoding c-myc, IL-2, IL-8, and LC-pTP (1, 48, 49, 50). Inhibition of HDACs by TSA or trapoxin alters transcription, either up or down, of approximately 2% of genes (48). Therefore, HDAC inhibitors regulate gene expression either positively or negatively in a gene-specific manner. Our study showed that VPA treatment represses adiponectin gene transcription in mature adipocytes. Interestingly, our study also revealed that VPA increased mouse adiponectin promoter-driven luciferase expression in 3T3-L1 fibroblasts but decreased it in differentiated 3T3-L1 adipocytes. This finding suggests that, similar to most promoters tested, hyperacetylation of histones up-regulates adiponectin promoter activity. However, the adipocyte-specific inhibition of adiponectin promoter activity by VPA suggests that VPA treatment inhibits adiponectin gene expression through an indirect mechanism. Therefore, we hypothesized that VPA treatment might alter signal pathway(s) or transcription factor(s) and thereby inhibit adiponectin gene transcription in mature adipocytes.
C/EBP and PPAR are major adipogenic transcription factors that transactivate many adipocyte-specific genes. Our previous study and other studies have demonstrated that C/EBP is required to fully activate adiponectin gene expression (32, 41, 51). SREBP1c also plays a role in regulating adiponectin gene transcription. Our current study showed that VPA treatment significantly decreased C/EBP protein levels in differentiated adipocytes. However, VPA treatment did not alter protein levels of PPAR1, PPAR2, or SREBP1. Furthermore, VPA treatment did not alter adiponectin expression in C/EBP-deficient adipocytes that ectopicly express PPAR2 (Fig. 7B). These studies strongly support our hypothesis that by reducing the level of the adipogenic transcription factor C/EBP, VPA inhibits adiponectin gene transcription in adipocytes.
C/EBP-responsive elements that are associated with C/EBP and C/EBP have been identified in the mouse adiponectin promoter (42) (our unpublished data). While we were preparing this manuscript, a new study demonstrated that inhibition of HDACs increases C/EBP acetylation as well as histone acetylation (37). The study demonstrated that elevated acetylation of C/EBP inhibits its binding to target DNA sequences (37). Although we did not test the acetylation levels of C/EBP, our quantitative ChIP assay showed the binding of C/EBP to the adiponectin promoter region is significantly reduced in VPA-treated adipocytes. Therefore, we conclude that VPA inhibits adiponectin gene transcription in adipocytes by reducing the amount of C/EBP protein and its binding to the adiponectin promoter. In contrast to our study, Lagace et al. (24) reported that 3 h of VPA treatment transiently reduced PPAR and SREBP1a mRNA in adipocytes, but prolonged exposure (6 or 12 h) had no significant effect on protein or mRNA levels of C/EBP, PPAR, and SREBP1a. We do not have any evidence to explain the discrepancy but speculate that differences in the adipocyte differentiation procedure and differentiation stages of the adipocytes might contribute. The protein samples used in our study were nuclear extract while total cell lysis was used in the study by Lagace et al. Thus, differences in sample preparation may also contribute to the discrepant results. Strikingly, VPA treatment (1 mM, 12 h) significantly increases C/EBP protein levels in FAO hepatoma cells (Qiao, L., and J. Shao, unpublished data). These data indicate that VPA regulates C/EBP gene expression in a tissue-specific manner. Nevertheless, further studies are warranted to investigate the mechanism(s) whereby VPA represses C/EBP expression in mature adipocytes.
In summary, our results demonstrate that VPA suppresses adiponectin gene expression in mice and differentiated adipocytes, and that the suppression is correlated with the inhibition of HDAC activity of VPA. VPA-suppressed adiponectin gene expression occurs at the transcriptional level. We provide evidence that VPA suppresses adiponectin gene expression via down-regulation of C/EBP and its binding at the adiponectin promoter.
Acknowledgments
We thank Dr. Lisa A. Cassis (Graduate Center for Nutritional Sciences, University of Kentucky, Lexington, KY) for comments on the manuscript and Dr. Paul MacLean (Department of Medicine, University of Colorado Health Sciences Center, Aurora, CO) for helpful discussion.
Footnotes
This work was supported by grants from the American Diabetes Association (1-04-JF-44; to J.Sh.).
First Published Online November 10, 2005
Abbreviations: aP2, Adipocyte-specific fatty acid-binding protein; C/EBP, CCAAT/enhancer-binding protein; ChIP, chromatin immunoprecipitation; GFP, green fluorescent protein; HDAC, histone deacetylase; 2M2P, 2-methyl-2-pentenoic acid; PPAR, peroxisome proliferator-activated receptor ; siRNA, small interfering RNA; SREBP1, steroid regulatory element binding protein-1; TSA, trichostatin A; TZD, thiazolidinedione; VPA, valproic acid; VPM, valpromide.
Accepted for publication October 31, 2005.
References
Gurvich N, Klein PS 2002 Lithium and valproic acid: parallels and contrasts in diverse signaling contexts. Pharmacol Ther 96:45–66
Genton P, Bauer J, Duncan S, Taylor AE, Balen AH, Eberle A, Pedersen B, Salas-Puig X, Sauer MV 2001 On the association between Valproate and polycystic ovary syndrome. Epilepsia 42:295–304
Devlin MJ, Yanovski SZ, Wilson GT 2000 Obesity: what mental health professionals need to know. Am J Psychiatry 157:854–866
Corman C, Leung N, Guberman A 1997 Weight gain in epileptic patients during treatment with valproic acid: a retrospective study. Can J Neurol Sci 24:240–244
Verrotti A, Basciani F, De Simone M, Trotta D, Morgese G, Chiarelli F 2002 Insulin resistance in epileptic girls who gain weight after therapy with valproic acid. J Child Neurol 17:265–268
Scherer PE, Williams S, Fogliano M, Baldini G, Lodish HF 1995 A novel serum protein similar to C1q, produced exclusively in adipocytes. J Biol Chem 270:26746–26749
Hu E, Liang P, Spiegelman BM 1996 AdipoQ is a novel adipose-specific gene dysregulated in obesity. J Biol Chem 271:10697–10703
Maeda K, Okubo K, Shimomura I, Funahashi T, Matsuzawa Y, Matsubara K 1996 cDNA cloning and expression of a novel adipose specific collagen-like factor, apM1 (AdiPose most abundant gene transcript 1). Biochem Biophys Res Commun 221:286–289
Nakano Y, Tobe T, Choi-Miura NH, Mazda T, Tomita M 1996 Isolation and characterization of GBP28, a novel gelatin-binding protein purified from human plasma. J Biochem (Tokyo) 120:803–812
Kim S, Moustaid-Moussa N 2000 Secretory, endocrine and autocrine/paracrine function of the adipocyte. J Nutr 130:3110S–3115S
Kahn BB, Flier JS 2000 Obesity and insulin resistance. J Clin Invest 106:473–481
Diez JJ, Iglesias P 2003 The role of the novel adipocyte-derived hormone adiponectin in human disease. Eur J Endocrinol 148:293–300
Stefan N, Stumvoll M 2002 Adiponectin—its role in metabolism and beyond. Horm Metab Res 34:469–474
Berg AH, Combs TP, Du X, Brownlee M, Scherer PE 2001 The adipocyte-secreted protein Acrp30 enhances hepatic insulin action. Nat Med 7:947–953
Fruebis J, Tsao TS, Javorschi S, Ebbets-Reed D, Erickson MR, Yen FT, Bihain BE, Lodish HF 2001 Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci USA 98:2005–2010
Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, Yamashita S, Noda M, Kita S, Ueki K, Eto K, Akanuma Y, Froguel P, Foufelle F, Ferre P, Carling D, Kimura S, Nagai R, Kahn BB, Kadowaki T 2002 Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 8:1288–1295
Yamauchi T, Kamon J, Ito Y, Tsuchida A, Yokomizo T, Kita S, Sugiyama T, Miyagishi M, Hara K, Tsunoda M, Murakami K, Ohteki T, Uchida S, Takekawa S, Waki H, Tsuno NH, Shibata Y, Terauchi Y, Froguel P, Tobe K, Koyasu S, Taira K, Kitamura T, Shimizu T, Nagai R, Kadowaki T 2003 Cloning of adiponectin receptors that mediate antidiabetic metabolic effects. Nature 423:762–769
Tomas E, Tsao TS, Saha AK, Murrey HE, Zhang Cc C, Itani SI, Lodish HF, Ruderman NB 2002 Enhanced muscle fat oxidation and glucose transport by ACRP30 globular domain: acetyl-CoA carboxylase inhibition and AMP-activated protein kinase activation. Proc Natl Acad Sci USA 99:16309–16313
Arita Y, Kihara S, Ouchi N, Takahashi M, Maeda K, Miyagawa J, Hotta K, Shimomura I, Nakamura T, Miyaoka K, Kuriyama H, Nishida M, Yamashita S, Okubo K, Matsubara K, Muraguchi M, Ohmoto Y, Funahashi T, Matsuzawa Y 1999 Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 257:79–83
Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M, Okamoto Y, Iwahashi H, Kuriyama H, Ouchi N, Maeda K, Nishida M, Kihara S, Sakai N, Nakajima T, Hasegawa K, Muraguchi M, Ohmoto Y, Nakamura T, Yamashita S, Hanafusa T, Matsuzawa Y 2000 Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol 20:1595–1599
Statnick MA, Beavers LS, Conner LJ, Corominola H, Johnson D, Hammond CD, Rafaeloff-Phail R, Seng T, Suter TM, Sluka JP, Ravussin E, Gadski RA, Caro JF 2000 Decreased expression of apM1 in omental and subcutaneous adipose tissue of humans with type 2 diabetes. Int J Exp Diabetes Res 1:81–88
Tsao TS, Lodish HF, Fruebis J 2002 ACRP30, a new hormone controlling fat and glucose metabolism. Eur J Pharmacol 440:213–221
Verrotti A, Basciani F, Morresi S, de Martino M, Morgese G, Chiarelli F 1999 Serum leptin changes in epileptic patients who gain weight after therapy with valproic acid. Neurology 53:230–232
Lagace DC, McLeod RS, Nachtigal MW 2004 Valproic acid inhibits leptin secretion and reduces leptin messenger ribonucleic acid levels in adipocytes. Endocrinology 145:5493–5503
Struhl K 1998 Histone acetylation and transcriptional regulatory mechanisms. Genes Dev 12:599–606
Gottlicher M, Minucci S, Zhu P, Kramer OH, Schimpf A, Giavara S, Sleeman JP, Lo Coco F, Nervi C, Pelicci PG, Heinzel T 2001 Valproic acid defines a novel class of HDAC inhibitors inducing differentiation of transformed cells. EMBO J 20:6969–6978
Phiel CJ, Zhang F, Huang EY, Guenther MG, Lazar MA, Klein PS 2001 Histone deacetylase is a direct target of valproic acid, a potent anticonvulsant, mood stabilizer, and teratogen. J Biol Chem 276:36734–36741
Nau H 1985 Teratogenic valproic acid concentrations: infusion by implanted minipumps vs conventional injection regimen in the mouse. Toxicol Appl Pharmacol 80:243–250
Schank JR, Liles LC, Weinshenker D 2005 Reduced anticonvulsant efficacy of valproic acid in dopamine -hydroxylase knockout mice. Epilepsy Res 65:23–31
Orlicky DJ, DeGregori J, Schaack J 2001 Construction of stable coxsackievirus and adenovirus receptor-expressing 3T3-L1 cells. J Lipid Res 42:910–915
Ross SA, Song X, Burney MW, Kasai Y, Orlicky DJ 2003 Efficient adenovirus transduction of 3T3-L1 adipocytes stably expressing coxsackie-adenovirus receptor. Biochem Biophys Res Commun 302:354–358
Qiao L, MacLean PS, Schaack J, Orlicky DJ, Darimont C, Pagliassotti M, Friedman JE, Shao J 2005 C/EBP regulates human adiponectin gene transcription through an intronic enhancer. Diabetes 54:1744–1754
Shao J, Qiao L, Janssen RC, Pagliassotti M, Friedman JE 2005 Chronic hyperglycemia enhances PEPCK gene expression and hepatocellular glucose production via elevated liver activating protein/liver inhibitory protein ratio. Diabetes 54:976–984
Chakrabarti SK, James JC, Mirmira RG 2002 Quantitative assessment of gene targeting in vitro and in vivo by the pancreatic transcription factor, Pdx1. Importance of chromatin structure in directing promoter binding. J Biol Chem 277:13286–13293
Hazel M, Cooksey RC, Jones D, Parker G, Neidigh JL, Witherbee B, Gulve EA, McClain DA 2004 Activation of the hexosamine signaling pathway in adipose tissue results in decreased serum adiponectin and skeletal muscle insulin resistance. Endocrinology 145:2118–2128
Lagace DC, Nachtigal MW 2004 Inhibition of histone deacetylase activity by valproic acid blocks adipogenesis. J Biol Chem 279:18851–18860
Duan H, Heckman CA, Boxer LM 2005 Histone deacetylase inhibitors down-regulate bcl-2 expression and induce apoptosis in t(14;18) lymphomas. Mol Cell Biol 25:1608–1619
Sanjo Y, Yokoyama T, Sato S, Ikeda K, Nakajima R 1999 Ergonomic automated anesthesia recordkeeper using a mobile touch screen with voice navigtion. J Clin Monit Comput 15:347–356
Gurvich N, Tsygankova OM, Meinkoth JL, Klein PS 2004 Histone deacetylase is a target of valproic acid-mediated cellular differentiation. Cancer Res 64:1079–1086
Das K, Lin Y, Widen E, Zhang Y, Scherer PE 2001 Chromosomal localization, expression pattern, and promoter analysis of the mouse gene encoding adipocyte-specific secretory protein Acrp30. Biochem Biophys Res Commun 280:1120–1129
Gustafson B, Jack MM, Cushman SW, Smith U 2003 Adiponectin gene activation by thiazolidinediones requires PPAR2, but not C/EBP-evidence for differential regulation of the aP2 and adiponectin genes. Biochem Biophys Res Commun 308:933–939
Park S-k, Oh S-Y, Lee M-Y, Yoon S, Kim K-S, Kim J-w2004 CCAAT/enhancer binding protein and nuclear factor-Y regulate adiponectin gene expression in adipose tissue. Diabetes 53:2757–2766
Ferin M, Morrell M, Xiao E, Kochan L, Qian F, Wright T, Sauer M 2003 Endocrine and metabolic responses to long-term monotherapy with the antiepileptic drug valproate in the normally cycling rhesus monkey. J Clin Endocrinol Metab 88:2908–2915
Lagace DC, Nachtigal MW 2003 Valproic acid fails to induce polycystic ovary syndrome in female rats. Prog Neuropsychopharmacol Biol Psychiatry 27:587–594
Wolden-Hanson T, Gidal BE, Atkinson RL 1998 Evaluation of a rat model of valproate-induced obesity. Pharmacotherapy 18:1075–1081
Chapman JB, Cutler MG 1984 Sodium valproate: effects on social behaviour and physical development in the mouse. Psychopharmacology (Berl) 83:390–396
Nusinzon I, Horvath CM 2005 Histone deacetylases as transcriptional activators Role reversal in inducible gene regulation. Sci STKE 2005:re11
Van Lint C, Emiliani S, Verdin E 1996 The expression of a small fraction of cellular genes is changed in response to histone hyperacetylation. Gene Expr 5:245–253
Takahashi I, Miyaji H, Yoshida T, Sato S, Mizukami T 1996 Selective inhibition of IL-2 gene expression by trichostatin A, a potent inhibitor of mammalian histone deacetylase. J Antibiot (Tokyo) 49:453–457
Huang N, Katz JP, Martin DR, Wu GD 1997 Inhibition of IL-8 gene expression in Caco-2 cells by compounds which induce histone hyperacetylation. Cytokine 9:27–36
Park B-H, Qiang L, Farmer SR 2004 Phosphorylation of C/EBP at a consensus extracellular signal-regulated kinase/glycogen synthase kinase 3 site is required for the induction of adiponectin gene expression during the differentiation of mouse fibroblasts into adipocytes. Mol Cell Biol 24:8671–8680(Liping Qiao, Jerome Schaack and Jianhua )