|Home | About | Journals | Submit | Contact Us | Français|
Acetylation is one of the most crucial post-translational modifications that affect protein function. Protein lysine acetylation is catalyzed by acetyltransferases, and acetyl-CoA functions as the source of the acetyl group. Additionally, acetyl-CoA plays critical roles in maintaining the balance between carbohydrate metabolism and fatty acid synthesis. Here, we sought to determine whether lysine acetylation is an important process for adipocyte differentiation. Based on an analysis of the acetylome during adipogenesis, various proteins displaying significant quantitative changes were identified by LC-MS/MS. Of these identified proteins, we focused on malate dehydrogenase 1 (MDH1). The acetylation level of MDH1 was increased up to 6-fold at the late stage of adipogenesis. Moreover, overexpression of MDH1 in 3T3-L1 preadipocytes induced a significant increase in the number of cells undergoing adipogenesis. The introduction of mutations to putative lysine acetylation sites showed a significant loss of the ability of cells to undergo adipogenic differentiation. Furthermore, the acetylation of MDH1 dramatically enhanced its enzymatic activity and subsequently increased the intracellular levels of NADPH. These results clearly suggest that adipogenic differentiation may be regulated by the acetylation of MDH1 and that the acetylation of MDH1 is one of the cross-talk mechanisms between adipogenesis and the intracellular energy level.
Protein lysine acetylation is a posttranslational modification that acts as a key regulator of cellular processes. Histone deacetylases (HDACs) and histone acetyltransferases are enzymes that catalyze deacetylation and acetylation of the ϵ-amino groups of lysine residues of proteins. Lysine acetylation was first discovered to occur in histones, and histone acetylation is important in controlling the structure and function of chromatin (1). A large number of studies have further shown the existence of acetylated nonhistone proteins, including transcription factors, hormone receptors, signal transducers, and chaperones. The reversible acetylation of nonhistone proteins modulates a wide variety of key cellular processes, such as apoptosis, survival, and proliferation. Recently, several groups have reported that many metabolic enzymes are highly acetylated (2–5). Such enzymes are involved in glycolysis, fatty acid metabolism, gluconeogenesis, the TCA cycle, and the urea cycle, and the acetylation of these proteins regulates their activity so that they can respond to the metabolic demands of cells.
Adipose tissue, which consists of loose connective tissue composed of adipocytes, is an important metabolic organ that functions in energy homeostasis (6). Adipocytes regulate physiologic processes, including glucose metabolism, angiogenesis, the inflammatory response, and reproductive functions through secreted adipokines. Adipocyte differentiation can contribute to the development of obesity via a positive energy balance (energy intake > energy expenditure) (7). Obesity leads to serious health problems all over the world. The coexistence of obesity, type II diabetes, dyslipidemia, and hypertension, known as metabolic syndrome, constitutes an increased risk for the development of cardiovascular diseases (8). In adipocytes, fatty acids are stored as triglycerides, which serve as the fuel for maintaining energy balance. Acetyl-CoA is the direct precursor in the synthesis of fatty acids, and all fatty acid carbons come from the acetyl group of acetyl-CoA. In addition, it has been reported that the adipogenic differentiation could be differentially regulated by HDAC inhibitors (9–11). These facts led us to hypothesize that the acetylation of proteins plays a key role in controlling adipogenesis.
Malate dehydrogenase (MDH) catalyzes the interconversion of malate and oxaloacetate in the mitochondrial membrane using the coenzyme NAD+/NADH (12). In eukaryotic cells, two MDH proteins exist and are classified according to their subcellular localization as MDH1 or MDH2 (13). In the mitochondrial matrix, MDH2 oxidizes malate to oxaloacetate through the citric acid cycle. The cytosolic form of the protein (MDH1) participates in the malate/aspartate shuttle. The shuttle allows for the overall balancing of the import and export of nitrogen, oxaloacetate, and the α-ketoglutarate intermediate in the cytosol and mitochondrial matrix (14). The malate required for antiport exchange can be exported by citrate, which accumulates in large amounts, instead of isocitrate. Accumulated citrate is transported from the mitochondria to the cytosol. Then, oxaloacetate and acetyl-CoA are synthesized in the cytosol. Cytosolic oxaloacetate is then converted to malate by MDH1, and acetyl-CoA is used for fatty acid synthesis. In mammals, MDH1 is highly expressed in many tissues, and its different roles are determined by its expression levels (15–17).
In this study, we attempted to determine whether lysine acetylation is an important process for adipocyte differentiation. Based on an analysis of the acetylome during adipogenesis, a number of proteins displaying significant quantitative changes were identified using LC-MS/MS. Among these proteins, we focused on MDH1. The acetylation level of MDH1 is significantly increased during adipogenic differentiation. Interestingly, the overexpression of MDH1 in 3T3-L1 preadipocytes induces enhanced adipogenesis in these cells. The introduction of mutations to putative lysine acetylation sites in MDH1 reduces the ability of the enzyme to promote adipogenic differentiation. Additionally, the acetylation of MDH1 significantly increases its enzymatic activity. These data suggest that the lysine acetylation of MDH1 is involved in the regulation of adipogenic differentiation.
The preadipocyte cell line, 3T3-L1, which is derived from mouse embryonic fibroblasts, was purchased from ATCC. The cells were cultured in growth medium (high glucose DMEM containing a 1% antibiotic-antimycotic solution and 10% bovine calf serum; Gibco-Invitrogen, Carlsbad, CA) at 37°C in a humidified atmosphere with 5% CO2. GP2-293 packaging cells were grown in DMEM containing a 1% antibiotic-antimycotic solution and 10% FBS. 3T3-L1 cells were induced to differentiate into mature adipocytes as described in our previous reports (18–21). Confluent 3T3-L1 cells were incubated in differentiation medium composed of DMEM, 10% FBS, and MDI (a differentiation cocktail of 0.5 mM 3-isobutyl-1-methylxanthine, 1 μM dexamethasone, and 10 μg/ml insulin [Sigma, US]). After 48 h, the medium was changed to a maintenance medium composed of DMEM, 10% FBS, and 10 μg/ml insulin. The medium was replenished every other day.
Cultured cells were washed twice with PBS and fixed with 10% formalin for 30 min at room temperature. The cells were then washed with distilled water and stained for 30 min at room temperature with 0.3% filtered Oil-Red-O solution in 60% isopropanol (Sigma, St. Louis, MO). The stained cells were washed with distilled water, and micrographs were obtained. To extract the incorporated Oil-Red-O dye, absolute isopropanol was added to the stained cell-culture dish, and the dish was shaken at room temperature for 30 min. Triplicate samples were read at 510 nm using a GeneQuant 1300 spectrophotometer (GE HealthCare, Uppsala, Sweden) (20, 21).
The proteins were extracted from 3T3-L1 cells using a lysis buffer (30 mM Tris [pH 8.5], 7 M urea, 2 M thiourea, and 4% 3-[(3-cholamidopropyl)-dimethylammonio propanesulfonate [CHAPS]). Protein samples to be separated on the same gel were pooled, and 2 volumes of cold acetone were added. After centrifugation, the precipitate was rehydrated using a rehydration buffer (7 M urea, 2 M thiourea, 4% CHAPS, 2% ampholyte, and 20 mM DTT), and isoelectric focusing was conducted using a Multiphor II apparatus (GE HealthCare). After focusing, each strip (a 7 cm immobilized pH nonlinear strip [pH 3–11]) was equilibrated twice for 15 min in 2.5 ml of equilibrium buffer. The equilibrium buffer contains 50 mM Tris (pH 8.8), 6 M urea, and 30% glycerol. During the second equilibrium step, 260 mM iodoacetamide was added to the equilibrium buffer. The IPG strips were then loaded and separated on a 10% acrylamide SDS-PAGE gel using a Mini-Protean system (Bio-Rad, Hercules, CA).
Sixty micrograms of total protein was used for the detection of lysine acetylated proteins during adipogenic differentiation. The samples were separated in the first and second dimension following previously described protocols (22, 23) and transferred to an Immunoblot PVDF membrane (Millipore). Membranes were blocked with 5% skim milk for 1 h and probed with a rabbit anti-acetyl-lysine antibody (ICP0380 Immunechem) diluted 1:2,000 in TBS with 5% skim milk for 4 h at 4°C as previously described (21).
The separated proteins in SDS-PAGE gels were visualized by Coomassie brilliant blue G-250 staining, and spots of interest were detected. Gel pieces were washed twice with 150 μl of 100 mM ammonium bicarbonate (pH 8.2) and 70% v/v acetonitrile and dried at 37°C for 20 min. Trypsin in 50 mM ammonium bicarbonate (20 μg/μl) was added to each gel piece, and gel pieces were incubated at 37°C for 2 h. Peptides were then extracted using a mixture containing 20 μl of 0.1% v/v trifluoroacetic acid and 70% acetonitrile. The peptides were purified using ZipTipsTM (Millipore) according to the manufacturer's instructions before LC-MS/MS spectrometric analysis. The peptides were analyzed using a Synapt High Definition mass spectrometer (Waters, Manchester, UK) as previously described (24, 25).
To construct 3T3-L1 cells that stably express a FLAG-tagged wild-type or mutant MDH1 protein, a retroviral infection system was used. For the expression of MDH1, DNA encoding the FLAG-tagged MDH1 was inserted into 3T3-L1 cells using the pRetroX-IRES-ZsGreen1 vector (Clontech). For virus production, GP2-293 cell lines were transfected using Lipofectamine 2000 (Gibco-Invitrogen). The details of the transfection and transduction methods have been described in our previous reports (18–21, 26). Infected cells were selected using a FACSAria cell sorter (BD Biosciences, San Jose, CA) and further maintained in growth medium. The ectopic expression of MDH1 was confirmed by Western blot analysis.
FLAG-tagged MDH1 was mutated using the EZchangeTM Site-Directed Mutagenesis kit (Enzynomics, KR). The putative acetylation sites in MDH1 were mutated to arginine or glutamine. Arginine is charged and abolishes acetylation, whereas glutamine is uncharged and may mimic the acetylated lysine. The sites mutated in this study were based on those reported by Zhao et al. (K118, K121, and K298) (5). We introduced the mutation into each site individually or at all three sites of MDH1.
Sequences from the wild-type and mutant MDH1 gene were amplified using the forward primer 5′- gggaattccatatgatgtctgaaccaatcagagtc-3′ containing an NdeI site (in bold) and the reverse primer 5′-ccgctcgagttaggcagaggaaagaaattcaaa-3′ containing a XhoI site (in bold) and inserted into the corresponding sites of the pET28a plasmid (Novagen, Madison, WI). Escherichia coli Rosetta DE3 cells (Novagen) carrying the respective expression plasmids were grown at 37°C in LB medium until the optical density at A600 reached 0.6–0.8. The expression of MDH1 and MDH1 mutants was induced by adding 1 mM IPTG at 18°C for 18 h. The cells were harvested, washed with lysis buffer (50 mM Tris [pH 7.5], 250 mM NaCl, 5% glycerol, and 1 mM β-mercaptoethanol), and lysed by ultrasonication. After centrifugation (29,820 g for 30 min), the supernatants were incubated with a cobalt affinity resin (TALON®, Clontech) on a rocker for 1 h at 4°C and washed with lysis buffer supplemented with 10 mM imidazole. The proteins were eluted from the metal affinity resin using a lysis buffer supplemented with 100 mM imidazole. The proteins were concentrated to 5 mg/ml and stored at −80°C.
The cells were washed three times with ice-cold PBS and were harvested in ice-cold 7 M urea lysis buffer containing a protease inhibitor cocktail (Roche). The protein concentrations were measured using a Bradford assay (Bio-Rad). Immunoblotting and immunoprecipitation were performed as described previously. The anti-MDH1 antibody was obtained from Santa Cruz, and the acetyl-lysine antibody was purchased from ImmuneChem Pharmaceuticals Inc. Anti-FLAG and anti-α-tubulin antibodies were purchased from Sigma. Anti-enolase1 and anti-histone H3 antibodies were obtained from Cell Signaling (Beverley, MA). The secondary antibodies were purchased from Abcam. The membranes were visualized using the SuperSignal West Pico Chemiluminescent Substrate kit (Pierce, Rockford, IL).
For immunoprecipitation, anti-acetyl-lysine antibody agarose beads (ImmuneChem Pharmaceuticals Inc.) were added to 200 μg of the lysates. The mixture was incubated overnight at 4°C. The immunoprecipitates were recovered by centrifugation at 2,500 g, washed five times with NP-40 lysis buffer, and analyzed as described above.
Cell pellets were lysed with NP-40 buffer supplemented with a protease inhibitor (Roche, Indianapolis, IN) and a KDAC inhibitor (10 µM trichostatin A, 10 mM nicotinamide, and 50 mM butyric acid; Sigma, US) (5). For immunoprecipitation, extracted proteins were incubated with anti-FLAG M2 agarose beads (Sigma) at 4°C overnight. The immunoprecipitated beads were washed 10 times with NP-40 buffer, and then the FLAG fusion proteins were eluted using a 3×FLAG peptide (Sigma). The concentration of the eluted proteins was measured using a Bradford assay and Coomassie blue G-250 staining. Purified human MDH1 that was overexpressed in 3T3-L1 cells was incubated with 0.2 mM oxaloacetate and 1 mM NADH in PBS (pH 7.5). The activity of MDH1 was determined by measuring the decrease in the fluorescence of NADH (Ex. 350 nm; Em. 470 nm). The reaction was monitored 15 times at 1-min intervals on a VictorTM X3 Multilabel plate reader (PerkinElmer, Waltham, MA).
Commercial NADP/NADPH assay kits (Abcam, Cambridge, UK) were used to detect NADPH and total NADP. The cells were washed three times with ice-cold PBS and placed in NADP/NADPH extraction buffer. The washed cells were extracted by two freeze (20 min on dry-ice)/thaw (10 min at room temperature) cycles, and then the cell lysates were dissolved in buffer using gentle sonication. The concentration of the eluted proteins was measured using a Bradford assay (Bio-Rad). To calculate the NADP/NADPH ratio, total NADP (NADPt) and NADPH-only were prepared, and an NADPH standard was made using a serial dilution of NADPH. The NADPH reaction was monitored on a microplate reader (Bio-Rad) for 1 to 4 h at 450 nm.
Many groups have reported that the adipogenic differentiation process could be differentially regulated depending on the type of HDAC inhibitors, suggesting that acetylation may be intimately associated with adipogenesis (9–11). As an initial step, 1D SDS-PAGE was conducted to determine whether acetylation patterns changed during adipogenesis (Fig. 1A). Whole cell proteins from 3T3-L1 cells derived from 0, 2, and 12 days after the induction of differentiation were used for analysis. It was confirmed that the level of aP2 was increased after adipogenic stimulation (Fig. 1B). The expression level of histone H3 was unchanged during differentiation, and it was thus used as a loading control. Our results clearly demonstrated that the acetylation pattern and protein expression were significantly altered during the differentiation of 3T3-L1 preadipocytes into adipocytes. Next, we performed two-dimensional gel electrophoresis (2-DE). The expression pattern of proteins with isoelectric points ranging between pH 3 and 11 was analyzed. The overall changes in the acetylation pattern (acetylome) were also investigated by Western blot analysis using an anti-acetyl-lysine antibody after 2-DE (Fig. 1C). As shown, the levels of lysine-acetylated proteins were generally increased after adipogenic differentiation. A total of 69 spots showing changes in acetylation during adipogenesis were detected by image analysis and visual confirmation; these spots were identified by LC-MS/MS analysis (Table 1). As expected, a variety of metabolic enzymes, including those involved in glycolysis, fatty acid metabolism, and the TCA cycle, were acetylated during adipogenesis.
To validate the results of the acetylome analysis, we assessed the expression and acetylation level of several of the identified proteins in 3T3-L1 cells using Western blot and immunoprecipitation analyses, respectively (Fig. 2A). MDH1, one of the identified candidates, showed an approximately 2-fold increase in expression during adipogenesis. The acetylation level of MDH1 was also dramatically enhanced (6-fold) during adipogenic differentiation (Fig. 2B). MDH2 also demonstrated this enhanced acetylation pattern, but enolase showed no changes in expression or acetylation level during adipogenesis. Generally, the Western blot and immunoprecipitation results correlated well with the 2-DE acetylome data (data not shown).
To examine if the acetylation of MDH1 occurred during adipogenesis, we assessed the expression and acetylation levels of MDH1 in other differentiation processes, such as myogenesis and osteogenesis. There was no significant change in expression and acetylation levels during myogenesis and osteogenesis (data not shown). These results suggest that the acetylation of MDH1 occurred in adipogenic differentiation. Next, to clarify the role of MDH1 acetylation in adipogenesis, 3T3-L1 cells were infected with the FLAG-tagged, full-length human MDH1 protein using a retroviral infection system (IRES-GFP). A control vector was used as the negative control. Infected cells were isolated by a FACS sorter, and most of the cells were GFP positive by fluorescence microscopy (Fig. 3A). The overexpression of MDH1 proteins was confirmed by Western blot analysis using anti-FLAG and anti-MDH1 antibodies (Fig. 3B). Infected cells were induced to differentiate into mature adipocytes, and after 8 days, lipid accumulation was assessed by Oil-Red-O staining of control vector- and MDH1-expressing cells (Fig. 3C). The results clearly showed that the overexpression of MDH1 significantly increased lipid accumulation when compared with the control vector (Fig. 3C, D). Consistently, the level of aP2 in the MDH1-infected cells was higher than that in the control vector-expressing cells (Fig. 3E). In addition, the expression levels of adipogenic markers, such as C/EBPα and PPARγ, were also increased upon ectopic expression of MDH1.
To examine the effect of MDH1 acetylation on adipogenesis, putative acetylation sites in MDH1 were mutated to arginine, which is charged but blocks acetylation, or to glutamine, which is uncharged and may mimic the acetylated lysine (2, 4, 5, 27). K118, K121, and K298 were chosen for mutation to arginine or glutamine. 3T3-L1 cells were then infected with these mutants of MDH1 using a retroviral system (IRES-GFP). The control vector was used as a negative control, and wild-type MDH1 was used as a positive control. Infected cells were isolated using a FACS sorter. Most of the infected cells were GFP positive by fluorescence microscopy (Fig. 4A). The overexpression of the mutated MDH1 proteins was confirmed by Western blot analysis using anti-FLAG and anti-MDH1 antibodies (Fig. 4B). Infected cells were cultured in differentiation medium for 8 days, and lipid accumulation was assessed by Oil-Red-O staining. Among the MDH1 mutants, the 3KR (in which the three putative acetylation lysine residues were replaced with arginine) and K118R mutants showed a significantly lower level of Oil-Red-O staining than the cells infected with wild-type MDH1 (Fig. 4C, D). These results clearly demonstrate that the mutation of the acetylation site(s) influences adipogenic differentiation, indicating that the acetylation of MDH1 is intimately involved in controlling adipogenic differentiation. We confirmed the continuous overexpression of MDH1 until the late stage of adipogenic differentiation. In addition, we found that differences in adipogenic differentiation correlated well with the expression level of aP2 (Fig. 4E, F). The acetylation levels of 3KR and K118R were considerably lower than those of wild-type MDH1 (Fig. 4G, H). Next, we measured the mRNA level of adipogenic markers, such as adiponectin, resistin, C/EBPα, and PPARγ, using real-time PCR. As expected, the expression levels of these adipogenic markers were significantly increased by introducing the MDH1. Additionally, 3T3-L1 cells carrying 3KR or K118R showed the reduced expression levels of adipogenic markers, compared with wild-type MDH1 (Fig. 4I). To rule out the possibility that the mutation of MDH1 affects its enzymatic activity, we expressed and purified the wild-type and mutant MDH1 proteins (3KR and K118R) using the E. coli expression system. Mutant MDH1s have similar activity to wild-type MDH1, indicating that the introduction of the mutation has no effect on enzymatic activity (data not shown).
An immunoblot analysis showed that the acetylation level of MDH1 increased during adipogenesis (Fig. 5A). Generally, acetylation modulates enzymatic function by altering enzymatic stability, catalytic activity, or protein-protein interactions. To explore the direct effect of acetylation on MDH1 activity, we purified the ectopically expressed wild-type MDH1 during adipogenesis (day 0 and day 8). MDH1 activity was increased by more than approximately 50% 8 days after differentiation when compared with the activity at day 0 (Fig. 5B). We then investigated the activity of the ectopically expressed wild-type and mutant MDH1 proteins after immunopurification. MDH1-3KR and MDH1-K118R showed significantly reduced enzymatic activity when compared with the wild-type and KQ mutant (Fig. 5C, D). Furthermore, the acetylation level of MDH1-3KR and MDH1-K118R considerably decreased when compared with wild-type MDH1 (Fig. 5E). These results clearly indicate that the acetylation level of MDH1 is enhanced during adipogenesis and that the enhanced acetylation of MDH1 increases its enzymatic activity.
Intracellular NADPH is an important source of reducing power for fatty acid synthesis. As mentioned above, MDH1 participates in the malate/aspartate shuttle that contributes the supply of acetyl-CoA and NADPH for fatty acid synthesis in the cytoplasm. To elucidate the mechanism of action of acetylated MDH1 on adipogenic differentiation, the NADP/NADPH ratio was assessed in cells 8 days after differentiation. The NADP/NADPH ratio was considerably decreased when wild-type MDH1 was overexpressed (Fig. 5F, G). In contrast, MDH1-3KR and MDH1-K118R showed a roughly similar ratio of NADP/NADPH to the control vector (Fig. 5F, G). These results strongly indicate that there is a relationship between the NADP/NADPH ratio and the enzymatic activity of MDH1, which is regulated by acetylation during adipogenesis.
Recent studies have suggested the existence of nonhistone protein acetylation. More than 2,000 acetylated proteins have been identified by mass spectrometric analysis in mammalian cells (2–5). The acetylated proteins are highly conserved in human and mouse tissues. The preferentially acetylated proteins participate in intermediate metabolic processes, such as glycolysis, gluconeogenesis, the TCA cycle, the urea cycle, fatty acid metabolism, nitrogen metabolism, and glycogen metabolism. Thus, it has been suggested that acetylation plays a key role in metabolic regulation. Protein acetylation is catalyzed by acetyltransferases that use acetyl-CoA as the acetylation donor. It has been shown that changes in the availability of acetyl-CoA can directly affect the acetylation status of critical substrates (5). In cells, acetyl-CoA enters the TCA cycle or is used to synthesize fatty acids depending on the energy requirements of the cell. In the case of fatty acid synthesis, acetyl-CoA is converted to malonyl-CoA and donates two carbon units (28). Therefore, it is speculated that there are interconnections between adipogenesis and protein lysine acetylation.
Although there is evidence to support the acetylation of histone and nonhistone proteins during adipogenesis (29–31), extensive studies have not been performed. In the present study, we conducted an acetylome analysis to identify the changes in the acetylation level of nonhistone proteins during adipocyte differentiation. The pattern of acetylated proteins is significantly different between preadipocytes and mature adipocytes (supplementary ). A total of 69 spots were detected by 2-DE Western blot analysis and identified by LC-MS/MS analysis (Table 1). Most of the identified acetylated proteins were involved in metabolic pathways, such as glycolysis, the TCA cycle, and fatty acid synthesis. In conjunction with earlier reports, these results suggest that acetylation is an extensive regulator of adipogenic differentiation (32–34). A role for acetylation in lipid metabolism can be inferred by the acetyl-CoA synthetase (30, 35), the bifunctional enzyme, enoyl CoA hydrolase/3-hydroxyl CoA dehydrogenase (2, 5, 36), SIRT3 (31), and long-chain acyl CoA dehydrogenase (37).
MDH1, one of the identified proteins that showed an altered acetylation pattern, is known to be a regulator that is required for calorie restriction (38). Thus, we focused on MDH1, which catalyzes the conversion of oxaloacetate into malate in the cytoplasm (13). Under our experimental conditions, the expression level of MDH1 was increased approximately 2-fold during adipocyte differentiation, and its acetylation level showed up to a 6-fold increase. This result indicates that the acetylation of MDH1 may play a crucial role in adipogenic differentiation. The ectopic expression of MDH1 in 3T3-L1 preadipocytes was associated with the induction of adipogenesis. Under our experimental conditions, approximately 12 days are needed for the cells to fully differentiate into adipocytes after the initiation of differentiation. However, the overexpression of MDH1 induced accelerated differentiation of the cells such that differentiation was completed only 8 days after differentiation was initiated. Several recent reports have identified possible acetylation sites in MDH1 by mass spectrometric analysis of mammalian cells (2, 4, 5). We thus further investigated the role of the acetylation of MDH1 in adipogenesis using MDH1 mutants in which putative MDH1 acetylation sites were replaced by arginine or glutamine. In previous studies, substituting glutamine for a lysine residue was found to mimic the acetylated form (39), whereas substituting arginine was found to mimic the deacetylated form of Foxo1 and MDH2 (5, 32, 40–42). Three acetylated lysine residues (K118, K121, and K298) were identified in MDH1 (5). Among the mutant MDH1 proteins that we constructed, only MDH1-3KR and K118R significantly suppressed the adipogenic differentiation of 3T3-L1 cells. These results show that a specific site (or sites) of MDH1 is acetylated during adipogenic differentiation and that the acetylation of MDH1 directly regulates the lipid metabolic pathway. To determine the direct effect of acetylation on enzymatic activity, we first confirmed the increased acetylation level of MDH1 during adipogenic differentiation. Consistent with endogenous MDH1 during adipogenesis, the acetylation level of ectopic MDH1 showed up to a 50% increase during adipogenesis, and this correlated with a dramatic increase in enzymatic activity. In contrast, KR mutants of MDH1 showed a decreased acetylation level and reduced enzymatic activity when compared with wild-type MDH1. Remarkably, the acetylation of the lysine 118 residue alone can control MDH1 activity during adipogenesis. Additionally, the acetylation of MDH1 regulates the intracellular ratio of NADP/NADPH during fatty acid synthesis. Adipose tissue uses glucose to produce triglyceride, which is the major form of fat (43). Glucose enters the glycolytic pathway to generate pyruvate and ATP. Pyruvate is converted to acetyl-CoA, which is used as citrate with oxaloacetate in the mitochondria. In adipose tissue, citrate is transported to the cytoplasm by the citrate shuttle when its concentration is high in the mitochondria. The cytoplasmic citrate disintegrates into acetyl-CoA and oxaloacetate. This reaction provides cytoplasmic acetyl-CoA that is used for fatty acid synthesis. Cytoplasmic oxaloacetate is converted to malate by MDH1, and the malate then rejoins the glycolytic pathway in a pyruvate form. NADPH is necessary for fatty acid synthesis as a reducing equivalent. Thus, we propose that MDH1 is involved in the citrate shuttle that produces two NADPH molecules per one cycle (44, 45) and that the activity of MDH1 is modulated by Lys-118 acetylation during adipogenesis. The pentose phosphate pathway also generates NADPH from glucose in most species. It has been reported that the majority of NADPH production in horse adipose tissue is due to the specific activity of glucose-6-phosphate dehydrogenase and 6-phospho-gluconate dehydrogenase (46). In contrast, the generation of NADPH from the malate cycle is higher than its generation from the pentose phosphate pathway in adipose tissue (47). Additionally, we speculated that adipogenesis as well as lipogenesis may be up-regulated by NADPH-dependent transcriptional regulators after increased intracellular NADPH level (48). In summary, depending on the cellular energy requirements, acetyl-CoA enters the TCA cycle or is used to synthesize fatty acids. When the cellular energy level is sufficient, excess acetyl-CoA is used to acetylate MDH1. The acetylation of MDH1 increases its enzymatic activity, and activated MDH1 supports acetyl-CoA and NADPH in fatty acid synthesis by accelerating the citrate shuttle (Fig. 6). In agreement with earlier reports, the present results indicate that the acetylation of MDH1 functions as a cross-talk mechanism between adipogenesis and the intracellular energy level.
The authors thank Professors Sayeon Cho and Byoung Chul Park for their continuous encouragement and helpful advice and Drs. Yoo-Sook Cho, Jeong-Ki Min, Doo Byoung Oh, and Seung-Wook Chi for carefully reading the manuscript and providing insightful comments.
This work was supported by grants from KRIBB, the Korea Research Council of Fundamental Science and Technology (NAP project), and the Research Program (grant nos. 2011-0027634, 2011-0027796 and 2011-0030028) of the Korea National Research Foundation.