Mitochondrial Glycerol Phosphate Acyltransferase Directs the Incorporation of Exogenous Fatty Acids into Triacylglycerol
2001; Elsevier BV; Volume: 276; Issue: 45 Linguagem: Inglês
10.1074/jbc.m103386200
ISSN1083-351X
AutoresR. Ariel Igal, Shuli Wang, María R. González-Baró, Rosalind Coleman,
Tópico(s)Adipose Tissue and Metabolism
ResumoThe mitochondrial isoform of glycerol-3-phosphate acyltransferase (GPAT), the first step in glycerolipid synthesis, is up-regulated by insulin and by high carbohydrate feeding via SREBP-1c, suggesting that it plays a role in triacylglycerol synthesis. To test this hypothesis, we overexpressed mitochondrial GPAT in Chinese hamster ovary (CHO) cells. When GPAT was overexpressed 3.8-fold, triacylglycerol mass was 2.7-fold higher than in control cells. After incubation with trace [14C]oleate (∼3 μm), control cells incorporated 4.7-fold more label into phospholipid than triacylglycerol, but GPAT-overexpressing cells incorporated equal amounts of label into phospholipid and triacylglycerol. In GPAT-overexpressing cells, the incorporation of label into phospholipid, particularly phosphatidylcholine, decreased 30%, despite normal growth rate and phospholipid content, suggesting that exogenous oleate was directed primarily toward triacylglycerol synthesis. Transiently transfected HEK293 cells that expressed a 4.4-fold increase in GPAT activity incorporated 9.7-fold more [14C]oleate into triacylglycerol compared with control cells, showing that the effect of GPAT overexpression was similar in two different cell types that had been transfected by different methods. When the stable, GPAT-overexpressing CHO cells were incubated with 100 μm oleate to stimulate triacylglycerol synthesis, they incorporated 1.9-fold more fatty acid into triacylglycerol than did the control cells. Confocal microscopy of CHO and HEK293 cells transfected with the GPAT-FLAG construct showed that GPAT was located correctly in mitochondria and was not present elsewhere in the cell. These studies indicate that overexpressed mitochondrial GPAT directs incorporation of exogenous fatty acid into triacylglycerol rather than phospholipid and imply that (a) mitochondrial GPAT and lysophosphatidic acid acyltransferase produce a separate pool of lysophosphatidic acid and phosphatidic acid that must be transported to the endoplasmic reticulum where the terminal enzymes of triacylglycerol synthesis are located, and (b) this pool remains relatively separate from the pool of lysophosphatidic acid and phosphatidic acid that contributes to the synthesis of the major phospholipid species. The mitochondrial isoform of glycerol-3-phosphate acyltransferase (GPAT), the first step in glycerolipid synthesis, is up-regulated by insulin and by high carbohydrate feeding via SREBP-1c, suggesting that it plays a role in triacylglycerol synthesis. To test this hypothesis, we overexpressed mitochondrial GPAT in Chinese hamster ovary (CHO) cells. When GPAT was overexpressed 3.8-fold, triacylglycerol mass was 2.7-fold higher than in control cells. After incubation with trace [14C]oleate (∼3 μm), control cells incorporated 4.7-fold more label into phospholipid than triacylglycerol, but GPAT-overexpressing cells incorporated equal amounts of label into phospholipid and triacylglycerol. In GPAT-overexpressing cells, the incorporation of label into phospholipid, particularly phosphatidylcholine, decreased 30%, despite normal growth rate and phospholipid content, suggesting that exogenous oleate was directed primarily toward triacylglycerol synthesis. Transiently transfected HEK293 cells that expressed a 4.4-fold increase in GPAT activity incorporated 9.7-fold more [14C]oleate into triacylglycerol compared with control cells, showing that the effect of GPAT overexpression was similar in two different cell types that had been transfected by different methods. When the stable, GPAT-overexpressing CHO cells were incubated with 100 μm oleate to stimulate triacylglycerol synthesis, they incorporated 1.9-fold more fatty acid into triacylglycerol than did the control cells. Confocal microscopy of CHO and HEK293 cells transfected with the GPAT-FLAG construct showed that GPAT was located correctly in mitochondria and was not present elsewhere in the cell. These studies indicate that overexpressed mitochondrial GPAT directs incorporation of exogenous fatty acid into triacylglycerol rather than phospholipid and imply that (a) mitochondrial GPAT and lysophosphatidic acid acyltransferase produce a separate pool of lysophosphatidic acid and phosphatidic acid that must be transported to the endoplasmic reticulum where the terminal enzymes of triacylglycerol synthesis are located, and (b) this pool remains relatively separate from the pool of lysophosphatidic acid and phosphatidic acid that contributes to the synthesis of the major phospholipid species. glycerol-3-phosphate acyltransferase acyl-CoA, cholesterol acyltransferase bovine serum albumin Chinese hamster ovary diacylglycerol acyltransferase enhanced green fluorescent protein fetal bovine serum fluorescein isothiocyanate lysophosphatidic acid minimal essential medium peroxisome proliferator activation receptor phosphate-buffered saline sterol regulatory element-binding protein thin layer chromatography 1,4-piperazinediethanesulfonic acid Excess accumulation of triacylglycerol in adipocytes is a major factor contributing to the current epidemic of obesity-related disorders, including type 2 diabetes, and excess accumulation of triacylglycerol in non-adipocytes has been linked to lipid-mediated apoptosis (1Unger R.H. Zhou Y.T. Orci L. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 2327-2332Crossref PubMed Scopus (376) Google Scholar, 2Zhou Y.-T. Grayburn P. Karim A. Shimabukuro M. Higa M. Baetens D. Orci L. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1784-1789Crossref PubMed Scopus (1079) Google Scholar). Although obesity research has focused on the storage of calories that are ingested in excess of energy needs, the observation that mice null for diacylglycerol acyltransferase are resistant to diet-induced obesity suggests the possibility that, independent of energy intake or expenditure, the amount of triacylglycerol stored can be regulated biochemically (3Smith S.J. Cases S. Jensen D.R. Chen H.C. Sande E. Tow B. Sanan D.A. Raber J. Eckel R.H. Farese R.V.J. Nat. Genet. 2000; 25: 87-90Crossref PubMed Scopus (743) Google Scholar). This remarkable observation further suggests that the synthesis of triacylglycerol can be specifically regulated and that regulation of the synthetic pathway could determine how much triacylglycerol is stored in cells. If this were true, enzymes in the biosynthetic pathway of triacylglycerol synthesis could provide targets for therapy. The pathway for triacylglycerol synthesis begins with the acylation of glycerol 3-phosphate at the sn-1 position by glycerol 3-phosphate acyltransferase (GPAT)1 (EC 2.3.1.15), the committed step for the synthesis of all glycerolipids. After acylation by GPAT to form lysophosphatidic acid, LPA acyltransferase acylates the sn-2 position to form phosphatidic acid, which is then hydrolyzed to form diacylglycerol. Phosphatidic acid is the precursor for the anionic phospholipids, phosphatidylinositol, phosphatidylglycerol, and cardiolipin, whereas diacylglycerol is the precursor for the quantitatively most prominent phospholipids, phosphatidylcholine, phosphatidylethanolamine, and phosphatidylserine, as well as for triacylglycerol. Most of the enzymes required for these synthetic steps are located in the endoplasmic reticulum, but LPA acyltransferase activity and a specific GPAT isoform are also located on the mitochondrial outer membrane (4Coleman R.A. Lewin T.M. Muoio D.M. Annu. Rev. Nutr. 2000; 20: 77-103Crossref PubMed Scopus (263) Google Scholar). The rate-limiting step of phosphatidylcholine synthesis, CTP:phosphocholine cytidylyltransferase, undergoes complex post-translational regulation by membrane lipid composition, subcellular location, and phosphorylation state (5Cornell R.B. Gross R.W. Advances in Lipobiology. 1. JAI Press Inc., Greenwich, CT1996: 1-38Google Scholar). It has been generally assumed that phospholipid synthesis is tightly regulated in accordance with requirements for membrane turnover and cell proliferation (6Jackowski S. J. Biol. Chem. 1996; 271: 20219-20222Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar), but that the synthesis of triacylglycerol is controlled primarily by the amount of fatty acid available. Thus, triacylglycerol synthesis is enhanced only when fatty acid synthesis or uptake exceeds the cellular need for phospholipid synthesis. This view is supported by cell culture studies showing that when the concentration of media fatty acid is low, it is primarily incorporated into phospholipid, and when fatty acid concentration is high, the formation of triacylglycerol is greatly favored (7Rosenthal M.D. Lipids. 1980; 15: 838-848Crossref PubMed Scopus (13) Google Scholar, 8Rosenthal M.D. Lipids. 1981; 16: 173-182Crossref PubMed Scopus (33) Google Scholar). In addition, overexpression of putative fatty acid transporters can increase triacylglycerol stores (9Schaffer J.E. Lodish H.F. Cell. 1994; 79: 427-436Abstract Full Text PDF PubMed Scopus (744) Google Scholar). On the other hand, the view that triacylglycerol storage is unregulated is inconsistent with the observation that DGAT knockout mice remain slim on a high fat diet even though they are able to synthesize triacylglycerol by a DGAT-independent mechanism (3Smith S.J. Cases S. Jensen D.R. Chen H.C. Sande E. Tow B. Sanan D.A. Raber J. Eckel R.H. Farese R.V.J. Nat. Genet. 2000; 25: 87-90Crossref PubMed Scopus (743) Google Scholar). Indirect evidence that triacylglycerol synthesis is regulated also comes from studies of the mitochondrial isoform of GPAT, which is transcriptionally controlled by diet changes that promote or depress the synthesis of triacylglycerol. For example, carbohydrate feeding or high insulin concentrations increase liver GPAT mRNA, whereas fasting or lack of insulin down-regulates GPAT mRNA (10Shin D.-H. Paulauskis J.D. Moustaid N. Sul H.S. J. Biol. Chem. 1991; 266: 23834-23839Abstract Full Text PDF PubMed Google Scholar), probably via insulin-mediated regulation of SREBP-1c (11Ericsson J. Jackson S.M. Kim J.B. Spiegelman B.M. Edwards P.A. J. Biol. Chem. 1997; 272: 7298-7305Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar, 12Shimomura I. Bashmakov Y. Ikemoto S. Horton J.D. Brown M.S. Goldstein J.L. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13656-13661Crossref PubMed Scopus (628) Google Scholar). Mitochondrial GPAT mRNA is also elevated 12-fold in pancreatic β-cells from obese diabetic Zucker fa/fa rats, implying a role for triacylglycerol synthesis in the observed β-cell steatosis (13Lee Y. Hirose H. Zhou Y. Esser V. McGarry J.D. Unger R.H. Diabetes. 1997; 46: 408-413Crossref PubMed Scopus (176) Google Scholar). Acute GPAT regulation has also been reported in mitochondria; consistent with the need to decrease triacylglycerol synthesis and increase energy-producing oxidation in the face of limited nutrients, 5-amino-4-imidazolecarboxamide riboside, a synthetic activator of AMP-activated kinase, the proposed sensor of cellular fuel deprivation, acutely decreases GPAT activity (14Muoio D.M. Seefield K. Witters L. Coleman R.A. Biochem. J. 1999; 338: 783-791Crossref PubMed Scopus (350) Google Scholar). Despite these data suggesting that mitochondrial GPAT regulates triacylglycerol synthesis in cells, this hypothesis has not been examined directly. To determine whether mitochondrial GPAT plays a specific role in triacylglycerol synthesis, we created a stable cell line in which the expression of mitochondrial GPAT could be regulated. Silica gel G plates were from Whatman. [2-3H]Glycerol and [1-14C]oleic acid were from PerkinElmer Life Sciences. Essentially fatty acid-free BSA, anti-FLAG M2 monoclonal antibody, and digitonin were from Sigma. Lipid standards were from Doosan Serdary Research Laboratories. Tissue culture supplies were from Corning, and culture medium and ultrafiltered FBS were obtained from Life Technologies, Inc. GPAT with a FLAG epitope (DYKDDDDK) at the C terminus was constructed by adding the FLAG coding sequence by polymerase chain reaction amplification in a two-step procedure. The plasmid pcDNA3.1 containing the entire GPAT open reading frame flanked by BamHI and XhoI sites was used as a template for the amplification. In the first step, the oligonucleotide (5′BamGPAT) ATCGCGGATCCACCATGGAGGAGT was used as the 5′ primer, and the oligonucleotideCATCATCATCCTTGTAGTCGAGCACGACGAAGCTGAGAATGTA (3′GPAT-F1), which adds 19 nucleotides encoding the FLAG epitope (shown in bold), was used as the 3′primer. The amplification product was used as a template for the second polymerase chain reaction step, using the same 5′ primer and the 3′ primer oligonucleotide CGCCTCTAGACTACTTGTCATCATCATCCTTGTAGTCGAGCA (3′Xba FL2), designed to add the last 5 nucleotides corresponding to the FLAG epitope, a stop codon, and a XbaI restriction site. The resulting second amplification product was digested with BamHI and XbaI and inserted in the multicloning site of the plasmid pcDNA3.1 to generate pcDNA3.1-GPAT-FLAG. Mlu1 and NheI sites flanking GPAT-FLAG were added by polymerase chain reaction, and the amplified GPAT-FLAG product was subcloned into the mammalian expression vector pBI-EGFP-tet (CLONTECH), a bidirectional response plasmid that allows simultaneous expression of both green fluorescent protein and GPAT-FLAG under the control of a single tetracycline (doxycycline)-responsive element. CHO cells were cotransfected with the pBI-EGFP/GPAT-FLAG construct and a pTet-Off plasmid (CLONTECH), which expresses a hybrid protein and activates transcription in the absence of doxycycline. When doxycycline is present, neither GPAT nor EGFP are expressed. To establish the double-stable transfected cell line, CHO cells in 35-mm dishes were cotransfected with pBI-EGFP/GPAT-FLAG (1 μg) and pTet-Off (2 μg) using Transfast transfection reagent (Promega). Twenty-four h after transfection, cells were selected with 600 μg/ml G418 for 2 weeks. Positive GFP clones were identified under UV light microscopy and diluted into a 96-well plate. After continued selection with G418 for 6–8 weeks, lines were selected that expressed green fluorescent protein in 100% of the cells. Cells were routinely cultured in MEM with 10% heat-inactivated FBS plus 100 units/ml penicillin and 100 μg/ml streptomycin (normal medium) at 34 °C, 5% CO2, and 100% humidity, in the presence or absence of doxycycline (2 μg/ml) to repress or induce, respectively, GPAT overexpression. The medium was changed every other day. Transient transfection and immunocytochemical analysis of GPAT expression were performed as described previously (15Lin S. Cheng D. Liu M. Chen J. Chang T.Y. J. Biol. Chem. 1999; 274: 23276-23285Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar) with modifications. CHO K1 cells were trypsinized and seeded in 100-mm tissue culture dishes containing six to eight glass coverslips. The cells were incubated at 37 °C until 70% confluence and then transfected with cationic liposomes (LipofectAMINE, Life Technologies, Inc.) with pcDNA 3.1 vector alone or with pcDNA3.1-GPAT-FLAG. After 27 h cells were washed with PBS and incubated for 15 min with 0.1 μm Texas Red, a mitochondrial marker (Mito-tracker, Molecular Probes) in culture medium, rinsed with PBS, and fixed with 2% formaldehyde at room temperature for 10 min. After washing with PBS, cells were permeabilized with 5 μg/ml digitonin in buffer A (0.3 m sucrose, 25 mmMgCl2, 0.1 m KCl, 1 mm EDTA, 10 mm PIPES buffer, pH 6.8) for 5 min on ice, incubated with a 1/1000 dilution of anti-FLAG M2 monoclonal antibody (Sigma) for 60 min, washed with PBS, and incubated for 45 min with a 1/200 dilution of fluorescein isothiocyanate (FITC)-conjugated secondary antibody: goat anti-mouse IgG (Santa Cruz Biotechnology). Cells were then washed in PBS, and the coverslips were mounted. Confocal microscopy was performed on a Zeiss LSH 210 fluorescence microscope equipped with excitation argon laser 488/514 to visualize cells stained with the Texas Red and FITC antibodies. Images were saved and analyzed using Photoshop v.5.5 and Image Processing Tool Kit to overlay multiband fluorescence images. CHO cells that had been grown continuously in either the presence or absence of doxycycline were washed with PBS, scraped into Medium 1 (250 mm sucrose, 10 mm Tris, pH 7.4, 1 mm EDTA, 1 mm dithiothreitol), homogenized with 10 up and down strokes in a Teflon-glass homogenizing vessel, and centrifuged at 100,000 × g for 1 h to obtain total particulate (membrane) fractions. GPAT activity was measured with 10–40 μg of protein after incubation for 15 min on ice in the absence (total GPAT) or presence (mitochondrial GPAT) of 2 mm N-ethylmaleimide. The 10-min assay at room temperature contained 2 mm [3H]glycerol 3-phosphate and 0.75 mm palmitoyl-CoA (16Coleman R.A. Bell R.M. J. Biol. Chem. 1980; 255: 7681-7687Abstract Full Text PDF PubMed Google Scholar). Microsomal GPAT activity was calculated as total activity minus N-ethylmaleimide-resistant activity. Acyl-CoA synthetase activity was assayed using 1–5 μg of protein, 50 μmol of [14C]palmitic acid, 10 mm ATP, 5 mm dithiothreitol, and 0.25 mm CoA for 10 min at 37 °C (17Banis R.J. Roberts C.S. Stokes G.B. Tove S.B. Anal. Biochem. 1976; 73: 1-8Crossref PubMed Scopus (21) Google Scholar). DGAT activity was assayed using 5–20 μg of protein, 0.2 mm sn-1,2-dioleoylglycerol, and 30 μm [3H]palmitoyl-CoA for 10 min at room temperature (18Coleman R.A. Reed B.C. Mackall J.C. Student A.K. Lane M.D. Bell R.M. J. Biol. Chem. 1978; 253: 7256-7261Abstract Full Text PDF PubMed Google Scholar). Proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane, which was probed with anti-FLAG M2 antibody (Sigma), washed extensively, and then probed with horseradish peroxidase-conjugated goat anti-mouse IgG (Pierce). GPAT-FLAG protein was visualized by a reaction with Supersignal chemiluminescent substrate (Pierce) and exposed to x-ray film. CHO cells (stably overexpressing GPAT) were seeded in 60-mm dishes, grown to near confluence in the continuous presence or absence of 2 μg/ml doxycycline, and incubated with either trace (∼3 μm) or 100 μm [14C]oleic acid (0.25 μCi/dish) in medium supplemented with 0.5% BSA. HEK293 (human embryonic kidney) cells in 60-mm dishes were transfected with pcDNA3.1-GPAT-FLAG or pcDNA3.1 (control) for 18 h, then incubated with trace (∼3 μm) [14C]oleic acid (0.25 μCi/dish) in medium supplemented with 0.5% BSA. At various time points, the medium was removed, and the cells were washed with 0.1% BSA in ice-cold PBS scraped in ice-cold methanol and H2O. Total lipids were extracted (19Bligh E.G. Dyer W.J. Can. J. Biochem. Physiol. 1959; 37: 911-917Crossref PubMed Scopus (43132) Google Scholar) and concentrated in a Savant SpeedVac concentrator. Neutral lipids were resolved by TLC using a solvent system consisting of hexane:ethylether:acetic acid (80:20:1; v/v). Phospholipid species were separated by TLC in a one-dimensional solvent system of chloroform:methanol:acetic acid:water (50:37.5:3.5:1.5; v/v). All samples were chromatographed in parallel with pure lipid standards. The 14C-labeled lipids were detected using a Bioscan Image System. Cell monolayers at 80% confluence were scraped from the dishes, washed twice with ice-cold PBS, resuspended in 0.25 m sucrose, 1 mm EDTA, 1 mg/ml protease inhibitor mixture (Sigma), and 10 mm Tris buffer, pH 7.4, and disrupted by several passages through a 22-gauge syringe needle. Proteins from total membranes from GPAT overexpressing CHO cells and controls were separated in SDS-PAGE and blotted onto nitrocellulose membranes. Immunoblots were performed using the corresponding polyclonal antibody against full-length (128 kDa) and mature (68 kDa) SREBP-1 and PPARγ (Santa Cruz Biotechnologies). Anti-rabbit horseradish peroxidase-conjugated antibody was used as secondary antibody. Immunoreactive bands were detected using a chemiluminescence detection kit (Pierce). Protein was measured using serum bovine albumin as the standard (20Lowry O.H. Rosebrough N.J. Farr A.L. Randall R.J. J. Biol. Chem. 1951; 193: 265-275Abstract Full Text PDF PubMed Google Scholar). Triacylglycerol was measured using a TGO-Trinder kit (Sigma), and phospholipid was measured by phosphate analysis (21Bartlett G.R. J. Biol. Chem. 1959; 234: 466-468Abstract Full Text PDF PubMed Google Scholar). Distribution of [14C]oleate in the sn-1 and sn-2 positions of phospholipids was examined by phospholipase A2 (Naja naja) (Sigma) hydrolysis of phospholipids labeled for 48 h with [14C]oleate as described previously (22Igal R.A. Wang P. Coleman R.A. Biochem. J. 1997; 324: 529-534Crossref PubMed Scopus (126) Google Scholar). Lipids extracts were chromatographed with hexane/ethyl ether/acetic acid (80:20:1; v/v) to separate unesterified fatty acids from total polar lipids and counted in the Bioscan Image System. Data are presented as means ± S.D. Differences between the transfected cell lines and the controls were analyzed by Student's t test. A stable Tet-Off CHO cell line was constructed that expressed both EGFP and mitochondrial GPAT with a FLAG-epitope at the C terminus. EGFP was used to ensure that every cell expressed the construct. When cells were grown in the absence of doxycycline, fluorescence microscopy showed expression of both EGFP and the GPAT-FLAG epitope in every cell; no fluorescence was observed when 2 μg/ml doxycycline was present (data not shown). Withdrawing doxycycline from the medium resulted in expression of GPAT protein at 6 days and a 3.3-fold increase in GPAT specific activity over the endogenous basal activity (Fig. 1). Adding doxycycline to the medium repressed GPAT activity and protein expression within 24 h, suggesting that GPAT protein is rapidly degraded (data not shown). A dose response showed that plasmid-mediated GPAT expression (ascertained by Western blot analysis of the FLAG epitope) was nearly completely blocked when as little as 0.05 μg/ml doxycycline was present in the medium for 24 h. To determine whether increased mitochondrial GPAT expression resulted in secondary effects on related enzymes of glycerolipid synthesis, we measured activities of microsomal GPAT, acyl-CoA synthetase, and diacylglycerol acyltransferase in total particulate preparations from the Tet-Off CHO cells grown with (control) or without (GPAT overexpressing) doxycycline (Table I). Mitochondrial GPAT specific activity was 3.8-fold higher in the overexpressing cells compared with controls, but no changes were observed in the specific activities of the other enzymes.Table IGlycerolipid synthetic activities in control cells and cells that over-express mitochondrial GPATMitochondrial GPATMicrosomal GPATAcyl-CoA synthetaseDiacylglycerol acyltransferasenmol/min/mg proteinControl0.11 ± 0.010.85 ± 0.2211.0 ± 2.50.37 ± 0.03(+doxycycline)11%(89%)GPAT0.42 ± 0.09ap < 0.01 compared to control.0.62 ± 0.2610.7 ± 3.10.36 ± 0.01(−doxycycline)(40%)(60%)Mitochondrial GPAT, microsomal GPAT, acyl-CoA synthetase, and diacylglycerol acyltransferase were assayed as described under "Experimental Procedures" in three independent total particulate preparations from cells that were grown continuously with (control) or without (GPAT) doxycycline. Data are expressed as mean ± S.D. Numbers in parentheses represent the percent of total GPAT specific activity.a p < 0.01 compared to control. Open table in a new tab Mitochondrial GPAT, microsomal GPAT, acyl-CoA synthetase, and diacylglycerol acyltransferase were assayed as described under "Experimental Procedures" in three independent total particulate preparations from cells that were grown continuously with (control) or without (GPAT) doxycycline. Data are expressed as mean ± S.D. Numbers in parentheses represent the percent of total GPAT specific activity. GPAT-overexpressing CHO cells in medium without added fatty acid contained 2.7-fold more triacylglycerol mass than control cells (Fig. 2). When the cells were incubated with 100 μm oleate to stimulate triacylglycerol accumulation, both groups of cells increased their neutral lipid content at similar rates throughout the 24-h incubation time, but the mass of triacylglycerol was always greater in the GPAT overexpressing cells (3.1-, 2.2-, and 1.6-fold higher at 3, 6, and 24 h, respectively). Because the final steps in the enzyme pathway that synthesizes triacylglycerol are located in the endoplasmic reticulum, it was surprising to find that overexpression of a mitochondrial enzyme resulted in excess triacylglycerol content. Thus, we wondered whether GPAT might be overexpressed in an aberrant location that could provide LPA directly for triacylglycerol biosynthesis. Because the EGFP expressed by GPAT overexpressing CHO cells would interfere with FITC excitation at 488 nm, we studied CHO cells that had been transiently transfected with an identical GPAT-FLAG construct. Mitochondrial GPAT activity at 28 h after transfection increased from 0.14 to 0.86 nmol/min/mg of protein. Anti-FLAG indirect antibody staining recognized the expressed GPAT (Fig. 3 b) and matched the specific mitochondrial marker signal (Fig. 3, c and d). Both the cells transfected with the empty vector and those that had not been exposed to the first antibody emitted fluorescence in the FITC band only at background levels. Thus, FITC staining was not due to nonspecific binding of the anti-FLAG antibody to mitochondria. When the overlapping image (Fig. 3 d) was subtracted from the GPAT localization image (Fig. 3 b) using Boolean operator transformation, the resulting image was completely dark, indicating that GPAT was present only in mitochondria and not in other locations. Because GPAT-overexpressing CHO cells accumulated more triacylglycerol mass, we measured the ongoing rate of triacylglycerol formation in the absence of added medium fatty acid. Cells were incubated with tracer amounts (∼3 μm) of [14C]oleic acid for up to 24 h. Cells that overexpressed GPAT showed a significantly higher incorporation of [14C]oleate into triacylglycerol (2.7-, 2.5-, and 3.4-fold higher at 3, 6, and 24 h, respectively) compared with control cells (Fig. 4 A). Overexpression of GPAT did not affect [14C]oleic acid incorporation into cholesterol esters (Fig. 4 B). Not only is GPAT the committed step for triacylglycerol synthesis, but it is also required for the synthesis of all the glycerophospholipids. Therefore we examined ongoing incorporation of low amounts of [14C]oleate (∼3 μm) into the different phospholipid species in the GPAT-overexpressing CHO cells and in controls (Fig. 5). At each time point up to 24 h, GPAT overexpressing cells showed 30% decreased incorporation of [14C]oleate into total phospholipids. [14C]Oleate-labeled phosphatidylcholine decreased 30–35% and [14C]PI plus [14C]phosphatidylserine decreased 40–50%, but [14C]phosphatidylethanolamine decreased relatively little (15%). Viewed as a whole, the control cells incorporated 4.7 times more label into phospholipid than triacylglycerol, but the GPAT overexpressing cells incorporated the 14C label equally into the two lipid species. With trace [14C]oleate plus 10% FBS (which contains 21–28 μm fatty acid (7Rosenthal M.D. Lipids. 1980; 15: 838-848Crossref PubMed Scopus (13) Google Scholar)), the cells are exposed to medium that contains ∼30 μmfatty acid. Thus, the incorporation data are surprising, because normally with concentrations of media oleate of less than 30 μm, four to five times more label is incorporated into phospholipid than triacylglycerol, as occurred in the control cells. To enable radiolabel to be distributed equally in the two lipid fractions in fibroblasts, total media oleate concentrations must be greater than 60 μm (8Rosenthal M.D. Lipids. 1981; 16: 173-182Crossref PubMed Scopus (33) Google Scholar). Diminished incorporation of [14C]oleate into phospholipid could result from a decrease in the rate of phospholipid synthesis or because exogenously added fatty acids are incorporated preferentially into triacylglycerol rather than phospholipid. We examined these possibilities by comparing the rate of cell division of control CHO cells and CHO cells that overexpressed GPAT. The rate of growth was identical for the two cell types (Fig. 6), phospholipid content was 0.42, 0.34, and 0.28 μg/μg of protein in GPAT, control, and vector-transfected cells, respectively, and these differences were not statistically significant. Thus, phospholipid synthesis and accretion appeared to be unaffected despite the diminished incorporation of [14C]oleate. Because mitochondrial GPAT prefers saturated over unsaturated fatty acids, it has been hypothesized that one of the enzyme's functions might be to direct the incorporation of saturated fatty acids into the sn-1 position of phospholipids (4Coleman R.A. Lewin T.M. Muoio D.M. Annu. Rev. Nutr. 2000; 20: 77-103Crossref PubMed Scopus (263) Google Scholar). To test this idea, GPAT overexpressing and control cells were incubated with [14C]oleate (∼3 μm). After 48 h total lipids were extracted and treated with phospholipase A2, which hydrolyzes the fatty acid at the sn-2 position of phospholipids. The [14C]oleate content of hydrolyzed (sn-2 fatty acids) and nonhydrolyzed (sn-1 acyl-lysophospholipid) was quantified. In both GPAT-overexpressing and control cells, 70% of the label was released, showing that there had been equivalent incorporation of oleate (30%) at the sn-1 position in control and GPAT overexpressing cells. This study indicates that a 3.8-fold increase in GPAT specific activity did not affect the positional distribution of a monounsaturated fatty acid. As shown in Figs. 4 and 5, when a low concentration of fatty acid was present in the media, GPAT overexpression in CHO cells greatly enhanced the incorporation of [14C]oleate into triacylglycerol while concomitantly decreasing inco
Referência(s)