Cloning and Functional Characterization of a Mouse Intestinal Acyl-CoA:Monoacylglycerol Acyltransferase, MGAT2
2003; Elsevier BV; Volume: 278; Issue: 16 Linguagem: Inglês
10.1074/jbc.m300139200
ISSN1083-351X
AutoresJingsong Cao, John Lockwood, Paul Burn, Yuguang Shi,
Tópico(s)Diet, Metabolism, and Disease
ResumoAcyl-CoA:monoacylglycerol transferase (MGAT) plays a predominant role in dietary fat absorption in the small intestine, where it catalyzes the first step of triacylglycerol resynthesis in enterocytes for chylomicron formation and secretion. Although the mouse small intestine exhibits the highest MGAT enzyme activity among all of the tissues studied, the gene encoding the enzyme has not been identified so far. In the present studies, we report the identification and characterization of a mouse intestinal MGAT, MGAT2. Transient expression of MGAT2 in AV-12, COS-7, and Caco-2 cells led to a more than 70-, 30-, and 35-fold increase in the synthesis of diacylglycerol, respectively. MGAT2 expressed in mammalian cells can catalyze the acylation of rac-1-, sn-2-, andsn-3-monoacylglycerols, and the enzyme prefers monoacylglycerols containing unsaturated fatty acyls as substrates. MGAT2 also demonstrates weak DGAT activity, which can be distinguished from its MGAT activity by detergent treatment that abolishes DGAT but not MGAT activity. We also analyzed the biochemical features of MGAT2 and demonstrated homogenate protein-, time-, and substrate concentration-dependent MGAT enzyme activity in transiently transfected COS-7 cells. Northern blot analysis indicates that the mouse MGAT2 is most abundantly expressed in the small intestine, suggesting that MGAT2 may play an important role in dietary fat absorption. Acyl-CoA:monoacylglycerol transferase (MGAT) plays a predominant role in dietary fat absorption in the small intestine, where it catalyzes the first step of triacylglycerol resynthesis in enterocytes for chylomicron formation and secretion. Although the mouse small intestine exhibits the highest MGAT enzyme activity among all of the tissues studied, the gene encoding the enzyme has not been identified so far. In the present studies, we report the identification and characterization of a mouse intestinal MGAT, MGAT2. Transient expression of MGAT2 in AV-12, COS-7, and Caco-2 cells led to a more than 70-, 30-, and 35-fold increase in the synthesis of diacylglycerol, respectively. MGAT2 expressed in mammalian cells can catalyze the acylation of rac-1-, sn-2-, andsn-3-monoacylglycerols, and the enzyme prefers monoacylglycerols containing unsaturated fatty acyls as substrates. MGAT2 also demonstrates weak DGAT activity, which can be distinguished from its MGAT activity by detergent treatment that abolishes DGAT but not MGAT activity. We also analyzed the biochemical features of MGAT2 and demonstrated homogenate protein-, time-, and substrate concentration-dependent MGAT enzyme activity in transiently transfected COS-7 cells. Northern blot analysis indicates that the mouse MGAT2 is most abundantly expressed in the small intestine, suggesting that MGAT2 may play an important role in dietary fat absorption. When total energy input is in excess of energy output, most excess calories are stored as triacylglycerol and deposited into adipose tissue. The excess storage of triacylglycerol in fat tissues causes obesity. There are two major sources of triacylglycerol syntheses. One is the endogenous pathway where triacylglycerol is resynthesized from circulating free fatty acids, and the other is from dietary fat absorption. The latter becomes a major contributing factor for the ongoing epidemic of obesity in people who consume western diets with high fat content. In mammals, the intestine plays a predominant role in fat absorption (1Phan C.T. Tso P. Front. Biosci. 2001; 6: D299-D319Crossref PubMed Google Scholar, 2Mansbach 2nd, C.M. Nevin P. J. Lipid Res. 1998; 39: 963-968Abstract Full Text Full Text PDF PubMed Google Scholar). Neutral fat or triacylglycerol is emulsified to fine lipid droplets and further enzymatically digested to monoacylglycerol and fatty acids, mainly by pancreatic lipase and its cofactor. Facilitated by bile salts, the produced monoacylglycerol and fatty acids are up-taken by enterocytes (3Westergaard H. Dietschy J.M. J. Clin. Invest. 1976; 58: 97-108Crossref PubMed Scopus (248) Google Scholar), where they are used to resynthesize triacylglycerol allowing for lipids to be transported into the circulation system (1Phan C.T. Tso P. Front. Biosci. 2001; 6: D299-D319Crossref PubMed Google Scholar). Diacylglycerol is a precursor for phospholipids and triacylglycerol synthesis, which play important roles in cellular signal transduction (4Ron D. Kazanietz M.G. FASEB J. 1999; 13: 1658-1676Crossref PubMed Scopus (553) Google Scholar), intestinal fat absorption (1Phan C.T. Tso P. Front. Biosci. 2001; 6: D299-D319Crossref PubMed Google Scholar, 5Bell R.M. Coleman R.A. Annu. Rev. Biochem. 1980; 49: 459-487Crossref PubMed Scopus (456) Google Scholar, 6Lehner R. Kuksis A. Prog. Lipid Res. 1996; 35: 169-201Crossref PubMed Scopus (254) Google Scholar), energy storage in muscle and adipose tissues (7Swanton E.M. Saggerson E.D. Biochim. Biophys. Acta. 1997; 1346: 93-102Crossref PubMed Scopus (22) Google Scholar), and lactation (5Bell R.M. Coleman R.A. Annu. Rev. Biochem. 1980; 49: 459-487Crossref PubMed Scopus (456) Google Scholar, 8Smith S.J. Cases S. Jensen D.R. Chen H.C. Sande E. Tow B. Sanan D.A. Raber J. Eckel R.H. Farese Jr., R.V. Nat. Genet. 2000; 25: 87-90Crossref PubMed Scopus (734) Google Scholar). Diacylglycerol is synthesized by two pathways, the glycerol phosphate pathway and the monoacylglycerol pathway (5Bell R.M. Coleman R.A. Annu. Rev. Biochem. 1980; 49: 459-487Crossref PubMed Scopus (456) Google Scholar, 6Lehner R. Kuksis A. Prog. Lipid Res. 1996; 35: 169-201Crossref PubMed Scopus (254) Google Scholar). The glycerol phosphate pathway is present in almost all tissues (6Lehner R. Kuksis A. Prog. Lipid Res. 1996; 35: 169-201Crossref PubMed Scopus (254) Google Scholar). During fat absorption by intestine, the monoacylglycerol pathway is the major route of diacylglycerol resynthesis because of the high influx of monoacylglycerol from breakdown of dietary triacylglycerol (9Coleman R.A. Haynes E.B. J. Biol. Chem. 1986; 261: 224-228Abstract Full Text PDF PubMed Google Scholar). In the intestinal enterocytes, the resynthesis of triacylglycerol is predominantly catalyzed by two types of enzymes: acyl-CoA:monoacylglycerol acyltransferase (MGAT) 1The abbreviations used are: MGATacyl-CoA:monoacylglycerol acyltransferasemMGATmurine MGATDGATacyl-CoA:diacylglycerol acyltransferasehDGAThuman DGAT1The abbreviations used are: MGATacyl-CoA:monoacylglycerol acyltransferasemMGATmurine MGATDGATacyl-CoA:diacylglycerol acyltransferasehDGAThuman DGAT and acyl-CoA:diacylglycerol acyltransferase (DGAT). Sequentially, MGAT catalyzes the joining of monoacylglycerol and fatty acyl-CoAs to form diacylglycerol, whereas DGAT catalyzes the final reaction in which fatty acyl-CoAs are covalently joined to diacylglycerol, producing triacylglycerol (10Coleman R.A. Lewin T.M. Muoio D.M. Annu. Rev. Nutr. 2000; 20: 77-103Crossref PubMed Scopus (261) Google Scholar). It was reported that MGAT, DGAT, and acyl-CoA synthase form a triacylglycerol synthase complex, which is distributed within intestinal microsomes (11Rao G.A. Johnston J.M. Biochim. Biophys. Acta. 1966; 125: 465-473Crossref PubMed Scopus (55) Google Scholar, 12Rao G.A. Johnston J.M. Biochem. Biophys. Res. Commun. 1966; 22: 696-700Crossref Scopus (2) Google Scholar, 13Lehner R. Kuksis A. J. Biol. Chem. 1995; 270: 13630-13636Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). acyl-CoA:monoacylglycerol acyltransferase murine MGAT acyl-CoA:diacylglycerol acyltransferase human DGAT acyl-CoA:monoacylglycerol acyltransferase murine MGAT acyl-CoA:diacylglycerol acyltransferase human DGAT Two mammalian DGAT enzymes and one MGAT enzyme have been identified so far (14Cases S. Smith S.J. Zheng Y.W. Myers H.M. Lear S.R. Sande E. Novak S. Collins C. Welch C.B. Lusis A.J. Erickson S.K. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13018-13023Crossref PubMed Scopus (875) Google Scholar, 15Cases S. Stone S.J. Zhou P. Yen E. Tow B. Lardizabal K.D. Voelker T. Farese Jr., R.V. J. Biol. Chem. 2001; 276: 38870-38876Abstract Full Text Full Text PDF PubMed Scopus (637) Google Scholar, 16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar). The identification and characterization of these gene products and thereafter the generation of knockout mice lacking these enzymes have greatly extended our understanding of lipid metabolism and their relevance with human diseases such as obesity (8Smith S.J. Cases S. Jensen D.R. Chen H.C. Sande E. Tow B. Sanan D.A. Raber J. Eckel R.H. Farese Jr., R.V. Nat. Genet. 2000; 25: 87-90Crossref PubMed Scopus (734) Google Scholar, 17Chen H.C. Ladha Z. Farese Jr., R.V. Endocrinology. 2002; 143: 2893-2898Crossref PubMed Scopus (31) Google Scholar, 18Chen H.C. Farese Jr., R.V. Trends Cardiovasc. Med. 2000; 10: 188-192Crossref PubMed Scopus (90) Google Scholar, 19Farese Jr., R.V. Cases S. Smith S.J. Curr. Opin. Lipidol. 2000; 11: 229-234Crossref PubMed Scopus (121) Google Scholar). These works also provide valuable molecular probes to further explore additional enzymes involved in lipid metabolism. A gene encoding MGAT1 was recently identified in mice where expression of MGAT1 was detected in the stomach, kidney, white and brown adipose tissue, and liver (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar). However, expression of MGAT1 was not detected in small intestine, where high levels of MGAT activity had been observed (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar, 20Tsujita T. Miyazaki T. Tabei R. Okuda H. J. Biol. Chem. 1996; 271: 2156-2161Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar), thereby implying other potential candidates for intestinal MGAT enzymes. In the present study, we identified and characterized a mouse intestinal MGAT gene (MGAT2) that is most abundantly expressed in small intestine. The murine MGAT2 demonstrates a strong MGAT activity and a weak DGAT activity, with the latter being sensitive to inactivation with detergent. A cDNA clone was identified in NCBI databases based on sequence homology to the known mouse MGAT1. According to flanking cDNA ends, primer pairs were designed to amplify the full-length coding region of MGAT2 from Marathon-Ready cDNA prepared from mouse small intestine (BD Biosciences Clontech, Palo Alto, CA). Amplification was performed by PCR using Pfu DNA polymerase (Stratagene, La Jolla, CA), resulting in a ∼1.1-kb cDNA product. The PCR product was cloned into the pPCR-script Amp SK(+) vector (Stratagene) and sequenced. A mammalian expression plasmid coding full-length mouse MGAT2 was engineered by unidirectional subcloning of the 1.1-kb cDNA fragment from the pPCR-script Amp SK(+) vector described above into the NotI and HindIII sites of pcDNA3.1/Hygro(−) mammalian expression vector (Invitrogen) according to the manufacturer's instruction. Transient transfection was performed on AV-12, COS-7, or Caco-2 cell lines using the mouse MGAT2 expression plasmids. Cells were maintained under the conditions recommended by the ATCC (Manassas, VA). A day before transfection, 2–4 million cells were subcultured into 20 × 150-mm plates, resulting in ∼70% confluence. Cells were transfected with 15 μg of DNA with the Fugene 6 mixture (Roche Molecular Biochemicals) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were harvested in ice-cold phosphate-buffered saline, pelleted by centrifugation, lysed, and assayed immediately or frozen in liquid N2 for later use. Cell pellets were homogenized in 20 mm NaCl with three short 10-s pulses from a Brinkmann Instruments Polytron homogenizer. The resultant homogenates were used to assess the activity of MGAT and DGAT in transfected mammalian cells. The protein concentration in homogenates was determined by a BCA protein assay kit (Pierce) according to the manufacturer's instructions. MGAT and DGAT activity was determined at room temperature in a final volume of 200 μl as previously described (14Cases S. Smith S.J. Zheng Y.W. Myers H.M. Lear S.R. Sande E. Novak S. Collins C. Welch C.B. Lusis A.J. Erickson S.K. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13018-13023Crossref PubMed Scopus (875) Google Scholar, 16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar). Briefly, MGAT activity was determined by measuring the incorporation of [14C]oleoyl moiety into dioleoylglycerol with [14C]oleoyl-CoA (American Radiolabeled Chemical Inc., St. Louis, MO) and acyl acceptors (rac-1-oleoylglycerol, sn-2-oleoylglycerol,rac-1-stearoylglycerol, andsn-3-stearoylglycerol). The incorporation of the [14C]oleoyl moiety into trioleoylglycerol with [14C]oleoyl-CoA and sn-1,2-dioleoylglycerol was measured to obtain DGAT activity. The acyl acceptors were introduced into the reaction mixture by liposomes. Briefly, acyl acceptors and phosphatidylcholine (molar ratio 1:5) were dissolved in chloroform and dried by vacuum. Liposomes were formed by adding water and shaking for 1 min. The reaction mixture contained 100 mm Tris/HCl, pH 7.4, 5 mm MgCl2, 1 mg/ml bovine serum albumin free fatty acids (Sigma), 200 mmsucrose, 25 μm [14C]oleoyl-CoA (50 mCi/mmol; American Radiolabeled Chemicals), 200 μm acyl acceptors in liposomes, and various amounts of cellular homogenate. Unless indicated elsewhere, these experimental conditions were used throughout the study. The reaction was initiated by adding protein homogenate and terminated by adding 4 ml of chloroform/methanol (2:1, v/v) after predetermined incubation periods. Lipids were extracted with chloroform/methanol (2:1, v/v) by vortexing for 30 s. After centrifugation to remove debris, aliquots of the organic phase containing lipids were dried under a N2 stream and separated by the Linear-K Preadsorbent TLC plate (Waterman Inc., Clifton, NJ) with hexane/ethyl ether/acetic acid (80:20:1, v/v/v). The separation was always performed under the conditions wheresn-1,2(2,3)- and sn-1,3-diacylglycerols were clearly resolved. Individual lipid moieties were identified by standards with exposure to I2 vapor. The TLC plates were exposed to x-ray film or a phosphor screen to assess the incorporation of 14C-labeled acyl moieties into respective lipid products. Phosphorimaging signals were visualized using a Storm 860 PhosphorImager (Amersham Biosciences) and quantitated using the ImageQuant software. The labeled products in TLC plates were also scraped into scintillation vials to quantitate the incorporation of radioactivity into lipid products by a Beckman LS 6500 scintillation system (Fullerton, CA). The isomers of monoacylglycerols (rac-1-oleoylglycerol, sn-2-oleoylglycerol,rac-1-stearoylglycerol, andsn-3-stearoylglycerol) were from Sigma. Diacylglycerols (sn-1,2-dioleoylglycerol,rac-1,2-dioleoylglycerol, sn-1,3- dioleoylglycerol, rac-1,2-distearoylglycerol, andsn-1,3-distearoylglycerol) and triacylglycerols (glycerol trioleate and glycerol tristearate) were purchased from Doosan Serdary Research Laboratories (Toronto, Canada). To analyze the tissue distribution pattern of mouse MGAT2 mRNA, multiple tissue poly(A)+RNA Northern blots from OriGene Technologies Inc. (Rockville, MD) or BioChain Institute Inc. (Hayward, CA) were hybridized with [α-32P]dCTP (3000 Ci/mmol; ICN Radiochemicals) labeled probes prepared from full-length cDNA of mouse MGAT2 using a random primer DNA labeling system (Invitrogen). Hybridization was carried out in ULTRAhyb (Ambion, Austin, TX) at 50 °C overnight, followed by three washes at 55 °C in 2× SSC buffer containing 0.1% SDS and 1 mm EDTA. Blots were stripped with boiling 1% SDS to remove radiolabeled probe and reprobed with β-actin cDNA as an internal control. The blots were exposed to Bio-Max MR film or a PhosphorImager screen to visualize the signals, which were quantitated using ImageQuant (Amersham Biosciences). Yen et al. (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar) recently reported the cloning and characterization of the first mammalian MGAT, MGAT1. Surprisingly, the mouseMGAT1 gene is not expressed in the small intestine, a tissue that exhibits the highest MGAT enzyme activity (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar); hence, the gene responsible for the intestinal MGAT activity remains to be identified. Using the mouse MGAT1 cDNA as a query sequence for blast analysis of the public genomic and expressed sequence tag data bases, we identified and cloned a MGAT candidate gene, MGAT2, by PCR amplification. The mouse MGAT2 gene predicts a 334-amino acid protein of 38.6 kDa and possesses at least one transmembrane region, as suggested by a 40-amino acid signal peptide at the N terminus (MVEFAPLLVPWERRLQTFAVLQWVFSFLALAQLCIVIFVG), and one potential tyrosine phosphorylation site. There is no obvious N-linked glycosylation site within the protein predicted by the sequence analysis. Mouse MGAT2 is localized at chromosome 7 at 7E1. Sequence comparison indicates that the mouse MGAT2 protein shares extensive homology (52.5% identity) with MGAT1, which spans the entire molecule (Fig. 1). The mouse MGAT2 also shares close similarity in hydrophobicity profile with MGAT1 (Fig.2). Although the MGAT2 shares significant sequence homology (47.5%) with the mouse DGAT2 as well, the homology does not cover the entire molecule (data not shown). Likewise, the MGAT2 protein also differs from the mouse DGAT2 in hydrophobicity profile (Fig. 2). These features predict the mouse MGAT2 as a novel member of the MGAT family of enzymes.Figure 2Comparison of hydrophobicity profiles among the mouse MGAT2, MGAT1, and DGAT2 proteins. The hydrophobicity profiles are assessed by Kyte-Doolittle analysis.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The tissue distribution of mouse MGAT2 mRNA was analyzed by Northern blot analysis (Fig. 3). A 1.8-kbMGAT2 mRNA transcript was detected in small intestine, kidney, adipose, and stomach (Fig. 3, A and B). Very low expression levels of MGAT2 were also detected in liver, skeletal muscle, and spleen. When normalized with the mRNA level of β-actin, the expression level was the highest in intestine, followed by kidney, adipose, stomach, liver, skeletal muscle, and spleen (Fig. 3C). In contrast, no observable level of expression was detected in murine brain, heart, lung, skin, testis, and thymus (Fig. 3A). To determine whether the protein encoded by MGAT2 cDNA possesses MGAT, DGAT, or both activities, MGAT2 was transiently expressed in AV-12, COS-7, and Caco-2 cells followed by in vitro enzyme assay analyses. Since MGAT2 shares the closest sequence homology with MGAT1, we first examined MGAT activity in cells transfected with MGAT2 cDNA. In all three cell lines tested, MGAT activity in protein homogenates from MGAT2-transfected cells was significantly higher than that observed in untransfected and mock-transfected (with empty vector) cells (Fig.4A) as assessed by a significant increase of radioactive 1,2(2,3)-dioleoylglycerol product when 2-monooleoylglycerol was used as acyl acceptor, thereby establishing the function of cloned MGAT2 as a synthetic enzyme of diacylglycerol. Radioactivity incorporated into 1,2(2,3)-diacylglycerol by homogenate prepared fromMGAT2-transfected AV-12, COS-7, and Caco-2 cells was 70-, 30-, or 35-fold higher, respectively, compared with that from the untransfected or mock-transfected cells (Fig. 4B). Since MGAT2 also shares sequence homology with the mouse DGAT2, we next analyzed DGAT activity using the same cell homogenates for the MGAT enzyme assays. As indicated in Fig. 4, C and D, the incorporation of 14C-acyl into triacylglycerol by homogenates from MGAT2-transfected AV-12, COS-7, and Caco-2 cells was 7-, 3-, and 1.5-fold higher than that observed in control or mock-transfected cells, respectively, indicating that MGAT2 also possesses weak DGAT activity. However, the detected DGAT activity is 10-fold lower when compared with the MGAT activity. To investigate substrate stereospecificity, we examined whether MGAT2 can catalyze the acylation of the other stereoisomers of monoacylglycerol. The experiments were conducted under conditions where a 50 μm concentration of each acyl acceptor (rac-1-oleoylglycerol, sn-2-oleoylglycerol,sn-3-stearoylglycerol, or rac-1-stearoylglycerol) was incubated with 25 μm [14C]oleoyl-CoA for 20 min. Cell homogenate was prepared from mock-transfected (as negative control), mouse MGAT2 (mMGAT2)-, or mouseMGAT1 (mMGAT1)-transfected COS-7 cells. As shown in Fig.5, MGAT2 can acylate each of four monoacylglycerol isomers to form diacylglycerols. The radioactive bands of diacylglycerols were quantitated, and the results were shown in Fig.5B. The major products were sn-1,3-diacylglycerol and sn-1,2(2,3)-diacylglycerol whenrac-1-monoacylglycerol and 2-monoacylglycerol were employed, respectively (Fig. 5). MGAT2 also can usesn-3-stearoylglycerol as substrate (Fig. 5, lane 7 versus lane 3). The chiral sn-1-monoacylglycerol was not included in the present studies due to the lack of commercially available compound. However, under the same concentration (50 μm), a similar amount of diacylglycerol was produced when rac-1-stearoylglycerols were used as substrate in comparison withsn-3-stearoylglycerol (Fig. 5, lane 8 versus lane 7), implying thatsn-1-stearoylglycerol can also be used as substrate of MGAT2. In addition, MGAT2 utilized monoacylglycerol containing unsaturated fatty acyls (rac-1-oleoylglycerol, C18:1) more efficiently than that of saturated fatty acyls (rac-1-stearoylglycerol, C18:0) (Fig. 5, lane 5 versus lane 8). mMGAT1 showed similar patterns of enzymatic activity to those of mMGAT2 (Fig.5, lanes 9–12), which is consistent with a previous report by Yen et al. (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar). We next directly compared the pattern of enzymatic activity of mMGAT2 with that of the human DGAT1 (hDGAT1) (14Cases S. Smith S.J. Zheng Y.W. Myers H.M. Lear S.R. Sande E. Novak S. Collins C. Welch C.B. Lusis A.J. Erickson S.K. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13018-13023Crossref PubMed Scopus (875) Google Scholar) under the same experimental conditions. When assessed for the MGAT enzyme activity, the murine MGAT2 demonstrated significantly higher MGAT activity than the human DGAT1 (15-fold versus 2-fold) as measured by incorporation of the [14C]oleoyl moiety into 1,2-diacylglycerol by cell homogenates from mMGAT2 and hDGAT1-transfected cells, respectively, as compared with mock-transfected cells (Fig. 6,lanes 3 and 4 versus lane 2, bottom bands). In contrast, the production of labeled triacylglycerol by cell homogenates from the hDGAT1-transfected cells was significantly higher than that from the mMGAT2-transfected cells (7.6-fold versus 1.8-fold) when compared with the negative controls (Fig. 6, lanes 3 and 4 versus lane 2, top bands). These data further confirm that MGAT2 is a novel member of the MGAT family of enzymes despite its weak DGAT activity. Furthermore, the MGAT activity of mMGAT2 can be separated from its DGAT activity by treatment with detergent. Solubilization of cell homogenates by Triton X-100 abolished the DGAT activity but not the MGAT activity of mMGAT2 (Fig. 6,lane 5 versus lane 4). We next examined the biochemical characteristics of MGAT activity in cell homogenate from MGAT2-transfected COS-7 cells. The dependence of the MGAT2 activity on the protein concentration in the reaction as well as the incubation time is shown in Fig. 7, A andB, respectively. In Fig. 7A, increased amounts of protein homogenate from MGAT2-transfected or 500 μg of that from mock-transfected cells were incubated with 200 μmSn-2-monooleoylglycerol and 25 μm[14C]oleoyl-CoA for 10 min, followed by lipid extraction and TLC assay. In Fig. 7B, 50 μg of protein homogenate was incubated with the above mentioned substrates for 0–60 min. MGAT activity levels of MGAT2, as assessed by the formation of radioactive 1,2(2,3)-dioleoylglycerol, increased proportionately with incubation times (Fig. 7B). The embedded values in Fig. 7, Aand B, represent the quantitative [14C]1,2(2,3)-diacylglycerol produced (arbitrary units/min) in the reaction. Again, the homogenate protein from mock-transfected cells showed only marginal activities (Fig.7A, last lane; Fig. 7B,last two lanes). We further examined the profile of MGAT2 activity at different time points after transfection as well as the kinetics of MGAT activity. The MGAT2 activity can be detected as early as 12 h post-transfection period as judged by a 15-fold increase in the synthesis of radioactive diacylglycerol as compared with the negative controls (Fig.8A). The MGAT activity reached a plateau at 24 h and began to decline at 72 h post-transfection (Fig. 8A). In contrast, cell homogenates from mock-transfected cells collected at 12 and 72 h post-transfection did not display any increase of MGAT activity (1.2versus 1.0 unit). We next analyzed the effect of substrates concentration on the enzyme activity of MGAT2. When the concentration of one component of MGAT substrates, eitherSn-2-monooleoylglycerol or oleoyl-CoA, was increased while the other was held constant (25 μm for oleoyl-CoA and 200 μm for Sn-2-monooleoylglycerol) in the reaction, the formation of dioleoylglycerol also increased over a wide range (0–66 μm for both substrates) (Fig.8B). Further increases in substrate concentration resulted in inhibition of the MGAT activity (Fig. 8B). It has been known for some time that MGAT enzyme activity is the rate-limiting step in the synthesis of triacylglycerol in the small intestine, which generates 80% of the triacylglycerol levels in chylomicrons (21Tso P. Crissinger K. Stipanuk M.H. in Biochemical and Physiological Aspects of Human Nutrition. W. B. Saunders Co., Philadelphia2000: 125-141Google Scholar). However, a clear understanding of the biosynthesis pathway of glycerolipids in intestinal mucosa had been hindered by the absence of the purified enzymes. Although a mouse MGAT (MGAT1) enzyme was recently identified by its sequence homology to the mammalian DGAT2 gene (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar), the mouse MGAT1 is not expressed in small intestine, where dietary absorption takes place (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar), implying the existence of another MGATgene. In the present studies, we have identified and characterized a gene encoding a MGAT enzyme from a mouse intestinal cDNA library by its sequence homology to mouse MGAT1 and hence named MGAT2. In addition to high sequence homology to MGAT1, the hydrophobicity profile of the mouse MGAT2 protein also shares striking similarity with that of the mouse MGAT1 but is quite different from that of the mouse DGAT2. Evidence that the mouse MGAT2 is a MGAT enzyme is provided by the findings that the MGAT2 expressed in three different mammalian cell lines catalyzed significantly the monocylglycerol synthesis as measured by a more than 30-fold increase in radiolabeled diacylglycerol than that from mock-transfected cells. Furthermore, the increased MGAT activity was dependent on the concentrations of either substrate, monoacylglycerol or oleoyl-CoA. Protein homogenate from MGAT2-transfected cells showed time- and dose-dependent responses with incubation time and amount of protein, whereas this phenomenon was not found in control cells under the same conditions. The MGAT activity coincides with the expression profile of theMGAT2 transgene, which can be detected as early as 12 h after transfection and reached a plateau at 48 h post-transfection. As observed with MGAT1 (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar), cell homogenates from three different mammalian cell lines transiently expressing MGAT2 also displayed significantly higher DGAT activity than that from mock-transfected cells. The DGAT activity is more pronounced with an extended incubation time. However, the DGAT activity is much weaker than the MGAT activity as assessed by the rate of formation of radioactive products (700 pmol of diacylglycerol/min/mg of protein versus 29 pmol of triacylglycerol/min/mg of protein in AV-12 cells) catalyzed by mouse MGAT2 in the assay. This is consistent with our observation that no significant levels of triacylglycerol over background were detected in the MGAT assay from MGAT2-transfected cells, although MGAT2 produced a significant amount of diacylglycerol (over 10 μm) during 10 min of incubation. Furthermore, when compared directly with the human DGAT1 enzyme for substrate specificity under the same experimental conditions, the murine MGAT2 demonstrated significantly higher MGAT activity than the human DGAT1 enzyme that exhibited primarily the DGAT activity, thus establishing the MGAT2 as a new member of the MGAT family of enzymes. A fatty acyl-CoA is required for the display of MGAT activity, since the incorporation of [3H]oleic acid into diacylglycerol by membrane protein from MGAT2-transfected cells is not significantly higher than that by membrane from wild type cells (data not shown), indicating that, similar to DGAT1, the activity of MGAT2 is CoA-dependent (14Cases S. Smith S.J. Zheng Y.W. Myers H.M. Lear S.R. Sande E. Novak S. Collins C. Welch C.B. Lusis A.J. Erickson S.K. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13018-13023Crossref PubMed Scopus (875) Google Scholar). Tsujita et al. (20Tsujita T. Miyazaki T. Tabei R. Okuda H. J. Biol. Chem. 1996; 271: 2156-2161Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar) reported a relatively high CoA-independent MGAT activity due to pancreatic lipase distributed within intestinal cells, which may serve as an alternative pathway for diacylglycerol synthesis in intestinal mucosa. The individual role of these two different pathways in intestinal fat absorption remains unknown and awaits further investigation. It was reported that MGAT and DGAT are integral membrane proteins, and application of detergent severely inhibited the activity, which makes it difficult to purify them, because tight lipid-protein interactions are necessary for their optimal activities (13Lehner R. Kuksis A. J. Biol. Chem. 1995; 270: 13630-13636Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Interestingly, our result shows that solubilization of MGAT2-transfected cells by Triton X-100, a nonionic detergent, resulted in a total loss of DGAT activity but did not affect the MGAT activity possessed by MGAT2. Thus, dependence on membrane integrity can be used to separate the DGAT activity from the MGAT activity of the murine MGAT2 enzyme. Using the intestinal mucosal microsomes, Paltauf and Johnston (22Paltauf F. Johnston J.M. Biochim. Biophys. Acta. 1971; 239: 47-56Crossref PubMed Scopus (26) Google Scholar) reported the utilization of a series of monoalkyl and monoacyl glycerols as substrates for glyceride synthesis. In their studies, bothsn-1- and sn-2-alkyl glycerols are acylated, whereas the sn-3-alkyl was a poor substrate. However, each of the sn-acyl monoglycerols could be utilized as substrate for the intestinal mucosa (22Paltauf F. Johnston J.M. Biochim. Biophys. Acta. 1971; 239: 47-56Crossref PubMed Scopus (26) Google Scholar). Therefore, there were observed differences between the sn-alkyl and acyl glycerols in their studies. Our data show that MGAT2 expressed in mammalian cells can catalyze the acylation of rac-1-, sn-2-, andsn-3-monoacylglycerol (Fig. 5). Studies with everted sacs of rat intestinal mucosa (23Breckenridge W.C. Kuksis A. Can. J. Biochem. 1975; 53: 1184-1195Crossref PubMed Scopus (17) Google Scholar, 24Breckenridge W.C. Kuksis A. Can. J. Biochem. 1975; 53: 1170-1183Crossref PubMed Scopus (20) Google Scholar) and a purified monoacylglycerol acyltransferase from rat intestinal mucosa (25Manganaro F. Kuksis A. Can. J. Biochem. Cell Biol. 1985; 63: 341-347Crossref PubMed Scopus (28) Google Scholar) indicated the acylation in sn-1- and sn-3-positions of thesn-2-monoacylglycerol occurred at about the same rate with a slight preference at sn-1-position. In contrast, Johnstonet al. (26Johnston J.M. Paultauf F. Schiller C.M. Schultz L.D. Biochim. Biophys. Acta. 1970; 218: 124-133Crossref PubMed Scopus (47) Google Scholar) reported that the intestinal microsomes greatly preferred to acylate the sn-1-position, compared with thesn-3-position. Further stereospecific analysis of 1,2(2,3)-diacylglycerol product would resolve the issue of whether mMGAT2 has a preference at the sn-1- orsn-3-position of 2-monoacylglycerol. Manganaro and Kuksis (25Manganaro F. Kuksis A. Can. J. Biochem. Cell Biol. 1985; 63: 341-347Crossref PubMed Scopus (28) Google Scholar) reported the purification and preliminary characterization of a 2-monoacylglycerol acyltransferase from rat intestinal villus cells, which gives a single band of 37,000 daltons on SDS-polyacrylamide gel electrophoresis. This is very close to the 38,600 daltons predicted by the mouse MGAT2. Additionally, both enzymes catalyzed efficient acylation of 2-monoacyglcerols. However, analysis of the amino acid composition of the mouse MGAT2 revealed significant differences between the two enzymes (data not shown). Thus, the purified enzyme might represent a novel isoform/homologue of the MGAT enzyme that remains to be identified. One of the unique features of MGAT2 is its selective tissue expression pattern, which is quite different from the previously reported twoDGAT genes (DGAT1 and DGAT2) andMGAT1 gene. The DGAT1 and DGAT2 genes are ubiquitously expressed, whereas the MGAT1 expression is absent in the small intestine (14Cases S. Smith S.J. Zheng Y.W. Myers H.M. Lear S.R. Sande E. Novak S. Collins C. Welch C.B. Lusis A.J. Erickson S.K. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13018-13023Crossref PubMed Scopus (875) Google Scholar, 15Cases S. Stone S.J. Zhou P. Yen E. Tow B. Lardizabal K.D. Voelker T. Farese Jr., R.V. J. Biol. Chem. 2001; 276: 38870-38876Abstract Full Text Full Text PDF PubMed Scopus (637) Google Scholar, 16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar). In contrast, the mouseMGAT2 gene is most abundantly expressed in small intestine, which is consistent with the previous report that the mouse small intestine exhibits the highest MGAT activity among all of the tissues examined (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar). Although it remains elusive, the data suggest that the mouse MGAT2 may play an important role in dietary fat absorption. A high level of MGAT2 expression was also detected in mouse kidney, a tissue where triacylglycerol synthesis and MGAT activity are reported (27Barac-Nieto M. Cohen J.J. Nephron. 1971; 8: 488-499Crossref PubMed Google Scholar, 28Trimble M.E. Life Sci. 1978; 22: 883-890Crossref PubMed Scopus (3) Google Scholar, 29Trimble M.E. Harrington Jr., W.W. Bowman R.H. Curr. Probl Clin. Biochem. 1977; 8: 362-370PubMed Google Scholar, 30Coleman R.A. Haynes E.B. J. Biol. Chem. 1984; 259: 8934-8938Abstract Full Text PDF PubMed Google Scholar). Additionally, MGAT2 expression is detected in the white adipocytes, implicating a role in fat storage. In contrast to abundantMGAT1 expression in the stomach (16Yen C.L. Stone S.J. Cases S. Zhou P. Farese Jr., R.V. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 8512-8517Crossref PubMed Scopus (151) Google Scholar), only a low level ofMGAT2 mRNA is detected in this tissue, which is consistent with the pattern of dietary fat absorption that starts in stomach and mainly occurs in small intestine (1Phan C.T. Tso P. Front. Biosci. 2001; 6: D299-D319Crossref PubMed Google Scholar). Consumption of a western diet enriched with a high fat content has been cited as one of the common causes of the ongoing epidemic of obesity. In a recent study conducted in Otsuka Long-Evans Tokushima fatty rat, which is an animal model of diabetes and obesity, increased small intestine MGAT activities were associated with the onset of the metabolic diseases (31Luan Y. Hirashima T. Man Z.W. Wang M.W. Kawano K. Sumida T. Diabetes Res. Clin. Pract. 2002; 57: 75-82Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar). In Zucker obese rats, although the specific activities of MGAT in the small intestine were not different from those in lean rats, the total activity of MGAT increased significantly in the obese rats because of the heavy and large intestinal segments (32Jamdar S.C. Cao W.F. Biochim. Biophys. Acta. 1995; 1255: 237-243Crossref PubMed Scopus (27) Google Scholar). The importance of triacylglycerol synthesis and storage is further underlined by recent characterization of mice deficient in one of the DGAT genes (8Smith S.J. Cases S. Jensen D.R. Chen H.C. Sande E. Tow B. Sanan D.A. Raber J. Eckel R.H. Farese Jr., R.V. Nat. Genet. 2000; 25: 87-90Crossref PubMed Scopus (734) Google Scholar). Inactivation of DGAT1 results in mice that are resistant to diet-induced obesity and other related metabolic abnormalities (8Smith S.J. Cases S. Jensen D.R. Chen H.C. Sande E. Tow B. Sanan D.A. Raber J. Eckel R.H. Farese Jr., R.V. Nat. Genet. 2000; 25: 87-90Crossref PubMed Scopus (734) Google Scholar), although a recent report indicates that the DGAT1 is not required for dietary fat absorption (33Buhman K.K. Smith S.J. Stone S.J. Repa J.J. Wong J.S. Knapp Jr., F.F. Burri B.J. Hamilton R.L. Abumrad N.A. Farese Jr., R.V. J. Biol. Chem. 2002; 277: 25474-25479Abstract Full Text Full Text PDF PubMed Scopus (194) Google Scholar). Our current study on MGAT2 has provided basic information for future work on the potential physiological and pathological roles of MGAT2 in dietary fat absorption and thus may present MGAT2 as an attractive drug target for obesity intervention. We thank Drs. Dod Michael and Craig Hammond for critically reading the manuscript and Dr. Jose Caro for encouragement and support of the studies.
Referência(s)