1 The triacylglycerol synthesis enzyme DGAT1 also catalyzes the synthesis of diacylglycerols, waxes, and retinyl esters § ,†, ,† 1, ,† Betty J. Burri, Chi-Liang Eric Yen, and Robert V. Farese, Jr.* Mara Monetti,* * ** † and Department of Gladstone Institute of Cardiovascular Disease,* Cardiovascular Research Institute, Medicine,** University of California, San Francisco, CA 94158; and Western Human Nutrition Research § United States Department of Agriculture, Davis, CA 95616 Center, Abstract The final step of triacylglycerol biosynthesis is ferred to as lipoxicity (1). Thus, an understanding of the catalyzed by acyl CoA:diacylglycerol acyltransferase (DGAT) molecular aspects of triacylglycerol biosynthesis is of great enzymes. The two known DGATs, DGAT1 and DGAT2, are importance. encoded by unrelated genes. Although both DGAT1 and The final step in triacylglycerol biosynthesis, the joining DGAT2 knockout mice have reduced tissue triacylglycerol of diacylglycerol and fatty acyl coenzyme A (CoA), is cata- Downloaded from contents, they have disparate phenotypes, prompting us to lyzed by acyl CoA:diacylglycerol O- acyltransferase (DGAT). investigate whether the two enzymes have unrecognized func- Two DGAT enzymes have been identified. DGAT1, a mem- differences. We now report that DGAT1 exhibits ad- tional ber of a gene family that includes acyl CoA:cholesterol ditional acyltransferase activities in vitro, including those of acyltransferase (ACAT) 1 and ACAT2 (2–4), is a protein of acyl CoA:monoacylglycerol acyltransferase (MGAT), wax mo- noester and wax diester synthases, and acyl CoA:retinol acyl- 500 amino acids in most species (5, 6). It is very hydro- transferase (ARAT), which catalyze the synthesis of diacyl- www.jlr.org phobic, has multiple predicted membrane-spanning do- glycerols, wax esters, and retinyl esters, respectively. These mains (5), and like ACAT1 (7), forms homodimers and activities were demonstrated in in vitro assays with mem- homotetramers (8). DGAT1 is expressed in most tissues, branes from insect cells or homogenates from COS7 cells with the highest expression levels in the small intestine, by guest, on May 8, 2019 overexpressing DGAT1. Wax synthase and ARAT activities testis, adipose tissue, mammary gland, and skin (5). DGAT2, were also demonstrated in intact COS7 cells expressing on the other hand, is a member of a larger gene family DGAT1. Additionally, cells and tissues from DGAT1-deficient whose members include acyl CoA:monoacylglycerol acyl- mice exhibited reduced ARAT activity, and the mice had in- creased levels of unesterified retinol in their livers on a transferase (MGAT) 1, MGAT2, MGAT3, and wax syn- Our findings indicate that DGAT1 can high-retinol diet. thases (9–16). DGAT2 enzymes of different species have utilize a variety of acyl acceptors as substrates in vitro and 400 amino acids, are less hydrophobic than DGAT1, and suggest that these activities may be relevant to the in vivo have only one or two predicted membrane-spanning do- —Yen, C-L. E., M. Monetti, B. J. Burri, functions of DGAT1. mains (10). DGAT2 is expressed highly in tissues that fig- The triacylglycerol synthesis enzyme and R. V. Farese, Jr. ure prominently in triacylglycerol metabolism, including DGAT1 also catalyzes the synthesis of diacylglycerols, waxes, liver and adipose tissues (10). Despite their similar enzy- 1502–1511. J. Lipid Res. 2005. 46: and retinyl esters. matic properties in in vitro assays, DGAT1 and DGAT2 share little, if any, sequence similarity (5, 10). Supplementary key words acyl CoA:diacylglycerol acyltransferase • monoacylglycerol wax ester retinol (vitamin A) • • The in vivo functions of DGAT1 and DGAT2 have been analyzed in gene-targeted mice deficient in either enzyme / (17, 18). DGAT1-deficient ( Dgat1 ) mice are viable, have modest reductions in tissue triacylglycerol content, and Triacylglycerols (triglycerides) are the primary molecules have normal plasma triacylglycerol levels (17). These mice of energy storage in eukaryotic organisms. In mammals, are protected from diet-induced and some genetic forms triacylglycerol biosynthesis is essential for normal physiol- of obesity through an increase in energy expenditure (17, ogy and is prominent in the adipose tissue, liver, small in- testine, and mammary gland, which all store or secrete tri- acylglycerols. However, excessive storage of triacylglycerols in human adipose tissue results in obesity, and in nonadi- Abbreviations: ACAT, acyl CoA:cholesterol acyltransferase; ARAT, pose tissues, it is associated with tissue dysfunction, re- acyl CoA:retinol acyltransferase; CoA, coenzyme A; DGAT, acyl CoA:di- acylglycerol acyltransferase; LRAT, lecithin:retinol acyltransferase; MEF, mouse embryonic fibroblast; MGAT, acyl CoA:monoacylglycerol acyl- Manuscript received 31 January 2005 and in revised form 7 April 2005. transferase; TLC, thin-layer chromatography. 1 To whom correspondence should be addressed. Published, JLR Papers in Press, April 16, 2005. e-mail: [email protected] DOI 10.1194/jlr.M500036-JLR200 This article is available online at http://www.jlr.org Journal of Lipid Research Volume 46, 2005 1502
2 tracted lipids were dried, separated by thin-layer chromatogra- 19). They exhibit increased sensitivity to insulin and lep- phy (TLC) with hexane-ethyl ether-acetic acid (80:20:1; v/v/v), tin and are protected from diet-induced insulin resistance visualized with iodine vapor, and identified by comparison with (19, 20). These phenotypic effects appear to be due in lipid standards. For experiments with radiolabeled substrates, part to altered endocrine function of the white adipose TLC plates were scanned with an imaging scanner (Bioscan AR- / tissue (21). Unexpectedly, mice also have abnor- Dgat1 2000; Washington, D.C.) or exposed to X-ray film, and bands malities in tissue-specific developmental processes. Their were scraped to assess the incorporation of radioactivity into lipid skin is characterized by atrophy of the sebaceous glands products. All assays were performed in duplicate, and experi- and hair loss, which is most prominent in postpubertal ments were repeated two to three times. MGAT (EC 126.96.36.199), DGAT (EC 188.8.131.52), acyl CoA:long- male mice. Their fur lacks wax diesters (22), which are the O- acyltransferase (wax synthase; EC 184.108.40.206), chain fatty alcohol major component of skin surface lipids in mice. Female / O- acyltransferase (ARAT; EC 220.127.116.11) activ- and acyl CoA:retinol Dgat1 mice exhibit a lactation defect and do not secret 14 C]palmitoyl CoA (25 M; ities were detected by incorporation of [ milk due to impaired development of the mammary epi- 20,000 cpm/nmol; Amersham Biosciences, Pis- specific activity, thelium (17, 23). This aspect of the phenotype appears to cataway, NJ) into diacylglycerol, triacylglycerol, wax monoester reflect the absence of DGAT1 both in the mammary epi- sn -2 and wax diester, and retinyl ester in the presence of added thelium and in the surrounding stromal tissue (23). sn -1,2 dioleoylglycerol, 1-hexadecanol, 1,2- monooleoylglycerol, The absence of DGAT2 has a more profound effect on retinol, respectively. Stereoisomers -trans hexadecandiol, and all / triacylglycerol metabolism. DGAT2-deficient ( ) Dgat2 and 13- -trans -trans cis reti- retinol, all and derivatives of retinol (all -trans retinoic acids) were from Sigma-Aldrich. The nal, and all mice have intrauterine growth retardation and die within dependence of MGAT, wax synthase, and ARAT activities on acyl hours after birth (18). They have severely reduced tri- acceptors as substrates was determined with assays using various acylglycerols and lack substrates for energy metabolism. sn -2 monooleoylglycerol, 1-hexa- concentrations of acyl acceptor ( They also lack essential fatty acids and specific skin lipids -trans retinol, respectively) 1,2-hexadecandiol, and all decanol, that are crucial for the function of the epidermal water 14 M [ C]palmitoyl CoA (specific activity, 20,000 cpm/ and 50 Downloaded from barrier (18). In yeast, the DGAT2 homolog Dga1p ac- nmol). The acyl acceptors were serially diluted beforehand to counts for the majority of triacylglycerol synthesis (24), maintain the percentage of acetone in each reaction at 5%. Like- whereas the DGAT1 homologs Are1p and Are2p account wise, the dependence of MGAT, wax monoester synthase, and ARAT activity on acyl donor as substrates was determined with as- for only a minor portion (25). says using various concentrations of palmitoyl CoA and the re- The disparate phenotypes of DGAT1- and DGAT2-defi- 3 -2 mo- M [ H] sn spective radiolabeled acyl acceptor (200 cient mice prompted us to investigate whether the two en- www.jlr.org 3,000 cpm/nmol; 100 M nooleoylglycerol, specific activity zymes have previously unrecognized differences in their 14 C]1-hexadecanol, specific activity 6,000 cpm/nmol; 50 M [ biochemical functions. In this study, we sought to deter- 3 -trans H]all 330,000 cpm/nmol; Amer- retinol, specific activity [ mine whether DGAT1 and DGAT2 have other substrates ican Radiolabeled Chemicals, St. Louis, MO). by guest, on May 8, 2019 and, more specifically, whether either DGAT can use addi- Expression studies in mammalian cells tional fatty acyl acceptors as substrates. The FLAG-tagged mouse DGAT1 cDNA was subcloned into a pcDNA3 expression vector and transfected into COS-7 cells with Fugene 6 (Roche Diagnostics; Chicago, IL). The human DGAT1 MATERIALS AND METHODS coding sequence (accession number NM_012079) was amplified from human liver cDNA (Clontech; Palo Alto, CA) by PCR Expression studies in insect cells (Takara Ex Taq; Panvera, Madison, WI) and was subcloned with- Mouse DGAT1 (accession number AF078752) and DGAT2 (ac- -Galactosi- out the FLAG tag into pIRESneo2 vector (Clontech). cession number AF384160) were tagged with an N-terminal FLAG dase, FLAG-tagged mouse ACAT2 (accession number AF078751), , epitope underlined) and expressed epitope (MGDYKDDDDG DGAT2, MGAT1 (accession number AF078752), and human (Sf9) insect cells by baculovirus infections Spodoptera frugiperda in MGAT2, (accession number AY157608) were expressed as con- as described (5, 10–12). Wild-type baculovirus infections were trols. Cells were harvested 48 h after transfection. Expression of used as a negative control. Expression of DGAT1 and DGAT2 FLAG-tagged proteins was analyzed by immunoblotting of ho- proteins was demonstrated by immunoblotting of membrane g of protein) with the anti-FLAG antibody. The mogenates (20 g) with an anti-FLAG M2 antibody (Sigma-Aldrich proteins (5 in vitro acyltransferase activities of total homogenates were as- Co.; St. Louis, MO). sayed with labeled palmitoyl CoA as described above for proteins expressed in insect cells. In vitro acyltransferase assays Wax and retinol esterification in cultured cells For initial screening of acyl aceptors, acyltransferase activities 14 M acyl acceptor as M acyl donor and 100 were assayed with 25 C]1-hexadecanol into wax monoester The incorporation of [ 3 described (11). Nonpolar acyl acceptors (diacylglycerol, mono- H]retinol into retinyl esters in cultured cells was mea- and of [ acylglycerol, 1-hexadecanol, 1,2-hexadecandiol, cholesterol, and sured in COS-7 cells overexpressing DGAT1. COS-7 cells were 5 retinol) were dissolved in acetone, which accounted for 5% of /well) for 24 h and transfected with plated in six-well dishes (10 l reactions. Retinol was dissolved in dimethylsulfoxide in 200 vectors containing cDNAs for DGAT1 or control enzymes; 24 h 14 some experiments, and results were identical to those obtained C]1- M [ later, cells were incubated in medium containing 5 with retinol dissolved in acetone. Reactions were started by add- 5,500 cpm/nmol) or 2.5 M hexadecanol (specific activity, 3 ing membranes (from insect cells) or total protein homogenates H]retinol (specific activity, 200,000 cpm/nmol) for 18 h. [ (from mammalian cells or mouse tissues) and stopped after 10 Neutral lipids were extracted, separated by TLC, visualized with a min (increases in reaction products were linear for up to 30 Bioscan imaging scanner, and scraped to assess the incorpora- min) by adding 4 ml of chloroform-methanol (2:1; v/v). The ex- tion of radioactivity into wax and retinyl esters. DGAT1 is a multifunctional enzyme 1503 Yen et al.
3 NC) (27). In one experiment, 12-week-old male wild-type and In vitro assay and labeling studies with mouse / Dgat1 mice were maintained on a high-vitamin A diet (contain- embryonic fibroblasts ing 160 IU/g of retinyl palmitate and 10 kcal% fat, D99102101; Mouse embryonic fibroblasts (MEFs) were established from Research Diets, New Brunswick, NJ) for 3 weeks, and tissue reti- DGAT1-deficient embryos and their wild-type littermates as de- noid levels were measured by Craft Technologies. scribed (10). The in vitro acyltransferase activities of total homog- enates from primar y MEFs (passage 3) were assayed with labeled palmitoyl CoA as described above for insect proteins. For label- ing of intact cells, primary MEFs were plated in triplicate (4 RESULTS 5 10 /100 mm dish) and, 24 h later, incubated with medium con- 3 M [ 70,000 cpm/nmol) H]retinol (specific activity, taining 5 DGAT1 expressed in insect cells possesses multiple for 18 h (similar results, with proportionally less incorporation of acyltransferase activities radioactivity, were observed after 6 h of incubation). Neutral lip- To examine the acyltransferase activities of DGAT1 and ids were analyzed as described above. DGAT2, FLAG-tagged versions of their murine cDNAs Measurements of retinol and retinyl esters were expressed in insect cells. These proteins migrated on Retinoid levels were measured with reverse-phase HPLC meth- SDS-PAGE at the expected molecular masses, and the lev- ods. For MEFs, cellular retinol levels were measured in three els of expression were comparable, as demonstrated by 6 10 cells from each cell line. Cells were cul- samples of 3–10 immunoblotting with an anti-FLAG antibody ( ). In Fig. 1A tured in media containing 10% fetal bovine serum that con- control experiments, membranes expressing either DGAT M retinol and 0.25 M retinyl esters. For mouse stud- tained 0.5 14 / incorporated more [ C]palmitoyl CoA into triacylglycer- mice (19 weeks old; Dgat1 96% ies, male wild-type and ols than did membranes from cells infected with wild-type C57BL/6J genetic background) were maintained on a regular chow diet (containing 10 kcal% fat, PicoLab 5053; Purina Mill, baculovirus (Fig. 1B), verifying that both expressed enzymes Brentwood, MO) and then either kept on chow or switched to a were active. As reported previously (11, 12), both DGAT1 high-fat diet (containing 45 kcal% fat, TD88137; Harlan, Madi- Downloaded from and DGAT2 also possess in vitro MGAT activity, incorpo- son, WI) for 1 week before tissues were harvested. The chow diet 14 rating [ C]palmitoyl CoA into diacylglycerol and triacyl- 10 IU/g of vitamin A (mainly retinyl acetate), and contained glycerols in the presence of monoacylglycerol (Fig. 1B). 20 IU/g of vitamin A (mainly reti- the high-fat diet contained We next explored whether DGAT1 or DGAT2 could uti- nyl palmitate). 14 lize other acyl acceptors as substrates in vitro. With [ C] Retinoid levels in MEFs and mouse livers were measured in palmitoyl CoA as the acyl donor, membranes expressing the laboratory of B. J. Burri (USDA, Davis, CA) (26) and levels in www.jlr.org DGAT1, but not those expressing DGAT2 or control viral diets and serum were measured by Craft Technologies (Wilson, by guest, on May 8, 2019 Expression of acyl CoA:diacylglycerol acyltransferase 1 (DGAT1) and DGAT2 in Sf9 insect cells. A: Immunoblots of insect cell mem- Fig. 1. g) branes. Expression of DGAT1 and DGAT2 was demonstrated by immunoblotting with an anti-FLAG antibody. Membrane proteins (5 from Sf9 cells infected with wild-type virus, FLAG-tagged DGAT1 (DGAT1), or FLAG-tagged-DGAT2 (DGAT2) recombinant baculoviruses were analyzed. B–D: Acyltransferase activities conferred by expression of DGAT1. Acyl CoA:monoacylglycerol acyltransferase (MGAT) and ncorpora- DGAT (B), wax monoester and diester synthase (C), and acyl CoA:retinol acyltransferase (ARAT) (D) activities were detected by i 14 tion of [ C]palmitoyl coenzyme A (CoA) into diacylglycerol (DAG), triacylglycerol (TAG), wax monoester (ME), wax diester (DE), and ret- inyl ester (RE) in the presence of 100 M added sn -2 monooleoylglycerol (MAG), sn -1,2 dioleoylglycerol (DAG, top label), 1-hexadecanol (Mono-ol), 1,2-hexadecandiol (Di-ol), and retinol (all and 13- cis retinol), respectively. FFA, free fatty acid. The lipid products were ex- -trans tracted, dried, separated by thin-layer chromatography, and detected by exposure to X-ray film as described. Arrows indicate the respective product(s) of each acyltransferase activity. In (C), the prominent acylation product when 1,2-hexadecandiol was provided is most likely the intermediate 2-hydroxylhecadecyl hexadecanoate (lower arrow). Chromatographs shown are representative of three independent expe ri- ments. 1504 Journal of Lipid Research Volume 46, 2005
4 proteins, catalyzed the incorporation of the acyl group all -trans retinol for ARAT activity). In each case, the acyl- into wax monoesters, wax diesters, and retinyl esters in the transferase activity increased with increasing acyl CoA presence of 1-hexadecanol, 1,2-hexadecandiol, and reti- C ). Again, the activity was highest Fig. 3A – concentrations ( nol, respectively (Fig. 1C, D). Both all with sn- 2-monoacylglycerol as the acyl acceptor, and the retinol and 13 -trans - levels of this MGAT activity did not reach plateau even cis retinol served as substrates for the DGAT1-mediated when the concentration of palmitoyl CoA was as high as retinol acyltransferase activity, but all- retinal and trans 200 M (Fig. 3A–C). Because these assays utilized radiola- retinoic acid did not, consistent with a requirement for a beled acyl acceptors as substrates, they provide strong evi- free hydroxyl group on the end of the retinoid side chain dence that DGAT1 possesses in vitro MGAT, wax synthase, (Fig. 1D). Cholesterol, glycerol, glycerol-3-phosphate, and and ARAT activities. 1-acylglycerol-3-phosphate did not serve as substrates (not shown). DGAT1 possesses wax synthase and ARAT activities in To examine whether the acyltransferase activities of mammalian cells DGAT1 depend on the presence of appropriate substrates, To determine whether DGAT1 possesses wax synthase we first varied the acyl acceptor concentrations in assays 14 and ARAT activities in mammalian cells, we transiently over- containing 50 M [ C]palmitoyl CoA. In general, enzy- expressed mouse DGAT1 and control enzymes in COS-7 matic activity increased with the concentration of each cells and assessed their enzymatic activities in cellular ho- acyl acceptor ( Fig. 2A – D ). Although these experiments mogenates. The expression of FLAG-tagged DGAT and con- were not designed to calculate apparent K , un- V and m max Fig. trol proteins was demonstrated by immunoblotting ( der the assay condition used, we found that the maximal 14 ). In control assays with diacylglycerol and [ 4A C]palmi- enzymatic activity measured was highest with monoacyl- toyl CoA as the tracer, homogenates of COS-7 cells overex- glycerol as the acceptor, a condition that reflects both pressing DGAT1 or DGAT2 incorporated more radioactiv- MGAT and DGAT activities (Fig. 2A). The apparent maxi- Downloaded from ity into triacylglycerols than did controls overexpressing mal activities observed with other acceptors were 10% of other enzymes, including -galactosidase (LacZ), ACAT2, that for monoacylglycerol but were still considerable (be- MGAT1, and MGAT2 (Fig. 4B). Homogenates of cells tween 0.5 and 1.5 nmol/min/mg protein) (Fig. 2A–D). 14 overexpressing ACAT2 incorporated more [ C]palmitoyl We also examined these acyltransferase activities at dif- CoA into cholesterol esters than did the other enzymes. ferent concentrations of palmitoyl CoA and a fixed con- With 1-hexadecanol as the acyl acceptor, only homoge- centration of radiolabeled acyl acceptor as tracers (200 M www.jlr.org 3 nates from cells expressing DGAT1 exhibited wax syn- [ H] sn- 2-monooleoylglycerol for MGAT activity, 100 M 3 14 trans retinol as thase activity (Fig. 4C). Similarly, with all- [ H] C]1-hexadecanol for wax synthase activity, and 50 M [ by guest, on May 8, 2019 Dependency of acyltransferase activities in mem- Fig. 2. brane expressing DGAT1 on acyl acceptor concentra- tions. A: Acylglycerol ac yltransferase (both DGAT and 14 MGAT) activities assessed by the amount of [ C]palmi- toyl CoA incorporated into both diacylglycerol and tri- acylglycerol in the presence of different concentrations of sn -2 monooleoylglycerol. B: Wax monoester synthase 14 activity assessed by the amount of [ C]palmitoyl CoA incorporated into wax monoester in the presence of dif- ferent concentrations of 1-hexadecanol. C: Wax diester 14 synthase activity assessed by the amount of [ C]palmi- toyl CoA incorporated into wax diester in the presence of different concentrations of 1,2-hexadecandiol. D: 14 ARAT activity assessed by the amount of [ C]palmitoyl CoA incorporated into retinyl ester in the presence of different concentrations of all -trans retinol. Values are the mean of duplicate measurements and are represen- tative of two experiments. 1505 Yen et al. DGAT1 is a multifunctional enzyme
5 cells overexpressing mouse DGAT1 and control enzymes ( Fig. 5A ). (We did not examine MGAT activity in intact cells overexpressing DGAT1 because monoacylglycerol, the MGAT reaction substrate, can be hydrolyzed before up- take by cells, and because diacylglycerol, the MGAT reac- tion product, can be converted to other lipids and does not accumulate in significant amounts.) Wax ester synthe- 10-fold sis was highest in cells overexpressing DGAT1 ( of LacZ controls; Fig. 5B). Cells expressing DGAT2, and to a lesser extent cells overexpressing MGAT1, also synthe- sized more wax esters than did control cells. Retinol ester- ification levels were highest in cells overexpressing DGAT1 ( 5-fold of LacZ controls; Fig. 5C). Cells overexpressing MGAT1 also exhibited higher levels of retinol esterfica- tion than did control cells. In addition, we examined COS-7 cells expressing hu- man DGAT1 without the FLAG tag epitope. Like mouse DGAT1, human DGAT1 exhibited wax synthase and ARAT activities both in in vitro assays and in intact cell esterifica- tion assays (not shown). These findings also showed that the presence of the FLAG tag at the N-terminus of DGAT1 did not affect these additional acyltransferase activities. Downloaded from Cells and tissues lacking DGAT1 exhibit reduced ARAT activity Because the genes encoding ARAT enzymes have not been identified, we further characterized the ARAT activ- www.jlr.org ity of DGAT1 by examining whether ARAT activity was di- minished in mouse cells and tissues lacking DGAT1. We / first examined whether primary MEFs from Dgat1 mice by guest, on May 8, 2019 have decreased in vitro ARAT activity. In a control experi- 14 ment with [ C]palmitoyl CoA as the acyl donor and diac- ylglycerol as the acceptor, homogenates from two indepen- / dent lines of Dgat1 MEFs exhibited less triacylglycerol synthesis than did homogenates from wild-type MEFs, as 14 expected (not shown). With [ C]palmitoyl CoA as the acyl donor and all- -retinol as the acyl acceptor, ARAT trans activity was detectable in homogenates of three different ). In contrast, ARAT activ- wild-type MEF cell lines ( Fig. 6A Fig. 3. Dependency of acyltransferase activities in membrane ex- ity was nearly absent in homogenates of two different lines pressing DGAT1 on acyl donor concentrations. A: MGAT activity as- / 3 of Dgat1 MEFs. sessed by the amount of [ H] sn -2 monooleoylglycerol incorporated into both diacylglycerol and triacylglycerol in the presence of dif- We next examined retinol esterification in intact wild- ferent concentrations of palmitoyl CoA. B: Wax monoester synthase / type and Dgat1 MEFs by measuring the accumulation 14 activity assessed by the amount of [ C]1-hexadecanol incorporated of radioactivity in retinyl esters after 18 h incubation with into wax monoester in the presence of different concentrations of 3 [ H]all -retinol. Levels of radiolabeled retinyl esters -trans 3 palmitoyl CoA. C: ARAT activity assessed by the amount of [ - H]all / in two different cell lines of Dgat1 MEFs were less than trans retinol incorporated into retinyl ester in the presence of dif- ferent concentrations of palmitoyl CoA. Values are the mean of du- one-half of those in wild-type MEF lines (Fig. 6B). We ex- plicate measurements and are representative of two experiments. cluded a dilution effect of the tracer as a cause of the re- / Dgat1 duced retinol esterification in MEFs by measur- / Dgat1 ing the retinol content in wild-type and MEFs. We found that the retinol levels were similar in wild-type the acceptor, homogenates from cells expressing DGAT1 / Dgat1 and 0.004 pmol/ 1.25 vs. 1.45 MEFs [1.68 exhibited ARAT activity (Fig. 4D). In some experiments, 6 10 cell (mean SD)] and that the basal cellular retinol homogenates from cells expressing MGAT1 also exhibited levels were 3 orders of magnitude lower than the levels of very low levels of ARAT activity. 6 labeled retinol taken up by cells (0.74–3.06 pmol/10 cell To determine whether DGAT1 exhibits wax synthase 6 vs. 2–3 nmol/10 cell). The residual retinol esterification and ARAT activities in intact cells, we measured the accu- / 14 Dgat1 in MEFs is presumably due to a nonARAT activ- mulation of radiolabeled substrates ([ C]1-hexadecanol 3 ity, possibly that of lecithin:retinol acyltransferase (LRAT), or [ retinol) in esterification products in COS-7 -trans H]all Journal of Lipid Research Volume 46, 2005 1506
6 Fig. 4. Expression of DGAT1 in mammalian cells confers wax synthase and ARAT activity in in vitro assays. A: Immunoblotting of FLAG-tagged proteins demonstrating expression of FLAG-tagged DGAT1 and control proteins. Cells were transfected with expression vectors containing cDNAs for -galactosidase (LacZ) or FLAG-tagged versions of acyl CoA:cholesterol acyltransferase 2 (ACAT2), DGAT1, DGAT2, MGAT1, or MGAT2. B–D: DGAT (B), wax synthase (C), and ARAT (D) activities in COS-7 cell homogenates conferred by expres- 14 sion of DGAT1. [ C]palmitoyl CoA was used for enzyme assays. Arrows indicate the incorporation of 14 [ C]palmitoyl CoA into triacylglycerol (TAG), wax monoester (ME), and retinyl ester (RE). CE, cholesterol Downloaded from ester. FFA, free fatty acid. Chromatographs shown are representative of three independent experiments. additional activities were not found for DGAT2, with the which uses lecithin (phosphatidylcholine) instead of fatty exception of increased wax synthesis in intact cells overex- acyl CoA as the acyl donor. pressing DGAT2, which was reported previously (15). We next examined in vitro ARAT activities in liver, tes- www.jlr.org / Thus, unlike DGAT2, which appears to function primarily Dgat1 tis, and kidney homogenates from wild-type and as a DGAT, DGAT1 possesses multiple acyltransferase ac- mice. In control assays, ACAT activities were similar in ho- / tivities. mogenates from wild-type and Dgat1 tissues ( ). Fig. 7A by guest, on May 8, 2019 / These results are consistent with the evolutionary his- In Dgat1 tissue homogenates, DGAT activity was re- tory of the two enzymes. The DGAT2 homolog in Saccharo- duced by more than 80% (Fig. 7B) and ARAT activity was myces cerevisiae , Dga1p, plays a major role in triacylglycerol reduced by 75–90% (Fig. 7C). synthesis (24, 28), whereas the DGAT1 homologs, Are1p Finally, we assessed whether retinol metabolism is per- / and Are2p, play minor roles (24). Instead, Are1p and Dgat1 turbed in mice. Mice were fed diets containing Are2p are involved in sterol esterification (29). Because different levels of retinol, and retinol and retinyl esters the relatives of DGAT1 in lower organisms were not pri- were measured in the liver, where the majority of retinol is marily involved in esterifying diacylglycerol, DGAT1 must 10 IU/g retinol), stored. In mice fed standard chow ( have acquired the ability to utilize diacylglycerol as a sub- the hepatic content of unesterified retinol was similar in / strate in higher organisms during evolution. Therefore, it wild-type and Dgat1 ). However, in mice Fig. 8A mice ( is plausible that the substrate recognition site of DGAT1 is 20 IU/g retinol) or a high-ret- fed either a high-fat diet ( less discriminating than the corresponding site of DGAT2. inol diet (160 IU/g retinol), unesterified retinol was 2- to / Moreover, DGAT1 may have also acquired the ability to 3-fold more abundant in Dgat1 livers (Fig. 8A). Inter- utilize other fatty acyl acceptors, such as monoacylglyc- estingly, the hepatic content of retinyl esters was not re- / erol, long-chain alcohols, and retinol. duced in Dgat1 livers from mice fed any of the diets, The demonstration that DGAT1 possesses MGAT activ- and in fact, retinyl ester levels were significantly higher in / ity, as noted briefly in reports of the cloning of MGAT Dgat1 livers of mice fed the high-fat diet (Fig. 8B). genes (11, 12), suggests that DGAT1 can catalyze the se- quential esterification of two fatty acyl moieties to convert monoacylglycerol to diacylglycerol and then to triacylglyc- DISCUSSION erol. This ability is consistent with a role for DGAT1 in the In this study, we show that mammalian DGAT1 pos- reesterification of monoacylglycerol and diacylglycerol sesses MGAT, wax synthase, and ARAT activities as well as generated by triacylglycerol hydrolysis in cells. Whether DGAT activity in in vitro assays of membranes of insect this is a function of DGAT1 in intact cells requires further cells or homogenates of mammalian cells overexpressing testing. DGAT1. Wax synthase and ARAT activities catalyzed by DGAT1 catalyzed the synthesis of both wax monoesters DGAT1 were demonstrated in intact cells, and ARAT activ- and wax diesters. The combination of wax monoester syn- ity was reduced in DGAT1-deficient cells and tissues. These thase and DGAT activities is similar to that of an unrelated DGAT1 is a multifunctional enzyme 1507 Yen et al.
7 Downloaded from www.jlr.org Mouse embryonic fibroblasts (MEFs) lacking DGAT1 ex - Fig. 6. by guest, on May 8, 2019 hibit reduced ARAT activity. A: Reduced in vitro ARAT activities in MEF homogenates lacking DGAT1. Primary MEFs were established from DGAT1-deficient embryos and their wild-type littermates as described (10). Each number represents an independent line o f 14 C]palmitoyl CoA was used for enzyme assays. Values MEF cells. [ are mean SD of four measurements. B: Decreased incorporation of retinol into retinyl esters in MEFs lacking DGAT1. Cells were plated in triplicate and, 24 h later, incubated in medium containing 3 M [ 5 H]retinol (specific activity, 70,000 cpm/nmol) for an ad- SD of measurements from three ditional 18 h. Values are mean Fig. 5. Expression of DGAT1 in mammalian cells increases esteri- plates of each clone and are representative of two repeated experi- fication of fatty acyl alcohol and retinol in intact cells. A: Immuno- ments. blotting of FLAG-tagged proteins demonstrating expression of FLAG- tagged DGAT1 and control proteins. B and C: Increased incorpora- tion of fatty acyl alcohol into wax esters and of retinol into retinyl esters in COS-7 cells overexpressing DGAT1. Cells were plated in tivity of DGAT1 may help to explain the deficiency of type six-well plates, transfected with DGAT1 or control proteins, and 24 h / 14 II wax diesters in the fur lipids of Dgat1 mice (22). M [ later, incubated in medium containing 5 C]1-hexadecanol 3 M [ (specific activity, 5,500 cpm/nmol) or 2.5 H]retinol (spe- However, because wax diesters are present in the fur of / cific activity, 200,000 cpm/nmol) for 18 h. Neutral lipids were leptin-deficient Dgat1 mice (22), other wax synthases separated by TLC, visualized with a Bioscan imaging scanner, and must exist. Our results show that DGAT2 overexpression scraped to assess the incorporation of radioactivity. Values repre- also increased wax monoester synthase activity in intact sent results from six pooled six-well dishes of each group. Similar cells, a finding that was recently reported (15), although results were observed in three independent experiments. we did not detect either wax monoester or diester syn- thase activity in in vitro assays of DGAT2. Additional wax wax synthase/DGAT found in bacteria (30). It is not synthases (members of the DGAT2 gene family) that cata- known whether, like DGAT1, the bacterial bifunctional lyze the synthesis of wax monoesters (15, 16), but not wax enzyme is active in forming wax diesters. The wax mo- diesters, were recently identified (C-L. Yen et al., unpub- noester synthase activity of DGAT1 agrees with a recent re- lished observations). port showing that DGAT1 overexpression in intact cells Although ARAT activity was reported decades ago (31– increases wax synthesis (15). The wax diester synthase ac- 34), the ARAT genes have not been identified. The cur- Journal of Lipid Research Volume 46, 2005 1508
8 Downloaded from Fig. 8. Retinol and retinyl ester content in the livers of DGAT1- / deficient mice. Wild-type and Dgat1 mice were fed a regular chow (containing 10 IU/g retinol), a high-fat diet (HF, 20 IU/g ret- www.jlr.org inol), or a high-vitamin A diet (HA, 160 IU/g retinol). Levels of ret- inol (A) and retinyl esters (B) were measured by HPLC methods. Values are mean SEM. (The number of mice is indicated in the bars.) * P 0.05 versus wild-type mice (two-way ANOVA followed by guest, on May 8, 2019 by Tukey test for multiple comparisons). Fig. 7. Reduced in vitro ARAT activities in liver, testis, and kidney bound to cytosolic retinol binding protein, whereas ARAT of DGAT1-deficient mice. Tissue homogenates were from wild-type prefers unbound retinol as a substrate. Therefore, ARAT / Dgat1 and mice fed a regular chow. The activities of ACAT (A), may play a role in retinol metabolism in tissues with low 14 DGAT (B), and ARAT (C) were assessed by the incorporation of [ C] levels of cytosolic retinol binding protein or when cellular M) into cholesterol esters, retinyl ester, and tri- palmitoyl CoA (25 retinol concentrations are relatively high (40), thereby acylglycerol, respectively. Values are mean SD of measurements protecting cells from toxicity resulting from excess unes- from four mice in each group. 0.001 versus wild-type mice P * (two-way ANOVA followed by Tukey test for multiple comparisons). terified retinol. Our data agree with this possible function for ARAT, because hepatic levels of unesterified retinol / Dgat1 were increased in mice challenged with high-ret- rent findings suggest that DGAT1, which is expressed in inol diets. The high levels of unesterified retinol in the liv- / tissues reported to have ARAT activity, contributes to Dgat1 ers of mice also raise the possibility that al- ARAT activity in vivo. DGAT1 exhibited ARAT activity in terations in retinoid metabolism may contribute to the in vitro assays and accounted for a large portion of ARAT phenotype of mice lacking DGAT1. Retinol is the precur- activity in the liver, testis, and kidney of mice. DGAT1 defi- sor for retinoic acids (all retinoic and its isomer 9 -cis -trans ciency also perturbed retinol metabolism in the livers of acids), which are ligands for the nuclear hormone recep- / Dgat1 mice. The physiological significance of ARAT, tors retinoic acid receptor and retinoid-X receptor (41– however, is not clear. Retinyl esters can also be synthesized 43), and retinoids affect sebaceous gland function (44, by LRAT (32, 35–38). LRAT has a significantly lower K 45), hair growth (46), and proliferation of mammary m 15–30 for retinol than does ARAT ( 2 M), sug- M vs. gland epithelial cells (47, 48), all of which are part of phe- / gesting that LRAT plays the major role in retinol storage notypic changes observed in Dgat1 mice (17, 22, 23). when cellular retinol concentrations are relatively low. In- Consistent with LRAT being the major enzyme respon- deed, mice lacking LRAT have severe reductions in retinyl sible for maintaining retinyl ester levels in mouse liver ester content in liver, blood, and eyes and exhibit abnor- (39), hepatic retinyl ester content was not reduced in / malities of the retinal pigmented epithelium (39). The Dgat1 livers. In fact, the hepatic retinyl ester content in / preferred substrate for LRAT is believed to be retinol Dgat1 mice on a high-fat diet was, paradoxically, in- DGAT1 is a multifunctional enzyme 1509 Yen et al.
9 erol acyltransferase expressed in the small intestine. J. Biol. Chem. creased. The explanation for this finding is unclear at 278: 18532–18537. present. It is possible that the increases in retinyl ester 13. Cheng, D., T. C. Nelson, J. Chen, S. G. Walker, J. Wardwell-Swan- content resulted from increased LRAT activity or de- son, R. Meegalla, R. Taub, J. T. Billheimer, M. Ramaker, and J. N. Feder. 2003. Identification of acyl coenzyme A:monoacylglycerol creased hydrolysis of retinyl esters. acyltransferase 3, an intestinal specific enzyme implicated in di- In summary, although both DGAT1 and DGAT2 share J. Biol. Chem. etary fat absorption. 13611–13614. 278: diacylglycerol as a common fatty acyl acceptor, the two en- 14. Cao, J., J. Lockwood, P. Burn, and Y. Shi. 2003. Cloning and func- zymes differ in their ability to use other acyl acceptors. tional characterization of a mouse intestinal acyl-CoA:monoacyl- glycerol acyltransferase, MGAT2. 13860–13866. J. Biol. Chem. 278: DGAT1, much more than DGAT2, can catalyze esterifica- 15. Cheng, J. B., and D. W. Russell. 2004. Mammalian wax biosynthe- tion reactions utilizing a variety of fatty acyl acceptors and sis. II. Expression cloning of wax synthase cDNAs encoding a mem- is, therefore, a less-discriminating enzyme than DGAT2. ber of the acyltransferase enzyme family. J. Biol. Chem. 279: 37798– 37807. Our findings indicate that DGAT1 is a multifunctional 16. Turkish, A. R., A. L. Henneberry, D. Cromley, M. Padamsee, P. acyltransferase, and these additional activities of DGAT1 Oelkers, H. Bazzi, A. M. Christiano, J. T. Billheimer, and S. L. Stur- may be relevant to its in vivo functions. ley. 2005. Identification of two novel human Acyl-CoA wax alcohol acyltransferases: members of the diacylglycerol acyltransferase 2 J. Biol. Chem. (DGAT2) gene superfamily. 14755–14764. 280: 17. Smith, S. J., S. Cases, D. R. Jensen, H. C. Chen, E. Sande, B. Tow, The authors thank S. Cases for providing MEF cells, T. Neidlinger D. A. Sanan, J. Raber, R. H. Eckel, and R. V. Farese, Jr. 2000. Obe- (USDA) for measuring retinoids, and G. Gildengorin (USDA) sity resistance and multiple mechanisms of triglyceride synthesis in for statistical analysis of these measurements; S. Erickson, Y. 25: Nat. Genet. 87–90. mice lacking DGAT. 18. Stone, S. J., H. Myers, B. E. Brown, S. M. Watkins, K. R. Feingold, Huang, R. Mahley, K. Weisgraber, and M. Zeiger for comments P. M. Elias, and R. V. Farese, Jr. 2004. Lipopenia and skin barrier on the manuscript; S. Ordway and G. Howard for editorial as- 279: 11767– abnormalities in DGAT2-deficient mice. J. Biol. Chem. sistance; and B. Taylor for manuscript preparation. This work 11776. was supported by National Institutes of Health Grant DK-56084 19. Chen, H. C., S. J. Smith, Z. Ladha, D. R. Jensen, L. D. Ferreira, (R.V.F.), an American Heart Association fellowship (C-L.E.Y.), L. K. Pulawa, J. G. McGuire, R. E. Pitas, R. H. Eckel, and R. V. Farese, Downloaded from Jr. 2002. Increased insulin and leptin sensitivity in mice lacking and the J. David Gladstone Institutes. J. Clin. Invest. acyl CoA:diacylglycerol acyltransferase 1. 1049– 109: 1055. 20. Chen, H. C., Z. Ladha, and R. V. Farese, Jr. 2002. Deficiency of acyl coenzyme a:diacylglycerol acyltransferase 1 increases leptin sensi- REFERENCES 143: tivity in murine obesity models. Endocrinology. 2893–2898. 21. Chen, H. C., D. R. Jensen, H. M. Myers, R. H. Eckel, and R. V. www.jlr.org 1. Unger, R. H., and L. Orci. 2002. Lipoapoptosis: its mechanism and Farese, Jr. 2003. Obesity resistance and enhanced glucose metabo- 202–212. 1585: Biochim. Biophys. Acta. its diseases. lismin in mice transplanted with white adipose tissue lacking acyl 2. Chang, C. C. Y., H. Y. Huh, K. M. Cadigan, and T. Y. Chang. 1993. 111: 1715–1722. CoA:diacylglycerol acyltransferase 1. J. Clin. Invest. Molecular cloning and functional expression of human acyl-coen- 22. Chen, H. C., S. J. Smith, B. Tow, P. M. Elias, and R. V. Farese, Jr. zyme A:cholesterol acyltransferase cDNA in mutant Chinese ham- 2002. Leptin modulates the effects of acyl CoA:diacylglycerol acyl- by guest, on May 8, 2019 20747–20755. 268: ster ovary cells. J. Biol. Chem. J. Clin. transferase deficiency on murine fur and sebaceous glands. 3. Cases, S., S. Novak, Y-W. Zheng, H. M. Myers, S. R. Lear, E. Sande, 175–181. 109: Invest. C. B. Welch, A. J. Lusis, T. A. Spencer, B. R. Krause, et al. 1998. 23. Cases, S., P. Zhou, J. Schillingford, B. Wiseman, J. Fish, C. S. Angle, ACAT-2, a second mammalian acyl-CoA:cholesterol acyltransferase. L. Hennighausen, Z. Werb, and R. Farese, Jr. 2004. Development Its cloning, expression, and characterization. J. Biol. Chem. 273: of the mammary gland requires DGAT1 expression in stromal and 26755–26764. 131: 3047–3055. Development. epithelial tissues. 4. Oelkers, P., A. Behari, D. Cromley, J. T. Billheimer, and S. L. Stur- 24. Oelkers, P., D. Cromley, M. Padamsee, J. T. Billheimer, and S. L. ley. 1998. Characterization of two human genes encoding acyl co- Sturley. 2002. The DGA1 gene determines a second triglyceride enzyme A:cholesterol acyltransferase-related enzymes. J. Biol. Chem. 8877–8881. J. Biol. Chem. synthetic pathway in yeast. 277: 273: 26765–26771. 25. Oelkers, P., A. Tinkelenberg, N. Erdeniz, D. Cromley, J. T. Bill- 5. Cases, S., S. J. Smith, Y-W. Zheng, H. M. Myers, S. R. Lear, E. heimer, and S. L. Sturley. 2000. A lecithin cholesterol acyltrans- Sande, S. Novak, C. Collins, C. B. Welch, A. J. Lusis, S. K. Erickson, ferase-like gene mediates diacylglycerol esterification in yeast. J. and R. V. Farese, Jr. 1998. Identification of a gene encoding an acyl Biol. Chem. 15609–15612. 275: CoA:diacylglycerol acyltransferase, a key enzyme in triacylglycerol 26. Burri, B. J., T. R. Neidlinger, A. O. Lo, C. Kwan, and M. R. Wong. Proc. Natl. Acad. Sci. USA. 13018–13023. 95: synthesis. 1997. Supercritical fluid extraction and reversed-phase liquid chro- 6. Buhman, K. K., H. C. Chen, and R. V. Farese, Jr. 2001. The en- matography methods for vitamin A and beta-carotene heterogeneous zymes of neutral lipid synthesis. J. Biol. Chem. 40369–40372. 276: J. Chromatogr. A. 762: 201–206. distribution of vitamin A in the liver. 7. Yu, C., J. Chen, S. Lin, J. Liu, C. C. Y. Chang, and T-Y. Chang. 1999. 27. Craft, N. E., T. B. Haitema, K. M. Garnett, K. A. Fitch, and C. K. Human acyl-CoA:cholesterol acyltransferase-1 is a homotetrameric Dorey. 2004. Carotenoid, tocopherol, and retinol concentrations 274: 36139–36145. enzyme in intact cells and in vitro. J. Biol. Chem. 156–162. 8: in elderly human brain. J. Nutr. Health Aging. 8. Cheng, D., R. L. Meegalla, B. He, D. A. Cromley, J. T. Billheimer, Sorger, D., and G. Daum. 2002. Synthesis of triacylglycerols by the 28. and P. R. Young. 2001. Human acyl-CoA:diacylglycerol acyltrans- acyl-coenzyme A:diacyl-glycerol acyltransferase Dga1p in lipid par- ferase is a tetrameric protein. Biochem. J. 359: 707–714. Saccharomyces cerevisiae. ticles of the yeast J. Bacteriol. 519–524. 184: 9. Lardizabal, K. D., J. T. Mai, N. W. Wagner, A. Wyrick, T. Voelker, 29. Yang, H., M. Bard, D. A. Bruner, A. Gleeson, R. J. Deckelbaum, G. and D. J. Hawkins. 2001. DGAT2 is a new diacylglycerol acyltrans- Aljinovic, T. M. Pohl, R. Rothstein, and S. L. Sturley. 1996. Sterol ferase gene family. Purification, cloning, and expression in insect Science. 1353–1356. 272: esterification in yeast: a two-gene process. with diacyl- Mortierella ramanniana cells of two polypeptides from 30. Kalscheuer, R., and A. Steinbuchel. 2003. A novel bifunctional wax glycerol acyltransferase activity. 38862–38869. 276: J. Biol. Chem. ester synthase/acyl-CoA:diacylglycerol acyltransferase mediates wax 10. Cases, S., S. J. Stone, P. Zhou, E. Yen, B. Tow, K. D. Lardizabal, T. ester and triacylglycerol biosynthesis in Acinetobacter calcoaceticus Voelker, and R. V. Farese, Jr. 2001. Cloning of DGAT2, a second ADP1. J. Biol. Chem. 278: 8075–8082. mammalian diacylglycerol acyltransferase, and related family mem- 31. Ross, A. C. 1982. Retinol esterification by rat liver microsomes. Ev- 276: 38870–38876. bers. J. Biol. Chem. idence for a fatty acyl coenzyme A:retinol acyltransferase. J. Biol. 11. Yen, C-L. E., S. J. Stone, S. Cases, P. Zhou, and R. V. Farese, Jr. 257: Chem. 2453–2459. 2002. Identification of a gene encoding MGAT1, a monoacylglyl- 32. Saari, J. C., and D. L. Bredberg. 1988. CoA- and non-CoA-depen- cerol acyltransferase. 8512–8517. 99: Proc. Natl. Acad. Sci. USA. dent retinol esterification in retinal pigment epithelium. J. Biol. 12. Yen, C-L. E., and R. V. Farese, Jr. 2003. MGAT2, a monoacylglyc- Chem. 263: 8084–8090. Volume 46, 2005 Journal of Lipid Research 1510
10 33. Helgerud, P., L. B. Petersen, and K. R. Norum. 1982. Acyl CoA:ret- and lactating mammary gland microsomes. Arch. Biochem. Biophys. inol acyltransferase in rat small intestine: its activity and some 500–508. 288: 23: 609–618. J. Lipid Res. properties of the enzymic reaction. 41. Petkovich, M., N. J. Brand, A. Krust, and P. Chambon. 1987. A hu- 34. Blomhoff, R., M. Rasmussen, A. Nilsson, K. R. Norum, T. Berg, man retinoic acid receptor which belongs to the family of nuclear W. S. Blaner, M. Kato, J. R. Mertz, D. S. Goodman, U. Eriksson, et 444–450. 330: Nature. receptors. al. 1985. Hepatic retinol metabolism. Distribution of retinoids, en- 42. Giguere, V., E. S. Ong, P. Segui, and R. M. Evans. 1987. Identifica- zymes, and binding proteins in isolated rat liver cells. J. Biol. Chem. Nature. tion of a receptor for the morphogen retinoic acid. 330: 260: 13560–13565. 624–629. 35. MacDonald, P. N., and D. E. Ong. 1988. A lecithin:retinol acyl- 43. Heyman, R. A., D. J. Mangelsdorf, J. A. Dyck, R. B. Stein, G. transferase activity in human and rat liver. Biochem. Biophys. Res. Eichele, R. M. Evans, and C. Thaller. 1992. 9-Cis retinoic acid is a 157–163. Commun. 156: high affinity ligand for the retinoid X receptor. 397–406. Cell. 68: 36. Yost, R. W., E. H. Harrison, and A. C. Ross. 1988. Esterification by 44. Zouboulis, C. C., B. Korge, H. Akamatsu, L. Q. Xia, S. Schiller, H. rat liver microsomes of retinol bound to cellular retinol-binding Gollnick, and C. E. Orfanos. 1991. Effects of 13-cis-retinoic acid, J. Biol. Chem. 263: 18693–18701. protein. all-trans-retinoic acid, and acitretin on the proliferation, lipid syn- 37. Schmitt, M. C., and D. E. Ong. 1993. Expression of cellular retinol- thesis and keratin expression of cultured human sebocytes in binding protein and lecithin-retinol acyltransferase in developing 792–797. 96: vitro. J. Invest. Dermatol. Biol. Reprod. rat testis. 972–979. 49: 45. Strauss, J. S., M. E. Stewart, and D. T. Downing. 1987. The effect of 38. Zolfaghari, R., and A. C. Ross. 2000. Lecithin:retinol acyltrans- Arch. Dermatol. 1538– 123: 13-cis-retinoic acid on sebaceous glands. ferase from mouse and rat liver. CDNA cloning and liver-specific 1541. 41: regulation by dietary vitamin A and retinoic acid. J. Lipid Res. 46. Bazzano, G., N. Terezakis, H. Attia, A. Bazzano, R. Dover, D. Fenton, 2024–2034. N. Mandir, L. Celleno, M. Tamburro, and S. Jaconi. 1993. Effect of 39. Batten, M. L., Y. Imanishi, T. Maeda, D. C. Tu, A. R. Moise, D. J. Invest. Dermatol. retinoids on follicular cells. 101 (Suppl. 1): 138–142. Bronson, D. Possin, R. N. Van Gelder, W. Baehr, and K. Palczewski. 47. Mehta, R. G., W. L. Cerny, and R. C. Moon. 1983. Retinoids inhibit 2004. Lecithin-retinol acyltransferase is essential for accumulation prolactin-induced development of the mammary gland in vitro. J. Biol. Chem. of all-trans-retinyl esters in the eye and in the liver. 23–26. 4: Carcinogenesis. 279: 10422–10432. 48. Kistler, A. 1986. Structure-activity relationship of retinoids on lobu- 40. Randolph, R. K., K. E. Winkler, and A. C. Ross. 1991. Fatty acyl loalveolar differentiation of cultured mouse mammary glands. Car- CoA-dependent and -independent retinol esterification by rat liver 7: 1175–1182. cinogenesis. Downloaded from www.jlr.org by guest, on May 8, 2019 DGAT1 is a multifunctional enzyme 1511 Yen et al.
The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Year 2021–2026 Passenger Cars and Light Trucks Draft Environmental Impact Statement July 2018 Docket No. NHTSA-2017-0069More info »
H. R. 5515 One Hundred Fifteenth Congress of the United States of America AT THE SECOND SESSION Begun and held at the City of Washington on Wednesday, the third day of January, two thousand and eighte...More info »
Programming Guide ZPL II ZBI 2 Set-Get-Do Mirror WMLMore info »