HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 36, 28083-28092, September 8, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/36/28083    most recent M003927200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chang, C. C. Y. Articles by Chang, T. Y. Search for Related Content PubMed PubMed Citation Articles by Chang, C. C. Y. Articles by Chang, T. Y. Social Bookmarking                      What's this?

Immunological Quantitation and Localization of ACAT-1 and ACAT-2 in Human Liver and Small Intestine^*

From the ^a Department of Biochemistry, Dartmouth Medical School, Hanover, New Hampshire 03755, the ^b Second Department of Pathology, Kumamoto University School of Medicine, Kumamoto 860-0811, Japan, the ^c Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire 03766, the ^d Andrology Laboratory, University of British Columbia, Vancouver V6T 2B5, Canada, the ^e Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, the ^f University of Pittsburgh Medical Center, T. E. Starzl Transplantation Institute, Pittsburgh, Pennsylvania 15261, the ^g Gladstone Institute of Cardiovascular Disease and the University of California, San Francisco, California 94141 , and the ^h Center de Recherche Pierre Fabre-17, Avenue Jean Moulin, 81106 Castres Cedex, France

Received for publication, May 9, 2000, and in revised form, June 7, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

By using specific anti-ACAT-1 antibodies in immunodepletion studies, we previously found that ACAT-1, a 50-kDa protein, plays a major catalytic role in the adult human liver, adrenal glands, macrophages, and kidneys but not in the intestine. Acyl-coenzyme A:cholesterol acyltransferase (ACAT) activity in the intestine may be largely derived from a different ACAT protein. To test this hypothesis, we produced specific polyclonal anti-ACAT-2 antibodies that quantitatively immunodepleted human ACAT-2, a 46-kDa protein expressed in Chinese hamster ovary cells. In hepatocyte-like HepG2 cells, ACAT-1 comprises 85-90% of the total ACAT activity, with the remainder attributed to ACAT-2. In adult intestines, most of the ACAT activity can be immunodepleted by anti-ACAT-2. ACAT-1 and ACAT-2 do not form hetero-oligomeric complexes. In differentiating intestinal enterocyte-like Caco-2 cells, ACAT-2 protein content increases by 5-10-fold in 6 days, whereas ACAT-1 protein content remains relatively constant. In the small intestine, ACAT-2 is concentrated at the apices of the villi, whereas ACAT-1 is uniformly distributed along the villus-crypt axis. In the human liver, ACAT-1 is present in both fetal and adult hepatocytes. In contrast, ACAT-2 is evident in fetal but not adult hepatocytes. Our results collectively suggest that in humans, ACAT-2 performs significant catalytic roles in the fetal liver and in intestinal enterocytes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Acyl-coenzyme A:cholesterol acyltransferase (ACAT)¹ is an integral membrane protein located in the endoplasmic reticulum. It catalyzes the formation of cholesteryl esters from long chain fatty acyl coenzyme A and cholesterol (1). ACAT is present in a variety of cells and tissues. In macrophages, ACAT participates in the cholesterol-cholesteryl ester cycle (2); chronic accumulation of cholesteryl esters in macrophages leads to foam cell formation characteristic of atherogenesis (review in ref. 3). In the liver, ACAT utilizes cholesterol as its substrate, which is either synthesized endogenously or acquired via the LDL receptor and chylomicron remnant receptor. Hepatic cholesteryl ester and triacylglycerol are important lipid ingredients of very low density lipoprotein (VLDL). In monkeys, hepatic ACAT may increase the atherogenicity of LDL by augmenting both its secretion from the liver and accumulation of cholesteryl oleate in plasma LDL (4). Depletion of hepatic cholesteryl ester with ACAT inhibitors decreases VLDL secretion in several model systems (5-8). In intestinal enterocytes, substrates of ACAT include dietary cholesterol, biliary cholesterol, endogenous cholesterol, or cholesterol acquired via the LDL receptor. In animal studies, ACAT inhibitors reduce dietary cholesterol absorption (9, 10). For these reasons, ACAT has been a pharmaceutical target for prevention of hypercholesterolemia and atherosclerosis.

In mammals, two ACAT isoenzymes have been identified, ACAT-1 and ACAT-2. The ACAT-1 gene was identified (11) by its ability to complement the ACAT deficiency of a CHO cell mutant AC29 (12). The human ACAT-1 gene originated from two different chromosomes (13). ACAT-1 is a homotetrameric protein (14), which spans the ER membrane seven times (15). It displays sigmoidal kinetics with cholesterol as its substrate, implying that ACAT-1 is an allosteric enzyme regulated by cholesterol (16). Homozygous ACAT-1 knockout mice exhibit markedly reduced amounts of cholesteryl esters in adrenal glands and peritoneal macrophages; however, liver ACAT activity is not reduced, suggesting that the structure of ACAT in liver may be different from ACAT-1 (17). Based on the ACAT-1 sequence, the ACAT-2 gene in monkeys, mice, and humans has been identified (18-20). ACAT-2 shares high homology with ACAT-1 near the C terminus but not near the N terminus. Tissue distribution studies show that the expression of ACAT-1 message and protein is ubiquitous (21-23), whereas the expression of the ACAT-2 message is restricted to the liver and small intestine (18-20). The studies in mice and monkeys have led experts to believe that in mammals ACAT-2 comprises the majority of ACAT activity in the liver and intestines (24-26). These results provide the rationale for selective inhibition of a given ACAT isoenzyme in a tissue-specific manner. Recently, the ACAT-1 knockout mouse has been used as a model in predicting the consequences of selective inhibition of ACAT-1 in humans (26, 27).

Results from studies in mice or monkeys may not directly apply to humans. To assess the functional significance of ACAT-1 in various human tissue homogenates, our laboratory developed a protein knockout approach (28). We used the detergent CHAPS to solubilize ACAT-1, followed by specific anti-ACAT-1 antibodies to remove ACAT-1. We then measured the residual ACAT activity in the immunodepleted supernatants. The results showed that in liver, adrenal glands, macrophages, and kidneys, most of the ACAT activity (80% or more) had been immunodepleted. In contrast, in the intestine most of the ACAT activity was largely resistant to immunodepletion (29). This finding suggests that in adult humans, a different ACAT may exist in the intestine. To address this issue further, as detailed in this paper, we prepared specific anti-ACAT-2 antibodies against the N-terminal region of ACAT-2. We used anti-ACAT-1 and anti-ACAT-2 antibodies to perform immunodepletion experiments in HepG2 cells, Caco-2 cells, human livers, and human intestines. We also conducted immunoblotting and immunohistochemical staining analyses in livers and intestines. Our studies show that in the adult human, ACAT-2 is the major ACAT isoform in the intestine, whereas ACAT-1 is the major ACAT isoform in the liver. These results provide the foundation for designing ACAT inhibitors targeted at specific human tissues.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials

Antipain, CHAPS, chymostatin, elastantinal, fetal bovine serum, leupeptin, oleoyl-CoA, poly-L-lysine, and taurocholate were purchased from Sigma. Pefabloc was from Roche Molecular Biochemicals. Reagent grade solvents were obtained from Fisher. Transwell inserts were from Costar (Cambridge, MA). Human intestinal cDNA library was purchased from CLONTECH. Multilink kit was from BioGenex Laboratories (San Ramon, CA). Microtome and OCT compound were obtained from Sakura Finetechnical Co. (Tokyo, Japan). The monoclonal antibodies against hACAT-1 (ACAT-1a) and hACAT-2 (ACAT204) were from Vancouver Biotech Ltd.

Cloning of Human ACAT-2 cDNA-- A 5'-rapid amplification of cDNA ends was performed using RNA from HepG2 cells as the templates and primer sequences that were derived from an EST clone (R10292) containing high regional homology to ACAT-1. A partial cDNA was obtained, which was designated as a candidate of human ACAT-2 cDNA. This retained the N-terminal and C-terminal regions but lacked three different internal coding regions. By using a human intestinal cDNA library as the template, appropriate primers (ACAT-2-N', 5'-ATG-GAG-CCA-GGC-GGG-GCC-CGT-CTG-CGT-CTG-3'; and ACAT-2-C', 5'-CTA-GGT-ATG-GCA-GGA-CCA-AGA-TCG-AGG-TGT-C-3') were used for polymerase chain reaction. Several positive clones were isolated, the largest clone of which contained 2,100 base pairs (sequence deposited in the GenBank data base; accession number AF099031). The nucleotide sequence was determined from both strands by the automated DNA primer cycle sequencing reaction. The sequencing reaction was repeated once for confirmation. One difference exists between our clone and the clone (accession number AF059203) reported by Oelkers et al. (20). Our clone encodes isoleucine (atc), instead of threonine (acc) at amino acid 254. For expression studies, the ACAT-2 coding region was subcloned into the EcoRV site of the mammalian expression vector pcDNA3-neo (Invitrogen). In addition, a DNA fragment encoding the 6-histidine tag followed by a T7 tag (16) was inserted at the N terminus of hACAT-2 cDNA by polymerase chain reaction. The constructs were verified by sequencing.

Cell Culture-- CHO cells were grown as monolayers in medium A (Ham's F-12 supplemented with 10% fetal bovine serum and 10 µg/ml gentamicin). The ACAT-deficient mutant AC29 was described previously (12). By using appropriate constructs for transfection into AC29 cells, we isolated stable clones that either expressed the untagged ACAT-1, the His-T7-tagged HisACAT-1 (16), the untagged ACAT-2, or the His-T7-tagged HisACAT-2. Stable transfectant clones were isolated using a procedure previously described (16). We used both untagged ACAT-2 and HisACAT-2 cells as the ACAT-2 enzyme source to perform various biochemical experiments (described in Fig. 2, see "Results"). The results showed that the addition of the His tag did not alter the enzymatic properties of ACAT-2. HepG2 cells were obtained from the ATCC. Two clones of Caco-2 cells, yielding similar results reported here, were obtained from ATCC or from Dr. Jeffrey Field at the University of Iowa. Both cell types were cultured at 37 °C in a 10% CO₂ incubator in Dulbecco's minimal essential medium with 20% fetal bovine serum, 4.5 g/liter glucose, and 2 mM L-glutamine, supplemented with penicillin (100 units/ml) and streptomycin (100 µg/ml).

Human Hepatocytes-- Eleven human hepatocytes (1 fetal, 4 child, and 6 adult) were collected at the Department of Pathology, University of Pittsburgh, and stored in liquid nitrogen in 12% Me₂SO. Two additional fetal hepatocytes (22 and 35 weeks) were from Anatomic Gift Foundation, Laurel, MD.

Human Tissues-- The liver and intestinal samples were obtained from organs that were removed from bodies 2-21 h postmortem and frozen in liquid nitrogen until use. Causes of deaths varied among donors. A total of 20 adult human livers and 21 adult intestines were collected from the Anatomic Gift Foundation, Dartmouth-Hitchcock Medical Center, Starzl Transplantation Institute of University of Pittsburgh Medical Center, and at Kumamoto University Medical School, Japan. Immunoblot analysis indicated that the ACAT-1 and ACAT-2 proteins were not degraded significantly among various liver samples. In contrast, significant degradations of ACAT-2 protein occurred in two-thirds of the intestinal samples. Therefore, only intestinal samples that did not exhibit degradation of ACAT-2 were used for immunodepletion and immunohistochemical staining studies. Nine fetal tissues (livers and intestine) were collected. Immunoblot analysis indicated that the ACAT-1 and ACAT-2 proteins were not degraded significantly among various fetal samples.

Antibodies

Targeted against hACAT-1-- The specific polyclonal antibodies DM10 and monoclonal antibody ACAT-1a were described previously (16, 28).

Targeted against hACAT-2-- The specific polyclonal antibodies from two rabbits (DM54 and DM56) were generated according to the procedure as described (28), except that the N-terminal region of hACAT-2 (amino acid residues 1-121) was expressed as a GST-ACAT2 fusion protein adduct. The antisera were subsequently purified by a protein A-Sepharose column, followed by a GST-ACAT2 fusion protein affinity column. The monoclonal antibody against hACAT-2 was generated according to the procedure as described (16), using the GST-ACAT2 fusion protein described above as the antigen. Media of one of the hybrid cell lines ACAT204 was collected and employed in the present study. Mouse antiserum against sucrase, HSI14 (32), was a gift from Dr. Andrea Quaroni at Cornell University.

Protease Inhibitor Mixture-- Various protease inhibitors were prepared as individual concentrated stocks and stored at 20 °C. The mixture was prepared freshly in appropriate buffer to reach final concentration of 1 mM EDTA, 1 µg/ml leupeptin, 1 µg/ml antipain, 40 µg/ml chymostatin, 1 mg/ml elstantinal, and 40 µg/ml Pefabloc.

ACAT Inhibitors-- F12511, ±CP-113818, DuP-128, and CI-976 were synthesized and purified to 99% purity by the Medicinal Chemistry Division of Pierre Fabre Research. CI-976, 2,2-dimethyl-N-(2,4,6-trimethoxyphenyl) dodecanamide; DuP-128, N'-(2,4-difluorophenyl)- N-[5-(4,5-diphenyl-1H-imidazol-2-ylthio)pentyl]-N-heptylurea; ±CP- 113818,N-[2,4-bis(methylthio)-6-methylpyridin-3-yl]-2-(hexylthio) decanamide; F 12511, (S)-2',3',5'-trimethyl-4'-hydroxy-a-dodecylthio-phenylacetanilide.

Methods

ACAT Enzyme Assay-- The assay was performed essentially as described previously (16). Unless indicated otherwise, cell extracts were solubilized with 2.5% CHAPS in the presence of 1 M KCl. The mixture, designated as solubilized cell extract, was mixed on ice with preformed taurocholate/cholesterol/PC-mixed micelles (with final concentration of taurocholate at 9.3 mM, PC at 11.2 mM, and cholesterol at 1.6 mM). The amount of mixed micelles solution was used in excess to dilute the detergent CHAPS presented in the solubilized enzyme, such that a final CHAPS/PC molar ratio of less than 0.4 was maintained. The ACAT assay was initiated by adding 20 µl of a solution containing 10 nmol of [³H]oleoyl-CoA (in 0.25 M Tris at pH 7.8) and 10 nmol of fatty acid-free bovine serum albumin to the enzyme-mixed micelles mixture. The reaction was carried out at 37 °C for up to 20 min.

Human Tissue Preparation for Immunoblotting-- About 0.4 g of each frozen tissue sample was homogenized at room temperature with 1 ml of lysis solution (10% SDS, 100 mM DTT in the presence of protease inhibitor mixtures). The homogenization was carried out using a Potter-Elvenhjem grinder at medium speed for 2 min, followed by passing the homogenates through a 3-ml syringe fitted with a 21-gauge needle 20 times. The homogenates were incubated at 37 °C for 20 min. The total protein concentration was determined, and tissue homogenates were analyzed with 10% SDS-PAGE. Proteins were electro-transferred from gels onto polyvinylidene difluoride membranes and immunoblotted with 0.8 µg/ml of various affinity-purified IgG as indicated.

Immunodepletion Analysis

Tissue Culture Cells-- Monolayers of cells were harvested by hypotonic shock at room temperature for 3 min followed by scraping (33). The scraped cells were thoroughly homogenized on ice with a stainless steel tissue grinder (Wheaton; Dura-Grind) in Buffer A (50 mM Tris and 1 mM EDTA at pH 7.8) at 2-5 mg of protein/ml and then solubilized by 2.5% CHAPS and 1 M KCl. The extract was incubated at 4 °C for 5 min and then centrifuged at 100,000 × g for 45 min at 4 °C to isolate the solubilized cell lysate from the nuclei and other particulate materials. The cell lysate (450 µl, with 2 mg of protein/ml) was incubated with 50 µl of Buffer A containing 12 µg of IgG in a microcentrifuge tube with constant shaking for 30 min at 4 °C. The immune complex was mixed further for 1 h after adding 100 µl of protein A-Sepharose (diluted 1:1 with Buffer A). The immunodepleted supernatants and the immunoprecipitates were separated by centrifugation at 14,000 × g for 1 min. To measure ACAT activity remaining in the immunodepleted supernatant, 24 µl of the supernatant was mixed with 220 µl of taurocholate/cholesterol/PC-mixed micelles at 4 °C. The ACAT assay was carried out at 37 °C for 10 min. To detect ACAT protein in the immunoprecipitates, the immunoprecipitates were washed with 3× 0.5 ml of PBS and then dissolved by adding 120 µl of 10% SDS at 37 °C for 10 min without thiol-reducing reagent. The lysate was loaded onto four identical 10% SDS-PAGE and then processed for immunoblots to detect the ACAT-1 and ACAT-2 proteins.

Human Tissues-- Frozen tissues were thawed on ice in Buffer A with protease inhibitor mixtures. Tissues were homogenized using a Potter-Elvenhjem grinder at medium speed for 5 min on ice and solubilized with 2.5% CHAPS and 1 M KCl at protein concentration of 10 mg/ml. After centrifugation of 100,000 × g for 45 min at 4 °C to isolate the solubilized homogenate, its protein concentration was determined; and the final concentration at 5-8 mg/ml was adjusted for immunodepletion analysis according to the procedure described under "Tissue Culture Cells" section.

Immunohistochemical Stainings of Human Tissues

Method A (Performed in Kumamoto, Japan)-- Liver and small intestinal samples were obtained from autopsy cases within 2 h postmortem. Tissues with visible pathological changes were excluded from further study. Seven samples were from adults (ages between 34 and 74 years old) and four samples were from fetuses (9-28 gestational weeks). Samples were fixed by periodate/lysine/paraformaldehyde for 4 h at 4 °C, rinsed with 0.05 mol/liter PBS containing graded series of sucrose (10, 15, and 20%), embedded in OCT compound, rapidly frozen using liquid nitrogen, and stored at 80° C. The frozen materials were cut into 5-µm-thick sections using cryostat (HM 500-M, MICROM, Waldorf, Germany) and mounted on poly-L-lysine-coated slides. The indirect immunoperoxidase method described previously (23) was used, using primary antibodies at 2 µg/ml.

Method B (Performed at Dartmouth Medical School)-- Wedge biopsy samples of 6 adult liver samples (ages between 12.7 and 59 years old) were quickly frozen and stored in liquid nitrogen at T. E. Starzl Transplantation Institute and then shipped to Dartmouth on dry ice. Four fetal livers and intestines (22-40 gestational weeks) were obtained from the Dartmouth-Hitchchock Center, quickly frozen at 80° C until use. Sections 4 µm thick were cut from neutral-buffered, formaldehyde-fixed, and paraffin-embedded tissue blocks, mounted on barrier slides, dried at 60° C, deparaffinized, and hydrated. A modification of the avidin-biotin-peroxidase complex technique (34) was used for histochemical staining, with anti-ACAT-1 or anti-ACAT-2 as primary antibodies at 2 µg/ml. After immunostaining, the slides were briefly counterstained with Mayer's hematoxylin (BioGenex Laboratories, CA).

Protein determination, SDS-PAGE, and immunoblotting analyses were performed as described previously (16). We found that similar to what as been reported for ACAT-1 (14, 28, 35), ACAT-2 solubilized in either CHAPS or in SDS tends to form higher molecular weight materials that are two or four times the apparent size of the ACAT-2 monomer within 24 h. The use of freshly prepared extract in the presence of 25-100 mM DTT effectively minimizes the formation of the aggregated material. Therefore, for immunoblot analysis, we routinely used freshly prepared cell extract in SDS and DTT, conduct SDS-PAGE, and evaluate ACAT-1 and ACAT-2 in their monomeric forms (50 and 46 kDa). For immunoprecipitation analysis, we avoid the use of DTT (which will cause dissociation of IgG into the heavy and light chains), and we analyze the immunoprecipitates by SDS-PAGE without thiol-reducing agent. Under this condition, more than 80% of the ACAT-1 and ACAT-2 are found in dimeric form (100 and 92 kDa, respectively).

Quantitation of immunoblots was performed using the Molecular Dynamics Computer Densitometer and ImageQuant software.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The Enzymatic Characteristics of ACAT-2-- In transient transfection studies, the ACAT-2 cDNA complemented the ACAT deficiency of CHO cell mutant AC29 (result not shown). A stable transfectant clone, designated as the ACAT-2 cell, has been isolated. Unlike AC29 cells but similar to the cells expressing HisACAT-1 (16), ACAT-2 cells express significant ACAT activity in vitro and contain abundant cholesteryl esters as cytoplasmic lipid droplets (Fig. 1). The specific activity of ACAT-2 expressed in this cell line ranged between 200 and 300 pmol/min/mg, which is similar to that of HisACAT-1 cells.

View larger version (74K): [in this window] [in a new window]   Fig. 1.   Nile Red-stained cells viewed with phase contrast (I) or fluorescence microscopy (II). AC29 (A), ACAT-1 (B), and ACAT-2 (C) cells were seeded in 25-cm2 tissue culture flasks at sparse density in medium A for 48 h, rinsed 4 times with Hanks' buffer, and stained for 6 min at room temperature with 100 ng/ml Nile Red. The flasks were briefly rinsed with Hanks' buffer and viewed with an inverted Nikon Diaphot microscope with epifluorescence (excitation 450-490 nm, emission 520 nm).

To study ACAT-2 in vitro, we used cell homogenates from ACAT-2 cells as the enzyme source. ACAT-2 can be optimally solubilized with the zwitterionic detergent CHAPS at 2-2.5% and high salt (1 M KCl) (Fig. 2A); the enzyme is more active when assayed in bile salt-based mixed micelles than in liposomes, with taurocholate being the best bile salt tested (Fig. 2B). Its cholesterol saturation curve is sigmoidal, although less so than the curve for ACAT-1 (Fig. 2C). The oleoyl-coenzyme A saturation curve is hyperbolic (Fig. 2D), with a K_m value of 14.8 µM, which is twice as high as the value for ACAT-1 (7.4 µM). Thus, the characteristics of ACAT-2 are similar to those of ACAT-1 described earlier (16). These results justify the use of the immunodepletion approach to assess the ACAT-1 and ACAT-2 activity present in the same cell extracts. We tested the sensitivities of ACAT-1 and ACAT-2 against four different ACAT inhibitors, each possessing a distinct structural characteristic (F 12511, ±CP 113818, Dup 128, and CI 976). The results showed that Dup 128 preferentially inhibited ACAT-2 than ACAT-1; the IC₅₀ for ACAT-2 is 0.022 µM and 0.11 µM for ACAT-1 (Fig. 3). This result is in agreement with our earlier finding that demonstrated that Dup 128 preferentially inhibited intestinal ACAT activity than liver ACAT activity (29). The other three ACAT inhibitors did not exhibit a large preference between the two isoenzymes (Fig. 3).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Biochemical characteristics of ACAT-2. A, solubilization of ACAT-2 by CHAPS in the presence of 1 M KCl. Cell extracts at 4 mg/ml were treated with CHAPS at the indicated concentrations with 1 M KCl. To 15 µl of detergent-treated cell extract, 136 µl of taurocholate/cholesterol/PC-mixed micelles were added. ACAT assay was performed at 37 °C for 5 min. B, ACAT-2 activity in cholesterol/PC-mixed micelles prepared with various bile salts at concentrations as indicated. Cholesterol and PC were at constant concentration of 1.6 and 11.2 mM, respectively. The symbol  indicates the ACAT activity of the enzyme assayed in cholesterol/PC vesicle (16). C, cholesterol substrate saturation curves of ACAT-1 and ACAT-2 in taurocholate/cholesterol/PC-mixed micelles. 17 µl of solubilized cell extract (see "ACAT Enzyme Assay") was added to 155 µl of mixed micelles containing 9.3 mM taurocholate, 11.2 mM PC and cholesterol at increasing concentration. ACAT assay was conducted as described in A. D, oleoyl-CoA substrate saturation curves of ACAT-1 and ACAT-2 in taurocholate/cholesterol/PC-mixed micelles. 17 µl of solubilized cell extract was added to 155 µl of taurocholate-PC mixed micelles containing 1.6 mM cholesterol (with 3H labeling at 0.2 × 106 cpm/reaction). The reaction was initiated by adding 20 µl of assay mixture containing increasing amounts of oleoyl-CoA pre-mixed with fatty acid-free bovine serum albumin at an equal molar ratio as described previously (16). The data shown in A-D are from one of three separate experiments.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Inhibition of ACAT-1 or ACAT-2 by various ACAT inhibitors. 14.6 ml of preformed taurocholate/cholesterol/PC-mixed micelles were incubated with 1.6 ml of solubilized cell extract that were prepared from either HisACAT-1 or HisACAT-2 cells, for 10 min at 4 °C. 165 µl of the above mixture was added into each tube that contained various ACAT inhibitors with final concentrations as indicated. The ACAT inhibitors were added from a stock solution in Me2SO. The final concentration of Me2SO was 0.24%. The mixture was incubated on ice for 30 min. The ACAT reaction was carried out at 37 °C for 20 min with duplicate assay tubes. The values presented here are the average of five experiments.

The Specificity of the Polyclonal Antibodies against Human ACAT-2 (DM54)-- We produced high titer, affinity-purified polyclonal anti-ACAT-2 antibodies (DM54 and DM56) against the N-terminal region of human ACAT-2 (see "Experimental Procedures"). Western blot analysis revealed that these antibodies specifically recognized a single 46-kDa protein band in AC29 cells expressing the untagged ACAT-2 (results not shown). We next prepared CHAPS-solubilized cell extracts from ACAT-1 or ACAT-2 cells, used anti-ACAT-1 (DM10 (28)) and/or an anti-ACAT-2 antibody (DM56) to perform the immunoprecipitation experiments, and then measured residual ACAT activities in the immunodepleted supernatants. Anti-ACAT-2 antibodies quantitatively immunodepleted the ACAT activity from ACAT-2 cells (Fig. 4A, lane 7) but not from ACAT-1 cells (Fig. 4A, lane 3). Control experiments showed that anti-ACAT-1 antibody quantitatively immunodepleted ACAT activity from ACAT-1 cells (Fig. 4A, lane 2) but not from ACAT-2 cells (Fig. 4, lane 6). Additional control experiments showed that the nonspecific anti-rabbit IgGs failed to deplete either ACAT-1 or ACAT-2 activity (Fig. 4A, lanes 1 and 5). In a parallel experiment, we identified the ACAT protein bands in the immunoprecipitates by Western blots, using monoclonal antibody against either ACAT-1 or ACAT-2. The results showed that anti-ACAT-2 immunoprecipitated the ACAT-2 (Fig. 4B-II, lanes 7 and 8) but not the ACAT-1 (Fig. 4B-II, lanes 3 and 4), whereas the anti-ACAT-1 immunoprecipitated ACAT-1 (Fig. 4B-I, lanes 2 and 4) but not ACAT-2 (Fig. 4B-I, lanes 6 and 8). The nonspecific IgGs failed to immunoprecipitate either ACAT-1 or ACAT-2 proteins (Fig. 4B-I and II, lanes 1 and 5).

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Immunodepletion of ACAT-1 or ACAT-2 in various cell types as indicated. A, ACAT activity remained in the supernatant after the immunodepletion. 12 × 106 cells were seeded in 150-cm2 dishes as described under "Experimental Procedures." Immunodepletion was carried out as described under "Immunodepletion Analysis." For ACAT-1 and ACAT-2 cells, the results shown are the averages of 5 experiments; for HepG2 cells, the results shown are the average of 10 experiments. The ACAT-specific activities (in pmol/min/mg) in various solubilized cell extracts were 372 ± 50 for ACAT-1 cells, 313 ± 60 for ACAT-2 cells, and 50 ± 10 for HepG2 cells. B, detection of ACAT-1 (B-I) or ACAT-2 (B-II) protein in the immunoprecipitates. One-fourth of the precipitates produced in the immunodepletion experiment described above were analyzed by immunoblotting.

ACAT-1 and ACAT-2 Contents in HepG2 Cells, in Human Hepatocytes, and in Human Livers-- We used extracts from HepG2 cells to perform immunodepletion experiments. Anti-ACAT-1 or anti-ACAT-2 immunodepleted 85% (Fig. 4A, lane 10) or 15% (Fig. 4A, lane 11) of total ACAT activity in HepG2 cells (n = 10). The use of both anti-ACAT-1 and anti-ACAT-2 IgGs essentially depleted all of the ACAT activity (Fig. 4A, lane 12). We analyzed the ACAT protein bands that were present in the immunoprecipitates, and we found that anti-ACAT-1 only immunoprecipitated ACAT-1 (Fig. 4B-I, lanes 10 and 12), whereas anti-ACAT-2 only immunoprecipitated ACAT-2 protein (Fig. 4B-II, lanes 11 and 12). These results show that in HepG2 cells, ACAT-1 and ACAT-2 do not form tightly bound hetero-oligomeric complex.

We next estimated the relative contents of ACAT-1 and ACAT-2 in human hepatocytes based on Western blotting; for quantitation purposes, we used the ACAT-1 and ACAT-2 in HepG2 cells as standards (Fig. 5). The results showed that the ACAT-1 signals were present at relatively constant levels in fetal (n = 3), child (n = 4), and adult hepatocytes (n = 6, ages 21-59) (Fig. 5A, lanes 1-11). Densitometric analysis showed that the intensities of ACAT-1 signals among individual hepatocytes did not vary by more than 3-fold. In contrast, the ACAT-2 signals were very strong in fetal hepatocytes (n = 3) but were diminished in child hepatocytes (age 2-15 year old, n = 4). Densitometric analysis showed that the signals in child hepatocytes were less than 10% of the signals found in fetal hepatocytes. A typical result is seen in Fig. 5B, lanes 1-5. In hepatocytes from adults, the ACAT-2 signals were further diminished to virtually undetectable levels (Fig. 5B, lanes 6-11).

View larger version (41K): [in this window] [in a new window]   Fig. 5.   The presence of ACAT-1 and ACAT-2 in human hepatocytes as analyzed by immunoblotting. The frozen hepatocytes were thawed on ice. The preserving medium was removed by centrifugation for 5 min at 1,000 rpm at room temperature. After rinsing with PBS twice, the pellets were dissolved in 10% SDS and 100 mM DTT with various protease inhibitors. 150 µg of cell extract from each hepatocyte was loaded into lanes 1-11. Ages of donors are indicated at bottom of each lane. 50 µg per lane of HepG2 cell extracts freshly solubilized in 10% SDS and 100 mM DTT was loaded onto the last two lanes on far right, to serve as standards for quantitation. Results in lane 1 (22-week-old fetal) and lane 2 (2-year-old newborn) are representative from three different donors.

ACAT-1 and ACAT-2 Contents in Human Enterocyte-like Caco-2 Cells, in Human Intestines, and in Human Livers-- Human Caco-2 cell is an established cell line that spontaneously differentiates and expresses various intestinal enterocyte-like properties (36, 37), upon reaching confluence in culture, and mimics immature human enterocytes (38). We monitored the relative contents of ACAT-1 and ACAT-2 in Caco-2 cells for up to 15 days after confluence. The results show that that in differentiating Caco-2 cells, ACAT-2 protein content increased 5-10-fold in 6 days (Fig. 6B), whereas ACAT-1 protein content remained relatively constant (Fig. 6A). Control experiments show that the expressions of the intestinal sucrase, an enzyme found only in intestinal enterocytes (39), increased dramatically in differentiating Caco-2 cells (Fig. 6C). Similar results were found when we seeded Caco-2 cells in tissue culture dishes or used two other polarized cell lines HT29 and T84 cells (data not shown). We next performed immunodepletion experiments, using extracts from Caco-2 cells grown after confluence at zero time or for 16 days. The results show that for zero time cells, anti-ACAT-1 or anti-ACAT-2 immunodepleted 85 or 15% total ACAT activity, respectively (n = 7, Fig. 7A, lanes 5-8). For the 16th day cells, anti-ACAT-1 or anti-ACAT-2 IgG immunodepleted 72 or 52% of total ACAT activity, respectively (Fig. 7A, lanes 1-4). We also found that after differentiation, the specific activity of total ACAT activity of Caco-2 cells increased by about 2-fold. The use of both anti-ACAT-1 and anti-ACAT-2 IgGs essentially depleted all of the ACAT activity (Fig. 7A, lanes 4 and 8). We analyzed the ACAT protein present in the immunoprecipitates, and we found that for both undifferentiated and differentiated Caco-2 cells, anti-ACAT-1 precipitated only ACAT-1 (Fig. 7B-I, top lanes 2, 4, 6, and 8), whereas anti-ACAT-2 IgG precipitated only ACAT-2 (Fig. 7B-II, bottom lanes 3, 4, 7, and 8).

View larger version (36K): [in this window] [in a new window]   Fig. 6.   The presence of ACAT-1 and ACAT-2 in differentiating Caco-2 cells as analyzed by immunoblotting. Caco-2 cells (18th to 22nd passage) were seeded in microporous membranes (Costar; 0.4-µm pore size and 24.5 mm diameter), at 5 × 106 density and grown as described under "Cell Culture." The time when cells grew to confluence was designated as the zero time. Medium was replaced every other day. At various times as indicated, cells were rinsed with 3× PBS and then quickly frozen at 80° C. HepG2 cells seeded in a similar manner were harvested at zero time and served as controls for the quantitation of ACAT-1 and ACAT-2. At the end of the time course, cells were harvested by adding 10% SDS containing 100 mM DTT and protease inhibitors to the cell monolayers; rubber scrapers were used to collect the cell homogenates. Cell homogenates were loaded at quantities as indicated onto three identical gels. Immunoblot analysis was performed using IgGs against ACAT-1, ACAT-2, and sucrase. Results are representative of two experiments.

View larger version (32K): [in this window] [in a new window]   Fig. 7.   Immunodepletion of ACAT-1 and/or ACAT-2 in Caco-2 cells grown at OT zero time or at 16th day post-confluence. 30 × 106 per dish of Caco-2 cells were seeded in 150-cm2 dishes. Media were replaced every other day. Cells at confluence (usually in 2 days) are designated as zero time (OT). Immunodepletion was carried out as described under "Immunodepletion Analysis." A, ACAT activity remained in the supernatant after the immunodepletion. The results shown are the average of two experiments. The ACAT-specific activities (in pmol/min/mg) in solubilized cell extracts were 18.9 ± 5.6 (n = 4) for zero time culture and 36.4 ± 10 (n = 4) for 16 days culture. B, detection of ACAT-1 (row B-I) or ACAT-2 (row B-II) protein band in the immunoprecipitates. One-fourth of the precipitates produced in the immunodepletion experiment described above were analyzed by immunoblotting as described under "Experimental Procedures."

We next performed immunodepletion experiments, using detergent-solubilized extracts from segments of small intestines and livers. In intestine, anti-ACAT-1 and anti-ACAT-2 IgGs depleted almost equal amounts of total measurable ACAT activity. In livers, anti-ACAT-1 or anti-ACAT-2 IgGs depleted 79 or 12.7% of total measurable ACAT activity (Fig. 8A, lanes 5-8). We analyzed the ACAT protein present in these immunoprecipitates, and we found that for both intestines and livers, anti-ACAT-1 IgG precipitated only ACAT-1 (Fig. 8B-I), whereas anti-ACAT-2 precipitated only ACAT-2 protein (Fig. 8B-II). These results, along with results shown earlier (Fig. 4B and Fig. 7B), clearly indicate that ACAT-1 and ACAT-2 do not form tightly bound hetero-oligomeric complexes in liver cells or in intestinal cells.

View larger version (49K): [in this window] [in a new window]   Fig. 8.   Immunodepletion of ACAT-1 and/or ACAT-2 in human intestines and livers. Immunodepletion was carried out as described under "Immunodepletion Analysis." A, ACAT activity remained in the supernatant after the immunodepletion. For intestinal tissue, the results shown are the average of 8 experiments using three different intestines. The ACAT-specific activities (in pmol/min/mg) in solubilized cell extracts were 20.5 ± 5.5 for sample 1, 7.2 ± 3.1 for sample 2, and 6.2 ± 2.7 for sample 3. For liver tissue, the results shown are the average of five experiments, using two different liver samples. The average total ACAT-specific activities (in pmol/min/mg) in the two solubilized liver cell extracts were 4. 0 ± 1.3. B, detection of ACAT-1 (row B-I) or ACAT-2 (row B-II) protein in the immunoprecipitates. One-fourth of the precipitates produced in the immunodepletion experiment described above were analyzed by immunoblotting.

Histochemical Stains of Livers and Intestines-- We used histochemical stainings to localize ACAT-1 and ACAT-2 in the adult liver samples; ACAT-1 signals were clearly present in hepatocytes (Fig. 9A, 1a-6a), with more intense signals in the periportal than in the pericentral zones. In addition, ACAT-1 signals were especially prominent in epithelial cells lining the hepatic ducts. A typical result is demonstrated in Fig. 9A, 1a. In contrast, the ACAT-2 signals were barely detectable in human livers (Fig. 9A, 1b-6b). To ensure that the histochemical signals in livers were derived from ACAT-1 and ACAT-2, we performed Western blot analyses, using HepG2 cell extracts to serve as controls (Fig. 9B, lanes on the far left). Results showed that ACAT-1 signals were easily demonstrated (Fig. 9B, top panel, lanes 1-6). In contrast, the ACAT-2 signals were barely detectable (Fig. 9B, bottom panel, lanes 1-6). These results also confirmed the results found in hepatocytes (Fig. 5). Additional histochemical stains showed that in fetal livers both ACAT-1 and ACAT-2 signals were clearly present in hepatocytes (Fig. 10A). In adult jejunum samples, both ACAT-1 and ACAT-2 were present, with ACAT-2 preferentially located at the apical half of the villi, whereas ACAT-1 was uniformly distributed along the vilus-crypt axis (n = 4). A typical result is shown in Fig. 10C. The same results were obtained using various duodenum (n = 3) or ileum (n = 3) samples (results not shown) or fetal intestinal samples (n = 4) (Fig. 10B). We conclude that both isoenzymes are present in the liver and the intestine; the major ACAT isoform in the adult hepatocytes is ACAT-1, and the major ACAT isoform at the apical region of the intestinal villi is ACAT-2.

View larger version (110K): [in this window] [in a new window]   Fig. 9.   Immunohistochemical staining (A) and Western blot (B) of human liver samples using anti-ACAT-1 or anti-ACAT-2. Livers used were of transplantation quality and came from Starzl Transplantation Institute, University of Pittsburgh Medical Center. The ages of donors ranged from 12 to 59 years old. Wedge biopsy samples from six livers were stored in liquid nitrogen. One-half of each sample was quickly fixed in 10% formaldehyde at room temperature and then processed for immunohistochemical staining (Method B described under "Immunohistochemical Stainings of Human Tissues"). The results are shown in A. The brown color denotes ACAT-1 or ACAT-2 signals; the dark blue color denotes the cell nuclei. Sections 1a-6a, results using anti-ACAT-1; sections 1b-6b, results using anti-ACAT-2. The other half of the liver samples was quickly dissolved by 10% SDS and 100 mM DTT in the presence of protease inhibitor mixture and used for Western blot analysis. The results are shown in B. For lanes 1-6, 150 µg of protein from each liver was loaded; for the lane on the far left, 50 µg of protein from HepG2 cell extract was loaded as standard for quantitation.

View larger version (118K): [in this window] [in a new window]   Fig. 10.   Immunolocalizations of ACAT-1 and ACAT-2 in the human fetal liver (A), fetal intestine (B), and adult intestine (C). Immunohistochemical stains were performed according to the procedures described in Method A under "Immunohistochemical Stainings of Human Tissues." The brown color denotes ACAT-1 or ACAT-2 signal; the green color denotes cell nuclei signal. A1, B1, and C1 are results using anti-ACAT-1 antibody (DM10). A2, B2, and C2 are results using anti-ACAT-2 antibody (DM54). A3 and C3 are results without using the primary antibodies and serve as negative controls. Results are representative of samples from four fetuses and seven adults. For adult intestines, the figures shown here are results using the autopsy samples; the same results have been confirmed by using paraffin-embedded biopsy specimen fixed in 10% formaldehyde within several minutes after removal from live patients. Arrowheads in C1 indicate the infiltrated macrophages.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

ACAT has been a visible target for pharmaceutical intervention of hypercholesterolemia and atherosclerosis. Two isoenzymes, ACAT-1 and ACAT-2, with similar but different enzymatic characteristics, exist in mammals, raising the hope for selectively targeting a given ACAT isoenzyme in a tissue-specific manner. If the drugs are to be used for treating human diseases, the rationale for selective targeting should be based on ACAT-1/ACAT-2 tissue distribution in humans. Our current and earlier studies (29) indicate that the quantitative distribution of ACAT-1 and ACAT-2 in humans is different from that in mice and monkeys. This finding is not surprising. Among various mammalian species, important differences in tissue expression levels also exist for other enzymes/proteins involved in cholesterol and lipoprotein metabolism, such as the cholesteryl ester transfer protein (40) and the catalytic subunit of the ApoB editing enzyme (41). Our findings imply that in adult humans, a selective ACAT-1 inhibitor should be more effective in the liver and in macrophages but less effective in the intestine, whereas a selective ACAT-2 inhibitor should be more effective in the intestines but less effective in the liver and macrophages.

In hepatocytes, cholesteryl esters along with triacylglycerols constitute the bulk of the neutral lipid core of VLDL. Multiple lipids are required for VLDL assembly. The transfer of triacylglycerol and cholesteryl ester to ApoB within the endoplasmic reticulum involves microsomal triglyceride transfer protein (for reviews, see Refs. 42-45). Both ACAT-1 and ACAT-2 are integral membrane proteins. Based on hydropathy plot analysis, it has been hypothesized that the putative active site of ACAT-1 and ACAT-2 may be located at opposite sides of the ER. This hypothesis predicts that ACAT-2 may directly participate in the lipoprotein synthesis and assembly, whereas ACAT-1 may play important roles in cellular cholesterol homeostasis and in lipid droplet formation (18). Experimentally, ACAT-1 has been shown to contain seven transmembrane domains, with its putative active site located at the cytoplasmic side of the ER membrane (15). The membrane topography of ACAT-2 is currently unknown. The results demonstrated in Fig. 1 in this report show that each isoenzyme is capable of producing large amount of cytoplasmic cholesteryl ester when expressed in CHO cells. Mechanistically, how lipids, including cholesteryl esters produced by ACAT, contribute to the lipoprotein synthesis/assembly in the adult liver is currently unclear. Whether one of these two isoenzymes plays a more direct role in the lipoprotein synthesis assembly process requires further investigation. Results using hamster hepatocytes suggest that the total cholesteryl ester mass, not the cholesteryl esters newly synthesized by ACAT, is a major factor in affecting the VLDL secretion rate (see Ref. 8 and a review in Ref. 45). We therefore favor the notion that it may be the total pool of cholesteryl esters produced by both ACAT-1 and ACAT-2 that serves as an important determinant in the VLDL secretion process.

The physiological significance of having two ACAT isoforms in the human liver and in the intestine is presently unclear. Our results show that hepatic ACAT-1 seems to be constitutively expressed before and after birth, whereas hepatic ACAT-2 is significantly expressed in fetal liver but is much diminished in the adult liver. In humans, the levels of mRNAs for hydroxymethylglutaryl-CoA synthase and prenyltransferase, two key enzymes in the cholesterol biosynthetic pathway, and for ApoA1, ApoAII, and ApoB, three key apoproteins in lipoprotein transport, are much higher in fetal liver than in adult liver (46). Thus, fetal liver probably exhibits hyperactive rates in endogenous cholesterol and lipoprotein synthesis. It would be advantageous to have two isoforms to provide higher ACAT activities for the protection against free cholesterol toxicity and for lipoprotein synthesis at high capacity. In the adult human liver, a minor but measurable amount (about 10-15%) of total ACAT activity can be attributed to ACAT-2. In the intestine, it is generally believed that dietary cholesterol absorption mainly takes place in the villi of the proximal end (47). The finding that ACAT-2 is more concentrated in the villus region, whereas ACAT-1 is distributed uniformly along the villus-crypt axis, implies that ACAT-2 may play a more important role in dietary cholesterol absorption. In addition, within a single intestinal enterocyte, various intracellular cholesterol-cholesteryl ester pools may exist potentially, in order to cope with various intracellular cholesterol trafficking activities that occur in a polarized manner. The compartmentalized cholesterol-cholesteryl ester pools, originally proposed by Stange et al. (31), may be associated with different ACAT isoenzymes. The HepG2 cells and Caco-2 cells, shown to contain ACAT-1 and ACAT-2 in proportions that closely resemble the in vivo situation, should serve as useful tools to further investigate the roles of these two isoenzymes in the liver and intestine.

	ACKNOWLEDGEMENTS

We thank Drs. Andrea Quaroni for antiserum against sucrase; Jeffrey Field for Caco-2 cells; Aaron Barchowsky for advice on fluorescence microscopy; and Leaf Huang for advice on human tissue availability. We thank Peter Seery and Barbara K. Schaeffer for providing the human tissues and Maudine Waterman for performing immunohistochemical staining. We also thank Yin Li, Akira Miyazaki, Rick Reid, Jonathan Cruz, Chunjiang Yu, Xiaohui Lu, Shigeki Sugii, and Walter Chang for stimulating discussions.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants HL 36709 and HL 60306 (to T.-Y. C.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

ⁱ To whom correspondence should be addressed. Tel.: 603-650-1622; Fax: 603-650-1128; E-mail: Ta.yuan.Chang@dartmouth.edu.

Published, JBC Papers in Press, June 8, 2000, DOI 10.1074/jbc.M003927200

	ABBREVIATIONS

The abbreviations used are: ACAT, acyl-coenzyme A:cholesterol acyltransferase; CHAPS, 3-[(3-cholamidopropyl)-dimethylammonil]-1-propane sulfonate; CHO, Chinese hamster ovary; Me₂SO, dimethyl sulfoxide; ER, endoplasmic reticulum; GST, glutathione S-transferase; PBS, phosphate-buffered saline; PC, phosphatidylcholine; PAGE, polyacrylamide gel electrophoresis; DTT, dithiothreitol; LDL, low density lipoprotein; VLDL, very low density lipoprotein.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. Iqbal, L. L. Rudel, and M. M. Hussain Microsomal Triglyceride Transfer Protein Enhances Cellular Cholesteryl Esterification by Relieving Product Inhibition J. Biol. Chem., July 18, 2008; 283(29): 19967 - 19980. [Abstract] [Full Text] [PDF]

				J. M. Brown, T. A. Bell III, H. M. Alger, J. K. Sawyer, T. L. Smith, K. Kelley, R. Shah, M. D. Wilson, M. A. Davis, R. G. Lee, et al. Targeted Depletion of Hepatic ACAT2-driven Cholesterol Esterification Reveals a Non-biliary Route for Fecal Neutral Sterol Loss J. Biol. Chem., April 18, 2008; 283(16): 10522 - 10534. [Abstract] [Full Text] [PDF]

				R. E. Temel, L. Hou, L. L. Rudel, and G. S. Shelness ACAT2 stimulates cholesteryl ester secretion in apoB-containing lipoproteins J. Lipid Res., July 1, 2007; 48(7): 1618 - 1627. [Abstract] [Full Text] [PDF]

				A. A. Spector and W. G. Haynes LDL Cholesteryl Oleate: A Biomarker for Atherosclerosis? Arterioscler. Thromb. Vasc. Biol., June 1, 2007; 27(6): 1228 - 1230. [Full Text] [PDF]

				A. Turkish and S. L. Sturley Regulation of Triglyceride Metabolism. I. Eukaryotic neutral lipid synthesis: "Many ways to skin ACAT or a DGAT" Am J Physiol Gastrointest Liver Physiol, April 1, 2007; 292(4): G953 - G957. [Abstract] [Full Text] [PDF]