| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,1,,
AstraZeneca R&D, Mölndal, Sweden; and
Wallenberg Laboratory for Cardiovascular Research, Göteborg University, Sweden
1Correspondence: Department of Integrative Pharmacology, AstraZeneca R&D, S-431 83 Mölndal, Sweden. E-mail: daniel.linden{at}astrazeneca.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: GPAT • stearoyl-CoA desaturase • ADRP • adenovirus • fatty liver
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The prevalence of fatty liver in the general population has been estimated to be between 14 and 34% (2
3
4)
, and the prevalence increases to 58–76% in obese subjects (3
, 5)
. Fatty liver is associated with several features of insulin resistance and hypertriglyceridemia (6
7
8)
, and has been proposed to be included as a component of the metabolic syndrome (8)
.
The first committed and rate-limiting step in glycerophospholipid and TAG biosynthesis is the formation of lysophosphatidic acid (LPA) from glycerol-3-phosphate catalyzed by glycerol-3-phosphate acyltransferase (GPAT) (9)
. Two unique enzymes display GPAT activity; one NEM-sensitive isoform associated with the endoplasmic reticulum (ER, microsomal GPAT) and a NEM-resistant isoform associated with the outer mitochondrial membrane (mtGPAT) (10)
. In addition, a NEM-resistant mtGPAT activity has been detected in mtGPAT-deficient mice, but this activity was absent in normal mice (11)
. In most tissues the mtGPAT constitutes 
10% of the total activity, but in the liver mtGPAT comprises up to 50% of the total activity (10)
.
MtGPAT expression and activity are influenced both by diet and obesity. Hepatic mtGPAT mRNA level is up-regulated by a high carbohydrate, fat-free diet and by insulin administration to streptozotocin-diabetic mice (12)
. The promoter of mtGPAT contains both sterol- and carbohydrate-responsive elements (13
, 14)
. We have shown that mtGPAT activity is increased in livers from both diet-induced obese mice and leptin-deficient ob/ob mice compared with their lean controls (15)
.
Overexpressing mtGPAT in Chinese hamster ovary (CHO) cells or primary cultures of rat hepatocytes dramatically increases lipid content and incorporation of labeled fatty acids into diacylglycerol (DAG), TAG and phospholipids (15
16
17)
. Overexpression of mtGPAT in rat hepatocytes results in a sharp reduction in fatty acid oxidation (15
, 17)
, indicating a competition for acyl-CoAs between mtGPAT and carnitine palmitoyltransferase I (CPT-I), both located on the outer mitochondrial membrane (18
, 19)
. Thus, mtGPAT can divert acyl-CoAs away from ß-oxidation and toward TAG biosynthesis. The opposite may occur during mtGPAT inhibition because mtGPAT-deficient (mtGPAT–/–) mice on a diet enriched in fat and sucrose have lower hepatic TAG content and increased levels of plasma ß-hydroxybutyrate compared with wild-type mice (20
, 21)
.
Enzymes that catalyze the final steps in TAG biosynthesis are located in the ER (22)
. Thus, the LPA formed by mtGPAT and/or phosphatidate produced by the subsequent enzymatic step must be transported to the ER for final TAG biosynthesis. Newly formed TAG from extracellular nonesterified fatty acids or de novo synthesized fatty acids usually are not directly available for the secretory pool (23)
. Rather, the stored TAG is hydrolysed to DAG, which is re-esterified to TAG before VLDL assembly (24
, 25)
. Thus, even though mtGPAT has been shown to increase TAG biosynthesis in cultured cells (15
16
17)
, the effect of increased mtGPAT expression in the liver of intact animals is not known.
In this study we investigated the in vivo effects of hepatic mtGPAT overexpression on liver lipid metabolism. Chow-fed male C57BL/6 mice were transduced with control (zsGreen) adenovirus or adenoviruses expressing either wild-type murine mtGPAT or a catalytically inactive variant in which an invariant aspartic acid residue was changed to glycine [D235G] in the active site of the enzyme (15
, 26
, 27)
. Overexpression of mtGPAT resulted in decreased ex vivo fatty acid oxidation and massive liver accumulation of TAG and DAG without affecting phospholipid or cholesterol ester levels. In addition, mtGPAT overexpression increased liver TAG secretion rate 2-fold and elevated plasma TAG and cholesterol levels. Thus, in vivo hepatic mtGPAT overexpression increases both storage and secretion of TAG leading to hepatic steatosis and dyslipidemia.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recombinant adenoviruses
The construction of recombinant replication-deficient adenoviral CMV promoter containing vectors (Ad) expressing either fully active wild-type mtGPAT (Ad-mtGPAT), or a catalytically inactive point-mutated [D235G] mtGPAT (Ad-mtGPATmut) has been described (15)
. As an additional control, Ad-zsGreen was used in some studies (15)
. After large-scale amplification in AD-293 cells, recombinant adenoviruses were purified by two rounds of CsCl density gradient ultracentrifugation. The purified virus stocks were desalted over 10DG columns and eluted in sterile PBS. Glycerol (15%) was added and the virus stocks were batched and stored at –80°C until use. Infectious viral titers were determined using the Adeno-X Rapid Titer Kit. All purified virus stocks were screened for possible wild-type virus contamination (28)
before injection.
Animal experiments
All procedures described in this article involving animals were conducted in accordance with accepted standards of humane animal care and approved by the Ethics Committee of Göteborg University. Seven-wk-old male C57BL/6 mice (Harlan Netherlands B.V.), caged individually were given free access to water and regular chow containing 12% fat, 62% carbohydrates, and 26% protein with a total energy content of 3.0 Kcal/g (R3 diet, Lactamin AB, Stockholm, Sweden). Animals were maintained under standardized conditions of temperature (21–22°C) and humidity (40–60%), with lights on between 0600 h and 1800 h. After 1 wk the mice were randomized by body weight and received a single dose (1.6x109 infectious units (ifu)) of Ad-mtGPAT, Ad-mtGPATmut, or Ad-zsGreen adenoviruses in 250 µL PBS or PBS alone by tail vein injections. Body weight and food consumption were recorded. Four or 11 days after virus administration, food was withdrawn 3 h before collection of blood by heart puncture of isoflurane anesthetized animals. Tissues were collected, snap frozen, and stored at –80°C. Plasma was routinely screened for presence of adenoviruses (28)
, but the blood clearance rate for all our recombinant adenoviruses was found to be very rapid and viruses were nondetectable in plasma 1 h post-transduction (data not shown), in accordance with a previous study (29)
.
Measurement of plasma metabolic variables
Plasma levels of TAG, cholesterol, glucose, and ALT activity were measured using commercially available kits. Plasma insulin levels were determined by a rat insulin RIA. Plasma apolipoprotein B (apoB) concentrations were determined with an electroimmunoassay as described previously (30)
. The size distribution profiles of serum lipoproteins were measured on 10 µL plasma from individual mice using a high-performance liquid chromatography system (HPLC), SMART, and a Superose 6 PC 3.2/30 column as described before (30)
.
Membrane isolation and measurement of GPAT activity
Liver samples were homogenized in sucrose solution (250 mM sucrose, 10 mM Tris, pH 7.4, 1 mM EDTA, and 1 mM DTT) and nuclei removed by centrifuging at 600 g for 15 min, 4°C. Supernatants were then spun at 8000 x g for 15 min at 4°C to pellet crude mitochondria or 100,000 g for 1 h 4°C to obtain both mitochondria and microsomes. The 100,000 g pellets were then tested for GPAT activity in the presence and absence of 2 mM N-ethylmaleimide (NEM) as described previously (15)
. The use of NEM sensitivity on a single high-speed spin pellet has been previously demonstrated to effectively differentiate mitochondrial (NEM resistant) from ER-associated (NEM-sensitive) GPAT activity (16
, 31
, 32)
.
Western blot
Crude mitochondria preparations supplemented with protease inhibitors (CompleteTM) were used to estimate the extent of mtGPAT expression using Western blot analyses and anti-mtGPAT antibodies directed against the murine amino acid sequence 312-326 as described before (15)
. For total liver protein isolation, liver tissue was homogenized in 0.05 M Tris-HCl, 2% sodium dodecyl sulfate, 15% sucrose, 5 mM dithiothreitol, and protease inhibitors (CompleteTM), followed by denaturation at 95°C and centrifugation at 100,000 x g for 1 h. Proteins (30 µg) in supernatant were used in Western blot as described before (33)
using either mouse FAS (diluted 1:500) or guinea pig ADRP (diluted 1:2000) antibodies, respectively. HRP-rabbit anti-mouse (diluted 1:7500) or HRP-rabbit anti-guinea pig (diluted 1:10 000) antibodies were used before development using Super Signal West Pico Chemiluminescent substrate.
Histology
Tissues were fixed in 4% buffered paraformaldehyde. For hematoxylin and eosin staining, tissues were dehydrated and embedded in paraffin. For Oil Red O staining, livers were embedded in Tissue-Tek OCT, snap frozen and stored at –80°C.
Measurement of liver lipid and malonyl-CoA mass
Measurement of liver lipids was performed essentially as described (34)
. After extraction of lipids (35)
, samples were dissolved in a mix of solutions A and B (see below) at a ratio of 9:1. The mobile phase was created by a combination of three solvent mixtures: A) heptane/tetrahydrofuran 99:1 (v/v), B) acetone/dichloromethane 2:1 (v/v), C) isopropanol/water 85:15 (v/v) containing 7.5 mM acetic acid and 7.5 mM ethanolamine. Samples were separated on an HPLC system using a Waters spherisorb, 5 µm silica column. Lipids were detected with a PL-ELS 1000 Evaporative Light Scattering Detector from Polymer Laboratories and HPLC Mix 41 was used as standard. Malonyl-CoA was quantified in mouse liver extracts (in 6% perchloric acid) by HPLC coupled to a mass spectrometer using electrospray ionization (MSQ, Thermo Electron). The extracts were directly separated on an Aquasil C18 column, using a gradient of 1% acetonitrile to 90% acetonitrile (all containing 10 mM ammonium acetate; flow rate 0.5 mL/min). The optimized settings for the MS detector were cone voltage: 120V, probe temp. 350°C, needle voltage: 3.0 kV. The signal was recorded in positive ion SIM mode at m/z = 854.2 ± 0.5, and it was found to be linear in the range of approx. 0.4–40 ng malonyl CoA.
Measurement of fatty acids in DAG and TAG
After extraction of lipids (35)
, DAG and TAG were isolated by thin-layer chromatography using a solvent system consisting of petroleum ether-diethyl ether-acetic acid (80:20:1, vol/vol/vol). DAG and TAG in TLC spots were dissolved in methyl tert-butyl ether and trans-esterified into fatty acid methyl esters (FAMEs) with 0.5 M sodium methoxide. FAMEs were extracted into hexane and analyzed using a HP 5890 gas chromatograph (Agilent, Palo Alto, CA, USA) equipped with a DB-23 column, using hydrogen as carrier gas, and a flame ionization detector. Individual FAMEs (16:0, 16:1 (9c), 18:0, 18:1(9c), 18:2(9c, 12c), 18:3(9c, 12c, 15c) and 20:0) were identified and quantified by comparison with FAME calibration samples (Larodan) using Chromeleon (Dionex, Synnyvale, CA, USA) chromatography software.
Mass spectrometric analysis of fatty acids in phospholipids
Extraction of lipids and sample preparation prior to mass spectrometric phospholipid analysis were performed as described previously (36)
. Analysis of individual phospholipids was performed by precursor ion scanning on a QSTAR XL QqTOF mass spectrometer (MDS Sciex, Concord, ON, Canada) as described (37)
equipped with an automated chip-based nanoelectrospray NanoMate HD system (Advion BioSciences, Ithaca, NY, USA). Briefly, 5 µL of sample solution (chloroform-methanol (1:2, v/v) containing 5 mM ammonium acetate) was delivered to the back plane of the nanoESI chip. In negative ion mode, electrospray process was initiated by applying 1.1 kV and 0.2 psi nitrogen head pressure to the sample in the pipette tip to ensure constant sample flow to the chip. For fatty acid scanning analysis (37)
of infused lipid extracts, precursor ion spectra were simultaneously acquired for 30–50 acyl anion fragments of fatty acid moieties, containing 12 to 22 carbon atoms and 0 to 6 double bonds. Collision energy was set at 40 eV in negative ion mode unless specified otherwise. Fragment ions were selected within an m/z window of 0.15 Th. Fatty acid scanning spectra were interpreted using a prototype of LipidProfiler 1.0 software (MDX Sciex). Individual phospholipids were determined by monitoring the ratio between peak areas of detected fatty acid signals, e.g., Fa1/Fa2.
Measurement of in vivo liver TAG secretion rate
TAG secretion rate in vivo was measured by intravenous administration of Triton WR-1339 (38)
. The animals were fasted for 4 h to avoid the influence of chylomicrons from the intestine. Thereafter, the mice were anesthetized using a combination of ketamine hydrochloride (77 mg/kg; Ketalar) and Xylazine (9 mg/kg; Rompun), and injected intravenously with Triton WR-1339 diluted in saline (200 mg/mL) via the jugular vein (500 mg/kg body weight). Blood samples were taken before the injection and 20, 40 and 60 min after Triton WR-1339 administration during anesthesia. Plasma TAG levels were analyzed as described above, and accumulation of TAG was calculated using published plasma volume in a normal male mouse (0.071 mL/g body weight) (38)
. Liver TAG secretion rate, expressed as µmol/min/kg body weight, was calculated from the slope of the curve.
Measurement of fatty acid oxidation
Fatty acid oxidation activity was measured similarly to that described (39)
. Fresh liver was homogenized in 0.25 M sucrose containing 1 mM EDTA and 1–10 mg homogenate was incubated in 0.2 mL assay medium (150 mM KCl, 10 mM HEPES (pH 7.2), 0.1 mM EDTA, 1 mM potassium phosphate buffer (pH 7.2), 5 mM malonate, 10 mM MgCl2, 1 mM carnitine, 0.5% BSA, 5 mM ATP, and 50 µM palmitic acid containing 1 µCi of [9,10(n)-3H]palmitic acid). The reaction was run for 30 min at 25°C and stopped by addition of fatty acid free BSA (1% final concentration) followed by addition of 11 µL 70% perchloric acid. The mixture was vortexed extensively and centrifuged 10,000 x g for 5 min and unreacted fatty acids in supernatant were removed by two extractions with n-hexane. Radioactive degradation products in the water phase were counted and expressed as pmol/min/µg protein or nmol/min/liver.
cDNA synthesis and quantitative RT-PCR
Total RNA was isolated from tissues with Trizol according to the manufacturer’s protocol. DNA-free was used to remove DNA from the RNA preparations. cDNA was synthesized using the Invitrogen SuperscriptTM synthesis system. Real-time RT PCR analysis was performed in presence or absence of reverse transcriptase using the ABI Prism 7900 Sequence Detection system (Applied Biosystems, Foster City, CA, USA). All reactions were analyzed in triplicate and the expression data was normalized to an endogenous control, acidic ribosomal phosphoprotein P0 (36B4). The relative expression levels were calculated according to the Eq. 2-dCt, where dCt is the difference in cycle threshold (Ct) values between the target and the 36B4 endogenous control. Specific primers and probes for each gene (Suppl. Table 1) were designed with Primer ExpressTM software (Applied Biosystems) and gene sequences available from GenBank database. Endogenous mtGPAT mRNA expression was measured using primers and probe directed against the 3'-untranslated region of the gene and exogenous mtGPAT mRNA expression was measured using primers and probe directed against the FLAG epitope of the construct (15)
.
Statistics
Values are expressed as means ± SE. Comparisons between groups were made by Student’s t test or 1-way ANOVA (ANOVA), followed by Tukey’s test between individual groups. Values were transformed to logarithms when appropriate.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
No green fluorescence or exogenous mtGPAT mRNA expression was detected 4 days after virus administration in soleus or quadriceps muscle, heart, brain, spleen, lung, kidney, or gonadal white adipose tissue after transduction with Ad-zsGreen or Ad-mtGPAT, respectively (data not shown). Ad-mtGPAT and Ad-mtGPATmut transduction markedly increased liver total mtGPAT mRNA levels compared with PBS or Ad-zsGreen-treated animals after 4 days (Fig. 1
A). Endogenous mtGPAT mRNA expression was not changed in the liver by mtGPAT overexpression (Fig. 1A
) and exogenous mtGPAT mRNA (overexpressing construct) was only detected when either Ad-mtGPAT or Ad-mtGPATmut were used (data not shown). MtGPAT protein expression in mitochondria was markedly increased after Ad-mtGPAT or Ad-mtGPATmut treatment as compared with PBS or Ad-zsGreen-treated animals (Fig. 1B
). Ad-mtGPAT treatment increased total GPAT activity by 60% and NEM-resistant mtGPAT activity 3.1-fold in the liver compared with Ad-zsGreen-treated mice (Fig. 1C
). Despite the high mtGPAT mRNA and protein expression in the liver after Ad-mtGPATmut treatment, mitochondrial GPAT activity was unaffected (Fig. 1C
). Thus, the mtGPAT overexpression is liver-specific and the single amino acid substitution [D235G] in Ad-mtGPATmut gives rise to an inactive enzyme in vivo in mice.
|
After 11 days, a less marked increased total mtGPAT mRNA (9.5-fold induction) and protein expression as well as mtGPAT activity (2.3-fold induction, P<0.01, Student’s t test) was observed in the liver after Ad-mtGPAT treatment compared with Ad-zsGreen-treated animals (data not shown). Thus, the overexpression of mtGPAT decreased between day 4 and 11.
Effects on liver histology and lipid content
Body weight gain and daily food intake were not different between Ad-mtGPAT and Ad-mtGPATmut-treated animals during the 4 day period after virus administration (Table 1
). After 4 days, Ad-mtGPAT-treated mice had 13% heavier livers than Ad-mtGPATmut-treated mice but there was no difference in gonadal white adipose tissue weight (Table 1)
. Eleven days after virus administration there were no differences in either liver or gonadal white adipose tissue weights (data not shown). Histological analysis of livers from Ad-mtGPAT-treated mice revealed the presence of numerous hepatocyte vacuoles that stained positive for lipids using oil red O staining compared with Ad-mtGPATmut-treated livers (Fig. 2
A). In addition, liver lipid mass was analyzed 4 and 11 days after virus administration using HPLC. Ad-mtGPAT treatment resulted in 12-fold increased liver TAG levels and 6.7-fold increased DAG levels compared with Ad-zsGreen transduction (Fig. 2B
). A less marked effect on liver TAG and DAG levels (3-fold and 1.7-fold, respectively, P < 0.05, Student’s t test) was seen 11 days after Ad-mtGPAT treatment compared with Ad-zsGreen-treated mice (data not shown), in line with a decreased mtGPAT overexpression. Adipocyte differentiation-related protein (ADRP) is expressed in hepatocytes and regarded as a sensitive marker for accumulation of intracellular lipid droplets (40)
. In line with the increased lipid droplet accumulation, mtGPAT but not mtGPATmut overexpression resulted in a massive protein expression of ADRP in the liver (Fig. 2C
).
|
|
MtGPAT overexpression did not affect liver content of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine, sphingomyelin, lysolecithin, cholesterol, or CE after 4 or 11 days (Suppl. Table 2, data not shown). Results indicate that hepatic mtGPAT overexpression directs fatty acids into DAG and TAG rather than phospholipids or CE in vivo in mice leading to hepatic steatosis. In contrast to Ad-zsGreen-treated mice, Ad-mtGPAT but not Ad-mtGPATmut-treated mice had elevated plasma ALT activity after 4 days (PBS=25.8±1.3 U/L, Ad-zsGreen=15.9±6.7 U/L, Ad-mtGPAT=934±249 U/L, and Ad-mtGPATmut=380±223 U/L), but there was no difference between the groups after 11 days (data not shown). Two individuals in the Ad-mtGPATmut group had high ALT levels after 4 days. However, these two individuals did not differ from PBS or Ad-zsGreen-treated mice in terms of liver or plasma lipid levels (data not shown). This indicates that the transient elevation in ALT after Ad-mtGPAT transduction is due to the transient hepatic steatosis and not to the viral exposure in itself.
Effects on fatty acid composition in DAG, TAG and phospholipids
Saturated fatty acids and especially palmitic acid (16:0) have been shown to be the preferred substrate for mtGPAT (41
42
43
44)
. As expected, both palmitic acid and stearic acid (18:0) content of DAG and TAG were increased in mtGPAT overexpressing livers (Fig. 3
A, B). Surprisingly, 18:1 was increased in DAG (2.3-fold) and both 16:1 and 18:1 were even more markedly increased in TAG (21- and 15-fold, respectively). Also 18:2 was increased in TAG after mtGPAT transduction (Fig. 3A, B
). Other quantified fatty acids were too low in abundance for quantitative analyses. To understand why the content of mono-unsaturated fatty acids in DAG and TAG increased we determined the expression of stearoyl-CoA desaturase-1 (SCD-1), the major enzyme that catalyses mono-unsaturation of fatty acids in liver (45)
. Indeed, SCD-1 mRNA expression was up-regulated 10-fold in the liver after mtGPAT overexpression (Fig. 3C
).
|
In parallel with the increased levels of the essential fatty acid 18:2 in TAG the amount of fatty acid combinations in PC (not in PE, data not shown) containing 18:2 or metabolites thereof (such as 20:3, 20:4, and 22:6) decreased (Fig. 3D
). This indicates a redistribution of 18:2 from PC toward TAG biosynthesis. In addition, the ratio between respective fatty acid signal (e.g., Fa1/Fa2) was calculated and used to determine changes in individual PE species. MtGPAT overexpression specifically enriched the individual 16:0/22:5, 18:0/18:2, 18:0/20:4 and 18:0/22:5 PE species over 22:5/16:0, 18:2/18:0, 20:4/18:0 and 22:5/18:0 PE (Fig. 3E
). Although the same trend was observed for PC, no statistically significant changes were observed (data not shown). Thus, although mtGPAT overexpression in vivo did not affect total amount of phospholipids, the saturation of the first position fatty acid in liver PE increased.
Effects on genes of importance for lipogenesis and carbohydrate metabolism
To investigate whether the hepatic steatosis after mtGPAT overexpression could influence expression of genes other than SCD-1 in lipogenesis, we determined the mRNA expression level of several important transcription factors involved with lipogenesis: sterol regulatory element binding protein-1 (SREBP-1) and peroxisome proliferator-activated receptor (PPAR)
1 and PPAR2. However, the expression of these genes was not influenced by mtGPAT overexpression (data not shown). In contrast to SCD-1 mRNA, none of the other investigated SREBP-1 downstream target genes; ATP citrate lyase, acetyl-CoA carboxylase 1 (ACC1) mRNA levels or fatty acid synthase (FAS) protein levels were influenced by mtGPAT overexpression (data not shown). In addition, mtGPAT overexpression did not affect liver malonyl-CoA levels (data not shown). Thus, results indicate that the lipid accumulation in the liver after mtGPAT overexpression is not secondary to increased de novo lipogenesis.
We also determined the mRNA expression levels of phosphoenolpyruvate carboxykinase and glucokinase, two enzymes involved in gluconeogenesis and glycolysis, respectively. Neither of these two genes was transcriptionally regulated after mtGPAT overexpression (data not shown).
Effects on liver fatty acid oxidation activity and liver TAG secretion rate
We have previously demonstrated that rat hepatocytes transduced with mtGPAT show decreased fatty acid oxidation (15)
. To test whether this occurs in vivo, total liver homogenates were assayed for ex vivo fatty acid oxidation. Hepatic overexpression of mtGPAT resulted in reduced liver fatty acid oxidation compared with Ad-mtGPATmut-treated mice both when related to amount of protein (Fig. 4
A) and total liver mass (Ad-mtGPAT (n=5) = 12.63 ± 0.83 nmol/min/liver vs. Ad-mtGPATmut (n=6) = 16.02 ± 0.71 nmol/min/liver; P <0.01, Student’s t test). MtGPAT overexpression did not affect mRNA levels of PPAR
or its downstream target genes: acyl-CoA oxidase (ACO), medium-chain acyl-CoA dehydrogenase (MCAD) or CPT-1 (data not shown). The results indicate that mtGPAT can divert acyl-CoAs from CPT-I and ß-oxidation and toward TAG biosynthesis in vivo.
|
Ad-mtGPAT treatment had no effect on plasma glucose or insulin levels (Table 2
). However, both plasma TAG and cholesterol levels were increased (27% and 20%, respectively) after 4 days in mice treated with Ad-mtGPAT compared with Ad-mtGPATmut (Table 2)
, indicating that the VLDL secretion from the liver may be affected by hepatic mtGPAT overexpression. Ad-mtGPATmut transduced mice did not differ from PBS injected mice in terms of plasma TAG and cholesterol levels (data not shown). In a separate experiment, mice received either Ad-mtGPAT or Ad-mtGPATmut and after 4 days they were fasted for 4 h and injected with Triton WR-1339 to determine the liver TAG secretion rate. Ad-mtGPAT-treated mice had 86% higher liver TAG secretion rate (Fig. 4B
) than AdmtGPATmut-treated mice.
|
In non-Triton WR-1339-treated animals, lipoprotein size liquid chromatography revealed no major differences in cholesterol lipoprotein distribution after mtGPAT overexpression (data not shown). In addition, mtGPAT overexpression did not affect plasma levels of apolipoprotein B (Table 2)
, showing that the number of apoB-containing particles in the circulation was not affected. Thus, in addition to a massive liver lipid accumulation, mtGPAT overexpression leads to an increased TAG secretion from the liver.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our findings are consistent with the assumption that mtGPAT competes with CPT-I for acyl-CoAs when driving TAG synthesis. No change in the amount of FAS or hepatic malonyl-CoA levels indicates that the decreased fatty acid oxidation is not secondary to increased fatty acid synthesis. The increased plasma TAG level after mtGPAT overexpression was explained by a nearly 2-fold increased hepatic TAG secretion rate. Thus, the increased cytosolic TAG pool associated with increased ADRP expression after mtGPAT overexpression was available for hydrolysis and subsequent TAG synthesis in ER for VLDL assembly in vivo.
The increase in mono-unsaturated fatty acids in response to mtGPAT overexpression suggests an unexpected link between SCD-1 and mtGPAT. While microsomal GPAT esterifies both saturated and unsaturated fatty acyl-CoAs equally well, mtGPAT has been shown to have a preference for saturated fatty acids and in particular palmitoyl-CoA (16:0) (41
42
43
44)
. Overexpression of mtGPAT increased the total amount of 16:0 and 18:0 in DAG and TAG. Additionally, 16:1 and 18:1 increased markedly in TAG in association with increased expression of SCD-1. The essential fatty acid 18:2 was redistributed from PC toward TAG possibly due to a markedly increased TAG biosynthesis in the liver in the context of unchanged food intake. Consistent with our results, mtGPAT-deficient mice show decreased 16:0 content in the sn-1 fatty acid position of both PC and PE (21)
, decreased 18:1 LPA and reduced SCD-1 expression in the liver (20)
. The mechanism for the up-regulation of SCD-1 expression after mtGPAT overexpression is unclear. However, it indicates a novel mechanism not involving PPAR or SREBP-1 regulation since the expression of these genes and other downstream target genes besides SCD-1 were unaffected.
The markedly increased DAG levels in the livers after mtGPAT overexpression are in line with studies in primary cultures of rat hepatocytes (15
, 17)
. The reason for the accumulation of DAG is not known, but indicates that acyl CoA:diacylglycerol acyltransferase activity is rate limiting for TAG synthesis in vivo. Recently, mtGPAT–/– mice were shown to have lower hepatic levels of DAG and to be protected from hepatic insulin resistance possibly due to lower DAG-mediated protein kinase C
activation in the liver compared with wild-type mice (46)
. Thus, the massive DAG accumulation after mtGPAT overexpression may lead to impaired hepatic insulin sensitivity.
MtGPAT overexpression did not influence liver CE content, consistent with our previous finding demonstrating no effect on CE biosynthesis after mtGPAT overexpression in primary hepatocytes (15)
. Also consistent with this observation, mtGPAT–/– mouse liver shows no change in CE content (21)
. These findings indicate that the TAG and CE biosynthesis pathways do not share the same pool of fatty acids in the liver. A recent in vitro study found that mtGPAT overexpression in primary cultures of rat hepatocytes markedly decreased both the amount and biosynthesis of CE in spite of unchanged acyl-CoA:cholesterol acyltransferase activity (17)
. In that study, mtGPAT activity was enhanced 13-fold (17)
. In the present in vivo study and our previous study in primary hepatocytes (15)
, mtGPAT activity was induced 3-fold, similar to the elevation seen after refeeding experiments in rats (3-fold increased mtGPAT activity) (47)
and in ob/ob mice (2.3-fold increased mtGPAT activity compared with lean controls) (15)
. Thus, the decreased amount and biosynthesis of CE (17)
may have been due to substrate shortage in the in vitro model due to the supra-physiological increase in mtGPAT activity. The decreased CE levels may also have influenced the apparent lack of effect of mtGPAT on TAG secretion (17)
, since hepatic CE content has been shown to be of importance for liver TAG secretion in vivo (48)
.
Several murine models of obesity and diabetes, including ob/ob and KKAy develop fatty livers that express enhanced levels of PPAR (49
, 50)
, a key regulator of adipocyte differentiation and lipid storage. Increased PPAR expression in the liver leads to increased expression of genes, associated with steatotic livers, including SREBP-1 and FAS (51
, 52)
. MtGPAT overexpression did not influence PPAR1, PPAR2, SREBP-1, ATP citrate lyase or ACC1 mRNA expression, FAS protein expression or malonyl-CoA levels, indicating that the hepatic steatosis occurred without up-regulation of de novo lipogenesis.
In conclusion, liver-specific overexpression of mtGPAT in vivo resulted in massive liver accumulation of TAG and DAG without affecting phospholipid or CE levels. In addition, mtGPAT overexpression decreased liver fatty acid oxidation, confirming in vivo that mtGPAT is a driving component of hepatic lipid accumulation associated with obesity and insulin resistance. MtGPAT overexpression also increased the liver TAG secretion rate associated with increased plasma TAG and cholesterol levels. Thus, in vivo hepatic mtGPAT overexpression increases both storage and secretion of TAG leading to hepatic steatosis and associated dyslipidemia. Since the prevalence of fatty liver is markedly high in obese subjects (3
, 5)
, liver mtGPAT, which is elevated in obesity, may be an attractive target for treatment of NAFLD and associated dyslipidemia and hepatic insulin resistance.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 3, 2005. Accepted for publication October 13, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Horikawa, T. Nomura, T. Hashimoto, and K. Sakamoto Elongation and Desaturation of Fatty Acids are Critical in Growth, Lipid Metabolism and Ontogeny of Caenorhabditis elegans J. Biochem., August 1, 2008; 144(2): 149 - 158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Q. Chen, M.-S. Kuo, S. Li, H. H. Bui, D. A. Peake, P. E. Sanders, S. J. Thibodeaux, S. Chu, Y.-W. Qian, Y. Zhao, et al. AGPAT6 Is a Novel Microsomal Glycerol-3-phosphate Acyltransferase J. Biol. Chem., April 11, 2008; 283(15): 10048 - 10057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Lelliott, A. Ljungberg, A. Ahnmark, L. William-Olsson, K. Ekroos, A. Elmgren, G. Arnerup, C. C. Shoulders, J. Oscarsson, and D. Linden Hepatic PGC-1 Overexpression Induces Combined Hyperlipidemia and Modulates the Response to PPAR{alpha} Activation Arterioscler. Thromb. Vasc. Biol., December 1, 2007; 27(12): 2707 - 2713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. Mitchell, K. Ekroos, S. J. Blanksby, A. J. Hulbert, and P. L. Else Differences in membrane acyl phospholipid composition between an endothermic mammal and an ectothermic reptile are not limited to any phospholipid class J. Exp. Biol., October 1, 2007; 210(19): 3440 - 3450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Huang, C. Yang, Y. Luo, C. Jin, F. Wang, and W. L. McKeehan FGFR4 Prevents Hyperlipidemia and Insulin Resistance but Underlies High-Fat Diet Induced Fatty Liver Diabetes, October 1, 2007; 56(10): 2501 - 2510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Nagle, J. An, M. Shiota, T. P. Torres, G. W. Cline, Z.-X. Liu, S. Wang, R. L. Catlin, G. I. Shulman, C. B. Newgard, et al. Hepatic Overexpression of Glycerol-sn-3-phosphate Acyltransferase 1 in Rats Causes Insulin Resistance J. Biol. Chem., May 18, 2007; 282(20): 14807 - 14815. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Gonzalez-Baro, |
