Acid Sphingomyelinase Deficiency Prevents Diet ...

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 284, NO. 13, pp. 8359 –8368, March 27, 2009 © 2009 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

Acid Sphingomyelinase Deficiency Prevents Diet-induced Hepatic Triacylglycerol Accumulation and Hyperglycemia in Mice* Received for publication, October 9, 2008, and in revised form, December 3, 2008 Published, JBC Papers in Press, December 11, 2008, DOI 10.1074/jbc.M807800200

Gergana M. Deevska‡, Krassimira A. Rozenova‡, Natalia V. Giltiay‡, Melissa A. Chambers‡1, James White§, Boris B. Boyanovsky‡, Jia Wei¶, Alan Daugherty储, Eric J. Smart§, Michael B. Reid‡, Alfred H. Merrill, Jr.¶, and Mariana Nikolova-Karakashian‡2 From the Departments of ‡Physiology, §Pediatrics, and 储Internal Medicine, University of Kentucky College of Medicine, Lexington, Kentucky 40536 and the ¶Department of Molecular Cell Biology, School of Biology and Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, Georgia 30332-0363

Obesity is the major risk factor associated with the development of glucose intolerance and type 2 diabetes. In obesity, large amounts of triacylglycerides (TAG)3 accumulate in liver, a

* This work was supported, in whole or in part, by National Institutes of Health Grant R01 AG 19223 (NIA to M. N. K.). This work was also supported by American Heart Association Scientist Development Grants 0130238N (to M. N. K.) and P20RR15592 (to E. J. S.) and LIPID MAPS Consortium Grant GM069338 (to A. H. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by Pre-doctoral Fellowship AHA0615263B. 2 To whom correspondence should be addressed: Dept. of Physiology, University of Kentucky College of Medicine, A. B. Chandler Medical Center, 800 Rose St., Lexington, KY 40536. Tel.: 859-323-8210; E-mail: [email protected]. 3 The abbreviations used are: TAG, triacylglyceride; ASMase, acid sphingomyelinase; DAG, diacylglycerol; DGAT, DAG acyltransferase; LDL, low density

MARCH 27, 2009 • VOLUME 284 • NUMBER 13

so-called “lean tissue,” and hepatic TAG accumulation is a better predictor for development of type 2 diabetes than body weight or body mass index (1). Liver plays a central role in the metabolism and storage of dietary fat (2). Depending on the hormonal and nutritional status of the organism, hepatocytes ␤-oxidize fatty acids derived from adipose tissue, chylomicron remnants, and other lipoprotein sources or reutilize them for the biosynthesis of TAG and other lipids. However, when dietary fat is in excess, TAG-rich droplets accumulate in the cytosol, resulting in “fatty liver“ and lipotoxicity (3, 4). Another lipid that accumulates is ceramide (5), which has been proposed to contribute to insulin resistance and lipotoxicity (6 – 8) because of suppression of IRS-1 phosphorylation in hepatocytes (9) and inhibition of Rac activation, Glut-4 translocation, and Akt-1 phosphorylation in muscle cells (10, 11). Ceramide is generated via de novo biosynthesis and turnover of complex sphingolipids such as sphingomyelin (SM). The ratelimiting step in the de novo pathway is catalyzed by serine palmitoyltransferase (SPT), which exhibits a high degree of specificity for the CoA-thioester of palmitic acid, the major saturated fatty acid found in the Western diet. Moreover, ceramide biosynthesis de novo has been shown to be influenced by the supply of palmitic acid in hepatocytes (12), muscle (8, 13, 14), and heart (15). Acid sphingomyelinase (ASMase), in turn, generates ceramide by hydrolysis of SM from the recycling/ endocytic pathway. ASMase might play a role in obesity because it is overexpressed in adipose tissue of ob/ob mice (16), and it appears to be involved in the pathogenesis of atherosclerosis (17), a disease which, similar to diabetes, is linked to obesity and to the consumption of diets rich in saturated fats. In contrast, patients with deficient ASMase activity (NiemannPick patients) maintain very low body weight (18). In this study we investigate the link between high fat diets, ceramide, and insulin resistance in vivo, and we provide evidence that ASMase plays a critical role in regulation of insulin sensitivity and blood glucose by modulating the partitioning of palmitic acid in sphingolipid and TAG pools in the liver.

lipoprotein receptor; VLDL, very LDL; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PS, phosphatidylserine; SPT, serine palmitoyltransferase; SM, sphingomyelin; ANOVA, analysis of variance; HPLC, high pressure liquid chromatography.

JOURNAL OF BIOLOGICAL CHEMISTRY

8359

Downloaded from www.jbc.org at University of Kentucky, on January 12, 2010

Acid sphingomyelinase plays important roles in ceramide homeostasis, which has been proposed to be linked to insulin resistance. To test this association in vivo, acid sphingomyelinase deletion (asmⴚ/ⴚ) was transferred to mice lacking the low density lipoprotein receptor (ldlrⴚ/ⴚ), and then offsprings were placed on control or modified (enriched in saturated fat and cholesterol) diets for 10 weeks. The modified diet caused hypercholesterolemia in all genotypes; however, in contrast to asmⴙ/ⴙ/ldlrⴚ/ⴚ, the acid sphingomyelinase-deficient littermates did not display hepatic triacylglyceride accumulation, although sphingomyelin and other sphingolipids were substantially elevated, and the liver was enlarged. asmⴚ/ⴚ/ldlrⴚ/ⴚ mice on a modified diet did not accumulate body fat and were protected against diet-induced hyperglycemia and insulin resistance. Experiments with hepatocytes revealed that acid sphingomyelinase regulates the partitioning of the major fatty acid in the modified diet, palmitate, into two competitive and inversely related pools, triacylglycerides and sphingolipids, apparently via modulation of serine palmitoyltransferase, a rate-limiting enzyme in de novo sphingolipid synthesis. These studies provide evidence that acid sphingomyelinase activity plays an essential role in the regulation of glucose metabolism by regulating the hepatic accumulation of triacylglycerides and sphingolipids during consumption of a diet rich in saturated fats.

Acid Sphingomyelinase and Diet-induced Hepatic Steatosis EXPERIMENTAL PROCEDURES Animals and Diets—A colony of ASMase-deficient mice (19) was maintained in Division of Laboratory Animal Resources of the University of Kentucky College of Medicine. ASMase deletion was bred to ldlr⫺/⫺ background by cross-breeding asm⫹/⫺ and ldlr⫺/⫺ (Ldlrtm1Her (C57Bl6), The Jackson Laboratory, Bar Harbor, ME). asm⫹/⫺/ldlr⫺/⫺ animals were maintained as a breeding heterozygous colony and used as a parental strain to generate asm⫺/⫺/ldlr⫺/⫺ and asm⫹/⫹/ldlr⫺/⫺ for the experiments. Eight-to-9-week-old male and female litter-matched animals were randomly placed either on a modified diet (TD.88137, adjusted calories diet, 42% from fat) (Harlan-Teklad, Indianapolis, IN) or continued on standard chow diet (2918, Teklad Global 18% protein rodent diet) (Harlan-Teklad)

Standard (chow) diet

High fat (modified) diet

g/kg

Protein DL-Methionine Carbohydrates (Starch) Crude oil (Saturated fats) (Palmitic acid) Cholesterol Crude fiber Mineral mix Calcium Vitamin mix

g/kg

189.0 3.5 573.0 (412.4) 60.0 (10.0) (7.6)

Protein DL-Methionine Sucrose (Corn starch) Milk fat (Saturated fats) (Palmitic acid) Cholesterol Cellulose Mineral mix Calcium Vitamin mix

38.0 22.2 10.0 3.0

195.0 3.0 341.5 (150.0) 210.0 (137.5) (36.0) 1.5 50.0 35.0 4.0 10.0

TABLE 2 Food and water consumption Mice were placed in metabolic cages for 7 days, and food and water intake and urine and feces excretion were measured daily. For each individual mouse, the values were averaged to calculate the mean daily value. Presented data are means for all mice in given group ⫾ S.D. (n ⫽ 3– 6 mice per group). ASMase genotype Food intake (g) Water intake (ml) Feces (g) Urine (ml)

Standard (chow) diet

High fat (modified) diet

(ⴙ/ⴙ)

(ⴚ/ⴚ)

(ⴙ/ⴙ)

(ⴚ/ⴚ)

2.02 ⫾ 0.50 2.78 ⫾ 1.10 1.68 ⫾ 0.59 0.30 ⫾ 0.05

2.26 ⫾ 0.68 2.35 ⫾ 0.64 1.19 ⫾ 0.42 0.29 ⫾ 0.10

3.08 ⫾ 0.54 2.23 ⫾ 0.39 0.42 ⫾ 0.08 0.87 ⫾ 0.12

3.95 ⫾ 0.67 2.18 ⫾ 0.39 0.51 ⫾ 0.12 1.10 ⫾ 0.25

VLDL

LDL

HDL

***

**

##

FIGURE 1. Effects of diet and genotype on serum cholesterol. asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ldlr⫺/⫺ mice were placed on a standard or modified diet for 10 weeks. A, lipoprotein cholesterol distribution. Serum VLDL, LDL, and high density lipoprotein cholesterol (HDL) were resolved by fast protein liquid chromatography. Data are mean ⫾ S.E. (n ⫽ 3 animals in each group). B and C, total esterified and non-esterified (free) cholesterol in serum. The values of individual mice are shown and represent the mean of triplicate measurements. Statistical significance of the main effects (***, p ⬍ 0.001; **, p ⬍ 0.005) and the interaction effect of genotype and diet (##, p ⬍ 0.01) are shown based on two-way ANOVA.

8360 JOURNAL OF BIOLOGICAL CHEMISTRY

VOLUME 284 • NUMBER 13 • MARCH 27, 2009


TABLE 1 Diet composition

and fed ad libitum for 10 weeks. Two to four animals per cage were housed in micro-isolation in a 12-h light/dark cycle at the University of Kentucky Animal Care Facility according to the criteria outlined in the University of Kentucky Animal Resources and Procedures Handbook. Body weight was monitored twice a week. Randomly chosen mice from each group were housed individually in metabolic cages for 7 days in the second half of the 10-week period to measure food and water intake, feces, and urine. At the end of the diet blood was withdrawn by heart puncture, and various organs were collected, flash-frozen in liquid nitrogen, and stored at ⫺80 °C until further processing. Histological Studies in Fat and Liver Tissues—Adipose tissue samples were fixed in 3.7% paraformaldehyde in phosphatebuffered saline and processed for paraffin embedding, slicing, and hematoxylin and eosin staining. Flash-frozen sections were used for hematoxylin and eosin and Oil Red-O staining of liver. Lipid Analysis—Total lipids were extracted as described previously (20). The lipid extracts were analyzed by TLC on silica gel 60 plates using chloroform:methanol:triethylamine:2-propanol: potassium chloride (0.25%) (30:9:18:25:6, by volume) as the developing solvent. The regions migrating with a standard ceramide (bovine brain) (Avanti Polar Lipids, Alabaster, AL) were scraped off plates, and the lipids were eluted. After the addition of an internal standard, N-hexanoyl-C20-sphinganine (Avanti Polar Lipids, Alabaster, AL), the ceramide mass was quantified by HPLC (21). Free sphingoid bases were quantified by HPLC using C20 sphinganine (Matreya Inc., Pleasant Gap, PA) as an internal standard (21). For quantification of individual phospholipid classes, the regions corresponding to standard SM, PC, PS, and PE were sprayed with 50% sulfuric acid and incubated at 190 –200 °C for 2–2.5 h. Inorganic phosphorus was measured according to Kahovkova and Odavic (22). Radiolabeled triacylglycerols, ceramide, and SM was quantified after lipid extraction in the presence of the respective carrier and separation by TLC using the following mobile phases: chloroform:acetic acid (94:6, by volume) for TAG; diethyl ether: methanol (99:1, by volume) for ceramide; and chloroform:

Acid Sphingomyelinase and Diet-induced Hepatic Steatosis

FIGURE 2. Analyses of adipose tissues. asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ldlr⫺/⫺ mice were placed on a standard or modified diet for 10 weeks. A and B, weight of epididymal fat (EF) and retroperitoneal (RPF) tissues. The values of individual animals are shown. The mean ⫾ S.D. is depicted on the side. C, histological changes in epididymal fat tissue. Fixed tissue was stained with hematoxylin and eosin and examined under a microscope. Scale bar corresponds to 200 ␮m. Statistical significance of the main effects is shown (***, p ⬍ 0.001) based on two-way ANOVA.


methanol:acetic acid:water (56:30:4:2, by volume) for SM. The radioactivity was counted on an LS 6500 scintillation counter (Beckman Instruments, Palo Alto, CA) after scraping the respective spots and correcting for quenching by the silica. The concentrations of TAG and total cholesterol were measured using commercially available kits. SPT Activity Assay—SPT activity was measured in microsomes prepared from 30% liver homogenates. After removal by centrifugation of the debris at 500 ⫻ g, and the heavy membrane fraction at 20,000 ⫻ g, microsomal membranes were pelleted at 105,000 ⫻ g for 1 h, resuspended in 100 mM Tris buffer (pH 7.4), and frozen for future use. The SPT activity was assayed using 3H-labeled L-serine and palmitoyl-CoA as exogenous substrates according to Dickson et al. (23). JOURNAL OF BIOLOGICAL CHEMISTRY

8361


FIGURE 3. Analyses of liver tissue. asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ldlr⫺/⫺ mice placed on a standard or modified diet for 10 weeks. A, weight of liver. The values of individual animals are shown. B and C, histological changes in liver. Flash-frozen liver tissue stained with hematoxylin and eosin (H&E) (B) or Oil Red-O (C). Scale bar corresponds to 100 ␮m. D, concentration of TAG. TAG levels were measured in total lipid extract of liver. Data are means ⫾ S.D. (n ⫽ 3 animals per group). Statistical significance of the main effects (***, p ⬍ 0.001) and the interaction effect of genotype and diet (##, p ⬍ 0.002; ###, p ⬍ 0.001) are shown based on two-way ANOVA.

Acid Sphingomyelinase and Diet-induced Hepatic Steatosis TABLE 3 Effect on body weight Changes in body weight at the beginning, middle, and end of the diet are shown. The “fold increase” is calculated for each animal before statistical analyses. Data are means ⫾ S.D. (n ⫽ 5–11 animals per group). Genotype/diet

a b

0 days

35 days

70 days

Body weight

Increase

Body weight

Increase

Body weight

Increase

g

-fold

g

-fold

g

-fold

Males asm⫹/⫹/ldlr⫺/⫺ standard asm⫺/⫺/ldlr⫺/⫺ standard asm⫹/⫹/ldlr⫺/⫺ high fat asm⫺/⫺/ldlr⫺/⫺ high fat

30.8 ⫾ 1.3 33.8 ⫾ 3.3 29.2 ⫾ 2.1 30.6 ⫾ 1.8

1.00 1.00 1.00 1.00

35.7 ⫾ 1.8 38.1 ⫾ 1.9 39.0 ⫾ 3.1 36.9 ⫾ 2.5

1.17 ⫾ 0.04 1.13 ⫾ 0.03 1.33 ⫾ 0.03a 1.20 ⫾ 0.03

38.1 ⫾ 2.6 38.8 ⫾ 3.4 43.2 ⫾ 2.9b 36.4 ⫾ 3.1

1.25 ⫹ 0.04 1.15 ⫹ 0.02 1.48 ⫾ 0.04a 1.17 ⫾ 0.05

Females asm⫹/⫹/ldlr⫺/⫺ standard asm⫺/⫺/ldlr⫺/⫺ standard asm⫹/⫹/ldlr⫺/⫺ high fat asm⫺/⫺/ldlr⫺/⫺ high fat

21.7 ⫾ 2.4 21.2 ⫾ 3.1 20.9 ⫾ 2.1 22.6 ⫾ 1.8

1.00 1.00 1.00 1.00

25.1 ⫾ 4.5 24.4 ⫾ 2.0 27.1 ⫾ 2.0 25.9 ⫾ 2.0

1.11 ⫾ 0.10 1.15 ⫾ 0.11 1.25 ⫾ 0.18 1.16 ⫾ 0.06

24.7 ⫾ 3.0 25.6 ⫾ 0.7 31.2 ⫾ 5.1b 26.5 ⫾ 3.6

1.15 ⫾ 0.04 1.21 ⫾ 0.15 1.49 ⫾ 0.19a 1.21 ⫾ 0.06

Statistical significance of the interaction effect is p ⬍ 0.001 based on two-way ANOVA. Statistical significance of the interaction effect is p ⬍ 0.05 based on two-way ANOVA.

Downloaded from www.jbc.org at University of Kentucky, on January 12, 2010 FIGURE 4. Effects on phospholipid and cholesterol concentration in liver. asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ldlr⫺/⫺ mice were placed on a standard or modified diet for 10 weeks. A–D, concentrations of SM (A), PS (B), PE (C), and PC (D). Levels of individual phospholipids were measured in total lipid extract of liver after separation on TLC. Data are means ⫾ S.D. (n ⫽ 3 animals per group). E, concentrations of esterified and free cholesterol. Data for each individual mouse are shown. Statistical significance of the main effects (***, p ⬍ 0.001; **, p ⬍ 0.01) and the interaction effect of genotype and diet (#, p ⬍ 0.05; ##, p ⬍ 0.01; ###, p ⬍ 0.001) are shown based on two-way ANOVA.

Cell Culture and Labeling Studies—HepG2 cells were maintained in minimum Eagles medium (Invitrogen) supplemented with 10% fetal bovine serum. Cells were radiolabeled with [3H]serine (10 mCi/mmol; Amersham Biosciences) for 24 h or with [3H]palmitic acid (50 mCi/mmol; American Radiochemical, St Louis, MO) for 3, 5, and 8 h. Palmitic acid was delivered to the cells as a complex with bovine serum albumin (2:1, by mol) at a low (0.1 mM) or high (1 mM) concentration. Desipra-


mine (50 ␮M) (Sigma), fumonisin B1 (25 ␮M) (Sigma), and myriocin (5 ␮M; Biomol International, Plymouth Meeting, PA) were added 1 h before the addition of the palmitic acid. Glucose and Insulin Tolerance Tests—Blood glucose was measured 1 week prior to the end of the diet. Mice were fasted overnight (14 –16 h); blood was collected from the tail, and glucose was measured using Accu-Chek active glucometer (Roche Diagnostics). The concentrations of glucose in the fed VOLUME 284 • NUMBER 13 • MARCH 27, 2009

Acid Sphingomyelinase and Diet-induced Hepatic Steatosis diet or genotype alone was confirmed by one-way comparisons as indicated in the legends. Throughout the figures, an asterisk symbolizes the significance of the main effect, and a number sign symbolizes the significance of the interaction effect.



8363


RESULTS Generation of Mice Deficient in ASMase (asm⫺/⫺) in LDL Receptor Null (ldlr⫺/⫺) Background—ldlr⫺/⫺ mice are responsive to diets rich in saturated fats and develop moderate to severe obesity and metabolic syndrome when fed a diet rich in saturated fats (24). To test the role of *** *** *** ASMase in hepatic fat accumulation, ASMase deficiency was generated in ldlr⫺/⫺ mice by cross-breeding asm⫺/⫹ and ldlr⫺/⫺ animals. asm⫺/⫹/ldlr⫺/⫺ offsprings were used as breeding pairs to maintain a new compound-deficient mouse ⫹/⫹ ⫺/⫺ ⫺/⫺ ⫺/⫺ FIGURE 5. Changes in de novo synthesis of sphingolipids in liver. asm /ldlr and asm /ldlr mice colony. The experiments described were placed on standard or modified diet for 10 weeks. A, SPT-specific activity. Activity of SPT was measured in in this study were done with litter-, isolated hepatic microsomes using [3H]serine as a substrate. Data are means ⫾ S.D. (n ⫽ 3). B, SPT1 protein. The age-, and gender-matched asm⫺/⫺/ abundance of the SPT1 subunit of SPT was determined by Western blotting using protein extracts from iso⫺/⫺ and asm⫹/⫹/ldlr⫺/⫺ anilated microsomal preparations. The abundance of the microsomal protein Bip is shown as loading control. ldlr C, quantification of the changes in SPT1. Intensity of the bands corresponding to SPT1 was normalized for the mals (8 –16 mice per group). levels of Bip. Data are average of two Western blots. D–F, hepatic concentrations of the main intermediates in ASMase Deficiency Does Not the de novo pathway of ceramide synthesis. The concentration of sphinganine, dihydroceramide, and ceramide in lipid extracts from liver homogenates was determined by HPLC and TLC/HPLC. Values are means ⫾ S.D. Affect Food Intake or High Fat Diet(n ⫽ 3). Statistical significance of the main effects (***, p ⬍ 0.001; **, p ⬍ 0.01, *, p ⬍ 0.05), and the interaction induced Increases in VLDL and LDL effect of genotype and diet (#, p ⬍ 0.05) are shown based on two-way ANOVA. Cholesterol—Male or female asm⫺/⫺/ ldlr⫺/⫺ and asm⫹/⫹/ldlr⫺/⫺ animals state were measured 24 h later after returning the animals to their were placed either on a diet enriched with saturated fats and respective diet. Glucose tolerance and insulin tolerance tests were cholesterol or a standard calorie-adjusted diet for 10 weeks. performed at the end of the diet. After overnight fasting (14 –16 h) Palmitic acid was the major saturated fat in the modified diet for glucose tolerance and 6 h for the insulin tolerance test, animals (Table 1). ASMase deficiency had no effect on food and water were injected intraperitoneally with D-glucose (1.5 mg/g body consumption, as well as urine volume and feces amount (Table weight) (Sigma) or bovine insulin (0.75 units/kg body weight) 2). Some diet-related differences were observed, and mice on a (Sigma), respectively. Blood glucose levels were determined prior modified diet ate significantly more and had a substantial increase in urine volume and a decrease in feces amount as to and 20, 60, 90, and 120 min after the injection. In vitro 2-deoxy-D-[1,2-3H]glucose uptake was measured in compared with those on normal diet. Furthermore, VLDL and excised paired extensor digitorum longus muscle. Briefly, mus- LDL cholesterol (Fig. 1A) and total serum cholesterol (Fig. 1, B cles were incubated at 37 °C in an oxygenated bath (95% O2 and and C) were similarly elevated by the modified diet in both 5% CO2) of Krebs/bicarbonate buffer (117 mM NaCl, 4.7 mM genotypes. asm⫺/⫺/ldlr⫺/⫺ Mice Do Not Accumulate Fat Either in the KCl, 2.5 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, and 24.6 mM NaHCO3 (pH 7.5)) containing 2 mM pyruvate. One muscle Adipose Tissue or in the Liver—When placed on a modified diet, of each pair was stimulated with bovine insulin (100 nmol/liter), asm⫹/⫹/ldlr⫺/⫺ mice accumulated a substantial amount of and the other was not. After 30 min, 1 mM [3H]glucose, 7 mM body fat. The weight of epididymal fat and retroperitoneal fat [14C]mannitol (PerkinElmer Life Sciences) was added to both pads, which represent subcutaneous and visceral fat respecmuscles for additional 10 min. The muscles were washed exten- tively, increased by 1.5–2-fold (Fig. 2, A and B). In contrast, sively, digested in 1 N NaOH, and heated for 10 min at 80 °C. The weight of adipose tissue of asm⫺/⫺/ldlr⫺/⫺ animals on a standard diet was significantly lower than that of asm⫹/⫹/ldlr⫺/⫺ glucose uptake was then quantified by scintillation counting. Statistical Analysis—Two-way ANOVA with subsequent Bon- mice and did not increase as a result of the modified diet (Fig. 2, ferroni test was used to determine the significance of changes in A and B). Histological analyses confirmed the lack of adipocyte multiple comparisons. When the interaction effect of genotype fat accumulation in asm⫺/⫺/ldlr⫺/⫺ mice (Fig. 2C). It should be and diet was statistically highly significant, the significance of the noted that these data are for male mice only and the differences



FIGURE 6. Effect of palmitic acid and ASMase activity on lipid metabolism in HepG2 cells. HepG2 cells were treated with ASMase inhibitor desipramine (50 ␮M) or with palmitic acid at low (0.1 mM) or high (1.0 mM) concentrations for 24 h. A, concentration of TAG. TAG content in total lipid extract from the cells was determined using a commercially available kit. Data are means ⫾ S.D. (n ⫽ 3). B and C, incorporation of [3H]palmitate into TAG (B) and ceramide (C). HepG2 cells were labeled with [3H]palmitate (50 mCi/mmol, for the indicated times). Levels of [3H]TAG and [3H]ceramide were quantified by scintillation counting after TLC separation. Data represent means ⫾ S.D. (n ⫽ 4). Statistical significance of the main effects (***, p ⬍ 0.001; **, p ⬍ 0.01, *, p ⬍ 0.05) and the interaction effect of palmitate and desipramine treatments (##, p ⬍ 0.01; #, p ⬍ 0.05) are shown based on two-way ANOVA.

diet led to a substantial increase in the in vitro measured SPT activity in both asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ldlr⫺/⫺ mice (Fig. 5A). This correlated with increases in the abundance of SPT1 protein (Fig. 5, B and C). The lack of ASMase did not affect these changes. The liver levels of key intermediates in the de novo synthesis of SM, namely sphinganine, dihydroceramide, and ceramide, increased in both asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ ldlr⫺/⫺ mice fed the modified diet (Fig. 5, D–F), consistent with the increase in SPT activities. However, the accumulation of these intermediates was much more substantial in asm⫺/⫺/ ldlr⫺/⫺ animals as compared with asm⫹/⫹/ldlr⫺/⫺ mice. The diet-induced increase in sphinganine concentration was around 4 pmol/mg䡠protein in asm⫹/⫹/ldlr⫺/⫺ mice and around 100 pmol/mg䡠protein in asm⫺/⫺/ldlr⫺/⫺ animals. For dihydroceramide, these increases were 70 and 340 pmol/mg䡠protein, and for ceramide increases were 0.8 and 5.8 nmol/mg䡠protein, respectively. In contrast, the levels of free sphingosine, which is generated only during ceramide hydrolysis but not synthesis, were similar in asm⫺/⫺/ldlr⫺/⫺ animals on standard and modified diet (228 ⫾ 15.4 pmol/mg䡠protein and 227.5 ⫾ 21.3 pmol/ mg䡠protein, respectively). The above observations thus agree with previous reports that a diet enriched in saturated fats has a potent stimulatory effect on the rate of sphingolipid synthesis, VOLUME 284 • NUMBER 13 • MARCH 27, 2009


observed in females were even greater; however, the data from females were not included in the analyses because the estrus cycle of each individual mouse was not taken into account. Genotype and diet both significantly affected liver size (not shown) and weight (Fig. 3A). Consistent with the hepatomegaly observed in Niemann-Pick patients, ASMase deficiency alone led to hepatic enlargement. In turn, asm⫹/⫹/ldlr⫺/⫺ mice on a modified diet also had enlarged liver as compared with those on a standard diet, indicative of fatty liver development. However, the increases in liver weight observed in asm⫺/⫺/ldlr⫺/⫺ animals on a modified diet were much more substantial (Fig. 3A). Two-way ANOVA analyses showed a diet and genotype interaction of p ⬍ 0.001. The effects of modified diet on the liver were further studied by histological analyses using hematoxylin and eosin (Fig. 3B) and Oil red O stains (Fig. 3C), as well as by direct measurements of TAG concentration (Fig. 3D). All three approaches confirmed the accumulation of TAG in asm⫹/⫹/ldlr⫺/⫺ but not in asm⫺/⫺/ldlr⫺/⫺ mice on a modified diet, thus eliminating the possibility of lipodystrophy. However, the livers of asm⫺/⫺/ ldlr⫺/⫺ animals exhibited large pale-blue/pink areas similar to ones described in patients with ASMase deficiency and attributed to the accumulation of SM and some glycerophospholipids. Notably, the number and the size of these areas were significantly increased by the modified diet (Fig. 3B). The body weight for each group was similar at the beginning of the diet (Table 3). Mice on the modified diet gain more weight than those on standard diet. Notably, during the first 5 weeks there were no significant genotype-related differences; after that, however, the mice on a modified diet continue to gain weight whereas the weight of the asm⫺/⫺/ldlr⫺/⫺ mice leveled off. Accumulation of Phospholipids and Sphingolipids in the Livers of asm⫺/⫺/ldlr⫺/⫺ Mice on High Fat Diet—In contrast to changes in TAG levels, liver content of the main phospholipids, phosphatidylserine (PS), phosphatidylethanolamine (PE), and phosphatidylcholine (PC), was significantly elevated in asm⫺/⫺/ldlr⫺/⫺ but not in asm⫹/⫹/ldlr⫺/⫺ mice on a modified diet (Fig. 4, B–D). The SM content elevated in both genotypes as a result of the diet; however, the increase was severalfold higher in asm⫺/⫺/ldlr⫺/⫺ mice than in asm⫹/⫹/ldlr⫺/⫺ mice (Fig. 4A). Consequently, the SM level in these mice surpassed that of PC, the most abundant phospholipid in normal cells. The hepatic concentration of both free (unesterified) and esterified cholesterol increased as a result of the high fat diet in all mice (Fig. 4E), apparently reflecting the elevated dietary cholesterol content. ASMase-deficient mice appeared to have elevated free cholesterol as compared with the respective wild type controls, which was likely because of the inhibitory effect that SM accumulation had on cholesterol esterification (25). Stimulation of de Novo Sphingolipid Synthesis by the Modified Diet—Next we sought to determine why ASMase deficiency leads to accumulation of SM rather than TAG in livers of mice on a modified diet. In asm⫹/⫹/ldlr⫺/⫺ mice, modified diet had no effect on the activity of ASMase, and thus the direct role of this enzyme seemed unlikely (data not shown). SPT is the enzyme catalyzing the initial rate-limiting step of the de novo synthesis of all complex sphingolipids, including SM. Modified


FIGURE 7. Role of the de novo pathway for ceramide synthesis. HepG2 cells were treated with SPT inhibitor myriocin (5 ␮M) and ASMase inhibitor desipramine (50 ␮M). Cells were labeled with [3H]palmitate (50 mCi/mmol, for the


indicated times) or with palmitic acid (PA) at low (0.1 mM) or high (1.0 mM) concentrations for 24 h. A–C, incorporation of [3H]palmitate into ceramide (A), SM (B), and TAG (C). Levels of the radiolabeled lipids were quantified by scintillation counting after TLC separation. Data represent means ⫾ S.D. (n ⫽ 3). The subsets of data from cells treated or not with myriocin were analyzed separately by two-way ANOVA. Statistical significance of the main effects (**, p ⬍ 0.01; *, p ⬍ 0.05) and the interaction effect of palmitate and desipramine treatments (#, p ⬍ 0.05) are shown. The statistical significance of the effects of myriocin alone was calculated based on Student’s t test by comparing corresponding groups (*, p ⬍ 0.05; **, p ⬍ 0.01).


8365


but the observations also suggest that the magnitude of the effect depends upon hepatic ASMase activity. ASMase Activity Regulates the Utilization of Exogenous Palmitic Acid for TAG and de Novo Sphingolipid Synthesis—The link between sphingolipid and TAG synthesis was investigated in HepG2 cells supplemented with palmitic acid at low (0.1 mM) or high (1.0 mM) concentrations. Palmitic acid is a common precursor for both TAG and de novo sphingolipid synthesis. It is also the main saturated fat in the diets commonly consumed in the Western world and has been implicated in the development of insulin resistance and lipotoxicity (26). In agreement with the in vivo observations reported above, incubation with high concentrations of palmitic acid (Fig. 6A) led to substantial increases in cellular TAG mass; however, these increases were attenuated by desipramine, an inhibitor, which abolished ASMase activity by more than 90% and had no apparent toxicity. The rate of 3H-labeled palmitic acid incorporation into TAG (Fig. 6B), ceramide (Fig. 6C), and, to a lesser extent, SM (data not shown and Fig. 7B) was also higher when the palmitate was present at high concentrations. These results provide direct evidence that increased supply of palmitic acid stimulates both the synthesis of TAG and sphingolipids. However, inhibition of hepatic ASMase activity reduced the incorporation of radiolabeled palmitate into TAG by 40% (Fig. 6B), while further increasing that in ceramide (Fig. 6, C and D) and SM (data not shown and see Fig. 7B). Desipramine-induced increases were also observed in palmitate incorporation into some glycerophospholipids, namely phosphatidylserine and phosphatidylethanolamine (data not shown). The effects of desipramine on the de novo synthesis of sphingolipids were further confirmed when [3H]serine was used as a precursor instead of palmitate (data not shown). Treatment with myriocin (Fig. 7) and fumonisin B1 (data not shown), which are specific inhibitors of SPT and dihydroceramide synthase, respectively, attenuated the desipramine effects on ceramide (Fig. 7A) and SM (Fig. 7B) labeling. These observations further supported the role of ASMase as a negative regulator of the rate of sphingolipid synthesis. The most interesting finding, however, was that the myriocin treatment also had an effect on TAG synthesis, and it blocked the desipramine-induced decrease in palmitate incorporation into TAG (Fig. 7C). Evidently, not only the rates of sphingolipid and TAG synthesis are inversely correlated, but also the decreased flux of palmitate toward TAG synthesis is caused by its increased utilization for sphingolipid synthesis. Glucose Regulation and Insulin Sensitivity in asm⫺/⫺/ ldlr⫺/⫺ Mice on High Fat Diet—Importantly, the lack of TAG accumulation in asm⫺/⫺/ldlr⫺/⫺ mice on a modified diet is paralleled by normal blood glucose regulation (Fig. 8). As can be seen in Fig. 8A, the blood glucose levels of both asm⫹/⫹/ldlr⫺/⫺





type 2 diabetes (3, 4, 27). Mice lacking the LDL receptor respond to such diets similar to humans; they develop metabolic syndrome (24), hypercholesterolemia, and atherosclerosis (28). Here we find that when placed on an experimental diet enriched in saturated fats, ldlr⫺/⫺ mice lacking ASMase exhibit normal regulation of blood glucose and normal insulin response, despite the same daily food and water uptake as wild type littermates. In contrast, the diet-induced hypercholesterolemia is not affected by the lack of ASMase. The amelioration of hyperglycemia and insulin resistance is accompanied by severe inhibition of dietinduced TAG accumulation in liver and adipose tissue. Liver plays a central role in the metabolism and storage of dietary fat and through the VLDL pathway is a major source of fat for other tissues. TAG synthesis is catalyzed by diacylglycerol (DAG) FIGURE 8. Regulation of blood glucose concentrations. asm⫹/⫹/ldlr⫺/⫺ and asm⫺/⫺/ldlr⫺/⫺ mice were fed a acyltransferase (DGAT), which standard or high fat diet (n ⫽ 6 –11 animals per group) for 10 weeks. A, blood glucose levels in fasted and fed state. Blood glucose concentrations were measured after 14 h of fasting (open stars) and 24 h later after adds a fatty acid from acyl-CoA to re-feeding with the respective diet ad libitum (solid stars). Changes for individual mice are shown. B, glucose DAG to produce TAG. There are tolerance test. Levels of blood glucose after administration of D-glucose (1.5 g/kg body weight, intraperitone- two genetically distinct forms of ally) to fasted animals are shown. Data are means ⫾ S.D. (n ⫽ 5–7 animals per group). Statistical significance of the main effect of the diet is shown (*, p ⬍ 0.05; **, p ⬍ 0.01) based on Bonferroni post test analyses. C, in vitro DGAT, DGAT1 and DGAT2, which glucose uptake in muscle. Excised paired extensor digitorum longus muscles were incubated with [3H]glucose/ play different roles in TAG synthe[14C]mannitol in the presence and absence of insulin. Data are means ⫾ S.D. (n ⫽ 3 animals per group). Statistical significance of the effect of insulin is shown for each muscle pair (*, p ⬍ 0.05; **, p ⬍ 0.01) according sis, despite similar biochemical to Student’s t test. D, in vivo insulin tolerance test. Levels of blood glucose after administration of insulin (0.75 characteristics and substrate speciunits/kg body weight, intraperitoneally) are shown. Data are means ⫾ S.D. (n ⫽ 4 animals per group). Statistical ficity. Increasing evidence had sugsignificance of the main effect of the diet (*, p ⬍ 0.05; **, p ⬍ 0.01) is shown, based on Bonferroni post test gested that the two enzymes are part analyses. of a mechanism that coordinates the regulation of TAG and glycerophosand asm⫺/⫺/ldlr⫺/⫺ on a standard diet are higher in fed than in pholipid synthesis in an apparently competitive manner. Delethe fasted state. Consistent with the development of hypergly- tion of DGAT1 has only a minor effect on TAG synthesis. Its cemia, asm⫹/⫹/ldlr⫺/⫺ mice on a modified diet exhibit similar overexpression is sufficient to elevate synthesis of TAG howglucose concentrations in the fasted and fed state (Fig. 8A). In ever; surprisingly, it also reduces that of glycerophospholipids, contrast, asm⫺/⫺/ldlr⫺/⫺ animals showed striking differences suggesting that the overexpressed enzyme uses a pool of DAG in blood glucose concentrations similar to those observed in initially designed for phospholipid synthesis (29). In contrast, animals on a standard diet. The ability of asm⫺/⫺/ldlr⫺/⫺ mice overexpression of DGAT2 has a much greater effect on TAG on a modified diet to clear glucose faster than asm⫹/⫹/ldlr⫺/⫺ accumulation than DGAT1, and its silencing inhibits hepatic mice is also evident by the enhanced clearance of intraperito- TAG synthesis (30). Notably, suppression of DGAT2, but not neally administered D-glucose (Fig. 8B). Finally, insulin toler- DGAT1, with antisense oligonucleotides reverses diet-induced ance tests administered in vitro in isolated muscle (Fig. 8C) and hepatic steatosis and insulin resistance (31). Thus, the two in vivo (Fig. 8D) provide direct evidence that asm⫺/⫺/ldlr⫺/⫺ DGAT forms seems to use two metabolically distinct and comanimals on a high fat diet are insulin-responsive and confirm petitive pools of DAG that are specifically designated for TAG further that the ASMase-deficient animals are protected or phospholipid syntheses (29). The experiments shown here imply that similar metabolic against diet-induced hyperglycemia and insulin resistance. interactions exist between sphingolipid and TAG synthesis. DISCUSSION It is noteworthy that the PC:ceramide phosphorylcholine Increased consumption of food rich in saturated fats and transferase reaction, which catalyzes SM synthesis, genercholesterol is a common trend in Western society. It leads to ates one molecule of DAG for each molecule of SM that is obesity that increases the risk of cardiovascular diseases and synthesized (Fig. 9). It is possible that extensive stimulation


of SM synthesis, as observed in ASMase null mice on saturated fat-enriched diet, might disrupt the DAG pools available for TAG synthesis. This is further supported by the spatial separation of the synthesis of SM that occurs in cisGolgi (32) and by the parallel increase in the levels of glycerophospholipids observed in asm⫺/⫺/ldlr⫺/⫺ mice fed a high fat diet and in hepatocytes with inhibited ASMase activity. Several studies have shown that increased supply of palmitic acid elevates de novo synthesis of sphingolipids (12), including SM, in isolated rat hepatocytes and C2C12 myotubes (5). This study finds similar elevation in de novo synthesis of sphingolipids in animals fed a palmitic acid-rich diet, which parallels the increases in TAG synthesis. Together, these studies provide clear evidence that dietary palmitate partitions between TAG and sphingolipid pools (Fig. 9). However, our study also finds that the palmitate flux through ceramide and TAG synthetic pathways is regulated in a competitive fashion. Palmitate-induced ceramide accumulation in muscle has been shown to inhibit insulin-induced translocation of Glut-4 MARCH 27, 2009 • VOLUME 284 • NUMBER 13

to the plasma membrane and seems to be required for the onset of hyperglycemia (33). At this point, it is not known whether ASMase deficiency had an effect on ceramide synthesis in muscle, comparable with that observed in liver. It seems, however, that diet-induced accumulation of ceramide may not be sufficient to offset glucose regulation because, at least in liver, it correlated with the onset of hyperglycemia only when accompanied by an accumulation of TAG. This might reflect the complexity of hepatic insulin response, which regulates the rate of glycogenolysis and gluconeogenesis, instead of the active glucose uptake. It seems that the capacity of liver to handle excess dietary fat and to accumulate TAG has a prominent role in the regulation of overall glucose homeostasis. Cellular ASMase activity emerges as a critical factor that determines the pathway for utilization of exogenous palmitate because of the following. (i) ASMase deficiency in mice augments dietary-induced increases in dihydroceramide and sphinganine, which are intermediates only in the de novo pathway; concomitantly, it abolishes TAG accumulation. (ii) Similarly, inhibition of ASMase activity in HepG2 cells stimulates de novo JOURNAL OF BIOLOGICAL CHEMISTRY

8367


FIGURE 9. Schematic representation of metabolic links between sphingolipids, glycerolipids, and TAG in liver. Pathways for the de novo SM synthesis from palmitate (solid green background), turnover of SM to ceramide and consequently to sphingosine and sphingosine-phosphate (striped green background), as well as synthesis of phospholipids and TAG from common precursor, phosphatidic acid (pink background), are shown. The experiments in this study suggest that in liver palmitic acid may partition between sphingolipid and TAG pools in a competitive manner that is influenced by the rate of lysosomal turnover of SM.


Acknowledgments—We sincerely thank Alexander Karakashian for thoughtful reading and correcting our manuscript; Dr. Richard Charnigo for advice regarding statistical analyses of the data; the Cardiovascular Research Discussion Group (University of Kentucky) for valuable comments; and Dr. Lisa Cassis for helping with isolation of fat tissue. REFERENCES 1. den Boer, M., Voshol, P. J., Kuipers, F., Havekes, L. M., and Romijn, J. A. (2004) Arterioscler. Thromb. Vasc. Biol. 24, 644 – 649 2. Coleman, R. A., and Lee, D. P. (2004) Prog. Lipid Res. 43, 134 –176 3. Unger, R. H. (2003) Trends Endocrinol Metab. 14, 398 – 403 4. McGarry, J. D. (2002) Diabetes 51, 7–18 5. Summers, S. A. (2006) Prog. Lipid Res. 45, 42–72 6. Pickersgill, L., Litherland, G. J., Greenberg, A. S., Walker, M., and Yeaman, S. J. (2007) J. Biol. Chem. 282, 12583–12589 7. Turpin, S. M., Lancaster, G. I., Darby, I., Febbraio, M. A., and Watt, M. J. (2006) Am. J. Physiol. 291, E1341–E1350 8. Schmitz-Peiffer, C., Craig, D. L., and Biden, T. J. (1999) J. Biol. Chem. 274, 24202–24210 9. Herschkovitz, A., Liu, Y. F., Ilan, E., Ronen, D., Boura-Halfon, S., and Zick, Y. (2007) J. Biol. Chem. 282, 18018 –18027 10. Ghosh, N., Patel, N., Jiang, K., Watson, J. E., Cheng, J., Chalfant, C. E., and Cooper, D. R. (2007) Endocrinology 148, 1359 –1366 11. JeBailey, L., Wanono, O., Niu, W., Roessler, J., Rudich, A., and Klip, A. (2007) Diabetes 56, 394 – 403 12. Merrill, A. H., Jr., Lingrell, S., Wang, E., Nikolova-Karakashian, M., Vales,


T. R., and Vance, D. E. (1995) J. Biol. Chem. 270, 13834 –13841 13. Chavez, J. A., and Summers, S. A. (2003) Arch. Biochem. Biophys. 419, 101–109 14. Powell, D. J., Turban, S., Gray, A., Hajduch, E., and Hundal, H. S. (2004) Biochem. J. 382, 619 – 629 15. Zhou, Y. T., Grayburn, P., Karim, A., Shimabukuro, M., Higa, M., Baetens, D., Orci, L., and Unger, R. H. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 1784 –1789 16. Samad, F., Hester, K. D., Yang, G., Hannun, Y. A., and Bielawski, J. (2006) Diabetes 55, 2579 –2587 17. Marathe, S., Kuriakose, G., Williams, K. J., and Tabas, I. (1999) Arterioscler. Thromb. Vasc. Biol. 19, 2648 –2658 18. Crocker, A. C., and Farber, S. (1958) Medicine (Baltimore) 37, 1–95 19. Horinouchi, K., Erlich, S., Perl, D. P., Ferlinz, K., Bisgaier, C. L., Sandhoff, K., Desnick, R. J., Stewart, C. L., and Schuchman, E. H. (1995) Nat. Genet. 10, 288 –293 20. Williams, R. D., Wang, E., and Merrill, A. H., Jr. (1984) Arch. Biochem. Biophys. 228, 282–291 21. Merrill, A. H., Jr., Wang, E., Mullins, R. E., Jamison, W. C., Nimkar, S., and Liotta, D. C. (1988) Anal. Biochem. 171, 373–381 22. Kahovcova, J., and Odavic, R. (1969) J. Chromatogr. 40, 90 –96 23. Dickson, R. C., Lester, R. L., and Nagiec, M. M. (2000) Methods Enzymol. 311, 3–9 24. MacDonald, M. L., Singaraja, R. R., Bissada, N., Ruddle, P., Watts, R., Karasinska, J. M., Gibson, W. T., Fievet, C., Vance, J. E., Staels, B., and Hayden, M. R. (2008) J. Lipid Res. 49, 217–229 25. Leventhal, A. R., Chen, W., Tall, A. R., and Tabas, I. (2001) J. Biol. Chem. 276, 44976 – 44983 26. Reynoso, R., Salgado, L. M., and Calderon, V. (2003) Mol. Cell. Biochem. 246, 155–162 27. Weinberg, J. M. (2006) Kidney Int. 70, 1560 –1566 28. Wu, L., Vikramadithyan, R., Yu, S., Pau, C., Hu, Y., Goldberg, I. J., and Dansky, H. M. (2006) J. Lipid Res. 47, 2215–2222 29. Bagnato, C., and Igal, R. A. (2003) J. Biol. Chem. 278, 52203–52211 30. Stone, S. J., Myers, H. M., Watkins, S. M., Brown, B. E., Feingold, K. R., Elias, P. M., and Farese, R. V., Jr. (2004) J. Biol. Chem. 279, 11767–11776 31. Choi, C. S., Savage, D. B., Kulkarni, A., Yu, X. X., Liu, Z. X., Morino, K., Kim, S., Distefano, A., Samuel, V. T., Neschen, S., Zhang, D., Wang, A., Zhang, X. M., Kahn, M., Cline, G. W., Pandey, S. K., Geisler, J. G., Bhanot, S., Monia, B. P., and Shulman, G. I. (2007) J. Biol. Chem. 282, 22678 –22688 32. Futerman, A. H., Stieger, B., Hubbard, A. L., and Pagano, R. E. (1990) J. Biol. Chem. 265, 8650 – 8657 33. Holland, W. L., Brozinick, J. T., Wang, L. P., Hawkins, E. D., Sargent, K. M., Liu, Y., Narra, K., Hoehn, K. L., Knotts, T. A., Siesky, A., Nelson, D. H., Karathanasis, S. K., Fontenot, G. K., Birnbaum, M. J., and Summers, S. A. (2007) Cell Metab. 5, 167–179 34. Laviad, E. L., Albee, L., Pankova-Kholmyansky, I., Epstein, S., Park, H., Merrill, A. H., Jr., and Futerman, A. H. (2008) J. Biol. Chem. 283, 5677–5684 35. Koval, M., and Pagano, R. E. (1990) J. Cell Biol. 111, 429 – 442 36. Worgall, T. S., Johnson, R. A., Seo, T., Gierens, H., and Deckelbaum, R. J. (2002) J. Biol. Chem. 277, 3878 –3885



sphingolipid synthesis while inhibiting palmitate-induced TAG synthesis and accumulation. (iii) Pharmacological inhibitors of the de novo sphingolipid synthesis ameliorate the effects of ASMase deficiency on both sphingolipid and TAG synthesis. The mechanism by which ASMase deficiency promotes increased de novo synthesis of sphingolipids is unclear. One possibility is that ASMase is part of a negative feedback mechanism for regulation of sphingolipid synthesis depending on the exogenous supply of SM. Indeed, a recent study has suggested that sphingosine 1-phosphate, a metabolite of the terminal turnover of sphingolipids might influence the rate of the de novo synthesis (34). Alternatively, ASMase deficiency has been implicated in deregulation of lipid trafficking (35) and transcriptional regulation of lipid metabolism, including the activity of sterol regulatory element-binding protein (36) that could have profound effects on coordinated regulation of sphingolipid and TAG synthesis. Nevertheless, these observations provide a novel insight into the mechanisms of diet-related hepatic steatosis and its role in insulin resistance, opening new possibilities for treatment of obesity-related pathologies.