Atherosclerosis and Lipoproteins |
From the Department of Pathology (T.S.W.), Institute of Human Nutrition (T.S.W., R.A.J., T.S., R.J.D.), and Department of Pediatrics (R.J.D.), Columbia University, New York, NY.
Correspondence to Dr Tilla S. Worgall, Columbia University, 630 W 168th Street, PH 1512, New York, NY 10032. E-mail tpw7{at}columbia.edu
| Abstract |
|---|
|
|
|---|
Methods and Results Experiments were performed in Chinese hamster ovary cells. Inhibition of ceramide synthesis with myriocin, cycloserine, or fumonisin decreases levels of transcriptionally active SREBP and reduces SRE-mediated gene transcription. When ceramide synthesis is increased through exogenous sphingosine or inhibition of sphingosine kinase, SRE-mediated gene transcription is increased. The important role of ceramide synthesis in SRE-mediated gene transcription is confirmed in LY-B cells that do not synthesize ceramide de novo. LY-B cells fail to increase SRE-mediated gene transcription in sterol depletion.
Conclusions Ceramide synthesis correlates with the generation of transcriptionally active SREBP and SRE-mediated gene transcription. Inhibition of ceramide synthesis decreases levels of transcriptionally active SREBP and SRE-mediated gene transcription. It is hypothesized that the process of ongoing ceramide synthesis contributes to the physiological processing of SREBP, perhaps affecting ER-to-Golgi trafficking. Taken together, modification of ceramide synthesis could be a novel target for drug development in the pharmacologic modification of SRE-dependent pathways.
Key Words: SREBP ceramide myriocin LY-Bsphingosine
| Introduction |
|---|
|
|
|---|
Cholesterol and unsaturated fatty acids are known regulators of transcriptional and posttranscriptional processing of SREBP. We recently reported that unsaturated fatty acids-mediated decreases in SRE-mediated gene transcription are linked to cellular sphingolipid metabolism.4 Ceramide, a metabolite of sphingomyelin hydrolysis, also regulates mature SREBP levels. Regulation occurs even in the presence of inhibitors of intracellular cholesterol trafficking, suggesting a cholesterol-independent effect.4
Ceramide has multiple roles ranging from lipid second messenger to the induction of apoptosis, cell growth, and differentiation.5,6 Cellular ceramide levels are generated either by de novo synthesis from serine and palmitoyl-CoA or through a recycling pathway of sphingolipid hydrolysis. Ceramide also has a role in intracellular protein trafficking, can inhibit coated vesicle formation and exocytosis in Chinese hamster ovary (CHO) cells,7 and can modulate endocytosis in mammalian cells.8 In yeast, ongoing ceramide synthesis is critical in the vesicular ER-to-Golgi transport of GPI-anchored proteins.911
We investigated the effect of ceramide synthesis on SREBP levels and SRE-mediated gene transcription. Because increased ceramide levels decrease mature SREBP protein levels and SRE-mediated gene transcription,4 it was anticipated that inhibition of ceramide synthesis should increase mature SREBP levels and SRE-mediated gene transcription. Contrary to this hypothesis, inhibition of ceramide synthesis decreases SRE-mediated gene transcription. Increasing ceramide synthesis correlates with increased mature SREBP levels and SRE-mediated gene transcription. Consequently, the effect of ceramide on its own synthesis was investigated and shown to be inhibitory, providing an explanation that increased ceramide levels and decreased ceramide synthesis inhibit SRE-mediated gene transcription. The role of ceramide synthesis in SRE-mediated gene transcription is supported by experiments in CHO cells that lack ceramide de novo synthesis (LY-B cells) because of a mutation in a subunit of serine-palmitoyl transferase.12,13 LY-B cells fail to increase SRE-mediated gene transcription when they are cholesterol-depleted. Ceramide increases levels for precursor SREBP but decreases levels for mature SREBP, suggesting a block in the generation of mature SREBP. Our data provide evidence that ceramide synthesis is an important regulatory factor in the maturation cascade of SREBP.
| Methods |
|---|
|
|
|---|
Plasmids
The pSyn-SRE luciferase reporter plasmid originates from the hamster HMG-CoA synthase promoter and contains three SRE elements (326 to 225 bp) and has been described before.14,15 The pWLNeo plasmid was obtained from Stratagene (La Jolla, Calif).
Cell Culture and Stable Transfections
Cells were grown in F12-nutrient mixture medium containing 10% fetal bovine serum, 1% glutamine (v/v), 1% penicillin/streptomycin (v/v), and 10% fetal bovine serum (v/v) at 37°C in humidified CO2 (5%). The generation of stable transfectants has been previously described.4
Enzyme Assays
Luciferase activity was analyzed in aliquots of cell extracts using a luminometer (Berthold LB 9501; Wallac, Gaithersburg, Md). Cells were lysed with 0.1% Triton X-100, 50 mmol/L Hepes, 10 mmol/L MgSO4, pH 7.7. Luciferase activity in relative light units was divided by protein content (mg/mL) for each extract.
Protein Determination
Cellular protein was determined by the Biorad method. BSA was used as a standard.
Cell Survival
In all conditions, cell survival was assessed using the 3-(4,5-Dimethylthiazol-2yl)-2 to 5 diphenyltetrazolium bromide (MTT) assay16 and trypan blue exclusion. Cytotoxicity was assessed by measuring LDH release using a colorimetric assay (Roche, Indianapolis, Ind) in which addition of 1% Triton X was used as a positive control. A 3/7 caspase assay17 was used to assess induction of apoptosis and staurosporine was used as a positive control.
Ceramide de Novo Synthesis
Cells growing in the experimental conditions were incubated during the last 1.5 hours with 3H-serine (1 µL/mL) to allow incorporation into ceramide.18,19 Then, cells were washed twice with PBS, 0.2% BSA, twice with phosphate-buffered saline, lysed in 400 µL, 250 mmol/L Tris-Cl, scraped, and transferred to glass tubes. Then, 1 mL of ice-cold methanol, 2 mL of chloroform, and 0.5 mL of 0.1 N HCl was added, vortexed, and spun at 800g for 10 minutes. The organic phase was washed with 3x2 mL 0.001 N HCl. Lipids were dried under N2. Alkaline hydrolysis was performed by incubation in 2 mL of 0.1 N KOH in methanol at 37°C for 1 hour. Lipids were reextracted by adding 2 mL of chloroform and 1.2 mL of balanced salt solution (135 mmol/L NaCl, 4.5 mmol/L KCl, 1.5 mmol/L CaCl2, 0.5 mmol/L MgCl2, 5.6 mmol/L glucose, 10 mmol/L Hepes, pH 7.2)/EDTA 100 mmol/L (1.08 mL/0.12 mL). After vortexing and centrifugation at 800g for 5 minutes, the lower phase was dried under N2.20,21 The extracted lipids were then dissolved in 50 µL chloroform/methanol (1:1), spotted on TLC plates (Merck Silicagel 60; Darmstadt, Germany), and chromatographed with chloroform-methanol 0.22% aqueous CaCl2 (60:35:8 v/v).22 Ceramide and sphingomyelin (dissolved at 1 µg/µL) were run as standards. The TLC plate was cut at the corresponding lipid spots, mixed with scintillation fluid (Ultima Gold; Packard Instrument Company, CT), and analyzed in a scintillation counter (Perkin Elmer Wallac, Gaithersburg, Md). Results were expressed in dpm/mg protein as a percentage of total counts.
Western Blot Analysis
Cells were incubated in control media (1% BSA) or with the respective conditions. After 8 hours, cells were scraped and pelleted at 1000g and resuspended in lysis buffer C (10 mmol/L Tris-Cl, 100 mmol/L NaCl, 1% SDS, pH 7.6) containing protease inhibitor CompleteTM (Roche Pharmaceuticals, Nutley, NJ). Two hours before harvesting, all cells had received 25 µg/mL N-acetyl-leucyl-leucyl-norleucinal (ALLN). An aliquot (30 µg of protein) was electrophoresed on a denaturing 7.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The monoclonal antibodies against SREBP-1 (BD Biosciences, San Jose, Calif) and actin (Sigma, St. Louis, Mo) and the peroxidase-labeled anti-mouse IgG (Amersham NIF 824) were used for Western blot analysis using the ECL method (Amersham, Arlington Heights, Ill).
Northern Bots
Cells were treated for 8 hours; 30 µg of total RNA were separated by 1.2% denaturing agarose/formaldehyde electrophoresis and transferred to Duralon UV-membranes (Stratagene, La Jolla, Calif). The cDNA probe for HMG-CoA synthase was obtained by RT-PCR using described primers15 and labeled by random priming (Stratagene Prime-It® Random priming labeling kit). The blot was hybridized in Quick-Hyb (Stratagene, La Jolla, Calif).
Data Analysis
Statistical significance was calculated by paired t tests. Results are given as mean±SD. All experiments were repeated on different days at least 3 times and each time in triplicate.
| Results |
|---|
|
|
|---|
Increased Exogenous and Endogenous Ceramide Decreases Ceramide Synthesis
Cells were incubated for 8 hours in the presence of C6-ceramide or C8-ceramide (20 µmol/L), DH-C6-ceramide (20 µmol/L), D-MAPP (20 µmol/L), an inhibitor of alkaline ceramidase, or PPMP (20 µmol/L), an inhibitor of glucosylceramide synthesis.23 As a negative control, cells were incubated with NB-DNJ (40 µmol/L), an inhibitor of glucosylceramide synthesis that does not increase ceramide levels.24 For the last 1.5 hours of incubation, 3H-serine was added as a label to determine ceramide synthesis. All conditions, except incubation with NB-DNJ, significantly (P<0.05) decreased ceramide synthesis measured by incorporation of 3H-serine into ceramide (Figure 1).
|
Ceramide Increases Levels of Precursor SREBP and Decreases Levels of Mature SREBP
Next, the effect of ceramide on SREBP levels was investigated by Western blot analysis. Incubation of CHO cells over 4 hours and 8 hours with C6-ceramide (20 µmol/L) increased cellular levels of precursor SREBP compared with controls at 4 hours and even more at 8 hours (Figure 2). At the same time, levels of mature SREBP decreased in the presence of C6 ceramide at 4 hours and at 8 hours compared with controls and compared with precursor SREBP of the same cell extract. To assure equal loading of the gel, the membrane was probed for actin.
|
Inhibition of Ceramide de Novo Synthesis Decreases SRE-Mediated Gene Transcription
The effect of decreased ceramide synthesis on SRE-mediated gene transcription was investigated next (Figure 3). CHO cells, stable transfectants for SRE-regulated reporter gene, were incubated for 8 hours with myriocin (1 µmol/L), a specific inhibitor of serine-palmitoyl transferase,25 cycloserine (500 mmol/L), another inhibitor of serine-palmitoyl transferase,26,27 or fumonisin B1 (10 µmol/L), an inhibitor of ceramide synthase.28 All 3 inhibitors significantly reduced SRE-mediated gene transcription by 8 hours (Figure 3). Cells were also incubated with PPMP (20 µmol/L), a glucosyltransferase inhibitor that increases intracellular ceramide levels and decreases ceramide synthesis as measured by incorporation of 3H-serine (Figure 1). As a negative control, cells were incubated with 40 µmol/L NB-DNJ, an inhibitor of glucosylceramide synthase that does not increase ceramide levels24 (Figure 3). The effects of myriocin were dose-dependent (Figure I, available online at http://atvb.ahajournals.org) and reversible within 8 hours and did not decrease the expression of a ß-gal control reporter gene (data not shown). All inhibitors shown did not affect cell survival at the concentrations used.
|
Increased Ceramide Synthesis Increases SRE-Mediated Gene Transcription
Sphingosine can increase ceramide synthesis.2932 Sphingosine levels can also be increased by DMS, an inhibitor of sphingosine-1-phosphate kinase.33,34 Cells were incubated with sphingosine (1.5 µmol/L) or DMS (1.5 to 5 µmol/L) for up to 6 hours. DMS dose-dependently increases SRE-mediated gene transcription (Figure 4) and levels of mature SREBP (Figure 4 inset). Sphingosine increases SRE-mediated gene transcription up to 2-fold (Figure 4). DMS dose-dependently increases incorporation of 3H-sphingosine label into ceramide by 70% within 5 hours (Figure II, available online at http://atvb.ahajournals.org).
|
LY-B Cells That Do Not Synthesize Ceramide de Novo Fail to Increase SRE-Mediated Gene Transcription in Sterol Depletion
The role of ongoing ceramide synthesis on SRE-mediated gene transcription was investigated in LY-B cells.12 In LY-B cells, a mutation in the LCB1 subunit of serine-palmitoyl transferase results in a complete lack of enzyme activity and inability to de novo synthesize any sphingolipid species. SRE-mediated gene transcription was first suppressed by incubation for 16 hours in the presence of cholesterol (10 µg/mL) and 25-OH cholesterol (1 µg/mL). Then cells were switched for 6 hours to medium containing 1% BSA. Control cells increased SRE-mediated gene transcription but LY-B cells failed to do so (Figure 5). Control experiments were performed to demonstrate that LY-B cells are able to increase SRE-mediated gene transcription once a precursor for ceramide synthesis is supplied. This time medium was switched to 1% BSA containing 5 µmol/L DMS after previous incubation with cholesterol. Within 6 hours, cells significantly increased SRE-mediated gene transcription. This indicates that when the block in ceramide synthesis is bypassed by DMS, SRE-mediated gene transcription increases as expected, reflecting the physiological response to cellular sterol depletion.
|
Inhibition of Ceramide de Novo Synthesis Decreases Levels of HMG-CoA Synthase mRNA
Northern blot analysis of HMG-CoA synthase was performed to confirm results obtained with SRE-reporter gene assays. HMG-CoA synthase is sensitively regulated by SREBP.35 We have previously shown that ceramide and D-MAPP decrease HMG-CoA synthase mRNA levels.4 Incubation with myriocin for 16 hours decreases HMG-CoA synthase mRNA levels to less than half (Figure 6).
|
| Discussion |
|---|
|
|
|---|
We had previously shown that increasing cellular levels of ceramide decreases SRE-mediated gene transcription independent of cellular cholesterol levels, the classical feedback inhibitor of SRE-mediated gene transcription.4 Therefore, the initial hypothesis was that decreased ceramide synthesis should increase SRE-mediated gene transcription. On the contrary, inhibition of ceramide synthesis decreases SRE-mediated gene transcription (Figure 3) and levels of mature SREBP. Myriocin, a very specific inhibitor of serine-palmitoyl transferase, inhibits SRE-mediated gene transcription even more than cholesterol (Figure 3). Synthesis of ceramide occurs de novo through serine-palmitoyl transferase or through a recycling pathway via sphingosine.32,36 SRE-mediated gene transcription and mature SREBP levels are increased when ceramide synthesis is increased by sphingosine either added exogenously or increased endogenously through inhibition of sphingosine kinase (Figure 4). Taken together, the data suggest that increasing the flux into ceramide synthesis either de novo or through a recycling pathway modifies SREBP proteolysis and SRE-mediated gene transcription. The role of both pathways is demonstrated in LY-B cells. These cells, mutated in a subunit of serine-palmitoyl transferase and unable to synthesize sphingolipids de novo12,13 fail to increase SRE-mediated gene transcription in sterol depletion but recover this ability when sphingosine is present in the incubation medium (Figure 5). Our data do not implicate sphingosine 1-phosphate in SREBP proteolysis because inhibition of sphingosine kinase (DMS) increases SREBP proteolysis, and phosphatidylethanolamine, a metabolite of sphingosine 1-phosphate lyase, does not affect SRE-mediated gene transcription (data not shown).
To explain the finding that decreasing ceramide synthesis (this article) or increasing cellular ceramide levels4 both decrease SRE-mediated gene transcription, we questioned whether ceramide inhibits its own synthesis as shown for short-chain ceramides, dihydroceramides and dihydroceramide analogues.37,38 The data demonstrate that all ceramide analogues and inhibitors that increase intracellular cellular levels (ie, DMAPP, an inhibitor of ceramidase that increases cellular ceramide levels39 or PPMP, an inhibitor of glucosylceramide synthase) decrease ceramide de novo synthesis (Figure 1). Importantly, NB-DNJ, a glucosylceramide synthase inhibitor shown not to increase ceramide levels, equally does not affect ceramide de novo synthesis (Figure 1) or SRE-mediated gene transcription (Figure 3). These data support the role of ceramide in SRE-mediated gene regulation and furthermore suggests that glucosylceramide does not regulate SRE-mediated gene transcription. Therefore, inhibition of its own synthesis can be a mechanism of ceramide-mediated decrease of SRE-mediated gene expression.
Our data show that addition of exogenous ceramide increases levels of precursor SREBP and decreases levels of mature SREBP (Figure 2). This could suggest that the processing of precursor SREBP to mature SREBP is inhibited, potentially at multiple cellular sites. Ceramide synthesis has been shown to be obligatory in the ER to Golgi trafficking in yeast.10,11,40 Of relevance, increased levels of ceramides inhibit the formation of coated vesicles in CHO cells,7 glycoprotein traffic through the secretory pathway,41 and decrease endocytosis in mammalian cells.8 These observations also make it unlikely that the effect of ceramide on SREBP trafficking is unique.
We previously reported that ceramide decreases mature SREBP levels and SRE-mediated gene transcription even in the presence of inhibitors of intracellular cholesterol movement.4 Further evidence of a cholesterol-independent regulation of SREBP is found in Drosophila melanogaster, where SREBP levels are only regulated by palmitic acid and phosphatidylcholine but not by cholesterol or unsaturated fatty acids.42 Palmitic acid determines the rate of long-chain sphinganine synthesis,30 which can be further metabolized to ceramide. Hence, ceramide synthesis may also contribute to SREBP regulation in Drosophila. In mammalian cells, SREBP formation and cleavage occur by a number of metabolic pathways that can be modified by diet or by therapeutic agents. These "regulators" include cholesterol,35,43 fatty acids,15,44,45 and, as we show herein, modification of ceramide synthesis.
| Acknowledgments |
|---|
This work was supported by the National Institutes of Health Grants HL 40404. T. S. Worgall was supported in part by NIH training grant T32DK07715 and a grant-in aid awarded by the American Heart Association (0255656N). T. Seo was supported in part by NIH training grant HL0734322. R. A. Juliano was supported in part by NIH training grant DK07715. The authors gratefully acknowledge Fannie Keyserman for her excellent technical assistance.
Received January 27, 2004; accepted February 26, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Krishnamurthy, S. Dasgupta, and E. Bieberich Development and characterization of a novel anti-ceramide antibody J. Lipid Res., April 1, 2007; 48(4): 968 - 975. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J Deckelbaum, T. S Worgall, and T. Seo n-3 Fatty acids and gene expression Am. J. Clinical Nutrition, June 1, 2006; 83(6): S1520 - 1525S. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Ozbay, A. Rowan, A. Leon, P. Patel, and M. B. Sewer Cyclic Adenosine 5'-Monophosphate-Dependent Sphingosine-1-Phosphate Biosynthesis Induces Human CYP17 Gene Transcription by Activating Cleavage of Sterol Regulatory Element Binding Protein 1 Endocrinology, March 1, 2006; 147(3): 1427 - 1437. [Abstract] [Full Text] [PDF] |
||||
![]() |
T.-S. Park, R. L. Panek, S. B. Mueller, J. C. Hanselman, W. S. Rosebury, A. W. Robertson, E. K. Kindt, R. Homan, S. K. Karathanasis, and M. D. Rekhter Inhibition of Sphingomyelin Synthesis Reduces Atherogenesis in Apolipoprotein E-Knockout Mice Circulation, November 30, 2004; 110(22): 3465 - 3471. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2004 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |