Objectives and background Recent fi ndings have implicated glycosphingolipids as modulators of insulin receptor activity. Studies with C57BL/6J ob/ob mice have shown that insulin sensitivity is enhanced by the synthetic hydrophobic iminosugar AMP-DNM (N-(5-adamantane1-yl-methoxy-pentyl)-deoxynojirimycin) that inhibits glucosylceramide synthase. Results Here we treated the liver hepatoma cell line HepG2 with AMP-DNM, resulting in a 70% reduction of glycosphingolipids, and we analyzed the effect on gene expression. Using whole human genome 44K oligonucleotide arrays, we identifi ed 89 genes that were signifi cantly (p<0.01) upor down regulated by AMP-DNM treated treatment. Of the 56 up-regulated genes, 17 were direct target genes for transcription factors sterol regulatory element-binding protein 1 (SREBP1) or SREBP2, which activate genes in the sterol biosynthesis pathway. An increase in cholesterol production rate confi rmed that the induction of SREBP target genes seen at the mRNA level resulted in activation of the cholesterol biosynthesis pathway. It is interesting to note that the cholesterol content of the cells did not increase. It is noteworthy that no effects were found on expression of genes related to cell receptor signaling pathways, neither on toxicity or cell growth. Conclusion Our fi ndings indicate that inhibition of glucosylceramide synthase with AMP-DNM leads to activation of SREBP target genes and synthesis of cholesterol in HepG2 cells. 111 C hpter 6 Introduction It has become increasingly clear that the lipid composition of cell membranes and especially of lipid rafts plays a major role in cell signaling processes. Lipid rafts or detergent-resistant membranes are specialized domains of cell membranes enriched in cholesterol, sphingomyelin, and glycosphingolipids (GSLs) (Lahiri and Futerman, 2007). The signaling capacity of various receptors such as the insulin receptor and the epidermal growth factor receptor that reside in these lipid rafts can be infl uenced by changing the glycosphingolipid composition of these membrane domains (Rebbaa et al., 1996; Kabayama et al., 2005; Inokuchi, 2006). This process was fi rst described by Bremer et al. (1986) who showed that epidermal growth factor-mediated signaling is inhibited by the ganglioside sialocyllactosylceramide (GM3). Furthermore, addition of GM3 to adipocytes in culture suppresses phosphorylation of the insulin receptor, resulting in reduced glucose uptake (Tagami et al., 2002). This could have important physiological consequences because glycosphingolipids have indeed been shown to accumulate in tissues from insulin-resistant rodents and humans (Summers and Nelson, 2005). Moreover, GM3 synthase-defi cient mice are protected from high-fat diet-induced insulin resistance (Yamashita et al., 2003). Thus, either blocking GM3 synthase or inhibiting the synthesis of excess glycosphingolipids could improve insulin signaling in vivo. We have described previously that N-(5-adamantane-1-ylmethoxypentyl)-deoxynojirimycin (AMP-DNM), an inhibitor of glucosylceramide synthase, specifi cally lowers glycosphingolipid levels without affecting ceramide levels in various cell models (Overkleeft et al., 1998; Aerts et al., 2003; Wennekes et al., 2007). AMP-DNM was found to reverse insulin resistance and normalize blood glucose levels in animal models of diabetes and obesity (Zucker fa/fa rats, diet-induced obese mice, and ob/ob mice). The compound also ameliorated lipotoxicity in kidney and pancreas (Aerts et al., 2007). Although these effects can be ascribed to improved insulin receptor function via modulation of glycosphingolipid levels by AMP-DNM, the question remains whether this is the only effect of AMP-DNM in vivo. Glycosphingolipids are associated with a wide range of functions, ranging from mediation of cell adhesion to modulation of signal transduction (Lahiri and Futerman, 2007). To be able to begin to understand the effects of AMP-DNM in vivo, where various hormones and organs are involved in maintaining glucose homeostasis, we set out to investigate the effects of established low glycosphingolipids at the cellular level in vitro. The liver is one of the major players in the physiology of diabetes and obesity, due to its gluconeogenic capacity and its role in lipoprotein metabolism. We monitored 112 the effect of AMP-DNM on total gene expression in human hepatoma HepG2 cells using genome wide microarray analysis. We found that treatment of HepG2 cells with AMP-DNM results in 70% reduction of glycosphingolipid content without an effect on expression of genes related to cell receptor signaling pathways, toxicity, or cell growth. However, we did observe a specifi c activation of sterol regulatory element-binding protein (SREBP) target genes, associated with the synthesis of lipids and cholesterol. Materials and Methods HepG2 Cell Culture. The human hepatic cell line HepG2 was obtained from American Type Culture Collection (Manassas, VA) and cultured in Dulbecco’s modifi ed Eagle’s medium/ Ham’s nutrient mixture F-12 (DMEM/HAMF-12) (Invitrogen, Carlsbad, CA), supplemented with 10% FCS at 10% CO2. For each experiment, cells were seeded in either 75or 25-cm2 tissue culture fl asks or tissue culture plates as indicated and grown to 60 to 70% confl uence. Cells were incubated in DMEM/HAMF-12 with 10μM AMP-DNM in 0.01% DMSO, or vehicle only. Incubation of control and AMPDNM-treated cells was continued for 0, 24, 48, or 72 h. AMP-DNM was synthesized as described previously (Overkleeft et al., 1998). Lipid Measurements Ceramide and neutral lipids levels were determined exactly as described previously (Groener et al., 2007). In short, cells were washed three times in PBS, lysed in 500μl of water, and further disrupted by sonication (three strokes) on ice. Lipids were extracted with 2 ml of chloroform/methanol [1:1 (v/v)] according to Folch et al. (1957) followed by deacylation in 500 μl of 0.1M NaOH in methanol using a microwave oven (CEM microwave Solids/Moisture System SAM-155; CEM, Matthews, NC). The deacylated lipids were derivatized for 30 min with the addition of 25 μl of O-phtaldehyde reagent to 50 μl of lipid mixture and separated with a highperformance liquid chromatography (HPLC) method. Gangliosides were detected as described recently (Ghauharali-van der Vlugt et al., 2008) by analysis of the acidic glycolipid fraction obtained by the Folch extraction. In short, the upper phase was desalted on a C18 Sep-Pak (Bakerbond) column, and the eluted gangliosides were digested with ceramide glycanase. The released oligosaccharides were labeled at their reducing end with the fl uorescent compound anthranilic acid (2-aminobenzoic 113 C hpter 6 acid), before analysis using normal-phase HPLC. For total cholesterol and cholesterol-ester determination, cells were washed in PBS, lysed in 500 μl of 1% Triton X-100 (v/v), and further disrupted by sonication. Cholesterol content in these samples was measured using the cholesterol oxidase reaction coupled to fl uorometric determination of hydrogen peroxide as described previously (Elferink et al., 1998). Microarray Expression Profi ling and Pathway Analysis Cells were incubated in 75-cm2 tissue culture fl asks for 48 h in either normal medium (n = 6 for reference RNA on the arrays), medium with vehicle (n=3 control samples), or medium with 10 μM AMP-DNM (n = 3 treated samples). Subsequently, RNA was extracted using the TRIzol RNA isolation method (Invitrogen), followed by a purifi cation step using RNA II NucleoSpin columns, including DNase treatment (Machery Nagel, Düren, Germany). RNA concentration and integrity were determined with a NanoDrop (Wilmington, DE) and Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA), respectively. The individual samples from the control and treated cells were used for probe synthesis and hybridized on separate arrays against a probe made from a pool of reference RNA. Probe synthesis and microarray hybridizations were performed by Service XS (Leiden, The Netherlands) using a direct labeling reaction and human whole genome 44K arrays (Agilent Technologies). Processed images were imported into the Rosetta Resolver database and analysis software (Rosetta Biosoftware, Seattle, WA). Statistical analysis was performed using false discovery rate-corrected p values, involving a recalculation of the p values using a Benjamini-Hochberg correction for multiple testing. Genes with a signifi cantly (p < 0.05) altered expression profi le were imported into pathway analysis software (MAPPfi nder version 2.0) and PubMed-mining software (Ingenuity Pathway Analysis version 2; Ingenuity Systems, Redwood City, CA) for further analysis. Expression-Level Analysis Quantitative gene analysis was used to confi rm differential gene expression of candidate transcripts found by microarray analysis. Total RNA (2 μg) was used for fi rststrand cDNA synthesis with SuperScript II reverse transcriptase and oligo(dT) primer (Invitrogen). The PCR primers were designed using Primer 3 (Rozen and Skaletsky, 2000), and specifi city was verifi ed by conventional PCR. Quantitative PCR was performed with the MyIQ real-time detection system (Bio-Rad, Hercules, CA) using the SYBR Green I reaction mix (Bio-Rad). PCRs were performed in duplicate and normalized to cyclophilin and acidic ribosomal phosphoprotein P0 (36B4). 114 Preparation of Nuclear Extracts and Western Blot Analysis HepG2 cells were incubated in DMEM/HAMF-12 with 10 μM AMP-DNM in 0.01% DMSO, or vehicle only for 48 h in 10-cm2 tissue culture dishes. Subsequently, cells were washed with ice-cold PBS. To prevent degradation of proteins, a cocktail of protease inhibitors was added to the lysis buffer (Complete; Roche Molecular Biochemicals, Indianapolis, IN). Nuclear and cytosolic extracts were prepared using a nuclear extract kit (NE-PER; Pierce Chemical, Rockford, IL) according to the manufacturer’s protocol. The protein concentration was measured with
[1]
M. Langeveld,et al.
Prominent increase in plasma ganglioside GM3 is associated with clinical manifestations of type I Gaucher disease.
,
2008,
Clinica chimica acta; international journal of clinical chemistry.
[2]
J. Seppen,et al.
The sterol transporting heterodimer ABCG5/ABCG8 requires bile salts to mediate cholesterol efflux
,
2007,
FEBS letters.
[3]
A. Futerman,et al.
The metabolism and function of sphingolipids and glycosphingolipids
,
2007,
Cellular and Molecular Life Sciences.
[4]
S. O’Rahilly,et al.
Pharmacological Inhibition of Glucosylceramide Synthase Enhances Insulin Sensitivity
,
2007,
Diabetes.
[5]
C. Hollak,et al.
HPLC for simultaneous quantification of total ceramide, glucosylceramide, and ceramide trihexoside concentrations in plasma.
,
2007,
Clinical chemistry.
[6]
H. Overkleeft,et al.
Development of adamantan-1-yl-methoxy-functionalized 1-deoxynojirimycin derivatives as selective inhibitors of glucosylceramide metabolism in man.
,
2007,
The Journal of organic chemistry.
[7]
J. Inokuchi.
Insulin resistance as a membrane microdomain disorder.
,
2007,
Biological & pharmaceutical bulletin.
[8]
A. McIntosh,et al.
Liver fatty-acid-binding protein (L-FABP) gene ablation alters liver bile acid metabolism in male mice.
,
2005,
The Biochemical journal.
[9]
I. Gelissen,et al.
Glycosphingolipid Accumulation Inhibits Cholesterol Efflux via the ABCA1/Apolipoprotein A-I Pathway
,
2005,
Journal of Biological Chemistry.
[10]
S. Summers,et al.
A role for sphingolipids in producing the common features of type 2 diabetes, metabolic syndrome X, and Cushing's syndrome.
,
2005,
Diabetes.
[11]
Y. Igarashi,et al.
TNFalpha-induced insulin resistance in adipocytes as a membrane microdomain disorder: involvement of ganglioside GM3.
,
2004,
Glycobiology.
[12]
A. Rebbaa,et al.
Ganglioside GM 3 inhibition of EGF receptor mediated signal transduction
,
2005
.
[13]
R. Deckelbaum,et al.
Ceramide Synthesis Correlates with the Posttranscriptional Regulation of the Sterol-Regulatory Element-Binding Protein
,
2004,
Arteriosclerosis, thrombosis, and vascular biology.
[14]
Jay D. Horton,et al.
Combined analysis of oligonucleotide microarray data from transgenic and knockout mice identifies direct SREBP target genes
,
2003,
Proceedings of the National Academy of Sciences of the United States of America.
[15]
R. B. Rawson.
The SREBP pathway — insights from insigs and insects
,
2003,
Nature Reviews Molecular Cell Biology.
[16]
C. Hollak,et al.
Biochemistry of glycosphingolipid storage disorders: implications for therapeutic intervention.
,
2003,
Philosophical transactions of the Royal Society of London. Series B, Biological sciences.
[17]
R. Proia,et al.
Enhanced insulin sensitivity in mice lacking ganglioside GM3
,
2003,
Proceedings of the National Academy of Sciences of the United States of America.
[18]
Hitoshi Shimano,et al.
Transcriptional activities of nuclear SREBP-1a, -1c, and -2 to different target promoters of lipogenic and cholesterogenic genes DOI 10.1194/jlr.M100417-JLR200
,
2002,
Journal of Lipid Research.
[19]
Rebecca A. Johnson,et al.
Unsaturated Fatty Acid-mediated Decreases in Sterol Regulatory Element-mediated Gene Transcription Are Linked to Cellular Sphingolipid Metabolism*
,
2002,
The Journal of Biological Chemistry.
[20]
Y. Igarashi,et al.
Ganglioside GM3 Participates in the Pathological Conditions of Insulin Resistance*
,
2002,
The Journal of Biological Chemistry.
[21]
J. Horton.
Sterol regulatory element-binding proteins: transcriptional activators of lipid synthesis.
,
2001,
Biochemical Society transactions.
[22]
S Rozen,et al.
Primer3 on the WWW for general users and for biologist programmers.
,
2000,
Methods in molecular biology.
[23]
A. Smith,et al.
UvA-DARE ( Digital Academic Repository ) Class III P-glycoproteins mediate the formation of lipoprotein X in the mouse
,
1998
.
[24]
H. Overkleeft,et al.
Generation of Specific Deoxynojirimycin-type Inhibitors of the Non-lysosomal Glucosylceramidase*
,
1998,
The Journal of Biological Chemistry.
[25]
M. Brown,et al.
Sphingomyelin depletion in cultured cells blocks proteolysis of sterol regulatory element binding proteins at site 1.
,
1997,
Proceedings of the National Academy of Sciences of the United States of America.
[26]
I. Shimomura,et al.
Differential expression of exons 1a and 1c in mRNAs for sterol regulatory element binding protein-1 in human and mouse organs and cultured cells.
,
1997,
The Journal of clinical investigation.
[27]
J. Schlessinger,et al.
Ganglioside-mediated modulation of cell growth. Specific effects of GM3 on tyrosine phosphorylation of the epidermal growth factor receptor.
,
1986,
The Journal of biological chemistry.
[28]
J. Folch,et al.
A simple method for the isolation and purification of total lipides from animal tissues.
,
1957,
The Journal of biological chemistry.