Table 1.
Primers and polymerase chain reaction (PCR) conditions.
(A) Non-quantitative PCR: the first step of the PCR reaction was at 94°C for 5 min to activate the polymerase, followed by the indicated number of cycles of denaturation at 94°C for 45 s, annealing for 60 s at the indicated temperature for each gene, extension at 72°C for 90 s and an additional extension at 72°C for 5 min after the last cycle. (B) Real-time PCR: the first step was 95°C for 3 min, followed by 45 cycles of denaturation (95°C for 10 s), annealing (10 s at the indicated temperature for each gene) and extension (72°C for 10 s).
Fig 1.
Levels of SREBP proteins (A) and sphingomyelin in total and plasma membranes (B) in human adipose tissue.
In total, 23 obese women were grouped into tertiles according to their fasting plasma insulin (FPI) concentrations. Mean FPI±SEM (mU/L): low, 10.5±2.1; medium, 17.6±1.3; and high, 36.2±4.1. The results are expressed as percentages of low FPI cells. Cell lysates were separated by SDS-PAGE and immunoblotted with antibodies reacting with both precursor (p122) and cleaved active (p68) proteins of SREBP-1 or SREBP-2 (100 μg of protein). Total and plasma membranes were prepared, and sphingomyelin concentrations were determined as described in the Materials and methods. *P<0.05; high or medium FPI compared with low FPI.
Fig 2.
Relations between SM accumulation and SREBP-1, Ras, ERK, CREB and PPARγ proteins in human subcutaneous adipose tissue.
R, coefficient of correlation; statistical significance was set at P<0.05. Membrane SM levels refer to total membrane levels.
Fig 3.
Changes in the membrane phospholipid contents of 3T3-F442A adipocytes.
On day 9 of differentiation, the culture medium was supplemented with or without 15 μM exogenous SMs, 10 mM GSH (N-SMase inhibitor) or 20 μM PPMP for 24 h. Control cells were incubated with vehicle. The effects of four exogenous SMs were determined: (i) three natural with different acyl chains, primarily palmitic acid (SM-PA), stearic and nervonic acids (SM-SNA), or lignoceric acid (SM-LA), and (ii) one synthetic, namely, N-lignoceroyl-D-erythro-sphingosylphosphorylcholine (syn-SM). Total and plasma membranes were prepared, and phospholipid concentrations were determined as described in the materials and methods. (A) Levels of sphingomyelin in total membranes. (B) Levels of major phospholipids in plasma membranes following 24 h (15 μM) incubation with SM-PA, SM-SNA and SM-LA. (C) Dose-response relation. The cells were treated for 24 h with SM-LA (0–960 μM) and the levels of sphingomyelin in plasma membranes were quantified. The results are expressed as μg of the studied phospholipid per mg protein, presented as percentages of control cells, and are the mean±SEM of three independent experiments, which were each performed in triplicate (A) or duplicate (B and C). *P<0.05; SM-treated cells compared with control cells. **P<0.01; SM-, GSH-, and PPMP-treated cells compared with control cells. SM, sphingomyelin; PI, phosphatidylinositol; PS, phosphatidylserine; PE, phosphatidylethanolamine; PC, phosphatidylcholine.
Table 2.
Markers of nuclear and microsomal proteins in the plasma membrane and caveolae fractions.
Equal amounts of protein (10 μg) were subjected to SDS/PAGE, immunoblotting and densitometric scanning (Gel Doc 2000 imaging system, Bio-Rad). The results are expressed as the amount of the marker protein relative to the indicated subcellular fraction (set to 100%). The experiments shown are representative of 3 independent experiments. Abbreviations: plasma membrane (PM) fraction; caveolae membrane/lipid raft (CM) fraction.
Table 3.
Subcellular distribution of sphingomyelin in 3T3-F442A adipocytes.
Cells were treated with SM-LA (15 μM) for the indicated incubation times. The results, which are expressed as μg of SM per mg protein, are presented as percentages of control cells and are the mean±SEM of three independent experiments. The SM content of plasma membranes was 30±2 μg/mg protein in control adipocytes. The caveolae fraction represents approximately one-third (0.36) of the plasma membrane [78].
Table 4.
Time-dependent response of membrane fluidity to sphingomyelin treatment.
Cells were treated with SM-LA (15 μM) or vehicle for the indicated incubation times. The cells were labeled with 1,6-diphenyl-1,3,5-hexatriene (DPH), and the fluorescence anisotropy of the probe was determined. The measurements were performed at 37°C for 2 min immediately after the addition of the fluorescent probes. Each value is the mean±SEM of three independent experiments. P<0.05, P<0.005 and P<0.001; SM-treated cells compared with control cells.
Table 5.
Sphingomyelin enrichment decreased membrane fluidity in 3T3-F442A adipocytes.
3T3-F442A adipocytes were treated with 15 μM exogenous natural SM, 10 mM GSH, 20 μM PPMP or vehicle for 24 h. The cells were labeled and the measurements were performed as indicated in Table 4. Each value is the mean±SEM of four independent experiments. P<0.0005 and P<0.0001; SM-, GSH-, and PPMP-treated cells compared with untreated cells.
Fig 4.
SREBP protein and mRNA expression in SM-enriched adipocytes.
Cells were differentiated into adipocytes and treated with SMs as indicated in Fig 3. (A) Immunoblot analysis of nuclear SREBP proteins. Nuclear extracts were separated by SDS-PAGE (30 μg of protein) and immunoblotted with SREBP-1 (H-160) and SREBP-2 (N-19) polyclonal antibodies raised against epitopes mapping at the N-terminus of SREBP. (B) Semi-quantitative RT-PCR analysis of SREBP-1 and SREBP-2 mRNA levels. Total RNAs were reverse-transcribed and amplified by PCR as described in the Materials and methods. Amplified products were separated on an agarose gel and stained with ethidium bromide. Representative blots are shown under the histograms. The results are expressed as percentages of SM-untreated cells and are the mean±SEM of three independent experiments, performed in duplicate, triplicate or quadruplicate. *P<0.05, **P<0.01, ***P<0.005 and ****P<0.001; SM-treated cells compared with untreated cells.
Table 6.
Real-time quantitative RT-PCR determination of SREBP-1 and SREBP-2 mRNA levels in SM-enriched adipocytes or unmodulated adipocytes.
3T3-F442A adipocytes were treated with 15 μM natural SM (SM-LA, 24 h), 10 mM GSH (24 h), 20 μM PPMP (24 h) or vehicle. The mRNA levels of the studied genes were determined. The results are expressed as-fold variations over respective controls (R) after normalization to β-actin, as indicated in the Materials and methods. R values superior or equal to 2 were considered positive regulation of gene expression, whereas values lower than 0.5 indicated negative regulation. The results presented are the means of 3 independent experiments, which were performed twice each in duplicate.
Fig 5.
Inhibition of ERK proteins, but not p38 MAPK and JNK, in SM-enriched adipocytes.
Adipocytes were treated for 24 h with the indicated concentrations of SM-LA (A and B) or 30 μM SM, 10 mM GSH, 20 μM PPMP or vehicle (C and D). Cell lysates (40 μg of proteins) were separated by SDS-PAGE and immunoblotted using antibodies recognizing (1) the phosphorylated (activated) forms of ERK (p44/42 Tyr204), p38 (Tyr180/Tyr182), JNK (Thr183/ Tyr185), and (2) total ERK, p38, and JNK proteins. (A,C) Representative blots of phosphorylated and total proteins are shown. (B) Quantitative variations in phospho-ERK/JNK/p38 amounts relative to total proteins. (D) Quantitative variations in total and phospho-ERK amounts: the blots of p42 and p44 ERKs shown in (C) were combined to quantify the amounts of total ERK1/2 proteins. The results are expressed as percentages of control cells and are the mean±SEM of four independent experiments, which were each performed in duplicate. *P<0.05, **P<0.01 and ***P<0.005; SM-, GSH-, and PPMP-treated cells compared with untreated cells.
Fig 6.
MEK1/2 selective inhibitor (PD98059) and SM reduce the phosphorylation of ERK and SREBP-1.
Cells were treated 24 h with 50 μM PD98059 with or without 30 μM SM-LA. Cell lysates were separated by SDS-PAGE and immunoblotted with affinity-purified: (1) monoclonal antibody raised against a sequence containing phosphorylated Tyr204 of ERK1/2 and polyclonal antibody raised against a peptide mapping subdomain X1 of ERK and (2) with polyclonal antibodies raised against SREBP-1 or SREBP-2, as indicated in Fig 4. Representative blots (A) and quantitative variations (B) are shown. The results are expressed as the percentages of maximum and are the mean±SEM of four (ERK and SREBP-1) and six (SREBP-2) independent experiments, which were each performed in duplicate. *P<0.05, **P<0.01 and ***P<0.005; SM/ PD98059-treated cells compared with control cells.
Fig 7.
Inhibition of Ras, Raf, MEK(A and B)and KSR (C and D) by SM in 3T3-F442A adipocytes.
Cells were treated for 24 h with the indicated concentrations of SM-LA. Cell lysates (30–60 μg of protein) were separated by SDS-PAGE and immunoblotted with antibodies recognizing (1) the phosphorylated (activated) forms of Raf-1 (Tyr340/341), MEK-1/2 (Ser218/222) and KSR-1 (Ser392) and (2) total pan-Ras (N/H/K-Ras), Raf-1, MEK1/2 and KSR-1. Quantitative variations in phosphorytated amounts relative to total proteins are shown. The Ras activity was determined by ELISA (60 μg of cell lysate protein): active GTP-bound state of Ras is detected and measured quantitatively through the addition of a monoclonal anti-Ras antibody that detects K-, H-, N- Ras isoforms. GTP-Ras is normalized to the total-Ras. The results are expressed as percentages of SM-untreated cells and are the mean±SEM of four independent experiments, which were each performed twice. *P<0.05, **P<0.01 and ***P<0.005; SM-treated cells compared with untreated cells.
Table 7.
Real-time RT-PCR determination of caveolins mRNA levels in SM-enriched adipocytes or unmodulated adipocytes.
3T3-F442A adipocytes were treated with 30 μM exogenous natural SM (SM-LA, 24 h), 10 mM GSH (24 h), 20 μM PPMP (24 h) or vehicle. The results are expressed as-fold variation over respective controls (R) after normalization to β-actin, as indicated in the Materials and methods. R values greater than or equal to 2 were considered positive regulation of gene expression, whereas values lower than 0.5 indicated negative regulation. The results presented are the means of 3 independent experiments, which were performed twice each in duplicate.
Fig 8.
Effects of exogenous sphingomyelins on caveolin expression.
Cells treated with SMs as indicated in Fig 3. (A) Total RNAs were reverse-transcribed and amplified by PCR as described in the materials and methods. (B, C, and D) Lysates were separated by SDS-PAGE and immunoblotted with (1) a polyclonal antibody (N-20) and (2) a monoclonal antibody (IgG1, clone 65) that recognize Cav-1 and Cav-2, respectively (25 μg of proteins). A representative western blot is shown under each series of histograms, representing the expression levels of caveolins proteins: cellular (B), total membrane (C) and plasma membrane (D). The results are expressed as percentages of SM-untreated cells and are the mean±SEM of three independent experiments, which were performed in duplicate for mRNA and in triplicate for proteins. *P<0.05 and **P<0.01; SM-treated cells compared with untreated cells.
Fig 9.
SMs modulate free cholesterol (CHOL) distribution in 3T3-F442A adipocytes.
Total and plasma membranes, and cytosol were isolated from differentiated adipocytes and the level of CHOL was determined as indicated in methods. (A) Cells were treated for 24 h with SMs as indicated in Fig 3. (B) Cells were treated with SM-LA (24 h) with the indicated concentrations. (C) Mature adipocytes were fixed, stained with 100 μg/ml filipin for 3 h in the dark and visualized by fluorescence microscopy. From the top to the bottom of each column, representative examples of control cells, SM-PA-, SM-SNA- and SM-LA-treated cells are shown. One single representative adipocyte is shown on the right in each case. Cyt: cytoplasm; N: nucleus; G/ER: Golgi system and endoplasmic reticulum. (D) The distribution of CHOL between plasma membranes and triglyceride droplets (TGD) was examined by treating the cells with SM-LA (30 μM) for the indicated time. The results are expressed as μg of CHOL per mg protein and as percentages of SM-untreated cells and are the mean±SEM of four (A,B) or three (C,D) independent experiments, which were performed in triplicate. *P < 0.05 and **P<0.01; SM-treated cells compared with untreated cells.
Fig 10.
Neutral sphingomyelinase-selective inhibitor GW4869 inhibits the phosphorylation of ERK and SREBP-1 proteins.
Cells were treated for 24 h with 20 μM GW4869 with or without 30 μM SM-LA. Cell lysates were separated by SDS-PAGE and immunoblotted with an affinity-purified (1) monoclonal antibody raised against a sequence containing phosphorylated Tyr204 of ERK1/2 and polyclonal antibody raised against a peptide mapping subdomain X1 of ERK (40 μg of protein) and (2) polyclonal antibody raised against epitope mapping at the N-terminus of SREBP-1 (80 μg of protein). Representative blots (A) and quantitative variations (B) are shown. The results are expressed as percentages of maximum and are the mean±SEM of four independent experiments. (C) Enrichment of the SM content in total membranes of 3T3-F442A adipocytes. Total membranes were prepared, and SM concentrations were determined as described in the Materials and methods. The results are expressed as μg of SM per mg protein, are presented as percentages of control cells, and are the mean±SEM of three independent experiments. *P<0.05, **P<0.01 and ***P<0.005; SM- and GW4869-treated cells compared with control cells.
Fig 11.
SM-enriched adipocytes accumulate glucosylceramide, unlike SM-unmodulated cells.
Cells were treated for 24 h with 30 μM SM-LA, 20 μM PPMP or vehicle. After lipid isolation, ceramide and glucosylceramide were quantified as described in the Materials and methods. The results are expressed in pmol of sphingosine per mg of protein. The results presented are the means of 3 independent experiments, which were performed in triplicate. The control group was expressed as 100, and the results are shown as percentages of the control cells. ***P<0.005, **P<0.01 and *P<0.05; SM-, GSH-, and PPMP-treated cells compared with control cells.
Fig 12.
Rosiglitazone reverses the effects of SM in 3T3-F442A adipocytes.
Cells were treated for 24 h with 6 μM rosiglitazone (Rosi) with or without 30 μM SM-LA. Cell lysates (80 μg of protein) were separated by SDS-PAGE and immunoblotted with an affinity-purified polyclonal antibody raised against (1) a peptide mapping within the alpha region of CREB-1 p43, (2) a short amino acid sequence containing phosphorylated Ser 133 of CREB-1, (3) amino acids 8–106 of PPARγ and (4) epitope mapping at the N-terminus of SREBP-1. Representative blots (A) and quantitative variations (B) are shown. The results are expressed as percentages of control cells and are the mean±SEM of four independent experiments, which were each performed in duplicate. *P<0.05, **P<0.01 and ***P<0.005; SM-, Rosi-, and SM+Rosi-treated cells compared with control cells.
Fig 13.
SM inhibits the insulin-induced expression of SREBP-1.
Differentiated adipocytes were treated with SM-LA (30 μM, 24 h) in the presence or absence of insulin (100 nM, 24 h). Immunoblot analysis of mature SREBP-1 proteins; Nuclear extracts were separated by SDS-PAGE (30 μg of protein) and immunoblotted with the indicated antibodies. Representative blots (A) and quantitative variations (B) are shown. The results are expressed as percentages of control cells and are the mean±SEM of five independent experiments, performed in duplicate. *P<0.05; SM compared to control. **P<0.01; Ins compared to control. §P<0.01 and §§P<0.005; SM+Ins compared to Ins.
Fig 14.
Hypothetical model depicting how sphingomyelin might regulate SREBPs in 3T3-F442A adipocytes.
Illustration of how an SM-initiated signal transduction pathway leads to SREBP modulation. Two cell models of membrane SM enrichment were investigated in this report (1) by adding exogenous sphingomyelins in the culture medium of differentiated cells or (2) by inhibiting N-SMase with GW4869 or GSH. In turn, SM contents increase in membranes (caveolae). Membrane SM accumulation promotes a rigid state of the membrane and caveolin accumulation and regulates the Ras-Raf-ERK MAP kinase pathway, inhibiting SREBP-1. By inhibiting KSR, which is the putative target of ceramide, SM acts to amplify the signal provided to the Raf/ERK MAP kinase cascade. In turn, by affecting SREBP-1, SM induces SREBP-2. The increased levels of SREBP-2 are consistent with increased CHOL synthesis. In SM-enriched adipocytes, a higher amount of CHOL is required to maintain raft and caveolae assembly. Caveolin transports the neo-synthesized CHOL toward the cell surface, thus increasing the caveolin-2 levels in the plasma membrane. In SM-depleted cells, the opposite mechanism could be effective.