JCS02941 Supplementary Material

Files in this Data Supplement:

• Supplemental Table 1 - Adobe PDF

• Supplemental Table 2 - Adobe PDF

• Supplemental Figure 1 -

  Fig. S1. PLD1 localizes to interior membranes while PLD2 is located at the plasma membrane. Antibodies to PLD 1 or 2 (as indicated ) were used in immunohistochemistry experiments to localize the two isoforms in NIH 3T3 cells as described (Du et al., 2004; Freyberg et al., 2001).

• Supplemental Figure 2 -

  Fig. S2. Ste-Mek1₁₃, an inhibitor of the phosphorylation and subsequent activation of ERK2, inhibits the formation of cytosolic lipid droplets. Cells were first treated with 50 μM Ste-Mek1₁₃ for 60 minutes and incubated with oleic acid together with 50 μM Ste-Mek1₁₃ for 30 or 60 minutes (as indicated). The cells were stained with Oil Red O and the total area of red pixels in the Oil Red O-stained droplets/cell was determined (mean ± s.e.m. of all cells present in 20 randomly selected pictures from each group; *P=0.008 vs. control (i.e. no Ste-Mek1₁₃ treatment); †P=0.025 vs. control (i.e. no Ste-Mek1₁₃ treatment) (t-test). Dose-response experiments (not shown) demonstrated that the maximal effect of Ste-Mek1₁₃ was reached at a concentration of 25 μM, with no further effect at 50 μM. Only a small effect was seen at 10 μM.

• Supplemental Figure 3 -

  Fig. S3. ERK2 siRNA does not influence the expression of ERK1. Immunoblot of ERK1 in NIH 3T3 cells transfected with siRNA to ERK2, or a control siRNA, as indicated.

• Supplemental Figure 4 -

  Fig. S4. Transfection of ERK1 and siRNA to ERK1 does not significantly influence the formation of lipid droplets in the cell. (A) Immunoblot of ERK1 in NIH 3T3 cells transfected with ERK1-YFP (i.e. ERK1 fused to YFP) or GFP as control. (B) The effect of transfection of NIH 3T3 cells with GFP or ERK1-YFP on the amount of Oil Red O-stained lipid droplets in the cell (mean ± s.d.; all cells in 20 randomly selected pictures). (C) Immunoblot of ERK1 and β-actin (control protein) in NIH 3T3 cells transfected with ERK1 siRNA or a control siRNA. (D) The effect of transfection of NIH 3T3 cells with a control siRNA or ERK1 siRNA on the amount of Oil Red O-stained lipid droplets in the cell (mean ± s.d.; all cells in 20 randomly selected micrographs).

• Supplemental Figure 5 -

  Fig. S5. The effect of p38 and JNK siRNA and inhibitors of JNK on the formation of lipid droplets in NIH 3T3 cells. (A) Immunoblot of p38 and GRP78 (control protein) in NIH 3T3 cells transfected with an siRNA to p38α or p38β, or a control siRNA. (B) The effect of transfection of NIH 3T3 cells with a control siRNA, or siRNA to p38α or p38β, on the amount of Oil Red O-stained lipid droplets in the cell (mean ± s.d.; all cells in 20 randomly selected pictures. (C) Immunoblot of JNK and GRP78 (control protein) in NIH 3T3 cells transfected with siRNA to JNK1 or JNK2, or with a control siRNA. (D) The effect of transfection of NIH 3T3 cells with a control siRNA or siRNA to JNK1 or JNK2 on the amount of Oil Red O-stained lipid droplets in the cell (mean ± s.d.; all cells in 20 randomly selected micrographs). (E) The effect of the JNK inhibitor I [H-GRKKRRQRRRPPRPKRPTTLNLFPQVPRSQDT-NH₂ (Barr et al., 2002; Bonny et al., 2001)] (5μM) on the amount of Oil Red O-stained lipid droplets in the cell (mean ± s.d.; all cells in 30 randomly selected micrographs).

• Supplemental Figure 6 -

  Fig. S6. ERK2 has no influence on the rate of biosynthesis of triglycerides or the rate of β-oxidation. (A) NIH 3T3 cells were transfected with ERK2 siRNA (squares) or a control siRNA (triangles) and treated with [³H]palmitic acid for the indicated period of time. After each period of incubation, the triglycerides were isolated by thin layer chromatography and the radioactivity determined (mean ± sd; 3 different culture dishes; NS). (B) The cells were transfected with ERK2 or a control vector and incubated for 2 hours with [³H]-palmitic acid and the amount of water-soluble radioactivity measured (see Materials and Methods) (mean ± s.d.; six different culture dishes; NS)

• Supplemental Figure 7 -

  Fig. S7. Subcellular fractionation of NIH3T3 cells after nitrogen cavitation. NIH 3T3 cells were incubated with oleic acid for 6 hours and then disrupted by nitrogen cavitation (800 psi over 30 minutes). The cell homogenate in 20% sucrose was layered on top of a sucrose gradient from 25-50% (in 10 mM Tris-HCl, pH 7.4) and overlaid with 5% sucrose in 10 mM Tris-HCl, pH 7.4. After centrifugation, the gradient was unloaded from the top in 1 ml fractions. The density of each fraction was determined (upper panel). Each fraction was blotted against ADRP (as a marker of lipid droplets), tubulin and vimentin (as markers of cytosolic proteins) and β-COP and golgin (as markers of the Golgi apparatus). Based on these results and the observation that the distribution of the ER marker (GRP 78; not shown) grossly overlapped with the Golgi markers, we collected the fractions marked in the figure as lipid droplets and microsomes.

• Supplemental Figure 8 -

  Fig. S8. The effect of overexpression and knockdown of PLD1 on the phosphorylation of ERK2. (A) NIH 3T3 cells were transfected with PLD1 or a vector encoding GFP (‘Control’) and the amount of phosphorylated ERK2 was determined by immunoblot 24 hours after the transfection. GRP78 was used as control protein. (B) Cells were transfected with siRNA to PLD1 or a control siRNA and the amount of phosphorylated ERK2 (and GRP78; as control protein) was determined by immunoblot.

• Supplemental Figure 9A
• Supplemental Figure 9BC
• Supplemental Figure 9DE
• Supplemental Figure 9FG
• Supplemental Figure 9HIJ -

  Fig. S9. Software used to analyze droplets. (A) Identification of a red pixel (upper frame) with BioPix software and the equation used (lower frame; see text). (B) Empirical determination of the coefficient used in the first equation. The coefficient was allowed to vary between 0 and 1, and the relationship between the droplets counted manually and by the BioPix software was investigated as a function of this variation. (C) The value obtained for the coefficient (0.8) was tested in a new experiment in which manually counted Oil Red O-stained droplets were compared to droplets counted by the software. (D) Dextran beads of different sizes injected into the cytosol of NIH 3T3 cells and subjected to confocal microscopy and analyzed by the software (bar: 1μM). (E) Fluorescent beads of known size were microinjected into the cytosol of NIH 3T3 cells and the cells were subjected to confocal microscopy (as in D). The software was used to determine the volume of the beads injected. There was a highly significant correlation between the expected volume and the volume determined (r=0.9997; P<0.001). (F) The relationship between the total volumes of cytosolic lipid droplets/cell determined after staining with Nile Red or Bodipy (x-axis) and Oil Red O (y-axis) (the results are based on all cells in 20 randomly selected pictures). Cells were treated with oleic acid for different times to generate different amounts of lipid droplets before they were stained as indicated, and viewed by confocal microscopy. The software was used to calculate the total volume of all droplets/cell (r=0.991; P<0.001). (G) The relationship between the total volume of Oil Red O-stained lipid droplets/cell (x-axis) and the corresponding total area of the Oil Red O-stained droplets/cell (the results are based on all cells from 20 randomly selected pictures). Cells were incubated with oleic acid for different times to obtain cells with different amounts of cytosolic lipid droplets, and then stained with Oil Red O. The software was used to calculate the total volume (after confocal microscopy) and the total area of droplets/cell (r=0.991; P<0.001). (H) The distribution (percentage of total amount) of droplets of different sizes after staining with Oil Red O (squares) and Nile red (triangles). NIH 3T3 cells were incubated with oleic acid for 2 hours and stained with Oil Red O or Nile Red. The size and total area of the droplets were analyzed with the software, and the proportion of the total area of droplets found within the different sizes was determined. (I) The relationship between the cellular triglyceride content and the total area of Oil Red O-stained lipid droplets determined using the software. NIH 3T3 cells were cultured in parallel in 10-cm² culture dishes or on coverslips in the presence of oleic acid for periods between 0 and 8 hours. The cells in the culture dishes were harvested and analyzed for total proteins and for the amount of triglycerides (after solvent extraction). The cells cultured on coverslips were stained with Oil Red O, and the cells in 20 randomly taken pictures were analyzed for total area of Oil Red O-stained lipid droplets/cell (r=0.93; P<0.001). (J) The relationship between the cellular triglyceride content and the total area of Oil Red O-stained lipid droplets determined using the software. NIH 3T3 cells were cultured in parallel in 10-cm² culture dishes or on coverslips, in the presence of oleic acid for periods between 0 and 2 hours. They were then analyzed as in panel I above (r=0.86; P=0.015)