Supplemental Material

This article contains the following supporting material:

• Figure 1 - Primary human hepatocytes were cultured on collagen-coated glass slides, treated by 10 μM ALLN and 10 mM 3-MA for 12 h, and labeled for ApoB (red) and neutral lipids (green). The hepatocyte showed ApoB-crescents (arrows) that appeared to be identical to those found in Huh7 and HepG2 cells, but their frequency was low. Bar, 10 μm.

• Figure 2 - Huh7 cells were cultured in the following conditions for 6 h and the percentage of cells showing ApoB-crescents were counted. The conditions examined are: control (10% FCS), 10% LPDS, 50 μM mevastatin/50 μM mevalonolactone, 10% LPDS plus 50 μM mevastatin/50 μM mevalonolactone, and 5 μg/ml brefeldin A. In comparison to the control, all treatments increased ApoB-crescents significantly. Results of three independent experiments were averaged; statistical significance was examined by Student^³s t-test (* p < 0.05, ** p < 0.01).

• Figure 3 - Samples from the same starting material were subjected to Western blotting analysis, and the relative amount of ApoB was estimated to quantitate the proportion of ApoB in the fraction #1 (or CLD fraction) to the total lysate. The total lysate was electrophoresed in 1/1, 1/10 or 1/100 of the equivalent to make the reaction intensity in the same range as the fraction #1. (A, B) ApoB in the fraction #1 increased drastically by ALLN treatment; its reaction intensity was between the 1/1 and 1/10 dilution of the total lysate. The control fraction #1 showed a weaker reaction than the 1/100 dilution of the total lysate. (C) The relative reaction intensity of fraction #1 to the total lysate was 0.84 ± 0.30% in the control to 15.29 ± 3.52% in the ALLN-treated cell (n = 3; *p < 0.005).

• Figure 4 - The 20S proteasome activity was measured using 100 μM N-succinyl-Leu-Leu-Val-Tyr-7-amino 4-methylcoumarin as the substrate.The reaction mixtures (100 μl) were incubated for 1 h or 3 h at 37°C, and fluorescence intensity was measured using an excitation wavelength of 355 nm and an emission wavelength of 460 nm. Relative activities are shown (fraction #8 at 1 h = 100%). The proteasomal activity of the cytosolic fraction was significantly lowered by mixing with the CLD fraction instead of a buffer solution.

• Figure 5 - Huh7 cells were transfected with either wild-type dynamin-2 (wt) cDNA or dominant-negative dynamin-2 mutant (K44A) cDNA, and examined 2 days later. For quantitation, results of three independent experiments were averaged, and statistical significance was examined using Student's t-test. (A) Uptake of rhodamine-transferrin (Sigma) was reduced significantly in cells expressing the K44A mutant in comparison to cells expressing wild-type dynamin-2. Quantitation was done by measuring the intensity of rhodamine fluorescence in transfected cells (n = 3; *p < 0.05). (B) In contrast, colocalization of ApoB and Lysotracker after 12 h of ALLN treatment was observed similarly in cells transfected with wt and K44A cDNA. Quantitation confirmed that ApoB colocalizing with Lysotracker increased significantly by ALLN treatment, even in cells transfected with the K44A cDNA (n = 3; * p < 0.05, ** p < 0.01); colocalization showed no statistical difference between cells expressing wt and K44A dynamin-2.

• Figure 6 - The LC3 labeling started to increase as early as 1 h after ALLN was given. The number of LC3-positive particles per cell was quantitated in Huh7 cells treated with 10 μM ALLN for 0 h, 1 h, 12 h, or with 10 μM ALLN and 10 mM 3-MA for 12 h. Results of three independent experiments were averaged, and statistical significance was examined using Student's t-test. (n = 3; *p < 0.01)