J. test (?p 0.01, ??p 0.001). The frequency of ApoB-crescents increased markedly when the cells were treated by proteasome inhibitors, ALLN or MG132. ALLN at 10 M increased the percentage of Huh7 cells showing ApoB-crescents from 10% at 0 h to nearly 50% at 12 h (Physique 1, B and C). The ALLN treatment also increased the ratio of CLDs bearing ApoB-crescents in a similar manner (Physique 1D). The increases were significant as early as 1 h after addition of ALLN. These observations suggest that ApoB-crescents are related to the ubiquitin-proteasome degradation of ApoB. MG132 at 50 M or ALLN at concentrations 20 M caused a similar increase but affected the overall cell shape. Therefore, we mainly used 10 M ALLN for the subsequent experiments. Notably, the frequency of ApoB-crescents decreased to the basal level after 24 h of ALLN treatment (Physique 1C), and ApoB labeling BRD-6929 in other locations increased as explained below. The increase of ApoB-crescents was also observed when cells were cultured in DMEM with 10% LPDS instead of 10% FCS (Supplemental Physique 2). The frequency of ApoB-crescent further increased by adding mevastatin and mevalonolactone to the LPDS medium to inhibit de novo synthesis of cholesterol without affecting isoprenylation of Ras family proteins. Mevastatin/mevalonolactone alone also induced a slight but significant increase of ApoB-crescents. These results showed that ApoB-crescents were created when lipid supply was not adequate and that they did not form only as a result of nonspecific stress response to the protease inhibition. Brefeldin A, which blocks BRD-6929 vesicular transport from ER to Golgi, induced a sharp increase of ApoB-crescents as expected. ApoB-Crescents Were Complementary to ADRP and TIP47 around CLDs Among the PAT proteins that coat CLDs in mammalian cells, ADRP and TIP47 are expressed in nonadipocytes (Miura test (?p 0.05, ??p 0.001). (E and F) Immunogold labeling of ALLN-treated Huh7 cells. Bars, 0.2 m. (E) Lysosomes (arrowheads) contained ApoB-positive electron-lucent particles (arrows) and adhered to the ApoB-crescent area adjacent to CLDs. (F) In some cases, the lysosomes made up of ApoB labeling (arrows) wrapped round the ApoB-crescent and the adjacent CLD. The limiting membrane of the lysosome is usually marked by arrowheads. Immunogold labeling of cryosections revealed that this lysosomal lumen contained electron-lucent components labeled for ApoB (Physique 5E, arrows). Interestingly, the lysosomes were usually seen in the vicinity of ApoB-crescents adjacent to CLD. In some cases, the limiting membrane of the lysosomes made up of ApoB labeling seemed to wrap round the ApoB-crescent and the adjacent CLD (Physique 5F). The absence of ApoB in the early endosome (Physique 5B) implied that this lysosomal ApoB was not derived from endocytosed VLDL. To further confirm this point, cDNA of dominant-negative dynamin-2 (K44A) was transfected, and its effect on the amount of the lysosomal ApoB was examined. In comparison with wild-type dynamin-2, the K44A mutant inhibited the uptake of rhodamine-transferrin, but it did not affect the apoB labeling that colocalized with Lysotracker (Supplemental Physique 5). The result showed that most ApoB in the lysosome was not caused by uptake of the secreted lipoprotein. ApoB-Crescents Are Processed by Autophagy The above-mentioned result as well as immunoelectron microscopy of ALLN-treated cells suggested that this Rabbit polyclonal to ABCB1 ApoB-positive particles in the lysosome were largely caused by autophagy. In fact, a previous study exhibited that inhibition of proteasomes in Huh7 cells induced autophagic vacuoles (Harada test (n = 3; ?p 0.001). (E and F) Percentage BRD-6929 of cells (E) and CLDs (F) showing ApoB-crescents in cells treated with 3-MA alone. The frequency of crescent-positive cells and CLDs reached a maximum at 12 h. Results of three impartial experiments were averaged; statistical difference from your control (0 h) was examined by Student’s test (E: ?p 0.01, ??p 0.001; F: p 0.05, ??p 0.001). The above-mentioned assumption was tested by inhibiting autophagy by 3-MA. The ALLN-induced increase in LC3 was suppressed when the medium contained 10 mM 3-MA (Physique 6A), which confirmed the effectiveness of the reagent. In cells treated with 3-MA at 12 h after the beginning of ALLN treatment, the decrease in ApoB-crescents between 12 and 24 h was suppressed (Physique 6D). The suppressive effect was more obvious when cells were treated with 3-MA and ALLN from the beginning (Physique 6D). This observation verified that autophagy is usually important in the processing of ApoB-crescents. Furthermore, it suggested that even when proteasomal function is usually normal, autophagy may function in ApoB degradation. This assumption proved correct because 3-MA alone caused an increase in ApoB-crescent-positive cells and CLDs (Physique 6, E and F). The ratio of ApoB-crescent-positive cells increased to more than 30% at 12 h after addition of 3-MA. The results of the present study indicate that proteasomal and autophagocytic systems collaborate to process.