Tenovin-6 impairs autophagy by inhibiting autophagic flux

Abstract

Tenovin-6 has attracted significant interest because it activates p53 and inhibits sirtuins. It has anti-neoplastic effects on multiple hematopoietic malignancies and solid tumors in both in vitro and in vivo studies. Tenovin-6 was recently shown to impair the autophagy pathway in chronic lymphocytic leukemia cells and pediatric soft tissue sarcoma cells. However, whether tenovin-6 has a general inhibitory effect on autophagy and whether there is any involvement with SIRT1 and p53, both of which are regulators of the autophagy pathway, remain unclear. In this study, we have demonstrated that tenovin-6 increases microtubule-associated protein 1 light chain 3 (LC3-II) level in diverse cell types in a time- and dose-dependent manner. Mechanistically, the increase of LC3-II by tenovin-6 is caused by inhibition of the classical autophagy pathway via impairing lysosomal function without affecting the fusion between autophagosomes and lysosomes. Furthermore, we have revealed that tenovin-6 activation of p53 is cell type dependent, and tenovin-6 inhibition of autophagy is not dependent on its regulatory functions on p53 and SIRT1. Our results have shown that tenovin-6 is a potent autophagy inhibitor, and raised the precaution in interpreting results where tenovin-6 is used as an inhibitor of SIRT1.

Figure 1

LC3-II is increased following tenovin-6 treatment. (a) A panel of different types of cells was treated with tenovin-6 for 24 h. LC3B, SQSTM1/p62 and β-actin were examined by western blotting. Tenovin-6 was used at 5 μM except for A549 cells where 10 μM was used. (b) A549 and Huh7 cells were treated with indicated concentrations of tenovin-6 for 24 h. LC3B, SQSTM1/p62 and β-actin were examined by western blotting. (c) A549 and Huh7 cells were treated with 10 or 5 μM tenovin-6, respectively, for the indicated times. LC3B, SQSTM1/p62 and β-actin were examined by western blotting. (d) A549 and Huh7 cells were treated with 10 and 5 μM tenovin-6, respectively, for 16 h, and examined for LC3B by immunofluorescence assay

Figure 2

The increase of LC3-II by tenovin-6 is ATG5/7 dependent. (a) LC3B, LC3A, SQSTM1/p62 and β-actin were examined by western blotting in ATG5/7 wild-type and knockout MEF cells after treatment with 5 μM tenovin-6 for 24 h. (b) At day 4 post-infection with shRNA lentiviruses specific for ATG5 (shATG5), A549 and Huh7 cells were treated with 10 and 5 μM of tenovin-6, respectively, for another 8 h. LC3B, ATG5 and β-actin were then examined by western blotting. (c) The mRNA levels of LC3B and SQSTM1/p62 in A549 and Huh7 cells were examined by RT-qPCR following treatment with 10 and 5 μM of tenovin-6, respectively, for 4 h and 16 h. *P<0.05, **P<0.01

Figure 3

Tenovin-6 prevents Torin 1-induced SQSTM1/p62 degradation. (a) LC3B, SQSTM1/p62 and β-actin were examined by western blotting following the indicated treatments for 16 h. The concentrations of the compounds used were 10 μM tenovin-6 (Tnv-6), 100 μM chloroquine (CQ), 25 nM bafilomycin A1 (Baf A1) and 250 nM Torin 1 for A549 cells; and 5 μM Tnv-6, 50 μM CQ, 25 nM Baf A1 and 250 nM Torin 1 for Huh7 cells. (b) SQSTM1/p62 was examined by immunofluorescence assay following the indicated treatments for 16 h. The concentrations of the compounds used were the same as a. (c) LC3B, SQSTM1/p62 and β-actin were examined by western blotting after the indicated treatments for 8 h. Tenovin-6 was used at 10 μM for A549 and 5 μM for Huh7, whereas Baf A1 was used at 25 nM for all cells

Figure 4

Tenovin-6 does not affect the fusion between autophagosome and lysosome. (a and b) Colocalization of LC3B with LAMP1 (a) and SQSTM1/p62 with LAMP1 (b) were examined by immunofluorescence assay following the treatment with the indicated agents for 16 h. The concentrations of compounds used were 10 μM tenovin-6, 100 μM chloroquine, 25 nM bafilomycin A1 and 250 nM Torin 1 for A549 cells; and 5 μM tenovin-6, 50 μM chloroquine, 25 nM bafilomycin A1 and 250 nM Torin 1 for Huh7 cells. (c and d) The percentages of overlapping fluorescence of LC3B with LAMP1 (c) and SQSTM1/p62 with LAMP1 (d) were analyzed by fluorescence intensity profiling using the NIS-Elements 4.5 Software. Overlapping fluorescence was defined by (the number of overlapping peaks)/(total green fluorescence peaks). *P<0.05, **P<0.01, ***P<0.001

Figure 5

Tenovin-6 affects the acidification of autolysosomes and impairs the hydrolytic activity of lysosomes. (a) Colocalization of mCherry and GFP in live stable mCherry-GFP-LC3B-expressing cells following the indicated treatments. A549 cells were treated with 10 μM tenovin-6, 100 μM chloroquine, 25 nM bafilomycin A1 and 250 nM Torin 1 for 4 h. Huh7 cells were treated with 5 μM tenovin-6, 50 μM chloroquine, 25 nM bafilomycin A1 and 250 nM Torin 1 for 8 h. (b) The percentage of overlapping fluorescence analyzed by fluorescence intensity profiling using the NIS-Elements 4.5 Software. Overlapping fluorescence was defined by (the number of overlapping peaks)/(total red fluorescence peaks). ***P<0.001. (c) LysoTracker staining in cells following the indicated treatments. A549 cells were treated with 10 μM tenovin-6 and 25 nM bafilomycin A1 for 2 h. Huh7 cells were treated with 5 μM tenovin-6 and 25 nM bafilomycin A1 for 2 h. (d) Magic Red staining following the treatment of A549 cells with 10 μM tenovin-6 or 25 nM bafilomycin A1 for 3 h in the presence or absence of 250 nM Torin 1

Figure 6

Inhibition of autophagy by tenovin-6 does not correlate with p53 activation. (a and b) Total p53 and its acetylation form at K382, its phosphorylation form at S15, SIRT1, SIRT2, Bax, Puma, p21, LC3B and β-actin in A549 (a) and Huh7 (b) cells following tenovin-6 treatment at the indicated time points examined by western blotting. Tenovin-6 was used at 10 μM for A549 cells and 5 μM for Huh7 cell. OCI-Ly1 cells treated with 2 μM doxorubicin (Doxo) for 7 h were used as a positive control for p53 acetylation. (c) Total p53 and its acetylation form at K382, LC3B and β-actin in ATG5 wild-type (ATG^+/+) and knockout (ATG^−/−) MEF cells were examined by western blotting after treatment with 5 μM tenovin-6 for 24 and 48 h

Figure 7

Inhibition of SIRT1/2 by knockdown or knockout does not cause LC3B-II accumulation. (a) LC3B, SQSTM1/p62, SIRT1, SIRT2 and β-actin were examined by western blotting in A549, Huh7 and OCI-Ly1 cells following SIRT1 knockdown. A549 and Huh7 cells were collected 4 days after transduction without selection, whereas OCI-Ly1 cells were selected in 2.5 μg/ml puromycin for 5 days before collection. (b) LC3B, SQSTM1/p62, SIRT1, SIRT2 and β-actin were examined by western blotting in SIRT1 wild-type (SIRT^+/+) and knockout (SIRT1^−/−) KMM cells after treatment with 2.5 μM and 5 μM tenovin-6 for 24 h. (c) LC3B, SQSTM1/p62, SIRT1, SIRT2 and β-actin were examined by western blotting in OCI-Ly1 cells following SIRT2 knockdown or SIRT1/2 double knockdown. Cells were collected 4 days after transduction without selection