Cell Death Triggered by the Autophagy Inhibitory Drug 3-Methyladenine in Growing Conditions Proceeds With DNA Damage

Abstract

Macroautophagy (hereafter autophagy) is a multistep intracellular catabolic process with pleiotropic implications in cell fate. Attending to its activation, autophagy can be classified into inducible or constitutive. Constitutive, or basal autophagy, unfolds under nutrient-replete conditions to maintain the cellular homeostasis. Autophagy inhibitory drugs are powerful tools to interrogate the role of autophagy and its consequences on cell fate. However, 3-methyladenine and various of these compounds present an intrinsic capacity to trigger cell death, for instance the broadly-employed 3-methyladenine. To elucidate whether the inhibition of basal autophagy is causative of cell demise, we have employed several representative compounds acting at different phases of the autophagic process: initiation (SBI0206965 and MHY1485), nucleation (3-methyladenine, SAR405, Spautin-1 and Cpd18), and completion (Bafilomycin A₁ and Chloroquine). These compounds inhibited the basal autophagy of MEF cultures in growing conditions. Among them, 3-methyladenine, SBI-0206965, Chloroquine, and Bafilomycin A₁ triggered BAX- and/or BAK-dependent cytotoxicity and caspase activation. 3-methyladenine was the only compound to induce a consistent and abrupt decrease in cell viability across a series of ontologically unrelated human cell lines. 3-methyladenine-induced cytotoxicity was not driven by the inhibition of the AKT/mTOR axis. Autophagy-deficient Fip200-/- MEFs displayed an increased sensitivity to activate caspases and to undergo cell death in response to 3-methyladenine. The cytotoxicity induced by 3-methyladenine correlated with a massive DNA damage, as shown by γ-H2A.X. This genotoxicity was observed at 10 mM 3-methyladenine, the usual concentration to inhibit autophagy and was maximized in Fip200-/- MEFs. In sum, our results suggest that, in growing conditions, autophagy acts as a protective mechanism to diminish the intrinsic cytotoxicity of 3-methyladenine. However, when the cellular stress exerted by 3-methyladenine surpasses the protective effect of basal autophagy, caspase activation and DNA damage compromise the cell viability.

Keywords: 3-methyladenine; apoptosis; autophagy inhibitor; basal autophagy; Ɣ-H2A.X.

Figure 1

Schematic illustration depicting the process of autophagy and the targets of the autophagy inhibitory drugs. mTOR is a serine/threonine protein kinase that regulates cell growth and anabolism. Associated to other proteins, mTOR forms a complex known as mTORC1, a sensor of the nutritional state of the cell. In growing conditions, mTORC1 constitutively blocks the “ULK initiation complex” and hence, autophagy. MHY1485 is an activator of mTOR that acts through a yet unknown mechanism. The “ULK initiation complex” contains ULK1 or its homolog ULK2, FIP200, and ATG13 and triggers the first step of autophagy known as Initiation. SBI-0206965 is an inhibitor of ULK1 and ULK2. The “ULK initiation complex” drives the formation of the precursors of the autophagosomes through the direct activation of the “VPS34 Nucleation Complex”, for instance by phosphorylating VPS34 (Vacuolar Protein Sorting 34) and BECLIN-1. Spautin-1 triggers the destruction of BECLIN-1 through the inhibition of two of its deubiquitinases. On the other hand, VPS34 is a class 3 phosphatidylinositol 3-phosphate-kinase (PI3KC3) accountable for the production of the phospholipid phosphatidylinositol 3-phosphate (PI3P) necessary for the recruitment of PI3P-binding proteins that lead the nucleation and elongation of the vesicles. 3-MA, its close analog Cpd18 and SAR405 are inhibitors of VPS34. In addition, a third complex consisting of ATG16L1–ATG5–ATG12 plays a key role in the “Nucleation and Elongation” phase. This complex orchestrates a conjugation similar to what E3-ubiquitin ligases do, but in this case, they catalyze the transfer of the “ubiquitin-like” LC3-I to the lipid phosphatidylethanolamine, giving rise to LC3-II. p62/SQSTM1 is an autophagy receptor that binds the ubiquitinylated cargo and directs it to the growing double-membrane autophagosome by interacting with LC3-II and other related proteins. Finally, the “Completion of the autophagic process” includes two phases: “Fusion” and “Degradation”. During “Fusion”, the mature autophagosome fuses with lysosomes, originating a new vesicle known as autolysosome. The activation of the H⁺ pumps triggers the activation of the lysosomal hydrolases, which are in charge of degrading the cargo. Bafilomycin A₁ inhibits the acidification of the autolysosome by blocking the vacuolar-type H⁺-V-ATPase while Chloroquine impairs the fusion of lysosomes with autophagosomes. Finally, products from degradation reach the cytosol through permeases and enter into metabolic circuitries. All drugs are depicted within squares.

Figure 2

Basal autophagy is blocked in response to autophagy inhibitory drugs. (A) MEFs treated for 6 h with the drugs at the concentrations stated in the panel were stained with monodansylcadaverine (MDC) and observed with a fluorescence microscope. Acidic vesicles are displayed as puncta. Bar = 20 μm. Representative images of at least two independent experiments are shown. (B) Protein extracts of MEFs treated for 3 and 12 h, as indicated in the panel, were analyzed by western blot. “12 h BafA₁” is a control and refers to MEFs treated for 12 h with this drug. Intensity of the LC3-II band, normalized to the loading of each lane (naphtol blue staining) and referred to “C” (untreated control) without BafA₁, was shown. Western blots are the result of a significant experiment out of three independent experiments. A histogram representing “% Basal Autophagic Flux (LC3-II _T+BafA1/LC3-II _T)” at 3 h (white bars) and 12 h (black bars) of treatment was calculated as reported in the “MATERIAL and METHODS” section. The percentages represent the quotient between LC3-II band intensities in “T” and “T+ BafA₁”. “T” is treatment and “T+ BafA₁” is treatment in the presence of BafA₁. Naphthol blue (NB) stained membrane served as a loading control. The percentage of basal autophagic flux is expressed as mean ± SEM of at least three independent experiments (n = 3). Student’s t-test *P < 0.05 and **P < 0.01.

Figure 3

Cells cultured in growing media undergo cell death in response to the autophagy inhibitory compounds, 3-methyladenine, SBI-0206965, Bafilomycin A₁ or Chloroquine. MEFs were treated with the drugs at the concentrations stated in the panel. (A) After 24 h, cell viability was measured by the cellular ability to reduce AB reagent. Bar value is the mean ± SEM (n = 3). Student’s t-test *P < 0.01, **P < 0.005 and ***P < 0.001. (B) After 48 h, the percentage of propidium iodide (PI)-positive cells (dead cells) was determined by flow cytometry. Bar value is the mean ± SEM (n = 3). Student’s t-test **P < 0.005 and ***P < 0.001.

Figure 4

Cytotoxic inhibitors of autophagy engage the mitochondrial pathway of apoptotic cell death. MEFs were treated with the autophagy inhibitory drugs at the concentrations stated in the panel. (A) After 24 h of 3-MA and BafA₁ treatment or after 48 h of SBI and CQ treatment, cells were stained with bisbenzimide 33342 and analyzed by fluorescence microscopy. Arrowheads point to the typical images of apoptotic nuclei. Bar = 40 μm. Images representative of several independent experiments are shown. (B) After 24 and 48 h of treatment, effector caspase activity (DEVDase activity) was quantified in arbitrary fluorescent units (a.f.u.). Bar value is the mean ± SEM (n = 3). Student’s t-test **P < 0.005 and ***P < 0.001. Protein extracts of MEFs treated with 3-MA, BafA₁, SBI, and CQ for 48 h in the presence (+) or absence (−) of 40 μM of q-VD-OPh were analyzed by western blot. Intensity of the cleaved caspase-3 or 120 kDa α-Fodrin bands, normalized to the loading of each lane, was shown. Naphthol blue (NB) stained membrane served as a loading control. The images represent one representative western out of three independent experiments. The antibodies used were (C) anti-caspase-3 and (D) anti-α-Fodrin. (E) MEFs were challenged with the cytotoxic inhibitors of autophagy for the time indicated in the panel in the presence (+QVD) or absence (–QVD) of the caspase inhibitor q-VD-OPh at 40 μM. Bar value is the mean ± SEM (n = 3). Student’s t-test ***P < 0.001. (F) wt MEFs and Bax–/–Bak–/–MEFs (DKO MEFs) were challenged with the cytotoxic autophagy inhibitory drugs for 48 h. Drug concentrations are displayed in the panel. Cell death was quantified by means of PI incorporation and flow cytometry. Data are expressed as mean ± SEM (n = 3). Student’s t-test *P < 0.01, **P < 0.005 and ***P < 0.001.

Figure 5

Effects of autophagy inhibitory drugs on the cellular viability in different human cell lines. HCT116, HEK293, HeLa, and SH-SY5Y cells were treated with the “Initiation” and “Nucleation” autophagy inhibitory drugs from previous experiments. Drug concentrations are indicated in the panel. Cell viability was measured by the AB reducing procedure after 24 h of treatment. Bar value is the mean ± SEM (n = 3). Student’s t-test **P < 0.005 and ***P < 0.001.

Figure 6

Fip200–/– MEFs display increased sensitivity to apoptosis triggered by 3-methyladenine or SBI-0206965, but not to MHY1485, Cpd18, SAR405 or Spautin-1. (A) Protein extracts from Fip200–/– and Fip200+/+ MEFs cultured in full media or Hank’s buffer without glucose (starvation) for 6h, either in the presence (+) or absence (–) of BafA₁, were analyzed by western blot. Autophagy was evaluated by LC3-II and p62. Intensity of these bands, normalized to the loading of each lane and referred to the values of these proteins in Fip200+/+ maintained in growing medium, was shown. Naphthol blue (NB) stained membrane served as a loading control. The image belongs to a representative image out of three independent experiments. (B) Fip200–/– and Fip200+/+ MEFs were challenged with the drugs at the concentrations indicated in the panel. After staining with PI, the percentage of dead cells was determined by flow cytometry. Bar value is the mean ± SEM (n = 3). Student’s t-test *P < 0.01 (C, D) Effector caspase activity (DEVDase activity) was quantified in arbitrary fluorescent units (a.f.u.) after 24 h of treatment with the compounds indicated in the panel. Bar value is the mean ± SEM (n = 3). Student’s t-test *P < 0.01 and ***P < 0.001.

Figure 7

Concentration-dependent cytotoxicity of 3-methyladenine and its structural derivative Cpd18. (A) Chemical structure of 3-MA and Cpd18 (images borrowed from Selleckchem and MerckMillipore webpages, respectively). (B) Western blot of MEFs protein extracts treated for 8 h with growing concentrations of 3-MA and Cpd18 in the presence (+) or absence (–) of 100 nM BafA₁ for the whole treatment. Intensity of the LC3-II band, normalized to the loading of each lane (naphtol blue staining) and referred to “0” without BafA₁, was shown. Western blots are a significant experiment out of three independent experiments. A histogram representing “% Basal Autophagic Flux (LC3-II _T+BafA1/LC3-II _T)” was calculated as reported in the “MATERIAL and METHODS” section. The percentages represent the quotient between LC3-II band intensities in “T” and “T+ BafA₁”. “T” is treatment and “T+ BafA₁”is treatment in the presence of BafA₁. Naphthol blue (NB) stained membrane served as a loading control. The percentage of basal autophagic flux is expressed as mean ± SEM of at least three independent experiments (n = 3). (C) MEFs were treated with the concentrations of 3-MA and Cpd18 indicated in the panel. After 48 h, the percentage of propidium iodide (PI)-positive cells (dead cells) was determined by flow cytometry. Bar value is the mean ± SEM (n = 3).

Figure 8

AKT/PKB and/or mTORC1 inhibition are not the leading mechanisms of 3-methyladenine-mediated cytotoxicity. (A) Protein extracts of MEFs subjected to growing concentrations of wortmannin either in the presence (+) or absence (–) of BafA₁ for 8 h were analyzed by western blot. Autophagy was evaluated by LC3-II. Naphthol blue (NB) stained membrane served as a loading control. The image belongs to a representative image out of two experiments. (B) Protein extracts of control untreated MEFs (C) or MEFs treated with 10 mM 3-MA (3-MA), 100 μM Wortmannin (Wn), 5 μM SAR (SAR), 0.5 mM Cpd18 (Cpd18) at 6 and 12 h were analyzed by western blot with antibodies against phospho-AKT (p-AKT) (Ser473 and 308), AKT1, p-ULK (Ser757), ULK, p-4E-BP1(Thr37/46), and 4E-BP1. Quantifications are the ratios between the intensity of phosphorylated proteins normalized to unphosphorylated proteins. Naphthol blue (NB) stained membrane served as a loading control. The images belong to a representative experiment out of three independent repetitions. (C) MEFs were treated for 48 h with the drugs at concentrations stated in the panel. The percentage of propidium iodide (PI)-positive cells (dead cells) was determined by flow cytometry. Bar value is the mean ± SEM (n = 3).

Figure 9

3-methyladenine triggers the phosphorylation of H2A.X at Ser139. Western blots probed with antibodies against phosphorylated Ser139-H2A.X (γ-H2A.X) and total H2A.X are shown. Quantifications are the ratios between the intensity of γ-H2A.X normalized to total H2A.X. Naphthol blue (NB) stained membrane served as a loading control. (A) Protein extracts of wt MEFs treated with 3-MA at 2.5, 5, and 10 mM for 24 h. Extracts of control cells are depicted as “0”. (B) Protein extracts of Fip200–/– and Fip200+/+ MEFs challenged during 24 h with 10 mM 3-MA or left untreated (C) Protein extracts of MEFs untreated (C) or treated for 24 h with 5 μM SAR, for 20 and 40 h with 20 μM SBI and for 12 and 24 h with 10 mM 3-MA, 100 nM BafA₁ and 50 μM CQ. Images in (A–C) are a representative experiment out of, at least, three independent experiments.