Fig 1: Lysosomal inhibition increases ATF4 accumulation but also limits ATF4 binding to the promoters of UPRmt genes.a Western blot analysis showing time-dependent changes of proteins in MEFs pretreated with DMSO control or ConA (200 nM) for 1 h, and then co-treated with Dox (30 μg/mL) for 0–8 h. All ConA-treated conditions were thus treated with ConA for a total time of 9 h. b Western blot analysis of MEFs treated with ConA (200 nM) or Torin1 (250 nM) for 0–24 h. c Western blot analysis of MEFs treated with cycloheximide (CHX), in the absence or presence of ConA (200 nM) for 0–150 min. d Schematic diagram of the ATF4 translational reporter, comprising the upstream open reading frames (uORF1 and uORF2) of the ATF4 5' untranslated region (5′ UTR) followed by HA-mScarlet tag replacing the ATF4 coding sequence, built on a lentiviral expression vector. The GFP control is directly driven by the cytomegalovirus (CMV) promoter. e Western blot analysis of MEFs stably expressing the ATF4 translational reporter and GFP control treated with or without Dox (30 μg/mL) or Tunicamycin (TM, 1.5 μg/mL) for 3 h, in the presence of DMSO control or ConA (200 nM). f Western blot analysis of HEK293T cells expressing control (ctrl), ATP6V0C (V0C) or ATP6V0D1 (V0D1) sgRNA, and treated with or without Dox (30 μg/mL) for 24 h. The pro and mature forms of Cathepsin B (CTSB) were as indicated. g ATF4 ChIP-qPCR analysis (n = 4 biologically independent samples) of the promoters of ATF4-targeted genes in MEFs pretreated with DMSO or ConA (200 nM) for 1 h, and then co-treated with or without Dox (30 μg/mL) for 3 h. Error bars denote SEM. Statistical analysis was performed by ANOVA followed by Tukey post hoc test (*P < 0.05; **P < 0.01; ***P < 0.001; N.S. not significant).
Fig 2: ATF4 is a direct phosphorylation target of mTOR in response to mitochondrial stress.a ATF4 phosphorylation recognized by a context-dependent (S*P) phosphorylation-specific antibody is increased by Rheb co-expression and inhibited by Torin1. HEK293T cells transfected with the indicated plasmids were immunoprecipitated (IP) with anti-Flag antibody and analyzed by western blot analysis. When applicable, Torin1 (250 nM) were added 2 h before harvest. TCL total cell lysate. b mTOR directly phosphorylates ATF4 in vitro. In vitro kinase assay was performed with recombinant GST-tagged human mTOR purified from baculovirus-infected insect cells and recombinant His-tagged human ATF4 with or without Torin1 (250 nM). Arrows indicate the mobility shifts likely separating the hyperphosphorylated and nonphosphorylated ATF4. c The ratios of the phosphorylated and nonphosphorylated peptides containing the phosphorylation sites of ATF4 from a kinase assay performed similar to (b), as determined by mass spectrometry. d The identified mTOR-targeted phosphorylation sites on ATF4 with the vertebrate orthologs aligned below, with numbering according to the amino-acid sequence of human ATF4 protein. NTD N-terminal domain, BD Basic domain, CLZ C-leucine zipper. The highly conserved putative TOR signaling (TOS) motif was also highlighted. e Validation of the two commercially available antibodies that specifically recognize ATF4 phosphorylation at Ser166 and Thr173, respectively. HEK293T cells transfected with the indicated plasmids were immunoprecipitated with anti-Flag antibody and analyzed by western blot assay. Torin1 (250 nM) was added 2 h before harvest. f Increased ATF4 Ser166 and Thr173 phosphorylation upon Dox treatment, which was inhibited by ConA and Torin1. Wild-type MEFs were pretreated with DMSO, ConA (200 nM) or Torin1 (250 nM) for 1 h, and then co-treated with or without Dox (30 μg/mL) for 2 h, immunoprecipitated with anti-ATF4 antibody and analyzed by western blot assay. g Increased ATF4 phosphorylation upon mitochondrial, but not ER stress inducers. Wild-type MEFs were with treated with Antimycin A (AntiA, 2 μM), Oligomycin (Olig, 2 μM), or Tunicamycin (TM, 1.5 μg/mL) for 2 h, immunoprecipitated with anti-ATF4 antibody and analyzed by western blot assay. A similar amount of immunoprecipitated ATF4 protein was loaded for different conditions to compare phosphorylation changes in (f, g).
Fig 3: An essential role of ATF4 and its phosphorylation by mTORC1 in UPRmt activation.a qRT-PCR results (n = 4 biologically independent samples) of wild-type (WT) and Atf4−/− MEFs treated with or without Dox (30 μg/mL) or Antimycin A (AntiA, 2 μM) for 24 h. b The OCR of WT or Atf4−/− MEFs at basal or after sequential addition of Oligomycin (Olig), FCCP and AntiA/Rotenone. The basal and maximum OCR was statistically analyzed (n = 6 biologically independent samples). c Atf4 knockout leads to disrupted mitochondrial network upon mitochondrial stress. MitoTracker staining of WT or Atf4−/− MEFs treated with or without Dox (30 μg/mL) or AntiA (2 μM) for 24 h. The average mitochondrial network perimeter and area were analyzed by ImageJ with a Mito-Morphology macro (n = 3 independent experiments). Scale bar, 10 μm. d Western blot analysis of Atf4−/− MEFs stably expressing empty vector (vector), the wild-type ATF4 (WT-ATF4), the phospho-defective mutant (5A-ATF4), and an ATF4 mutant carrying a point mutation of the bulky phenylalanine residue 94 in the TOS motif to alanine (F94A-ATF4). e qRT-PCR results (n = 4 biologically independent samples) of Atf4−/− MEFs stably expressing vector, wild-type, 5A or F94A forms of ATF4, treated with or without Dox (30 μg/mL) for 24 h. Error bars denote SEM. Statistical analysis was performed by two-tailed unpaired Student’s t test in (b), or by ANOVA followed by Tukey post hoc test in (a, c, e) (**P < 0.01; ***P < 0.001; N.S. not significant).
Fig 4: Models of how mitochondrial stress and ER stress activate mechanistically different pathways, involving the lysosomes, v-ATPase/mTORC1, ATF4, and/or ribosomes, to concordantly activate the UPRmt and the UPRER.Left: in response to mitochondrial stress, mTORC1 is activated at the lysosomal surface and EIF2α phosphorylation is mildly increased, leading to a moderate increase in ATF4 translation. Meanwhile, activated mTORC1 directly phosphorylates ATF4, leading to increased ATF4 binding to the promoters of UPRmt genes and UPRmt activation. Right: in response to ER stress, mTORC1 activity is suppressed but the EIF2α phosphorylation is robustly increased, leading to an increase in translation and nuclear accumulation of ATF4, and to the subsequent activation of the ER UPR (UPRER). CQ chloroquine, ConA/BafA1 Concanamycin A/Bafilomycin A1.
Fig 5: mTORC1-mediated ATF4 phosphorylation sustains mitochondrial homeostasis and protects cells from ROS-associated cell death upon mitochondrial stress.a MitoTracker staining of Atf4−/− MEFs stably expressing wild-type (WT), 5A or F94A forms of ATF4, with or without Dox (30 μg/mL) or AntiA (2 μM) treatment for 24 h. The average mitochondrial network perimeter and area were analyzed by ImageJ with a Mito-Morphology macro (n = 3 independent experiments). Scale bar, 10 μm. b The OCR of Atf4−/− MEFs stably expressing WT, 5A or F94A forms of ATF4, after DMSO control (ctrl) or AntiA (2 μM) treatment for 24 h. The basal and maximum OCR was analyzed (n = 6 biologically independent samples). c Flow cytometry analysis of the mitochondrial superoxide (MitoSOX) intensity of Atf4−/− MEFs stably expressing WT, 5A or F94A forms of ATF4, after DMSO control or AntiA (2 μM) exposure for 48 h. The percentages of MitoSOX-positive cells were analyzed (n = 3 independent experiments). d Representative bright field photographs of Atf4−/− MEFs stably expressing empty vector, WT, 5A or F94A forms of ATF4, grown with or without the antioxidant β-mercaptoethanol (β-ME) or AntiA (2 μM) for 96 h. Mean percentages (±SEM) of the survival ratio of cells are indicated (n = 3 independent experiments). Error bars denote SEM. Statistical analysis was performed by ANOVA followed by Tukey post hoc test (**P < 0.01; ***P < 0.001; N.S. not significant).
Supplier Page from Abcam for Recombinant Human ATF-4 protein (His tag N-Terminus + Calmodulin tag N-Terminus)