CHOP induces activating transcription factor 5 (ATF5) to trigger apoptosis in response to perturbations in protein homeostasis - PubMed (original) (raw)

CHOP induces activating transcription factor 5 (ATF5) to trigger apoptosis in response to perturbations in protein homeostasis

Brian F Teske et al. Mol Biol Cell. 2013 Aug.

Abstract

Environmental stresses that disrupt protein homeostasis induce phosphorylation of eIF2, triggering repression of global protein synthesis coincident with preferential translation of ATF4, a transcriptional activator of the integrated stress response (ISR). Depending on the extent of protein disruption, ATF4 may not be able to restore proteostatic control and instead switches to a terminal outcome that features elevated expression of the transcription factor CHOP (GADD153/DDIT3). The focus of this study is to define the mechanisms by which CHOP directs gene regulatory networks that determine cell fate. We find that in response to proteasome inhibition, CHOP enhances the expression of a collection of genes encoding transcription regulators, including ATF5, which is preferentially translated during eIF2 phosphorylation. Transcriptional expression of ATF5 is directly induced by both CHOP and ATF4. Knockdown of ATF5 increases cell survival in response to proteasome inhibition, supporting the idea that both ATF5 and CHOP have proapoptotic functions. Transcriptome analysis of ATF5-dependent genes reveals targets involved in apoptosis, including NOXA, which is important for inducing cell death during proteasome inhibition. This study suggests that the ISR features a feedforward loop of stress-induced transcriptional regulators, each subject to transcriptional and translational control, which can switch cell fate toward apoptosis.

PubMed Disclaimer

Figures

FIGURE 1:

FIGURE 1:

Genome-wide analysis of CHOP-dependent genes regulated by proteasome inhibition. (A) The scatter plot is the log ratio of WT (_x_-axis) and _CHOP_−/− MEF cells (_y_-axis) treated with 1 μM MG132 for 8 h versus no stress. This comparative analysis of gene transcripts illustrates all significant probe sets (p < 0.05) induced or repressed upon MG132 treatment. (B) Pie charts generated by IPA analysis, which highlight functional classes of _CHOP_-dependent genes. (C) IPA comparison of _CHOP_-dependent genes classified by role in cells.

FIGURE 2:

FIGURE 2:

CHOP is required for enhanced ATF5 mRNA levels. WT or _CHOP_−/− MEF cells were treated with 1 μM MG132 (A) or 1 μM thapsigargin (TG; B) for 8 h, and the levels of ATF4, CHOP, ATF5, and GADD34 mRNAs were determined by qPCR. *Statistical significance (p < 0.05) with respect to the untreated control. #Significant difference (p < 0.05) between cell types. n.d., not detectable.

FIGURE 3:

FIGURE 3:

CHOP facilitates transcriptional activation of ATF5 by binding to CARE elements in the ATF5 promoter. (A) Diagram of potential CARE sequences present in the ATF5 promoter region, as determined by in silico analysis. The values −2.5 and −3.0 kb are relative to the transcriptional start site. Cells were treated with MG132 (B), thapsigargin (TG; C), or no treatment (NT), and chromatin immunoprecipitation was performed with antibodies specific for ATF4, CHOP, or nonspecific IgG as a negative control. qPCR was performed, and the enrichment relative to input DNA is shown for each of three genomic regions. The regions amplified by primer sets P1 and P2 include the possible CARE sequences, whereas the primer set designated P3 amplified a region located in exon 3 that does not contain a CARE site and serves as a negative control.

FIGURE 4:

FIGURE 4:

CHOP is required for ATF5 protein expression in response to proteasome inhibition. (A) WT and _CHOP_−/− MEF cells were treated with 1 μM MG132 for up to 24 h as indicated. Time 0 indicates no stress treatment. Levels of eIF2α∼P, eIF2α, ATF4, CHOP, ATF3, ATF5, caspase-3 (CASP-3 and the cleaved version), cleaved PARP, and actin were determined by immunoblot analysis using antibody specific to the protein of interest. (B) _CHOP_−/− cells and those stably expressing CHOP from an FRT locus were exposed to 1 μM MG132 for up to 24 h. Levels of CHOP, ATF4, ATF5, and actin proteins were determined by immunoblot analysis. Levels of ATF5 (C) and GADD34 (D) mRNAs were measured by qPCR in WT, _CHOP_−/−, and rescued _CHOP_−/− cells that were stably expressing CHOP. Cells were treated with 1 μM MG132 for up to 12 h or not treated (NT). (E) HEPA1-6 cells stably expressing a control scrambled (Scr) shRNA or a _CHOP_-targeted shRNA were treated with 1 μM MG132 for 8 h. In lanes 3–5, the RNA Consortium (TRC) ID number indicates the last two digits for the validated _CHOP_-targeted shRNA identification number obtained from Sigma-Aldrich. Levels of CHOP, ATF5, and actin proteins were determined by immunoblot analysis. (F) WT and _CHOP_−/− MEF cells were treated for 24 h with 1 μM MG132 or no treatment (NT) and subjected to FACS analyses using annexin V–FITC/PI staining. *Statistical significance (p < 0.05) compared with the untreated control. #Significant difference (p < 0.05) between WT and _CHOP_−/− cell types.

FIGURE 5:

FIGURE 5:

ATF5 facilitates apoptosis in response to proteasome inhibition. (A) RNA was isolated from MEF cells stably expressing a scrambled shRNA or an ATF5-targeted shRNA, and ATF5 mRNA levels were determined by qPCR relative to non-shRNA–expressing control. The ID number indicates the last two digits for the shRNA. The percentages of ATF5 knockdown were determined relative to a non-shRNA–expressing control. (B) WT, _ATF5_-KD1, and _ATF5_-KD2 MEF cells were treated with MG132 for the indicated time points. Levels of ATF4, CHOP, ATF5, cleaved caspase-9, cleaved caspase-3, cleaved PARP, and actin proteins were measured by immunoblot analysis using specific antibodies. (C) Cell treatment regimen for the data shown in D. Cells were cultured with 1 μM MG132 for 0, 1, 3, 6, 12, or 18 h as indicated by the black bars, and after the stress treatment, cells were cultured in the absence of MG132, as indicated by the gray bars, for a total of 24 h. (D) Measurements of survival of WT, _CHOP_−/−, _ATF5_-KD1, and _ATF5_-KD2 MEF cells were determined by the MTT assay, and the percentage of cell survival is normalized to the untreated controls for each cell type.

FIGURE 6:

FIGURE 6:

Genome-wide analysis of gene expression indicates that ATF5 is required for activation of proapoptotic target genes. Results from a microarray analysis of WT and ATF5-KD2 cells treated with MG132 or no stress indicated those genes that significantly required ATF5 for full induction in response to proteasome inhibition. These ATF5-target genes were then analyzed for functional groupings and compared with those genes determined in our microarray analysis to be statistically dependent on CHOP for full induction in response to MG132 treatment. (A) Pie charts generated by IPA analysis, highlighting functional classes of _ATF5_-dependent genes. (B) Venn diagram indicating the number of genes that required only CHOP for full induction response to MG132 treatment, those genes dependent on both CHOP and ATF5, and genes dependent on only ATF5. (C) Comparative bar graph showing that GADD34 required only CHOP for full induction in response to proteasome inhibition, examples of genes whose induction showed a statistically significant dependence on both CHOP and ATF5, and those genes requiring only ATF5; values represent mean fluorescence intensity (MFI) in WT and _ATF5_-KD2 cells upon MG132 treatment or no treatment (NT).

FIGURE 7:

FIGURE 7:

shRNA knockdown of NOXA protects cells from stress induced by MG132 treatment. (A) WT and _ATF5_-KD2 MEF were treated with 1 μM MG132 for 8 h or no treatment (NT). The mRNA levels for CHOP, ATF5, NOXA, TXNIP, and APAF1 were measured by qPCR. In addition, NOXA mRNA was measured in _ATF5_-KD1 cells as indicated. *Statistical significance (p < 0.05) with respect to the untreated control. #Difference between cell types. (B) MEF cells expressing an shRNA targeting NOXA or scramble control were treated for 8 h with MG132. The last two digits in the validated shRNA identification number are listed. Relative NOXA mRNA levels were measured by qPCR. *Statistical significance (p < 0.05). (C) WT, _CHOP_−/−, and _NOXA_-KD MEF cells were treated with 1 μM MG132 for 24 h, and cell death was determined as the percentage of viable cells as measured by MTT assay upon MG132 treatment compared with untreated controls. (D) WT and _TXNIP-_KD MEF cells were treated with 1 μM MG132 for 12 h, and the relative level of TXNIP mRNA was measured by qPCR. (E) Death of WT and TXNIP-KD cells was measured as the percentage of viable cells as determined by MTT assays after 24 h of MG132 treatment compared with untreated controls.

FIGURE 8:

FIGURE 8:

The ISR features a network of ATF4, CHOP, and ATF5 transcription factors in a feedforward loop that controls cell fate. Different stress conditions that disrupt protein homeostasis induce phosphorylation of eIF2, leading to enhanced expression of the bZIP transcription factors ATF4, CHOP, and ATF5 through increased translation, as indicated by the green arrows. ATF4 and CHOP bind the ATF5 promoter and serve to enhance the transcription of ATF5, and CHOP and ATF5, individually or in combination, induce the transcription of target genes (blue arrows). These target gene products affect feedback control, apoptosis, and inflammation, which together can determine cell fate. ATF4 may also serve as a direct contributor to the transcription of some of these target genes, as illustrated by the finding that ATF4, as well as CHOP, can directly bind to the GADD34 promoter to promote transcription (Ma and Hendershot, 2003; Marciniak et al., 2004; Kilberg et al., 2012).

References

    1. Angelastro JM, Mason JL, Ignatova TN, Kukekov VG, Stengren GB, Goldman JE, Greene LA. Downregulation of activating transcription factor 5 is required for differentiation of neural progenitor cells into astrocytes. J Neurosci. 2005;25:3889–3899. - PMC - PubMed
    1. Baird TD, Wek RC. Eukaryotic initiation factor 2 phosphorylation and translational control in metabolism. Adv Nutr. 2012;3:307–321. - PMC - PubMed
    1. Bruhat A, Averous J, Carraro V, Zhong C, Reimold AM, Kilberg MS, Fafournoux P. Differences in the molecular mechanisms involved in the transcriptional activation of the CHOP and asparagine synthetase genes in response to amino acid deprivation or activation of the unfolded protein response. J Biol Chem. 2002;277:48107–48114. - PubMed
    1. Chen C, Pan YX, Dudenhausen EE, Kilberg MS. Amino acid deprivation induces the transcription rate of the human asparagine synthetase gene through a timed program of expression and promoter binding of nutrient-responsive basic region/leucine zipper transcription factors as well as localized histone acetylation. J Biol Chem. 2004;279:50829–50839. - PubMed
    1. Chen JJ. Regulation of protein synthesis by the heme-regulated eIF2alpha kinase: relevance to anemias. Blood. 2007;109:2693–2699. - PMC - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources