Early growth response-1 is a regulator of DR5-induced apoptosis in colon cancer cells.

BACKGROUND: Tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) induces tumour cell apoptosis by binding to death receptor 4 (DR4) and DR5. DR4 and DR5 activation however can also induce inflammatory and pro-survival signalling. It is not known how these different cellular responses are regulated and what the individual role of DR4 vs DR5 is in these processes. METHODS: DNA microarray study was carried out to identify genes differentially expressed after DR4 and DR5 activation. RT-PCR and western blotting was used to examine the expression of early growth response gene-1 (Egr-1) and the proteins of the TRAIL signalling pathway. The function of Egr-1 was studied by siRNA-mediated knockdown and overexpression of a dominant-negative version of Egr-1. RESULTS: We show that the immediate early gene, Egr-1, regulates TRAIL sensitivity. Egr-1 is constitutively expressed in colon cancer cells and further induced upon activation of DR4 or DR5. Our results also show that DR4 mediates a type II, mitochondrion-dependent apoptotic pathway, whereas DR5 induces a mitochondrion-independent, type I apoptosis in HCT15 colon carcinoma cells. Egr-1 drives c-FLIP expression and the short splice variant of c-FLIP (c-FLIP(S)) specifically inhibits DR5 activation. CONCLUSION: Selective knockdown of c-FLIP(S) sensitises cells to DR5-induced but not DR4-induced apoptosis and Egr-1 exerts an effect as an inhibitor of the DR5-induced apoptotic pathway, possibly by regulating the expression of c-FLIP(S).

Differential activation of JNK1 isoforms by TRAIL receptors modulate apoptosis of colon cancer cell lines.

Tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) induces apoptosis on binding to its receptors, death receptor 4 and 5 (DR4, DR5). TRAIL can also activate c-Jun N-terminal kinase (JNK) through the adaptor molecules, TNF receptor-associated factor 2 (TRAF2) and receptor-interacting protein (RIP). The role of JNK in TRAIL-induced tumour cell apoptosis is unclear. In this study, we demonstrate that JNK is activated by TRAIL in colon cancer cells. Inhibition of JNK with L-JNKI reduced rhTRAIL-induced cell death but enhanced cell death induced by selective activation of DR4 or DR5. This difference was unrelated to receptor internalisation or differential activation of c-Jun, but activation of different JNK isoforms. Our data demonstrate that JNK1, but not JNK2 is activated by rhTRAIL in the examined colon cancer cell lines. Although rhTRAIL activated both the long and short isoforms of JNK1, selective activation of DR4 or DR5 led to predominant activation of the short JNK1 isoforms (JNK1alpha1 and/or JNK1beta1). Knockdown of JNK1alpha1 by shRNA enhanced apoptosis induced by TRAIL, agonistic DR4 or DR5 antibodies. On the other hand, knockdown of the long JNK1 isoforms (JNK1alpha2 and JNK1beta2) had the opposite effect; it reduced TRAIL-induced cell death. These data indicate that the short JNK1 isoforms transmit an antiapoptotic signal, whereas the long isoforms (JNK1alpha2 or JNK1beta2) act in a proapoptotic manner.

Distinct effects of high-glucose conditions on endothelial cells of macrovascular and microvascular origins.

Recent studies implicate hyperglycemia as an important cause of macrovascular and ocular complications in diabetes mellitus. In this study, the authors examined the effect of high glucose on macrovascular and microvascular endothelial cell viability and apoptosis in culture. Human aortic endothelial cells (HAECs) and human retinal endothelial cells (HRECs) were exposed to normal-glucose conditions (NG) and high-glucose conditions (NG supplemented with 25 mM D-glucose) for 72 h in vitro. D-Mannitol was used as an osmotic control. Cell viability was assessed by methlythiazolydiphenyltetrazolium bromide (MTT) assay, and induction of apoptosis was assessed by Hoechst staining. Statistics were analyzed by paired t tests. In HAECs, cell viability was decreased by 12.9% in high-glucose conditions, and apoptotic cells were significantly increased by 77%. However, in HRECs, cell viability was increased by 14.9% in high-glucose conditions, and apoptotic cells were significantly decreased by 33.3%. Mannitol did not show any effect on cell survival or apoptosis ruling out an osmotic effect. High-glucose conditions reduce cell viability and induce apoptosis in HAECs, which may contribute to macrovascular complications associated with diabetes. In contrast, high-glucose increases viability in HRECs and inhibits apoptosis, which may contribute to the development of diabetic retinopathy.

Decoy receptors block TRAIL sensitivity at a supracellular level: the role of stromal cells in controlling tumour TRAIL sensitivity.

Tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) is a death ligand cytokine known for its cytotoxic activity against malignantly transformed cells. TRAIL induces cell death through binding to death receptors DR4 and DR5. The inhibitory decoy receptors (DcR1 and DcR2) co-expressed with death receptor 4 (DR4)/DR5 on the same cell can block the transmission of the apoptotic signal. Here, we show that DcRs also regulate TRAIL sensitivity at a supracellular level and thus represent a mechanism by which the microenvironment can diminish tumour TRAIL sensitivity. Mathematical modelling and layered or spheroid stroma-extracellular matrix-tumour cultures were used to model the tumour microenvironment. By engineering TRAIL to escape binding by DcRs, we found that DcRs do not only act in a cell-autonomous or cis-regulatory manner, but also exert trans-cellular regulation originating from stromal cells and affect tumour cells, highlighting the potent inhibitory effect of DcRs in the tumour tissue and the necessity of selective targeting of the two death-inducing TRAIL receptors to maximise efficacy.

Caspases: their intracellular localization and translocation during apoptosis.

The activation of the caspase family of proteases has been detected in numerous cell systems and appears to function as a common pathway through which apoptotic mechanisms may operate. Caspases are synthesized as precursors (pro-caspases) and are converted into mature enzymes by apoptotic signals. The effects of caspases in apoptosis are accomplished by the cleavage of numerous proteins located in different intracellular compartments. In the present study we have addressed the question of the subcellular localization of different pro- and active caspases as well as several other proteins, such as Apaf-1, calpain and DFF, which also play important roles in the apoptotic process. We found that at least three pro-caspases (pro-caspases-2, -3 and -9) were present in both the mitochondrial and cytosolic fractions of untreated Jurkat T lymphocytes. Only pro-caspase-2 was found in the nuclear fraction. Pro-caspases-7 and -8 were found only in the cytosolic fraction. In apoptotic cells, caspases-3, -8 and -9 were present in the cytosolic fraction, whereas caspases-3 and -9 were also found in the mitochondrial fraction and caspase-7 in the microsomal fraction. Caspases-2 and -3 were present in the nuclear fraction. The selective localization of pro-caspases in different subcellular compartments may play an important, but yet unknown, role in their activation. The translocation of active caspases to other subcellular compartments appears to be critical for the development of the apoptotic process.

Cleavage of the calpain inhibitor, calpastatin, during apoptosis.

Calpain activity is thought to be essential for the execution of apoptotic cell death in certain experimental models. In the present study, the physiological inhibitor of calpain, calpastatin, was found to be cleaved in three different apoptotic systems. The 110-120 kDa calpastatin protein of Jurkat T-lymphocytes and U937 monocytic leukemia cells was cleaved to a 65-70 kDa form after the induction of apoptosis with anti-CD95 monoclonal antibody, staurosporine or TNF. Cleavage of calpastatin in apoptotic cells occurred simultaneously with the cleavage of the DNA repair enzyme, poly(ADP-ribose) polymerase. The caspase inhibitors VAD-cmk and IETD-fmk prevented calpastatin cleavage in all three systems. Calpain inhibitor I, however, suppressed calpastatin cleavage only during TNF-induced apoptosis. Other protease inhibitors, such as lactacystin and pepstatin A, did not confer any significant protection against apoptotic calpastatin cleavage. The results from in vitro incubations with cell lysates and purified enzymes showed that calpain I, calpain II and recombinant caspase-3, all cleaved calpastatin, with varying efficiency. In conclusion, the results of the present study suggest that caspases may cleave calpastatin and thus, regulate calpain activity during apoptotic cell death.

The ability to cleave 28S ribosomal RNA during apoptosis is a cell-type dependent trait unrelated to DNA fragmentation.

Discrete cleavages within 28S rRNA divergent domains have previously been found to coincide with DNA fragmentation during apoptosis. Here we show that rRNA and DNA cleavages can occur independently in apoptotic cells, i.e. that the previously observed correlation is likely to be coincidental. In HL-60 cells, apoptosis with massive DNA fragmentation could be induced without any signs of rRNA cleavage. The opposite situation; rRNA cleavage without concomitant internucleosomal DNA fragmentation, was found in okadaic acid-treated Molt-4 cells. Other leukemia cell lines underwent apoptosis either without (K562 and Molt-3) or with (U937) both forms of polynucleotide cleavage. In K562 cells transfected with a temperature-sensitive p53 mutant, internucleosomal DNA fragmentation but not 28S rRNA cleavage was inducible by wild-type p53 expression. The absence of apoptotic rRNA cleavage in some cell types suggests that this phenomenon is tightly regulated and unrelated to DNA fragmentation or a presumed scheme for general macromolecular degradation in apoptotic cells.

Endoplasmic reticulum stress induces ligand-independent TNFR1-mediated necroptosis in L929 cells.

Endoplasmic reticulum (ER) stress-induced cellular dysfunction and death is associated with several human diseases. It has been widely reported that ER stress kills through activation of the intrinsic mitochondrial apoptotic pathway. Here we demonstrate that ER stress can also induce necroptosis, an receptor-interacting protein kinase 1 (RIPK1)/RIPK3/mixed lineage kinase domain-like protein (MLKL)-dependent form of necrosis. Remarkably, we observed that necroptosis induced by various ER stressors in L929 cells is dependent on tumor necrosis factor receptor 1 (TNFR1), but occurs independently of autocrine TNF or lymphotoxin α production. Moreover, we found that repression of either TNFR1, RIPK1 or MLKL did not protect the cells from death but instead allowed a switch to ER stress-induced apoptosis. Interestingly, while caspase inhibition was sufficient to protect TNFR1- or MLKL-deficient cells from death, rescue of the RIPK1-deficient cells additionally required RIPK3 depletion, indicating a switch back to RIPK3-dependent necroptosis in caspase-inhibited conditions. The finding that ER stress also induces necroptosis may open new therapeutic opportunities for the treatment of pathologies resulting from unresolved ER stress.

Triggering and modulation of apoptosis by oxidative stress.

Cell survival requires multiple factors, including appropriate proportions of molecular oxygen and various antioxidants. Although most oxidative insults can be overcome by the cell's natural defenses, sustained perturbation of this balance may result in either apoptotic or necrotic cell death. Numerous, recent studies have shown that the mode of cell death that occurs depends on the severity of the insult. Oxidants and antioxidants can not only determine cell fate, but can also modulate the mode of cell death. Effects of oxidative stress on components of the apoptotic machinery may mediate this modulation. This review will address some of the current paradigms for oxidative stress and apoptosis, and discuss the potential mechanisms by which oxidants can modulate the apoptotic pathway.

Detection of pro-caspase-3 in cytosol and mitochondria of various tissues.

Caspases are a family of cysteine proteases of critical importance in the apoptotic cell death process. They are normally present as zymogens (pro-caspases) in the cytoplasm of vertebrate and other organisms. In this study we have shown that pro-caspase-3 is localized to cytosol and mitochondria of various rat tissues (brain, heart, kidney, liver, spleen and thymus). Although the majority of pro-caspase-3 was localized in the cytosol, the amount of mitochondrial pro-caspase-3 was significant. The ratio of cytosolic and mitochondrial pools of pro-caspase-3 appeared to vary between different tissues. The higher amount of mitochondrial pro-caspase-3 was found in thymus and spleen, i.e. tissues in which spontaneous apoptosis plays an important role. Our findings provide further support for mitochondrial localization of pro-caspase-3 and the critical role of this organelle in apoptosis.

Thermotolerance and cell death are distinct cellular responses to stress: dependence on heat shock proteins.

We tested the hypothesis that heat shock protein (Hsp) induction and cell death are mutually exclusive responses to stress. Despite activation of heat shock transcription factor 1 at temperatures ranging from 40 to 46 degrees C, Hsp72 and Hsp27 were not induced above 42 degrees C. Moreover, cells underwent apoptosis at 44 degrees C and necrosis at 46 degrees C, with mitochondrial cytochrome c release at both temperatures. However, only apoptosis was associated with caspase activation. Treatment of cells with z-VAD-fmk prior to heat shock at 44 degrees C failed to restore Hsp induction despite inhibition of heat-induced apoptosis. Furthermore, accumulation of Hsps after incubation at 42 degrees C rendered the cells resistant to apoptosis. These results suggest that lack of Hsp induction is the cause rather than the consequence of cell death.

Antioxidant-mediated inhibition of the heat shock response leads to apoptosis.

We examined the hypothesis that reactive oxygen species (ROS) contribute to the induction of heat shock proteins (hsps) during stress response. Exposure of HL-60 human myelocytic cells to 42 degrees C induced both hsp72 and hsp27. In the presence of the antioxidant molecules pyrrolidine dithiocarbamate or 1,10-phenanthroline induction of hsp72 and 27 was significantly decreased, while N-acetyl-L-cysteine caused a slight reduction. Prevention of hsp induction was associated with heat sensitization and increased caspase activity, indicating that the cells were undergoing apoptosis. These data suggest that ROS contribute to the induction of hsps and furthermore, that hsp induction and apoptosis are mutually exclusive events within the same cell.

Hsp27 protects mitochondria of thermotolerant cells against apoptotic stimuli.

Enhanced cell survival and resistance to apoptosis during thermotolerance correlates with an increased expression of heat shock proteins (Hsps). Here we present additional evidence in support of the hypothesis that the induction of Hsp27 and Hsp72 during acquired thermotolerance in Jurkat T-lymphocytes prevents apoptosis. In thermotolerant cells, Hsp27 was shown to associate with the mitochondrial fraction, and inhibition of Hsp27 induction during thermotolerance in cells transfected with hsp27 antisense potentiated mitochondrial cytochrome c release after exposure to various apoptotic stimuli, despite the presence of elevated levels of Hsp72. Caspase activation and apoptosis were inhibited under these conditions. In vitro studies revealed that recombinant Hsp72 more efficiently blocked cytochrome c-mediated caspase activation than did recombinant Hsp27. A model is presented for the inhibition of apoptosis during thermotolerance in which Hsp27 preferentially blocks mitochondrial cytochrome c release, whereas Hsp72 interferes with apoptosomal caspase activation.

Hsp27 inhibits cytochrome c-mediated caspase activation by sequestering both pro-caspase-3 and cytochrome c.

Mitochondrial cytochrome c release in response to pro-apoptotic signals leads to the formation of a cytochrome c/Apaf-1/procaspase-9 complex (the apoptosome) and resultant activation of caspase-9 and caspase-3. Here we demonstrate that the molecular chaperone, Hsp27, inhibits this cytochrome c-mediated activation of caspase-3. Immunodepeletion of Hsp27 from cytochrome c-activated cytosols resulted in decreased caspase activity. Furthermore, immunoprecipitation of Hsp27 resulted in the coprecipitation of both cytochrome c and procaspase-3. In reciprocal experiments, immunoprecipitation of both procaspase-3 and cytochrome c resulted in coprecipitation of Hsp27, indicating two independent interactions. These results point to Hsp27 mediating its inhibition of procaspase-3 activation through its ability to sequester both cytochrome c and procaspase-3, and thus prevent the correct formation/function of the apoptosome complex.

Interleukin-1, interleukin-8, tumour necrosis factor alpha and interferon gamma stimulate DNA synthesis but have no effect on apoptosis in small-intestinal cell lines.

OBJECTIVES: Cytokines stimulate lymphocyte cell proliferation and affect cell division in several other cell types. Helicobacter pylori-induced gastritis and coeliac disease are characterized by an increased cell proliferation in association with an increased production of proinflammatory cytokines, which could contribute to these cell kinetic changes. Our aim is to examine in vitro whether cytokines usually present in the gastrointestinal mucosa affect DNA synthesis and apoptosis in a rat and a human small-intestinal cell line. METHODS: IEC-6 and FHs-74 cells were incubated for 24 h with 10(-13)-10(-9) M of tumour necrosis factor alpha (TNF-alpha), interleukin-1beta (IL-1beta), interleukin-2 (IL-2), interleukin-6 (IL-6), interleukin-8 (IL-8), transforming growth factor-beta (TGF-beta) and interferon gamma (IFN-gamma). IEC-6 cells were also incubated with 10(-13)-10(-9) M of interleukin-1alpha (IL-1alpha) and 10(-8) M of interleukin-1 receptor antagonist (IL-1ra). The cells were labelled with 3H-methyl thymidine for the final 4 hours, and then processed for autoradiography. DNA synthesis was evaluated by the labelling index (LI%). Apoptosis was evaluated in IEC-6 cells by changes in membrane lipid asymmetry using annexin-V binding to externalized phosphatidylserine (flow cytometry) and by estimating the caspase activity. RESULTS: TNF-alpha, IL-1beta, IL-8 and IFN-gamma significantly and markedly increased the LI, even at low concentrations (P< 0.0001), in both IEC-6 and FHs-74 cells, as did IL-1alpha in IEC-6 cells. TGF-beta significantly reduced the LI in both cell lines (P< 0.0001), whereas IL-2, IL-6 and IL-1ra did not affect DNA synthesis significantly. None of IL-1beta, IL-8, TNF-alpha or IFN-gamma affected apoptosis in IEC-6 cells. CONCLUSION: TNF-alpha, IL-1alpha, IL-1beta, IL-8 and IFN-gamma stimulated DNA synthesis in a human and a rat small-intestinal cell line. The cytokines exert their mitogenic action directly on the intestinal cells via specific receptors. Our findings indicate that pro-inflammatory cytokines may participate in the regulation of the gastrointestinal epithelial cell proliferation in health and disease.

Nerve growth factor-mediated inhibition of apoptosis post-caspase activation is due to removal of active caspase-3 in a lysosome-dependent manner.

Nerve growth factor (NGF) is well characterised as an important pro-survival factor in neuronal cells that can inhibit apoptotic cell death upstream of mitochondrial outer membrane permeabilisation. Here we addressed the question of whether NGF can also protect against apoptosis downstream of caspase activation. NGF treatment promoted a rapid reduction in the level of the p17 subunit of active caspase-3 in PC12 cells that had been induced to undergo apoptosis by various cytotoxins. The mechanism involved TrkA-dependent activation of extracellular signal-regulated kinase (ERK1/2) but not phosphatidylinositol 3-kinase (PI3K)/Akt, and de novo protein synthesis. Involvement of inhibitor of apoptosis proteins (IAPs) and proteasomal degradation were ruled out. In contrast, inhibition of lysosome function using chloroquine and concanamycin A reversed NGF-induced removal of p17. Moreover, in NGF-treated cells, active caspases were found to be localised to lysosomes. The involvement of macroautophagy and chaperone-mediated autophagy were ruled out. Taken together, these findings suggest an anti-apoptotic mechanism by which NGF induces removal of active caspase-3 in a lysosome-dependent manner.

RIPK1 promotes death receptor-independent caspase-8-mediated apoptosis under unresolved ER stress conditions.

Accumulation of unfolded proteins in the endoplasmic reticulum (ER) causes ER stress and results in the activation of the unfolded protein response (UPR), which aims at restoring ER homeostasis. However, when the stress is too severe the UPR switches from being a pro-survival response to a pro-death one, and the molecular mechanisms underlying ER stress-mediated death have remained incompletely understood. In this study, we identified receptor interacting protein kinase 1 (RIPK1)-a kinase at the crossroad between life and death downstream of various receptors-as a new regulator of ER stress-induced death. We found that Ripk1-deficient MEFs are protected from apoptosis induced by ER stressors, which is reflected by reduced caspase activation and PARP processing. Interestingly, the pro-apoptotic role of Ripk1 is independent of its kinase activity, is not regulated by its cIAP1/2-mediated ubiquitylation, and does not rely on the direct regulation of JNK or CHOP, two reportedly main players in ER stress-induced death. Instead, we found that ER stress-induced apoptosis in these cells relies on death receptor-independent activation of caspase-8, and identified Ripk1 upstream of caspase-8. However, in contrast to RIPK1-dependent apoptosis downstream of TNFR1, we did not find Ripk1 associated with caspase-8 in a death-inducing complex upon unresolved ER stress. Our data rather suggest that RIPK1 indirectly regulates caspase-8 activation, in part via interaction with the ER stress sensor inositol-requiring protein 1 (IRE1).

Resistance to TRAIL in non-transformed cells is due to multiple redundant pathways.

Tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) is a cytokine and a selective inducer of apoptosis in a range of tumour cells, but not in normal, untransformed cells. A large number of chemotherapeutics as well as biological agents are being tested for their potential to sensitise resistant tumour cells to TRAIL as a means to broaden the range of tumours treatable with TRAIL. However, because of the incomplete understanding of the mechanism(s) underlying TRAIL resistance in non-malignant cells, it is unpredictable whether the effect of these sensitisers will be restricted to tumour cells or they would also sensitise non-transformed cells causing unwanted toxicity. In this study, we carried out a systematic analysis of the mechanisms driving TRAIL resistance in non-transformed cells. We found that cellular FLICE-like inhibitory protein, anti-apoptotic B-cell lymphoma 2 proteins, and X-linked inhibitor of apoptosis protein were independently able to provide resistance to TRAIL. Deficiency of only one of these proteins was not sufficient to elicit TRAIL sensitivity, demonstrating that in non-transformed cells multiple pathways control TRAIL resistance and they act in a redundant manner. This is contrary to the resistance mechanisms found in tumour cell types, many of them tend to rely on a single mechanism of resistance. Supporting this notion we found that 76% of TRAIL-resistant cell lines (13 out of 17) expressed only one of the above-identified anti-apoptotic proteins at a high level (≥1.2-fold higher than the mean expression across all cell lines). Furthermore, inhibition or knockdown of the single overexpressed protein in these tumour cells was sufficient to trigger TRAIL sensitivity. Therefore, the redundancy in resistance pathways in non-transformed cells may offer a safe therapeutic window for TRAIL-based combination therapies where selective sensitisation of the tumour to TRAIL can be achieved by targeting the single non-redundant resistance pathway.

Perk-dependent repression of miR-106b-25 cluster is required for ER stress-induced apoptosis.

Activation of the unfolded protein response sensor PKR-like endoplasmic reticulum kinase (Perk) attenuates endoplasmic reticulum (ER) stress levels. Conversantly, if the damage is too severe and ER function cannot be restored, this signaling branch triggers apoptosis. Bcl-2 homology 3-only family member Bim is essential for ER stress-induced apoptosis. However, the regulatory mechanisms controlling Bim activation under ER stress conditions are not well understood. Here, we show that downregulation of the miR-106b-25 cluster contributes to ER stress-induced apoptosis and the upregulation of Bim. Hypericin-mediated photo-oxidative ER damage induced Perk-dependent cell death and led to a significant decrease in the levels of miRNAs belonging to miR-106b-25 cluster in wild-type (WT) but not in Perk⁻/⁻ MEFs. Further, we show that expression of miR-106b-25 and Mcm-7 (host gene of miR-106b-25) is co-regulated through the transcription factors Atf4 (activating transcription factor 4) and Nrf2 (nuclear factor-erythroid-2-related factor 2). ER stress increased the activity of WT Bim 3'UTR (untranslated region) construct but not the miR-106b-25 recognition site-mutated Bim 3'UTR construct. Overexpression of miR-106b-25 cluster inhibits ER stress-induced cell death in WT but did not confer any further protection in Bim-knockdown cells. Further, we show downregulation in the levels of miR-106b-25 cluster in the symptomatic SOD1(G86R) transgenic mice. Our results suggest a molecular mechanism whereby repression of miR-106b-25 cluster has an important role in ER stress-mediated increase in Bim and apoptosis.

PERK is required at the ER-mitochondrial contact sites to convey apoptosis after ROS-based ER stress.

Endoplasmic reticulum stress is emerging as an important modulator of different pathologies and as a mechanism contributing to cancer cell death in response to therapeutic agents. In several instances, oxidative stress and the onset of endoplasmic reticulum (ER) stress occur together; yet, the molecular events linking reactive oxygen species (ROS) to ER stress-mediated apoptosis are currently unknown. Here, we show that PERK (RNA-dependent protein kinase (PKR)-like ER kinase), a key ER stress sensor of the unfolded protein response, is uniquely enriched at the mitochondria-associated ER membranes (MAMs). PERK(-/-) cells display disturbed ER morphology and Ca(2+) signaling as well as significantly weaker ER-mitochondria contact sites. Re-expression of a kinase-dead PERK mutant but not the cytoplasmic deletion mutant of PERK in PERK(-/-) cells re-establishes ER-mitochondria juxtapositions and mitochondrial sensitization to ROS-mediated stress. In contrast to the canonical ER stressor thapsigargin, during ROS-mediated ER stress, PERK contributes to apoptosis twofold by sustaining the levels of pro-apoptotic C/EBP homologous protein (CHOP) and by facilitating the propagation of ROS signals between the ER and mitochondria through its tethering function. Hence, this study reveals an unprecedented role of PERK as a MAMs component required to maintain the ER-mitochondria juxtapositions and propel ROS-mediated mitochondrial apoptosis. Furthermore, it suggests that loss of PERK may cause defects in cell death sensitivity in pathological conditions linked to ROS-mediated ER stress.