The ISR promotes cellular survival signaling by negative regulation of cell death pathways, such as apoptosis. For instance, as a consequence of ER stress, PERK‐induced activation of the ISR results in the expression of cIAP1 and cIAP2 (cellular inhibitor of apoptosis proteins) in tumor and non‐tumor cells (Hamanaka et al., 2009; Hu et al., 2004; Warnakulasuriyarachchi et al., 2004). Previously, it was demonstrated that restoration of the function of cIAP1 or cIAP2 in PERK^−/− murine embryonic fibroblasts during ER stress delays the early onset of ER stress-induced caspase activation and apoptosis seen in these cells (Figure 2) (Hamanaka et al., 2009).

ATF4 has also been demonstrated to facilitate anti-neoplastic agent bortezomib‐induced upregulation of anti‐apoptotic myeloid cell leukemia-1 (Mcl-1) protein, which is an anti-apoptotic Bcl-2 family protein that plays essential roles in multiple myeloma survival and drug resistance in many tumor types (Figure 2) (Hu et al., 2012).

It has been shown that both MCL‐1 and cIAPs can suppress apoptosis at different points in the apoptosis pathway that are upstream and downstream of the release of cytochrome c from the mitochondria. Mitochondrial cytochrome c plays a dual function in controlling both cellular energetic metabolism and apoptosis. It has been shown that, upon interacting with apoptotic protease activating factors (Apaf), cytochrome c can trigger the activation cascade of caspases once it is released from the mitochondria into the cytosol (Cai et al., 1998).

It has also been reported that miR‐211 is a pro-survival microRNA that regulates CHOP expression in a PERK‐dependent manner and thus PERK can mediate a pro‐survival function by suppressing a stress‐dependent expression of CHOP consequently leading to re‐establishment of cellular homeostasis before the initiation of apoptosis (Chitnis et al., 2012). In addition to its beneficial roles in restoring homeostasis, these ISR mechanisms may also contribute to tumor development. For example, an increased miR‐211 expression, found to be PERK-dependent, and was reported in mammary carcinoma and mouse models of human B‐cell lymphoma (Figure 2) (Chitnis et al., 2012).

Cancer cells use multiple stress response pathways such as the integrated stress response (ISR), cytosolic heat shock response (HSR), and unfolded protein response (UPR) mediated by organelles such as the endoplasmic reticulum (ER) and mitochondria to respond exogenous and endogenous or environmental stresses to evade apoptosis, ensure survival, proliferation, metastatic potential, and maintain cellular homoeostasis (OʼMalley et al., 2020). For example, to evade apoptosis and ensure survival, cancer cells may utilize the mitochondrial unfolded protein response (UPRmt) pathway and associated key proteins including chaperones HSP10, HSP60, and mtHSP70 and proteases ClpP and LONP1 to eliminate proteotoxic stress (Figure 2) (OʼMalley et al., 2020). Notably, upregulation of HSP60 expression and its upstream regulator ATF5 has been shown to enhance the apoptotic threshold in cancer cells resulting in therapeutic resistance in many cancer types. ATF-5 has been reported to regulate expression of Egr-1, BCL-2, and MCL1 to mediate proliferation and survival in cancer (Dluzen et al., 2011; Liu et al., 2011; Karpel-Massler et al., 2016).

Moreover, in addition to the genes mentioned above many other genes activated in response to ISR (Costa-Mattioli and Walter, 2020), including those encoding ATF4, ATF5 (Zhou et al., 2008); CHOP (C/EBP-homologous protein) (Palam et al., 2011); GADD34 (Growth Arrest And DNA-Damage-Inducible 34) (Lee et al., 2009); and in neurons, OPHN1 (Oligophrenin-1) (Di Prisco et al., 2014), other genes such as IBTKα (the α isoform of inhibitor of Bruton’s tyrosine kinase) (Baird et al., 2014) and NUPR1 (Nuclear protein-1), also play important roles in cell survival. NUPR1 has been found to play an important role in cell stress and stress-related apoptosis (Martin et al., 2021) and inactivation of NUPR1 promotes cell death by coupling ER-stress responses with necrosis (Santofimia-Castaño et al., 2018). More evidences suggest that ATF4 initiates the activity of transcription factor NUPR1. NUPR1 regulates the expression of several metabolic stress‐responsive genes, in particular, genes required in cell cycle regulation and DNA repair, as such, NUPR1 also is regarded as pro‐survival factors (Figure 2) (Jin et al., 2009; Hamidi et al., 2012).

Another gene activated during the ISR is the IBTKα which is activated during ER stress. IBTKα is a major substrate adaptor for protein ubiquitination and is an essential pro‐survival factor (Baird et al., 2014).

Likewise, eIF2α mediated translational repression has been suggested in activated B cell NF‐κB pathway induction as a mechanism to protect cells against ER stress (Deng et al., 2004). In a recent study, a pharmacologically activable version of PERK was used to uncouple eIF2α phosphorylation from stress and it was determined that eIF2α phosphorylation is both required and adequate to activate both NF‐κB DNA binding and an NF‐κB reporter gene (Deng et al., 2004). Also, HRI has been shown to be involved in NF‐κB activation (Abdel-Nour et al., 2019). This study found that the eIF2α kinase HRI controls NOD1 (Nucleotide-binding oligomerization domain-containing protein 1) signalosome folding and activation through a process requiring eIF2α, ATF4, and the heat shock protein HSPB8 (Abdel-Nour et al., 2019). Moreover, HRI/eIF2α signaling pathway was shown to be required for signaling downstream of the innate immune mediators including NOD2, MAVS (Mitochondrial antiviral-signaling protein), and TRIF (TIR-domain-containing adapter-inducing interferon-β) but dispensable for signaling pathways that rely on MyD88 (Myeloid differentiation primary response 88) or STING (Stimulator of interferon genes) (Figure 2) (Abdel-Nour et al., 2019).

Autophagy is a highly regulated eukaryotic cellular pathway that plays a major role in the lysosomal degradation of cytoplasmic unfolded proteins, peptides, damaged organelles or cytosolic components while also serving as a means to replenish depleted amino acids for building proteins and to provide energy to a starved cell. Autophagy can be activated by a variety of cellular stresses such as nutrient or growth factor deprivation, hypoxia, reactive oxygen species, DNA damage, protein aggregates, damaged organelles, or intracellular pathogens (Pakos-Zebrucka et al., 2016; Clementi et al., 2020; Akman et al., 2021). Autophagy can be activated both via specific, stimulus-dependent manner and more general, stimulus-independent signaling pathways to coordinate different phases of autophagy.

The ISR can modulate cell survival and cell death pathways through the activation of autophagy and the phosphorylation of eIF2α at S51 appears to be essential for stress‐induced autophagy (Pakos-Zebrucka et al., 2016). Autophagy can be integrated with other cellular stress responses through parallel stimulation of autophagy and other stress responses by specific stress stimuli, through dual regulation of autophagy and other stress responses by multifunctional stress signaling molecules, and/or through mutual control of autophagy and other stress responses.

The cell death pathways are complex and can be exploited by cancer therapeutic agents (Carneiro and El-Deiry, 2020). When stress persists and cells are unable to reach homeostasis despite the activation of stress response pathways, ATF4 can induce the transcriptional activation of apoptotic genes encoding CHOP (DDIT3) (Harding et al., 2000), TRB3 (Tribbles homolog 3) (Ohoka et al., 2005), and pro-apoptotic BH3-only proteins including PUMA (p53 upregulated modulator of apoptosis), Noxa (Phorbol-12-myristate-13-acetate-induced protein 1) and BIM (Bcl-2 Interacting mediator of cell death), thus leading to cell death (Galehdar et al., 2010; Altman et al., 2009; Puthalakath et al., 2007). ATF4 has been shown to regulate Noxa at the transcriptional level and this leads to the induction of apoptosis (Sharma et al., 2018; Núñez-Vázquez et al., 2021). Overall, through the induction of ATF4, this transcription factor appears to mainly trigger the intrinsic apoptosis by modulating the expression of pro- and anti-apoptotic BCL-2 family members. Interestingly, in the case of CHOP activation, induction of DR5 (Death receptor 5) mediated apoptosis appeared to be DR5 ligand binding independent and involving the engagement of FADD (Fas-associated protein with death domain) and caspase-8 (Figure 2) (Lu et al., 2014; Li et al., 2015).

Additional stresses such as those resulting from decreased mitochondrial translation (Sasaki et al., 2020) as well as the generation of reactive oxygen species (Kasai et al., 2019) have been shown to induce ATF4 expression. In the case of sustained mitochondrial deficiency, ATF4 response has been reported to lead to p53-mediated apoptosis (Evstafieva et al., 2014). Reactive oxygen species generated by Fenretinide treatment in neuroblastoma cells activates ATF4 leading to the induction of Noxa ultimately leading to apoptosis (Nguyen et al., 2019). In multiple myeloma cells, sensitivity to bortezomib treatment was associated with higher expression of ATF4 and loss of its expression lead to lower levels of Noxa, CHOP and DR5 (Narita et al., 2015). Recent work from our lab has also implicated ATF4 responsible for the induction of p53-target genes PUMA, Noxa, NAG-1(Nonsteroidal anti-inflammatory drug-activated gene-1)and DR5 upon treatment with prodigiosin analogue PG3-Oc (Figure 2) (Tian et al., 2021).

The aforementioned studies involve the induction of the ISR machinery in addition to distinct components of autophagy, cell cycle, and/or apoptosis pathway. This reflects the complexity of the interplay of these cellular pathways which remains underscored and likely to be context-dependent. Recent work has focused on post-translational modifications of ATF4 and how these affect the transcriptional control and cellular response. ATF4 has numerous sites that can be post-translationally modified including phosphorylation at various threonine and serine sites, methylation at arginine 239, and ubiquitination and acetylation at lysine residues (Wortel et al., 2017). These post-translational modifications affect ATF4 protein stability, activation and interaction with other proteins. In the case of apoptosis, methylation at arginine 239 by methyl transferase PRMT1 was found to be associated with the transcription of genes related to apoptosis (Yuniati et al., 2016). Further insight into ATF4 activation may shed light on understanding the context of how these transcription factors respond to stress and the biological outcome they ultimately trigger in both normal and cancer cells. Importantly, this will aid the intervention of novel therapies, the use of the ISR as potential biomarker for predicting therapy response and the combination of therapies that induce ATF4-mediated apoptosis. An example of therapy combination has been observed in in vivo neuroblastoma preclinical models with the BCL-2 inhibitor Venetoclax and Fenretinide (Nguyen et al., 2019). This studied combination highlighted the use of BCL-2 expression as a biomarker for neuroblastoma patients. A separate study in multiple myeloma suggested the use of ATF4 as a predictive therapy response biomarker for bortezomib and dexamethasone combination treatment (Narita et al., 2015). These studies exemplified the clinical translational applicability of exploiting the ISR in cancer therapy and highlight its warrant understanding to predict cancer types that will benefit from ISR modulating therapies.

The ISR plays different roles in tumorigenesis and tumor progression in different types of tumors. Hypoxia is a common phenomenon in solid tumors. It may induce apoptosis of tumor cells or tumor cells may develop the ability to adapt to the hypoxia or anoxic environment. Hypoxia can induce ISR gene expression in transformed mouse embryonic fibroblasts and the activated ER stress response confers resistance to apoptosis induced by hypoxia and thus facilitates tumor growth (Ameri et al., 2004). ISR mediator ATF4 is induced by anoxia in breast cancer cell lines (Ameri et al., 2004). The activated ISR plays an essential role in the adaptation to hypoxic stress allowing tumor cell survival under stress and is associated with resistance to therapy (Blais et al., 2004; Rouschop et al., 2013).

It was found that loss of extracellular matrix (ECM) attachment stimulates ISR signaling in vitro. And the activation of ISR further plays a critical role in resistance to anoikis and is required for metastasis (Dey et al., 2015). The ISR also has impact on the tumor microenvironment. Tumor cells undergoing ER stress can transmit ER stress to myeloid cells contributing to a pro-inflammatory tumor microenvironment, thus facilitating tumor progression (Mahadevan et al., 2011).

The role of ISR may be complex in tumors. In medulloblastoma, the ISR is activated, and the decreased ISR via gene manipulation attenuates medulloblastoma formation. Moderately enhanced ISR by gene manipulation noticeably increased the incidence of medulloblastoma, whereas a strongly enhanced ISR significantly decreased the incidence of medulloblastoma in vivo. Thus, the ISR plays dual roles in medulloblastoma formation (Stone et al., 2016).

Activation of the ISR is correlated with resistance to chemotherapy in pancreatic cancer and BRAF-mutated melanoma. Gemcitabine can induce ISR and the antiapoptotic pro-survival factors via the ISR pathway in pancreatic cancer cell line and the combination of gemcitabine + ISRIB which inhibits ISR induce more apoptosis in vivo (Palam et al., 2015). In BRAF-mutated melanoma, chronic ER stress involving induction of the ISR signaling pathway activates autophagy which contributes chemoresistance (Corazzari et al., 2015).

Triggering ISR can be a therapeutic strategy against cancer, since the ISR can induce apoptosis. ONC201 kills solid tumors by triggering ISR-dependent ATF4 activation and activation of the TRAIL-DR5 apoptotic pathway (Kline et al., 2016). In breast cancer, GBM and DMG cell lines, ONC201 induces ISR, TRAIL-DR5 and ultimately apoptosis (Zhang et al., 2021). The apoptosis increases with the enhancement of ISR induction by tazemetostat. The knockdown of ATF4 in GBM cell line reduced the apoptosis induced by ONC201 and the combination of ONC201 with tazemetostat or vorinostat remarkably. Therefore, induction of ISR can play an essential role in cell death of cancer cells. Apoptosis induced by ISR activation was also observed in AML cells (Ishizawa et al., 2016).

The combination of mitochondrial uncoupler niclosamide ethanolamine and dopamine receptor antagonist domperidone or TCAs induces ISR and leas to apoptosis in multiple cancer cell lines including CRC, GBM (Glioblastoma multiforme) and PDAC (Pancreatic ductal adenocarcinoma) cell lines (Hartleben et al., 2021). Even without inducing apoptosis, the ISR is induced by ONC201 in cancer cells exhibiting decreased cell proliferation (Kline et al., 2016).

The ISR contributes to drug sensitivity of cancer cells. Activation of the ISR in HER2+breast cancer contributes the sensitivity to Trastuzumab in vivo. Increased expression of the ISR mediator eIF2α-P predicts a better response of patients with HER2+ metastatic breast cancer to Trastuzumab therapy (Darini et al., 2019). Proteasome inhibitors are known to activate the ISR and lower expression of ISR markers thus implicating shorter progression-free survival in multiple myeloma (Obeng et al., 2006).

It was reported that ISR promotes the expression of potential target for immunotherapy (Obiedat et al., 2020). Thus, ISR may play a role in cancer immunotherapy.

On the one hand, activation of ISR plays a role in cancer therapy. On the other, Inhibition of ISR activation can increase the vulnerability of cancer cells. BCR-ABL inhibition prevents activation of ISR in K562 cell line derived from a chronic myeloid leukemia (CML) patient and makes the tumor cells more vulnerable to metabolic stress (Kato et al., 2018). Summaries of the mentioned cases and drugs can be found in the Table 1, Table 2 and Figure 3.

The ISR takes a dual role in cell survival and cell death. Enhance or inhibition of ISR signaling via targeting ISR components is a promising strategy for cancer therapy (Figure 3). Among the components in ISR signaling, eIF2α is a core component and an important focused for cancer therapy.