IRE1 (Inositol-requiring enzyme 1) is a transmembrane protein of ER (type I), with serine/threonine kinase activity, detecting ER stress through its N-terminal luminal domain, starting also the most common UPR signalling pathway ^[1][22]. IRE1 has two isoforms: IRE1α and IRE1β.

IRE1α is the most studied and is present in all eukaryotic cells. In case of unfolded proteins accumulation, IRE1 oligomerizes in the ER lumen and it starts the autophosphorylation ^[2][23].

After the activation, IRE1 cuts XBP1 mRNA, causing a shift in its codon reading frame, this condition triggers the formation of a new C-terminal domain which includes an active domain of transactivation, sXBP1 ^[3][4][5][6][4,5,24,25].

sXBP1, in turn, provokes the upregulation of UPR-related genes implicated in protein folding, and in translocation to the ER and ERAD ^[7][8][26,27].

IRE1 enlists TRAF2, and triggers ASK1 ^[9][28]. ASK1 causes the activation of JNK and p38, a MAPK ^[9][10][28,29].

Then, JNK molecules move to the membrane of mitochondria leading to Bim activation and the inhibition of Bcl-2. On the other hand, the phosphorylation of p38 MAPK provokes the activation of a transcriptional factor CHOP, causing the increasing of Bim and DR5 expression, at the same time decreasing Bcl-2 expression, this condition leads to the apoptosis initiation ^[11][12][30,31].

Bax and Bak can bind to IRE1 and trigger it, interact with IP₃R inducing the release of Ca²⁺ from the ER ^[13][32].

The attenuation of mRNA translation is induced by an ER-resident transmembrane protein called PERK (protein kinase R-like endoplasmic reticulum kinase), which works as sensor of ER stress and contains a luminal domain similar to IRE1 ^[14][15][33,34].

Usually, PERK is linked to GRP78, and, after its activation, it blocks the entrance of newly synthesized proteins into the ER (which is already stressed). After the estrangement from GRP78, PERK forms a dimer and induces its autophosphorylation and activation ^[16][35].

This condition is possible for the inactivation of elF2 by the phosphorylation of serine 51 ^[17][13]. The inhibition of elF2α is caused by a guanine nucleotide exchange factor complex that leads elF2 to its active GTP-bound form ^[18][36], this last reduces the excess of misfolded proteins and mitigates ER stress ^[19][11].

Moreover, elF2 phosphorylation plays another role; in fact, it allows the translation of UPR-dependent genes, for example ATF4, codifying for different upstream open reading frames ^[20][21][37,38]. ATF4 triggers the expression of ER stress target genes (like CHOP, growth arrest, GADD34 and ATF3) ^[22][23][39,40].

ATF6 (activating transcription factor 6) is a transmembrane protein of ER (type II), in case of ER stress conditions, dissociates from GRP78 (the release from GRP78 enables ATF6) and moves to the Golgi apparatus for additional proteolytic processing ^[24][25][41,42]. Into the Golgi, two enzymes called site proteases-1 and 2 (S1P-S2P) make the proteolytic cleavage of the full length ATF6 (90 kDa) ^[26][27][43,44].

Then, the cleaved N-terminal cytosolic domain of 50-kDa bZIP (cytosolic basic leucine zipper) moves to the nucleus binding to CRE (the ATF/cAMP response elements) and ERSE-1 to induce the transcription of some target protein like GRP78, XBP-1, and CHOP ^[28][29][45,46]. Thereby, during a prolonged ER stress, CHOP can be activated by IRE1, PERK, and ATF6, and then it leads to apoptosis ^[30][31][47,48].

The cleaved ATF6 translocate into the nucleus and works as an active transcription factor to upregulate proteins that improve ER folding capacity (for example two chaperons like GRP78 and GR94), and folding enzymes (for example PDI) ^{[28][32][33][34]}[45,49,50,51].

GRP78, also known as BiP, is a member of the HSP70 family. It is localized on the membrane of ER of all eukaryotes ^[35][52]. GRP78 is composed of 654 amino acids. It edits the folding and assembly, and also avoids the transport of protein (or subunits of them) that are misfolded ^[36][37][38][53,54,55]. The expression of GRP78 is increased in case of ER stress. For example, GRP78 is upregulated during

GRP78 is soluble in water, with only small hydrophobic parts; however, these last components are essential for its function, for example to recognize the unfolded proteins addressed either to the degradation or refolding mechanisms ^[40][57]. GRP78 possesses two domains called ABD (or NBD) locates at the N-terminal and SBD at the C-terminal ^[41][58].

There is 60% homology between GRP78 and the HSP70 family, and in particular, the most conserved domains are ABD and SBD. The most conserved sequence of the HSP70 family belongs to ABD domain ^[35][41][42][52,58,59]. However, GRP78 differs in protein expression regulation, nevertheless it is an HSP70 family A member, and owns the properties of abnormal protein binding in case of stress. In addition, the protein synthesis inhibitor cycloheximide is effective on GRP78 ^[43][44][60,61] (cycloheximide can inhibit protein synthesis in eukaryotic cells) ^[45][62].

The state of ER protein folding is checked by three ER-localized transmembrane UPR signal sensors, such as PERK, IRE1, and the transcription factor ATF6.

Under homeostatic conditions, these UPR sensors are maintained inactive by the interaction of their luminal domains with GRP78.

In response to ER stress, accumulated unfolded proteins sequester GRP78 from the UPR sensors, which promotes the activation of IRE1, PERK, and ATF6 and the induction of GRP78 ^[6][46][25,63].

UPR acts through the transitional (at transcriptional and transcriptional level) attenuation of the enzymes folding, the induction of ER chaperones, and ERAD involved proteins to relieve protein aggregation in the ER as an adaptive response.

In case of important and extended ER stress, the UPR triggers some apoptotic pathways ^[33][50]. ER stress induces conformational change in ER membrane through the proapoptotic BH3 protein; moreover, it allows Ca²⁺ transmigration to the cytosol, which provokes the activation of m-calpain and, subsequently, the cleavage and the activation of procaspase 12 and caspase cascade ^{[47][48][49][50]}[64,65,66,67].

CHOP, one of the principal UPR downstream effectors, inhibits Bcl-2, provoking growth arrest and the activation of GADD34 and ERO1, and all these elements promote apoptosis ^[51][52][68,69].

The upregulation of GADD34 by CHOP causes a feedback inhibition of elF2α phosphorylation; as a consequence, in cell death and survival, the role of CHOP may be context dependent. This phenomenon could allow the restoration of translation, which could be positive; however, if translation continues also in ER-stress conditions, the accumulation of anomalous proteins may compromise the ER folding ability, causing the death of cells.

After IRE1 activation, JNK is bound by IRE1 which recruits TRAF2; this condition causes both the release of the procaspase 12 from the ER and the activation of apoptosis signal-regulating kinase 1 and JNK ^[9][53][28,70]. Additionally, PUMA and NOXA are activated by ER stress, and this condition provokes BAX and BAK activation and apoptosis ^[54][71].

Activation and maintenance of representative UPR pathways in cells treated with low concentrations of chemical ER stress inducers, showed that survival, in case of stress, is reached through intrinsic mRNA instabilities and proteins that induce apoptosis ^[55][72].

Autophagy is a catabolic process, leading to several effects, in particular, the degradation and recycling of cytosolic, ageing, or misfolded proteins and excess or faulty organelles. ER stress provokes autophagy and promotes cell survival; in fact, in case of starving conditions, it enables the intracellular resources utilization ^[56][73].

In cells where GRP78 is reduced by small interfering RNA (siRNA), the ER structure is compromised, and autophagosome formation under ER-stress or starvation is suppressed ^[57][74].

The decrease of ER Ca²⁺ causes protein misfolding and chronic mitochondrial Ca²⁺ overload, which lead to apoptosis through Bcl-2-dependent pathway ^[58][75]. The localization and oligomerization of pro-apoptotic Bcl-2 proteins, Bax and Bak, are triggered by ER stress, promoting Ca²⁺ release from the ER into the cytosol ^[33][50], through IP₃Rs and RyRs ^[50][59][67,76], which are linked to the apoptotic signal transduction mechanism ^[60][61][62][77,78,79].

When [Ca²⁺]_cyt increases, it leads to the activation of Ca²⁺-dependent cysteine protease m-calpain, which is involved in several intracellular processes, for example apoptosis, cell cycle progression, differentiation, and signal transduction ^[63][64][80,81].

m-calpain is well-known to cleave and trigger the ER-resident procaspase-12 ^[47][65][64,82], involved in the ER stress-induced cell death pathway in differentiated PC12 cells ^[66][83]. The activation of caspase-12 leads to the activation of procaspase-9 and then to the activation of caspase-3 apoptotic mechanism ^[48][65].

The increasing of cytosolic Ca²⁺ causes its uptake into the mitochondrial matrix, inducing a depolarization of the inner mitochondrial membrane and a perturbation of the outer membrane permeability ^[16][35]. This leads to the cytochrome c release and to the activation of the apoptosome by Apaf-1, causing apoptosis ^[67][84].

The most important actor in the regulation of the ER stress-induced apoptosis is CHOP ^[68][85]. CHOP is a basic leucine zipper-containing transcription factor, suppressing the expression of Bcl-2 and triggering the transcription of many genes favouring apoptosis ^[52][69][69,86]. The release of procaspase-12 from TRAF2 (and then its activation) is triggered by the association of IRE1/TRAF2 and ER stress ^[70][71][87,88]. The activated caspase-12 results in apoptosis activation ^[53][70].

The ER is closely associated with mitochondria through mitochondria associated membranes (MAM), leading to a strong functional interaction between these two organelles ^[72][89]. For example, the acute ER stress impairs mitochondrial function in rat and mouse hearts, while inhibition of mitochondrial respiration by using rotenone or antimycin A also increases the ER stress ^[73][90].

ER is a key site of intracellular calcium storage, and ER stress leads to intracellular calcium overload by disrupting calcium homeostasis ^[74][91]. ER stress-mediated calcium overload contributes to mitochondrial damage. Furthermore, in response to Ca²⁺ overload, the ER stress also rises ROS production directly through NADPH oxidase 4, and then through impairment of mitochondrial electron transport ^[75][92].

Taken together, these data point out the existence of a positive feedback loop between mitochondrial dysfunction and ER stress.

Moreover, the activation of initial phase of UPR, i.e., the pathways to solve the ER stress by expanding the ER, upregulating chaperones and by causing a (temporary) translation stop, is accompanied by ER morphology changes and the consolidation of ER and mitochondria contact sites.

In fact, considering that the folding of newly proteins is one of most energy-requiring processes, this strengthening of the contact sites makes sense. In ER stress-induced by tunicamycin, an important number of mitochondria relocate towards the perinuclear ER. These mitochondria exhibit a transmembrane potential increase, Ca²⁺ uptake rise, higher ATP production, and increased oxygen consumption ^[76][93]. Therefore, tightening the ER-mitochondrial contacts could favour the increase in intracellular ATP level necessary to sustain the pro-survival ER machinery.

Conversely, if ER stress is too severe and cannot be resolved by the UPR, the signalling pathways initiated by the ER stress sensors will turn on a lethal signal, ultimately causing cell death, typically in an apoptotic way ^[77][94]. Several studies have also validated a direct relationship between changes in MAM components and deregulated Ca²⁺ transfer and apoptotic sensitivity during ER stress ^[78][79][95,96].

Likewise, new data suggested that the MAM could be involved not only in Ca²⁺ signals, but also in some toxic lipid oxidation products to the mitochondria ^[80][97]. This phenomenon could be very important as lipid peroxides may be responsible for spreading ROS signalling, to control mitochondrial Ca²⁺ uptake and cell death choice after ER stress ^[81][98].

Additionally, to increase importance of MAM involvement in ER stress, recent studies displayed that ER-mitochondria contacts may be crucial in the formation of autophagosomes ^[82][99].

One of tumorigenesis hallmark is the unrestrained growth of the transformed cells; cancers are continuously challenged by a limited oxygen and nutrients supply due to inadequate vascularisation. Moreover, some hematopoietic cancers often show that secretory proteins have increased production, such as multiple myeloma cells producing immunoglobulins ^[83][100]. All these circumstances induce an ER stress and UPR activation.

Different cancer types are linked to ER protein folding machinery disturbance, demonstrating as the correct folding process is a key for signalling pathway proteins ^[84][101]. A huge amount of evidence demonstrates how the UPR is a crucial process which is very important for cancer cells to keep malignancy and drug resistance.

UPR is a very active research area because different components are triggered or repressed in malignant cancers ^[85][102]. Recent indications suggest that UPR molecular components could be useful as prognostic and diagnostic markers for cancer progression and response to chemotherapeutics ^[77][94].

In a great variety of cancers, UPR pathways are activated, and they are essential for tumour microenvironment creation and maintenance. In fact, in cancer cells a great number of events can take place, such as:

-

over-expression of XBP1;

-

activation of ATF6;

-

phosphorylation of eIF2α;

-

induction of ATF4 and CHOP

-

upregulation of GRP78;

-

upregulation of glucose-regulated protein 94 (GRP94, also known as gp96 or HSP90b1);

-

upregulation of GRP170 ^[86][87][103,104].

From animal studies, it was discovered that XBP1 is important for tumour growth in vivo. In fact, in Xbp1-/- and Xbp1-knockdown mice cells, cancers cannot develop ^[88][105]. In addition, ER stress can cause anti-apoptotic reactions. After XBP1 activation, glycogen GSK3b ^[89][106] induces p53 phosphorylation, which causes an increasing of its degradation and avoids p53 dependent apoptosis for cancer cells. Furthermore, during ER stress, NFκB is activated and induces anti-apoptotic responses ^[90][107].

An important role is played by heat shock proteins; in fact, they help cancer cell adaptation against stress associated with oncogenesis both to repair damaged proteins (protein refolding) and to degrade them. Moreover, heat shock proteins are involved in cell proliferation and in resistance to different anti-tumoral drugs that cause apoptosis. For example, HSP90 interfaces with different key proteins in inducing prostate cancer progression, comprised both wild-type and mutated AR, HER2, ErbB2, Src, Abl, Raf, and Akt ^[91][92][108,109].

In a great variety of tumours, GRP78 is largely expressed at high levels conferring resistance against therapies in both proliferating and dormant cancer cells. Animal models with reduced GRP78 level shown significant impediments in tumour growth.

GRP78 mediated-cancer progression has been explained though three major mechanisms:

-

activation of cancer proliferation;

-

suppression of apoptosis;

-

stimulation of angiogenesis ^[93][94][110,111].

ER stress has been involved in different levels of tumour development. The current idea is that, during early tumorigenesis and before angiogenesis occurs, activation of the UPR causes the cell cycle arrest in G1 phase and the activation of p38, both of which induce a dormant state.

ER stress also promotes anti-apoptotic NF-κB and silences p53-dependent apoptotic signals. If the balance of early cancer development takes off against cell death, ER stress can further induce the aggressive growth of cancer cells by enhancing their angiogenic ability. For example, the induction of GRP170 (which is a BiP-like protein that acts as a chaperone for VEGF ^[95][112]) causes an increasing of VEGF secretion.

Emerging data indicate that agents disturbing UPR pathway may be utilized as promising anticancer drug. However, due to the dual role of the UPR in cell survival/death, and depending on the type of cancer, molecules that either provoke severe ER stress and cell death or compounds blocking the pro-survival role of disturbed UPR of tumour cells could be used either alone or in combination with conventional anticancer treatments ^[96][113].

In different kinds of cancer, the expression of GRP78 is often elevated if compared to healthy tissues. This has been noticed in different diseases such as hepatocellular carcinoma ^[97][114], gliomas ^[98][115], prostate ^[99][116], and gastric cancers ^[100][117]. However, a study on lung cancer showed a link between the overexpression of GRP78 and better prognosis ^[101][118]. Despite this, GRP78 results overexpressed in more cases, and, for this reason, the consensus is that its expression is linked to poor prognosis and strong tumour aggressiveness.

For example, it is well-known that an important expression of GRP78 is correlated with poor prognosis and lymph node metastasis in gastric tumours ^[100][117]; moreover, these data are reinforced by preclinical studies which show how the silencing of GRP78 reduces invasion in vitro and tumour growth and metastasis in vivo ^[100][117].

In prostate tumours, high GRP78 activity is linked to a reduction of patient survival ^[99][116], while, in breast tumours ^[102][103][119,120], a shorter time to recidivism is associated with high GRP78 expression ^[103][120].

In addition, a GRP78 overexpression has been described both in tumour with acquired anti-oestrogen resistance and in oestrogen-receptor positive breast cancer cells ^[102][119]. However, a study reported a correlation between oestrogen receptor positivity and GRP78 or XBP1 expression, showing that the oestrogen upgrade can promote the expression of GRP78 and XBP1 ^[104][121].

Moreover, GRP78 might contribute significantly to therapy resistance. In glioma cells and breast cancer cells, the inhibition of GRP78 expression improves sensitivity to chemotherapy ^[98][105][115,122]; in particular, in resistant breast cancer cells, it re-establishes sensitivity to anti-oestrogens ^[102][119]. On the other hand, GRP78 overexpression induces resistance to chemotherapeutics ^{[97][98][104][106]}[114,115,121,123] and anti-oestrogens ^[97][102][114,119].

GRP78-mediated therapy resistance is not a general rule; in renal carcinoma cells, GR78 knockdown causes resistance to chemotherapy ^[107][124].

Nevertheless, GRP78 is a possible target for cancer therapies. Chen et al. described the use of the GRP78-promoter to induce the expression of an apoptotic pathway ^[108][109][125,126]. Another strategy is the direct attack against GRP78 expressing cells. For example, EGCG ((−)-epigallocatechin-3-gallate), a green tea polyphenol, can block GRP78 binding on its ATP-binding domain. A treatment with EGCG can damage glioma cells resistance to temozolomide ^[98][115]. EGCG is also effective to reduce the proliferation of mesothelioma cell lines, overexpressing GRP78 ^[110][127].

An alternative is versipelostatin, a transcriptional inhibitor of GRP78, which causes selective death of glucose-deprived cells and stops cancer development in vivo ^[111][128]. XPB1 and ATF4 expression are repressed by the treatment with this compound, but only during glucose deprivation, not after treatment with tunicamycin or A23187. Combining versipelostatin with cisplatin induces growth inhibition ^[112][129].

The evidence indicates that GRP78 plays different roles in cancer cells, not only in the ER ^[112][129]. In fact, some isoforms of GRP78 have been detected in the nucleus, mitochondria, cytosol, and also in the ER membrane ^{[106][113][114][115][116]}[123,130,131,132,133].

Expression of GRP78 on the cell surface could be a great therapeutic target using specific therapeutic antibodies. An example is PAT-SM6, an autoantibody isolated from a gastric cancer patient ^[116][133]. PAT-SM6 is directed against a particular isoform of GRP78, showing the ability to inhibit the development of gastric carcinoma in vivo ^[117][134].

Moreover, this antibody can induce specific apoptosis in multiple myeloma cells, while it is safe against non-malignant cells ^[118][135]. A phase I clinical trial shown that PAT-SM6 could also be effective against melanoma ^[119][136].

A comparison between malignant tissue and normal tissue shows different expression of ATF4, the downstream factor of PERK ^[120][137]. In neoplastic tissues, the expression of ATF4 is higher than in normal tissues; so, this factor plays a crucial role in the response against chemotherapy. In addition, expression of ATF4 correlates with cisplatin resistance in lung cancer cell lines ^[121][138]. In the case of an ATF4 overexpression, it could induce multidrug resistance against cisplatin, doxorubicin, etoposide, irinotecan, and vincristine, but not to 5-fluorouracil ^[121][122][138,139]. The resistance to cisplatin causes an intracellular glutathione level increase ^[122][139]. On the contrary, ATF4 knockout cells show a decreased glutathione biosynthesis and higher sensitivity to anti-cancer treatments.

However, the UPR is helpful not only to improve chemotherapy; in fact, it can also have a significant impact on the radiotherapy efficacy. In breast cancer, radiotherapy triggers the PERK-pathway of the UPR, and this is induced by increased PERK protein levels, phosphorylated eIF2α, ATF4, and LAMP3 ^[123][140]. Moreover, in vivo studies on rat intestinal epithelial cells show how irradiation promotes three important consequences:

-

GRP78 expression;

-

eIF2α phosphorylation;

-

XBP1 splicing ^[124][141].

According to experimental data, this stress response is not linked to ATF6 ^[124][141]. Only during knockdown of PERK, ATF4 or LAMP3 can enhance the sensitivity of breast cancer cells to radiotherapy ^[123][140]. Furthermore, a knockdown of GADD34, an essential element to prolong eIF2α phosphorylation, causes radio-resistance; on the contrary, a treatment with a pharmacological PERK inhibitor (GSK2606414) provokes radio-sensitization ^[123][140].

Another PERK inhibitor (GSK2656157) shows the ability to suppress the development of xenografted tumours through the reduction of the vascular density ^[125][126][142,143]. Tamoxifen is a well-known anti-oestrogen drug which can induce UPR components, such as GRP78 and LAMP3 ^[104][123][121,140].

In addition to radio-sensitization of breast cancer cells, knockdown of LAMP3 sensitizes cells to tamoxifen ^[123][140]. In breast tamoxifen-tolerant cancer, cells have been noted an increased expression of LAMP3 which, if silenced, provokes a decreasing of tamoxifen-resistance ^[123][140].

Another arm of UPR is upregulated in breast cancer cells: XBP1-pathway ^[127][144]. In fact, in this kind of tumour, XBP1 is co-expressed with the oestrogen receptor. XBP1 overexpression causes the independent growth of oestrogen receptor positive cells regardless of the presence of the hormones. Furthermore, this condition makes cells more resistant to the anti-oestrogen drugs such as tamoxifen and faslodex ^[127][144]. According to data, higher levels of spliced XBP1 are linked to more aggressive breast tumours and poor prognosis ^[128][145].

There are different molecules which can induce ER stress and some of them have the potential to be used as anti-cancer therapies. For example, eeyarestatin 1 is a small molecule which induces ER stress by preventing ER-associated degradation ^[129][146]. Eeyarestatin 1 shows a synergistic behaviour with bortezomib and is selective against cancer cells. Furthermore, tunicamycin-induced ER stress can increase sensitivity of some cancers to therapies, for example: breast cancer cells sensitivity to radiotherapy ^[130][147] and ovarian cancer cells sensitivity to cisplatin and carboplatin ^[131][148].

Other data have shown different situations where ER-stress can induce resistance to chemotherapy ^[132][133][149,150]. Tunicamycin, through induction of GRP78, can significantly reduce the apoptosis induced by chemotherapy ^[134][151] and, as a consequence, GRP78 silencing can reduce the sensitivity to tunicamycin effects.

Thus, depending on the treatment given, pharmacological induction of ER stress can be effective both for promoting sensitivity and inducing resistance to anti-tumoral therapies.

Another important study led by Ledoux et al. shows that via the UPR, glucose withdrawal induces the expression of P-glycoprotein in hepatic cancer cells ^[135][152], enhancing the efflux of chemotherapeutic drugs.

Some conditions (hypoxia and detachment from the extracellular matrix) and compounds (A23187, tunicamycin, thapsigargin and brefeldin A) which cause ER-stress can induce both the UPR and the autophagy ^[136][137][153,154].

Autophagy is an important way to reduce ER stress and to allow survival of cells ^{[56][136][138]}[73,153,155] but this effect was discovered only in cancer cells, not in normal cells ^[136][153].

As shown by Kouroku et al., polyglutamine aggregates induce ER stress leading to autophagy as a degradation mechanism ^[139][156]. In this case, the activation of autophagy is provoked by the PERK-arm of the UPR. Dominant-negative PERK and mutated non-phosphorylatable eIF2α avoid the conversion of LC3-I to LC3-II, while the phosphorylation of eIF2α induces ATG12 expression ^[139][156].

In addition, in neoplastic conditions, a resistance against therapies is provoked by ER stress-induced autophagy.

Pharmacological-induced ER stress, before the cisplatin therapy, can induce autophagy and confer resistance against cisplatin-induced apoptosis ^[140][157], whereas in breast cancer, PERK-dependent autophagy is induced by radiotherapy ^[130][141][147,158]. It is possible to sensitize cells to radiotherapy though pharmacological inhibition of autophagy or PERK-pathway silencing ^[123][141][140,158].

Moreover, autophagy can be induced also by tamoxifen treatment, and this process is mediated by ATF4-induced LAMP3 ^[142][143][159,160] or by GRP78-dependent inhibition of mTOR ^[102][119]. Another possible treatment of breast cancer is bortezomib, that leads to an ATF4-dependent increase in LC3B and autophagy, favouring bortezomib-resistance ^[144][161].

Other reports have indicated that not the PERK-arm, but the IRE1-arm is responsible for UPR-mediated autophagy. In neuroblastoma, ER stress caused by amino acid starvation, thapsigargin, or tunicamycin promotes autophagy ^[56][73]. However, this condition can be blocked by IRE1 silencing or a pharmacological treatment with a JNK inhibitor.

Interestingly, cells lacking PERK or ATF6 induce autophagy similar to wild-type cells, suggesting how autophagy is independently triggered.

UPR-independent pathway can also induce autophagy in response to ER stress. ER stress caused by thapsigargin or tunicamycin triggers autophagy via protein kinase Cθ (PKCθ) ^[145][162]. PKCθ activation takes place independently from UPR sensors. ER stress-activated autophagy can be blocked by PKCθ silencing or its pharmacological inhibition but, this condition can be prevented by chelating intracellular Ca²⁺, while the kinase is not reactive to amino acid starvation.

Furthermore, when cytosolic Ca²⁺ levels increase, AMPK is triggered by CAMKK-β, this condition leads to ER-stress which causes the inhibition of mTOR ^[145][162], inducing autophagy.