Normal tissue cells need to undergo a series of changes before they can become cancer cells, such as genetic changes, or activation or inactivation of certain signaling pathways (8). For example, stearoyl CoA desaturase (SCD1) and proto-oncogene SEC61 translocon subunit γ can promote ER homeostasis, avoid the production of misfolded proteins and inhibit the occurrence of tumors (63). Paraoxonase 2 (PON2), a paraoxonase protein, consists of lactone hydrolases with different substrate specificities (64). When PON2 is located on the nuclear membrane and ER, it can increase the stability of the ER and protect cancer cells from adverse environmental conditions and chemotherapy (64). Hypoxia-stimulated galectin-1 (Gal1) is an effective regulator of GBM cell migration and an angiogenic molecule (65). Additionally, the reduction of Gal1 weakens the expression levels of seven genes related to chemical resistance: ORP150, BNIP3L, HERP, TRA1, GADD45B, GRP78 and CYR61 (65). The absence of cytochrome P450 17A1 can induce the occurrence of ER stress and reactive oxygen species (ROS) generation by regulating secretion-associated Ras-related GTPase 1 (66). Furthermore, heat-shock protein 27 and 90 can decrease the levels of cytochrome C, caspase3, caspase9 and caspase12 in GBM cells (67,68). Additionally, cyclophilin B is a prolyl isomerase residing in the ER, and its absence can damage the ER structure (69).

In addition, knockdown of cAMP-responsive element-binding protein 3 induces cell apoptosis by increasing p-PERK, p-eIF2α, ATF4, Bax and caspase3 (70). Reversion inducing cysteine rich protein with kazal motifs (RECK) is a key suppressor gene in regulating cancer cell invasion and metastasis (71). Highly expressive RECK can modulate ER stress by binding to or sequestering GRP78 to activate p-eIF2α (71). Tumor necrosis factor receptor-associated protein 1 can markedly induce the occurrence of ER stress by activating ATF4 (72). In addition, neural precursor cells can migrate to advanced astrocytoma via the release of the vanillin receptor [transient receptor potential vanillin subfamily member 1 (TRPV1)] (73). TRPV1 induces GBM cell death via ER stress (73). Therefore, drugs that target these proteins located on the ER membrane or stabilizing the structure of the ER can be developed.

The redox environment of the ER determines the fate of proteins entering the ER, and the level of redox signaling mediators also regulates the level of ROS (74). ROS can induce the occurrence of ER stress through redox signaling mediators, such as NADPH-P450 reductase, protein disulfide isomerase-endoplasmic reticulum oxidoreductase 1, NADPH oxidase 4, glutathione/glutathione disulfide and calcium (74,75).

In addition, ROS-mediated hypoxia-inducible factor 1α serves an important role in promoting tumor microenvironment, anti-apoptosis and drug resistance in GBM (76). Mitochondrial PTEN-induced kinase 1 (PINK1), a regulator of the Warburg effect, is a negative regulator of proliferation in GBM cells (77). PINK1 can inhibit ROS generation and cell proliferation through FOXO3a (77). This finding highlights the importance of the balance between PINK1 and ROS in normal cells and cancer cells (77). Additionally, luteolin, a common dietary flavonoid, can induce a lethal ER stress pathway and mitochondrial dysfunction by increasing the intracellular ROS levels (78). Dihydroartemisinin (DHA) induces cell apoptosis through mitochondrial membrane depolarization, cytochrome'C and caspase 9 (79). Furthermore, the cytotoxicity of DHA can increase the expression levels of GRP78, GADD153, eIF2α and caspase12 (79).

One of the main obstacles for tumor progression is that cancer cells grow in a low-oxygen environment (80). Hypoxia can stimulate the adaptive response to promote cell proliferation and survival, as well as angiogenesis (81). However, some researchers have illustrated that hypoxia can also cause cell apoptosis or necrosis (81,82). Consequently, under hypoxic conditions, understanding the decision-making process that regulates cell death, adaptation and resistance for treatment is crucial. Hypoxia has been demonstrated to induce ER stress (83). Additionally, hypoxia can induce the cell surface exposure of calreticulin, a hallmark of immunogenic cell death (84,85).

Studies have demonstrated that after knock out of the gene encoding endothelin-1, the expression levels of endothelin receptor type B, endothelin 1, endothelin converting enzyme 1 and endothelin receptor type A are upregulated, which will increase the sensitivity of GBM cells to hypoxia-induced ER stress (86,87). Furthermore, under hypoxic conditions, the expression levels of snail family transcriptional repressor 2 and mesoderm specific transcript are upregulated, which will amplify the inhibitory effect of IRE1 on various genes (88). Hypoxia also leads to upregulation of IGFBP6, IGFBP7, IGFBP10/CYR61, WISP1 and WISP2, and downregulation of IGFBP9/NOV at the mRNA level (89). IRE1 markedly downregulates IGFBP7, IGFBP10/CYR61, WISP1 and WISP2, and upregulates IGFBP9/NOV, which shows that ER stress is an essential part of malignant GBM cell proliferation (90).

The glucose metabolism allows energy to be oxidized by its carbon bonds and then used in the form of ATP. The final product of glucose can be lactate or carbon dioxide (91). In the 1920s, Otto Warburg and his colleagues found that tumors were absorbing a lot of glucose compared with surrounding tissues (92). Furthermore, in the presence of oxygen, glucose can also be fermented to produce lactic acid through aerobic glycolysis (92). Subsequently, in 1929, the British biochemist Herbert Crabtree confirmed Warburg's findings and further revealed that the respiratory intensity of tumors was variable, and numerous tumors exhibited this phenomenon (93). Additionally, Racker developed his theory of the origin of the Warburg effect in terms of intracellular pH imbalance and ATPase activity defects (94). Research on genetics and pharmacology demonstrated that the Warburg effect is necessary for cancer cell proliferation (95). Studies have demonstrated that the direct and indirect result of cancer-causing mutations is the reprogramming of cancer cell metabolism (96,97). Obtaining the necessary nutrients from the nutrient-deficient environment is a common feature of tumor metabolism (98). Cancer cells can use these nutrients to maintain viability and build new biomass (98). Therefore, tumors are considered to be a metabolic disease (99).

Low glucose means that cancer cells collect less energy. GRP78 can be upregulated in a low-glycemic state, and enables GBM cells to survive by inducing autophagy (100). Additionally, in the low-glycemic stage, p-PERK and cleavage of ATF6 are upregulated, which indicates that low glucose can lead to ER stress, activation of caspase, cell dysfunction, cell arrest and cell death (101-103). Metformin is the first-line drug for type 2 diabetes, and it can induce cancer cell death via the ASK1/phorbol-12-myristate-13-acetate-induced protein 1 and ROS/ASK1/JNK signaling pathways (104). The aforementioned can also indicate that the ketogenic diet can partially inhibit cancer cell proliferation.

In tumor tissues, due to increased glycolytic activity, slow blood circulation and insufficient blood supply of cancer cells, the pH of certain tumors (astrocytoma and squamous cell carcinoma) is <6.0, while the normal physiological pH is 7.3 (105). The low outflow of potentially toxic metabolic waste and a low influx of metabolites further enhance acidic conditions where cancer cells are located (106). The metastasis of solid tumors to acidic sites can promote tumor growth, invasion, angiogenesis, immunosuppression and chemotherapy resistance (107). However, there are conflicting reports on the effects of the acidic environment on cancer cell physiology. For example, low pH can promote tumor development; however, certain studies have demonstrated that low pH can also induce cancer cell apoptosis (108,109). For example, acidic stress may lead to upregulation of Src activity, VEGF and MMPs, which is conducive to cancer cell survival and metastasis (110,111). Acid-mediated apoptosis is considered to be the result of caspase activation (109). The acidic environment can regulate GBM cell proliferation and radiosensitivity (112). Additionally, acidity induces the expression of eIF2α, IRE1α, ATF6, GADD153 and caspase12 (113,114). This indicates that, in the brain tissue, the acidic environment may kill cancer cells via the ER stress pathway.

Certain drugs that affect ER calcium balance, such as the thapsigargin (115), calcium ionophore A23187 (116) and calcium ion chelator EGTA (117), can lead to ER stress. Calcitonin C (Cal-C) is a photoactivated inhibitor (118). Cal-C binds to protein kinase C and other protein regulatory domains with diacylglycerol/phorbol ester binding sites (118). It may damage calcium imbalance in GBM cells, and then result in ER stress (119). 2,9-diazaspiro[5.5]undecane depletes intracellular Ca²⁺ stores (120). Cyclophilin/calcineurin inhibitor Cyclosporine A has certain apoptotic characteristics, causes numerous cytoplasmic vacuoles, and increases immunizing ER stress and autophagy markers, such as PERK, IRE1α, GRP78, GADD153 and LC3-II (118). Salinomycin is a polyether ionophore antibiotic and induces cell apoptosis through ER stress and autophagy (121). Nonsteroidal anti-inflammatory drugs can release additional Ca²⁺ to induce ER stress, thereby preventing cell transformation and slowing proliferation (122). Monensin can destroy calcium homeostasis and overcome TNF-related apoptosis-inducing ligand (TRAIL) resistance in GBM cells via ER stress, and thus it is currently considered an anticancer drug (123). Additionally, other polyether antibiotics, such as narasin, salinomycin, lasalocid A and nigericin, can also overcome TRAIL resistance in GBM cells via ER stress (123).

ER stress can give rise to changes in lipid metabolism; however, certain evidence suggests that dysfunctional lipid metabolism may activate UPR, regardless of whether there is a misfolded protein in the ER lumen (124). Subsequently, after UPR, genes involved in lipid metabolism will be upregulated (125). The brain is generally considered to be one of the most fat-rich organs, and the lipid content itself can affect various clinically relevant behavioral indicators (126). These bioactive lipids include steroids, diacyl glycerol, sphingolipids, phosphatidylinositol phosphate, phosphatidylcholine and polyunsaturated fatty acids (126). Furthermore, saturated fatty acids have the effect of promoting ER stress, while unsaturated fatty acids counteract this effect (126). Saturated fatty acids, such as palmitic acid and stearic acid, are known inducers of ER stress in various cell types, such as liver and breast cancer cells, and can regulate cell survival and apoptosis signals (126). SCD1, a downstream gene of sterol regulatory element-binding protein (SREBP), mediates lipid desaturation, which has also been found to be a critical determinant of cancer cell survival (127). SREBPs have important roles in regulating lipid metabolism and mediate lipid synthesis in GBM cells (127). Loss of SREBP and lipid synthesis can block GBM cell proliferation in xenograft models (128). In addition, SREBP ablation is also accompanied by the activation of IRE1α and PERK (129). These findings indicate that proliferating cells need to establish a balance between their proliferation rate and unsaturated lipid supply to prevent ER stress. Normal cells can regulate their proliferation rate in response to nutrient availability and retain a pool of unsaturated lipid, which allows cells to maintain homeostasis and avoid ER stress (129). However, with the rapid proliferation of cancer cells, if the exogenous unsaturated lipids are limited, the cancer cells will experience ER stress, eventually leading to cell death (129).

GRP78 predominantly resides in the ER lumen within normal cells, and most of the research on GRP78 has focused on cytosolic or total GRP78 (131,132). However, in tumor microenvironments where GRP78 expression is upregulated, GRP78 also localizes to the surface of GBM cell membranes (133). GRP78 may influence not only GBM cells, but also the surrounding microenvironmental vasculature (133). Therefore, it has been gradually revealed that some compounds, such as epigallocatechin 3-gallate (EGCG), honokiol, celecoxib and bortemozib, can inhibit the growth of glioma by inhibiting GRP78 (133). IRE1 is the main mediator of the UPR (134). When cancer cells trigger ER stress in an unfavorable environment, the IRE1 signal can be an adaptive mechanism (135). However, the Food and Drug Administration-approved compounds methotrexate, cefoperazone, folinic acid and fludarabine phosphate, as inhibitors of IRE1, hinder the adaptation mechanism (135). In addition, flavokawain B, a natural kava chalcone, exhibits potent anti-tumor activity in various cancer types, such as lung cancer cells (136) and gastric cancer cells (134). It induces protective autophagy by targeting the ATF4-GADD153-AKT-mTOR signaling pathway (137). Piperlongumine preferentially kills high-grade glioma (HGG) cells but has little effect on normal brain cells (138). It induces ROS generation and disrupts protein folding in the ER by increasing the oxidative deactivation of peroxide reduction 4 to activate the ER stress pathway in HGG cells (138). Therefore, piperlongumine can be regarded as an effective drug for the treatment of GBM.

There are still numerous compounds that can induce cell apoptosis via the ER stress pathway in GBM cells. For example, Isochaihulactone, a natural compound extracted from the Chinese traditional herb Nan-Chai-Hu, can disrupt ER homeostasis in GBM cells (139). The novel resorcinol derivatives [2,4-bis (4-fluorophenylacetyl) resorcinol (BFP)] can increase some characteristic ER stress markers, such as GRP78, IRE1, eIF2α and GADD153, in human GBM cell lines (U251 and U87), and a mouse GBM cell line (C6 cells) (140). In addition, treatment with BFP can increase ROS generation and downstream caspase activation, such as caspase12, caspase9 and caspase7 (140). Cannabinoids can inhibit the epithelial-mesenchymal transition of several tumors in rats and mice, and enhance tumor immune surveillance (141). Therefore, cannabinoids may be considered as potential anticancer drugs. In 2003, cannabinoids were used to explore the anticancer mechanism (142). The results demonstrated that the main mediator of cannabinoid is the stress-regulating protein p8 (also designated as a candidate for metastasis 1) (142). Further research revealed that p8 has an apoptotic effect by upregulating ER stress-related genes ATF4 and GADD153 (142). Additionally, Shikonin (one of the main active ingredients of Chinese herbal medicine Lithospermum erythro-rhizon) (143), fatsioside A (a novel baccharane-type triterpene glycoside) (144), garlic compounds (diallyl sulfide and diallyl disulfide) (145), apigenin, (-)-epigallocatechin, and genistein (146,147), desipramine (a tricyclic antidepressant) (148), curcumin (149,150), xanthatin (a natural sesquiterpene lactone purified from Xanthium strumarium L.) (151), honokiol (a cell-wall component of M. grandiflora) (152), sinomenine hydrochloride (the main biologically active alkaloid isolated from Leymus chinensis) (152), radicol (a novel trinorguaiane type sesquiterpene) (153), phenyl isothiocyanate (a member of the isothiocyanate family) (154,155), redox organoruthenium compound 11 (RDC11; one of the most active compounds among the novel ruthenium-derived compounds) (156) and obtusaquinone [OBT; a natural compound from the heartwood of Dalbergia retusa (cocobolo)] (157) have also been reported to induce ER stress (Table I). Among them, after injecting GBM cells into the mouse cranial cavity, RDC11 and OBT can reduce tumor progression, and improve the survival rate of the mouse (156,157). Therefore, whether other natural compounds can pass through the blood-brain barrier at the individual level requires in-depth research. In addition, there are also numerous reports on other tumors, which suggest that natural compounds can induce cell apoptosis through ER stress. For example, aspirin can induce multiple myeloma (MM) cell apoptosis by inhibiting Blimp1, activating the ATF4/CHOP apoptotic pathway (158). Valosin containing protein (p97/VCP) is an ER-associated protein, and novel p97/VCP inhibitor induces ER stress and apoptosis in both bortezomib-sensitive and -resistant MM cells (159). Furthermore, there are numerous other compounds that can also affect cancer cell proliferation through ER stress in other tumors, such as 18βH (a semisynthetic derivative of -glycyrrhetinic acid) in breast cancer cells (MCF-7 and MDA-MBA-231) (160), resveratrol (a natural polyphenol compound) (161) and 2-pyrazine-PPD (a novel dammarane derivative) (162) in gastric cancer, sothiocyanates (natural compounds abundant in cruciferous vegetables) in non-small cell lung cancer cells (163), and honokiol (a hydroxylated biphenyl natural product) in prostate cancer, melanoma, lung cancer, leukemia and colorectal cancer (164). Therefore, whether these natural compounds can induce ER stress and be used for the treatment of GBM in vivo still requires basic verification and clinical trials.

GBM stem cells (GSCs) can be considered key drivers of tumor growth, aggressiveness and therapy resistance in GBM (8). PERK is a well-characterized switch between survival and death during persistent ER stress and mediates cell death through induction of GADD153 (165). A previous study has demonstrated that ER stress aggravation targets GSCs, and PERK directly regulates SOX2 downregulation at the protein level to induce GSC differentiation, independent from eIF2α/ATF4 signaling (165). Furthermore, ionizing radiation potentiates ER stress, which reduces proliferation in a PERK-dependent manner (166). Adding PERK inhibitor, ER stress inducer (2DG) and GADD34 phosphatase inhibitor (Sal003) to irradiated GBM cells can reduce cell viability (166). In addition, other highly selective PERK inhibitors may provide a ground-breaking, anticancer treatment strategy in a PERK-dependent manner. 7-Methyl-5-(1-2,3-dihydro-1H-indo l-5-yl)-7H-pyrrolo[2,3-d]pyrimidin-4-amine (GSK2606414) is an oral, effective and selective PERK inhibitor, which inhibited the growth of a human tumor xenograft in mice by inducing ER stress (167). Therefore, GSK2606414 treatment may be considered as an effective drug therapy for GBM. Small-molecule inhibitor 42215 of PERK can markedly induce apoptosis after treatment of cancer cells (168). ATF4 is the master regulator of the cellular stress response and the core regulator of the PERK-eIF2α signaling pathway (169). Kurarinone (Extract of S. flavescens Roots) activates ATF4 to induce cancer cell apoptosis (169). A mixture of sixteen previously selected small molecules ('active mixture'; AM16: l-arginine, l-tyrosine, l-histidine, l-tryptophan, l-methionine, l-phenylalanine, adenine, l-(-)-malic acid, 2-deoxy-d-ribose, orotic acid, d-(+)-mannose, hippuric acid, pyridoxine, d-biotin, (-)-riboflavin and l-ascorbic acid) can upregulate ATF4 and GADD153 to induce cell apoptosis (170). Furthermore, salicylaldehyde analogs (MK0186893) and umbelliferones (4 µ8c) can be used as IRE1 RNase inhibitors, and have shown promise as a potential therapeutic strategy to counteract disease pathogenesis associated with overactive IRE1 signaling (171,172). Palmitoylation inhibitors (2-bromopalmitate, cerulenin and tunicamycin) induce cell death by promoting the accumulation of XBP1 (173). In addition, >10,000 non-toxic compounds that may activate IRE1-dependent XBP1 splicing through a mechanism independent of binding the IRE1 kinase active site have been identified by a high-throughput screening approach in July 2020 (174). This undoubtedly provides more drugs for the treatment of tumors via ER stress and requires in-depth exploration and screening.

In addition to specific small molecule modulators of UPR pathways and their potential use in developing anticancer therapies, numerous small molecule compounds have been demonstrated to induce GBM cell apoptosis (Table I). For example, NEO214 (rolipram-perillyl alcohol conjugate) is produced by the covalent linkage of carbamate linkage and cyclopropanol (175). It can cross the blood-brain barrier, and induce cell death via ER stress and the death receptor 5/TRAIL/TNF superfamily member 10 signaling pathway (175). Therefore, NEO214 may be considered as a potential clinical antitumor drug. Asiatic acid (AsA) is a natural small molecule, and it is widely used to cure various neurological disorders. AsA also induces ER stress by increasing GRP78, IRE1α and calpain, to damage a cellular organization in human GBM cells (LN18, U87MG and U118MG) (176). Notably, AsA can cross the blood-brain barrier (12). Therefore, AsA may be considered as a potential clinical antitumor drug. In addition, 2-amino-N-acetamide (176), NEO212 (a combination of TMZ and perillyl alcohol) (177), NEO100 (a high-purity and high-quality polychlorohydrin) (178), Compound-7g (179), platinum thiopyridine (II) complex (180), endothelial monocyte activating polypeptide II (181), berberine (an isoquinoline quaternary alkaloid isolated from a variety of medicinal plants) (182), N-methyl-4-isoleucine-cyclosporine (a small molecule cyclophilin binding inhibitor) (183), celecoxib (a non-steroidal anti-inflammatory drug) (184), the silent mating type information regulation 2 homolog activator (R)-N-(2-(3-((3-Hydroxypyrrolidin-1-yl)methyl)imidazo[2,1-b] thiazol-6-yl)phenyl)-2-naphthamide (185), neuro-steroid, 5-androstene 3β,17α diol (186), minocycline (187) and ursolic acid (188-190) can also induce GBM cell apoptosis via the ER stress pathway. Whether these small molecules can be used in clinical trials needs to be examined in animal models to ensure that they can cross the blood-brain barrier and reduce damage to normal cells.

In addition to compounds, some viruses, bacteria, and calcium activators or inhibitors have been reported to induce GBM cell apoptosis via ER stress (Table I). For instance, ovitriol A (a fungal sesterterpene from Bipolaris oryzae) can induce paralysis-like cell death in human glioma cell lines (T98G, U251MG, U343, U373MG and A172), accompanied by the expansion of the ER (191). Unconjugated bilirubin may cause bilirubin neurotoxicity (192). It can also cause cell death t by increasing GADD153 in U87MG cells (192). Rhabdovirus is an important regulator of rhabdovirus-mediated cytotoxicity and mediates the ER stress response pathway to induce cell apoptosis (193). Chikungunya virus (CHIKV), an old-world alphavirus, can induce DNA fragmentation, loss of mitochondrial membrane potential, poly(ADP-ribose) polymerase (PARP) cleavage, nuclear enrichment and visible cytopathic effects in a dose- and time-dependent manner (194).

The focus of tumor research has always been the optimization of treatment strategies for malignant tumors. Silica nanoparticles (SiNPs) are such a strategy, which is rapidly developing into a promising tool for cancer diagnosis, imaging and treatment (195). SiNPs lead to impaired mitochondrial function, ROS generation and cell death by elevating levels of ER stress genes, including GRP94, GRP78, GADD153 and cyclooxygenase-2 (COX2) (195). Yessotoxin (YTX) is a polycyclic ether compound produced by dinoflagellate and accumulates in filter-fed shellfish (196). YTX can upregulate p-PERK, p-eIF2α, s-XBP1 and GADD153 in human glioma cell lines (SF539, SF295 and SNB75) (196). Additionally, YTX induces cell cycle arrest and increases cholesterol and polar lipid content in glioma cells (196). In addition, amiodarone is a widely used antiarrhythmic drug (197). Amiodarone can inhibit a variety of ion channels, including Na⁺/Ca²⁺ exchangers, L-type Ca²⁺ channels and Na⁺ channels, and increases the intracellular Ca(2+) level and GADD153 expression (197). Although viruses, bacteria, and calcium activators or inhibitors can induce cancer cell apoptosis through ER stress, their safety still needs to be considered.

Various studies on GRP78 have also demonstrated that GRP78 has an important role in recurrent GBM and tumor progression after initial treatment (198,199). Of particular importance is TMZ, the standard-of-care chemotherapeutic treatment for GBM. TMZ has been demonstrated to result in activation of the UPR in GBM cells, inducing increased levels of UPR markers, GRP78 and GADD153 (100). Therefore, certain drugs can be used in combination with TMZ to enhance the sensitivity of GBM to TMZ by increasing ER stress (Table II). For example, bufothionine is extracted from the skins and parotid venom glands of the toad Bufo bufo gargarizans Cantor (200). Bufothionine can synergize with TMZ to exert an anti-growth effect by triggering ER stress (200). PI3K/mTOR signaling is ubiquitous in GBM (201). XL765 (Voxtalisib/SAR245409), an effective dual inhibitor of PI3Ks and mTOR, inhibits the proliferation of GBM cells by inducing ER stress-dependent apoptosis (201). The combination of XL765 and TMZ can achieve improved therapeutic effects in A172, U87MG and T98G cells (201). Fluoxetine (FLT), as a drug widely used in cancer-related depression, has strong anticancer effects in different types of cancer cells, such as human ovarian granulosa tumor COV434 cells, and SKBR3 and MCF-7 breast cancer cells (202). The combination of FLT and TMZ can also induce the activation of ATF6, PERK, eIF2α, ATF4 and GADD153 (202). The upregulation of prolyl 4-hydroxylase-β polypeptide (P4HB) expression is associated with the increase of the IC₅₀ of TMZ, and is relatively upregulated in resistant GBM cells (203). Targeting P4HB blocks its protective function and makes GBM cells sensitive to TMZ (203). Chloroquine (CQ), a quinoline-based antimalarial drug, can kill the plasmodium falciparum parasite in the red blood cell stage by blocking the acidic food vacuole heme to detoxify (198). Under physiological pH conditions, CQ has unique chemical properties and is a weak base that easily crosses the lipid bilayer of cells (198). The combination of TMZ and CQ can trigger cell death by enhancing the formation of LC3B-II, the accumulation of polyubiquitinated proteins, GADD153 and the cleavage of PARP (198,204,205). N,N-[(8-hydroxyquinoline)methyl]-substituted benzylamine (JLK1486) is a novel type of ER stress inducer (206). The combined use of TMZ and JLK1486 can cause long-term ER stress in human GBM cell lines (U87MG, A172 and T98G) by increasing the levels of GRP78, ATF4 and GADD153 (206). In addition, celecoxib is a selective inhibitor of COX2. Increasing reports have described that this drug has powerful anti-proliferation and pro-apoptotic effects without the obvious involvement of COX2 (207,208). Celecoxib causes ER stress by leaking calcium from the ER into the cytoplasm (207). The combination of bortezomib and celecoxib can increase the expression levels of ER stress markers GRP78 and GADD153, and cause the activation of c-jun (207). In addition, whether other compounds enhance the sensitivity of GBM to TMZ by enhancing ER stress needs to be verified.