Dihydroartemisinin-induced unfolded protein response feedback attenuates ferroptosis via PERK/ATF4/HSPA5 pathway in glioma cells
- 207 Downloads
Dihydroartemisinin (DHA) has been shown to exert anticancer activity through iron-dependent reactive oxygen species (ROS) generation, which is similar to ferroptosis, a novel form of cell death. However, whether DHA causes ferroptosis in glioma cells and the potential regulatory mechanisms remain unclear.
Effects of DHA on the proliferation, cell death, ROS and lipid ROS generation as well as reduced gluthione consumption were assessed in glioma cells with or without ferroptosis inhibitor. The biological mechanisms by which glioma cells attenuate the pro-ferroptotic effects of DHA were assessed using molecular methods.
DHA induced ferroptosis in glioma cells, as characterized by iron-dependent cell death accompanied with ROS generation and lipid peroxidation. However, DHA treatment simultaneously activated a feedback pathway of ferroptosis by increasing the expression of heat shock protein family A (Hsp70) member 5 (HSPA5). Mechanistically, DHA caused endoplasmic reticulum (ER) stress in glioma cells, which resulted in the induction of HSPA5 expression by protein kinase R-like ER kinase (PERK)-upregulated activating transcription factor 4 (ATF4). Subsequent HSPA5 upregulation increased the expression and activity of glutathione peroxidase 4 (GPX4), which neutralized DHA-induced lipid peroxidation and thus protected glioma cells from ferroptosis. Inhibition of the PERK-ATF4-HSPA5-GPX4 pathway using siRNA or small molecules increased DHA sensitivity of glioma cells by increasing ferroptosis both in vitro and in vivo.
Collectively, these data suggested that ferroptosis might be a novel anticancer mechanism of DHA in glioma and HSPA5 may serve as a negative regulator of DHA-induced ferroptosis. Therefore, inhibiting the negative feedback pathway would be a promising therapeutic strategy to strengthen the anti-glioma activity of DHA.
KeywordsDihydroartemisinin Ferroptosis GPX4 Endoplasmic reticulum stress HSPA5 Glioma
CCAAT/enhancer-binding protein homologous protein
Glutathione peroxidase 4
Heat shock protein family A (Hsp70) member 5
Inositol-requiring transmembrane kinase/endoribonuclease 1α
Reduced β-nicotinamide adenine dinucleotide phosphate
Protein kinase R-like ER kinase
Reactive oxygen species
Unfolded protein response
Glioma is the most common primary brain tumor in both children and adults, and accounting for 7% of cancer-related death before the age of 70 years . Despite significant improvement in treatment for glioma patients, the survival remains not optimistic, particularly for those with glioblastoma. Patients with newly diagnosed glioblastoma exhibit one-year survival rate of 35.7% and five-year survival rate of 4.7%, with generally poor responses to most of current available therapeutic modalities . Therefore, there is an urgent need to develop novel effective therapies with low toxicity for the treatment of glioma based on molecular studies.
Regulated cell death modalities are emerging as important players in physiological conditions and diseases, in which caspase-dependent apoptosis is the mostly studied regulated cell death form . In recent years, growing evidence has indicated that regulated cell death mechanisms beyond apoptosis also play important roles in the molecular processes controlling the demise of cells. Ferroptosis is a recently recognized form of regulated cell death characterized by the accumulation of lipid peroxidation products and lethal reactive oxygen species (ROS) derived from iron metabolism, which is genetically, biochemically and morphologically distinct from other forms of cell death . An increasing number of studies have implicated that ferroptosis is involved in various human diseases, including neurodegenerative diseases, ischemia-reperfusion injury, and kidney degeneration . In addition, a number of cancer cells have been proven to be susceptive to ferroptosis inducers. Moreover, ferroptosis inducers such as erastin can improve the efficacy of chemotherapeutic reagents, including temozolomide, cisplatin, ara-C and doxorubicin . Therefore, inducing and enhancing ferroptosis may be a promising strategy for cancer therapy.
Artemisinin (ART), the active principle of the Chinese medicinal herb Artemisia annua L, and its derivatives have been widely used to treat schistocidal malaria infection with high efficiency and low side effects . Dihydroartemisinin (DHA), a main active derivative of ART, has been demonstrated to exert preferentially cytotoxic effects toward human malignancies from various tissues, including breast, lung carcinoma, colorectal cancers as well as lymphoma [8, 9, 10, 11]. Recent studies have shown that the anti-cancer activity of ART derivatives including DHA highly relies on the iron-mediated cleavage of endoperoxide bridge and subsequent ROS generation . Since most cancer cells have higher iron uptake than normal cells, DHA may exert selective toxicity toward cancer cells [12, 13]. Moreover, Ooko et al. have shown that treatment of ART derivatives caused significant changes in the expression of iron-related genes in 60 National Cancer Institute cell lines, indicating the ferroptosis-inducing activity of ART derivatives . Indeed, Lin et al. have preliminarily revealed that DHA induces ferroptosis in head and neck carcinoma cells . However, whether DHA causes ferroptosis in other types of cancer including glioma remains unclear.
The sensitivity to ferroptosis is tightly linked to numerous biological processes, including metabolisms of iron, membrane lipids and amino acid and the redox regulation system . Recent studies have demonstrated that glutathione peroxidase 4 (GPX4), heat shock protein family A (Hsp70) member 5 (HSPA5), heat shock protein beta-1 (HSPB1), nuclear factor erythroid 2-related factor 2 (NRF2) and α6β4 integrin function as negative regulators of ferroptosis by limiting ROS production or reducing cellular iron uptake [16, 17, 18, 19, 20], whereas reduced β-nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, heme oxygenase-1 and p53 act as positive regulators of ferroptosis [4, 21, 22]. However, the roles of these molecular players in ferroptosis appear to be drug- or cell type-specific, and thus regulatory mechanisms of ferroptosis induced by other reagents including DHA are still unknown.
Endoplasmic reticulum (ER) stress in solid tumors results from a dysregulation of protein synthesis, folding, secretion, or posttranslational modification, which can be triggered by environmental stimuli such as nutrient deprivation, hypoxia, chronic viral infection, oxidative stress and anticancer drugs . To cope with ER stress, cancer cells initiate an evolutionarily conserved signaling process known as the unfolded protein response (UPR), which is coordinated by three ER transmembrane-bound sensors: inositol-requiring transmembrane kinase/endoribonuclease 1α (IRE1α), activating transcription factor 6 (ATF6) and protein kinase R-like ER kinase (PERK). HSPA5, also termed GRP78 or BiP, is a molecular chaperone located in the ER and acts as a master regulator of UPR . A number of studies have demonstrated that HSPA5 promotes cell survival and drug resistance under ER stress conditions . Recently, Zhu et al. have reported that HSPA5 promotes ferroptosis resistance in pancreatic ductal adenocarcinoma cells . However, whether HSPA5 regulates ferroptosis in glioma cells remains unclear. In this study, we provided the first evidence that DHA induced ferroptosis in glioma cells. Furthermore, we found that DHA also induced the expression of HSPA5 via PERK-ATF4 pathway, which attenuated the sensitivity to ferroptosis through GPX4. Moreover, inhibition of HSPA5 or PERK showed a synergistic anticancer effect with DHA in glioma. These data provided a novel preclinical evidence for the combination of DHA and HSPA5 inhibition in the treatment of glioma.
Materials and methods
Antibodies and reagents
Antibodies against HSPA5 (ab21685), CHOP (ab11419), ATF4 (ab184909), GPX4 (ab125066) and xCT (ab60171) were purchased form abcam (Cambridge, MA, USA), ATF6 (#65880), PERK (#5683), and anti-HA tag (#3724 s) antibodies from Cell Signaling Technology (Danvers, MA, USA) and anti-β-actin antibody from Santa Cruz Biotechnology (Dallas, TX, USA). DHA (D7439), ferrostatin-1 (Fer-1, SML0583), liproxstatin-1 (Lip-1, SML1414), GSK2606414 (PERK Inhibitor I, PERKi, 516,535), epigallocatechine gallate (EGCG, 324880) and deferoxamine (DFO, 252750) were supplied by Merck KGaA (Darmstadt, Germany).
Patient sample and cell culture
Histopathologically-confirmed malignant glioma samples were obtained from patients undergoing surgical treatment at the First Affiliated Hospital of Zhengzhou University. Within 1–3 h after surgical removal, tumors were washed in PBS and enzymatically dissociated into single cells. Patient-derived glioma cells were maintained in astrocyte basal medium (Lonza), supplemented with 10% fetal bovine serum (Carlsbad, CA, USA), L-glutamate, ascorbic acid, rhEGF, insulin, and gentamicin sulphate/amphotericin B according to the manufacturer’s recommendations. This study was approved by the Ethical Committee of Zhengzhou University, and written informed consent was obtained from all participants. All study procedures were carried out in accordance with the ethical standards of the Helsinki Declaration.
Glioma cell lines U251 and U373 and mouse hippocampal neuronal cell line HT22 were purchased from the American Type Culture Collection (Manassas, VA, USA). U251 and U373 cell lines cultured in DMEM medium supplemented with 10% Gibco fetal bovine serum. All cells were kept at 37 °C in a 5% CO2 humidified incubator. All cells were Mycoplasma free and authenticated by short tandem repeat DNA profiling analysis.
Cell viability assay
Cell viability was assessed using a Promega Cell Titer 96 Aqueous One Solution (G3580, Madison, WI, USA) as previously described . Briefly, cells were seeded in 96-well plates (500 per well) in 100 μl DEME medium for five replicates. On the day of assay, 20 μl Cell Titer 96 Aqueous One Solution were added into the medium and incubated at 37 °C for 4 h. The absorbance at 490 nm (OD490) was measured on a microplate spectrophotometer. Values are expressed ratio to the vehicle-treated control.
Colony formation assay
Log-phase growing cells were seeded into 6-well plates at a density of 800 cells/ well and cultured for 14 days. Then cells were washed, fixed and stained with 0.5% crystal violet. Colonies that contained > 50 stained cells were classed as clones. Colony-forming efficiency was calculated as a ratio of the number of colonies formed to the number of cells seeded and calibrated to that of untreated cells as previously described .
Cell death analysis
Cell death was analyzed using Annexin-V/7AAD (BD Pharmingen) on a FACS Calibur flow cytometer (BD Biosciences, San Jose, CA, USA). Cells undergoing cell death were analyzed by counting the cells that stained positive for Annexin V-FITC or/and 7-ADD.
Intracellular ROS analysis
ROS was stained with Dihydroethidium (DHE, Merck KGaA) and detected using flow cytometry according to our previous protocol . Briefly, cells were trypsinized and washed, and then incubated with 1.25 μM DHE for 30 min at 37 °C in dark. Fluorescence was detected on a FACS Calibur™ flow cytometer at the emission wave length of 610 nm.
Lipid ROS analysis
Lipid ROS was stained with C11-BODIPY 581/591 (D3861, ThermoFisher Scientific, Shanghai, China) and analyzed using flow cytometry as previously described . Briefly, cells were trypsinized, incubated in Hanks Balanced Salt Solution (HBSS) with 2 μM DHE for 10 min at 37 °C in dark and the resuspended in fresh HBSS. Oxidation of the polyunsaturated butadienyl portion of the dye resulted in a shift of the fluorescence emission peak from ~ 590 nm to ~ 510 nm that could be detected on the FACS Calibur™ flow cytometer.
Malondialdehyde (MDA) determination
To assay lipid peroxidation, the malondialdehyde (MDA) lipid peroxidation microplate assay (Merck KGaA) was used according to manufacturers’ instructions as previously described . In brief, cells were collected by trypsinization, lysed, and reacted with thiobarbituric acid. Fluorescence was measured on a Micro-plate Reader (Bio-Rad, Hercules, CA, USA) using excitation and emission filters of 532 and 590 nm, respectively. Lipid peroxidation levels were normalized to protein concentration.
Intracellular reduced form gluthione (GSH) and its oxidized form (GSSG) were assessed using a GSH/GSSG Ratio Detection Assay Kit II (Abcam) according to the manufacturer’s instructions. Briefly, 50 μl of GSH Assay Mixture (for GSH) or Total GSH Assay Mixture (for total gluthione) solution was added to 50 μl of cell lysate sample in a 96-well plate and then incubated from light for 30 min. The output was measured on a fluorescence microplate reader at Ex/Em = 490/520 nm and concentrations of GSH and total gluthione were calculated using corresponding standard curves. GSSG concentration was calculated as following: GSSG = (total gluthione - GSH)/2.
Cells were lysed in RIPA lysis buffer, and lysates were harvested by centrifugation (12,000 rpm) at 4 °C for 30 min. Western blot was performed according to protocols as previously described , using β-actin as the internal control. Briefly, 20 μg of the protein sample was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and then transferred to a polyvinylidene fluoride membrane. After blocking nonspecific binding sites for 60 min with 5% non-fat milk, the membrane was incubated with the primary antibody at 4 °C overnight. Membranes were washed three times with tris buffered saline with 1‰ tween-20 and incubated with horseradish peroxidase-conjugated secondary antibody at 37 °C for 1 h. After 3 washes, the bands were detected by an enhanced chemiluminescence system (WBKLS0500, Merck KGaA). Band density was measured using ImageJ software (National Institutes of Health, Bethesda, MD) and standardized to that of β-actin. Antibodies against proteins and dilution multiples are as following: HSPA5 (1:2000), CHOP (1:1000), ATF4 (1,1000), GPX4 (1:2000), HA tag (1:2000) and β-actin (12000).
RNA extraction and real-time PCR
Total RNA was extracted using TRIzol reagent (15596026, ThermoFisher Scientific) according to the manufacturer’s protocol. Then, 2 μg of RNA was reverse transcribed into first-strand cDNA by M-MLV Reverse Transcriptase (M1705, Promega, Madison, WI, USA) according to the manufacturer’s instructions. Gene-specific amplification was performed in an ABI 7900HT real-time PCR system (Life Technologies, Carlsbad, CA, USA) as previously described . The reaction system contained 0.5 μl cDNA, 7.5 μl 2× SYBR Green PCR master mix (4309155, ThermoFisher Scientific), and 200 nM of the appropriate oligonucleotide primers. All measurements were performed in triplicate, after undergoing the following reaction cycle: preheat at 95 °C for 10 min, 40 cycles of 95 °C for 30 s, and 60 °C for 1 min. The melting curve was measured at 95 °C for 15 s, 60 °C for 15 s, and 72 °C for 15 s. The Ct (threshold cycle) value of each sample was measured during exponential amplification, and was calculated from threshold cycles with the SDS 2.3 software. The relative expression of mRNA was normalized to β-actin (2-ΔΔCt method). Primers were used as following: HSPA5 forward: 5′-CACGGTCTTTGACGCCAAG-3′, HSPA5 reverse: 5′-CCAAATAAGCCTCAGCGGTTT-3′; β-actin forward: 5′-CATGTACGTTGCTATCCAGGC-3′, β-actin reverse: 5′-CTCCTTAATGTCACGCACGAT-3′.
Gene and siRNA transfection
cDNA of full-length open reading frame of HSPA5 (NM_005347.4) and GPX4 (NM_002085.4) was amplified, digested with Xho I and Hind III enzymes and cloned into pLVPT expression vector. GPX4 siRNAs (sc44465) were purchased from Santa Cruz. siRNAs against HSPA5, ATF4 and negative control (NC) were synthesized by Shanghai GenePharma (Shanghai, China). siRNA sequences were designed as following: NC, 5′-AAGGUGGUUGUUUUGUUCACU-3′; HSPA5, 5′-AAGGUUACCCAUGCAGUUGTT-3′(siHSPA5–1), 5′-AGAUUCAGCAACUGGUUAAAGTT-3′ (siHSPA5–2), 5′-GAAAUCGAAAGGAUGGUUAAUTT-3′ (siHSPA5–3); ATF4 siRNA, 5-GCCUAGGUCUCUUAGAUGATT-3′ (siATF4–1), 5′-UCCCUCAGUGCAUAAAGGA-3′ (siATF4–2). ATF6 siRNA: 5′-UAAACUGAUAAUUGGUUGCdTdT-3′ (siATF6–1), 5′-UACACUUGUAGCUCACUCCCUGAGU-3′ (siATF6–2); PERK siRNA: 5′-UAAAGGUGCUUUCAAUAAAUCCGG-3′ (siPERK-1), 5′-GCAUCUGCCUGGUUACUUA-3′ (siPERK-2), 5′-CCAGAGAAGUGGCAAGAAA-dTdT-3′ (siPERK-3); IRE1α (ERN1) siRNA: 5′-GAUGUCCCACUUUGUGUCC-3′ (siIRE1–1), 5′-GGAGAGAAGCAGCAGACUU-3′ (siIRE1–2). Gene or siRNAs were transfected into cells using Lipofectamine 2000 reagent (11,668,019, ThermoFisher Scientific) as previously described .
GPX4 activity assay
The activity of GPX4 was determined using a HT Glutathione Peroxidase Assay Kit (Trevigen, Inc. Gaithersburg, MD, USA). The assay principle was based on the oxidation of GSH to GSSG catalyzed by GPX4, which was then coupled to the recycling of GSSG back to GSH by glutathione reductase using NADPH as a reductant. The rate of decrease in NADPH absorbance measured at 340 nm during the oxidation of NADPH to NADP+ was directly proportional to GPX4 activity and thus was used to indicate GPX4 activity.
Xenograft tumor model
All of the procedures of animal experiments were approved by the Ethics Committee of Zhengzhou University and performed in accordance with the Association for Assessment and Accreditation of Laboratory Animal Care guidelines (http://www.aaalac.org). Specific pathogen-free athymic nude BALB/c mice (4–6 weeks old) were obtained from Guangdong Experimental Animal Centre (Guangzhou, China). To generate murine subcutaneous tumors, cells (for U251: 2× 106 cells; for U373: 2× 106 cells) were suspended in 0.2 ml PBS and injected into the flanks of mice (n = 6/group). Tumor volume was measured once every 3 days using calipers. Tumor volume was estimated by the following formula: length × width2 × π/6 . The mice whole body weight was measured every 2–3 days after inoculation. At the end of the experiment, all mice were killed by intraperitoneal injection of 200 mg/kg pentobarbital.
Data were expressed as mean ± SD (standard deviation) from three independent experiments except animal experiments. Differences between groups were estimated using the Student’s t-test. All these analyses were performed using IBM SPSS Statistics 22.0 software (Armonk, NY, USA) and a two-tailed value of P < 0.05 was considered statistically significant.
For combination treatment of PERKi (PERK Inhibitor I, GSK2606414), EGCG and/or DHA, cell viability assay data were converted to fraction of growth affected by the individual drug or the combination treated cells compared with untreated cells and analyzed using CalcuSyn software (Biosoft, Ferguson, MO, USA) to determine whether the combination was synergistic. This program is based upon the Chou-Talalay equation , which calculates a combination index (CI). The general equation for the classic isobologram is given by: CI = (D)1/(Dx)1 + (D)2/(Dx)2. Where Dx indicates the dose of one compound alone required to produce an effect, (D)1 and (D)2 are the doses of compounds 1 and 2, respectively, necessary to produce the same effect in combination. From this analysis, the combined effects of the two compounds can be summarized as follows: CI < 1, CI = 1, CI > 1 indicate synergistic, additive and antagonistic effects, respectively .
DHA showed anticancer effects in glioma cells
DHA induced ferroptosis in glioma cells
Unfolded protein response attenuated anti-tumor effects of DHA via PERK/ATF4 pathway
Inhibition of PERK/ATF4 signaling enhanced DHA-induced ferroptosis in glioma cells
ATF4-induced HSPA5 expression protected glioma cells against DHA-induced ferroptosis
Inhibitor of HSPA5 synergistically enhanced anti-tumor effects of DHA
HSPA5 antagonized DHA-induced ferroptosis by increasing GPX4 in glioma cells
Besides to its antimalarial usage, DHA has been considered as a promising anticancer reagent for decades. The anticancer activity of DHA has been shown to involve a series of cellular procedures, including oxidative stress response, DNA damage and repair, cell death induction, inhibition of metastasis and angiogenesis, as well as pro-oncogenic signaling suppression . Although the mechanisms by which DHA induces cell death have not been fully elucidated, iron-dependent ROS generation by cleavage of endoperoxide moiety has been accepted as the crucial procedure for its cytotoxicity . Moreover, treatment with either ferrous iron or transferrin can enhance the cytotoxicity of DHA toward cancer cells, which could be reversed by iron chelator or anti-transferrin receptor antibody [33, 34]. These findings strongly hint that DHA may exert an anticancer effect through ferroptosis. Indeed, several studies have shown that ART derivates including DHA can induce ferroptosis in head and neck carcinoma as well as ovarian cancer cells [15, 35]. In line with these finding, we for the first time demonstrated that DHA induced ferroptosis in glioma cells, based on the increased ROS generation and lipid peroxidation as well as GSH exhaustion, which could be inhibited by the iron complexant DFO or lipid peroxidation inhibitor ferrostatin-1 or liprostatin-1. These data suggest that ferroptosis may be an important anticancer mechanism of DHA. In contrast, most studies up-to-date have shown that DHA induce apoptosis in cancer cells. For example, Qin and colleagues have found that DHA induces loss of mitochondrial membrane potential, cytochrome c release, caspase activation and phosphatidylserine externalization in liver cancer cells . Other studies have demonstrated that DHA can regulate various signaling pathway to mediate caspase-dependent and -independent apoptosis [37, 38, 39]. Therefore, whether DHA-induced ferroptosis and apoptosis share some common regulatory mechanisms and when they separate each other need further investigation.
ER stress is a condition in which misfolded/unfolded proteins accumulate in the ER lumen, which overwhelms cellular proteostasis. Numerous factors can cause ER stress, including oncogene activation, hypoxia, nutrient deprivation, acidosis, anticancer treatment and oxidative imbalance . As a strong oxidative source, ferroptosis has been shown to induce ER stress in several human cancer cells [17, 18, 39]. As a result of accumulation of misfolded/unfolded proteins, UPR is activated through PERK, IRE1α and ATF6 pathways to restore proteostasis and oxidative balance. In this study, we found that inhibition of PERK by either siRNA or PERKi significantly enhanced DHA-induced lipid peroxidation and glioma cell death, which could be abolished by DFO. However, IRE1α or ATF6 silencing did not enhance DHA-induced glioma cell death. These findings suggest that UPR through PERK pathway may protect glioma cells from DHA-induced ferroptosis. However, the biological roles of IRE1α and ATF6 signaling in DHA-induced ER stress need further investigation.
As a key transcription factor of ER stress signaling, ATF4 translation is selectively enhanced by phosphorylated eIF2α after PERK activation . ATF4 controls the expression of genes that promote cell survival and stress adaption, including amino acid transport and metabolism, antioxidant response, autophagy and protein folding . Tumor cells frequently exploit ATF4 to reduce the stress resulting from rapid proliferation and nutrient limitation inside a growing tumor mass . Previous studies have demonstrated that ATF4 is involved in chemoresistance via transcriptional regulation of membrane transporters and enzymes required for GSH biosynthesis in cancer cells . Moreover, PERK-mediated ATF4 activation has exhibited protective roles against erastin-induced ferroptosis in pancreatic cancer cells . In line with these findings, we found that ATF4 silencing significantly enhanced DHA-induced lipid peroxidation and glioma cell death like PERKi, suggesting that PERK-induced ATF4 activation may exert a protective effect against ferroptosis in glioma cells. Although most of its target genes are involved in various salvage pathways that promote cell survival, ATF4 can also exert tumor-suppressive effects through inducing apoptosis, cell cycle arrest and senescence under persistent unresolved stress conditions , suggesting ATF4 might play dual role in cell death regulation possibly in a context-dependent manner. Further investigations are needed to elucidate how ATF4 switches its functions between cancer protection against cell death and tumor suppression in specific context.
Under ER stress, a wide range of enzymes and chaperons are induced to exert cellular protection. HSPA5, as a prominent chaperone protein that is overexpressed in a wide range of cancer cells , can be induced under ER stress conditions such as lipid peroxidation and promote cancer progression and therapy resistance . HSPA5 transcription can be activated by the binding of its ATF/CRE sequence with ATF4 . In this study, we found DHA-induced HSPA5 expression could be attenuated by ATF4 silencing, indicating that ATF4 may promote HSPA5 expression under DHA-induced ER stress. Moreover, HSPA5 silencing enhanced DHA-induced ferroptosis, while HSPA5 overexpression had an opposite effect in glioma cells, which is similar to the fidings in pancreatic cancer cells . These data suggested that ATF-induced HSPA5 upregulation might be a crucial molecule for cell adaption in ER stress caused by ferroptosis.
Accumulation of lipid peroxidation products is recognized as a central mediator of ferroptosis . GPX4, the only peroxidase that eliminates lipid peroxides and protects membrane fluidity using GSH as its cofactor, has been implicated as an essential regulator of ferroptosis . Previous studies have demonstrated that direct inhibition of GPX4 using RSL3 or GPX4 knockdown can cause ferroptosis . Moreover, some ferroptosis inducers such as erastin and FIN56 have been shown to cause GPX4 degradation [46, 47]. In contrast, we found that DHA increased GPX4 expression by HSPA5 upregulation to exert a cell-protective effect against ferroptosis, while silencing GPX4 by siRNA enhanced the pro-ferroptotic effect of DHA. These data are similar to the findings of in erastin-treated pancreatic cancer cells , suggesting that ER stress-induced GPX4 activation by HSPA5 might serve as a negative feedback loop during ferroptosis.
Despite great advances in combined treatment regimen consisting of maximum resection with concurrent chemoradiotherapy, malignant glioma remains a highly lethal disease . Temozolomide, an alkylating agent, is widely used as the first-line chemotherapeutic drug for glioblastoma treatment . However, the efficacy of temozolomide is often limited by the development of resistance . Recent studies have demonstrated that ferroptosis inducers foster the cytotoxicity of temozolomide toward glioma cells , implying that ferroptosis inducers such as DHA could enhanced the antitumor effects of temozolomide. Previous studies have demonstrated that HSPA5 expression is elevated in glioma specimens and silencing HSPA5 increases temozolomide chemosensitivity in malignant glioma cell lines , suggesting the potential roles of UPR in glioma chemoresistance. Considering the crucial roles of PERK, ATF4 and HSPA5 during UPR and subsequent anti-ferroptotic effects, it is reasonable to enhance anticancer effects by targeting UPR pathways. In this study, we found that either inhibiting or silencing PERK, ATF4 or HSPA5 could enhance the pro-ferroptotic effects of DHA. However, in the absence of DHA, knockdown of these three genes did not observably induce cell death and the production of ROS and MDA in glioma cells. This could be explained by the fact that the anticancer effects of DHA relies on carbon-centered free radicals and ROS generated by the reaction of its endoperoxide bond with iron . In the absence of DHA, the basal levels of ROS and MDA in glioma cells are very low and could not be potentiated by knockdown of PERK, ATF4 and HSPA5. Moreover, UPR is activated and exerts cell protection under ER stress such as DHA treatment instead of normal conditions, which is in line with previous findings [17, 51, 52]. Therefore, interfering this negative feedback by siRNA or small molecules such as PERKi or EGCG would be a promising strategy to enhance the anticancer efficacy of ferroptosis inducers, including DHA.
In summary, we provided the evidence of ferroptosis as a novel anticancer mechanism of DHA in glioma, which could be neutralized by PERK-ATF4-HSPA5-GPX4 pathway activation under ER stress. Therefore, inhibition ER stress or subsequent PERK-ATF4-HSPA5-GPX4 pathway might enhance the anticancer activity of DHA by increasing ferroptosis in glioma.
We would like to thank the Academy of Medical Sciences of Zhengzhou University Translational Medicine Platform for their kind help and support to this work.
Study conception and design: YC, WD, ZZ; data analysis: YM, DW, BH, YS; cell and molecular experiment: XZ, QM, YM, LZ, YS, HJ; animal experiment: HL; manuscript drafting: YC, YM, XZ, WD; manuscript revising: YC, XZ, JX, ZZ. All authors reviewed and approved the final manuscript.
This work was supported by the National Natural Science Foundation of China (81772643, 81772803, 81972479, 81871877, 31501132, 81402281 and 81402187), Henan Medical Program (201602072), Fujian Provincial Department of Science & Technology (2017 J01363), Health and Family Planning Commission of Fujian Province (2017-ZQN-86), Scientific and Technological Planning Project of Guangzhou City (201805010002 and 201904010038) and Natural Science Foundation of Hainan (20158360).
Ethics approval and consent to participate
This study was approved by the Ethical Committee of Zhengzhou University. Written informed consent was obtained from all participants. All human study procedures were carried out in accordance with the ethical standards of the Helsinki Declaration. All of the procedures of animal experiments were and performed in accordance with the Association for Assessment and Accreditation of Laboratory Animal Care guidelines (http://www.aaalac.org).
Consent for publication
The authors declare that they have no competing interests.
- 26.Wang DD, Jin Q, Wang LL, Han SF, Chen YB, Sun GD, Sun SF, Sun SW, Wang T, Liu FJ, et al. The significance of ENAH in carcinogenesis and prognosis in gastric cancer. Oncotarget. 2017;22:8(42).Google Scholar
- 44.Vladykovskaya E, Sithu SD, Haberzettl P, Wickramasinghe NS, Merchant ML, Hill BG, McCracken J, Agarwal A, Dougherty S, Gordon SA, et al. Lipid peroxidation product 4-hydroxy-trans-2-nonenal causes endothelial activation by inducing endoplasmic reticulum stress. J Biol Chem. 2012;287(14):11398–409.CrossRefGoogle Scholar
- 48.Hombach-Klonisch S, Mehrpour M, Shojaei S, Harlos C, Pitz M, Hamai A, Siemianowicz K, Likus W, Wiechec E, Toyota BD, et al. Glioblastoma and chemoresistance to alkylating agents: involvement of apoptosis, autophagy, and unfolded protein response. Pharmacol Ther. 2018;184:13–41.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.