A role of metallothionein-3 in radiation-induced autophagy in glioma cells

Although metallothionein-3 (MT3), a brain-enriched form of metallothioneins, has been linked to Alzheimer’s disease, little is known regarding the role of MT3 in glioma. As MT3 plays a role in autophagy in astrocytes, here, we investigated its role in irradiated glioma cells. Irradiation increased autophagy flux in GL261 glioma cells as evidenced by increased levels of LC3-II but decreased levels of p62 (SQSTM1). Indicating that autophagy plays a cytoprotective role in glioma cell survival following irradiation, measures inhibiting autophagy flux at various steps decreased their clonogenic survival of irradiated GL261 as well as SF295 and U251 glioma cells. Knockdown of MT3 with siRNA in irradiated glioma cells induced arrested autophagy, and decreased cell survival. At the same time, the accumulation of labile zinc in lysosomes was markedly attenuated by MT3 knockdown. Indicating that such zinc accumulation was important in autophagy flux, chelation of zinc with tetrakis-(2-pyridylmethyl)ethylenediamine (TPEN), induced arrested autophagy in and reduced survival of GL261 cells following irradiation. Suggesting a possible mechanism for arrested autophagy, MT3 knockdown and zinc chelation were found to impair lysosomal acidification. Since autophagy flux plays a cytoprotective role in irradiated glioma cells, present results suggest that MT3 and zinc may be regarded as possible therapeutic targets to sensitize glioma cells to ionizing radiation therapy.

Patients with glioblastoma, the most malignant form of glioma, usually succumb within a few years after the diagnosis despite aggressive chemo and radiotherapies following surgery 1 . Such disheartening therapeutic outcomes desperately call for more effective measures. However, untoward biological characteristics of glioma such as aggressive local invasion and resistance to anticancer therapy, present a daunting obstacle in achieving this goal.
Dysregulation of autophagy has been implicated in cancer cell proliferation, invasion, high histological grades, and poor prognosis in a wide spectrum of cancers. These findings highlight the possibility that autophagy-related molecules may prove useful as prognostic markers and/or therapeutic targets. In addition, many of anticancer drugs along with ionizing radiation (IR) are known to activate autophagy in cancer cells. The functional significance of autophagy activation associated with cancer therapeutics is still controversial. A growing body of evidence supports the possibility that autophagy plays a self-defensive cytoprotective role against toxicity induced by anti-cancer therapies [2][3][4] , although opposing data favoring the cytotoxic effect of autophagy are also available 2,3 . Despite these controversies, the current consensus is that the modulation of autophagy may have significant effects in anti-cancer therapeutics. Although a number of intrinsic or extrinsic pathways of autophagy have been delineated, the final destination of all these pathways is the lysosome, where degradation of cargo contents occurs. For this reason, diverse attempts have been made to maximize cancer cell death following the treatments through reducing the rate of autophagy flux with drugs acting on lysosomes 4,5 . For instance, currently, several clinical trials are investigating the effect of the combination treatment with lysosomal inhibitors such as chloroquine (CQ) or hydroxychloroquine (HCQ) and various anticancer drugs or IR in various cancer types (http://clinicaltrials. gov).
Recent evidence indicates that aberrancy in zinc homeostasis is linked to a variety of cancers including breast, esophageal, gastric, colon, nasopharyngeal, prostate, bladder, and skin cancers 6 . As regard to the mechanistic roles of zinc dyshomeostasis in cancer cell proliferation, well known effects of zinc on transcriptional regulation of genes involved in cell growth, proliferation, and apoptosis, and on the oxidative stress have been suggested as such. Since zinc is involved in so many cellular processes, there exist a number of zinc binding proteins in cells. Of many zinc-binding proteins, metallothioneins (MTs) are regarded as the major regulators of cellular zinc. Of four isoforms (MT1-4), the ubiquitously expressed MT1 and MT2 have been most extensively investigated and frequently found to be overexpressed in many cancers such as breast, ovarian, renal, prostate, lung and colorectal cancers as well as soft tissue sarcomas, and associated with poor patient prognosis 6,7 . On the contrary, relatively little is known regarding the role of MT3, a CNS-enriched Mt isoform, in cancers including glioma. Unlike MT1 and MT2, the status of MT3 expression has been found rather inconsistent among different cancer types, upregulated in some cancers (i.e. breast, prostate, urinary bladder, and lung cancers) [8][9][10][11][12] , while downregulated in others (i.e. gastric and esophageal cancers and leukemia) [13][14][15] . Moreover, data regarding the role of MT3 are still scant and contradictory, some suggesting malignant phenotype-promoting 8,9,11,12 or others antitumor effect 15 of MT3. As for glioma, induced expression of MT3 as well as MT1 and MT2 following arsenic trioxide treatment on U87-MG glioblastoma cells was demonstrated 16 , which may be postulated as a potential mechanism for glioma resistance. Importantly, Mehrian-Shai et al. 17 recently reported that, in glioblastoma patients, high expression of including MT3 was associated with poor patient survival whereas low MT levels corresponded to good prognosis, suggesting the prognostic implications of Mts in glioma.
Since MT3 was first identified as a neuronal growth inhibitory factor (GIF) that was deficient in brain extracts of Alzheimer's disease 18 , altered MT3 expression has been also reported in various neurological disorders such as Parkinson's disease, Amyotrophic lateral sclerosis (ALS), Down syndrome, and Creutzfeld-Jakob disease 19 . The role of MT3 in the CNS pathologies appears to be either neuroprotective or cytotoxic depending on the experimental models. The neuroprotective effect of MT3, which is presumably mediated by its metal chelating and antioxidative abilities, was observed in epileptic brain injury, cortical cryoinjury, and a mutant superoxide dismutase 1 mouse model of ALS [20][21][22] . On the other hand, the cytotoxic effect of MT3 has been also demonstrated; intracellular zinc released from MT3 may trigger neuronal and astrocytic cell death [23][24][25] . We have previously demonstrated that MT3 plays a key role in regulating the function of lysosomes in astrocytes, in a zinc-and actin-dependent manner, which effects are not shared by MT1 and MT2 23 . Of note, free zinc may contribute to autophagic flux in neurons and astrocytes, as evidenced by the result that oxidative stress induces accumulation of zinc ions in autophagosomes and lysosomes, and the inhibition of zinc accumulation by chelators or MT3 silencing blocks the increases in autophagy flux 23,26,27 . Although detailed information as to how zinc increases autophagy flux, one possible mechanism may be that lysosomal zinc accumulation is correlated with subsequent lysosomal acidification 28 . Lysosomal acidity has been found to be defective in neurodegenerative conditions such as Alzheimer's disease, in which autophagy flux is arrested 29 .
Based on these findings, we hypothesized that glioma cells, as a cancerous counterpart of glial cells of developing brains, might utilize this novel cellular mechanism involving MT3 and zinc as a route for circumventing the toxicity of IR. Here, we explored this intriguing possibility.

Increases in autophagy flux in irradiated GL261 glioma cells.
To assess the autophagy flux in GL261 glioma cells after irradiation, we measured levels of LC3-II, a marker for autophagy activation, and p62 (SQSTM1), a marker for autophagic/lysosomal degradation. Immunoblots showed that levels of LC3-II gradually increased in GL261 glioma cells following irradiation at 2 Gy, reaching its peak level at 4 h and decreasing thereafter (

Inhibition of autophagy decreases clonogenic survival of irradiated GL261 cells.
To determine the functional significance of autophagy flux increased by irradiation, autophagy flux in irradiated cells was inhibited with pharmacologic inhibitors, BA or 3-methyladenine (3MA), or with siRNA against Beclin1 (beclin). BA or 3MA was given immediately after irradiation, whereas siRNA was given 24 h prior to the irradiation. In each experiment, effective inhibition of autophagy flux was confirmed by an increase in p62 levels on Western blot analysis ( Fig. 2a-c, see Supplementary Fig. 2(a-c) for the original blot). In colony forming assays, as compared with vehicle alone or negative-transfected controls (NC), cells treated with the drugs or beclin siRNA exhibited significantly lesser survival after irradiation ( Fig. 2d-f). Since colony forming assay measures total cell number reflecting not only cell death but cell growth, we measured LDH efflux from dead cells as more direct measure for cell death. Also, we examined other human-derived glioma cell lines, SF295 and U251 cells. In all glioma cell lines, irradiation-induced cell death (LDH release) was increased by above measures inhibiting autophagy flux (Suppl. Fig. 6).

siRNA knockdown of MT3 blocks lysosomal degradation of autophagic vacuoles in irradiated GL261 cells and decreases cell survival.
We have previously demonstrated that MT3 plays a regulatory role in autophagy flux in primary cultured astrocytes 23 . To examine the role of MT3 in GL261 glioma cell line, we downregulated MT3 using siRNA (Fig. 3a). Compared with GL261 cells that were transfected with NC siRNA, GL261 cells that were transfected with MT3 siRNA, showed increases in the levels of both LC3-II and p62 after irradiation (Fig. 3b,c, see Supplementary Fig. 3(b,c) for the original blot), suggesting that autophagy flux was inhibited distal to the autophagosome formation in the MT3-downregulated cells. To further confirm this, we observed RFP-LC3-transfected GL261 cells stained with Lysotracker Green under fluorescence microscope. In MT3 siRNA-transfected cells, both RFP-LC3 fluorescence and Lysotracker fluorescence were increased in lysosomes (Fig. 3d), consistent with the possibility that lysosomal degradation of autophagic vacuole-associated RFP-LC3 was impaired. Subsequent survival analysis demonstrated that MT3 knockdown decreased the survival of GL261 glioma cells after irradiation (Fig. 3e).
Accumulation of labile zinc in lysosomes following irradiation, which is attenuated by MT3 knockdown. As MT3 releases zinc upon various stimuli such as oxidative stress, the fact that MT3 knockdown reduced autophagy flux in irradiated GL261 cells raised a possibility that MT3-dependent zinc dynamics may be involved in this phenomenon. In line with our previous observation of zinc accumulation in lysosomes of neurons and astrocytes under oxidative stress 26,27 , irradiation of GL261 cells induced an elevation of labile zinc levels inside cells within 2 h as visualized with FluoZin-3 fluorescence microscopy; double staining of these cells with Lysotracker Red revealed that, most of zinc fluorescence was localized inside lysosomes (Fig. 4a). As expected from our previous study in cultured cortical astrocytes 23 , knockdown of MT3 markedly attenuated the increase in labile zinc following irradiation (Fig. 4b). In addition, treatment with tetrakis-(2-pyridylmethyl)ethylenediamine (TPEN), a cell-permeant zinc chelator, almost completely blocked the increase in free zinc levels in lysosomes following irradiation (Fig. 4b).
Chelation of intracellular zinc blocks lysosomal degradation of AVs in irradiated GL261 cells and decreases cell survival. The two findings that labile zinc accumulates in lysosomes or autolysosomes following irradiation and that knockdown of MT3, a probable source of zinc, blocks degradation of AVs imply that zinc may have a direct effect on the degradative function of lysosomes, and consequently on cell survival. To test this hypothesis, we examined the effects of depleting or adding labile zinc on these events. In TPEN-treated (zinc-depleted) cells, changes similar to those seen in MT3-downregulated cells were observed. First, the levels of both LC3-II and p62 markedly increased after irradiation (Fig. 5a,b, see Supplementary Fig. 4(a,b) original blot). Second, striking accumulation of dilated RFP-LC3 (+) AVs in lysosomes was observed (Fig. 5c). These findings indicate that lysosomal degradation of AVs was reduced, and hence autophagy flux was arrested as in the case of MT3-knockdown cells. The clonogenic survival of TPEN-treated cells were also reduced after irradiation (Fig. 5d). In contrast to TPEN treatment, addition of zinc chloride (Zn) in the media, increased autophagy flux and lysosomal degradation as indicated by decreased levels of both LC3-II and p62 together with attenuated accumulation of AVs after irradiation (Fig. 5a-c). As expected, raising intracellular zinc with zinc treatment that increased autophagy flux, resulted in the increased cell survival following the irradiation (Fig. 5d).

Impaired lysosomal acidification by MT3 knockdown and zinc chelation in irradiated GL261
cells. In diverse cases of arrested autophagy, lysosomal acidification has been found to be abnormal. To examine this possibility, changes in lysosomal pH were assessed using a fluorescent lysosomal pH indicator dye Lysosensor Yellow/Blue dextran. Consistent with the result showing that MT3 knockdown induced arrested autophagy in GL261 glioma cells, following irradiation MT3 deficient glioma cells exhibited lysosomal pH shifting toward more alkaline direction as compared with that in Negative control transfected cells (Fig. 6a). Similar changes ("alkalinization") in lysosomal pH following irradiation were observed in zinc-depleted (TPEN-treated) www.nature.com/scientificreports www.nature.com/scientificreports/ or NH 4 Cl-treated glioma cells (Fig. 6b). These results support the possibility that zinc and MT3 play a critical role in the maintenance of acidity of lysosomal lumens in these cells.

Discussion
While autophagy has been implicated as an important factor in glioma cell biology, its precise role in radio-sensitivity of glioma has been still controversial as to the end result being beneficial or harmful. The current study presents novel insights into this issue. First, present results demonstrated that MT3, possibly by releasing zinc that enters lysosomes, is a key regulator of autophagy flux that changes in response to irradiation. Second, thusly enhanced autophagy flux may be cytoprotective and increase the survival of glioma cells after irradiation. Therefore, downregulation of MT3 or zinc chelation at the time of irradiation treatment may provide a beneficial effect to glioma patients by blocking the autophagy flux and increasing glioma cell death. Mechanistically, it appears that irradiation releases zinc from MT3, which in turn causes lysosomal acidification and increases autophagy flux and lysosomal degradation. These results are in line with the glioma clinical trials with the lysosomal acidification inhibitors, CQ and HCQ 30 , which have shown a modest efficacy [31][32][33][34][35] . Present results further support this idea, and suggest that the strategy aiming at the blockade of lysosomal acidification, especially targeting MT3 and/or zinc, may prove useful to find a supportive measure to current glioma therapy.
In the present study, to examine the effect of autophagy on irradiation-induced death of GL261 cells, we modulated the autophagy flux by several different methods. First, we used standard autophagy inhibitors such as 3MA, BA, and siRNA against beclin. All these measures inhibited autophagy flux, and resulted in the increased death of GL261 glioma cells following irradiation. Two other human glioma cell lines, SF295 and U251 cells, showed the same effects. Next, we examined the effects of MT3 knockdown with siRNA as well as altering intracellular zinc levels with TPEN or zinc. As MT3 knockdown or zinc depletion produced similar results as the above autophagy inhibitors, we concluded that MT3 and zinc play a key role in positively modulating the autophagy flux.
MTs are a family of small (6-7 kd), intracellular, cysteine-rich proteins with a number of thiol (-SH) groups enabling them to bind metals such as zinc and copper as well as toxic heavy metals. In oncogenesis, MTs are generally regarded to function as tumor suppressors in normal cells via chelating carcinogenic metal ions and maintaining redox homeostasis. On the other hand, MTs are upregulated in diverse cancer cells, and their levels are positively correlated with the degree of malignancy and the resistance to anticancer therapy 36,37 . For example, high levels of MTs confer resistance to platinum chemotherapeutics, and hence are regarded as a marker for poor prognosis 6,7 . The upregulation of MTs may be a response of cancer cells to metabolic stresses associated with aberrant cancer cell biology, and may function as a defense against anti-cancer therapeutics.
There are 4 isoforms of MTs. Of these, MT3 is expressed mainly in the CNS. With approximately 70% sequence homology to other MTs, MT3 contains a unique conserved sequence TCPCP motif at positions 5-9 in the N-terminus. A characteristic conformational organization of the β domain by virtue of the TCPCP motif of MT3 in its zinc-bound form is suggested to provide a potential interface for protein-protein interactions, which may be responsible for the distinct biological functions of MT3 such as neuronal growth inhibitory activity 38,39 . Other biological functions of MT3 include roles in glycolytic metabolism, protein chaperone and scaffolding functions, metal transport/buffering, and redox signaling 25,40,41 . We have previously demonstrated that MT3 plays a role in the maintenance of autophagy flux in cultured cortical astrocytes, likely by regulating lysosomal functions 23 . Similar to the findings obtained in primary astrocytes, in the present study, siRNA knockdown of MT3 in GL261 glioma cells induced arrested autophagy following irradiation, likely due to impaired lysosomal degradation. Our results indicated that release of free zinc from MT3 may be a key event herein, since zinc chelation produced similar effects. Both MT3 knockdown and zinc chelation induced significant alkalinization of lysosomes, which should inhibit acidic hydrolase activities in lysosomes. Conversely, supplementation of zinc induced lysosomal reacidification. Although it is unclear how intracellular zinc release and its accumulation in lysosomes help acidify lysosomes, one plausible speculation is that a certain Zn 2+ /H + antiporter may be functioning in lysosomes. Hence, zinc inside lysosomes may provide energy for proton to enter the lysosomal lumen. Alternatively, zinc may indirectly upregulate the expression of lysosomal proton pumps such as the vacuolar-type ATPase. These intriguing possibilities remain to be tested.
The cytoprotective effect of MT3 in glioma cells has been reported by others. For instance, elevated MT3 expression in U87-MG glioblastoma cells than in normal astrocytes and other glioma cell line was correlated with an inactive conformational change of p53, resulting in attenuated apoptosis 17 . Hence, MT3 may contribute to glioma survival through diverse ways including increases in autophagy flux through lysosomal acidification. Further studies will be needed to elucidate the comprehensive mechanisms underlying the cytoprotective effect of MT3 on glioma resistance.
In summary, we demonstrated the roles of MT3 and zinc in radiation-induced autophagy and radioresistance in glioma cells. MT3 appears to be a key regulator for autophagy flux via zinc-dependent lysosomal acidification, and hence contributes to resistance of glioma cells to irradiation treatment. As such, MT3 may prove to be a suitable target in improving the efficacy of irradiation treatment in gliomas. irradiation. Cells were irradiated using a 6-MV photon beam linear accelerator system (Varian Medical Systems, Palo Alto, CA, USA). For radiation dose titration, a single-dose of 2, 5 or 10 Gy was tested for autophagy induction and cell survival. Having identified that 2 Gy of radiation was sufficient for autophagy induction and optimal for cell survival analysis (data not presented), 2 Gy was used for experiments. plate colony forming assay. Cells were seeded in six-well plates at a density of 100 (for controls) or 200 cells (for irradiated cells) per well and allowed to attach overnight before irradiation. Fourteen days after irradiation, cells were fixed with 4% paraformaldehyde and stained with Giemsa staining reagent (Sigma). The number of colonies with > 50 cells was counted under a dissecting microscope. The cell survival fraction or relative survival rate was expressed in terms of the plating efficiency in control groups: the number of colonies formed after treatment divided by both the number of cells seeded and the plating efficiency. Each assay was performed in triplicate.
LDH efflux assay for cell death. Cells death was quantitatively assessed by measuring lactate dehydrogenase (LDH) activity released into the culture medium from damaged cells, as described previously 43 . LDH concentrations were normalized to the mean concentration of cells exposed to 400 μM H 2 O 2 (100%) after subtracting the mean value of sham-washed control cells (0%).
Western blot analysis. Cells were washed with ice-cold PBS and lysed in PRO-PREP protein extraction solution (iNtRON, Sungnam, Korea). Cell lysates were separated on 12-15% SDS-PAGE gel and transferred to polyvinylidene difluoride membranes. The membranes were blocked and probed with antibodies against LC3, www.nature.com/scientificreports www.nature.com/scientificreports/ p62, beclin1 or β-actin for 12 h at 4 °C. After washing, blots were incubated with appropriate HRP-conjugated secondary antibodies (Cell Signaling) for 1 h at room temperature, and were washed. The blots were developed with SuperSignal West Pico Chemiluminescent Substrate kit (Thermo Scientific, Waltham, MA, USA). The band intensity was analyzed using ImageJ (NIH, Bethesda, MD, USA). All experiments were triplicated.
Real-Time reverse transcriptase PCR. Total RNA was extracted from GL261, SH295 and U251 cells with Qiazol Reagent (Qiagen, Hilden, Germany). ImProm-2 reverse transcriptase kit (Promega, Madison, WI, USA) was used to generate cDNA according to the manufacturer's instructions. Samples were subjected to real time PCR amplification using forward and reverse primers for MT3, LightCycler 480 SYBR Green 1 Master and LightCycler 480 machine (Roche, Mannheim, Germany). The following thermocycle conditions were used: an initial cycle at 95 °C for 5 min; 50 cycles, each at 95 °C for 10 s, 60 °C for 20 s, and 72 °C for 20 s; and a final cycle at 95 °C for 15 min. Melting curve analysis was performed on all PCR products to ensure that specific PCR products were generated. Data were analyzed using the comparative cycle threshold method. The levels of mRNA were normalized to those of the housekeeping gene Gapdh (glyceraldehyde-3-dehydrogenase), and expressed as relative fold changes. confocal microscopy. 1 × 10 5 GL261 cells were seeded onto a 0.1 mg/mL poly-L-lysine-coated coverslip in 4-well plates and incubated for 12 h. After irradiation, cells were stained with 100 nM Lysotracker Red DND-99 or Lysotracker Green DND-26 (Invitrogen, Carlsbad, CA, USA) for 1 h or 10 μM FluoZin-3 (Invitrogen) for 30 min before fixation. For detection of changes in lysosomal pH, cells were stained with 0.5 mg/mL Lysosensor Yellow/Blue dextran (Invitrogen), a fluorescent lysosomal pH indicator dye, for 1 h. Then, cells were washed with PBS twice and were fixed with 4% paraformaldehyde. For counterstain, the fixed cells were stained with 4′,6-diamidino-2-phenylindole (DAPI). Coverslips were mounted with Fluoromount G (Southern Biotech, Birmingham, AL, USA), and were examined by LSM 710 confocal microscope (Carl Zeiss, Dublin, CA, USA).

RFP-LC3 plasmid transfection. The RFP-LC3 expression plasmid was a generous gift from Drs. Maria
Colombo and Michel Rabinovitch (Universidad Nacional de Cuyo, Mendoza, Argentina). Plasmid transfection was performed using Effectene (Qiagen) according to the manufacturer's instructions. After 24 h transfection, cellular expression of RFP-LC3 was confirmed by Western blot analysis.
Statistics. All data presented are representative from at least 3 independent experiments. The numeric data are presented as means ± SEMs and evaluated using the Student t tests. P < 0.05 was considered statistically significant.