Arginine deprivation induces prolonged autophagy activation and ROS-dependent cell death in glioblastoma multiforme Authors: Maya Abdelkhalek, Yasmin El-Jawhari, Maha Shaheen, Rawan Saad, Majd Sarieddine, Nour El-Husseini, Mirvat El-Sibai, Ralph J Abi-Habib Published: 2026 ABSTRACT AbstractBackgroundArginine deprivation was shown to be selectively cytotoxic to GBM cells with cell death being caspase-independent, non-apoptotic. In this study, we assess the impact of prolonged arginine deprivation on GBM cells, the potential activation of autophagy, the mechanism of its activation, and its impact on cell cytotoxicity.MethodsGBM cell lines A172 and U251 were used in this study and arginine deprivation was induced by a cobalt-substituted, PEGylated human arginase I [HuArgI (Co)-PEG5000].ResultsArginine deprivation induced prolonged cytotoxicity up to 120 hours post-treatment. Cytotoxicity persisted despite cells overexpressing ASS-1. Arginine deprivation also induced a marked and sustained activation of autophagy, starting at 24 hours and lasting up to 120 hours post-treatment. Autophagy was shown to induce cell death since its inhibition by chloroquine significantly decreased cytotoxicity, indicating that the extensive activation of autophagy is resulting in cell death of GBM cells. Moreover, ROS was shown to accumulate in treated cells and neutralizing ROS using NAC did not affect the autophagic response but led to a complete reversal of cytotoxicity, demonstrating that death by autophagy is dependent on ROS generation.ConclusionArginine deprivation leads to sustained activation of autophagy and subsequent ROS-dependent cell death in GBM cells.Keywords: Glioblastoma multiforme (GBM), arginine auxotrophy, arginine-deprivation, [HuArgI (Co)-PEG5000], ASS-1, autophagy, ROS, autophagic cell death Introduction Glioblastoma multiforme (GBM) is characterized by high mortality and low survival rates secondary to poor prognosis and the lack of effective treatment options. Hence, there is an urgent need for novel, effective, and selective therapeutic strategies for targeting GBM [1–4]. One such approach consists of targeting arginine auxotrophy in GBM cells. Arginine is a semi-essential amino acid synthesized from L-citrulline and aspartate, using the urea cycle enzymes argininosuccinate synthetase 1 (ASS-1) and argininosuccinate lyase (ASL) [5,6]. We and others have shown that a large number of tumor types, including GBM, lack ASS-1 expression leading to dependence on extracellular sources of arginine, hence becoming auxotrophic for this amino acid and unable to survive its deprivation. In addition, we have shown that even tumors that do express ASS-1 do not do so at levels that allow them to meet their needs and are still dependent on extracellular source of arginine [7–9]. One of the most common approaches used to degrade extracellular arginine is the human arginase I enzyme, which degrades extracellular arginine to ornithine and urea. Native human L-arginase I (HuArgI) was first used for targeting tumor arginine auxotrophy, but it showed limited potency in patients secondary to a very short serum half-life and a low saturation constant, which was due to the rapid loss, in serum, of the two Mn2+ ion cofactors of the enzyme [5]. The native enzyme was modified by replacing the Mn2+ ions in the active site with Co2+ ions, which resulted in an enhanced serum stability and a higher catalytic rate. Addition of polyethylene glycol (PEG), further increased serum stability and decreased immunogenicity, leading to the final PEGylated, Cobalt-substituted version of the enzyme [HuArgI (Co)-PEG5000] [10–12]. We and others, have demonstrated the effectiveness and selectivity of arginine deprivation in many tumor types including pancreatic carcinoma, colorectal carcinoma (CRC), hepatocellular carcinoma, acute lymphoblastic leukemia, ovarian carcinoma, breast carcinoma, acute myeloid leukemia (AML), renal cell carcinoma, melanoma, prostate cancer and GBM, confirming the potential of this novel therapeutic approach for selectively targeting arginine-auxotrophic tumors [11,13–27]. In a previous study, we have shown that GBM cell lines were auxotrophic for arginine and that [HuArgI (Co)-PEG5000]-induced arginine deprivation was selectively cytotoxic to GBM cells while sparing normal fetal glial cells. Treatment of GBM cells with [HuArgI (Co)-PEG5000] seemed to induce non apoptotic cell death in addition to cell cycle arrest in the surviving fraction of cells [22]. We had also demonstrated that completely auxotrophic GBM cells, e.g., A172, do not express ASS-1 and are not rescued by the addition of excess exogenous arginine, while partially auxotrophic cells, e.g., U251, do express ASS-1 and are rescued by the addition of exogenous L-citrulline with both being sensitive to arginine deprivation. In addition, knockdown of ASS1, in the ASS1-expressing U251 cells, shifted their phenotype into that of completely auxotrophic cells [22]. Furthermore, we had shown that arginine-deprivation may lead to the activation of autophagy at very early time points with no clear indication as to the impact of its activation [22]. In this study, we aim to investigate the long-term impact of arginine deprivation on the cytotoxicity, and ASS-1 expression of GBM cells, as well as on the extent of autophagy activation and its potential contribution to cell death following [HuArgI (Co)-PEG5000]-induced arginine deprivation. Arginine deprivation is a promising approach for the selective targeting of tumors that exhibit arginine auxotrophy. Studies have demonstrated potency and selectivity of arginine deprivation in several tumor types, including GBM, which we were the first to investigate. It has been established that autophagy is activated in tumors following arginine deprivation, however, the extent of activation as well as its impact on cell survival and the potential mechanism through which it potentially induces cell death are still unknown. In this study, we show that arginine deprivation induces a sustained activation of autophagy in GBM cells for up to 5 days post-treatment. We also show that the activation of autophagy is not protective but leads to cell death through a ROS-dependent mechanism. This is critical because it provides the opportunity of combination therapy of arginine deprivation with other autophagy-inducing treatments in patients with GBM. Materials and methods Expression and purification of [HuArgI (Co)-PEG5000] PEGylated recombinant Arginase I cobalt [HuArgI (Co)-PEG5000] was provided by Aeaglea Biotherapeutics (Austin, TX, USA). Cell lines Human Glioblastoma cell lines A172 and U251 MG were obtained from the American Type Culture Collection (A172 ATCC, CRL-1620) and from the European Collection of Authenticated Cell Cultures (U251 MG, ECACC, catalog # 09063001). Both cell lines were grown in DMEM (Lonza, Basel, Switzerland) culture media supplemented with 10% Fetal Bovine Serum (FBS) (Sigma-Aldrich) and 100 U/mL penicillin/streptomycin (Biowest) at 37 °C/5% CO2. Cytotoxicity assay The cytotoxicity assay was carried out as described previously [13]. Briefly, aliquots of 104 cells/ml in culture medium, were plated (100 µL/well) in a flat-bottom 96-well plate (Corning Inc. Corning, NY). This was followed by the addition of 50 µL of the drug solution in media to each well to yield concentrations ranging from 10−7 to 10–13 M. The different conditions were set up in triplicates for each concentration. Chloroquine (CQ) was added at a concentration of 50 µM and N-Acetylcysteine (NAC) was added at a concentration of 0.82 g/L. Following a 24, 48, 72, 96, or 120 h incubation at 37 °C/5% CO2, 40 µl of XTT cell proliferation reagent (Roche, Basel, Switzerland) were added to each well and the plates were incubated for another 4h. Absorbance was then read at 450 nm using a Varioskan Flash plate reader (Thermo Fisher Scientific, Waltham, MA). Percent maximal fluorescence (Percent Control), were plotted against the log of concentration and a non-linear regression with a variable slope sigmoidal dose-response curve was generated using GraphPad Prism 10 software (GraphPad Software, San Diego, CA) and the IC50 (inhibitory concentration 50) of [HuArgI (Co)-PEG5000] alone or in the presence of CQ, or NAC were calculated. To convert the raw data to percent maximal fluorescence (Percent control), the raw fluorescence values of replicates are first averaged, then converted to a percentage of the maximal fluorescence value, which is that of untreated cells. ASS-1 expression Expression of ASS-1 was determined as described previously [13]. Briefly, aliquots of 1 × 106 cells/well in 2 mL cell culture medium were plated in a flat-bottom 6-well plate (Corning Inc. Corning, NY). Cells were either left untreated or were treated with [HuArgI (Co)-PEG5000] at a concentration of 10−7 M. Plates were incubated for 24, 48, 72, 96 and 120 hours at 37 °C/5% CO2. Cells were then collected and incubated in antibody binding buffer consisting of 0.05% Triton-X 100 along with a 1/200 dilution of anti-ASS-1 rabbit monoclonal antibody (Cell Signaling Technology, catalog # 54846) for 1 h at 37 °C, followed by a 30-min incubation with a 1/200 dilution of a FITC-conjugated goat anti-rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, catalog # sc-2012). Isotype control consisted of cells incubated with a Rabbit monoclonal antibody IgG (Cell Signaling Technology, catalog # 3900) and a FITC-conjugated goat anti-rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, catalog # sc-2012). Cells were washed once and read using a C6 flow cytometer (BD Accuri, Ann Arbor, MI). ASS-1 expression was determined using the ratio of fluorescence intensity (RFI) between the mean fluorescence intensity (MFI) of the stained cells and the MFI of the isotypic control. RFI ≥ 2.0 was considered positive. Autophagy assay Autophagosomes were detected using the CytoID assay on flow cytometry as described previously [17]. Briefly, aliquots of 1x106 cells/well in 2 mL cell culture medium were plated in a flat-bottom 6-well plate (Corning Inc. Corning, NY). [HuArgI (Co)-PEG5000] was added at a concentration of 10−7 M, with or without chloroquine (50 µM) or NAC (0.82 g/L). Plates were incubated for 24, 48, 72, 96 and 120 hours at 37 °C/5% CO2. Assay was carried out according to the company’s manual (ENZO product manual - CYTO-ID® Autophagy Detection Kit, 2016 - Catalog No. ENZ-51031 - Part D. Live Cell Analysis by Flow Cytometry). Results were read on a C6 flow cytometer (BD Accuri, Ann Arbor, MI). ULK1 phosphorylation Phosphorylation of ULK1 was detected as described previously [17]. Briefly, cells were plated in flat-bottom 12-well plates (5 × 105 cells/well) (Corning Inc.) and left untreated or treated with 10−7 M [HuArgI (Co)-PEG5000]. Plates were then incubated at 37 °C with 5% CO2 and harvested at 24, 48, 72, 96, and 120 hours. Cells were fixed in 4% formaldehyde for 15 minutes and permeabilized with ice-cold 90% methanol for a minimum of 10 minutes on ice. Cells were incubated in antibody dilution buffer containing 0.5% BSA, along with an anti-P-ULK1 rabbit monoclonal antibody (Cell Signaling Technology, catalog # 14202) (1:100) for 1 hour at room temperature. This was followed by a 30-minute incubation with a 1:100 dilution of a FITC-conjugated goat anti-rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, catalog # sc-2012). Isotype control consisted of cells incubated with a Rabbit monoclonal antibody IgG (Cell Signaling Technology, catalog # 3900) and a FITC-conjugated goat anti-rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, catalog # sc-2012). Samples were analyzed using a C6 flow cytometer (BD Accuri, Ann Arbor, MI). Phosphorylation of ULK1 was determined using the ratio of fluorescence intensity (RFI) between the MFI of the stained cells and the MFI of the isotype control. RFI > 2.0 was considered positive. Western blot Western blot was carried out as described previously [15]. Briefly, protein samples were separated on SDS-PAGE gels (10%). Proteins were then transferred to PVDF membranes (Bio-Rad), that were then blocked with 5% Bovine Serum Albumin (BSA) (Sigma-Aldrich A2153) for 1 hour at room temperature. Membranes were subsequently incubated with primary Rabbit monoclonal antibodies against Atg13 (1/1000) and phospho-Atg13 (1/1000) overnight at 4 °C or for 2 hours at room temperature (Cell Signaling Technology, catalog # 2983, 5536, 13468, and 43533, respectively). Membranes were then washed and incubated with Anti-Rabbit IgG HRP Conjugated secondary antibody (1:2500) (Promega, catalog # W401B) for 1 hour at room temperature. Bands were visualized by adding enhanced chemiluminescence western-blotting detection reagents (ECL, Bio-Rad). Development was done using Bio-Rad Molecular Imager® (ChemiDoc™XRS+) Imaging System. The levels of protein expression were compared by densitometry using the ImageJ software. Intracellular ROS assay Detection of cytosolic ROS was carried out as indicated by the protocol of the CellROXTM Deep Red Flow Cytometry Assay Kit (Invitrogen). Briefly, Aliquots of 5x105 cells/well, in 1 mL cell culture medium, were plated in a flat-bottom 12-well plate (Corning Inc. Corning, NY). Untreated cells and cells treated with 10−7 M [HuArgI (Co)-PEG5000], 200 μM TBHP (positive control), and 3 mM NAC (negative control) were incubated for 24, 48, and 72h, at 37 °C/5% CO2. Cells were then harvested and incubated with 500 nM of the CellROX® Deep Red reagent and incubated for 60 minutes at 37 °C/5% CO2. Samples were then analyzed by flow cytometer (Accuri C6, BD Biosciences), using 635-nm excitation for the CellROX® Deep Red reagent. Fluorescence emission was collected by the FL4-H channel Statistical analysis All experiments were carried out in triplicates. The mean and standard error of mean were calculated using GraphPad Prism 10 software (GraphPad Software, San Diego, CA). The combined averages of both cells were calculated and graphs plotted using GraphPad Prism. Statistical analysis was carried out using unpaired Welch’s t test. Significance was determined with p < .05. Results Arginine deprivation induces prolonged cell death and overexpression of ASS1 In this study, we first sought to determine the impact of prolonged arginine deprivation on cell survival and expression of ASS-1. Prolonged arginine deprivation led to sustained cell death and to the overexpression of ASS-1, for up to 120 hours post-treatment (Figure 1). The IC50 values of A172 and U251 cells decreased over time to reach a minimum at 120 hours post-treatment (Table 1, Figure 1(A)). Combined average MFI values for ASS1 staining combined showed a statistically significant increase at 72-, 96- and 120-hours post-treatment, compared to untreated cells (p = .0126, .0198, and .0455, respectively), indicating that ASS1 is overexpressed in GBM cells following prolonged arginine deprivation (Figure 1(B,C)). Cytotoxicity of GBM cell lines A172 and U251 following prolonged arginine deprivation (A). Average ASS-1 levels, expressed as mean fluorescence intensity (MFI), for each condition (isotype, untreated and treated cells) at 24, 48, 72, 96, and 120 hours post-treatment. Stars indicate a significant increase in ASS-1 expression compared to both isotype control and untreated cells (B) ASS-1 expression of A172 and U251 cell lines at 72 hours post-treatment. MFI is on the X-axis (FL1-H) and forward scatter is on the Y-axis (C). Table 1.Cytotoxicity of [HuArgI (Co)-PEG5000] to A172 and U251 cell lines at 24, 48, 72, 96, and 120 hour time points.Cell lineshArgI (24 hours) IC50 (pmol/L)hArgI (48 hours) IC50 (pmol/L)hArgI (72 hours) IC50 (pmol/L)hArgI (96 hours) IC50 (pmol/L)hArgI (120 hours) IC50 (pmol/L)A172>10,000176656273242U251>10,000792479535221 Cytotoxicity of [HuArgI (Co)-PEG5000] to A172 and U251 cell lines at 24, 48, 72, 96, and 120 hour time points. This data demonstrates that in GBM cells treated with [HuArgI (Co)-PEG5000] overexpression of ASS-1 does not affect sensitivity of cells to arginine deprivation. We hypothesize that this is due to the marked and sustained activation of autophagy, which initiates an autophagic cell death program that is not affected by the subsequent overexpression of ASS-1. Arginine deprivation leads to sustained activation of autophagy To investigate whether arginine deprivation leads to sustained autophagy activation in GBM cells, we examined autophagosome accumulation on flow cytometry following treatment with [HuArgI (Co)-PEG5000] for up to 120 h (Figure 2). Arginine deprivation led to the accumulation of autophagosomes starting at 48 h and lasting up to 120 h post-treatment, compared to untreated cells, as illustrated by the significant increase in the combined average MFI of treated cells compared to controls (p = .0417, .0138, .0324, and .0348, respectively), indicating that arginine deprivation induces a robust and sustained activation of autophagy in GBM cells, lasting for up to 5 days post-treatment (Figure 2(A)). The activation of autophagy in GBM cells was also confirmed by showing autophagosome accumulation following arginine deprivation using immunofluorescence (Figure 2(C)). In addition, the downstream autophagy inhibitor Chloroquine (CQ), was used alone or in combination with [HuArgI (Co)-PEG5000], to illustrate the further accumulation of autophagosomes in GBM cells subjected to arginine deprivation and in which autophagy has been blocked (Figure 2(B,C)). Autophagosome accumulation in GBM cells, expressed as mean fluorescence intensity (MFI), for control untreated cells and cells treated with [HuArgI (Co)-PEG5000] at 24, 48, 72, 96, and 120 hours post-treatment. Stars indicate a significant increase in autophagosome accumulation compared to untreated cells (A). Autophagosome accumulation in A172 and U251 cell lines at 72 hours post-treatment in untreated cells and cells treated with CQ, [HuArgI (Co)-PEG5000] and a combination of both. MFI is on the X-axis (FL1-H) and forward scatter is on the Y-axis (B). Immunohistochemistry staining of autophagosomes in A172 cells untreated and treated with CQ, [HuArgI (Co)-PEG5000], or a combination of both, at 24, 48, and 72 hours, post-treatment (C). To further confirm the sustained activation of autophagy, we assessed the activation of the upstream autophagy modulators Atg13 and ULK by examining their phosphorylation levels using Western blot and flow cytometry (Figure 3). Figure 3(A) shows a significant decrease in the phosphorylation of Atg13, at both 72 and 96 hours post-treatment, as evidenced by a decrease in the average ratio of phosoho-Atg13 to Atg13 in treated compared to untreated cells (p = .0025 and .0013 at 72 and 96 hours, respectively). Similarly, Figure 3(B) shows a significant decrease in the phosphorylation of ULK, starting at 48 hours and lasting up to 120 hours post-treatment, as evidenced by the significant decrease in the combined average of MFI of treated compared to untreated cells (p = .028, .0048, .0096, and .012 at 48, 72, 96 and 120 hours post-treatment, respectively). Western blot of Atg13 and phosphor-Atg13 on protein extracts of A172 and U251 untreated and treated with [HuArgI (Co)-PEG5000] for up to 96 hours (Atg13) (A). Histogram of the ratio of p-Atg13 over Atg13 showing a significant decrease in the phosphorylation of Atg13 at 72- and 96-hours post-treatment. Stars indicate statistical significance (A). ULK phosphorylation levels expressed as mean fluorescence intensity (MFI), for untreated cells and cells treated with [HuArgI (Co)-PEG5000] at 24, 48, 72, 96, and 120 hours post-treatment. Double stars indicate significantly lower ULK phosphorylation in treated compared to untreated cells (B). ULK phosphorylation is also shown in A172 and U251 cell lines at 24- and 48-hours post-treatment in isotype controls, untreated cells and cells treated with [HuArgI (Co)-PEG5000]. MFI is represented on the X-axis (FL1-H) and forward scatter is represented on the Y-axis. Last panel on the right is a histogram representing the overlayed phospho-ULK MFI of the isotype control (red), untreated cells (black) and cells treated with [HuArgI (Co)-PEG5000] (blue) (B). Cytotoxicity of [HuArgI (Co)-PEG5000] alone and in combination with chloroquine (CQ) (50 µM) to A172 and U251 cells at 48, 72, 96, and 120 hours (C). These findings confirm that arginine deprivation leads to a robust and sustained activation of autophagy, lasting up to 120 hours following arginine deprivation. Activation of autophagy contributes to GBM cell cytotoxicity To investigate the impact of the sustained activation of autophagy on GBM cell viability following arginine deprivation, we inhibited autophagy, using the downstream autophagy inhibitor chloroquine (CQ), in cells treated with [HuArgI (Co)-PEG5000]. Co-incubation with CQ induced a marked decrease in the cytotoxicity of [HuArgI (Co)-PEG5000]-induced arginine deprivation to GBM cells at all-time points, with IC50 values increasing by several folds in cells exposed to the co-treatment compared to cells treated with [HuArgI (Co)-PEG5000] alone (Figure 3(C)). This shows that the activation of autophagy, following arginine deprivation, contributes to cell death in GBM cells, hence potentially resulting in death by autophagy. Autophagy-dependent cell death is mediated by ROS Amino acid deprivation has been shown to lead, though different mechanisms, to the generation of cytosolic reactive oxygen species (ROS) that would eventually contribute to cell death, if not effectively neutralized. To investigate whether the activation of autophagy and/or the resulting cell death following arginine deprivation are ROS dependent, we first tested the potential increase in intracellular ROS levels following treatment, then we investigated the impact of the ROS scavenger, N-acetyl cysteine (NAC) on cells deprived of arginine and assessed its effect on autophagosome accumulation and on the resulting cell death. Arginine deprivation of GBM cells led to the accumulation of intracellular ROS at 48- and 72-hours post-treatment. The combined average of both the percentage of positive ROS cells and the MFI was significantly higher in cells treated with [HuArgI (Co)-PEG5000] compared to untreated cells (p = .039 and .0001 for the percentage of positive cells and p = .012 and .0027 for MFI, at 48 and 72 hours, respectively. Co-incubation with NAC did not lead to any significant decrease in the extent of autophagosome accumulation, compared to cells treated with [HuArgI (Co)-PEG5000] alone (Figure 4(A,B)), potentially indicating that autophagy activation is independent of ROS generation, which may also indicate that the generation of ROS could be occuring downstream of the activation of autophagy. However, addition of NAC to cells deprived of arginine led to the reversal of cytotoxicity and the complete rescue of GBM cells from cell death induced by arginine deprivation (Figure 4(C)). This indicates that autophagic cell death that occurs following arginine deprivation in GBM cells is mediated through the generation of a consequent pool of ROS that potentially takes place downstream of the activation of autophagy and upstream of cell death. Intracellular ROS accumulation in GBM cells following treatment with [HuArgI (Co)|-PEG5000], expressed as both percent positive cells (left panel) and mean fluorescence intensity (MFI) (right panel). Stars indicate a significant increase in ROS accumulation. The bottom histograms show ROS accumulation in treated cells (red) compared to controls (black) in A172 and U251 cells. (A) Autophagosome accumulation in GBM cells, expressed as MFI, for control untreated cells and cells treated with [HuArgI (Co)-PEG5000] alone or a combination of [HuArgI (Co)-PEG5000] and N-acetylcysteine (NAC) at 24, 48, 72, 96, and 120 hours post-treatment (B). Autophagosome formation in A172 and U251 cell lines. Control untreated cells (left panel) and cells treated with [HuArgI (Co)-PEG5000] (10−7 M) (middle panel) or treated with [HuArgI (Co)-PEG5000] and NAC (0.82 g/L) (right panel) at 24, 48, 72, 96, and 120-hours post-treatment. In the histogram (right panel), control cells are in black, cells treated with [HuArgI (Co)-PEG5000] are in red and cells treated with the combination with NAC are in blue (C). Cytotoxicity of A172 and U251 cells treated with [HuArgI (Co)-PEG5000] alone or in combination with N-acetylcysteine (NAC) (0.82 g/L) (D). Discussion The use of arginine depleting enzymes for the treatment of a range of different tumors is increasingly studied in vitro, in vivo, and in clinical trials [12,28–32]. The full extent of arginine auxotrophy in GBM cells in response to [HuArgI (Co)-PEG5000]-induced arginine deprivation was previously established in our lab. We had shown that GBM cell lines were either partially or completely auxotrophic for arginine depending on the extent to which they express ASS-1. Both completely auxotrophic and partially auxotrophic cells were sensitive to [HuArgI (Co)-PEG5000]-induced arginine deprivation but only partially auxotrophic cells could be rescued by the addition of excess exogenous L-citrulline [22]. We had also shown that arginine deprivation was selectively cytotoxic to GBM cells since it had no effect on the SVG p12 human fetal glial cell [22]. This demonstrated the potential use of [HuArgI (Co)-PEG5000]-induced arginine deprivation as a selective targeted therapy for GBM. When analyzing cell death, we found that cytotoxicity induced by [HuArgI (Co)-PEG5000] is non-apoptotic, and caspase-independent as indicated by the absence of caspase activation in GBM cells treated with [HuArgI (Co)-PEG5000] [22]. These results are consistent with what we obtained in other types of tumors including pancreatic cancer, AML and ovarian cancer [13,15,19]. Other groups have shown signs of apoptotic cell death following arginine deprivation, however, this was shown in different tumor types, such as hepatocellular carcinoma and ALL, and thus the difference in the mechanism of cell death can be attributed to the difference in the tumor types investigated [6,10]. However, in the previous study, we did not test the prolonged impact of arginine deprivation, nor did we determine the impact of prolonged arginine deprivation in ASS-1 expression in GBM cells [22]. Therefore, in this study we decided to investigate the long-term cytotoxicity of arginine deprivation and its impact on ASS-1 expression in GBM cells. In addition, given that autophagy is known to be activated in response to amino acid deprivation, we aimed to investigate the activation of autophagy and its impact on GBM cells following arginine deprivation by [HuArgI (Co)-PEG5000]. Arginine deprivation led to a marked, sustained and prolonged activation of autophagy that lasted up to 120 hours post-treatment, indicating that arginine deprivation in GBM cells induces long term activation of autophagy. Given that arginine deprivation significantly increases the autophagic flux, and because of the absence of any signs of death by apoptosis, we sought to investigate the role played by autophagy in the cytotoxic response of GBM cells to [HuArgI (Co)-PEG5000]. Inhibition of autophagy, using the downstream autophagy inhibitor chloroquine, led to a decrease in cell sensitivity to [HuArgI (Co)-PEG5000]-induced arginine deprivation, starting at 72 h and lasting up to 120 h post treatment. This shows that prolonged autophagy activation following arginine deprivation contributes to cell death, hence, the mechanism of cell death induced by [HuArgI (Co)-PEG5000] in GBM may be death by autophagy. These results are consistent with the results that we and other groups have obtained in several tumor types showing that autophagy is mechanism of cell death in several tumor types following different treatment conditions, including arginine deprivation [19,22–35]. However, another set of published data has documented a protective role for autophagy which is activated by several therapeutic agents including arginine depleting enzymes [36–39]. This paradox in the role of autophagy activation, whether protective or deleterious, is not resolved yet, with publications reporting both cases in different tumor types and in response to different treatments. This difference in outcomes could be related to the extent of autophagy activation and the difference in the cellular context [37,40,41]. Furthermore, in order to investigate the mechanism of autophagy-mediated cell death in GBM cells following arginine deprivation, we aimed to investigate the potential involvement of reactive oxygen species (ROS) in the mechanism of cell death. We first showed the accumulation of cellular ROS at 48 and 72 hours in treated cells, compared to controls. Then we showed that the neutralization of ROS, using the ROS-scavenger NAC, does not affect the activation of autophagy in cells deprived of arginine while significantly reducing their sensitivity to arginine deprivation. This may indicate that ROS generation occurs downstream of the activation of autophagy and upstream of cell death, opening the possibility that cell death in GBM cells, following arginine deprivation, is mediated by the accumulation of cellular ROS. Moreover, accumulating data are showing the ability of various cancer cells to overexpress ASS-1 following arginine depletion [11,12,27,42,43]. Similarly, we have shown in our study that GBM cells overexpress ASS-1 following arginine deprivation induced by [HuArgI (Co)-PEG5000]. It is well established that ASS-1 overexpression is highly correlated with poor prognosis and considered as a resistance mechanism to treatments including arginine deprivation [11,12,38,42,44]. However, our results show sustained cell death in GBM cells that lasted up to 5 days following arginine restriction, hence being unaffected by the over expression of ASS-1. This indicates that when autophagy is extensively activated for a prolonged period, the increased ASS-1 levels cannot rescue the cell from death by autophagy. Given that the extent of autophagy activation seems to be critical to the effectiveness of arginine deprivation then combining this approach with any therapeutic that leads to a further activation of autophagy, such as an mTOR inhibitor, would lead to increased potency, hence a more pronounced tumor shrinkage. This may impact the future development of arginine deprivation, and other therapeutic approaches based on amino acid deprivation, by combining them to approaches that increase the flux of autophagy potentially leading to synergistic effects. Conclusion In this study, we have shown that [HuArgI (Co)-PEG50000]-induced arginine deprivation leads to a marked and sustained activation of autophagy in GBM cells. We have also shown that autophagy is not protective but leads to ROS-dependent cell death. Supplementary Material Supplemental Material Funding Statement This study was funded by the intramural resources of the Lebanese American University (President Intramural Research Fund (PIRF) Project). Ethical declaration This study does not include the use of animals or the participation of human subjects. Therefore, it did not require prior IRB or ACUC approval. Author contributions Maya Abdelkhalek: Investigation, Methodology, Data curation, Writing – original draft. Yasmin El-Jawhari: Investigation, Methodology, Data curation. Maha Shaheen: Investigation, Data curation, Project administration. Rawan Saad: Investigation, Methodology, Majd Sarieddine: Investigation, Methodology. Nour El-Husseini: Investigation, Methodology. Mirvat El-Sibai: Conceptualization, Formal analysis, Writing review & editing. Ralph Abi-Habib: Conceptualization, Funding acquisition, Project administration, Supervision, Formal analysis, Writing – review & editing Disclosure statement The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose. Data availability statement The authors confirm that the data supporting the findings of this study are available within this article’s figures and any additional data or raw data can be provided by the corresponding author upon request. References Associated Data Supplementary Materials Supplemental Material Data Availability Statement The authors confirm that the data supporting the findings of this study are available within this article’s figures and any additional data or raw data can be provided by the corresponding author upon request.