Glioblastoma (GBM), the most lethal primary brain tumor, harbors glioma stem cells (GSCs) that not only initiate and maintain malignant phenotypes but also enhance therapeutic resistance. Although frequently mutated in GBMs, the function and regulation of PTEN in PTEN-intact GSCs are unknown. Here we found that PTEN directly interacts with MMS19 and competitively disrupts MMS19-based cytosolic iron-sulfur (Fe-S) cluster assembly (CIA) machinery in the differentiated glioma cells (DGCs). Interrogation of GSCs, when compared with their matched DGCs, revealed that PTEN is specifically succinated at cysteine (C) 211 in GSCs. Isotope tracing coupled with mass spectrometry analysis confirmed that fumarate, generated by adenylosuccinate lyase (ADSL) in de novo purine synthesis pathway which is highly activated in GSCs, promotes PTEN C211 succination. This modification abrogates the interaction between PTEN and MMS19, thereby reactivating CIA machinery pathway in GSCs. Functionally, inhibiting PTEN C211 succination through re-expressing PTEN C211S mutant, depleting ADSL, or consuming fumarate by N-acetylcysteine (NAC), an FDA-approved prescription drug, impairs GSC maintenance. Importantly, re-expressing PTEN C211S or treating with NAC sensitizes GSC-derived brain tumors to temozolomide and irradiation, the standard-of-care treatments for GBM patients, by retarding CIA machinery-mediated DNA damage repair. These findings reveal an immediately practicable strategy to target GSCs for treating GBMs by combined therapy with repurposing NAC.

Mice

To identify the potential tumor formation ability of GSCs and DGCs, luciferase-expressing cells (1 × 10⁴) were injected intracranially into 4-week-old female athymic old nude mice (BALB/cNj-Foxn1nu/Gpt, Strain NO. D000521, GemPharmatech) as previously described (25). Each group contained five mice. Mice were fed autoclaved food and water and maintained in a specific pathogen-free facility. Tumor volumes were monitored by detecting the flux activity using a bioluminescence imaging system at different time points.

To teste whether inhibiting PTEN C211sc sensitizes GSCs to TMZ treatment, PTEN-depleted-MGG8 or MES28 GSCs expressing with or without WT Flag-PTEN or Flag-PTEN C211S were transfected with firefly luciferase. 1 × 10⁴ cells were injected intracranially into 4-week-old female athymic old nude mice. Each group contained five mice. 7 days after the injection, mice were treated with PBS or TMZ (20 mg/kg) by intraperitoneal injection. To teste whether inhibiting PTEN C211sc sensitizes GSCs to radiation treatment, PTEN-depleted-MGG8 or MES28 GSCs expressing with or without WT Flag-PTEN or Flag-PTEN C211S were transfected with firefly luciferase. 1 × 10⁴ cells were injected intracranially into 4-week-old female athymic old nude mice. Each group contained five mice. On day 9 and 11, mice receive 5 Gy radiation treatment. Tumor volumes were monitored by detecting the flux activity using a bioluminescence imaging system at different time points.

To explore the function of NAC sensitizing chemotherapy of brain tumors, 1 × 10⁴ luciferase-expressing MGG8 and T3264 GSCs were injected intracranially into 4-week-old female athymic old nude mice. Each group contained five mice. 7 days after the injection, mice were treated with PBS, TMZ (20 mg/kg), NAC (200 mg/kg), or TMZ plus NAC by intraperitoneal injection. To explore the function of NAC sensitizing radiation therapy of brain tumors, 1 × 10⁴ luciferase-expressing MGG8 and T3264 GSCs were injected intracranially into 4-week-old female athymic old nude mice. Each group contained five mice. 7 days after the injection, mice were treated with NAC (200 mg/kg) by intraperitoneal injection. On day 9 and 11, mice receive 5 Gy radiation treatment. Tumor volumes were monitored by detecting the flux activity using a bioluminescence imaging system at different time points.

Cell Culture

GSCs (MGG8, T3264, CW839, T2907, GSC23, MES28, MES20, T3028 and CW738) were derived from human specimens as previously described (PMID: 30948495). Details of these patients are restricted by the institutional requirements. GSCs were maintained in Neurobasal medium (Life Technologies) supplemented with B27, L-glutamine, sodium pyruvate, 10 ng/ml basic fibroblast growth factor and 10 ng/ml epidermal growth factor (R&D Systems). For inducing DGCs, Fetal Bovine Serum (FBS) was added into culture medium for 7 days, and culture medium was changed every other day. The GSC phenotype was validated by stem cell marker expression (SOX2, Olig2 and GFAP) and tumor propagation in vivo.

Plasmids and lentiviral transduction

Polymerase chain reaction (PCR)-amplified full-length or truncated human FH, Flag-PTEN, and HA-MMS19 were cloned into the pLVX vector. Human CIAO1 and CIAO2B were cloned into pColdI (His) vector. PTEN C211S was generated using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). shRNA-resistant (r) PTEN was constructed by introducing nonsense mutations in shRNA-targeting sites as previously described (10). shRNA targeting PTEN (5’-GCCAGCTAAAGGTGAAGATATAT-3’) was inserted into the pGIPZ vector as previously described (10). Lentiviral clones expressing shRNAs against human MMS19 (TRCN0000096384), human ADSL (TRCN0000078271), human PRPS1 (TRCN0000010123), human PPAT (TRCN0000031614) or a control shRNA were purchased from Sigma-Aldrich (St. Louis, MO, USA). 293FT cells were used to generate lentiviral particles through co-transfection of the packaging vectors pCMV-dR8.2 and pCI-VSVG using a standard calcium phosphate transfection method in Neurobasal complete medium as previously described (25).

Immunoprecipitation and Immunoblotting Analysis

Protein samples were extracted from transfected GSCs using IP cell lysis buffer (50 mM Tris-HCl, pH 7.5, 0.01% SDS, 1% Triton X-100, 150 mM NaCl, 1 mM dithiothreitol, 0.5 mM EDTA, 100 μM PMSF, 100 μM leupeptin, 1 μM aprotinin, 100 μM sodium orthovanadate, 100 μM sodium pyrophosphate, and 1 mM sodium fluoride). Immunoblot and immunoprecipitation analyses were performed using indicated antibodies as previously described (58).

Purification of Recombinant Proteins

Expression of His-CIAO1 and CIAO2B were induced in bacteria, and protein purification was performed as previously described (58). Briefly, 6xHis-tagged recombinant proteins were cultured in 250 ml of lysogeny broth (LB) medium until the OD reached 0.6. 0.5 mM. Isopropyl β-D-1-thiogalactopyranoside (IPTG) was used to induce protein expression overnight at 16 ℃. Then, bacteria were collected and lysed. The cell lysates were loaded onto a Ni-NTA column, washed with five column volumes of 20 mM imidazole, and eluted with 250 mM imidazole. Purified proteins were desalted using 10-kDa cut-through spin columns by washing with PBS.

Neurosphere Formation Assay

In vitro limiting dilution was performed to measure neurosphere formation as previously described. Briefly, decreasing numbers of GSCs (100, 50, 25, 10, 2) per well were plated into 96-well plates. The presentation and numbers of neurospheres in each well were recorded. Extreme limiting dilution analysis was performed using software available at http://bioinf.wehi.edu.au/software/elda as previously described (5).

Quantitative RT-PCR

Total cellular RNA was isolated using Trizol reagent (Sigma Aldrich). The qScript cDNA Synthesis Kit (Quanta BioSciences) was used for reverse transcription into cDNA. Applied Biosystems 7900HT cycler using SYBR-Green PCR Master Mix (Thermo Fisher Scientific) was employed to perform quantitative real-time PCR. The primers used in this study were described in Table S4.

Cell Proliferation Analysis

A total of 1000 GSCs suspended in 200 μL Neurobasal medium were plated in a 96-well plate. CellTiter-Glo (Promega, Madison, WI, USA) was used to measure cell proliferation according to the manufacturer’s instructions. All data were normalized to those of day 1 and presented as mean ± SD from three independent experiments.

⁵⁵Fe Incorporation Assay

⁵⁵Fe incorporation assays were performed as previously described (21). Briefly, 16 mM ⁵⁵FeCl3 (Perkin Elmer) was incubated in 100mM HCl, 63 μM nitrilotriacetic acid (NTA), and 20 mM HEPES pH to 6.0 with Tris followed by titration to pH 7.0 with 100 mM NaOH to obtain ⁵⁵Fe-NTA. 1 × 10⁵ cells were seeded into 6-well plates followed by 2 mCi/mL ⁵⁵Fe-NTA treatment for 18 hours. Then, cells were washed, collected, and lysed using RIPA buffer (150 mM NaCl, 5 mM EDTA pH 8.0, 50 mM Tris-HCl pH 8.0, 1% NP-40 (v/v), 0.5% sodium deoxycholate (w/v), 0.1% SDS (w/v), 1 mM DTT, and 1X protease inhibitor cocktail (Roche)). MMS19 targeting proteins were Immunoprecipitated and ⁵⁵Fe incorporation was measured by scintillation counting. Data were normalized to cellular protein concentrations.

Fumarate Quantification

Intracellular fumarate levels were quantified using Fumarate Assay Kit (MAK060, Sigma-Aldrich) according to manufacturer’s instructions. For intracellular fumarate measurement, 1 × 10⁶ cells were lysed using fumarate assay buffer from the kit, followed by centrifuging at 13,000 g for 10 min to remove insoluble material. For intracranial fumarate measurement, mice were perfused with PBS. The tumor samples in mouse brains were collected and chopped into pieces. 4 mg tissues were rapidly homogenized in 10 mL of Fumarate Assay Buffer, followed by centrifuging at 13,000 g for 10 min to remove insoluble material. Set up the Master Reaction Mix using fumarate assay buffer, fumarate developer, and fumarate enzyme mix from the kit. Add 100 μL Master Reaction Mix to each well of plates from the kit, followed by incubation for 30 min at room temperature. Measure the absorbance at 450 nm and calculate fumarate levels. Data were normalized to cellular protein concentrations.

Mass Spectrometry Analysis

PTEN succination site was determined as previously described (10). Briefly, Flag-PTEN protein was immunoprecipitated from two GSC cultures (MGG8 and T3264) and digested by dithiothreitol (5 mM) for 30 min at 56 ℃ followed by alkylation with iodoacetamide (11 mM) for 15 min at room temperature in a dark environment. Then, protein was diluted to concentration less than 2 M using TEAB (100 mM). Finally, chymotrypsin (chymotrypsin : protein = 1 : 50) was used to digest protein twice. After digestion, the protein peptides were analyzed by MALDI-TOF/TOF MS (MALDI-7090, Shimadzu Kratos). The total MS/MS data was compared against SwissProt Database by the following parameters: chymotrypsin digestion allowing up to 1 missed cleavage, fixed modifications of cysteine (carbamidomethylation), variable modifications of methionine (oxidation) and cysteine (succination), precursor peptide tolerance of 0.05 Da, and MS/MS tolerance of 0.2 Da. Analysis results with e values less than 0.01 was considered as positive identifications.

To confirm the results that purine synthesis fuels PTEN C211sc through ADSL, MGG8 GSCs expressing with control shRNA or ADSL shRNA were washed with aspartate-free medium and incubated in fresh medium containing ¹³C-aspartate (1 mM) for 24 h. ¹³C-labelled C211sc of PTEN was analyzed by MS as described above.

To determine the abundance of intracellular IMP, AMP, GMP, and GSH, approximately 1 × 10⁵ GSCs were seeded in 10 cm dishes in triplicate. Cells were lysed, extracted in 90/9/1 (v/v/v) acetonitrile/water/formic acid and subjected to high-resolution mass spectrometry. Pure samples of IMP, GMP, AMP and GSH were purchased from Sigma-Aldrich. Samples were centrifuged and supernatants were dried using Termovap Sample Concentrator. Samples were then resolved in ammonium acetate (10 mM) containing 0.2% ammonium hydroxide. Samples were injected into a Luna NH2 column (P/N 00B-4378-B0; 5 μM, 50 × 2.0 mm; Phenomenex, Torrance, CA) heated to 35°C with mobile phase A (0.77 g NH4OAc, 1.25 mL NH4OH, 25 mL ACN, and 300 µL acetic acid [HAc] dissolved in 500 mL water) and mobile phase B (acetonitrile). Using a flowrate of 0.3ml/min, the elution program was: 0.1 min, 85% B; 3 min, 30% B; 12 min, 2% B; 15 min, 2% B; and 16–28 min, 85% B. Data were acquired using a Thermo Orbitrap Fusion Tribrid Mass Spectrometer via Selected Ion Mode (SIM) electrospray positive mode.

EC₅₀ Measurement

GSCs were exposed to TMZ with increasing concentrations from 2.9 to 1500 μM. Cell viability was measured at 24 h after treatment. The EC₅₀ of TMZ for GSCs was calculated using GraphPad software.

Combination Effect Analysis

The synergistic effect of TMZ and NAC combination was evaluated by a calculation of CI according to the Chou-Talalay method. Data were analyzed using CompuSyn software (CompuSyn Inc.): CI = 0.85 to 0.9, slight synergism; CI = 0.7 to 0.85, moderate synergism; CI = 0.3 to 0.7, synergism; CI = 0.1 to 0.3, strong synergism; CI < 0.1, very strong synergism.

Immunohistochemical (IHC) Analysis

Sections of paraffin-embedded xenografted tumors were stained with Ki-67, γ-H2AX, PTEN, and PTEN C211sc, respectively. The percentage of Ki-67 and γ-H2AX positive cells was quantified in five randomly selected fields using Image Pro Plus software (Media Cybernetics). The staining of PTEN and PTEN C211sc were quantified according to the percentage of positive cells and staining intensity as described previously. The staining intensity was scored on a scale of 0-3: 0, negative; 1, weak; 2, moderate; and 3, strong. The proportion scores were assigned to the sections: 0 if 0% of tumor cells exhibited positive staining, 1 for 0 to 1% positive cells, 2 for 2% to 10% positive cells, 3 for 11% to 30% positive cells, 4 for 31% to 70% positive cells, and 5 for 71% to 100% positive cells. The intensity and proportion scores were then added to obtain a total score ranging from 0 to 8 as described before.

TUNEL Analysis

GSCs-derived tumors were cut into 4 mm slices. The rate of apoptotic cells in tumors was analyzed using the TUNEL BrightGreen Apoptosis Detection Kit (Vazyme) according to the manufacturer’s instructions.