Patient-derived GBM tumor cell lines and culture conditions

Patient-derived GBM cell lines (HROG02, HROG05, HROG52, HROG63, GBM03, GBM06, GBM14, and GBM15) were established in our lab from patients with primary (HROG02, HROG52, GBM03, GBM06, GBM14, and GBM15) or recurrent (HROG05, HROG63) GBM at WHO °4 (Table 1). Patient consent was obtained in all cases. All procedures were approved by the Ethics Committee of the Rostock University Medical Center, University of Rostock (Ethikkommission an der Medizinischen Fakultät der Universität Rostock, St.-Georg-Str. 108, 18055 Rostock, Germany; reference number II HV A 2009/34 and A 2018-0167) following generally accepted guidelines for the use of human material. Detailed information about the cell lines is given in [30]. Cells were cultured in 2D and 3D (HROG05 and HROG63) using Dulbecco’s Modified Eagle medium supplemented with the nutrient mixture F-12 containing 10% FCS, L-glutamine (6 mmol/l), and 1% of the antibiotics penicillin/streptomycin (all from Pan Biotech, Aidenbach, Germany). The incubation took place at 37 °C in a humidified atmosphere of 5% CO2. In the 2D experiments, GBM cells were seeded into 96- (short-term treatment), 24- (invasion assay), or 6- (long-term treatment) well plates (Greiner Bio-One, Kremsmünster, Austria). 3D-spheroids were induced by incubation for 72–96 h with full medium in 96-well ultra-low-attachment (ULA) plates (Greiner Bio-One, Kremsmünster, Austria).

Table 1 clinical data of GBM patients’ from which tumors were obtained for cell line establishmentMithramycin A (MitA) and S. pyogenes arginine deiminase (SpyADI)

The therapeutic effects of the agents MitA (Cayman Chemical, Michigan, USA) and SpyADI were examined. The Arginine Deiminase from S. pyogenes (SpyADI, 35 mU/ml)) was heterologously expressed in E. coli DHα and purified as described in [31]. For functional and combination assays, MitA was used at IC20 (4 nM). The cells were treated simultaneously (SIM) and sequentially (SEQ) for two therapy cycles (144 h) with MitA and SpyADI.

Treatment protocols and viability assays

Cell viability of 2D-cultures was assessed by calcein acetoxymethyl ester (Calcein-AM) (Biomol GmbH, Hamburg, Germany) staining. GBM cells were seeded in plates and incubated overnight. To assess MitA susceptibility, cells were treated for 1 × 72 h or 2 × 72 h with increasing doses (ranging from 2 to 100 nM). Control cells were left untreated. Thereafter, Calcein-AM (4 mM) was added and incubated for 20 min (37 °C, 5% CO2). The fluorescence analysis was performed on a multiwell-plate reader (Tecan Reader Infinite M200, Tecan Group AG, Männedorf, Switzerland) at an excitation/emission of 485/535 nm. Additionally, ten cycles of long-term therapy (72 h each) were performed. The cells were treated in mono- and combination therapy, either SIM or SEQ. For the latter, cells received five cycles SpyADI followed by five cycles MitA. Cell viability was assessed by Calcein-AM assay and crystal violet (0.2%) staining (Sigma-Aldrich, St. Louis, USA). The viability of the 3D-spheroid cultures was analyzed using CellTiter-Glo 3D cell viability assay (Promega, Walldorf, Germany) following the manufacturer’s instructions. The luminescence signal was measured with the GloMax Microplate luminometer (Promega, Walldorf, Germany).

γ-Irradiation

After 24 h of treatment, GBM cells were irradiated with 2 Gy using an IBL 637 (CIS Bio-International, Codolet, France). Following the irradiation, the medium was replaced by fresh medium, followed by 72 h incubation at 37 °C and 5% CO2. This treatment cycle was repeated once. Thereafter, cell viability was measured using the Calcein-AM assay. Double-strand breaks (DSB) were assessed with γ-H2AX staining in 8 Well chamber µ-slides (ibidi, munich, Germany) as described before [32].

2D- and 3D-Invasion-model

A modified Boyden chamber technique (Greiner Bio-One, Kremsmünster, Austria) with Matrigel-coated membranes (Corning, Amsterdam, Netherlands) was used to examine the invasive behavior of GBM cells after treatment. Before seeding, cells were cultured for 24 h in a serum-free medium. After that, cells were seeded with serum-free medium supplemented with the cytostatic agents in the Matrigel-coated inserts (ThinCerts, 8 µm, Greiner Bio-One). To stimulate the cells to migrate through the membrane, the chamber below was filled with full medium. After 72 h, invading cells were quantified by WST-1 staining (1:10 in serum-free medium, Merck KGaA, Darmstadt, Germany). The analysis was done after 3 h incubation using the Tecan Reader at an excitation/emission of 480/650 nm. To document the invasiveness of tumor spheroids after treatment, 96-ULA well plates (Greiner Bio-One, Kremsmünster, Austria) were placed on ice after 4 days of sphere formation; half of the medium was removed, and cytostatic drugs were added including EGF (1%, ImmunoTools, Friesoythe, Germany) to stimulate the invasion into U-bottom wells containing ice-cold matrigel (Corning, Amsterdam, The Netherlands). The spheroids were monitored for 10 days and images were taken on days 0, 5, and 10 using the Leica microscope DMI 4000B (Leica, Heidelberg, Germany).

Immunogenic cell death, senescence and apoptosis/necrosis assay

GBM cells were treated for 72 h and stained for 30 min at 4 °C with an anti-Calreticulin-antibody (1:50, Cell Signaling Technology, Danvers, USA). Then, cells were stained with a secondary FITC-conjugated donkey-anti-rabbit antibody (1:50, BioLegend, San Diego, USA). Calreticulin translocation was quantified using the flow cytometer FACS Calibur (BD Biosciences, New Jersey, USA) at an excitation/emission of 495 nm/525 nm.

To detect senescent cells after treatment, β-Galactosidase staining was done. This staining detects the enzyme β-Galactosidase at pH 6, which is characteristic of senescent cells. A commercially available kit (Cell Signaling Technology, Leiden, The Netherlands) was used following the manufacturer’s instructions. The medium was removed and cells were washed, fixed in fixative solution (15 min, RT), and incubated with β-galactosidase staining solution at 37 °C overnight. The development of blue color as an indicator of senescent cells was analyzed by using a microscope. The number of senescent cells was quantified concerning the total cell number per high power field (HPF). Additional stainings for specific senescence markers were done using CDKN2A/p16INK4a antibody (JC8) (1:50, Santa Cruz), Alexa Fluor® 488 p21 Waf1/Cip1 (1:300, Cell Signaling), and Alexa Flour® 594 anti-p53 antibody (1:50, Biolegend) as described before [33].

For detecting apoptotic/necrotic cells, a flow cytometry-based assay was used as described before [32]. Briefly, early and late apoptotic cells were detected by either Yo-Pro-1 or Yo-Pro-1/propidium iodide (PI) positivity. Necrotic cells were defined as Yo-Pro-1 negative/PI positive.

MMP & autophagy

The assays MitoTracker CMXRos (20 nM), LysoTracker Green DND-26 (50 nM), and ER-Tracker Blue-White DPX (500 nM) were prepared according to the manufacturer’s instructions (Cell Signaling Technology, Thermo Fisher Scientific). After 72 h of treatment, Mitochondria and ER were stained for 30 min at 37 °C, and slides were washed twice. Acidic lysosomes were stained prior to analysis. Images were taken using fluorescence microscopy (Leica DMI 4000B).

ER stress and stemness marker

Cells were fixed with 2% paraformaldehyde (PFA) w/o methanol (15 min, Thermo Fisher Scientific, Darmstadt, Germany), permeabilized, and blocked with 0.5% Triton X-100 (Thermo Fisher Scientific, Waltham, USA) in 2% BSA (PAN-Biotech, Aidenbach, Germany) for 60 min. ER stress markers included: Alexa 647 anti-ATF-4 antibody (B-3, 1:50, Santa Cruz), Alexa 594 anti-calnexin antibody (AF18, 1:50, Santa Cruz), and Alexa 488 anti-cytochrome c (1:50, Biolegend). For stemness, antibody mixtures, either containing anti-GFAP (1:200, Alexa Fluor 594, BioLegend, San Diego, USA) and anti-A2B5 (1:200, Alexa Flour 647, BioLegend, San Diego, USA), or anti-Oct-4 (1:500, Alexa Fluor 647, BioLegend, San Diego, USA) and anti-Nanog (1:500, Alexa Flour 488, BioLegend, San Diego, USA) were added and staining was done at 4 °C overnight. The next day, GFAP/A2B5-antibody mix was stained with Phalloidin green (1:50, BioLegend, San Diego, USA). Nuclei were counterstained with DAPI (1:1.000, Biomol, Hamburg, Germany) and cells were analyzed using a Zeiss microscope Axio Observer 7 (Zeiss, Oberkochen, Germany).

Spectral flow cytometry

Functional analysis was done by spectral flow cytometry using two in-house designed multicolour panels. Panel 1 was used to study apoptosis, necrosis, proliferation, and autophagy. Panel 2 examined viability, methuosis, and immune regulation. For this purpose 0.5 × 106 cells were taken/panel and processed. All procedures were performed using staining buffer (PBS, 2 mM EDTA, 2% BSA).

Panel 1: Membrane permeabilization was done as first step (True Nuclear Transcription Factor Buffer Set, Biolegend, San Diego, California, United States, True-Nuclear™ 1X Fix concentrate, 45 min, RT). Then, the True-Nuclear™ 1X Perm Buffer (Biolegend, San Diego, California, United States) was added, cells were washed (350 × g, 5 min) and stained with antibodies for intracellular staining (in 100 µl True-Nuclear™ 1X Perm Buffer): V450 rat anti-histone H3 (1:40, BD Biosciences, Heidelberg, Germany), and PE/Cyanine7 mouse anti-H2A.X phospho (clone: 2F3, 1:40, Biolegend, San Diego, California, United States). Staining was done for 30 min at RT, reaction was stopped with True-Nuclear™ 1X Perm Buffer, followed by two washing steps (350 × g, 5 min). Cells were finally resuspended in 0.35 ml staining buffer.

Panel 2: Extracellular staining was done for 20 min at RT in staining buffer (in 100 µl): PerCP/Cyanine5.5 anti-human CD274 (1:62.5, Biolegend), PE/Cyanine7 mouse anti-human CD325 (1:25, N-Cadherin, clone: 8C11, Biolegend), and APC-Vio 770 mouse anti-human CD324 (1:15, E-Cadherin, clone 67A4, Biolegend). Afterwards, cells were washed two times followed by membrane permeabilization (BD Transcription Factor Buffer Set, BD, 1 × Fix/Perm Working solution, 45 min, 4 °C). Then, the 1 × Perm/Wash Buffer (BD) was added, cells were washed (350 × g, 5 min) and were stained with antibodies for intracellular staining (in 100 µl 1 × Perm/Wash Buffer): Alexa Fluor 647 mouse anti-human Glut1 (1:500, BD) and Alexa Fluor 700 mouse anti-human Glut4 (1:100, clone: # 925932, R&D, Minneapolis, Minnesota, USA). Staining was done for 30 min at room temperature, reaction was stopped with 1 × Perm/Wash Buffer, followed by two washing steps (350 × g, 5 min). Cells were finally suspended in 0.35 ml staining buffer.

All measurements were done on a spectral flow cytometer (Cytek Aurora, Cytek Biosciences, Fremond, California, United States) in the Core Facility for Cell Sorting and Cell Analysis, University Medical Center Rostock, Rostock, Germany.

Quantitative real-time PCR

RNA was isolated using the RNeasy Mini Kit (Qiagen). 1 µg mRNA and 50 ng random Hexamer Primer (Bioron, Ludwigshafen am Rhein, Germany) were incubated for 10 min at 70 °C. Sample mixes were completed with 5 × RT buffer complete, dNTPs, and 200 units reverase. cDNA was synthesized for 120 min at 45 °C followed by inactivation of the reverase for 10 min at 70 °C. 25 ng cDNA were used for quantitative real-time PCR with the SensiFAST Probe Lo-ROX Kit (Bioline, Memphis, Tennessee, USA). Predesigned or in-house designed Taqman gene expression assays were used: 6-FAM-3'BHQ-1 cMyc (Hs00153408_m1), SP1: 5' HEX-TCGGGGGATCCTGGCAAAAAGAAACA-3’BHQ-1, for 3’-AAGACAGTGAAGGAAGGGGC-5’, rev 3’-GCCATACACTTTCCCACAGC-5’, AHR: 5' HEX-GAGCTTCTTTGATGTTGCATTAAAATCCTCCCCT-3'BHQ-1, for 3’- TAGGCTCAGCGTCAGTTACC-5’, rev 3’- CTGGCCTCCGTTTCTTTCAG-5’. Self-designed 6-FAM-3'BHQ-1 b-Actin was used to detect β-actin as a housekeeping gene. The reaction was performed in the light cycler Viia7 (Applied Biosystems, Foster City, USA) with the following PCR conditions: 95 °C for 10 min, 40 cycles of 15 s at 95 °C, and 1 min at 60 °C. All reactions were run in triplicates. The mRNA levels of target genes were normalized to mRNA levels of b-Actin. The expression level of each sample was considered by calculating 2−ΔCT (ΔCt = Cttarget – CtHousekeeping gene), followed by 2−ΔΔCT quantification, taking values of untreated controls as calibrator.

Statistics

All values are given as mean ± SEM or mean ± SD. Statistical evaluation was performed using GraphPad PRISM 8 software (GraphPad Software, San Diego, CA, USA). Each experiment was done in at least three independent biological replicates. To perform statistical evaluation, one-way or two-way ANOVA (Bonferroni’s or Tukey’s multiple-comparison test) was used. The criterion for significance was taken to be P < 0.05. Significant differences are marked as follows: * vs. control; # vs. monotherapy; $ vs. SEQ-combination. The bliss independence model was used for calculating effects in the combination approach (SIM vs. SEQ).