Fig 2: ROS upregulate AKT phosphorylation. (A) Downregulation of ROS by GSH in CCTL14 hESCs. Cells in CM+ were treated for 1 h with 10 mM GSH in 5% O2 after media change. Normalized data from three different experiments (with at least 10 individual measurements in each) are presented as mean ± SD (Ctrl, 1 ± 0.0490, n = 33; GSH, 0.9187 ± 0.0349, n = 31). Statistical significance was calculated using a two-tailed Mann–Whitney test (***p < 0.0001). (B) Upregulation of ROS by H2O2 in CCTL14 hESCs. Cells in CM+ were treated for 1 h with 0.5 mM H2O2 in 5% O2 after media change. Normalized data from three different experiments (with at least 10 individual measurements in each) are presented as mean ± SD (Ctrl, 1 ± 0.0088, n = 32; H2O2, 1.321 ± 0.0404, n = 34). Statistical significance was calculated using a two-tailed Mann–Whitney test (***p < 0.0001). (C) WB analysis of ROS-induced changes in AKT and ERK1/2 phosphorylation and their quantification in hPSCs. Cells were cultivated in CM+ for 24 h in 21 and 5% O2, respectively, then treated with 10 mM GSH and 0.5 mM H2O2 for 1 h in fresh media. PCNA was used as a loading control. Data in graphs are presented as mean ± SD (pAKT in 21% O2: Ctrl, 1 ± 0; GSH, 0.622 ± 0.155, n = 10; H2O2, 2.105 ± 0.621, n = 6; pAKT in 5% O2: Ctrl, 1 ± 0; GSH, 0.703 ± 0.257, n = 9; H2O2, 2.379 ± 0.703, n = 8; pERK1/2 in 21% O2: Ctrl, 1 ± 0; GSH, 1.069 ± 0.344, n = 8; H2O2, 1.260 ± 0.667, n = 6; pERK1/2 in 5% O2: Ctrl, 1 ± 0; GSH, 1.083 ± 0.160, n = 8; H2O2, 1.793 ± 0.952, n = 6). Statistical analysis was conducted using one sample t-test (theoretical mean = 1) with statistical significance taken where **p < 0.01, ***p < 0.001. (D) WB analysis of PI3K-independent effect of ROS on AKT phosphorylation. PI3K activity was inhibited using wortmannin (WRT, 1 μM/1 h) and ROS were induced using H2O2 (1 mM/1 h). Data in graph are presented as mean ± SEM (–/H2O2, 2.10 ± 0.54, n = 4; WRT/–, 0.107 ± 0.071, n = 4; WRT/H2O2, 0.149 ± 0.299, n = 4; –/–, 1 ± 0). Please refer to Raw Images 1 for uncropped WB and to Supplementary Data 1 for source data used to generate the graphs shown in the figure. WB, western blot; ROS, reactive oxygen species; GSH, glutathione; Ctrl, control; CM+, conditioned media with FGF2 (10 ng/ml); AKT, protein kinase B; pAKT, phosphorylated protein kinase B; ERK1/2, extracellular signal-regulated kinase 1 and 2; pERK1/2, phosphorylated extracellular signal-regulated kinase 1 and 2; PCNA, proliferating cell nuclear antigen; WRT, wortmannin.

Fig 3: PTEN activity is downregulated by ROS. (A) WB analysis of PTEN expression and phosphorylation dependent on oxygen concentration. Cells were cultivated in CM+ and respective oxygen concentrations for 24 h and additional 2 h after fresh CM+ change. Vinculin (Vin) was used as a loading control. Graphs represent mean ± SD (PTEN: 21% O2, 1 ± 0; 5% O2, 0.977 ± 0.134, n = 10; pPTEN: 21% O2, 1 ± 0; 5% O2, 0.988 ± 0.189, n = 10). Statistical significance was calculated using one sample t-test (theoretical mean = 1). (B) WB analysis of the impact of ROS on PTEN expression and phosphorylation. Cells were cultivated in CM+ for 24 h and an additional hour after fresh CM+ change and GSH (10 mM) and H2O2 (0.5 mM) treatment. PCNA was used as a loading control. Graphs represent mean ± SEM (PTEN: Ctrl, 1 ± 0; GSH, 0.958 ± 0.078, n = 4; H2O2, 1.046 ± 0.167, n = 4; pPTEN: Ctrl, 1 ± 0; GSH, 1.042 ± 0.058, n = 4; H2O2, 1.099 ± 0.071, n = 4). Statistical significance was calculated using one sample t-test (theoretical mean = 1). (C) The silencing of PTEN expression upregulates AKT phosphorylation more substantially in 5% compared to the 21% O2 environment. Cells cultivated in 21% O2 in CM+ were transfected with an esiRNA targeting PTEN for 48 h, and respective cells were transferred to 5% O2 24 h before the harvesting. PCNA was used as a loading control. Graph represents mean ± SD (21% O2: Ctrl, 1 ± 0; Scr, 1 ± 0.274, n = 6; esiPTEN, 1.116 ± 0.225, n = 6; 5% O2: Ctrl, 0.851 ± 0.15, n = 6; Scr, 1.129 ± 0.211, n = 6; esiPTEN, 1.445 ± 0.305, n = 6). Statistical significance was calculated using paired two-tailed t-test with significance taken where **p < 0.01. (D) AKT phosphorylation after PTEN silencing is more pronounced in GSH-treated cells compared to the control in CCTL14 hESCs. Cells cultivated in 21% O2 in CM+ were transfected with esiRNA targeting PTEN. Respective cells were treated with GSH (10 mM) 1 h prior to harvesting. PCNA was used as a loading control. Graph represents mean ± SD (21% O2: Ctrl, 1 ± 0; Scr, 1.166 ± 0.268, n = 5; esiPTEN, 1.416 ± 0.495, n = 5; GSH: Ctrl, 0.63 ± 0.172, n = 5; Scr, 0.605 ± 0.131, n = 5; esiPTEN, 1.225 ± 0.229, n = 5). Statistical significance was calculated using paired two-tailed t-test with significance taken where *p < 0.05. (E) WB analysis of PTEN oxidation in hPSCs. Cells were treated with 10 mM GSH and 0.5 mM H2O2 for 1 h and harvested in native lysis buffer. Oxidized PTEN migrated faster on 8% gel due to H2O2-induced disulfides. (F) WB analysis of cells treated with SHIP2 inhibitor. Cells were cultivated in CM+ in 21% O2. Twenty-four hours prior to the treatment, CM– was administered, and the cells were transferred to 5% O2. The next day, fresh CM– was administered, and the cells were treated with FGF2 (10 ng/ml) and AS1938909 (AS19, 10 µM) for 0 or 2 h. PCNA was used as a loading control. Graph represents mean ± SD (21% O2: 0 h, 1 ± 0; 2 h, 2.069 ± 0.86, n = 4; 5% O2: 0 h, 0.979 ± 0.314, n = 6; 2 h, 2.089 ± 0.853, n = 4). Statistical significance was calculated using one sample t-test (theoretical mean = 1). (G) WB analysis of PP2A and PP1 inhibition with okadaic acid in CCTL14 hESCs. Cells were cultivated in CM+ in 21% O2 for 24 h; then, CM– was administered, and the cells were transferred to 5% O2. The following day, fresh CM– was administered once again, and the cells were treated for 2 h with FGF2 (10 ng/ml) and okadaic acid (OKA) in 0.5 and 10 nM concentrations, specific for PP2A and PP1 inhibition, respectively. PCNA was used as a loading control. Graphs represent mean ± SEM (pAKT: 21% O2: FGF2/–, 1 ± 0; –/–, 0.704 ± 0.037, n = 3; FGF2/0.5, 0.862 ± 0.120, n = 3; –/0.5, 0.793 ± 0.064, n = 3; FGF2/10, 0.759 ± 0.083, n = 3; –/10, 0.77 ± 0.064; 5% O2: FGF2/–, 0.781 ± 0.086, n = 3; –/–, 0.61 ± 0.039, n = 3; FGF2/0.5, 0.656 ± 0.12, n = 3; –/0.5, 0.576 ± 0.024, n = 3; FGF2/10, 0.609 ± 0.96, n = 3; –/10, 0.633 ± 0.001; pERK1/2: 21% O2: FGF2/–, 1 ± 0; –/– 0.299 ± 0.028, n = 3; FGF2/0.5, 1.136 ± 0.034, n = 3; –/0.5, 0.315 ± 0.039, n = 3; FGF2/10, 1.109 ± 0.116, n = 3; –/10, 0.328 ± 0.053; 5% O2: FGF2/–, 1.05 ± 0.051, n = 3; –/–, 0.304 ± 0.056, n = 3; FGF2/0.5, 1.146 ± 0.138, n = 3; –/0.5 0.348 ± 0.075, n = 3; FGF2/10, 1.134 ± 0.198, n = 3; –/10, 0.316 ± 0.094). Statistical significance was calculated using one sample t-test (theoretical mean = 1). Please refer to Raw Images 1 for uncropped WB and to Supplementary Data 1 for source data used to generate the graphs shown in the figure. WB, western blot; ROS, reactive oxygen species; GSH, glutathione; Ctrl, control; CM+, conditioned media with FGF2 (10 ng/ml); scr, scrambled short interfering RNA; esiRNA, endoribonuclease-prepared short interfering RNA; AKT, protein kinase B; pAKT, phosphorylated protein kinase B; ERK1/2, extracellular signal-regulated kinase 1 and 2; pERK1/2, phosphorylated extracellular signal-regulated kinase 1 and 2; PCNA, proliferating cell nuclear antigen; PTEN, phosphatase and tensin homolog deleted on chromosome 10; pPTEN, phosphorylated phosphatase and tensin homolog deleted on chromosome 10; Vin, Vinculin; PP1, protein phosphatase 1; PP2A, protein phosphatase 2A; AS19, AS1938909; OKA, okadaic acid.

Fig 4: Oxygen-induced ROS upregulate phosphorylation of FGFR1 and PI3K p85. (A) Representative WB analysis and quantification of ROS-induced changes in FGFR1 phosphorylation. Cells were cultivated in CM+ for 24 h in 21 and 5% O2, respectively, then treated with 10 mM GSH and 0.5 mM H2O2 for 1 h in fresh media. ß-Actin was used as a loading control for pFGFR1 and Lamin B1 for FGFR1. Data obtained on CCTL12, CCTL14, and AM13 hPSCs were used in the densitometry analysis. Data are presented as mean ± SEM (21% O2: Ctrl, 1 ± 0, n = 7; GSH, 0.7211 ± 0.0690, n = 6; H2O2, 1.530 ± 0.1669, n = 5; 5% O2: Ctrl, 0.6562 ± 0.0977, n = 7; GSH, 0.5748 ± 0.0801, n = 5; H2O2, 2.2038 ± 0.3416, n = 5). Statistical analysis was conducted using one sample t-test (theoretical mean = 1) with statistical significance taken where *p < 0.05. (B) WB analysis of exogenous FGF2-independent ROS-induced FGFR1 phosphorylation. Cells kept in 21 or 5% O2 were starved for 24 h and subsequently treated with FGF2 (10 ng/ml) and H2O2 (1 mM) for 1 h. PCNA was used as a loading control. Data in graph are presented as mean ± SEM (21% O2: FGF2/–, 1 ± 0; –/–, 0.803 ± 0.199, n = 3; FGF2/H2O2, 1.845 ± 0.379, n = 3; –/H2O2, 1.511 ± 0.365, n = 3; 5% O2: FGF2/–, 0.797 ± 0.033, n = 3; –/–, 0.617 ± 0.075, n = 3; FGF2/H2O2, 1.608 ± 0.243, n = 3; –/H2O2, 1.621 ± 0.524, n = 3). Statistical analysis was conducted using paired one-tailed t-test with significance taken where *p < 0.05. (C) Representative WB analysis of ROS-induced changes in PI3K p85 phosphorylation in CCTL14 hESCs. Cells were cultivated in CM+ for 24 h in 21 and 5% O2, respectively, then treated with 10 mM GSH and 0.5 mM H2O2 for 1 h in fresh media. PCNA was used as a loading control. Data obtained on CCTL12, CCTL14, and AM13 hPSCs were used in the densitometry analysis. Data are presented as mean ± SEM (21% O2: Ctrl, 1 ± 0, n = 7; GSH, 0.7335 ± 0.0323, n = 6; H2O2, 4.907 ± 1.657, n = 7; 5% O2: Ctrl, 0.6401 ± 0.0787, n = 6, GSH, 0.5416 ± 0.0731, n = 6; H2O2, 2.335 ± 0.9237, n = 5). Statistical analysis was conducted using one sample t-test (theoretical mean = 1) with statistical significance taken where **p < 0.01 and ***p < 0.001. Please refer to Raw Images 1 for uncropped WB and to Supplementary Data 1 for source data used to generate the graphs shown in the figure. WB, western blot; hESCs, human embryonic stem cells; ROS, reactive oxygen species; GSH, glutathione; Ctrl, control; hPSCs, human pluripotent stem cells; CM+, conditioned media with FGF2 (10 ng/ml); PCNA, proliferating cell nuclear antigen; FGFR1, fibroblast growth factor receptor 1; pFGFR1, phosphorylated fibroblast growth factor receptor 1; PI3K p85, p85 subunit of phosphatydylinositol-4,5-bisphosphate 3-kinase; pPI3K p85, phosphorylated p85 subunit of phosphatydylinositol-4,5-bisphosphate 3-kinase.

Fig 5: ISA27 inhibits cell proliferation and induces apoptosis in tumor tissues.14 days from starting treatment, tumors were harvested from mice of different ISA27 treatment groups (IP = intraperitoneal treatment; IT = intratumor treatment) and processed for histological and Western blotting analysis. A) Histological analysis: apoptosis and cell proliferation were evaluated by analysis of cleaved caspase 3 and PCNA levels by immunohistochemistry in paraffin embedded sections of tumors. Upper panel shows representative images of immunohistochemistry analysis. ISA27-treated tumors show increased cleaved caspase 3 levels and reduced cell proliferation. Cleaved caspase 3 and PCNA levels in tumors were quantified from digital images. Results are shown in graph as percent of cleaved caspase 3 and PCNA expression respect to controls (*P<0.05 vs control). B) Western blotting analysis: apoptosis and cell proliferation were evaluated by analysis of cleaved caspase 3 and phosphorylated histone H3 (pH3) levels by Western blotting in whole lysates of tumors using specific antibodies. Densitometric analysis shows that ISA27 induced a significant increase in cleaved-caspase-3 levels and a significant reduction in pH3 levels (*P<0.05 vs control).