Fig 2: BCAAs sustain the malignant proliferation of GBM tumors. (A) The survival percentage of nude mice orthotopically transplanted with BCAT1 wild-type and BACT1-deficient LN229 cells, respectively. There was a significant difference in the OS time between these two groups (p = 0.02). There were 16 biologically independent mice in each group. (B) Representative images of H&E staining of orthotopically transplanted tumors. (C) Quantification of tumor size of LN229 cells transplanted orthotopically in a nude mouse model. All data are shown as the mean ± SEM. Mean ± SEM, two-tailed T test, p =0.0003. (D) Representative images of IHC staining of BCAT1, GCLM, and TXNRD1 in orthotopically transplanted tumors. (E) Positive correlation between the staining density of BCAT1 and TXNRD1. (Pearson correlation: r = 0.7088, p =0.0099). (F) Representative images of BCAT1 staining in GBM samples. Negative (-), weakly positive (+), moderately positive (++), and strongly positive (+++). (G) Statistical analysis of BCAT1 IHC integral optical density in 90 GBM samples (IDH wild-type) and 14 diffuse astrocytoma samples (grade 4, IDH mutant). (H) Kaplan–Meier survival analysis showed that GBM patients with higher BCAT1 expression had shorter OS than those with lower BCAT1 expression (p = 0.045, log-rank test). (I) Representative images of GCLM staining in GBM samples. Negative (-), weakly positive (+), medium positive (++), and strong positive (+++). (J) Statistical analysis of GCLM IHC integral optical density in 90 GBM samples (IDH wild-type) and 14 diffuse astrocytoma samples (grade 4, IDH mutant). (K) Kaplan–Meier survival analysis showed that GCLM expression showed no correlation with OS in GBM samples (p = 0.3052, log-rank test). (L) Representative images of TXNRD1 staining in GBM samples. Negative (-), weakly positive (+), moderately positive (++), and strongly positive (+++). (M) Statistical analysis of TXNRD1 IHC integral optical density in 90 GBM samples (IDH wild-type) and 14 diffuse astrocytoma samples (grade 4, IDH mutant). In GBM samples, TXNRD1 was significantly higher in the high BCAT1 group than in the low BCAT1 group (p = 0.001). (N) Kaplan–Meier survival analysis showed that GBM patients with higher TXNRD1 expression had shorter OS than those with lower BCAT1 expression (p = 0.0103, log-rank test). All data are shown as the mean ± SEM. The p values were determined by a two-tailed t-test. Scale bar, 100 µm. *** p < 0.001.

Fig 3: DOT1L mediates LDHA-induced BCAT1 expression in PGBM cells under ROS accumulation. (A) GBM cells were treated with H2O2 and supplemented with NAC for blotting DOT1L, H3K79me2, and BCAT1, as indicated. (B) The a-HB from GBM cells was analyzed with LC-MS/MS, and the relative abundance (by metabolite peak area) was normalized to cell numbers. (C) H3K79 methylation levels and the expression of DOT1L and BCAT1 were analyzed in GBM cells upon a-HB treatment for 24 h. Ctrl, control. (D) H2O2 promotes BCAT1 expression and H3K79 methylation level in GBM cells, which are diminished by NAC. GAPDH was used as a loading control. (E) DOT1L mediates BCAT1 expression. GAPDH was used as a loading control. (F) LDHA knockdown by siRNA significantly inhibited the expression of BCAT1 after H2O2 treatment in GBM cells. (G,H) Positive correlations were found between the mRNA levels of LDHA and BCAT1 (Pearson correlation ? was 0.5202; p < 0.01) and DOT1L and BCAT1 (Pearson correlation ? was 0.5202; p < 0.01) in 549 TCGA GBM samples. (I) Representative IHC images for BCAT1 and LDHA in GBM samples. In case 1, both BCAT1 and LDHA were positive. In case 2, the expression of LDHA was negative, and BCAT1 was weakly positive. Scale bar, 100 µm. (J) A positive correlation was found between the IRS of LDHA and BCAT1 (Pearson correlation ? is 0.5202; p < 0.01). The results are from three independent experiments. IRS: immunostaining scores. All data are shown as the mean ± SEM. The p values were determined by a two-tailed t-test. Values of p < 0.05 were considered statistically significant. * denotes p < 0.05. Scale bar, 100 µm.