GPX1 confers resistance to metabolic stress in BCR/ABL-T315I mutant chronic myeloid leukemia cells

Abstract

Chronic myeloid leukemia (CML) harboring BCR/ABL-T315I mutation has been a challenging obstacle for targeted therapy due to the acquired resistance to tyrosine kinase inhibitor (TKI)-based therapy. Thus, it is especially urgent to investigate more effective therapeutic targets to overcome T315I-induced resistance. Here, we reported that BCR/ABL-T315I mutant CML cells possessed a long-term proliferative capacity and tolerance to metabolic stress. Importantly, we also found that selenoamino acid metabolism was increased in the bone marrows of BCR/ABL-T315I patients compared with non-T315I patients by GSEA from RNA-Seq data. Indeed, GPX1 was highly expressed in T315I mutant cells, while knockout of GPX1 significantly suppressed cell proliferation and triggered apoptosis under glucose-deprived condition. GPX1 knockout showed decreased cell metabolism signaling as well as mitochondrial gene expression by RNA-Seq. Mechanistically, GPX1 maintained mitochondrial activity and oxygen consumption rate (OCR), retaining mitochondrial redox homeostasis and oxidative phosphorylation (OXPHOS). Additionally, mercaptosuccinic acid (MSA), a GPX inhibitor, inhibited CML colony formation and induced cell apoptosis under glucose-free condition. Therefore, GPX1 is a promising therapeutic target to overcome drug resistance induced by the T315I mutation, which provides a novel approach for BCR/ABL-T315I CML treatment by disturbing mitochondrial OXPHOS.

Fig. 1. BCR/ABL-T315I CML cells are tolerant to metabolic stress.

The bone marrows of 3 BCR/ABL-T315I CML patients and 3 non-T315I patients were collected and subjected to RNA-Seq. GSEA-Hallmark (a) and GSEA-KEGG (b) analyses were performed. The number of BCR/ABL wildtype and BCR/ABL-T315I cells were counted for three consecutive days under the condition of normal culture system (c), glucose deprivation (d) or glutamine deprivation (e). f KBM5 cells and KBM5-T315I cells were stained with CFSE and subjected to flow cytometry. The CFUs of KBM5 and KBM5-T315I (g) as well as 32D-p210 and 32D-T315I (h) cells under the condition of nutritional deficiency were shown (Con: Control; Low: 1/3 concentration of glucose and glutamine). Colony numbers and diameters between KBM5 and KBM5-T315I cells (i), as well as 32D-p210 and 32D-T315I cells (j), under the condition of nutritional deficiency, were compared. Data are expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2. BCR/ABL-T315I CML cells display high OXPHOS.

a BCR/ABL wildtype and BCR/ABL-T315I cells were treated with 2-DG for 48 h, and CCK-8 assay was performed. b RNA-Seq data from BCR/ABL-T315I and non-T315I samples were analyzed by GSEA. ATP level (c) and mitochondrial membrane potential (d) of BCR/ABL and BCR/ABL-T315I cells cultured under the condition of galactose replenishment were determined. OCR of BCR/ABL and BCR/ABL-T315I cells were measured (e) and basal respiration, maximal respiration, as well as ATP production, was shown (f–h). i Heatmap of mitochondrial genes in CML patients from RNA-Seq was displayed. j Western blot was used to confirm the expression of mitochondrial proteins. Data are expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 3. GPX1 is highly expressed in BCR/ABL-T315I cells.

GSEA (a) and heatmap of GPXs (b) from RNA-Seq data of BCR/ABL-T315I CML and non-T315I patients were shown. c The expression level of GPX1 was detected by western blot. d GPX1 distribution in cytoplasm and mitochondria of BCR/ABL and BCR/ABL1-T315I cells was measured. e GPXs enzyme activity of BCR/ABL and BCR/ABL1-T315I cells was determined. f KBM5-T315I and 32D-T315I cells were cultured under glucose or glutamine deprivation for 48 h, and the GPX1 protein level was determined by western blot. Cells were treated with different concentrations of hydrogen peroxide for 48 h, and GPX1 expression (g) as well as cell viability (h) was measured. Data are expressed as mean ± SD. *P < 0.05, **P < 0.01.

Fig. 4. Knockout of GPX1 disturbs mitochondrial metabolism.

a GPX1-knockout efficiency in KBM5-T315I and 32D-T315I cells was determined by western blot. b Control and GPX1-knockout cells were subjected to cell counting under the condition of glucose deprivation. GSVA (c) and GSEA (d) were performed by RNA-Seq data from control and GPX1-knockout of KBM5-T315I cells. e Mitochondrial activity was detected by flow cytometry by MitoTracker Deep Red staining. f Apoptosis of control and GPX1-knockout cells under the condition of glucose deprivation was shown. Data are expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 5. MSA overcomes BCR/ABL-T315I-induced drug resistance.

a KBM5-T315I and 32D-T315I cells were treated with MSA for 48 h, and the cell viability was determined. b T315I cells were treated with MSA for 48 h under glucose deprivation and cell counting was conducted. Colony formation of KBM5-T315I and 32D-T315I cells with 0.075 mM MSA treatment was measured (c), and the number and diameter of CFUs were shown (d). e, f KBM5 and KBM5-T315I cells were treated with MSA for 48 h under glucose deprivation, and apoptosis was conducted by flow cytometry. T315I cells were treated with MSA, and GPXs enzyme activity (g), OCR (h) and ATP level (i) were determined. j T315I cells were treated with MSA for 48 h and subjected to Western blot. Data are expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.