Targeting glutamine metabolism in multiple myeloma enhances BIM binding to BCL-2 eliciting synthetic lethality to venetoclax

Abstract

Multiple myeloma (MM) is a plasma cell malignancy that is largely incurable due to development of resistance to therapy-elicited cell death. Nutrients are intricately connected to maintenance of cellular viability in part by inhibition of apoptosis. We were interested to determine if examination of metabolic regulation of BCL-2 proteins may provide insight on alternative routes to engage apoptosis. MM cells are reliant on glucose and glutamine and withdrawal of either nutrient is associated with varying levels of apoptosis. We and others have demonstrated that glucose maintains levels of key resistance-promoting BCL-2 family member, myeloid cell leukemic factor 1 (MCL-1). Cells continuing to survive in the absence of glucose or glutamine were found to maintain expression of MCL-1 but importantly induce pro-apoptotic BIM expression. One potential mechanism for continued survival despite induction of BIM could be due to binding and sequestration of BIM to alternate pro-survival BCL-2 members. Our investigation revealed that cells surviving glutamine withdrawal in particular, enhance expression and binding of BIM to BCL-2, consequently sensitizing these cells to the BH3 mimetic venetoclax. Glutamine deprivation-driven sensitization to venetoclax can be reversed by metabolic supplementation with TCA cycle intermediate α-ketoglutarate. Inhibition of glucose metabolism with the GLUT4 inhibitor ritonavir elicits variable cytotoxicity in MM that is marginally enhanced with venetoclax treatment, however, targeting glutamine metabolism with 6-diazo-5-oxo-l-norleucine uniformly sensitized MM cell lines and relapse/refractory patient samples to venetoclax. Our studies reveal a potent therapeutic strategy of metabolically driven synthetic lethality involving targeting glutamine metabolism for sensitization to venetoclax in MM.

Figures

Figure 1

MM cell lines and CD138+ MM primary patient cells are variably sensitive to glucose or glutamine deprivation. (a) MM cell lines and (b) patient cells were cultured in the presence of glucose (5 mm) and glutamine (2 mm) (Control) or in the absence of glucose (NG), absence of glutamine (NGlut) or absence of both nutrients (NG+NGlut) for 72 h (cell lines) or 48 h (patient cells). Cell viability was assessed by AnnexinV/DAPI staining and Flow cytometry. Data (a) is mean ± s.e.m. (n = 3). Control vs NG: P-value <0.001 (***) in all cell lines except U266 which is >0.05. Control vs NGlut: P-value <0.001 in all cell lines except U266 <0.1 (*) and L363 <0.01 (**).

Figure 2

Glucose or glutamine deprivation regulates expression of pro and anti-apoptotic BCL-2 proteins. MM cell lines cultured in the absence or presence of 5 mm glucose or 2 mm glutamine or both nutrients for 24 h were evaluated for expression of BCL-2 proteins. Cellular lysates were analyzed for expression of indicated proteins or GAPDH (loading control) by immunoblot analyses. One of two representative experiments is shown. NG, media without glucose; NGlut, media without glutamine.

Figure 3

Glucose or glutamine deprivation increase binding of pro-apoptotic BIM to BCL-2. L363 (a), KMS11 (b) and JJN3 cells (c) were cultured in the absence or presence of 5 mm glucose or 2 mm glutamine with or without (0.5 μm) ABT-199 for 24 h. Immunoprecipitates of MCL-1, BCL-2 and BCL-xL prepared from cellular lysates obtained from treated cells were evaluated for bound BIM, MCL-1, BCL-2 and BCL-xL by immunoblotting. Densitometric analysis of the BIM immunoblot was performed using IMAGE J software (imagej.nih.gov) and percent BIM bound to MCL-1, BCL-2 and BCL-xL presented in the corresponding stacked bar graph. NG, media without glucose; NGlut, media without glutamine. Representative blots from a minimum of three independent experiments are presented.

Figure 4

Glucose or glutamine deprivation sensitizes MM cell lines to the BH3 mimetic ABT-199. MM cell lines were cultured in the absence or presence of 5 mm glucose (a) or 2 mm glutamine (b) with or without the BH3 mimetic ABT-199 (0.5 μM) for 72 h. Cell viability was assessed by AnnexinV/DAPI staining and flow cytometry. NG, media without glucose; NGlut, media without glutamine. Data (a and b) are mean ± s.e.m. (n = 3). P-value of NG vs NG+ABT data is >0.05 (ns) in all cell lines except JJN3 and AMO1 where P-value is <0.001 (***). P-value of NGlut vs NGlut+ABT is <0.001 (***) in KMS18, L363 and U266, <0.01 (**) in MM.1S and JJN3, <0.1 (*) in KMS11 and >0.05 in RPMI-8226 and AMO1.

Figure 5

Metabolic supplementation with DMK reverses glutamine deprivation-elicited sensitivity to ABT-199. (a–d) MM cell lines AMO1 (a), KMS11 (b), MM.1S (c) and L363 (d) grown in absence or presence of glutamine treated with or without 0.5 μm ABT-199 were supplemented with indicated concentrations of dimethyl alpha ketoglutarate (DMK) for 18–24 h. Cells were harvested and evaluated for viability by AnnexinV/DAPI staining and flow cytometry. (e) Cellular lysates prepared from treated cells from d were evaluated for BIM, MCL-1, BCL-2 and BCL-xL protein expression with GAPDH evaluated as a loading control. (f) Cells from d were utilized for co-IP of MCL-1, BCL-xL and BCL-2 to evaluate BIM binding. MCL-1, BCL-xL and BCL-2 protein levels in IPs are also evaluated as loading controls. (g) L363 cells grown in presence or absence of DON treated with or without ABT-199 for 18 h were subjected to co-IP of MCL-1, BCL-xL and BCL-2 to evaluate BIM binding. Distribution of BIM binding is presented in bar graph to the right of f and g. One of two representative co-IP results is presented in f and g. NG, media without glucose; NGlut, media without glutamine. Data (a–d) is mean ± s.e.m. of n ≥ 3; (e–g) co-IPs and immunoblots are representative of two independent experiments. P-value of NGlut+ABT vs NGlut+ABT+DMK in a–d is <0.0001 in KMS11 (b), MM.1S (c) and L363 (d) indicated as ‘****’ and P<0.01 in AMO1 (a) indicated as ‘**’.

Figure 6

Evaluation of Ritonavir (targeting glucose transport) or DON targeting glutamine to ABT-199 sensitivity in MM. MM cell lines were cultured with 40 μm ritonavir (a) or 0.1 mm DON (c) with or without ABT-199 (0.5 μm) or in combination for 72 h. MM patient samples were cultured with ritonavir (20 μm) (b) or DON (1 mm or 0.1 mm), as indicated. (d) for 48 h with viability assessed by AnnexinV/DAPI staining. (e) Model showing the impact of glutamine deprivation on BCL-2 family members and the basis for synthetic lethality to ABT-199. This model is based on the L363 co-IP under glutamine deprivation. Data in a are mean ± s.e.m. (n = 3) and b are mean ± s.e.m. (n = 2). P-values of RIT vs RIT+ABT: >0.05 in RPMI-8226, KMS18, KMS11 and U266, <0.1 in AMO1, <0.01 in JJN3, <0.001 in L363 and MM.1S. P-value of DON vs DON+ABT is <0.001 in all cell lines except RPMI-8226. co-IP, co-immunoprecipitation.