Aldehyde dehydrogenase inhibition combined with phenformin treatment reversed NSCLC through ATP depletion

Abstract

Among ALDH isoforms, ALDH1L1 in the folate pathway showed highly increased expression in non-small-cell lung cancer cells (NSCLC). Based on the basic mechanism of ALDH converting aldehyde to carboxylic acid with by-product NADH, we suggested that ALDH1L1 may contribute to ATP production using NADH through oxidative phosphorylation. ALDH1L1 knockdown reduced ATP production by up to 60% concomitantly with decrease of NADH in NSCLC. ALDH inhibitor, gossypol, also reduced ATP production in a dose dependent manner together with decrease of NADH level in NSCLC. A combination treatment of gossypol with phenformin, mitochondrial complex I inhibitor, synergized ATP depletion, which efficiently induced cell death. Pre-clinical xenograft model using human NSCLC demonstrated a remarkable therapeutic response to the combined treatment of gossypol and phenformin.

Keywords: NSCLC; aldehyde dehydrogenase; cancer metabolism; gossypol; phenformin.

Figure 1. ALDH1L1 is highly increased in NSCLC

(A) Expression of ALDH1L1 in lung cancer cell lines was measured by multiple reaction monitoring mass spectrometry (MRM-MS). (B) Assessment of ALDH1L1 in IMR90 and H226 by immunofluorescence staining. Scale bar = 50 μm. (C) Representative immunohistochemical staining of ALDH1L1 in normal and cancerous lung tissue. Scale bar = 100 μm. Expression of ALDH1L1in cancerous (Cancer) and normal lung type I and II pneumocytes (Control). *p < 0.001, n = 57 for each case. (D) ALDH1L1 catalyses 10-formyltetrahydrofolate to THF with by-product of NADH in the serine-folate pathway. DHF, dihydrofolate; THF, tetrahydrofolate; SHMT1, serine hydroxymethyltransferase 1; DHFR, dihydrofolate reductase; CH₂-THF, 5,10-Methylenetetrahydrofolate; MTHFD1, methylenetetrahydrofolate dehydrogenase1.

Figure 2. Effect of ALDH1L1 over expression on induction of NADH and ATP production

(A–C) EKVX and H23 cells were transfected with plasmid expressing ALDH1L1 for 24 h and incubated with siRNA of DHFR for 24 h. (D–F) EKVX and H23 cells were transfected with plasmid expressing DHFR for 24 h and incubated with siRNA of ALDH1L1 for 24 h. Effect of DHFR siRNA or ALDH1L1 siRNA on NADH/NAD+ (A, D) and ATP (B, E) was analysed. Immunoblot analysis was performed to confirm DHFR expression, ALDH1L1 knockdown (C, F). Data are representative of the mean and standard deviation three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared to vehicle control.

Figure 3. ALDH participates significantly in ATP synthesis using NADH through oxidative phosphorylation that requires the malate-aspartate shuttle in NSCLC

(A–C) Effect of ALDH1L1 knockdown in the presence and absence of 10 mM malate on NADH/NAD+ (A) and ATP (B) was analysed. (D) Malate-aspartate shuttle for NADH transportation into the mitochondrial matrix. MAT, malate-α-ketoglutarate transporter; GAT, glutamate-aspartate transporter; OAA, oxaloacetate; α-KG, α-ketoglutarate. (E–G) Effect of GOT2 knockdown in the presence and absence of 10 mM malate on NADH/NAD+ (E) and ATP (F) was analysed. (H–J) Effect of MDH2 knockdown in the presence and absence of 10 mM malate on NADH/NAD+ (H) and ATP (I) was analysed. Immunoblot analysis was performed to confirm knockdowns of ALDH1L1, MDH2, GOT2 (C, G, J). Data are representative of the mean and standard deviation three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared to vehicle control.

Figure 4. Phenformin treatment combined with siRNA of ALDH1L1 synergistically reduced ATP production

(A, B) After incubation with ALDH1L1 siRNA, cells were treated with 100 μM phenformin for 24 h before assessment of NADH/NAD+ (A), ATP (B). (C) Immunoblot analysis was performed to confirm ALDH1L1 knockdown. (D) Tetramethylrhodamine ethyl ester (TMRE) staining was performed to examine changes in mitochondrial membrane potential. Data are representative of the mean and standard deviation of three independent experiments. **p < 0.01, ***p < 0.001 compared to vehicle control.

Figure 5. Gossypol inhibits NSCLC cell growth by decreasing ATP production

(A) Structure of gossypol. MW = 518.563. (B) Effect of gossypol on NSCLC cell proliferation as determined by the SRB assay. (C) Cells were treated with the indicated inhibitors and then the level of ATP was determined. (D, E) The levels of NADH/NAD+ (D) and ATP (E) were measured after EKVX and A549 cells were treated with 1 and 10 μM gossypol for 24 h. (F) Combined treatment of 10 μM of gossypol with 100 μM phenformin showed synergistic inhibition of cell growth in NSCLC, as determined by the SRB assay. Data are representative of the mean and standard deviation of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared to vehicle control.

Figure 6. Combination of gossypol and phenformin remarkably reduced NADH and ATP production with down regulation of mitochondrial action potential

(A) Cells were treated as indicated for 24 h, stained for TMRE, and analysed by flow cytometry and live cell imaging. Data are representative of the mean and standard deviation three independent experiments. (B) H23 was stained for TMRE after 24 h treatment of gossypol, phenformin, and combination. Scale bar = 20 μm. (C) Effect of gossypol treatment on metabolites from various metabolic pathways in A549 cells. Relative pool sizes of metabolites by targeted LC-MS/MS upon gossypol treatment for 24 h. Data are representative of the mean and standard deviation three independent experiments. *p < 0.05 compared to vehicle control.

Figure 7. Gossypol combined with phenformin reversed NSCLC through induction of cell death via ATP depletion

(A) Effect of gossypol (10 μM), phenformin (100 μM), or combined treatment for 48 h on cell death as determined by flow cytometric analysis and ATP production by ATP assay. (B) IVIS imaging of A549-luciferase xenograft tumors was taken. When the tumor mass volume reached to 100 mm³, the mice were treated orally 7 days/week. (n = 7 per each group). The dissemination of cells into lungs was monitored by IVIS (Xenogen). The data were expressed as photon flux (photons/s/cm²/steradian), and photon flux for each measurement is represented by a colour scale. (C) Tumor volumes were determined as described in Materials and Methods. (D) Proposed model for the role of ALDH in NSCLC metabolism. ALDH1L1 plays a key role in ATP synthesis through NADH production. NADH produced in the cytosol needs to be transported into the mitochondria by the malate-aspartate shuttle for ATP production. Data are representative of the mean and standard deviation three independent experiments. **p < 0.01, ***p < 0.001 compared to vehicle control.