Dual targeting of BCL-2 and MCL-1 in the presence of BAX breaks venetoclax resistance in human small cell lung cancer

Abstract

Background: No targeted drugs are currently available against small cell lung cancer (SCLC). BCL-2 family members are involved in apoptosis regulation and represent therapeutic targets in many malignancies.

Methods: Expression of BCL-2 family members in 27 SCLC cell lines representing all known four SCLC molecular subtypes was assessed by qPCR, Western blot and mass spectrometry-based proteomics. BCL-2 and MCL-1 inhibition (venetoclax and S63845, respectively) was assessed by MTT assay and flow cytometry and in mice bearing human SCLC tumours. Drug interactions were calculated using the Combenefit software. Ectopic BAX overexpression was achieved by expression plasmids.

Results: The highest BCL-2 expression levels were detected in ASCL1- and POU2F3-driven SCLC cells. Although sensitivity to venetoclax was reflected by BCL-2 levels, not all cell lines responded consistently despite their high BCL-2 expression. MCL-1 overexpression and low BAX levels were both characteristic for venetoclax resistance in SCLC, whereas the expression of other BCL-2 family members did not affect therapeutic efficacy. Combination of venetoclax and S63845 resulted in significant, synergistic in vitro and in vivo anti-tumour activity and apoptosis induction in double-resistant cells; however, this was seen only in a subset with detectable BAX. In non-responding cells, ectopic BAX overexpression sensitised to venetoclax and S63845 and, furthermore, induced synergistic drug interaction.

Conclusions: The current study reveals the subtype specificity of BCL-2 expression and sheds light on the mechanism of venetoclax resistance in SCLC. Additionally, we provide preclinical evidence that combined BCL-2 and MCL-1 targeting is an effective approach to overcome venetoclax resistance in high BCL-2-expressing SCLCs with intact BAX.

Fig. 1. BCL-2 expression pattern poorly correlates with venetoclax resistance in SCLC cell lines.

BCL-2 protein levels in SCLC cell lines, grouped according to SCLC molecular subtypes, from a proteomics analysis (N = 27). b BCL-2 levels of SCLC cell lines (N = 28) deriving from densitometric analysis of Western blots relative to GAPDH and representative Western blot pictures. Bars are shown as mean ± SD of 4–5 replicates. c BCL-2 mRNA levels (N = 28) relative to GAPDH measured by qPCR. d IC₅₀ values of venetoclax, calculated from dose–response curves after 72 h, assessed by MTT assay. The black line (4 µM) indicates the threshold between sensitive and resistant cells. e Spearman correlation between BCL-2 expression levels deriving from Western blot analysis and venetoclax IC₅₀ values (blue: SCLC-A, red: SCLC-N, yellow: SCLC-P, green: SCLC-Y). Black arrows show “outliers”. f BCL-2 expression deriving from Western blot analyses in SCLC cell lines categorised into venetoclax sensitive and resistant groups, according to the 4 µM threshold. Each dot represents the mean of one cell line. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 2. Venetoclax resistance is characterised by increased MCL-1 and decreased BAX levels.

Normalised log2 intensity of a MCL-1 and b BAX deriving from proteomic analysis of cell lines categorised into BCL-2 high-venetoclax sensitive (venS^BCL2high) and BCL-2 high-venetoclax resistant (venR^BCL2high) groups. c Pearson correlation between MCL-1 expression levels (deriving from Western blot analysis) and venetoclax IC₅₀ values. d Spearman correlation between BAX expression levels (deriving from Western blot analysis) and venetoclax IC₅₀ values (in c and d: blue: SCLC-A, red: SCLC-N, yellow: SCLC-P, green: SCLC-Y). e Dose–response curves of H146 and H146venR cells after 72 h, treated with venetoclax at the indicated doses. f Expression levels in H146venR cells relative to H146 deriving from Western blot analysis. Bars are shown as mean ± SD. g IC₅₀ values of S63845, calculated from dose-response curves after 72 h, assessed by MTT assay. The black line (2 µM) indicates the threshold between sensitive and resistant cells. h IC₅₀ values of venetoclax and S63845 (blue: SCLC-A, red: SCLC-N, yellow: SCLC-P, green: SCLC-Y). i BAX expression deriving from Western blot analysis in double sensitive (S/S), venetoclax sensitive (vS/R), S63845 sensitive (vR/S) and double resistant (R/R) cells. Each dot represents the mean of one cell line. ANOVA and Tukey’s multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 3. Venetoclax and S63845 synergise in a subset of SCLC cell lines in vitro and in vivo.

a Dose–response curves of SCLC cell lines after 72 h treatment with venetoclax and S63845 at the indicated doses, deriving from MTT assays and b HSA synergy/antagonism plots generated by the Combenefit software. Data are shown as mean ± SD. c Quantification of colony formation assays and representative pictures of SHP77 cells treated with DMSO (Co), 2.5 µM venetoclax (Ven), 2.5 µM S63845 (S6) or a combination of both (Ven+S6). Data are shown as mean ± SD of three biological replicates (ANOVA and Tukey’s multiple comparison test). d Tumour volume over time and tumour weight at day 24. SCID mice (n = 4–7 per group) were subcutaneously injected with SHP77 cells and treated with vehicle (Co), 100 mg/kg venetoclax 5× per week (Ven), 25 mg/kg S63845 2× per week (S6) or a combination of both (Ven+S6). Percentage of e TUNEL or f Ki67-positive cells in xenografts. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 4. Venetoclax and S63845 induce apoptosis in sensitive cells.

a Western blot of cleaved PARP and Caspase 3 in SCLC cells 1 or 2 days after treatment with 2.5 µM venetoclax (Ven), S63845 (S6) or a combination of both. Bars show densitometry relative to GAPDH of 2 biological replicates as mean ± SD. b Cells were treated with 2.5 µM venetoclax or a S63845, or a combination of both. Apoptosis was measured after 48 h by propidium iodide (PI) or AnnexinV staining via flow cytometry. Data are shown as mean + SD of 3 biological replicates. The high number of dead cells in the control groups is due to the non-adherent properties of the cell lines. c Representative flow cytometry plots after 48 h.

Fig. 5. BAX is required for synergistic interactions between venetoclax and S63845.

a Max. synergy score deriving from HSA/Combenefit analysis. The red line indicates the threshold between additive (<15, Add) and synergistic (>15, Syn), as indicated by the black line, resulting in 5 cell lines per group. b Expression profiles deriving from Western blot analysis between the synergistic (Syn) and additive/antagonistic (Add) cell line groups. Each dot represents the mean of one cell line. c Dose–response curves of H1048 cells transfected with eGFP (H1048^eGFP) or BAX (H1048^BAX) after 72 h treatment with venetoclax and S63845 at the indicated doses, deriving from MTT assays (left) and corresponding HSA synergy/antagonism plots generated by the Combenefit software (right). Data are shown as mean ± SD. d Max. synergy score and e synergism sum score, deriving from HSA analysis. f Colony formation assays of H1048^eGFP and H1048^BAX after 10 days of incubation with DMSO (Co), 1 µM venetoclax (Ven), 1 µM S63845 (S6) or a combination of both (Ven+S6) (ANOVA and Dunnett’s multiple comparison test), *p < 0.05, **p < 0.01, ***p < 0.001.