Bcl-XL but Not Bcl-2 Is a Potential Target in Medulloblastoma Therapy

Abstract

Medulloblastoma (MB) is the most common solid tumour in children and, despite current treatment with a rather aggressive combination therapy, accounts for 10% of all deaths associated with paediatric cancer. Breaking the tumour cells' intrinsic resistance to therapy-induced cell death should lead to less aggressive and more effective treatment options. In other tumour entities, this has been achieved by modulating the balance between the various pro- and anti-apoptotic members of the Bcl-2 family with small molecule inhibitors. To evaluate the therapeutic benefits of ABT-199 (Venetoclax), a Bcl-2 inhibitor, and ABT-263 (Navitoclax), a dual Bcl-XL/Bcl-2 inhibitor, increasingly more relevant model systems were investigated. Starting from established MB cell lines, progressing to primary patient-derived material and finally an experimental tumour system imbedded in an organic environment were chosen. Assessment of the metabolic activity (a surrogate readout for population viability), the induction of DNA fragmentation (apoptosis) and changes in cell number (the combined effect of alterations in proliferation and cell death induction) revealed that ABT-263, but not ABT-199, is a promising candidate for combination therapy, synergizing with cell death-inducing stimuli. Interestingly, in the experimental tumour setting, the sensitizing effect of ABT-263 seems to be predominantly mediated via an anti-proliferative and not a pro-apoptotic effect, opening a future line of investigation. Our data show that modulation of specific members of the Bcl-2 family might be a promising therapeutic addition for the treatment of MB.

Keywords: ABT-199 (Venetoclax); ABT-263 (Navitoclax); Bcl-2; Bcl-XL; apoptosis; chemotherapy; medulloblastoma.

Figure 1

Bcl-XL, but not Bcl-2, is a promising potential therapeutic target in medulloblastoma (MB) cell lines. (a) Summary of cell line characteristics. Abbreviations: SHH, sonic hedgehog; wt, wild type; amp, amplified; mut, mutant; m, male. (b) Immunoblotting of Bcl-2 family protein expression in MB cell lines. Extracts of four different MB cell lines were subjected to immunoblotting for Bcl-2 family proteins. β-actin served as loading control and the leukaemia cell line RS4;11 served as positive control for Bcl-2 and Mcl-1 expression. Anti-apoptotic members of the Bcl-2 family are marked by an asterisk. (c) The inhibitor of Bcl-2, ABT-199, does not show a therapeutic effect on MB cell lines. Cells were stimulated with 10 nM ABT-199 in the presence of either reduced serum concentrations (to 1.5% FCS for Daoy cells or 5% FCS for all others), or the chemotherapeutic drug doxorubicin (5 nM for Daoy cells or 10 nM for all others), or vincristine (1 nM for Daoy cells or 0.5 nM for all others) for either 24 (early) or 96 (late) h, as described in the Materials and Methods section. Left: Cell death was assessed by the surrogate readout of DNA fragmentation, evaluated via flow cytometric analysis of PI-stained nuclei. Right: The number of live cells was counted using the CASY1 DT cell counter. Using Bliss analysis, synergistic (green), additive (yellow) or antagonistic (red) effects were determined, while grey indicates that Bliss analysis was mathematically not possible to perform. Shown in (b) is a representative example of two independent experiments, while (c) summarizes at least three independent experiments performed in triplicate.

Figure 2

The Bcl-XL/Bcl-2 inhibitor ABT-263 shows therapeutic potential in medulloblastoma (MB) cell lines. (a) Monotreatment with ABT-263 affects the metabolic activity of MB cell lines. The MB cell lines were seeded and stimulated with a serial dilution of ABT-263 for 24 and 96 h. The relative viability of the respective populations was normalized to the solvent-treated control population. The boxed column represents the concentration selected for further experiments (1 µM). (b) Combining ABT-263 and different stressors can enhance apoptosis. The seeded cell lines were cultured in the presence of 1 µM ABT-263, reduced serum concentrations (to 1.5% FCS for Daoy cells or 5% FCS for all others), doxorubicin (5 nM for Daoy cells or 10 nM for all others) and vincristine (1 nM for Daoy cells or 0.5 nM for all others), as well as combinations of them (concentrations described in the Materials and Methods section). Percentage of specific DNA fragmentation was determined 24 and 96 h after stimulation by flow cytometric analysis of PI-stained nuclei. (c) Combining ABT-263 and different stressors can reduce living cell numbers. The experimental setup was similar to (b). This was followed by assessing the total living cell number 24 and 96 h after stimulation by CASY1 DT measurement. Data was normalized to a solvent-treated control population for each time point. (d) The inhibitor of Bcl-XL/Bcl-2, ABT-263, shows a therapeutic effect on MB cell lines. Similarly to the data shown in Figure 1b, the effects of combining ABT-263 with different stressors on apoptosis (left) and living cell number (right), as shown in (b,c), respectively (early, 24 h; late, 96 h), were analysed for synergism. Using Bliss analysis, synergistic (green), additive (yellow) or antagonistic (red) effects were determined, while grey indicates that Bliss analysis was mathematically not possible to perform. Shown in (a) are the mean and SD of at least three independent experiments with six individual data points. In (b) are the mean and SD of three independent experiments performed in triplicates, and in (c) are the mean and SD of at least four independent experiments performed in triplicates, while (d) is a mathematical summary of (b,c).

Figure 2

The Bcl-XL/Bcl-2 inhibitor ABT-263 shows therapeutic potential in medulloblastoma (MB) cell lines. (a) Monotreatment with ABT-263 affects the metabolic activity of MB cell lines. The MB cell lines were seeded and stimulated with a serial dilution of ABT-263 for 24 and 96 h. The relative viability of the respective populations was normalized to the solvent-treated control population. The boxed column represents the concentration selected for further experiments (1 µM). (b) Combining ABT-263 and different stressors can enhance apoptosis. The seeded cell lines were cultured in the presence of 1 µM ABT-263, reduced serum concentrations (to 1.5% FCS for Daoy cells or 5% FCS for all others), doxorubicin (5 nM for Daoy cells or 10 nM for all others) and vincristine (1 nM for Daoy cells or 0.5 nM for all others), as well as combinations of them (concentrations described in the Materials and Methods section). Percentage of specific DNA fragmentation was determined 24 and 96 h after stimulation by flow cytometric analysis of PI-stained nuclei. (c) Combining ABT-263 and different stressors can reduce living cell numbers. The experimental setup was similar to (b). This was followed by assessing the total living cell number 24 and 96 h after stimulation by CASY1 DT measurement. Data was normalized to a solvent-treated control population for each time point. (d) The inhibitor of Bcl-XL/Bcl-2, ABT-263, shows a therapeutic effect on MB cell lines. Similarly to the data shown in Figure 1b, the effects of combining ABT-263 with different stressors on apoptosis (left) and living cell number (right), as shown in (b,c), respectively (early, 24 h; late, 96 h), were analysed for synergism. Using Bliss analysis, synergistic (green), additive (yellow) or antagonistic (red) effects were determined, while grey indicates that Bliss analysis was mathematically not possible to perform. Shown in (a) are the mean and SD of at least three independent experiments with six individual data points. In (b) are the mean and SD of three independent experiments performed in triplicates, and in (c) are the mean and SD of at least four independent experiments performed in triplicates, while (d) is a mathematical summary of (b,c).

Figure 3

Characterisation of patient-derived medulloblastoma (MB) stem-like cells and short-term differentiated cells. (a) Schematic workflow of obtaining patient-derived MB cells. During therapeutically indicated surgery and following patient’s (or legal guardian’s) consent, a tumour sample surplus to histological characterisation and archiving was split into two. While one element was stored as reference material, the remaining tissue was further dissected and cleaned from blood cells. After several washes, fragments were sieved, and single cells/small cell clumps were cultured under non-adherent conditions in “stem cell” medium. Neurospheres that emerged after prolonged culturing were stored and characterised with reference to original tumour histology. Adherent, differentiated progeny were cultured from neurospheres via permitted adhesion and change of culture conditions, using “differentiation” medium. These differentiated populations were used for a maximum of ten generations to ensure genetic stability. Details are presented in the Materials and Methods section. Shown here are also exemplary pictures of the stem cell-like MB cells grown as free-flowing spheres and their adherently cultured differentiated progeny. Scale bar: 50 μm. (b) Neuropathology specimen of the patient’s tumour stained as indicated. The H&E staining shows small, round and densely packed cells. GFAP staining is negative as for most MB tumours, while a relatively high count of Ki-67 positive areas indicates an increased amount of cycling cells. MAP2 confirms the CNS origin of the tumour, while ß-catenin indicates that the tumour exhibits SHH activity. In addition, Alcian and PAS was used to detect polysaccharides, elevated in cancers. Taken together, these data suggest a medulloblastoma belonging to the SHH subgroup. Abbreviations: H&E—haematoxylin and eosin stain; GFAP—Glial Fibrillary Acidic Protein; Ki-67—Antigen KI-67; MAP2—Microtubule-Associated Protein 2; ß-catenin—βeta-catenin; Alcian & PAS—Alcian blue stain and Perodic acid–Schiff stain, SHH–sonic hedgehog. Scale bar: 200 µm. (c) Expression protein profile in MB stem cell-like cells and differentiated cells shown as a heat map. Protein extracts of a stem cell-like cell population (SC) and its differentiated progeny (DC) were subjected to immunoblotting for three different types of protein profile assays, as described in the Materials and Methods section. Following visualisation, membranes were densitometrically evaluated and bioinformatically processed. The map shows relative expression (blue, higher than average; red, lower than average; as described in the Materials and Methods section) and represents the mean of two independent experiments.

Figure 3

Characterisation of patient-derived medulloblastoma (MB) stem-like cells and short-term differentiated cells. (a) Schematic workflow of obtaining patient-derived MB cells. During therapeutically indicated surgery and following patient’s (or legal guardian’s) consent, a tumour sample surplus to histological characterisation and archiving was split into two. While one element was stored as reference material, the remaining tissue was further dissected and cleaned from blood cells. After several washes, fragments were sieved, and single cells/small cell clumps were cultured under non-adherent conditions in “stem cell” medium. Neurospheres that emerged after prolonged culturing were stored and characterised with reference to original tumour histology. Adherent, differentiated progeny were cultured from neurospheres via permitted adhesion and change of culture conditions, using “differentiation” medium. These differentiated populations were used for a maximum of ten generations to ensure genetic stability. Details are presented in the Materials and Methods section. Shown here are also exemplary pictures of the stem cell-like MB cells grown as free-flowing spheres and their adherently cultured differentiated progeny. Scale bar: 50 μm. (b) Neuropathology specimen of the patient’s tumour stained as indicated. The H&E staining shows small, round and densely packed cells. GFAP staining is negative as for most MB tumours, while a relatively high count of Ki-67 positive areas indicates an increased amount of cycling cells. MAP2 confirms the CNS origin of the tumour, while ß-catenin indicates that the tumour exhibits SHH activity. In addition, Alcian and PAS was used to detect polysaccharides, elevated in cancers. Taken together, these data suggest a medulloblastoma belonging to the SHH subgroup. Abbreviations: H&E—haematoxylin and eosin stain; GFAP—Glial Fibrillary Acidic Protein; Ki-67—Antigen KI-67; MAP2—Microtubule-Associated Protein 2; ß-catenin—βeta-catenin; Alcian & PAS—Alcian blue stain and Perodic acid–Schiff stain, SHH–sonic hedgehog. Scale bar: 200 µm. (c) Expression protein profile in MB stem cell-like cells and differentiated cells shown as a heat map. Protein extracts of a stem cell-like cell population (SC) and its differentiated progeny (DC) were subjected to immunoblotting for three different types of protein profile assays, as described in the Materials and Methods section. Following visualisation, membranes were densitometrically evaluated and bioinformatically processed. The map shows relative expression (blue, higher than average; red, lower than average; as described in the Materials and Methods section) and represents the mean of two independent experiments.

Figure 4

The Bcl-XL/Bcl-2 inhibitor ABT-263, but not the Bcl-2 inhibitor ABT-199, shows therapeutic potential in primary medulloblastoma (MB) cells. (a) The inhibitor of Bcl-2, ABT-199, does not show a therapeutic effect on the metabolism of primary MB cells. The stem cell-like cells (SC322, left) or their differentiated progeny (DC322, right) were cultured in the presence of ABT-199, reduced growth factor/serum concentrations, doxorubicin or two concentrations of vincristine as well as ABT-199 combined with the different stressors (concentrations as indicated). Metabolic activity was assessed after 24 and 96 h and normalized to the solvent-treated control population. (b) The inhibitor of Bcl-XL/Bcl-2, ABT-263, shows a therapeutic effect on the metabolism of primary MB cells. The experimental setup was similar to 4A. Metabolic activity was assessed after 24 and 96 h and normalized to the solvent-treated control population. (c) ABT-263 shows a therapeutic effect on cell death induction of primary MB cells. The stem cell-like cells (SC322, left) or their differentiated progeny (DC322, right) were cultured in the presence of 1 µM ABT-263, reduced growth factor/serum concentrations (as indicated), 5 nM doxorubicin or 0.5 nM vincristine as well as ABT-263 combined with the different stressors. Percentage of specific DNA fragmentation was determined 24 and 96 h after stimulation by flow cytometric analysis of PI-stained nuclei. (d) ABT-263 shows a therapeutic effect on living cell number of primary MB cells. The experimental setup was similar to 4C. The total living cell number was assessed 24 and 96 h after stimulation by CASY1 DT measurement and normalized to solvent-treated control population for each time point. (e) Combining ABT-263 and various stressors frequently elicits a synergistic effect in primary MB cells. The effects of combining ABT-263 with different stressors on apoptosis (left) and living cell number (right), as shown in (c) and (d), respectively (early, 24 h; late, 96 h), were analysed for synergism. Using Bliss analysis, synergistic (green), additive (yellow) or antagonistic (red) effects were determined, while grey indicates that Bliss analysis was mathematically not possible to perform. In (a–d) columns represent mean and SD of three independent experiments performed in triplicates.

Figure 4

The Bcl-XL/Bcl-2 inhibitor ABT-263, but not the Bcl-2 inhibitor ABT-199, shows therapeutic potential in primary medulloblastoma (MB) cells. (a) The inhibitor of Bcl-2, ABT-199, does not show a therapeutic effect on the metabolism of primary MB cells. The stem cell-like cells (SC322, left) or their differentiated progeny (DC322, right) were cultured in the presence of ABT-199, reduced growth factor/serum concentrations, doxorubicin or two concentrations of vincristine as well as ABT-199 combined with the different stressors (concentrations as indicated). Metabolic activity was assessed after 24 and 96 h and normalized to the solvent-treated control population. (b) The inhibitor of Bcl-XL/Bcl-2, ABT-263, shows a therapeutic effect on the metabolism of primary MB cells. The experimental setup was similar to 4A. Metabolic activity was assessed after 24 and 96 h and normalized to the solvent-treated control population. (c) ABT-263 shows a therapeutic effect on cell death induction of primary MB cells. The stem cell-like cells (SC322, left) or their differentiated progeny (DC322, right) were cultured in the presence of 1 µM ABT-263, reduced growth factor/serum concentrations (as indicated), 5 nM doxorubicin or 0.5 nM vincristine as well as ABT-263 combined with the different stressors. Percentage of specific DNA fragmentation was determined 24 and 96 h after stimulation by flow cytometric analysis of PI-stained nuclei. (d) ABT-263 shows a therapeutic effect on living cell number of primary MB cells. The experimental setup was similar to 4C. The total living cell number was assessed 24 and 96 h after stimulation by CASY1 DT measurement and normalized to solvent-treated control population for each time point. (e) Combining ABT-263 and various stressors frequently elicits a synergistic effect in primary MB cells. The effects of combining ABT-263 with different stressors on apoptosis (left) and living cell number (right), as shown in (c) and (d), respectively (early, 24 h; late, 96 h), were analysed for synergism. Using Bliss analysis, synergistic (green), additive (yellow) or antagonistic (red) effects were determined, while grey indicates that Bliss analysis was mathematically not possible to perform. In (a–d) columns represent mean and SD of three independent experiments performed in triplicates.

Figure 4

The Bcl-XL/Bcl-2 inhibitor ABT-263, but not the Bcl-2 inhibitor ABT-199, shows therapeutic potential in primary medulloblastoma (MB) cells. (a) The inhibitor of Bcl-2, ABT-199, does not show a therapeutic effect on the metabolism of primary MB cells. The stem cell-like cells (SC322, left) or their differentiated progeny (DC322, right) were cultured in the presence of ABT-199, reduced growth factor/serum concentrations, doxorubicin or two concentrations of vincristine as well as ABT-199 combined with the different stressors (concentrations as indicated). Metabolic activity was assessed after 24 and 96 h and normalized to the solvent-treated control population. (b) The inhibitor of Bcl-XL/Bcl-2, ABT-263, shows a therapeutic effect on the metabolism of primary MB cells. The experimental setup was similar to 4A. Metabolic activity was assessed after 24 and 96 h and normalized to the solvent-treated control population. (c) ABT-263 shows a therapeutic effect on cell death induction of primary MB cells. The stem cell-like cells (SC322, left) or their differentiated progeny (DC322, right) were cultured in the presence of 1 µM ABT-263, reduced growth factor/serum concentrations (as indicated), 5 nM doxorubicin or 0.5 nM vincristine as well as ABT-263 combined with the different stressors. Percentage of specific DNA fragmentation was determined 24 and 96 h after stimulation by flow cytometric analysis of PI-stained nuclei. (d) ABT-263 shows a therapeutic effect on living cell number of primary MB cells. The experimental setup was similar to 4C. The total living cell number was assessed 24 and 96 h after stimulation by CASY1 DT measurement and normalized to solvent-treated control population for each time point. (e) Combining ABT-263 and various stressors frequently elicits a synergistic effect in primary MB cells. The effects of combining ABT-263 with different stressors on apoptosis (left) and living cell number (right), as shown in (c) and (d), respectively (early, 24 h; late, 96 h), were analysed for synergism. Using Bliss analysis, synergistic (green), additive (yellow) or antagonistic (red) effects were determined, while grey indicates that Bliss analysis was mathematically not possible to perform. In (a–d) columns represent mean and SD of three independent experiments performed in triplicates.

Figure 5

Combining the Bcl-XL/Bcl-2 inhibitor ABT-263 and vincristine shows superior therapeutic potential over single treatment in a chorioallantoic membrane (CAM) assay. (a) The surface area of double treated tumours appears to be reduced compared to mock or single treated entities. A 20:80 mixture of medulloblastoma stem cell-like cells and their differentiated progeny (SC/DC322) was seeded on the CAM of fertilized chicken eggs and were treated daily for four consecutive days with the indicated substances via local surface application (ABT-263: 5 µM, vincristine: 2.5 nM). After the aforementioned time span, the tumour surface area of all tumours (as shown on the right) was determined as indicated (bottom left). While a trend towards reduction is clearly visible, the reduction of surface area was statistically not significant, as determined by a one-way ANOVA test (p-value 0.0623). (b) Combining ABT-263 and vincristine has an effect on tumour morphology and cycling cells, but not necessarily on apoptosis induction. Shown are representative sections taken from the tumour centre of an individual tumour per treatment group, stained for haematoxylin and eosin (H&E), haematoxylin and Ki-67 (Ki-67) or haematoxylin and cleaved Caspase-3 (cCaspase-3). Scale: bars indicate 1000 µm, except in the columns labelled “(5x)”; here, bars indicate 200 µm. (c) Combining ABT-263 and vincristine increases the fraction of non-cycling cells. A histological evaluation of cells progressing through the cell cycle (Ki-67 positive) was performed. + indicates some positive cells clearly detectable, ++ indicates the majority of cells are positive, +++ almost all cells are positive; n/a, not available. Fields marked in yellow indicate the tumours depicted in (b). (d) High concentrations of ABT-263 can induce apoptosis. Depicted is a section of a tumour treated with ABT-263 only. Clearly visible is the positive staining for cleaved Caspase-3 (cCaspase-3) in the top layers of the section. These are the layers to which ABT-263 was directly applied and then allowed to diffuse through the tumour. Scale bar indicates 200 µm.

Figure 5

Combining the Bcl-XL/Bcl-2 inhibitor ABT-263 and vincristine shows superior therapeutic potential over single treatment in a chorioallantoic membrane (CAM) assay. (a) The surface area of double treated tumours appears to be reduced compared to mock or single treated entities. A 20:80 mixture of medulloblastoma stem cell-like cells and their differentiated progeny (SC/DC322) was seeded on the CAM of fertilized chicken eggs and were treated daily for four consecutive days with the indicated substances via local surface application (ABT-263: 5 µM, vincristine: 2.5 nM). After the aforementioned time span, the tumour surface area of all tumours (as shown on the right) was determined as indicated (bottom left). While a trend towards reduction is clearly visible, the reduction of surface area was statistically not significant, as determined by a one-way ANOVA test (p-value 0.0623). (b) Combining ABT-263 and vincristine has an effect on tumour morphology and cycling cells, but not necessarily on apoptosis induction. Shown are representative sections taken from the tumour centre of an individual tumour per treatment group, stained for haematoxylin and eosin (H&E), haematoxylin and Ki-67 (Ki-67) or haematoxylin and cleaved Caspase-3 (cCaspase-3). Scale: bars indicate 1000 µm, except in the columns labelled “(5x)”; here, bars indicate 200 µm. (c) Combining ABT-263 and vincristine increases the fraction of non-cycling cells. A histological evaluation of cells progressing through the cell cycle (Ki-67 positive) was performed. + indicates some positive cells clearly detectable, ++ indicates the majority of cells are positive, +++ almost all cells are positive; n/a, not available. Fields marked in yellow indicate the tumours depicted in (b). (d) High concentrations of ABT-263 can induce apoptosis. Depicted is a section of a tumour treated with ABT-263 only. Clearly visible is the positive staining for cleaved Caspase-3 (cCaspase-3) in the top layers of the section. These are the layers to which ABT-263 was directly applied and then allowed to diffuse through the tumour. Scale bar indicates 200 µm.