Figure 1.
Cytotoxic effect of coibamide A on human glioblastoma cells.
(A) Cell death in U87-MG (upper panels) and SF-295 (lower panels) glioma cells after a 3-day exposure to coibamide A (20 nM). Cell death was documented by morphological evaluation of vehicle-(DMSO; panels a and c) or coibamide A-treated (panels b and d) cells using light microscopy. (B) Concentration-response profile for coibamide A-induced cytotoxicity in U87-MG and SF-295 cells. Glioma cells were treated with increasing concentrations of coibamide A (2.3 to 230 nM) for 3 days. Cytotoxicity was determined by MTT assay with the viability of control cells defined as 100%. Dose-response data represent mean viability ± SE (n = 3 wells per treatment) from a comparison that was repeated in at least four independent experiments.
Figure 2.
Coibamide A induces a time- and concentration-dependent decrease in plasma membrane integrity.
(A) Trypan Blue exclusion test of cell viability in U87-MG cultures treated with or without coibamide A (20 nM) or vehicle (DMSO) over 4 days. (B) Concentration-response relationship of LDH efflux in U87-MG cells treated with or without coibamide A or vehicle for 4 days (C) Trypan Blue exclusion test of cell viability and (D) concentration-response relationship of LDH efflux in SF-295 cells treated with or without coibamide A or vehicle (as above) for 4 days. Trypan Blue exclusion profiles represent mean cell counts ± SD of a representative time course performed in triplicate and repeated at least three times in each cell line. Histograms represent mean LDH efflux ± SE of at least three independent experiments performed in triplicate (**P<0.01 and ***P<0.001, coibamide A vs. untreated control).
Figure 3.
Coibamide A induces activation of caspase-3,7 and apoptosis in a cell type-specific manner.
Concentration-response relationship for coibamide A-induced caspase-3,7 activation in (A) U87-MG and (B) SF-295 cells exposed to coibamide A (10–300 nM) or vehicle (DMSO) for 48 h. Coibamide A-induced caspase-3,7 activation in (C) U87-MG and (D) SF-295 cells exposed to coibamide A (10–300 nM) for 72 h. Caspase-3,7 activity was measured at the end of each treatment using a luminescence-based assay coupled with automated luminescence detection. Histograms represent mean ± SE of up to four independent determinations (**P<0.01 and ***P<0.001, coibamide A vs. vehicle control). (E) Immunoblot analysis of PARP1 in coibamide A-treated (100 nM) U87-MG and SF-295 cells lystates. Adherent and detached (Det) cells were harvested over 4 days and examined for expression of the large 89 kDa fragment of PARP1, a marker of apoptosis, plus actin as a loading control. Immunoblot is representative of an experiment repeated at least three times with similar results.
Figure 4.
The broad-spectrum caspase inhibitor Z-VAK-FMK inhibits LDH release but does not prevent coibamide A-induced cell death.
(A) Time course of LDH efflux from U87-MG cells over a 4 day exposure to coibamide A (300 nM) in the presence or absence of Z-VAD-FMK (100 µM). (B) Concentration-response relationship of coibamide A-induced cell death in the presence or absence of Z-VAD-FMK (100 µM). (C) Time course of LDH efflux and, (D) concentration-response relationship of coibamide A-induced cell death in SF-295 cells measured in the presence or absence of Z-VAD-FMK (100 µM). Histograms represent mean LDH efflux ± SE of three independent experiments performed in triplicate (**P<0.01 and ***P<0.001, coibamide A vs. control). Cytotoxicity was determined by MTT assay after 3 days of exposure to coibamide A (300 nM) with the viability of control cells defined as 100%; cell viability in the presence of Z-VAD-FMK alone (100 µM) is represented by an open circle. Dose-response data represent mean viability ± SE (n = 3 wells per treatment) from a comparison that was repeated in three independent experiments.
Figure 5.
Linearized analogs of coibamide A are not cytotoxic.
(A) Structures of coibamide A (1), and two linearized derivatives: dehydrated seco acid (2), and seco acid (3). (B and C) Comparison of cell viability in human U87-MG glioblastoma and mouse Neuro-2A neuroblastoma cells, respectively, following exposure to the coibamide A (20 nM) parent structure, linear forms 2 and 3 (both 30 µg/mL, ∼230–260 nM), or vehicle (DMSO) for 4 days. (D) Trypan Blue exclusion test of cell viability over 4 days in SF-295 cultures treated with or without the coibamide A (20 nM), linear forms 2 and 3 (both 30 µg/mL, ∼230–260 nM), or vehicle (DMSO). Histograms represent mean cell viability ± SE of at least three independent MTT assays performed in triplicate (*P<0.05 and ** P<0.01, coibamide A vs. control). Trypan Blue exclusion profile represents mean cell counts ± SD of a representative time course performed in triplicate that was repeated three times with similar results.
Figure 6.
Coibamide A induces biochemical and morphological features of autophagy.
(A) Time course of endogenous LC3 expression in U87-MG cell lysates treated with coibamide A (20 nM) for up to 48 h, relative to vehicle (DMSO) or rapamycin- (100 nM)-treated control cells. Cell lysates were subjected to immunoblot analysis with an antibody recognizing both forms of LC3: cytosolic LC3-I and the autophagosome-associated LC3-II. Anti-GAPDH was used as a loading control. Data are representative of a time course repeated at least three times. (B) Monodansylcadaverine (MDC) staining of U87-MG cells treated with vehicle (DMSO; panel a), coibamide A (20 nM; panel b), or rapamycin (100 nM; panel c) for 48 h. After treatment cells were stained with MDC (50 mM) for 10 min. Images are representative of a pattern of staining observed in at least three independent experiments. Scale bar = 10 µm. (C) Electron micrographs of adherent U87-MG cells treated with or without coibamide A (20 nM). Ultrastructure of representative control (panel a) and treated U87-MG cells at 16 (panel b) and 48 h (panels c to e). By 48 h coibamide A-treated cells showed empty vacuoles that could be distinguished from autophagosomes (A) with a discernible double membrane. The nucleus (N), mitochondria (M), lysosome (L) and scale bars are indicated.
Figure 7.
Coibamide A induces autophagosome accumulation in U87-MG cells.
(A) Immunoblot analysis of endogenous LC3 in U87-MG cell lysates treated with vehicle (control) or coibamide A (30 nM) for 4 h, each in the presence or absence of bafilomycin A1 (10 nM) for the final 1 h of treatment. (B) Quantitation of LC3-II levels as shown in panel A. Histogram represents mean LC3-II intensity, normalized to tubulin, from immunoblots generated in three independent experiments. (C) Immunocytochemistry of U87-MG cells treated with vehicle (control) or coibamide A (30 nM) for 4 h, each in the presence of absence of bafilomycin (10 nM) for the final 1 h of treatment. Cells were stained for endogenous LC3 (red) and DAPI (blue). Scale bar, 10 µm. (D) Quantitation of mean LC3 puncta per cell (n = 32 cells) from three independent experiments.
Figure 8.
Coibamide A does not block ligand-induced degradation of EGFR.
(A) Immunoblot analysis of EGFR expression up to 4 h after the addition of ligand (EGF) in the presence of vehicle (control), coibamide A (30 nM) or bafilomycin A1 (100 nM). On the day of the experiment U87-MG cells were starved of serum, to promote membrane localization of EGFR, and at t = 0 were chased with EGF (100 ng/mL) and cycloheximide (25 µg/mL), with or without coibamide A, or bafilomycin A1. Cells were lysed at the times indicated and processed for immunoblot analysis. (B) Quantitation of ligand-induced EGFR degradation from three independent experiments as described in A. Graph represents % EGFR expression, following the addition of EGF, as a function of incubation time in the presence of vehicle, coibamide A or bafilomycin A1. EGFR intensity was normalized to tubulin and determined relative to EGFR level at t = 0.
Figure 9.
Coibamide A induces autophagy in an ATG5-dependent manner.
(A) Immunoblot analysis of endogenous LC3 in wild-type and ATG5 null MEF cell lysates treated with vehicle (control) or coibamide A (30 nM) for 4 h, each in the presence or absence of bafilomycin (10 nM) for the final 1 h of treatment. (B) Quantitation of LC3-II levels in wild-type MEF cells as shown in panel A. Histogram represents mean LC3-II intensity, normalized to tubulin, from immunoblots generated in three independent experiments. (C) Immunocytochemistry of wild-type MEF cells treated with vehicle (control) or coibamide A (30 nM) for 4 h, each in the presence of absence of bafilomycin A1 (10 nM) for the final 1 h of treatment. Cells were stained for endogenous LC3 (red) and DAPI (blue). Scale bar, 10 µm. (D) Quantitation of mean LC3 puncta per cell (n = 32 cells) from three independent experiments.
Figure 10.
Coibamide A induces mTOR-independent autophagy.
Coibamide A-induced signaling is mTOR-independent and can be distinguished from starvation- and AZD 8055-induced autophagy in MEF and U87-MG glioblastoma cells. Cells were incubated in EBSS starvation medium, or treated with or without coibamide A (30 nM) or the dual mTOR inhibitor AZD 8055 (300 nM) in standard nutrient-rich medium for 4 h. Following treatment cells were lysed and subjected to immunoblot analysis. The phosphorylation status of residues considered to be major targets of the mTOR pathway was assessed: phospho-ULK1 (Thr-757), phospho-p70 S6 Kinase (Thr-389), phospho-S6 ribosomal protein (Ser-235/236), and phospho-4E-BP1 (Thr-37/46) are shown relative to total ULK1, S6K1, S6 ribosomal protein and 4-E binding protein, respectively. Tubulin served as a loading control. Results are representative of at least three independent experiments.
Figure 11.
Coibamide A induces cell death in the absence of ATG5 and intrinsic Apoptotic activating factor 1 (Apaf-1).
(A) Comparison of coibamide A-induced cytotoxicity in wild-type, ATG5 null and Apaf-1 null MEFs. MEFs cells were treated with increasing concentrations of coibamide A (0.1 to 300 nM) for 48 h. Cytotoxicity was determined by MTT assay with the viability of control cells defined as 100%. Dose-response data represent mean viability ± SE (n = 3 wells per treatment) from a comparison that was repeated in at least three independent experiments. (B) Immunoblot analysis of autophaphagy (LC3) and apoptotic (PARP1 and caspase-3) markers following treatment of wild-type, ATG5-null and Apaf-1-null MEFs with coibamide A (30 nM). Adherent and detached (Det) cells were harvested after 24 h and examined for expression of: the large 89 kDa fragment of PARP1, full length and cleaved caspase-3, LC3 isoforms I and II plus tubulin as a loading control. Immunoblot is representative of an experiment repeated at least three times with similar results.