Primary acute lymphoblastic leukemia cells are susceptible to microtubule depolymerization in G1 and M phases through distinct cell death pathways

Abstract

Microtubule targeting agents (MTAs) are widely used cancer chemotherapeutics which conventionally exert their effects during mitosis, leading to mitotic or postmitotic death. However, accumulating evidence suggests that MTAs can also generate death signals during interphase, which may represent a key mechanism in the clinical setting. We reported previously that vincristine and other microtubule destabilizers induce death not only in M phase but also in G1 phase in primary acute lymphoblastic leukemia cells. Here, we sought to investigate and compare the pathways responsible for phase-specific cell death. Primary acute lymphoblastic leukemia cells were subjected to centrifugal elutriation, and cell populations enriched in G1 phase (97%) or G2/M phases (80%) were obtained and treated with vincristine. We found death of M phase cells was associated with established features of mitochondrial-mediated apoptosis, including Bax activation, loss of mitochondrial transmembrane potential, caspase-3 activation, and nucleosomal DNA fragmentation. In contrast, death of G1 phase cells was not associated with pronounced Bax or caspase-3 activation but was associated with loss of mitochondrial transmembrane potential, parylation, nuclear translocation of apoptosis-inducing factor and endonuclease G, and supra-nucleosomal DNA fragmentation, which was enhanced by inhibition of autophagy. The results indicate that microtubule depolymerization induces distinct cell death pathways depending on during which phase of the cell cycle microtubule perturbation occurs. The observation that a specific type of drug can enter a single cell type and induce two different modes of death is novel and intriguing. These findings provide a basis for advancing knowledge of clinical mechanisms of MTAs.

Keywords: Bcl-2 proteins; EndoG; acute lymphoblastic leukemia; apoptosis; apoptosis-inducing factor; autophagy; cell death; microtubule depolymerization; parylation.

Figure 1

Analysis of DNA content of phase-purified ALL cells. Asynchronous ALL-5 cells were subjected to centrifugal elutriation and pools corresponding to cells in different cell cycle phases obtained, as described in Experimental procedures. Cells were stained with propidium iodide and subjected to flow cytometry to determine DNA content as an indicator of cell cycle phase purity. A, Asynchronous cells; B, G1 phase (2N DNA); C, G2/M phases (4N DNA). The results shown are representative of numerous (>40) experiments.

Figure 2

Vincristine-induced death of G2/M but not G1 phase ALL cells is associated with phosphorylation of pro-survival Bcl-2 proteins. G1 and G2/M phase ALL cells were isolated by centrifugal elutriation and treated with vehicle (0.1% DMSO) or 100 nM vincristine (VCR) for the times (h) indicated. Asynchronous KB-3 cells were also treated as indicated. Whole cell extracts were prepared and subjected to immunoblot analyses for PARP, total Mcl-1, phospho-Mcl-1, total Bcl-xL, phospho-Bcl-xL, total Bcl-2, or GAPDH as a loading control. Molecular mass standards are indicated on the left; molecular masses of the individual proteins are given in Experimental procedures. Images are representative of three independent experiments. ALL, acute lymphoblastic leukemia; PARP, poly (ADP)-ribose polymerase.

Figure 3

Vincristine treatment strongly activates Bax in G2/M but not G1 phase ALL cells. Asynchronous ALL-5 cells were treated with 100 nM ABT-263 for 6 h as a positive control for Bax activation. G1 and G2/M phase ALL-5 cells were isolated using centrifugal elutriation and treated with vehicle (0.1% DMSO) or 100 nM VCR for the times indicated. Samples were lysed under nondenaturing conditions, and lysates were subjected to immunoprecipitation for active Bax using 6A7 antibody, which recognizes the active conformation of Bax. Immunoprecipitated samples (IP) and whole cell extracts (WCEs) were analyzed by immunoblotting for Bax, PARP, cyclin B, or GAPDH as a loading control. Mock immunoprecipitation (No Ab Ctrl, upper right lane) was conducted as described in Experimental procedures. ALL, acute lymphoblastic leukemia; VCR, vincristine.

Figure 4

Vincristine activates caspase-3 more strongly in G2/M than G1 phase cells. G1 phase and G2/M phase ALL-5 cells were isolated using centrifugal elutriation and treated with vehicle (0. 1% DMSO) or 100 nM VCR for the times indicated. Protein extracts were prepared and caspase-3 activity determined as described in Experimental procedures. Data shown are mean ± S.D. (n ≥ 3) with p values indicated. ALL, acute lymphoblastic leukemia; VCR, vincristine.

Figure 5

Differential effects of caspase inhibition on DNA fragmentation in G1 versus G2/M phase cells after vincristine treatment. A, G1 or B, G2/M phase ALL-5 cells, isolated via centrifugal elutriation, were treated with 100 nM VCR, 100 μM Z-VAD-FMK, or vehicle (0.1% DMSO), alone or in combination, as indicated. Cells were harvested and stained with propidium iodide and analyzed for DNA content as described in Experimental procedures. Data shown represent percent of cells with sub-G1 DNA content (mean ± SD, n = 3). p values are indicated; n.s. = not significantly different. ALL, acute lymphoblastic leukemia; VCR, vincristine.

Figure 6

Analysis of DNA fragmentation by gel electrophoresis. A, analysis of small fragments. DNA was prepared as described in Experimental procedures and samples (8 μg) separated by 2% agarose gel electrophoresis. Lanes are as indicated and as follows: M, DNA standards, with molecular masses indicated at left; 1, G1 phase cells, vehicle (0.1% DMSO), 24 h; 2, G1 phase cells, vehicle, 48 h; 3, G1 phase cells, 100 nM VCR, 24 h; 4, G1 phase cells, 100 nM VCR, 48 h; 5, G2/M phase cells, vehicle, 24 h; 6, G2/M phase cells, 100 nM VCR, 24 h; 7, asynchronous cells, 100 nM ABT-263, 24 h. B, analysis of large fragments. DNA was prepared as described in Experimental procedures and samples (0.2 μg) separated by pulsed field gel electrophoresis. Lanes as in panel A. Gels were stained with ethidium bromide. ALL, acute lymphoblastic leukemia; VCR, vincristine.

Figure 7

Vincristine induces loss of mitochondrial transmembrane potential in both G1 and G2/M phase ALL cells. Changes in mitochondrial membrane potential were determined by measuring JC-1 fluorescence, as described in Experimental procedures. Panels A–C show fluorescent traces, with respect to time of incubation with JC-1, derived from the following treatment conditions. In all cases, 0.1% DMSO was used as the vehicle control and 10 μM FCCP as a positive control for decay of fluorescence. A, asynchronous ALL-5 cells treated with ABT-263 for 60 min; B, G1 phase cells treated with 100 nM VCR for 24 h; C, G2/M phase cells treated with 100 nM VCR for 12 h. Panels D and E show results (mean ± S.D., n = 3, with p values indicated) normalized to that of FCCP (100% loss of mitochondrial membrane potential) derived from tracings at 120 min incubation with JC-1 for either: D, G1 phase cells, or E, G2/M phase cells, treated with 100 nM VCR or 0.1% DMSO for the times indicated. ALL, acute lymphoblastic leukemia; FCCP, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone; VCR, vincristine.

Figure 8

Death of G1 phase ALL cells in response to vincristine is associated with parylation and nuclear translocation of AIF and EndoG.A, non-nuclear (NN) and nuclear (NE) extracts were prepared from control and VCR (100 nM, 16 h)-treated ALL cells in G1 phase as described in Experimental procedures, and 20 μg protein/lane analyzed by immunoblotting for the indicated proteins. WCE, whole cell extract. B, G1 phase ALL cells were treated with vehicle (0.1% DMSO), 100 nM VCR, or 3 μM olaparib, alone and in combination, for 16 h, and non-nuclear and nuclear extracts prepared. Extracts (40 μg protein) were subjected to immunoblotting using an antibody that recognizes PAR polymers. C, nuclear extracts from A were subjected to immunoblotting for AIF and PARP as indicated. D, effect of olaparib on VCR-induced DNA fragmentation. ALL-5 cells were treated with vehicle, 100 nM VCR, or 3 μM olaparib, alone and in combination, for 24 h or 48 h, as indicated, and sub-G1 DNA content determined by propidium iodide staining and flow cytometry, as described in Experimental procedures. Results shown are mean ± S.D., n = 3; n.s. = not significantly different. E, non-nuclear and nuclear extracts from A were subjected to immunoblotting for EndoG, PARP, and GAPDH, as indicated. For panels A–C, and E, molecular mass standards are shown at the left. AIF, apoptosis-inducing factor; ALL, acute lymphoblastic leukemia; EndoG, endonuclease G; PARP, poly (ADP)-ribose polymerase; VCR, vincristine.

Figure 9

Vincristine treatment of G1 phase ALL cells is associated with autophagy.A, representative electron micrographs of G1 phase ALL-5 cells treated with vehicle (0.1% DMSO) or 100 nM VCR for 12 h. MT, mitochondrion; ER, endoplasmic reticulum; AP, autophagosome; AL, autolysosome; #, undigested cytoplasmic content; ##, undigested cytoplasmic content with same electron density as mitochondrion. B, quantitation of autophagic vesicle area. Autophagic vesicle area, normalized to cytoplasmic area and expressed as a ratio, was determined for a total of eight images of control and nine images of VCR-treated G1 phase ALL-5 cells. Results are displayed as a scatter plot with mean and SD indicated. C, effect of the autophagy inhibitor bafilomycin A1 (BAF) on VCR-induced DNA fragmentation. G1 phase ALL-5 cells were treated with vehicle, 100 nM VCR, or 3 nM bafilomycin A1, alone and in combination, for 36 h, as indicated, and sub-G1 DNA content determined by propidium iodide staining and flow cytometry, as described in Experimental procedures. Results shown are mean ± SD, n = 3. ALL, acute lymphoblastic leukemia; VCR, vincristine.