Evaluation of an Actinomycin D/VX-680 aurora kinase inhibitor combination in p53-based cyclotherapy

Abstract

p53-Based cyclotherapy is proving to be a promising approach to palliate undesired effects of chemotherapy in patients with tumours carrying p53 mutations. For example, pre-treatment of cell cultures with Nutlin-3, a highly-selective inhibitor of the p53-mdm2 interaction, has been successfully used as a cytostatic agent to protect normal cells, but not p53-defective cells, from subsequent treatment with mitotic poisons or S-phase specific drugs. Here we sought to evaluate whether low doses of Actinomycin D (LDActD), a clinically-approved drug and potent p53 activator, could substitute Nutlin-3 in p53-based cyclotherapy. We found that pre-treatment with LDActD before adding the aurora kinase inhibitor VX-680 protects normal fibroblasts from polyploidy and nuclear morphology abnormalities induced by VX-680. However, and although to a lower extent than normal fibroblasts, tumour cell lines bearing p53 mutations were also protected by LDActD (but not Nutlin-3) from VX-680-induced polyploidy. We also report that a difference between the response of p53 wild-type cells and p53-defective cells to the LDActD/VX-680 sequential combination is that only the former fail to enter S-phase and therefore accumulate in G1/G0. We propose that drugs that incorporate into DNA during S-phase may perform better as second drugs than mitotic poisons in cyclotherapy approaches using LDActD as a cytostatic agent.

Figure 1. LDActD's effects on cell survival are p53-dependent

HCT116 cells with wild-type p53 or knockout for p53 were treated with ActD for 72 hours, after which they were fixed and stained with Giemsa.

Figure 2. Effects of ActD and nutlin-3 on p53 levels

MCF7 cells were treated with 2 nM ActD, 4 nM ActD or 5 μM Nutlin-3 for the indicated time points. Whole cell extracts were prepared and p53 was detected by Western Blotting using the DO1 antibody. Actin was detected as a loading control.

Figure 3. Protection of normal human dermal fibroblasts from VX-680-induced polyploidy

(A) Pretreatment with low dose ActD decreases the proportion of 8N cells caused by VX-680 in NHDF cells. ActD was added 24 hours post-seeding. VX-680 was added 48 hours post-seeding. Cells were harvested 96 hours after seeding. (B) NHDFs were treated with 4 nM ActD for 72 hours, fixed and stained with Giemsa (top panel). In the bottom panel, ActD was removed from the medium and cells were allowed to proliferate for 6 days before Giemsa staining.

Figure 4. Protection of normal human dermal fibroblasts from VX-680-induced nuclear abnormalities

(A and B) NHDF cells were treated for 48 hours with VX-680 alone (labelled as no preinc.), or treated for 48 hours with VX-680 after preincubating with the indicated amounts of ActD for 24 hours. Thereafter cells were fixed, stained Giemsa (A) and counted (B). Arrows in (A) indicate cells with abnormal nuclei. (C and D) NHDF cells treated as in A and B were allowed to recover for 6 days in drug-free medium. The growth of the cultures in drug-free medium is quantified in panel (C) by counting the total number of cells per field before and after the recovery period. The average number of cells per field with aberrant nuclei before and after the recovery period is shown in panel (D). Error bars correspond to standard deviations.

Figure 5. Induction of p53 by ActD in normal fibroblasts

(A) NHDF cells were treated with the indicated amounts of ActD for 4 or 12 hours. p53 and p21 were analysed by Western blotting. α-tubulin was detected as a loading control. (B) Example of nuclear abnormalities in NHDF cells expressing a dominant negative form of p53. The percentage of cells with this sort of aberrant nuclei was 2.5 fold higher than in the corresponding cells with intact p53. Severely damaged cells, such as the one in the center of this picture, only appeared in the fibroblasts with dominant negative p53.

Figure 6. Effect of the LDActD/VX-680 combination on p53-mutant tumour cells

Cell-cycle distribution analyses of MDAMB231 (A) and MDAMB468 (B) cells treated with the indicated drugs as described in Figure 3A. The proportion of 2N, 4N and 8N cells not synthesising DNA (i.e. not incorporating BrdU) is given.

Figure 7. Effect of the LDActD/VX-680 combination on p53-null tumour cells

Cell-cycle distribution analysis of HCT116 p53-null cells treated with the indicated drugs as described in Figure 3A. The proportion of 2N, 4N and 8N cells not synthesising DNA (i.e. not incorporating BrdU) is given.

Figure 8. Effect of the Nutlin-3/VX-680 combination on p53-mutant tumour cells

MDAMB231 cells were treated sequentially with Nutlin-3 and VX-680 as described in Figure 6A for the ActD/VX-680 combination. Cell-cycle distribution analysis was carried out by FACS 72 hours after Nutlin-3 addition.

Figure 9. Recovery of p53 deficient cells from the LDActD/VX-680 combination

Actinomycin D does not improve the rate of recovery from VX-680 treatment in tumour cells with MDAMB231 cells with mutant p53 (top plates) nor in HCT116 p53-null cells (bottom plates). Cells were treated as described in Figure 3A, fixed and stained with Giemsa. In the top panels, drugs were removed from the medium and cells were allowed to proliferate for 8 days in fresh medium.

Figure 10. Effect of the LDActD/VX-680 combination on p53-wild type tumour cells

HCT116 cells (p53 wild-type) were treated as described in Figure 7 and cell cycle distribution analysis was carried out by FACS.