In spite of their different mechanisms of action, the ability of both LDActD and Nutlin-3 [11] to decrease cell growth is highly dependent on wild-type p53. This was clearly shown in a recent publication [17] and corroborated here for LDActD with a colony-growth assay comparing HCT116 colon carcinoma cells expressing wild-type p53 with HCT116 cells lacking p53 (Figure 1).

According to the existing literature, there are no clear differences between the effects of Nutlin-3 and LDActD on the p53 pathway. Phosphorylation of p53 on serine 15 is not substantially increased in response to exposure to either, whereas phosphorylation of p53 on serine 46 is increased to a similar extent by both agents [17]. Phosphorylation of p53 at serine 392 was initially observed after LDActD treatment, but not following Nutlin-3 treatment [17]. A more detailed analysis showed that phosphorylation at this site is also detectable in response to Nutlin-3 [25]. However, because of the nature of their p53-activation mechanisms, some differences in the speed at which p53 is induced by Nutlin-3 and ActD could be expected. As shown in Figure 2, Nutlin-3 treatment quickly increases p53 levels, its performance being much faster than ActD's in the same experiment. According to current understanding, ActD's effect on mdm2 requires that nucleoli are disrupted and that there are sufficient molecules of ribosomal proteins to inhibit mdm2. This is perhaps a much longer process than directly inhibiting mdm2 in order to stabilise p53. Whether this difference in the speed at which the p53 protein is increased is also partly due to differences in the permeability and/or stability of the compounds cannot yet be ruled out. Alternatively, Nutlin-3 and ActD may increase p53 in certain cell subpopulations only, i.e. in cells at particular stages of the cell cycle. These considerations may have repercussion for the use of ActD in a cyclotherapy setting.

It has been recently shown that pretreatment with Nutin-3 protects normal keratinocytes from polyploidy caused by the small molecule aurora kinase inhibitor VX-680 (or MK-0457) [11]. Building upon the experience with the Nutlin-3/VX-680 combination, we carried out similar experiments using normal human dermal fibroblasts (NHDFs) pretreated with LDActD, instead of Nutlin-3, and subsequently exposed to VX-680. VX-680 is a potent aurora kinase inhibitor tested for the treatment of solid tumours and leukemia. Clinical trials were stopped due to electrocardiograph QTc interval prolongation in one patient. Nevertheless, because cyclotherapy studies using Nutlin-3 have been performed in combination with VX-680, we proceeded with the LDActD/VX-680 combination as this facilitates comparisons between Nutlin-3 and LDActD as potential cytostatic agents in p53-based cyclotherapy.

Activity of aurora kinases peaks during the G2 and M phases of the cell cycle and, accordingly, VX-680's killing effect is mainly by induction of endoreduplication in the absence of mitosis resulting in polyploidy and, eventually, cell death by mitotic catastrophe [26]. Normal dividing cells are susceptible to this effect and indeed proliferating NHDFs respond to VX-680 growth inhibition (Figure 1 supplemental data). Within 48 hours of treatment there was a significant increase in the number of polyploid 8N cells in cultures treated with 100-500 nM VX-680 (Figure 3A and Figure 2 supplemental data). At the higher doses of VX-680, the proportion of 8N cells diminished in association with an increase in the number of sub-diploid population. In order to ensure on-target effects of the compound we chose to work at a concentration of VX-680 of 100 nM.

At 1.2 and 1.6 nM, ActD had a weak cell-cycle inhibitory effect on NHDF cells that became more evident with a slightly higher (2 nM) concentration (Figure 3A). The number of dead cells with a sub-diploid DNA content was only marginally increased by these doses of ActD. Interestingly, NHDF cells are able to resume proliferation upon removal of ActD (Figure 3B).

Preincubation of the NHDF cells with 1.2, 1.6 and 2 nM ActD before VX-680 treatment markedly diminished the proportion of polyploid 8N cells (and to some extent the number of sub-diploid cells) in a dose-dependent manner, demonstrating that LDActD's cytostatic effect can protect normal cells from the effects of aurora kinase inhibition. A marked difference in the number of cells in G1 was also observed between VX-680 treated (8.3% of cells in G1) and ActD/VX-680 treated cells (up to 38.2% of cells in G1) (Figure 3A). This suggests that at least part of the protection of NHDFs by LDActD from VX-680 is due to inhibition of the G1/S transition by LDActD. The importance of this point is highlighted in the Discussion section.

We also quantified the number of NHDF cells with aberrant nuclear morphologies induced by VX-680 (Figure 4). In the cultures treated with VX-680 only, a large proportion of the cells have abnormal nuclei. However, this number was markedly lower in the cultures that had been preincubated with LDActD (Figures 4A and 4B). Moreover, NHDF cells recovered from the ActD/VX-680 combination and proliferated after removal of the compounds (Figure 4C). In these recovery experiments, the number of cells with aberrant nuclei in the cultures treated with VX-680 alone as well as with the LDActD/VX-680 combination did rise after the recovery period. However, the number of abnormal cells remained comparatively low in the cultures recovering from exposure to the LDActD/VX-680-combination regimen (Figure 4D).

Finally, Figure 5A shows that in NHDFs, LDActD clearly increased the levels of p53. As expected from the results in Figure 2, long-term exposure to low doses of ActD is required to increase p53 levels. The levels of the p53 downstream transcriptional target p21, a well-established cell-cycle arresting protein were also increased by LDActD. This suggests that the protective effect of LDActD in NHDF cells could be due to the activation of p53 by this compound. We also observed that cells with abnormal nuclei are more frequent in NHDF cells expressing a dominant negative form of p53 than in the corresponding control NHDFs (Figure 5B).

The next question in this study was whether preincubation with LDActD affected the outcome of VX-680 treatment in tumour cells with defects in the P53 gene. For this analysis we chose to work with two human breast cancer cell lines carrying p53 mutations frequent in cancers. These cell lines, MDAMB231 and MDAMB468 express transcriptionally inactive p53 with mutations R280K and R273H, respectively. In both cell lines, as in normal fibroblasts, VX-680 treatment alone led to an increase in 8N-polyploid cells (Figure 6). However, and unexpectedly, LDActD pre-treatment at the doses required to protect normal fibroblasts also reduced the population of VX-680-induced 8N cells in these two tumour cell types. The effect of LDActD in these tumour cells differs from the effect in normal fibroblasts in that most MDAMB231 and MDAMB468 cells accumulate in a tetraploid state, whereas co-treatment of NHDF cells also leads to a rise in the proportion of cells with a 2N DNA content. In fact, it can be seen that exposure to LDActD on its own led to a small accumulation of cells with 4N DNA content in both MDAMB231 and MDAMB468 cultures. It may be that this small effect of LDActD becomes more apparent in cultures treated with a compound that, like VX-680, impairs the G2/M transition. A similar response to LDActD combined with VX-680 was observed in HCT116 p53-null cells (Figure 7), strongly supporting that the effect seen in mutant-p53 tumour cell lines is not dependent on the presence of mutant p53. Unlike ActD, Nutlin-3 had no effect on the cell-cycle progression of p53 mutant tumour cells either on its own or in combination with VX-680 (Figure 8). This is in line with the high selectivity of Nutlin-3 as an inhibitor of the p53/mdm2 interaction.

However, despite the ability of LDActD to weaken VX-680-induced polyploidy in cells with defective p53, when cells are allowed to recover from treatment upon drug removal, the cell density in the LDActD/VX-680-treated and VX-680-treated cultures is similar after the recovery period (Figure 9).

In conclusion, we observed that, even at low doses, ActD has p53-independent effects, including a p53-independent G2/M arresting effect that becomes accentuated in the presence of an aurora kinase inhibitor. Therefore, we cannot exclude that LDActD could protect a proportion of tumour cells with defective p53 from the toxicity of this mitotic poison.

NHDF cells were bought from Promocell. NHDF cells expressing a dominant negative form of p53 were obtained by stable transfection with an expression vector for the murine DD-p53 dominant-negative mutant and neomycin resistance as described [28]. HCT116 and HCT116 p53-null cells were a kind gift from Bert Vogelstein (Baltimore). H1299 and MCF-7 cells were bought from ATCC. MDAMB231 and MDAMB468 cells were bought from CRUK. Actinomycin D was purchased from Sigma and stock solution was 2 mg/ml in Ethanol. VX-680 was a gift from Phil Cohen (Dundee) and stock solution was 10 mM in DMSO.

(±)-Nutlin-3, referred to as Nutlin-3 here, was purchased from Cayman Chemicals. It must be noted that this product comprises two enantiomers, active (-)-Nutlin and inactive (+)-Nutlin. Therefore, 5 μM Nutlin-3 corresponds to 2.5 μM of the active enantiomer. The stock solution was 50 mM in DMSO.

Fluorescence Activated Cell Sorting (FACS) was used to analyse the cell-cycle distribution. Cells were seeded in 6-well tissue culture coated plates at required density on Day 1 and compounds were added 24 hours post-seeding at required doses on Days 2 (cytotastic agent) and 3 (VX-680). After required treatment times, Bromodexoyuridine (BrdU), supplied by Sigma, was added to cells for 30 minutes at final concentration of 30 μM. Cells were subsequently harvested using trypsinisation, spun down and resuspended in 1 ml of PBS. Cells were then fixed in 3 ml of ice-cold ethanol and stored for at least 1 hour at 4°C. Cells were pelleted and resuspended in 2 ml of warm pepsin at 1 mg/ml (Sigma) in 30 mM HCl pH1.5, for 30 minutes at 37°C with gentle shaking. Cells were then pelleted by centrifugation and resuspended in 1 ml of 2M HCl for 20 minutes at RT. Cells were washed in excess PBS before washing with the antibody buffer (0.5% w/v BSA, 0.5% v/v Tween-20 in PBS) and then resuspended in 200 μl of mouse primary antibody to BrdU (Becton Dickinson) in antibody buffer for 1 hour at RT. After washing with PBS, the samples were resuspended in 200 μl of secondary anti-mouse IgG FITC conjugated antibody (Sigma) and incubated at RT for 30 minutes in the dark. The cells were then washed with PBS and collected by centrifugation and finally resuspended in 500 μl of PBS containing 25 μg/ml propidium iodide (Invitrogen) to stain DNA. Samples were analysed using a Becton Dickinson FACScan operated by the CELLQuest software.

Cells were seeded in 6-well plates at the required density on Day 1 and treated as indicated. Cells were fixed with methanol for 7 minutes after which methanol was removed and plates left to dry. Cells were stained with 5% Giemsa for at least 1 hour before being washed with distilled water to remove excess stain. Plates were left to dry and then the morphology of the cell nuclei was assessed using a Zeiss light microscope under the 10X or 32X objective.

Cells grown in 6-well plates were harvested by first washing them twice in 2 ml of ice-cold PBS before being lysed using 1 x Novex NuPage LDS sample buffer (LDS), supplied by Invitrogen. All samples were then scraped into separate 1.5 ml centrifuge tubes. Cell lysates were heated to 90°C for 5 minutes, sonicated twice for 10 seconds, and centrifuged at 13200 rpm for 5 minutes. Protein concentrations were then normalised after determining the protein concentration using the BCA assay kit (Perbio). Once levelled, DTT was added to a final concentration of 0.1 M.

Proteins were separated on NuPAGE TM 4-12% Bis-Tris gels in NuPAGE TM MOPS SDS running buffer (Invitrogen) at 130V for approximately 100 minutes. Proteins were transferred to Immobilon-P membranes (Millipore) for 90 minutes at 35 V with 1 x NuPAGE TM transfer buffer (Invitrogen). Membranes were incubated at RT for 45 minutes in blocking solution (PBS/ 0.1 % v/v Tween-20/ 5% w/v non-fat dried milk) with gentle shaking. Following blocking, membranes were incubated with the appropriate primary antibody (diluted in blocking solution) for 1 hour at RT or overnight at 4°C. Membranes were washed 4 times for 5 minutes at RT in PBS/0.1 % Tween-20, before incubating with the appropriate HRP-conjugated secondary antibody (Dako) diluted in the blocking solution for 45 minutes at RT. Membranes were washed a further 5 times for 10 minutes each in PBS/0.1 % v/v Tween-20, and then a further 2 washes for 5 minutes in PBS at RT before detection of chemiluminescent signals with ECL solutions according to manufacturer's instructions (Amersham Biosciences).