The Activity of the Novel Aurora Kinase B Inhibitor AZD1152 in Acute Myeloid Leukaemia Cells.

Background: Aurora kinases (A, B and C) play a critical role in regulating mitosis and cell division. Aurora kinase A is required for correct spindle assembly while aurora kinase B is involved in histone H3 phosphorylation, chromosome segregation, and cytokinesis. Emerging data suggest these proteins are implicated in the survival and proliferation of both haematologic and solid malignancies, and therefore represent novel therapeutic targets. Several aurora kinase inhibitors have been described, mainly inhibiting both A and B kinases, although inhibition of aurora kinase B is more closely associated with the effects seen with these compounds. We have therefore investigated the activity of AZD1152, a novel, specific inhibitor of aurora kinase activity, with selectivity for aurora B.

Methods: Drug activity was studied in a panel of leukaemic cell lines (HL-60, MV411, THP-1, U937) and in primary AML cells (n=4) at concentrations of 0–1000 nM AZD1152 for up to 168 h. Cell number and percent viability were determined using a Viacount assay on a Guava PCA-96 analyser, cell cycle distribution (including apoptotic and polyploid (>4n) cells) by PI staining with flow cytometry, and the phosphorylation of an aurora kinase B substrate histone H3 by fluorescence immunocytochemistry using a phospho-specific antibody.

Results: AZD1152 displayed time- and concentration-dependent activity in cell lines and primary cells. A clear decrease in phospho histone H3 positive cells was apparent by 48 h, with <1% positive cells at 50 nM AZD1152 in cell lines compared with 8–10% in control cells and <0.5% positive cells in primary cultures at 100 nM AZD1152 compared with 2 - 4.5% in control cultures. In all cell lines studied a cell population with >4n DNA content by flow cytometry was apparent at 48 h (18–60% cells), confirmed by karyotype analysis which showed a variable but marked increase in chromosome number. This population had typically reduced, with a corresponding increase in apoptotic cells, by 96 h. In THP-1 cells, in which little effect on cell viability was observed, this >4n population increased to 80% by 96 h and persisted out to 168 h, with little apoptosis. In primary cells an increase in the G2/M population was evident in some cultures at 48 h, with only small changes in cells with >4n DNA content. In cell lines 10 nM AZD1152 resulted in a 28–65% reduction in cell number and a 6–35% reduction in cell viability after 48 h. At 100 nM these values increased to 65–78% and 6–42%, respectively. Notably little further increase in activity was observed at 1000 nM. By 96 h 100 nM AZD1152 reduced cell number by >80% in all cell lines, and cell viability by >67% in 3 lines (17% reduction in THP-1 cells). In primary AML cells AZD1152 over 48 h had little effect on cell number, but by 120 h this had decreased by >50% at 100 nM AZD1152, with a small effect on cell viability. Activity with 48 h exposure to drug followed by 72 h drug free was similar to that seen with a continuous 120 h exposure.

Conclusions: The specific aurora kinase inhibitor AZD1152 demonstrates antiproliferative and apoptotic activity in leukaemic cell lines and primary cells at concentrations associated with changes in the phosphorylation of an aurora kinase B substrate protein.