Abstract
Context: Novel therapeutic strategies targeting cell cycle regulation are attractive for multiple myeloma (MM) because of the increased proliferative index in advanced drug-resistant disease. Having previously studied the role of Cdk inhibition in MM, we have looked at the cell cycle-related polo kinases (PLKs), because their expression is associated with adverse prognosis in solid tumors. We report preclinical studies on the anti-MM activity of the PLK1/2/3 small molecule inhibitor BI2536.
Methods/Results: We tested 39 human tumor cell lines by MTT colormetric assay, including MM (n=26), T-ALL (n=8), solid tumors (n=5), as well normal human tissues. BI2536 exhibited activity in the low nano-molar range with IC50 values <10 nM for the most sensitive cells lines, which included several MM lines. BI2536 exhibited minimal activity against normal PBMCs, unstimulated or PHA-stimulated, with IC50 values greater than the highest dose tested (i.e. 80 nM). Myeloma cell lines were further tested in the presence and absence of exogenous IL-6 (10ng/mL) and IGF-1 (50ng/mL) and exhibited the ability to overcome the cytokine-induced resistance observed with other anti-MM agents (e.g. Dex or Doxo). Interestingly, several stromal responsive myeloma cell lines, including MM.1S, MM.1R, H929 and INA-6 were more sensitive to BI2536 in the presence of HS-5 stromal cells compared to the stromal unresponsive cell line OPM2, which was equally sensitive in the presence and absence of stromal cells. In addition, myeloma cell lines co-cultured with osteoclasts (OC) exhibited comparable activity in the presence and absence of OCs. Cell cycle analysis showed that treatment with BI2536 causes rapid G2/M arrest and increased G0/G1 phase events in KMS18 cells. Mechanistic studies revealed that Akt, MAPK, cyclin B1, cyclin D1 and cdk1 levels decrease in response to BI2536 treatment, while caspase-3 and PARP are cleaved within 8 hrs of drug treatment at 20 nM. Interestingly, Notch and phospho-histone H3 levels increased in response to treatment. Gene expression profiling analysis further validated the finding that BI2536 functions distinctly from other anti-MM agents, since there was not an effect on transcriptional signatures of proteasome, NF-kB or IRF4 activity following BI2536 treatment in KMS18 cells. In addition, higher doses of BI2536 preferentially killed side-population cells (SP cells) compared to the main population (MP), as shown by Hoechst staining. Importantly, immunohistochemisty revealed that MM.1S cells treated with BI2536 were unable to recruit alpha-tubulin to mitotic centrosomes and form bipolar spindles, which is compatible with the role of polo kinases in mitotic spindle formation. We also evaluated a series of combinations of this agent with conventional (e.g. dexamethasone, doxorubicin) and novel (e.g. bortezomib) anti-MM agents. No evidence of antagonism with any of these anti-MM agents was observed, indicating that combinations of BI2536 may be feasible in clinical settings with current anti-myeloma regimens.
Conclusion: Proteins pivotal for cell cycle progression represent promising targets for treating highly proliferating tumors. Treatment of MM with a PLK inhibitor provides evidence that polo kinases are promising targets for MM therapy. Importantly, BI2536 activity was enhanced in the presence of stromal cells, providing evidence that this class of compounds will be active in the tumor microenvironment.
A proteolytic fragment of Mcl-1 exhibits nuclear localization and regulates cell growth by interaction with Cdk1