Early changes in rpS6 phosphorylation and BH3 profiling predict response to chemotherapy in AML cells.

Blasts from different patients with acute myeloid leukemia (AML) vary in the agent(s) to which they are most responsive. With a myriad of novel agents to evaluate, there is a lack of predictive biomarkers to precisely assign targeted therapies to individual patients. Primary AML cells often survive poorly in vitro, thus confounding conventional cytotoxicity assays. The purpose of this work was to assess the potential of two same-day functional predictive assays in AML cell lines to predict long-term response to chemotherapy. (i) Ribosomal protein S6 (rpS6) is a downstream substrate of PI3K/akt/mTOR/ kinase and MAPK kinase pathways and its dephosphorylation is also triggered by DNA double strand breaks. Phospho-rpS6 is reliably measurable by flow cytometry and thus has the potential to function as a biomarker of responsiveness to several therapeutic agents. (ii) A cell's propensity for apoptosis can be interrogated via a functional assay termed "Dynamic BH3 Profiling" in which mitochondrial outer membrane permeabilization in drug-treated cells can be driven by pro-apoptotic BH3 domain peptides such as PUMA-BH3. The extent to which a particular cell is primed for apoptosis by the drug can be determined by measuring the amount of cytochrome C released on addition of BH3 peptide. We demonstrate that phospho-rpS6 expression and PUMA-BH3 peptide-induced cytochrome C release after 4 hours both predict long term chemoresponsiveness to tyrosine kinase inhibitors and DNA double strand break inducers in AML cell lines. We also describe changes in expression levels of the prosurvival BCL-2 family member Mcl-1 and the pro-apoptotic protein BIM after short term drug culture.

Predicting effective pro-apoptotic anti-leukaemic drug combinations using co-operative dynamic BH3 profiling.

The BH3-only apoptosis agonists BAD and NOXA target BCL-2 and MCL-1 respectively and co-operate to induce apoptosis. On this basis, therapeutic drugs targeting BCL-2 and MCL-1 might have enhanced activity if used in combination. We identified anti-leukaemic drugs sensitising to BCL-2 antagonism and drugs sensitising to MCL-1 antagonism using the technique of dynamic BH3 profiling, whereby cells were primed with drugs to discover whether this would elicit mitochondrial outer membrane permeabilisation in response to BCL-2-targeting BAD-BH3 peptide or MCL-1-targeting MS1-BH3 peptide. We found that a broad range of anti-leukaemic agents-notably MCL-1 inhibitors, DNA damaging agents and FLT3 inhibitors-sensitise leukaemia cells to BAD-BH3. We further analysed the BCL-2 inhibitors ABT-199 and JQ1, the MCL-1 inhibitors pladienolide B and torin1, the FLT3 inhibitor AC220 and the DNA double-strand break inducer etoposide to correlate priming responses with co-operative induction of apoptosis. ABT-199 in combination with pladienolide B, torin1, etoposide or AC220 strongly induced apoptosis within 4 hours, but the MCL-1 inhibitors did not co-operate with etoposide or AC220. In keeping with the long half-life of BCL-2, the BET domain inhibitor JQ1 was found to downregulate BCL-2 and to prime cells to respond to MS1-BH3 at 48, but not at 4 hours: prolonged priming with JQ1 was then shown to induce rapid cytochrome C release when pladienolide B, torin1, etoposide or AC220 were added. In conclusion, dynamic BH3 profiling is a useful mechanism-based tool for understanding and predicting co-operative lethality between drugs sensitising to BCL-2 antagonism and drugs sensitising to MCL-1 antagonism. A plethora of agents sensitised cells to BAD-BH3-mediated mitochondrial outer membrane permeabilisation in the dynamic BH3 profiling assay and this was associated with effective co-operation with the BCL-2 inhibitory compounds ABT-199 or JQ1.