Abstract 1024

Poster Board I-46

The treatment of acute myeloid leukemia (AML) still represents a challenge. As the growth and survival of AML cells are enhanced by the deregulation of signaling pathways such as PI3K, ERK/MAPK or mTOR, major efforts are made to develop targeted molecules against these activated pathways. The mTOR serine/threonine kinase belongs to two separate complexes, mTORC1 and mTORC2. The mTORC1 complex, defined by the interaction between mTOR and raptor, is inhibited by rapamycin in most models. It controls mRNA translation through the phosphorylation of the physiological translation repressor 4E-BP1. A constitutive activation of mTORC1 is usually detected in primary AML cells, indicating this pathway as a major target for AML therapy. However, inhibition of mTORC1 with rapamycin has only modest effects in this disease (Recher, Beyne-Rauzy et al., Blood 2005; Xu, Thompson et al.,Blood 2005). Accordingly, we recently showed that the translation process is deregulated and rapamycin resistant in AML, although specifically targeting translation with the 4E-BP1 mimetic 4EGI-1 markedly impairs AML cells survival (Tamburini, Green et al., Blood 2009). The mTORC2 complex is defined by the interaction between rictor and mTOR and is generally reported as resistant to rapamycin. An emerging role for mTORC2 has recently been rose in cancer biology, including in AML, in part through its ability to activate the oncogenic kinase Akt by phosphorylation on S473 (Zeng, Sarbassov dos et al., Blood 2007). In AML, the PI3K and mTOR activities are closely connected: inhibition of mTORC1 activates an IGF-1/IGF-1R autocrine loop that overactivates PI3K but mTORC1 activity is PI3K-independent (Tamburini, Chapuis et al., Blood 2008). These results strongly suggest a rational for simultaneous inhibition of PI3K and mTOR to enhance the killing of AML cells (Park, Chapuis et al., Leukemia 2008; Tamburini, Chapuis et al., Blood 2008). The NVP-BEZ235 compound (Novartis®) is an ATP-competitive inhibitor specific of PI3K and mTOR kinases, actually in phase I/II trial in oncology. We tested for the first time NVP-BEZ235 in 3 human AML cell lines (MV4-11, MOLM-14 and OCI-AML3) and in primary AML samples (n=10). This compound inhibited PI3K and mTOR signaling without off-target effects against other kinases usually activated in AML cells (ERK1/2 and STAT5). In contrast to rapamycin, NVP-BEZ235 dose-dependently inhibited the catalytic activity of mTOR as attested by inhibition of the mTOR autophosphorylation site on S2481. Consequently, NVP-BEZ235 fully inhibited the rapamycin-resistant phosphorylation of the translation repressor 4E-BP1 on T37/46, T70 and S65 residues. This resulted in a marked inhibition of mRNA translation in AML cells, attested by (i) an inhibition of the assembly of the translation initiating complex eIF4F, (ii) a shift from large to small polysomes in NVP-BEZ235-treated AML cells and (iii) a decreased expression of the oncogenic cap-dependant proteins Bcl-XL, c-myc and cyclinD1. Moreover, to study the effects of NVP-BEZ235 on mTORC2 activity independently of Akt, we investigated the regulation of the mTORC2 substrate paxillin (Jacinto, Loewith et al., Nat Cell biol 2004). We first confirmed that paxillin Y118 residue was a substrate for mTORC2 activity in primary AML cells as the siRNA-mediated rictor knockdown dramatically inhibited this phosphorylation event. In primary AML samples, NVP-BEZ235 inhibited paxillin Y118 phosphorylation in a dose-dependent way and this closely correlated to the inhibition of Akt S473 phosphorylation. Targeting the catalytic activity of mTOR represses therefore both mTORC1 and mTORC2 activity. Accordingly, the NVP-BEZ235 compound had marked anti-leukemic activity in AML. Indeed, it strongly reduced the proliferation of AML cells and blocked the cell-cycle as attested by a decrease of AML cells in S-phase and an accumulation in G1-phase. Finally, NVP-BEZ235 induced an important apoptotic response in AML blast cells but its effects on normal hematopoiesis were limited ex vivo. Overall, NVP-BEZ235 strongly inhibited protein synthesis and induced the killing of AML cells through the catalytic inhibition of the oncogenic PI3K and mTOR kinases. Given these results, NVP-BEZ235 may be tested urgently in clinical trials in AML.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution