Molecular targeting of the oncogene eIF4E in acute myeloid leukemia (AML): a proof-of-principle clinical trial with ribavirin.

The eukaryotic translation initiation factor eIF4E is elevated in 30% of malignancies including M4/M5 subtypes of acute myeloid leukemia (AML). The oncogenic potential of eIF4E arises from its ability to bind the 7-methyl guanosine (m(7)G) cap on mRNAs, thereby selectively enhancing eIF4E-dependent nuclear mRNA export and translation. We tested the clinical efficacy of targeting eIF4E in M4/M5 AML patients with a physical mimic of the m(7)G cap, ribavirin. Among 11 evaluable patients there were 1 complete remission (CR), 2 partial remissions (PRs), 2 blast responses (BRs), 4 stable diseases (SDs), and 2 progressive diseases (PDs). Ribavirin-induced relocalization of nuclear eIF4E to the cytoplasm and reduction of eIF4E levels were associated with clinical response. Lack of response or relapse coincided with continued or renewed nuclear localization of eIF4E. This first clinical study to target eIF4E in human malignancy demonstrates clinical activity and associated molecular responses in leukemic blasts. This trial is registered at ClinicalTrials.gov (NCT00559091).

Molecular dissection of Meis1 reveals 2 domains required for leukemia induction and a key role for Hoxa gene activation.

The Hoxa9 and Meis1 genes represent important oncogenic collaborators activated in a significant proportion of human leukemias with genetic alterations in the MLL gene. In this study, we show that the transforming property of Meis1 is modulated by 3 conserved domains, namely the Pbx interaction motif (PIM), the homeodomain, and the C-terminal region recently described to possess transactivating properties. Meis1 and Pbx1 interaction domain-swapping mutants are dysfunctional separately, but restore the full oncogenic activity of Meis1 when cotransduced in primary cells engineered to overexpress Hoxa9, thus implying a modular nature for PIM in Meis1-accelerated transformation. Moreover, we show that the transactivating domain of VP16 can restore, and even enhance, the oncogenic potential of the Meis1 mutant lacking the C-terminal 49 amino acids. In contrast to Meis1, the fusion VP16-Meis1 is spontaneously oncogenic, and all leukemias harbor genetic activation of endogenous Hoxa9 and/or Hoxa7, suggesting that Hoxa gene activation represents a key event required for the oncogenic activity of VP16-Meis1.

Molecular interactions involved in HOXB4-induced activation of HSC self-renewal.

HOXB4 overexpression induces unique in vivo and in vitro expansion of hemopoietic stem cells (HSCs) without causing leukemia. Very little is known about the molecular basis underlying HOXB4-induced HSC self-renewal. We now report the in vitro proliferation and in vivo expansion capacity of primary bone marrow (BM) cells engineered to overexpress selected HOXB4 point mutants lacking either the capacity to directly bind DNA (HOXB4(A)), or to cooperate with members of the PBX family (HOXB4(W-->G)) in DNA binding. The DNA binding-incompetent HOXB4 mutant failed to enhance the proliferation activity of transduced BM populations in vitro and HSC expansion in vivo. In contrast, the HOXB4(W-->G) mutant conferred a pronounced in vitro proliferation advantage to the transduced BM populations, and dramatically enhanced their in vivo regenerative potential. We also demonstrate a correlation between HOXB4 protein levels and in vitro proliferative capacity of primary BM cells. Our observations thus suggest that the capacity of HOXB4 to induce HSC expansions is DNA-binding dependent and does not require direct HOX/PBX interaction, and sets the stage for identifying HOXB4-dependent targets involved in HSC expansion.

In vitro expansion of hematopoietic stem cells by recombinant TAT-HOXB4 protein.

Hematopoietic stem cells (HSCs) can self-renew extensively after transplantation. The conditions supporting their in vitro expansion are still being defined. Retroviral overexpression of the human homeobox B4 (HOXB4) gene in mouse bone marrow cells enables over 40-fold expansion of HSCs in vitro. To circumvent the requirement for retroviral infection, we used recombinant human TAT-HOXB4 protein carrying the protein transduction domain of the HIV transactivating protein (TAT) as a potential growth factor for stem cells. HSCs exposed to TAT-HOXB4 for 4 d expanded by about four- to sixfold and were 8-20 times more numerous than HSCs in control cultures, indicating that HSC expansion induced by TAT-HOXB4 was comparable to that induced by the human HOXB4 retrovirus during a similar period of observation. Our results also show that TAT-HOXB4-expanded HSC populations retain their normal in vivo potential for differentiation and long-term repopulation. It is thus feasible to exploit recombinant HOXB4 protein for rapid and significant ex vivo expansion of normal HSCs.

Signal transduction by several KIT juxtamembrane domain mutations.

Mutations of KIT receptor tyrosine kinase are found in the majority of patients with mastocytosis and in most gastrointestinal stromal tumors. Oncogenic KIT mutations in GISTs are located in the KIT juxtamembrane domain (JMD), while codon 816 in the KIT kinase domain is mutated in systemic mastocytosis. We describe and characterize a mutation in the KIT-JMD named Kdelta27. We show that Kdelta27 mutant is constitutively dimerized and phosphorylated. Kdelta27 ectopic expression renders both the Ba/F3 cell line and primary cultures of bone marrow mast cells independent of cytokines for proliferation and cell survival. The classical signaling pathways activated by wild-type KIT upon ligand stimulation are constitutively activated by Kdelta27 and other JMD mutations. However, a side-to-side comparison revealed differences between the wild-type and JMD mutations. First, in vitro kinase assays reveal a change in peptide substrate specificity. Second, STAT proteins are preferentially phosphorylated by KIT mutants. Third, inhibitors of KIT kinase are more efficient on JMD mutations than on WT KIT. We conclude that Kdelta27 is a new oncogenic KIT mutation showing constitutive activation of downstream signaling pathways, and suggest that specific pathways are activated by oncogenic KIT.

The competitive nature of HOXB4-transduced HSC is limited by PBX1: the generation of ultra-competitive stem cells retaining full differentiation potential.

We previously showed that HOXB4 is a potent stimulator of hematopoietic stem cell (HSC) proliferation in vivo and ex vivo. As a result, HOXB4 overexpressing HSCs are 20- to 50-times more competitive than untransduced cells when transplanted into mice. By knocking down the expression of PBX1 (PBX1(K.D.)) in HOXB4 overexpressing cells, we now present the possibility of generating HSCs that are >20-times more competitive than those that overexpress HOXB4. The differentiation activity of these cells appears intact, since they competitively contributed to the reconstitution of normal myeloid and lymphoid compartments in vivo. We also show that the in vivo expansion of HOXB4-PBX1(K.D.)-expressing HSCs regenerated normal stem cell pools and did not lead to HSC levels above those detected in unmanipulated mice. The vigorous competitive nature of these cells in vivo compared to HOXB4-transduced HSCs suggests the existence of a distinct, non-cell autonomous mechanism that limits the expansion of HOXB4-transduced hemopoietic stem cells in mice.