Targeting the MLL complex in acute leukemia

Chromosomal rearrangements leading mostly to fusion oncoproteins of the Mixed Lineage Leukemia (MLL) gene occur in about 10% of all patients with acute leukemia and are often associated with poor clinical outcome, emphasizing the need for new treatment modalities. The MLL protein forms a ternary complex with the lens epithelium-derived growth factor (LEDGF/p75 also known as PSIP1) and another protein MENIN. Previous work has shown that LEDGF/p75 contributes to the association of the MLL multi-protein complex to chromatin. In addition, LEDGF/p75 is known for acting as a tether of the human immunodeficiency virus 1 (HIV-1) pre-integration complex to chromatin, and previous works has demonstrated that expression of the C-terminal fragment fused to eGFP (eGFP-LEDGF/p75325-530) impaired HIV-1 replication. Here, we explored this strategy to selectively interfere with the leukemogenic activity of MLL-fusion proteins. We found that expression of the LEDGF/p75325-530 fragment impaired the clonogenic growth of MLL-fusion gene transformed human and mouse leukemic cell lines, without affecting the growth of immortalized control cells, or normal lineage marker-depleted murine bone marrow cells. Expression of LEDGF/p75325-530 was associated with downregulation of the known MLL target Hoxa9 and associated with impaired cell cycle progression. Structure-function analysis revealed two small eGFP fused LEDGF/p75 peptide-sized fragments, LEDGF/p75424-435 and LEDGF/p75375-386 were able to phenocopy these effects. LEDGF/p75325-530 and the smaller active peptides were all able to disrupt the LEDGF/p75-MLL interaction. Expression of LEDGF/p75325-530 or the LEDGF/p75375-386 fragment increased the latency period to disease development in vivo in a mouse bone marrow transplant model of MLL-AF9 induced acute myeloid leukemia (AML). From these studies we concluded that small peptides disrupting the LEDGF/p75-MLL interface have selective anti-leukemic activity providing a direct rationale for the design of small molecule inhibitors targeting this interaction.
Intensive biochemical and structural analysis led by our collaborators (J. De Rijck, K. Cermakova, KU LEUVEN, Belgium) further allowed the identification of two MLL-LEDGF/p75 targetable interfaces. Indeed a recently resolved partial structure revealed a potentially drugable hydrophobic pocked stabilizing the MLL-MENIN-LEDGF/p75 interface. Interestingly, our IBD-derived LEDGF/p75424-435 fragment was targeting this interface that seems to be dependent on the interaction with MENIN and has been
successfully targeted by MENIN small molecule inhibitors. As the available X-ray data represented only a partial structure of the LEDGF/p75-MLL-MENIN complex, our collaborators used NMR spectroscopy to identify an additional MLL-LEDGF/p75 interface, which partially overlaps with the binding site of known LEDGF/p75 interactors including the HIV-1 integrase. They proved that binding of the HIV-1 integrase or MLL to LEDGF/p75 is mutually exclusive and seems to be dependent of MENIN. Importantly, the newly defined interface was directly targeted by expression of the LEDGF/p75375-386 IBD-derived fragment we previously defined. We were then able to show that the clonogenic growth of primary murine MLL-AF9 expressing leukemic blasts was selectively impaired upon overexpression of a LEDGF/p75 binding cyclic peptide CP65, known to bind the IBD and disrupt its interaction with HIV-1 integrase. Thus collectively our data shows that this newly defined protein-protein interface represents a new target for the development of therapeutics against HIV-1 replication as well as LEDGF/p75-dependent MLL fusion oncoprotein driven leukemic disorders.
Intensive research efforts led by many groups resulted in the definition of multiple possibilities for potential interference with the leukemogenic MLL fusion protein complex. We therefore also started to explore whether targeting of the MLL complex at different nodes could result in synergistic effects to efficiently impair MLL-fusion mediated leukemia. Previous studies have shown that the histone H3 lysine 79 (H3K79) methyltransferase DOT1L is essential for MLL-fusion driven leukemogenesis. Selective small molecule DOT1L inhibitors have been generated and are currently entering first clinical trials. However, when given in monotherapy, work in mouse models has shown limited and slow responses to these compounds in many MLL-rearranged leukemia models. Collaborators from the Novartis Institute for Biomedical Research (NIBR, Basel) have performed shRNA screens in MLL-rearranged cell lines to identify sensitizing targets for DOT1L inhibitors and found that knockdown of several MLL complex components, including LEDGF/p75 significantly enhanced anti-leukemic responses. In absence of any available pharmacological agents targeting LEDGF/p75, they decided to test syngertistic activity of the DOT1L inhibitor EPZ004777 in combination with the MENIN inhibitor MI-2-2 previously shown to induce growth arrest and differentiation in MLL-rearranged leukemia cells. We therefore tested the effects of combination of MI-2-2 and EPZ004777 in our MLL-AF9 mouse AML model and found that transient and non-lethal exposure to the combination of these compounds was sufficient to permanently disable their leukemogenic infiltrating
potential in vivo mainly through induction of a rapid and effective differentiation of MLL-AF9 expressing leukemic blasts. Altogether, these results suggested that the EPZ004777/MI-2-2 combination might deliver synergistic and durable anti-leukemia effects in MLL-rearranged AML.
All together this work suggests that the MENIN-MLL-LEDGF/p75 complex offers several interfaces for selective therapeutic targeting for MLL-rerranged acute leukemia. In addition, our results suggest that co-targeting the MLL complex at different nodes could pave the way for selective and efficient novel therapeutic approaches for leukemic disorders mediated by MLL fusion oncoproteins. Such approaches need to be clinically tested in particular whether they will be able to overcome early relapse of the disease that is the “Achilles heel” of current applied polychemotherapeutic strategies.