Identifying Small Molecules That Overcome HoxA9-Mediated Differentiation Arrest in Acute Myeloid Leukemia

Abstract 3513

AML in adults is a devastating disease with a 5-year survival rate of 25%. We lack new treatments for AML, and the chemotherapy standard of care remains unchanged in thirty years. One success story in the treatment of AML has been the discovery of drugs that trigger the differentiation of leukemic blasts in the small subset of patients with acute promyelocytic leukemia. However, differentiation therapy is unfortunately not available for the remaining 90% of non-APL acute myeloid leukemia patients.

Understanding and targeting the mechanism of differentiation arrest in AML has been under investigation for more than four decades. There is growing evidence to support the role of the homeobox transcription factors in normal hematopoietic differentiation as well as malignant hematopoiesis. The persistent, and inappropriate, expression of the homeobox gene HoxA9 has been described in the majority of acute myeloid leukemias. This implicates HoxA9 dysregulation as a common pathway of differentiation arrest in myeloid leukemias and suggests that by understanding and targeting this pathway, one might be able to overcome differentiation arrest.

In cultures of primary murine bone marrow, constitutive expression of HoxA9 blocks myeloid differentiation and results in the outgrowth of immature myeloid cell lines. The mechanism by which HoxA9 causes differentiation arrest is not known and no compounds exist that inhibit HoxA9. We developed a murine cell line model in which the cells were blocked in differentiation by a conditional version of HoxA9. In this system, an estrogen-dependent ER-HoxA9 protein was generated by fusion with the estrogen receptor hormone-binding domain. When expressed in cultures of primary murine bone marrow, immortalized myeloblast cell lines can grow indefinitely in the presence of stem cell factor and beta-estradiol. Upon removal of beta-estradiol, and inactivation of HoxA9, these cell lines undergo synchronous and terminal myeloid differentiation.

We took advantage of an available transgenic mouse model in which GFP was expressed downstream of the lysozyme promoter, a promoter expressed only in mature neutrophils and macrophages. Cell lines derived from the bone marrow of this lysozyme-GFP mouse were GFP-negative at baseline and brightly GFP-positive upon differentiation. In this manner, we generated a cell line with a built-in reporter of differentiation.

These cells formed the basis of a high-throughput screen in which cells were incubated with small molecules for a period of four days in 384-well plate format. The cells were assayed by multi-parameter flow cytometry to assess for toxicity and differentiation. Compounds that triggered green fluorescence were scored as “HITS” and their pro-differentiation effects confirmed by analysis of morphology and cell surface markers. Given the availability of cells and the simple and reliable assay, we performed both a pilot screen of small molecules at The Broad Institute as well as an extensive screen of the NIH Molecular Libraries Small Molecule Repository. The screen of more than 350,000 small molecules was carried out in collaboration with the University of New Mexico Center for Molecular Discovery.

We have identified one lead class of compounds - prostacyclin agonists – capable of promoting myeloid differentiation in this cell line model of AML. Using a parallel cell line derived from a prostacyclin receptor knock-out mouse, we confirmed that activity was due to signaling through the prostacyclin receptor. The role of prostacyclin signaling in myeloid differentiation has not been previously described. Analysis of gene expression demonstrated that the expression of the prostacyclin receptor is seen in ∼60% of in primary human AML samples. This is a potentially exciting finding as prostacyclin agonists (e.g. treprostinil) are clinically relevant as well as FDA-approved. Their potential role in the treatment of acute myeloid leukemia is unknown.

Here we present the details of our high-throughput flow cytometry system and preliminary identification of pro-differentiation agents in AML. If successful, we anticipate that one of these small molecules may offer insight into a mechanism for overcoming differentiation arrest, and may also translate into a novel, clinically relevant treatment for acute myeloid leukemia.