Identification of CHD4 As a Potential Therapeutic Target of Acute Myeloid Leukemia

Acute myeloid leukemia (AML) is characterized by impaired myeloid differentiation of hematopoietic progenitors, causing uncontrolled proliferation and accumulation of immature myeloid cells in the bone marrow. Rearrangements of the mixed lineage leukemia (MLL) gene are common aberrations in acute leukemia and occur in over 70% in childhood leukemia and 5-10% in leukemia of adults. MLL rearrangements encode a fusion oncogenic H3K4 methytransferase protein, which is sufficient to transform hematopoietic cells and give rise to an aggressive subtype of AML. Leukemia where the MLL fusion oncogene is expressed is characterized by dismal prognosis and 30-60% of 5-years overall survival rate.

The current standard treatment for AML is chemotherapy and in certain cases bone marrow transplantation. However, chemotherapy causes severe side effects on normal cells and an increased risk of relapse. Consequently, discovery of novel drug targets with better efficacy and low toxicity are needed to improve treatment of AML.

In this study, we aimed to identify genes that are required for growth of AML cells and that encode proteins that potentially could be used as therapeutic targets. To do this, we performed high-throughput RNAi screening covering all annotated human genes and the homologous genes in mice, using barcoded lentiviral-based shRNA vectors. Stable loss-of-function screening was done in three AML cell lines (two human and one murine AML cell lines) as well as in a non-transformed hematopoietic control cell line. The candidate genes were selected based on that shRNA-mediated knockdown caused at least a 5-fold growth inhibition of leukemic cells and that the individual candidates were targeted by multiple shRNAs.

The chromodomain Helicase DNA binding protein 4 (CHD4), a chromatin remodeler ATPase, displayed the most significant effect in reduced AML cell proliferation upon inhibition among the overlapping candidate genes in all three AML cell lines. CHD4 is a main subunit of the Nucleosome Remodeling Deacetylase (NuRD) complex and has been associated with epigenetic transcriptional repression. A recent study has shown that inhibition of CHD4 sensitized AML cells to genotoxic drugs by chromatin relaxation, which increases rate of double-stranded break (DSB) in leukemic cells.

To verify whether CHD4 is exclusively essential for AML with MLL rearrangements, we inhibited CHD4 expression with two independent shRNAs in various AML cell lines with and without MLL translocations. In vitro monitoring of growth and viability indicated that knockdown of CHD4 efficiently suppressed growth in all tested cell lines, suggesting that CHD4 is required in general for growth of leukemic cells. To test the effect of CHD4 inhibition in normal hematopoiesis, we pursued knockdown of CHD4 and monitored effects in hematopoiesis using colony formation assays of human CD34⁺ cells. The results demonstrated that CHD4 knockdown had minor effects in colony formation as well as growth and survival of normal hematopoietic cells.

Furthermore, to explore whether inhibition of CHD4 can prevent AML tumor growth and disease progression in vivo, we have generated a mouse model for AML. By transplanting AML cells transduced with shRNA against CHD4 into recipient mice, we showed that shRNA-mediated targeting of CHD4 not only significantly prolonged survival of AML transplanted mice but also in some cases completely rescued some mice from development of the disease. Collectively, these data suggested that CHD4 is required for AML maintenance in vivo.

Next, to determine whether suppression of CHD4 can inhibit cell growth of different subpopulations and subtypes of AML, we performed loss of function studies of CHD4 on patient-derived AML cells ex vivo. Loss of CHD4 expression significantly decreased the frequency of leukemic initiating cells in different subtypes AML patient samples. In further in vivo studies using a xeno-tranplantation model for AML, we demonstrated that shRNA-mediated inhibition of CHD4 significantly reduced the frequency of leukemic cells in the marrow 6 weeks after transplantation.

Taken together our results demonstrated the critical and selective role of CHD4 in propagation of patient-derived AML cells as well as in disease progression in mouse models for AML. We believe that CHD4 represents a novel potential therapeutic target that can be used to battle AML.