Acute myeloid leukaemia (AML) is the most common form of leukaemia in adults. Although of the order of 75-85% of patients will achieve complete remission after induction chemotherapy, long-term survival is still relatively low. Despite the progress in the rational design of drugs in disorders such as chronic myeloid leukaemia, AML lacks a single specific pathogenomic event to act as a drug target. Interferon regulatory factor 1 (IRF1) is a member of a family of related proteins that act as transcriptional activators or repressors. IRF1 and its functional antagonist IRF2 originally discovered as transcription factors regulating the interferon-beta (IFN-beta) gene, are involved in the regulation of normal haematopoiesis and leukaemogenesis. IRF1 appears to act as a tumour suppressor gene and IRF2 as an oncogene. IRF1 acts to repress IRF2 function through the repression of cyclin-dependent kinase (CDK) inhibitor p21WAF1 critical for cell growth control. It appears that the tumour suppression function of IRF1 is abolished by IRF2. This review focuses on the interaction between IRF1 and IRF2 in myeloid development and leukaemogenesis, particularly in relation to the Ras signalling pathway. IRF2 may be a viable and specific therapeutic target in human leukaemia.

Abstract. Interferon regulatory factor (IRF) 1 and its functional antagonist IRF2 were originally discovered as transcription factors that regulate the interferon-ß gene. Control of cell growth has led to the definition of IRF1 as a tumour suppressor gene and IRF2 as an oncogene. Clinically, approximately 70% of cases of acute myeloid leukaemia demonstrate dysregulated expression of IRF1 and/or IRF2. Our previous studies have shown that human leukaemic TF-1 cells exhibit abnormally high expression of both IRF1 and IRF2, the latter acting to abrogate IRF1 tumour suppression, making these cells ideal for analysis of down-regulation of IRF2 expression. A novel G418 screening protocol was developed and used for identifying effective siRNA that targets IRF2 (siIRF2). Using optimized siIRF2 in leukaemic TF-1 cells, IRF2 was downregulated by approximately 70% at both mRNA and protein levels. Phenotypically, this resulted in growth inhibition associated with G 2 /M arrest as well as induction of polyploidy, differentiation and apoptosis. In contrast to these results, siIRF2 targeting did not affect normal haematopoietic stem/progenitor cell growth. These results indicate the potential utility of IRF2 inhibition as a therapeutic approach to cancer.

Activating mutations of the Ras genes occur at high frequency in acute myeloid leukemia (AML). We have previously shown that expression of mutant N-ras(N-rasm) in murine hematopoietic stem cells is sufficient to induce a myeloid malignancy that resembles human AML(Mackenzie et al. Blood, 1999, 93, 2043–2056). In a ’humanised’ NOD/SCID mouse model N-rasm induced a pre-leukemic condition characterised by myeloid proliferation of human hematopoietic progenitor cells in the bone marrow of recipient mice (Shen et al. Exp. Hematol., 2004, 32: 852–860). Even though Ras usually acts as a dominant transforming oncogene, in primary cells and some cancer cell lines, Ras inhibits cell growth. We have previously shown that ectopic expression of N-rasm in leukemia U937 and K562 cells leads to growth suppression (Passioura et al. Cancer Res. 2005, 65, 797–804). The expression profile induced by N-rasm in these cells included the up-regulation of transcription factor Interferon Regulatory Factor1 (IRF1) and activation of cdk inhibitor p21WAF. IRF1 was previously defined as a tumour suppressor, and as such is a target of oncogenic mutations in AML. Antisense suppression of IRF1 prevented N-rasm induced growth suppression and up-regulation of p21WAF1. These results defined a novel tumour suppressive response to oncogenic N-rasm in leukemia cells. A retroviral cDNA library screen for genes that counteract N-rasm-induced growth suppression identified the gene for the Interferon Regulatory Factor2 (IRF2), and as confirmation of the screen, over-expression of IRF2 in leukemia U937 cells acted to inhibit N-rasm-induced growth suppression (Passioura et al. Oncogene. 2005; 24: 7327–36). IRF2 is known for its oncogenic properties and can antagonise IRF1-mediated tumour suppression. In addition, IRF2 is often up-regulated in primary leukemia samples. Here we show that IRF2 gene suppression using RNA interference acts to suppress the growth of leukemia TF-1 cells bearing N-ras mutation in codon 61 and expressing high levels of IRF1 and IRF2 and low level of p21Waf1. IRF2 down-regulation confirmed at RNA (quantitative RT-PCR) and protein (Western analysis) levels resulted in up-regulation of p21Waf1 and G2/M- rather than G1/S-growth arrest. In addition, increased polyploidisation that results from discoordinated DNA synthesis in mitotically arrested cells, was observed. In addition, IRF2-down-regulation significantly reduced clonogenic growth of the leukemic blasts. Cell growth of normal hematopoietic progenitor cells that express low levels of both IRF1 and IRF2, however, was not affected by IRF2 targeting. IRF2 targeting is currently being examined in primary AML samples in an animal model of AML. We suggest that IRF2 suppression can be used for ex vivo purging of leukemia cells in the autologous stem cell transplantation setting. To the best of our knowledge, specific IRF2 inhibition in cancer cells as a potential therapeutic approach has not been tested to date. IRF2 suppression may prove to be a novel therapeutic target for leukemia therapy.