RUNX1-ETO induces an anti-leukemic response that is overcome in acute myeloid leukemia development /
- Author(s): De Kelver, Russell Christopher
- et al.
Acute myeloid leukemia (AML) is the most prevalent form of adult leukemia. The most common chromosomal translocation associated with this malignancy is t(8;21)(q22;q22), which generates the fusion protein RUNX1-ETO (AML1-ETO), a transcription factor that both activates and represses target gene transcription. However, RUNX1-ETO alone is insufficient to cause leukemia. Herein we examine the mechanisms of t(8;21)-associated AML development. We identify the N-CoR binding site in the NHR4 domain as a critical factor inhibiting the leukemic potential of RUNX1 -ETO. When this site is removed through mutation or deletion, there is an attenuation of both RUNX1-ETO transcriptional repression and negative cellular effects, such as induction of apoptosis and cell cycle block, and the fusion protein is able to induce leukemia in a murine model. This supports the idea that the degree of transcriptional dysregulation by RUNX1-ETO is an important factor in AML development. Therefore, we next examined gene classes differentially regulated by RUNX1-ETO and its leukemogenic isoforms RUNX1-ETO9a and RUNX1-ETO-W692A. We thereby identify type I interferon signaling as being strongly upregulated by RUNX1-ETO but less so by its leukemogenic isoforms. We further determine that interferons inhibit RUNX1-ETO-induced increases in self- renewal in vitro and that knockout of the type I interferon receptor significantly accelerates RUNX1-ETO9a- induced leukemia development in vivo. This identifies interferon signaling as one mechanism of RUNX1-ETO-induced cellular toxicity that must be overcome in leukemia development and supports the use of interferons in the treatment of AML. Finally, we examine the regulation of Alox5 by RUNX1-ETO9a. Although Alox5 has been shown to be critically required by BCR-ABL to induce chronic myeloid leukemia, our data demonstrate that it is not required in vivo for the induction of AML by RUNX1-ETO9a. However, in studying its regulation we identify KLF6 as a novel regulator of RUNX1-ETO9a transcriptional activation. KLF6 itself is strongly upregulated in primary human t(8;21)+ AML patient samples as well as in leukemia cell lines by both RUNX1-ETO and RUNX1-ETO9a. Furthermore, KLF6 interacts with both wildtype RUNX1 and the t(8;21) fusion proteins. This identifies KLF6 as a novel mediator of t(8;21) target gene regulation, providing a new mechanism for RUNX1-ETO transcriptional control