New role for AML1/ETO in leukemogenesis

In this issue of Blood, Krejci and colleagues suggest that the transcription factor fusion protein, AML1/ETO (also known as RUNX1/MTG8), which is known to block differentiation in myeloid cells, may also increase DNA damage, adding to its functions in leukemogenesis. The AML1/ETO fusion protein was originally identified 15 years ago as the protein product of the t(8;21) translocation associated with M2 acute myeloid leukemia (AML).¹  The protein acts as a transcriptional regulator by recruitment of nuclear corepressors leading to inhibition of the core binding factor complex, which includes the AML1 protein.²  This inhibition leads to a block in myeloid differentiation, but in vivo, AML1/ETO expression is not sufficient to generate acute leukemia.³  In their article, Krejci and colleagues study the expression of AML1/ETO in primary human CD34 cells. They demonstrate that cells expressing AML1/ETO induce expression of a DNA damage response that is p53 dependent, and suggest that the induction of a p53 response is associated with improved prognosis of t(8;21)⁺ AML.

The authors use retroviral constructs to express AML1/ETO in primary human CD34⁺ cells. As they have previously demonstrated,⁴  expression of AML1/ETO “immortalizes” these cells so that they can be studied over the course of several weeks in culture. Consistent with previous results,⁵  they make the observation that AML1/ETO expression leads to transcriptional down-regulation of genes involved in base excision repair of DNA damage, particularly the genes 8-oxoguanine glycosylase gene (OGG1), and polymerase epsilon gene (POLE).⁵  Importantly, they demonstrate that binding of AML1/ETO is through a RUNX1 binding site, suggesting that regulation of DNA damage-response genes may be a physiologic function of RUNX1. Curiously, although base excision repair is primarily involved in the repair of point mutations in DNA, the authors then demonstrate that AML1/ETO-expressing cells have an increase in DNA double-strand breaks and an up-regulation of p53 expression. Down-regulation of p53 expression with a siRNA approach in AML1/ETO-expressing cells makes the cells more sensitive to ionizing radiation and other DNA-damaging agents. Overall, the study makes a convincing argument that AML1/ETO expression leads to an altered response to spontaneous and induced DNA damage.

How are we to interpret these data in the context of a growing body of data about the multiplicity of functions of AML1/ETO and data about the role of DNA damage responses in hematologic malignancies? The answer to this is unknown, but the data do not support the idea that oncogenes such as AML1/ETO are simple stimulators of cell growth. If these complex proteins lead to development of a “super cell,” they do so by inducing the cells to develop mechanisms to survive stress (a transforming boot camp of sorts). Expression of AML1/ETO actually induces apoptosis in some cells.⁶  The presence of DNA double-strand breaks in these cells may represent the induction of replication stress and DNA breaks during S phase. The data leave unanswered the question of the necessity of this DNA damage-induced stress response for cellular transformation. However, the well-described structure of the AML1/ETO fusion may allow for structure-function studies that will better define the necessities and sufficiencies for transformation induced by this fascinating oncoprotein.