28 September 2021
Paving the way for treatment of AML patients with EVI1 gene expression
Amongst AML (acute myeloid leukemia) patients, a subgroup in which the EVI1 gene is overexpressed has a very poor outcome, with low response to treatment and a high need for new treatment strategies. The activation of this oncogene is caused by chromosomal rearrangements on the long arm of chromosome 3. The exact mechanism was however unclear. Previous work from Oncode Investigator Ruud Delwel’s group at Erasmus MC showed that in both inv(3)/t(3;3) as well as other 3q26 rearranged AMLs, an enhancer is hijacked and drives oncogenic expression of EVI1. Now two new manuscripts of his group report on the mechanisms that cause overexpression of this EVI1 oncogene, which may provide new angles to identify strategies to modulate EVI1 expression for treatment of AML. The first paper was published online this May in Cancer Discovery and investigates in more detail how the hijacked GATA2 enhancer leads to EVI1 activation. The second study, published in Nature Communications today, focuses on the mechanism whereby EVI1 hijacks hematopoietic super-enhancers, using t(3;8) as a model.
“Acute Myeloid Leukemia is not one disease. Different subtypes with distinct abnormalities can be distinguished based on different genetic defects” says Oncode Investigator Ruud Delwel. “For many years the focus of our group has been on one of the most aggressive forms of AML, that is to say AML with overexpression of the EVI1 oncogene. We previously uncovered the mechanism of enhancer hijacking in this AML subgroup. In these two recent papers, we further shed light on the aberrant function of the regulatory elements driving oncogene expression” he explains.
In the Cancer Discover publication, the team applied an unbiased CRISPR/Cas9 enhancer scan in the inv(3) MUTZ3 cell line to uncover transcription factor binding elements essential for driving EVI1transcription in the hijacked GATA2 enhancer. This unbiased scan pinpointed a single regulatory element of approximately 1 kb of open chromatin that is critically required for aberrant EVI1 expression. This element contained a DNA binding motif for the transcription factor MYB, which specifically occupied this site at the translocated allele. This motif was required for EVI1 expression but dispensable for GATA2. Peptidomimetic blockade of MYB resulted in downregulation of EVI1 but not of GATA2. The data showed an essential role for MYB in the regulation of EVI1 expression via the translocated GATA2 enhancer.
”Tumor growth in AML patients with inv(3)/t(3;3) is dependent on EVI1activation. This subtype of AML is characterized by its aggressive course and is refractory to therapy. We show a novel paradigm in which chromosomal aberrations enable the activation of critical regulatory elements that are non-functional at their endogenous loci. This knowledge provides a rationale to develop new compounds to selectively interfere with oncogenic enhancer activity” says Leonie Smeenk, lead author of the Cancer Discovery publication.
Overexpression of EVI1 is found not only in in AML with inv(3)/t(3;3), but also with other 3q26 translocations. Previously, the team showed that AMLs with other rearrangements of 3q26, such as t(3;6) or t(3;8), also involve enhancer hijacking by hematopoietic super-enhancers. This led to the hypothesis of a common oncogenic mechanism for overexpression of EVI1 in those leukemias. For the manuscript published today in Nature Communications, they tested this hypothesis and studied the mechanism of the t(3;8) translocation in more detail.
To test the hypothesis, the team used CRISPR-Cas9 technology and mimicked a chromosomal rearrangement found in patients which lead to EVI1overexpression. In this model, a MYC super-enhancer is translocated to the long arm of chromosome 3. Using this model, the team then identified regions in the MYC enhancer critical for enhancer-promoter interaction and demonstrated that a single CTCF binding site near the EVI1 promoter is crucial for EVI1 activation through enhancer-promoter interaction. The study showed that the CTCF protein, a master regulator of 3D genome architecture, is also involved in enhancer hijacking by EVI1 in other 3q26 rearrangements, including inv(3)/t(3;3).
“These findings are important because AMLs with 3q26 rearrangements are insensitive to any current form of chemotherapy, but highly depend on the expression of EVI1. Understanding what mechanisms drive the overexpression of EVI1 will pave the way to the discovery of targeted therapies that overcome the chemo resistance of these leukemias” says Sophie Ottema, lead author of the paper published today in Nature Communications. “Furthermore, our novel t(3;8) K562 model is a useful tool in compound screens aimed at discovering drugs that successfully downregulate EVI1” she adds.
More insights are to be expected in the future. “We now generated two models that I believe can be of importance to find the regulators of aberrant transcription and uncover drugs able to interfere with oncogene overexpression. Using a drug repurposing library obtained through Oncode we will carry out the first screen using our EVI1-reporter models” says Ruud Delwel regarding the next steps. Besides testing compounds to interfere with EVI1 expression in a large-scale screen, the team is also exploring a rational approach based on the knowledge generated in these studies. Given the preference of MYB for the rearranged allele, they are investigating strategies to selectively target EVI1 via MYB while sparing the wild type allele. The inv(3)/t(3;3) and the t(3;8) models generated in both studies are indispensable tools in this endeavour.