AUTHORS
Le'Mark Russell, Lynn White2, Haley Schmidt, PhD3, Abby Green, MD4, Jeffrey Bednarski, MD, PhD5
AFFILIATIONS
1 University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, United States, Washington University in Saint Louis, Saint Louis, Missouri, United States
2 Washington University in Saint Louis, Saint Louis, Missouri, United States
3 Washington University in Saint Louis, Saint Louis, Missouri, United States
4 Washington University in Saint Louis, Saint Louis, Missouri, United States
5 University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, United States, Washington University in Saint Louis, Saint Louis, Missouri, United States
ABSTRACT
The ETV6-RUNX1 translocation is the most prevalent genomic aberration in pediatric precursor B cell acute lymphoblastic leukemia (pre-B ALL). Despite this, the molecular mechanism underlying how ETV6-RUNX1 leads to malignancy remains incompletely understood. ETV6-RUNX1 is a fusion protein that arises in hematopoietic progenitor cells as a translocation of the ETV6 and RUNX1 loci resulting in the N-terminal regulatory domain of ETV6 fused to the full length RUNX1. ETV6-RUNX1 translocations arise in utero during fetal hematopoiesis. Although ETV6 is expressed in all hematopoietic cells, leukemias with ETV6-RUNX1 translocations are exclusively of pre-B cell origin and occur uniquely in children, and not adults. Thus, B cell-specific and age-specific factors contribute to ETV6-RUNX1-driven leukemic transformation.
To investigate mechanisms of leukemogenesis, we used a conditional murine model for expressing ETV6-RUNX1 in which exons 6-7 of Etv6 have been replaced by a floxed stop cassette followed by the RUNX1 gene (Etv6RUNX1). When crossed with VavCre (VavCre:Etv6RUNX1), the stop cassette is removed and RUNX1 is spliced onto Exon 5 of Etv6 in hematopoietic cells. This experimental system mirrors human disease with expression of ETV6-RUNX1 controlled by endogenous Etv6 regulatory elements and expression in hematopoietic cells, including B cells. To investigate age-specific effects of ETV6-RUNX1 expression on pre-B cells, we isolated pre-B cells from fetal liver (E17.5) and adult (8-week-old) bone marrow of control Etv6RUNX1 (which do not express the transgene) and VavCre:Etv6RUNX1 mice then cultured them in IL-7 containing media. IL-7 promotes proliferation and expansion of pre-B cells. ETV6-RUNX1-expressing pre-B cells from adult mice had reduced expansion, proliferation, and viability compared to control adult pre-B cells. In contrast, ETV6-RUNX1-expressing pre-B cells from fetal liver had similar expansion and viability as fetal pre-B cells from control mice. Thus, expression of ETV6-RUNX1 is differentially deleterious to adult pre-B cells whereas fetal pre-B cells tolerate the fusion and continue to proliferate.
To further assess the effect of ETV6-RUNX1 on adult pre-B cells, we performed bulk RNA-sequencing on sorted large (proliferating) and small (G1-arrested) pre-B cells from adult control and VavCre:Etv6RUNX1 mice. Expression of ETV6-RUNX1 profoundly altered pre-B cell transcriptional programs. Specifically, pre-B cells expressing ETV6-RUNX1 had aberrant expression of inflammation-associated genes including Hmox1, Bcl6, and Spic. We next asked whether ETV6-RUNX1 affects gene expression through distinct mechanisms in early B cells. In this regard, pre-B cells uniquely undergo immunoglobulin (Ig) receptor gene recombination. This process is mediated by the RAG endonuclease which generates double-stranded DNA breaks (DSBs) necessary for Ig gene assembly. RAG-mediated DSBs activate the DNA damage response kinase ATM, which in turn triggers a B cell-specific transcriptional program that coordinates pre-B cell development. We hypothesized that RAG DSB-mediated signaling influenced effects of ETV6-RUNX1 on pre-B cell gene expression. To test this, we expressed ETV6-RUNX1 in RAG-deficient and ATM-deficient pre-B cells, which do not generate RAG DSBs or activate DSB-dependent signaling, respectively. ETV6-RUNX1-mediated induction of inflammation-associated genes did not occur in pre-B cells lacking RAG or ATM. These findings demonstrate that ETV6-RUNX1 drives inflammatory gene programs in pre-B cells in a manner dependent on RAG DSB-mediated cellular signals.
Our results show that ETV6-RUNX1 synergizes with B cell-specific DNA damage signaling to mediate gene expression changes in pre-B cells, which differentially impacts B cell development in an age-specific manner. Ongoing work will define the mechanisms underlying differential activity of ETV6-RUNX1 in fetal versus adult pre-B cells and the cooperative activity of RAG-induced DNA break signals and ETV6-RUNX1 in regulating expression of inflammatory genes in pre-B cells. Together these studies establish new insight into mechanisms that promote leukemic transformation in ETV6-RUNX1-expressing pediatric pre-B ALL.
DISCLOSURES
Le'Mark Russell: No relevant conflicts of interest to declare.