Ebf1 in DNA repair and leukemogenesis

In this issue of Blood, Prasad et al provide evidence for a new role for the B-lineage transcriptional regulator early B-cell factor 1 (Ebf1) during early B-cell development and B-cell acute lymphoblastic leukemia (B-ALL).¹ 

Transcriptional regulators such as Ebf1 have been shown to support B-cell differentiation by coordinating the expression of downstream transcription and signaling factors with the activation of the immunoglobulin gene recombination machinery.²  They also preserve B-lineage fidelity by suppressing expression of signaling and transcription factors that promote alternate fates latent in these early precursors. The outcome of this complex transcriptional process is the generation of the humoral arm of the immune system through the production of a diverse repertoire of properly selected mature B cells capable of antibody production on antigenic challenge. Loss-of-function mutations in EBF1 and other B-lineage transcriptional regulators, such as PAX5, are associated with B-ALL in humans.^3-5  Notably, the leukemogenic effect of these mutations manifests at the heterozygous level with differentiation only partly affected, suggesting that other dose-dependent functions of these factors in B-cell precursors are contributing to the disease state.

It might seem surprising that impaired function of EBF1 would lead to B-ALL because genetic studies in mice have shown that Ebf1 supports proliferation of B-cell precursors and mature B cells.^6,7  Prasad et al identify a significant reduction in homologous recombination (HR) DNA repair genes in the gene expression profiles of primary Ebf1^+/− mouse B-cell precursors (see figure). Of these downregulated genes, Rad51 is a key component of the HR DNA repair machinery that is loaded to resected single-stranded DNA ends at double-strand (ds)DNA breaks.⁸  Rad51, through its interactions with BRCA2 and BRCA1, forms a nucleoprotein filament that promotes strand invasion and DNA repair synthesis.⁹  Reduction in the activity of these genomic caretakers can leave dsDNA lesions unrepaired, compromising chromosome stability. This can lead to developmental defects by reducing survival of differentiating intermediates. However, in the case of the Ebf1^+/− B-cell precursors, expression of the prosurvival gene Bcl2 was increased, preventing apoptotic deletion of the mutant cells that have accumulated DNA damage (see figure).

Heterozygosity for Ebf1 gave rise to leukemias at a very low rate (87% survival at 40 weeks), but when combined with Pax5 heterozygosity, there was a strong increase in the leukemic potential (25% survival at 40 weeks) of B-cell precursors. Gene expression and genomic analysis of the leukemic clones revealed that, although leukemias share a B-cell precursor phenotype, their otherwise heterogeneous genetic profile indicates the contribution of different leukemia-promoting genes and affiliated pathways to this process. These findings are consistent with a role of Ebf1 in regulating DNA repair rather than repressing a specific growth-promoting pathway. Under this scenario, deficiency in Ebf1 compromises HR DNA repair, causing the generation of a garden variety of dsDNA breaks in B-cell precursors that normally lead to cell death. However, the increase in prosurvival genes allows these cells to persist and fuel leukemic transformation. Although these Ebf1 dose-dependent events in B-cell precursors are not overtly leukemogenic, the combination with Pax5 haploinsufficiency dramatically increases leukemic potential. It is proposed that by further stalling B-cell differentiation at a highly proliferative and recombination active stage, Pax5 haploinsufficiency allows the selection and expansion of precursors carrying appropriate DNA mutations that support aberrant cell growth (see figure).

Interestingly, the loss in expression of DNA repair factors was not maintained in the Ebf1^+/−Pax5^+/− leukemic clones, indicating selection against this leukemia-initiating mechanism once the leukemia was established. This could also suggest that another mechanism is in play, and one should examine whether the rise in leukemic clones from the double-mutant B-cell precursors can take place in the absence of DNA damage after Rad51 reconstitution. A drop in leukemic potential after Rad51 re-expression would conclusively demonstrate that loss in HR DNA repair was the main driving force of leukemic transformation of the Ebf1^+/−Pax5^+/− B-cell precursors. Another point of interest is whether Pax5 collaborates with Ebf1 in the regulation of the HR DNA repair pathway as it does in the regulation of many lymphoid genes.¹⁰  Expression studies and DNA damage assays on primary double-mutant B-cell precursors should settle the question of cooperation between these 2 key B-lineage transcription factors on the HR DNA repair pathway. Are Ebf1 and Pax5 indeed targeting distinct molecular pathways or are they also contributing to the same DNA repair pathway in a semiredundant manner? In either case, one might be able to harness the defect in DNA repair exhibited by these cells by interfering with the gain in survival of the leukemia-initiating cells. Insight into the Ebf1 and Pax5 dose-dependent mechanisms operating in normal B-cell precursors can provide new perspective for treatment of human leukemias that harbor mutations in these factors.

The current findings by Prasad et al indicate that lymphoid lineage-specific transcription factors are not only important for promoting gene expression programs supporting differentiation but also serve the need for increased surveillance of genomic integrity during the proliferative and recombination-active stages of lymphocyte development. Ebf1 is a key lineage-specific guardian of genomic stability at these early stages of B-cell differentiation.