The Philadelphia (Ph) chromosome stems from a reciprocal translocation between chromosomes 9 and 22. On a gene level, this rearrangement places upstream domains from the Bcr gene from chromosome 22 in juxtaposition with the downstream tyrosine kinase domains of Abl, from chromosome 9. The Ph chromosome is found in virtually all cases of chronic myeloid leukemia (CML) and approximately 5 percent of pediatric and 25 percent of adult acute lymphoblastic leukemia (Ph+ ALL). An intriguing mystery is how Bcr-Abl causes both a myeloid (CML) and lymphoid (Ph+ ALL) leukemia, and why does chronic-phase CML usually progress to myeloid blast crisis, but sometimes to lymphoid blast crisis?

In the mid-1990s, researchers described the gene Ikaros, a member of a family of zinc-finger-containing transcription factors. Like many such genes, it possessed DNA domains involved in forming homo- and heterodimer formation, with internal DNA domains coding for zinc fingers. Ikaros undergoes several splice variations, and it is thought that the mix of these splice variants influences Ikaros function. It appears that in these various isoforms, certain exons seemed essential for normal lymphocyte development, whereas a shift in splicing was associated with lymphoid malignancy.1-3 

Previously, in vitro experiments suggested that Bcr-Abl induction causes a shift in Ikaros splicing, encouraging the non-DNA binding isoforms. However, a more recent elegant paper highlights the role of DNA structural changes and the expression bias of Ikaros isoforms, demonstrating that actual deletions occur in the gene, eliminating the key DNA-binding domains. In a genome-wide screen of deletions and additions in ALL, Mullighan, et al. found an exceptionally high loss of the IKZF1 gene (coding for Ikaros) in Ph+ ALL cases. Of the 43 Ph+ cases studied, 36 (84 percent) had the IKZF1 deletion. In 19 cases, the deletion was restricted to the region coding Ikaros exons 3-6 (designated the Ik6 variant), which code for the DNA binding domain. The authors then examined 159 ALL cases looking for Ik6 mRNA expression, and found it only in cases that harbored the exon 3-6 deletion, strongly suggesting that Ik6 expression was based on DNA structural changes, rather than a splicing variation. Chronic-phase CML was free of alterations in the IKZF1 site. However, four of 15 blast crisis samples showed a IKZF1 deletion, including two/three lymphoid blast crisis. Sequence analysis of the IKZF1 deletion suggested that exons 3-6 were possibly deleted by aberrant work of the lymphoid RAG-mediated recombination machinery, which functions normally in differentiation to create V(D)J rearrangements. Thus, the study implicates non-DNA-binding Ikaros variants associated with the pathogenesis of Ph+ lymphoid malignancy.

A complementary paper by Iacobucci, et al. suggests that the Ik6 variant may be important in understanding the resistance of Ph+ ALL to tyrosine kinase inhibition. At diagnosis, nearly 50 percent of Ph+ ALL expressed only the non-DNA-binding Ik6 isoform. In vitro studies in cell lines suggested that the Ik6 variant was associated with an impaired apoptotic response to TKI exposure, increased DNA synthesis, and more robust colony growth.

These studies suggest an important role of Ikaros in the pathogenesis of Ph+ lymphoid malignancies. Many interesting questions remain. What is the link between Bcr-Abl and Ikaros? If the Ik6 isoform causes a loss of Ikaros transcription regulation, what genes are deregulated? Would their correction be therapeutic? One suspects that this Ikaros has wings that can withstand the heat of investigation (sorry, readers, I couldn't resist).

Competing Interests

Dr. Radich indicated no relevant conflicts of interest.