Altered Transcription and Replication Driving MYC-Induced B and T Cell Leukemias

MYC is over-expressed by many cancers, yet its oncogenic mechanisms are incompletely understood. MYC is central to acute lymphoblastic leukemia (ALL) - the most common and second most lethal pediatric malignancy, and ALL afflicts even more adults. Much of MYC's oncogenic function is attributed to its role as a transcription factor, but MYC has been shown to deregulate DNA replication independent of transcription. As master regulator of transcriptomes and epigenomes, we predict that MYC impacts both biologic features in both ALL types, B- and T-ALL. We hypothesize that MYC alters both RNA expression and DNA replication (the ordered spatio-temporal process where genomic domains replicate in either early or late S-phase) in B- and T-ALL, and that these perturbations-some shared, others unique to one ALL type-drive leukemogenesis.

Our project utilizes a unique double-transgenic rag2:hMYC, lck:GFP zebrafish ALL model that we established, which is the only animal model that develops both highly penetrant B- and T-ALL. In this model, B- and T-ALL are induced by human MYC (hMYC) that is regulated by a zebrafish (Danio rerio) rag2 promoter. Because B and T lymphoblasts each express rag2, both lineages over-express MYC, inducing B- and T-ALL. The differential activity of the D. rerio lck promoter (regulating GFP) causes B cells to fluoresce dimly and T cells to fluoresce brightly, permitting identification of B- vs. T-ALL by fluorescent microscopy and FACS-purification. Thus, we can compare B- and T-ALL in an isogenic background.

We have collected 30 ALL samples (18 T-ALL, 12 B-ALL) and completed two types of analyses on 12 T-ALL and 3 B-ALL. Using RNA-seq, we established gene expression profiles (GEP) for both ALL types; principal component analysis and other clustering algorithms demonstrate B- and T-ALL are distinct. Although we analyze the entire transcriptome, we prioritize genes conserved in humans to focus on translatable targets. To assess DNA replication, we generated Replication Timing (RT) profiles by first FAC-sorting ALL cells based on cell cycle phase (G1, S, G2; defined by DNA content) and then performing whole-genome sequencing to generate RT profiles for the same ALL analyzed by RNA-seq. We identified differentially replicating regions by comparing RT of B-vs. T-ALL, revealing many loci where replication reproducibly shifts from early-to-late, or late-to-early, based on ALL type.

Overall, despite their shared genetic driver (MYC) , we found RT differences that distinguish B- vs. T-ALL in ~30% of the genome. Most differences occur in large chromosomal domains, suggesting abnormal chromatin structure in ALL. An additional unexpected result was that many ALL G1 samples had read count differences across large chromosomal regions, indicating the presence of aneuploidies/large CNAs. Several were recurrent and lineage-specific (i.e., exclusive to B- or T-ALL).

Together, our data demonstrate differences in RNA transcription, DNA replication, and regions of genomic instability that are lineage-specific, despite a shared MYC oncogene that drives both B- and T-ALL. We will next determine which deranged loci are also perturbed in human ALL, with an overarching goal of finding prognostic biomarkers and therapeutic targets. MYC hyper-activity occurs in ~70% of human malignancies. Thus, MYC is crucial to virtually all cancer biology, making our findings likely to inform not only the mechanisms that drive ALL, but also other cancers where MYC is oncogenic.