The Pias-Like Coactivator ZMIZ1 Drives C-MYC Induction in Collaboration with NOTCH1 in T-Cell Acute Lymphoblastic Leukemia.

Abstract 2380

Activating NOTCH1 mutations are found in 50–60% of human T-cell acute lymphoblastic leukemia (T-ALL) samples. In mouse models, these mutations generally fail to induce leukemia. Cooperating oncogenes must be recruited by NOTCH1 to fully induce leukemia. Murine insertional mutagenesis screens previously implicated ZMIZ1 as a possible NOTCH1 collaborator in leukemia (Uren et al., Cell, 2008; Dupuy et al., Nature, 2005; Berquam-Vrieze et al., Blood, 2011). ZMIZ1 is a transcriptional co-activator of the Protein Inhibitor of Activated STAT (PIAS)-like family. It shares a zinc finger domain, the MIZ domain, with PIAS proteins. The MIZ domain mediates interactions with DNA-binding transcription factors and sumoylation. Previously, we showed that ZMIZ1 promotes T-ALL in collaboration with leukemia-associated NOTCH1 alleles in mouse models. ZMIZ1 and activated NOTCH1 were co-expressed in a subset of human patients. Genetic ZMIZ1 inhibition slowed leukemic cell growth and overcame resistance of some T-ALL cell lines to NOTCH inhibitors. ZMIZ1 may be a new clinically relevant oncogene. Here we sought to determine the downstream target genes of ZMIZ1 in leukemia. Validation of gene expression profiling data identified C-MYC and IL7RA as downstream targets of ZMIZ1. Targeting the C-MYC or IL-7 pathways using genetic and pharmacological inhibitors partly phenocopied the growth inhibitory effects we previously saw with ZMIZ1 inhibition. In order to determine whether these genes are direct or indirect targets of ZMIZ1, we generated an estrogen fusion protein, ZMIZ1-ER. ZMIZ1-ER induced C-MYC and IL7RA expression in the presence of tamoxifen, but failed to induce these genes with the addition of cycloheximide. These data suggest that C-MYC and IL-7RA are indirect targets. Like the PIAS proteins, ZMIZ1 appeared to have a broad effect on transcription to exert its functions. We next sought to elucidate the biochemical mechanism of ZMIZ1. Ectopic expression of ZMIZ1 or NOTCH1 had weak effects on endogenous c-Myc expression and failed to rescue a C-MYC-dependent T-ALL cell line after withdrawal of ectopic C-MYC. In contrast, ZMIZ1 in combination with NOTCH1 dramatically induced C-MYC expression by several fold and rescued the C-MYC dependent cell line. ZMIZ1 enhanced the ability of even weak NOTCH1 mutants to induce C-MYC, suggesting a mechanism by which ZMIZ1 may increase resistance to NOTCH inhibitors. ZMIZ1 did not influence C-MYC expression post-transcriptionally. It functioned primarily as a transcriptional activator. Although both C-MYC and IL7RA are both NOTCH1 target genes, ZMIZ1 did not directly interact with NOTCH1 or influence the expression of several other NOTCH1 target genes such Ptcra, Hes1, Dtx1, and Cd25. Thus, ZMIZ1 did not pan-activate NOTCH signaling. Based on bioinformatic analysis, we generated mutants that deleted individual domains of ZMIZ1. All mutants expressed at high levels by Western blot. Deletion of the transcriptional activation domain or the N-terminal domain (NTD) abolished the ability of ZMIZ1 to induce c-Myc and drive proliferation. Surprisingly, deletion of the PAT-like, Proline-rich, and MIZ domains or all three domains simultaneously had no effect on ZMIZ1 function. The 120-amino acid NTD has a predicted helical structure without significant sequence homology to any known domain. It is not found in ZMIZ2 or PIAS proteins. In summary, the mechanism of ZMIZ1 appears to be novel, indirect, transcriptional, and independent of canonical NOTCH and PIAS functions. Our study demonstrates the importance of characterizing genetic collaborations between parallel leukemic pathways that may be therapeutically targeted. They also raise new inquiries into potential NOTCH-ZMIZ1 collaboration in a variety of C-MYC-driven cancers.