NUP98-HBO1-fusion generates phenotypically and genetically relevant chronic myelomonocytic leukemia pathogenesis.

Chronic myelomonocytic leukemia (CMML) constitutes a hematopoietic stem cell (HSC) disorder characterized by prominent monocytosis and myelodysplasia. Although genome sequencing has revealed the CMML mutation profile, the mechanism of disease development remains unclear. Here we show that aberrant histone acetylation by nucleoporin-98 (NUP98)-HBO1, a newly identified fusion in a patient with CMML, is sufficient to generate clinically relevant CMML pathogenesis. Overexpression of NUP98-HBO1 in murine HSC/progenitors (HSC/Ps) induced diverse CMML phenotypes, such as severe leukocytosis, increased CD115+ Ly6Chigh monocytes (an equivalent subpopulation to human classical CD14+ CD16- monocytes), macrocytic anemia, thrombocytopenia, megakaryocyte-lineage dysplasia, splenomegaly, and cachexia. A NUP98-HBO1-mediated transcriptional signature in human CD34+ cells was specifically activated in HSC/Ps from a CMML patient cohort. Besides critical determinants of monocytic cell fate choice in HSC/Ps, an oncogenic HOXA9 signature was significantly activated by NUP98-HBO1 fusion through aberrant histone acetylation. Increased HOXA9 gene expression level with disease progression was confirmed in our CMML cohort. Genetic disruption of NUP98-HBO1 histone acetyltransferase activity abrogated its leukemogenic potential and disease development in human cells and a mouse model. Furthermore, treatment of azacytidine was effective in our CMML mice. The recapitulation of CMML clinical phenotypes and gene expression profile by the HBO1 fusion suggests our new model as a useful platform for elucidating the central downstream mediators underlying diverse CMML-related mutations and testing multiple compounds, providing novel therapeutic potential.

A novel immunotoxin reveals a new role for CD321 in endothelial cells.

There are currently several antibody therapies that directly target tumors, and antibody-drug conjugates represent a novel moiety as next generation therapeutics. Here, we used a unique screening probe, DT3C, to identify functional antibodies that recognized surface molecules and functional epitopes, and which provided toxin delivery capability. Accordingly, we generated the 90G4 antibody, which induced DT3C-dependent cytotoxicity in endothelial cells. Molecular analysis revealed that 90G4 recognized CD321, a protein localized at tight junctions. Although CD321 plays a pivotal role in inflammation and lymphocyte trans-endothelial migration, little is known about its mechanism of action in endothelial cells. Targeting of CD321 by the 90G4 immunotoxin induced cell death. Moreover, 90G4 immunotoxin caused cytotoxicity primarily in migratory endothelial cells, but not in those forming sheets, suggesting a critical role for CD321 in tumor angiogenesis. We also found that hypoxia triggered redistribution of CD321 to a punctate localization on the basal side of cells, resulting in functional impairment of tight junctions and increased motility. Thus, our findings raise the intriguing possibility that endothelial CD321 presented cellular localization in tight junction as well as multifunctional dynamics in several conditions, leading to illuminate the importance of widely-expressed CD321 as a potential target for antitumor therapy.