RIZ1 Is Downregulated during CML Progression and Displays Tumor Suppressor Properties in CML Cell Lines.

Transition to CML blast crisis is characterized by the continued presence of the Philadelphia chromosome and the acquisition of additional molecular and chromosomal alterations. Loss of heterozygosity at chromosome region 1p36 is a frequent finding in CML blast crisis. RIZ1 is located at 1p36 and frequently undergoes deletion, rearrangements, and loss of heterozygosity in a variety of cancers. Taken together, these data suggest that decreased RIZ1 expression may contribute to CML progression. We used immunohistochemistry to analyze RIZ1 expression in matched bone marrow biopsy specimens from CML chronic phase patients that progressed to accelerated phase or myeloid blast crisis. Similar to control bone marrow, strong cytoplasmic and nuclear RIZ1 expression was observed during chronic phase in all cases. However, RIZ1 expression was found to be markedly decreased in matched bone marrow biopsy specimens obtained during myeloid blast crisis in five patients. RIZ1 expression was maintained in the immature cells of two CML patients, one in accelerated phase with 15% myeloid blasts and the other in myeloid blast crisis, indicating that low RIZ1 expression is not an inherent property of immature hematopoietic cells. To confirm this, we analyzed G-CSF mobilized peripheral blood by flow cytometry and found RIZ1 to be expressed in mature myeloid and CD34+ cells. Transient transfection of myeloid blast crisis cell lines K562, YN-1, ERY-1, and JURL-MK1 with a RIZ1 expression plasmid (pRIZ1) increased the number of cells undergoing early and late apoptosis and reduced viability by 20–80% within 24 hours. RIZ1 effects on erythroid differentiation were assessed in K562, YN-1, and ERY-1 as these cell lines express low levels of hemoglobin, reflecting their myeloid/erythroid progenitor phenotype. As transient RIZ1 expression in these cells is too toxic to measure erythroid differentiation, we modified K562 to express less toxic levels of RIZ1 by stably integrating RIZ1 under the control of a CMV promoter (K562+RIZ1). As determined by benzidine staining, stable expression of RIZ1 in K562+RIZ1 increased erythroid differentiation compared to K562 alone. To confirm that RIZ1 is responsible for the enhanced erythroid differentiation, we transfected K562+RIZ1 as well as ERY-1 and YN-1 (which have higher endogenous RIZ1 levels than K562), with a plasmid that expresses RIZ1 shRNA (pRIZ1shRNA). Expression of pRIZ1shRNA in K562+RIZ1 reduced RIZ1 protein expression and erythroid differentiation to levels similar to that observed in K562 and decreased erythroid differentiation in ERY-1 and YN-1. RIZ1 effects on JURL-MK1 differentiation were assessed by measuring CD33 and CD117 as the expression of these proteins decreases during myeloid differentiation. Transient transfection of JURL-MK1 with pRIZ1 decreased CD33 and CD117 expression as assessed by flow cytometry. In summary, our study demonstrates that RIZ1 expression is frequently reduced in myeloid blast crisis. Furthermore, RIZ1 expression decreases cell proliferation, increases apoptosis, and enhances differentiation in cell line models of CML myeloid blast crisis. Taken together, our results build upon previous observations that a putative CML tumor suppressor gene is present at 1p36 and suggest that deregulation of RIZ1 expression may contribute to CML progression.