ABSTRACT: We recently revealed that myeloid master regulator PU.1 directly represses metallothionein (MT)-1G through its epigenetic activity, but the functions of MT-1G in myeloid differentiation remain unknown. To clarify this, we established MT-1G-overexpressing acute promyelocytic leukemia NB4 (NB4MTOE) cells, and investigated whether MT-1G functionally contributes to all-trans retinoic acid (ATRA)-induced NB4 cell differentiation. Real-time PCR analyses demonstrated that the inductions of CD11b and CD11c and reductions in myeloperoxidase and c-myc by ATRA were attenuated in NB4MTOE cells. Since G1 arrest is a hallmark of ATRA-induced NB4 cell differentiation, we observed a decrease in G1 accumulation, as well as decreases in p21WAF1/CIP1 and cyclin D1 inductions, by ATRA in NB4MTOE cells. Nitroblue tetrazolium (NBT) reduction assays revealed that the proportions of NBT-positive cells were decreased in NB4MTOE cells in the presence of ATRA. Microarray analyses showed that the changes in expression of several myeloid differentiation-related genes (GATA2, azurocidin 1, pyrroline-5-carboxylate reductase 1, defensin-4, C-X3-C motif receptor 3, matrix metallopeptidase -8, S100 calcium-binding protein A12, neutrophil cytosolic factor 2 and oncostatin M) induced by ATRA were disturbed in NB4MTOE cells. Collectively, overexpression of MT-1G disturbs the proper differentiation of myeloid cells. The present study provides evidence that expression analysis of MT-1G in acute promyelocytic leukemia patients may be a good prediction marker to estimate the efficacy of ATRA. Cell culture and generation of MT-1G-overexpressing cells: To generate MT-1G-overexpressing cells and their control cells, the MT-1G expression vector and its parental pcDNA 3.1/myc-His(-) version A vector (Invitrogen) were transfected using a CLB-Transfection device (Lonza, Basel, Switzerland). NB4 clones stably transfected with the vectors were isolated by limiting dilution and selection with 400 µg/ml of neomycin in RPMI (Gibco BRL, Rockville, MD) containing 10% heat-inactivated fetal bovine serum (HIFBS). Cells were cultured under 5% CO2 at 37°C in a humidified atmosphere. Microarray analyses: MT-1G-overexpressing (NB4MTOE) cells and their control cells were seeded at a density of 1×105 cells/ml and treated with 1 µM all-trans retinoic acid (ATRA). The cells were harvested after 72 h and total cellular RNA was isolated from control (NB4pcDNA4, 6, 7) cells and NB4MTOE (NB4MT22, 23, 25) cells using an RNA Mini Purification Kit (Qiagen, Miami, FL) according to the manufacturer’s protocol. Aliquots containing 10 µg of RNA from each sample of control cells were mixed and used as controls. Similarly, 10 µg of RNA from each sample of NB4MTOE cells were mixed and used as NB4MTOE cells. The samples were subjected to microarray analyses using a CodeLink Human 54K Whole Genome Bioarray (Filgen, Nagoya, Japan).