such as embryonic stem cells and neural stem cells, express approximately 40-60% of the total genes in a genome; in contrast, most differentiated cells express only 10-20% of the total genes (1,4). Further, as we demonstrated, the rise of these methylation activities at the beginning of differentiation subsided after NSCs migrated and proceeded with advanced differentiation, indicating that the DNA methylation activities were low in mature neural cells in which house-keeping and functional activity genes are active. These observations indicate that the transient rise in DNMT activities for DNA methylation is a required program prior to differentiation. Our high throughput MeDIP-chip analysis of genome wide DNA methylation agrees that there is actively increased DNA methylation as well as decreased DNA methylation in genes with moderate methylation level in the NSCs going through differentiation (Manuscript in preparation). This hypothesis is supported by the retardation of migration and differentiation of NSCs by DNA methylation inhibitor AZA. This mechanism of AZA-inhibited gene methylation has proven to be a useful strategy to increase transgene expression following transplantation and removal of AZA (19). In the AZA treated NSCs, a disruption of 3me-H3K27 marks and DNMT1 distribution was also observed, which appears to be in line
