As expected from a transcriptional activator, the group of genes bound by MYCN but not MeCP2 had significantly higher levels of expression than those bound by MeCP2 alone. Groups of genes bound in combination with MYCN and MeCP2 had median expression that was intermediate to those bound by only MYCN or MeCP2. These results suggest that both proteins can act as modulators of transcription. Previous ChIP-chip analysis of MeCP2 binding sites within the neuroblastoma cell line SH-SY5Y by Yasui et al. revealed that the majority of MeCP2 binding sites were unmethylated, occurred outside of CpG islands, and that downstream genes were actively expressed . These results challenged the dominant model of MeCP2 as a functional repressor of transcription, acting by binding to methylated CpG dinucleotides, facilitating the recruitment of corepressors and chromatin remodeling complexes . Our results from the promoter arrays also indicate that MeCP2, as well as MYCN, bind to a relatively small proportion of hypermethylated promoters . Interestingly, similar analysis of the custom PD 0332991 citations tilling array, which contained a high proportion of inter/intragenic DNA sequences, revealed a much higher association with regions of hypermethylation for both MYCN and MeCP2. . The higher frequency of MYCN/MeCP2 co-localization at hypermethylated non-promoter regions is consistent with both proteins playing a role in the maintenance of chromatin structure. An intriguing possibility is that MYCN and MeCP2 can cooperate to regulate gene expression by altering higher-order chromatin structure. Several recent studies using chromosome conAbmole BioScience GDC-0879 formation capture techniques have shown that distally located cis-regulatory elements can be brought into proximity to the promoter of active genes, indicating that a chromatin loop forms to allow the adjacent association of both elements. A study of the imprinted Dlx5/Dlx6 locus using 3C combined ChIP with an anti-MeCP2 antibody, demonstrated the MeCP2 was required for the formation of a silencing chromatin loop, and that Mecp2- deficient mice had increased expression of Dlx5 and Dlx6 as a result of aberrant loop formation. The influence of MeCP2 in the formation of chromatin loops does not seem to be limited to transcriptional silencing, however, as gene expression and ChIPchip analysis of several genes, including JUNB, within the NB cell line SH-SY5Y, indicate that these genes are modulated by distal and proximal MeCP2 binding sites .