Next, the effect of 6 weeks of CUMS and continuous IMI treatment on the binding of MeCP2 to the Gdnf promoter was analyzed in the vSTR ( Figure 4I). ChIP analysis revealed that CUMS significantly increased MeCP2 binding check details to the Gdnf promoter in both

BALB and B6 mice, and continuous IMI treatment reversed this effect in stressed BALB mice. There was no significant difference in the binding of MeCP2 to the Bdnf promoter II region, which was assessed as a control. These results indicate that CUMS enhances the binding of MeCP2 to the Gdnf promoter in both mouse strains. We next investigated the functional role of methylated CpG site 2 on Gdnf expression in Neuro2a cells. Treatment of these cells with 5-aza-2′-deoxycytidine, an inhibitor of DNA methylation, reduced the methylation level at the Gdnf promoter ( Figure S8A) and concomitantly increased Gdnf mRNA expression ( Figure S8B). Next, the promoter activity

of a CpG site 2-specific methylated Gdnf luciferase reporter gene was investigated. We found that CpG site 2-specific methylation resulted in an approximately 68% decrease in reporter activity when MeCP2 and HDAC2 were cotransfected into Neuro2a cells ( Figure S8C). Previous reports have indicated that the high-affinity binding of MeCP2 to methylated DNA requires a run of four or more Glycogen branching enzyme A/T bases adjacent to the methylated CpG site ( Klose et al., ON-01910 mw 2005). We found two runs of A/T motifs located downstream of CpG site 2 ( Figure S8D). To test the role of these motifs on Gdnf promoter activity, wild-type and mutant reporters were constructed for the A/T motifs in CpG site 2 (m1, m2, and m3; Figure S8D). Then, the promoter activity of the CpG site 2-specific methylated and nonmethylated luciferase

reporters was measured using cotransfection experiments with MeCP2 and HDAC2 in Neuro2a cells ( Figure S8E). We found that in nonmethylated conditions, there was no mutation effect on reporter activity by cotransfection with MeCP2 and HDAC2, whereas in the specific methylation of CpG site 2, the reporter activities of wild-type and m1 and m2 mutants, but not m3 mutant, were affected by HDAC2 and MeCP2 overexpresson. These results suggest that the A/T motifs adjacent to CpG site 2 are critically involved in the MeCP2-HDAC2-mediated silencing of Gdnf transcription. Furthermore, we found that among the MBDs, MeCP2 was the most potent repressor of the CpG site 2-specific methylated reporter vector ( Figure S8F). Together with the results observed in vivo, these findings suggest that the methylation of CpG site 2 is important for the epigenetic repression of Gdnf expression.