Identifying the key functions of MeCP2 via genetic manipulation in mice

Abstract

MeCP2 was identified by its ability to bind DNA in a methylation-specific manner. Yet, how it interprets the DNA methylome remains unclear. Several mechanisms have been proposed, including a role in transcriptional repression. MeCP2 is highly abundant in the brain, and loss-of-function mutations result in a neurological disorder called Rett syndrome (RTT). Strikingly, RTT-causing missense mutations are almost all located in either the methyl-CpG-binding domain (MBD) or a region that has been shown to bind the NCoR/SMRT co-repressor complex (NID). This suggests that the MBD and the NID are the key functional domains in MeCP2, and that the role of MeCP2 is to form a ‘bridge’ between chromatin and the co-repressor complex to regulate gene expression. To test this ‘bridge’ hypothesis, I have made an allelic series of knock-in mice with truncated forms of MeCP2 to determine whether the other regions are dispensable for protein function. The three other regions of MeCP2 (the N-terminus before the MBD, the Intervening region between the MBD and the NID, and the C-terminus after the NID) were deleted in a step-wise manner to produce progressively smaller truncated proteins. Knock-in mice which lack just the N-terminus or both the N- and C-termini are phenotypically normal. Therefore, these regions, which together make up 46% of the protein sequence, are dispensable for MeCP2 function in vivo. Additional deletion of the Intervening region, retaining only 34% of the original sequence, results in mild RTT-like symptoms in the knock in mice. This is likely to be caused by this protein’s decreased stability and reduced ability to bind the NCoR/SMRT complex in the brain. The most severely truncated protein is nevertheless able to reverse the Mecp2-null phenotype when reactivated after the onset of symptoms. Together, these findings strongly support the ‘bridge’ hypothesis.