Back

UHRF1-SRA recognizes symmetric non-CG methylated DNA through dual-flip out of 5-methyl cytosines

Nakarakanti, N. K.; Abhishek, S.; Deeksha, W.; Eerappa, R.

2019-12-19 biophysics
10.1101/2019.12.17.880419 bioRxiv
Show abstract

Non-CG DNA methylation (non-mCG) is enriched in the genome of brain neurons and germline cells. Non-mCG is differentially distributed on neuronal X-chromosome in males and females. Accumulation of non-mCG during postnatal brain development correlates with reduced gene expression and inactivation of distal regulatory elements, and allele specific gene regulation. Recently, UHRF1 has been found to contribute to de novo non-CG methylation, however, whether UHRF1 could recognize non-mCG is not known. Here, we have demonstrated through calorimetric measurements that the SRA domain of UHRF1 can recognize mCH and fully-mCHG, types of non-mCG. Furthermore, our ITC binding analyses with methylated CG DNA (mCG) revealed 6-fold decrease in binding affinity for fully-mCG compared to hemi-mCG and, despite symmetrical 5mCs, stoichiometry of 1:1 for UHRF1 SRA binding to fully-mCG indicates UHRF1 may not form stable complex with fully-mCG DNA. In contrast, UHRF1 SRA recognizes fully-mCHG with a stoichiometry of 2:1 protein to DNA duplex, and has tighter binding compared to fully-mCG. Crystal structure of UHRF1 SRA bound to 5mC containing DNA in fully-mCHG context revealed dual flip-out mechanism of 5mC recognition. Altogether, this study indicates that UHRF1 SRA also recognizes non-mCG DNA, besides known hemi-mCG DNA and exhibits contrasting mechanisms for hemi-mCG and fully-mCHG DNA recognition. These findings may open a new window to investigate the biological function of non-CG methylation recognition by the UHRF1.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.