An engineered chromatin protein with enhanced preferential binding of H3K27me3 over H3K9me3

Abstract

The human genome is organized within the nucleus as chromatin, which is largely comprised of histone proteins that assemble on DNA into nucleosome complexes. Histone post-translational modifications (PTMs) are dynamic chromatin features that signal distinct gene expression states and modulate important cellular functions like cell differentiation. Histone binding domains (HBDs) from chromatin reader-effector proteins are being used as new tools to target and track histone PTMs in living cells. HBDs bind histones through multi-contact interactions that may confer more specificity than antibodies, but are hard to study because of their weaker affinity in vitro. To explore the large HBD design space, we developed the Cell-Free Histone-Binding Immunoassay (CHIA) where interactions between cell-free-expressed HBD proteins and immobilized biotinylated histone peptides are measured in an ELISA-style assay. We showed that the number of functional CBX8 polycomb chromodomains (PCD) in a fusion protein scales with H3K27me3 binding. We tackled the challenge of engineering a high affinity HBD that distinguishes H3 A-A-R-K27me3-S from a similar region on the same histone, T-A-R-K9me3-S. Previously reported K33E and Q9D CBX7 PCD variants bound with high affinity to H3K27me3, and bound as strongly with H3K9me3. In contrast, the K33E substitution enhanced CBX8 PCD binding to K27me3 with minimal K9me3 binding. To determine if the CBX8 hydrophobic clasp (V10 and L49) supports K27me3 specificity we tested hydrophobic substitutions, and observed increased affinity and strong specificity for H3K27me3. These results will enable more robust sensing of H3K27me3 for applications such as histone PTM-detection, and cell engineering.