Molecular Determinants for Targeting Heterochromatin Protein 1-Mediated Gene Silencing: Direct Chromoshadow Domain–KAP-1 Corepressor Interaction Is Essential

Abstract

The KRAB domain is a highly conserved transcription repression module commonly found in eukaryotic zinc finger proteins. KRAB-mediated repression requires binding to the KAP-1 corepressor, which in turn recruits members of the heterochromatin protein 1 (HP1) family. The HP1 proteins are nonhistone chromosomal proteins, although it is unclear how they are targeted to unique chromosomal domains or promoters. In this report, we have reconstituted and characterized the HP1–KAP-1 interaction using purified proteins and have compared KAP-1 to three other known HP1 binding proteins: SP100, lamin B receptor (LBR), and the p150 subunit from chromatin assembly factor (CAF-1 p150). We show that the chromoshadow domain (CSD) of HP1 is a potent repression domain that binds directly to all four previously described proteins. For KAP-1, we have mapped the CSD interaction region to a 15-amino-acid segment, termed the HP1BD, which is also present in CAF-1 p150 but not SP100 or LBR. The region of KAP-1 harboring the HP1BD binds as a monomer to a dimer of the CSD, as revealed by gel filtration, analytical ultracentrifugation, and optical biosensor analyses. The use of a spectrum of amino acid substitutions in the human HP1α CSD revealed a strong correlation between CSD-mediated repression and binding to KAP-1, CAF-1 p150, and SP100 but not LBR. Differences among the HP1 binding partners could also be discerned by fusion to a heterologous DNA binding domain and by the potential to act as dominant negative molecules. Together, these results strongly suggest that KAP-1 is a physiologically relevant target for HP1 function.

ACKNOWLEDGMENTS

We thank Gabriela Canziani of the Biosensor/Interaction Analysis Structural Biology Cores Group (University of Pennsylvania; Irwin Chaiken, Director, and Jerry Salem, Manager) for the basic surface plasmon resonance training, experiment optimization, and data analysis. We are also grateful to P. B. Singh, H. J. Worman, J. C. Eissenberg, and G. Maul for supplying plasmids. We acknowledge the many helpful discussions from members of the F.J.R. laboratory and W. J. Fredericks, H. Peng, and D. C. Schultz for critical reading of the manuscript. We also thank Sandra L. Harper for help in creating the analytical ultracentrifugation figures.

M.S.L. was supported by Wistar basic cancer research training grant CA 09171. D.W.S. was supported by grants CA 74294 and CA 66671. F.J.R. is supported in part by National Institutes of Health grants CA 52009, CA 10815 (core grant), DK 49210, and GM 54220; ACS grant NP-954; the Irving A. Hansen Memorial Foundation; the Mary A. Rumsey Memorial Foundation; and the Pew Scholars Program in the Biomedical Sciences.