THE HELIX-TURN-HELIX AS A SCAFFOLD FOR CHIMERIC NUCLEASE DESIGN

ABSTRACT

De novo design is a powerful tool to investigate the active site of enzymatic metalloproteins, in a smaller, defined model system. It is also a way to build or combine activity and selectivity in unique ways, not seen biologically. We are utilizing protein design to build artificial endonucleases, and to investigate fundamental questions of metallonuclease structure and function. We have focused on designing peptide constructs comprising geometrically similar turns from unrelated proteins, in particular the Ca-binding EF-hand motif of calmodulin and the helix-turn-helix motif (HTH) of engrailed homeodomain. By substituting the calcium-binding (and thus lanthanide-binding) loop in place of the “turn” of engrailed HTH, hydrolytically active, DNA-binding constructs were created. The NMR solution structure of one La-binding chimera (P3W), calculated based on NOE volume integrals, demonstrated that the 33-mer peptide retains the parental helix-turn-helix structure when bound to lanthanide ions. The binding affinities of the chimeras for Ln(III) ions are in the low µM regime, typical for EF-hand sequences, despite the significant changes in flanking sequence. Importantly, the Ln(III) chimeras are catalytically competent, able to hydrolyze phosphate esters including DNA, and were found to bind and cleave DNA with sequence preference. Thus, these designed HTH/EF-hand chimeras represent the first examples of small peptidic artificial nuclease with sequence discrimination, and show that the HTH is a robust scaffold on which to build novel metallopeptide constructs. This review describes the design and characterization of Ln-binding HTH/EF-hand chimeras.

ACKNOWLEDGMENTS

SJF wishes to thank Profs. Kenneth N. Raymond, Jacqueline K. Barton and Sture Forsén for inspiration (fundamental, influential and coincidental, respectively) leading to this work. She also wishes to thank her co–workers, whose efforts made this research possible. This work was supported by an NSF-CAREER grant (CHE-0093000), the University of Iowa Carver Research Foundation, and the American Cancer Society (IN-122U). JTW was supported by a Center for Biocatalysis and Bioprocessing NIH Predoctoral Training Grant (T32 GM-08365). The authors thank Dr. William R. Kearney for assistance with Figure 4.

Notes

†Current address: Schering-Plough Research Institute, Kenilworth, NJ, USA.