A consensus zinc finger peptide: design, high-affinity metal binding, a pH-dependent structure, and a His to Cys sequence variant

A single zinc finger peptide, ProTyrLysCysProGluCysGlyLysSerPheSerGlnLysSerAspLeuValLysHisGlnAr~ ThrHisThrGly, has been designed with the use of a data base of 131 zinc finger sequences. Studies indicated that this peptide binds metal ions such as Zn2+ and Co2+ and folds in their presence. The affinity of this peptide for metal ions is greater than that demonstrated for any other zinc finger peptide characterized to date. Nuclear magnetic resonance studies revealed that the zinc complex of this peptide adopts a structure similar to that predicted and observed for other zinc finger domains. In addition, these studies led to the discovery that the latter of the histidine residues can be protonated and dissociated from the metal center with only local loss of structure. This histidine residue can also be replaced with a cysteine residue to yield a peptide that has a (Cy~)~(His) rather than a (Cys)*(Hi~)~ metal binding site. Introduction In recent years, a large class of proteins has been discovered that is characterized by the presence of one or more sequences that closely approximate the form (Tyr,Phe)-X-Cys-X2,4-CysX3-Phe-X,-Leu-X2-His-XS,,-His where X represents relatively variable amino acids.' Each of these sequences appears to form a small domain (often termed a 'zinc finger" domain2) organized around a zinc ion tetrahedrally coordinated by the cysteine and histidine residues. Where it has been determined, proteins that contain these domains have been shown to be specific nucleic acid binding proteins. Peptides corresponding to single domains of this type have been used to characterize the folding, metal-binding, and structural properties of such domains although these peptides have been found to be incapable of site-specific interactions with DNA.s1o It has been demonstrated that these peptides are largely unfolded in the absence of appropriate metal ions but fold to unique three-dimensional structures in their presence. These peptides have been very useful for structural studies of zinc finger domains by two-dimensional nuclear magnetic resonance (NMR) methods, as they are small (approximately 30 amino acids) yet fold in aqueous solution. Such N M R studies on several different zinc finger peptides have revealed structures that are quite similar to although not identical with one a n ~ t h e r . ~ ~ ~ , ' * ~ J ~ Moreover, the experimentally determined structures are strikingly similar to a structure predicted for the zinc finger domains on the basis of the observation of recurring metal-chelating substructures in crystallographically characterized metalloproteins.lI The structure, which consists of two strands and a helix, is shown schematically in Figure 1. The number of zinc finger sequences that are known is large and growing rapidly.' This is true for several reasons. First, as noted above, these sequences often occur as large tandem arrays, with as many as 37 such sequences being found in a single deduced protein sequence. Second, a large number of genes involved in developmental control are being cloned and sequenced and many of these are found to contain zinc finger sequences. Third, it has been possible to intentionally clone zinc finger encoding regions on the basis of hybridization to other zinc finger encoding DNA fragments. Finally, methods based on the polymerase chain reaction are being developed that may make detection and characterization of zinc finger sequences even more facile.I2 When the project described herein was initiated, a total of 131 sequences from 18 protein^,'^-^* tabulated in Table I, were known. *To whom correspondence should be addressed at The Johns Hopkins University School of Medicine. This sequence data base contains a great deal of information. As one illustration of this point, consider the distribution of proline (1) Berg, J. M. Annu. Rev. Biophys. Biophys. Chem. 1990,19,405 and (2) Miller, J.; McLachlan, A. D.; Klug, A. EMBO J . 1985, 4, 1609. (3) Frankel, A. D.; Berg, J. M.; Pabo, C. 0. Proc. Natl. Acad. Sci. U.S.A. (4) Pirraga, G.; Horvath, S. J.; Eisen, A,; Taylor, W. E.; Hood, L.; Young, references therein.