Artificial di-iron proteins: solution characterization of four helix bundles containing two distinct types of inter-helical loops

Abstract

Peptide-based models have an enormous impact for the development of metalloprotein models, as they seem appropriate candidates to mimic both the structural characteristics and reactivity of the natural systems. Through the de novo design of four-helix bundles, we developed the DF (Due Ferri) family of artificial proteins, as models of di-iron and di-manganese metalloproteins. The goal of our research is to elucidate how the electrostatic environment, polarity and solvent accessibility of the metal-binding site, influence the functional properties of di-iron proteins. The first two subsets of the DF protein family, DF1 and DF2, consist of two non-covalently associated helix-loop-helix motifs, which bind the di-metal cofactor near the center of the structure. The DF2 subset was designed to improve the properties of DF1: DF2 and DF2t have several changes in their sequences to improve solubility and metal ion access, as well as a change in the loop connecting the two helices. In order to evaluate how these changes affect the overall structure of the model proteins, we solved the NMR structures of the di-Zn(II) complexes of DF2 and DF2t, and compared these structures with those recently obtained from X-ray crystallography. Further, we examined the thermodynamic consequences associated with the mutations, by measuring the stability of DF2t in the presence of different metal ions, and comparing the results with the data already obtained for DF2. Taken together, analysis of all the data showed the importance of the turn conformation in the design and stability of four-helix bundle.

Prediction of the functional class of metal-binding proteins from sequence derived physicochemical properties by support vector machine approach

“…High-affinity metal binding sites in some proteins are located at sequence segments with specific amino acid composition[63], and specific sequence motifs have been used for predicting metal-binding proteins[64,65]. It was also found that polarity and solvent accessibility of the binding site influences the functional properties of metal-binding proteins[66]. Therefore, our prediction results are consistent with these experimental findings.…”

Engineering model proteins for Photosystem II function

“…Some of these simple designs have been used to bind hemes (Gibney and Dutton 2001), bacteriochlorophylls (Noy et al 2006), flavins (Sharp et al 1998), iron-sulphur clusters , quinones (Li et al 2005;Mennenga et al 2006;Hay et al 2007), amino acid radicals , carboxy-bridged dimetal complexes (Laplaza and Holm 2001;Maglio et al 2005;Wade et al 2006) and very recently the di-iron Fe 2 S 2 (CO) 6 complex which is a model of a hydrogenase catalytic site (A. Jones, P. L. Dutton, personal communication). In an elegant study of the de novo 4-helix bundle, heme-binding proteins, the spectroscopic and electrochemical properties of the bound heme could be varied to match the corresponding properties of the different natural heme proteins by systematically modifying the electrostatic interactions and protonation states of the residues neighbouring the heme (Shifman et al 2000;Gibney et al 2001).…”

Catalysis by Peptide‐Based Enzyme Models

“…This hairpin is repeated with pseudo-C 2 symmetry to form the second half of the site. DeGrado and Lombardi used this information to guide the design of a family of minimal models of diiron proteins -the Due-Ferri (DF) series [125][126][127][128][129][130][131][132]. The DF family is characterized by a four-helix bundle, which can be formed by individual helices, or by the dimerization of a helical hairpin, or by a single chain of amino acids.…”