Dynamic Similarity of the Unfolded States of Proteins L and G ²,‡ Vijay R. Singh, §,| Michaela Kopka, §,| Yujie Chen, § William J. Wedemeyer, §,⊥ and Lisa J. Lapidus* ,§ Departments of Physics and Astronomy and Biochemistry and Molecular Biology, Michigan State UniVersity, East Lansing, Michigan 48824 ReceiVed February 7, 2007; ReVised Manuscript ReceiVed June 15, 2007 ABSTRACT: The formation of specific intramolecular contacts has been studied under a range of denaturing conditions in single domains of the immunoglobulin-binding proteins L and G. Although they share no significant sequence similarity and have dissimilar folding pathways, the two domains have a similar native fold. Our measurements show that the rates of forming corresponding contacts in the unfolded states of both proteins are remarkably similar and even exhibit similar dependence on denaturant concentration. The unfolded proteins were modeled using Szabo, Schulten, and Schulten (SSS) theory as wormlike chains with excluded volume; when combined with our experimental data, the SSS analysis suggests that the unfolded state becomes uniformly more compact and less diffusive (i.e., rearranges more slowly) with decreasing denaturant concentrations. The unfolded state of proteins is often characterized as a completely random, non-interacting polymer (1). However, Levinthal proposed the paradox that if a polypeptide chain were allowed to randomly sample conformations with a residence time of less than a nanosecond, it would take astronomically long for any protein to fold (2). Initially researchers responded to this paradox by hypothesizing that parts of the protein would exhibit “flickering” local structure that would become stabilized upon associating with other such parts (3-5) and that the native fold would be quickly locked in once such assembly had led to a roughly correct topology (6). Later work on energy landscape theory suggested that the top of the folding funnel is flat and featureless, implying that the unfolded protein has no such “flickering” structure (7). Recent experimental work with a variety of techniques has challenged the idea that the unfolded states of proteins are generic non-interacting polymers. Much of the experimental investigations of unfolded proteins have focused on residual structure mea- sured by NMR under highly denaturing conditions (8, 9) which may give the misleading picture that these chains are not highly flexible. On the other hand, small-angle X-ray scattering (SAXS 1 ) and fluorescence resonance energy transfer (FRET) have been used to measure the overall radius of gyration of the unfolded protein, which generally increases with denaturant and can be modeled by a random walk chain (10-12). Thus a picture is emerging of the unfolded state that is still quite random but is biased by transient interactions (13). What has been missing is the time scale on which this random, transiently interacting chain reorganizes. New spectroscopic techniques have been developed to probe the internal dynamics of the unfolded state (14-16). In particular, the rate of forming a specific intramolecular contact can be monitored by the triplet quenching of a tryptophan side chain by a cysteine side chain (17). Un- structured peptides of (Ala-Gly-Gln) n repeats can be well modeled as wormlike chains with persistence lengths of 1-2 peptide bonds and an excluded volume radius of 2-4Å(18, 19). Comparing measurements of the same peptides in 0 and 6 M GdnHCl, the average chain dimension was smaller and the intramolecular diffusion coefficient was higher in aqueous solutions. A similar study of the cold shock protein of Thermotoga maritima also suggested that the unfolded state gradually became more compact as the denaturant concentra- tion was reduced, consistent with FRET observations (20); however, the degree of compaction was not well quantified (21). Another study using fluorescence quenching coupled with fluorescence correlation spectroscopy investigated in- tramolecular diffusion in intestinal fatty acid binding protein (IFABP) under a variety of solvent conditions and also observed a similar compaction (22). More recently, a single molecule FRET study on the natively unfolded yeast prion also observed compaction with decreasing denaturant con- ² M.K. was supported by Michigan State University start-up funds to W.J.W. V.R.S. and Y.C. were supported by Michigan State University start-up funds to L.J.L. The research of L.J.L. was supported in part by a Career Award at the Scientific Interface from the Burroughs Wellcome Fund. ‡ L.J.L. and W.J.W. designed the research; M.K. carried out the protein mutagenesis, expression, and purification; M.K. and V.R.S. measured the intramolecular contact rates and analyzed the data to obtain k R and kD; V.R.S., M.K., and Y.C. took the equilibrium unfolding data and assessed the mutant folding stability; L.J.L., V.R.S., and W.J.W. carried out the computational simulations using GPL software written by W.J.W. All the authors contributed significantly to writing the paper. * Corresponding author. Phone: (517) 355-9200 x2211. Fax: (517) 353-4500. E-mail: lapidus@msu.edu. § Departments of Physics and Astronomy. | These authors contributed equally to the research and can be considered coauthors. ⊥ Departments of Biochemistry and Molecular Biology. 1 Abbreviations: SAXS, small-angle X-ray scattering; FRET, fluo- rescence resonance energy transfer; Tris, tris(hydroxymethyl)ami- nomethane hydrochloride; GdnHCl, guanidinium hydrochloride; CD, circular dichroism; IPTG, isopropyl--D-thiogalactopyranoside; SSS, Szabo, Schulten, and Schulten. 10046 Biochemistry 2007, 46, 10046-10054 10.1021/bi700270j CCC: $37.00 © 2007 American Chemical Society Published on Web 08/08/2007