Probing Denatured State Conformational Bias in a Three-Helix Bundle, UBA(2), Using a Cytochrome c Fusion Protein

Abstract

Previous work with the four-helix-bundle protein cytochrome c' from Rhodopseudomonas palustris using histidine-heme loop formation methods revealed fold-specific deviations from random coil behavior in its denatured state ensemble. To examine the generality of this finding, we extend this work to a three-helix-bundle polypeptide, the second ubiquitin-associated domain, UBA(2), of the human DNA excision repair protein. We use yeast iso-1-cytochrome c as a scaffold, fusing the UBA(2) domain at the N-terminus of iso-1-cytochrome c. We have engineered histidine into highly solvent accessible positions of UBA(2), creating six single histidine variants. Guanidine hydrochloride denaturation studies show that the UBA(2)-cytochrome c fusion protein unfolds in a three-state process with iso-1-cytochrome c unfolding first. Furthermore, engineered histidine residues in UBA(2) strongly destabilize the iso-1-cytochrome c domain. Equilibrium and kinetic histidine-heme loop formation measurements in the denatured state at 4 and 6 M guanidine hydrochloride show that loop stability decreases as the size of the histidine-heme loop increases, in accord with the Jacobson-Stockmayer equation. However, we observe that the His27-heme loop is both more stable than expected from the Jacobson-Stockmayer relationship and breaks more slowly than expected. These results show that the sequence near His27, which is in the reverse turn between helices 2 and 3 of UBA(2), is prone to persistent interactions in the denatured state. Therefore, consistent with our results for cytochrome c', this reverse turn sequence may help to establish the topology of this fold by biasing the conformational distribution of the denatured state.

Denatured State Conformational Biases in Three-Helix Bundles Containing Divergent Sequences Localize near Turns and Helix Capping Residues

“…After purification (QIAquick PCR purification kit, Qiagen), the PCR product was cloned into pRbs_BTR1fuse using the unique EcoRI and NgoMIV restriction sites of the vector, as described previously. 43 The resulting pRbs_BTR1(UBA1_Cc) plasmid was sequenced (Genomics Core Facility, University of Montana), confirming the presence of the fusion protein gene.…”

“…The first ubiquitin-associated domain, UBA (1), which comprises residues 162−204 of HHR23A, was inserted between Phe(−3) and Lys(−2) of the yeast iso-1-Cytc domain using the unique EcoRI and NgoMIV restriction sites of the pRbs_BTR1fuse vector. 43 The pGEX-2T plasmid containing the UBA(1) gene was a gift from Juli Feigon. 74 The gene was amplified by polymerase chain reaction (PCR) with primers that put EcoRI and NgoMIV restriction sites, respectively, at the 5′ and 3′ ends of the gene (Table S1).…”

“…Expression and purification of pWT and single histidine variants of UBA(1)−iso-1-Cytc used previously described methods. 43 In brief, the pRbs_BTR1(UBA1_Cc) plasmid containing the desired UBA(1)−iso-1-Cytc variant was transformed into competent phage T1-resistant BL21(DE3) Escherichia coli cells (New England Biolabs) following the manufacturer's recommended method. After suspending transformed cells from LB medium agar plates containing ampicillin into 0.5 mL of sterile 2xYT media, the cell suspension was used to inoculate 1 L of 2xYT cultures in 2.8 L Fernbach flasks containing 100 mg of ampicillin.…”

“…42 Our lab also has shown that thermodynamically significant biases occur near turn sequences of helical bundles. 4,17,43,44 Recent work on the DSE has focused on the denaturant dependence of R g as measured by small-angle X-ray scattering (SAXS) versus end-to-end distances, R ee , as measured by fluorescence resonance energy transfer (FRET). 45−55 In general, both SAXS and FRET agree that ν ∼ 0.6 in 6 M guanidine hydrochloride (GuHCl) and 8 M urea and that there is a gradual decrease in ν as denaturant concentration is lowered, with more pronounced effects occurring below ∼2−3 M denaturant concentration.…”

“…The kinetics of loop breakage for this same subset of His−heme loops in the denatured state is also slower than expected for a random coil. 66 The portions of the primary structure containing this thermodynamic and kinetic bias in the DSE localize near interhelical turns or loops in the tertiary structures of Cytc′ and UBA(2), 4,43,44 suggesting that residues near interhelical turns form unusually stable contacts in the DSE in these folds. Furthermore, molecular dynamics (MD) simulations of Cytc′ under denaturing conditions revealed that the persistence of turns in this four-helix bundle is stabilized by hydrophobic side chains, particularly aromatic side chains.…”