Charges in Hydrophobic Environments: A Strategy for Identifying

Article pubs.acs.org/biochemistry

Charges in Hydrophobic Environments: A Strategy for Identifying Alternative States in Proteins Aaron C. Robinson,† Ananya Majumdar,‡ Jamie L. Schlessman,§ and Bertrand García-Moreno E*,† †

Department of Biophysics and ‡Biomolecular NMR Center, Johns Hopkins University, 3400 N Charles St, Baltimore, Maryland 21218, United States § Chemistry Department, United States Naval Academy, 572M Holloway Rd MS 9B, Annapolis, Maryland 21402, United States S Supporting Information *

ABSTRACT: In the V23E variant of staphylococcal nuclease, Glu-23 has a pKa of 7.5. At low pH, Glu-23 is neutral and buried in the hydrophobic interior of the protein. Crystal structures and NMR spectroscopy experiments show that when Glu-23 becomes charged, the protein switches into an open state in which strands β1 and β2 separate from the βbarrel; the remaining structure is unaffected. In the open state the hydrophobic interior of the protein is exposed to bulk water, allowing Glu-23 to become hydrated. This illustrates several key aspects of protein electrostatics: (1) The apparent pKa of an internal ionizable group can reflect the average of the very different pKa values (open ≈4.5, closed ≫7.5) sampled in the different conformational states. (2) The high apparent dielectric constant reported by the pKa value of internal ionizable group reflects conformational reorganization. (3) The apparent pKa of internal groups can be governed by large conformational changes. (4) A single charge buried in the hydrophobic interior of a protein is sufficient to convert what might have been a transient, partially unfolded state into the dominant state in solution. This suggests a general strategy for examining inaccessible regions of the folding landscape and for engineering conformational switches driven by small changes in pH. These data also constitute a benchmark for stringent testing of the ability of computational algorithms to predict pKa values of internal residues and to reproduce pH-driven conformational transitions of proteins.

T

not compatible with the hydrophobic interior of proteins.18 By engineering 100 variants of SNase with Lys, Arg, Asp, and Glu at internal positions, we have shown previously that specialized structural adaptations are not required for highly stable proteins to tolerate buried ionizable groups.20−23 The majority of SNase variants with internal ionizable groups are stable, and their crystal structures are almost indistinguishable from those of the parent protein.24,25 However, systematic optical and NMR spectroscopy experiments have shown that ionization of the internal groups can promote structural reorganization of varying amplitude.20,24−28 This reorganization is of interest as it can be the dominant influence that determines the pKa of the buried ionizable group and can also explain the origins of the relatively high apparent dielectric constant reported by internal ionizable groups.29 Conformational rearrangement in response to the creation of charge in hydrophobic environments is of interest for the mechanistic insight it can shed on energy transduction processes, for example, in ATPase and bacteriorhodopsin.30,31 Here we demonstrate that the introduction of charge into the hydrophobic interior of a protein by ionization of a buried ionizable group can convert what might have been only a

he folding of globular, water-soluble proteins is a highly cooperative process. Protein sequences have evolved such that their folding landscapes are usually dominated by fully folded and unfolded states;1,2 in general, partially unfolded proteins are substantially less stable than either fully folded or unfolded proteins and only transiently populated, if at all.3−7 An important consequence of this effective suppression of folding intermediates is that it minimizes the availability of aggregation-prone, partially folded species, thereby ensuring the success of the folding reaction.6−10 Despite their low occurrence, partially unfolded states are of great interest for the insight they contribute into the origins of folding cooperativity,11 folding mechanisms,12,13 functional roles in energy transduction processes, 14 and the genesis and propagation of aggregation and misfolding diseases.8−10 Nevertheless, direct structural characterization of partially unfolded proteins remains challenging because the equilibrium population of these species is usually insignificant.4 Here we present a general strategy for stabilizing partially unfolded states by introducing ionizable residues in the hydrophobic core of a protein and changing the pH to charge the buried residues. Ionizable residues buried in the hydrophobic protein interior are relatively rare. When present, they often play essential roles in energy transduction processes, usually involving H+ or e− transfer reactions15−17 or catalysis.18,19 Internal groups often titrate with anomalous pKa values because charged species are © XXXX American Chemical Society

Received: August 16, 2016 Revised: December 8, 2016

A

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

a partially unfolded (open) state in which the interior of the βbarrel, which forms the main hydrophobic core of SNase, is exposed to bulk solvent owing to the displacement of strands β1 and β2 by as much as 18 Å away from the rest of the βbarrel (Figure 1B). In the open state Glu-23 is fully exposed to bulk water (Figure 1D). The open state appears to be stabilized by crystal contacts: a single direct contact between the backbone N of Gly-20 and the side chain Oε2 of Glu-57 in a symmetry related molecule and several water mediated interactions between the backbone carbonyls of Ala-17, Asp19, and Asp-21 to the side chains of Glu-57, Tyr-54, and Asn138 (Figure S3). The exposure of the hydrophobic core of SNase to bulk water is presumably offset by gains made in hydrating the charged carboxylic group of Glu-23. Surprisingly, in the open state, the rest of the protein is unperturbed, even the region lining the exposed hydrophobic core. The Cα RMSD between the folded, closed structure in which Glu-23 is buried, and the structure of the open state in which it exposed to solvent, decreased from 2.6 to 0.3 Å when the β1-β2 region comprising residues 15−24was omitted. Structural Reorganization Monitored by NMR Spectroscopy. The crystal structure alone cannot confirm that the open state observed in Δ+NVIAGLA/V23E is the dominant population in solution at high pH (e.g., the open state could be a minor state, but more competent to crystallize under these conditions). NMR spectroscopy experiments were used to confirm that the open state observed in the crystal structure of the Δ+NVIAGLA/V23E variant at pH 8 is indeed the dominant population in solution. 1H−15N HSQC spectra of Δ+NVIAGLA/V23E acquired below pH 5.7 were consistent with the fully folded protein at acidic pH; specifically, all resonances in the β1-β2 region were present (Figure 2A). Titration in increments of 0.3−0.4 pH units yielded virtually identical behavior for all resonances between Δ+NVIAGLA and its V23E variant, except for the residues in the β1-β2 region, a short segment of β3 and the hydrophobic face of α1 in the V23E variant. These residues displayed significant line broadening with increasing pH and became undetectable between pH 6.6 and 7.0. The concerted nature of the broadening event and the spatial proximity of the residues involved suggest a conformational change driven by the ionization of Glu-23, which is the only ionizable group in this region of the protein that titrates over this pH range. Above pH 7.6 the effects of pH on the remaining resonances were comparable between Δ+NVIAGLA and its V23E variant showing no evidence of any further conformational change resulting from the ionization of Glu-23. Similar pH-dependent behavior was observed in the HSQC spectra of Δ+PHS/V23E (Figure 2B) confirming that the structural consequences observed were the result of the ionization of Glu-23 and not of any differences in the two SNase reference proteins. It was possible to observe the carboxyl group of Glu-23 directly in the Δ+NVIAGLA/V23E variant using the 13Cdetected CBCGCO experiment (Figure S4) that correlates the Cβ/γ of Asp/Glu with the side chain carboxyl (Cγ/δ) carbon(37) (Figure 3). In addition to the surface carboxylic groups observed in SNase, an upfield-shifted peak (Cγ = 31.4 ppm, Cδ = 174.2 ppm) absent in the spectra of Δ+NVIAGLA was observed in its V23E variant from pH 4.0 to 5.7. Standard (H)CC(CO)NH-TOCSY and two-dimensional (2D) CBCGCO-TOCSY experiments assigned this new peak unambiguously to Glu-23 in the protonated state.37−39 The

marginally populated, partially unfolded state into the dominant form in solution. This effect is driven by the imperative for charge created in the hydrophobic interior of a protein to gain access to water. This observation suggests an effective strategy useful to identify the structures and to measure the energetics of alternative states of proteins that populate generally inaccessible regions of the folding landscape. It also suggests a strategy for the design of protein conformational switches driven by small changes in pH. Finally, these results illustrate that the pKa values of buried ionizable groups can be governed less by the population of transient states that might constitute 1% or less of the total population,32,33 but rather by alternative structural states that become the dominant structural state of the protein under conditions where the internal ionizable group is charged.26,34−36

■

RESULTS Crystal Structures. Two highly stable variants of SNase were used in this study: Δ+PHS and Δ+NVIAGLA. The Δ+NVIAGLA variant differs from the previously described Δ+PHS form of SNase37 by four point mutations (D21N, T33V, T41I, and S59A) that increase the global thermodynamic stability of the protein, but otherwise have no major effects on the structure of the protein (Figures S1 and S2). The structure of Δ+PHS/V23E was solved previously at pH 6 to a resolution of 1.9 Å.38 Under these conditions, the protein adopts a fold indistinguishable from that of the reference protein, Δ+PHS (Figure 1A), thereby sequestering a neutral Glu-23 in a hydrophobic microenvironment (Figure 1C). Here we report the structure of Δ+NVIAGLA/V23E at pH 8 to a resolution of 1.4 Å (Table S1). In this structure the protein is in

Figure 1. Ribbon diagrams of Δ+PHS/V23E (A) at pH 6 (PDB ID: 3QOL) and Δ+NVIAGLA/V23E (B) at pH 8 (PDB ID: 3TME). Residues 15−24, which adopt alternate conformations in the two structures, are colored red. The microenvironment of Glu-23 in Δ+PHS (C) and Δ+NVIAGLA (D) backgrounds of SNase. Hydrophobic side chains in the main hydrophobic core are shown as balls and sticks in yellow. Polar side chains are colored white. Glu23 is colored blue. The internal water molecule observed in the Δ+PHS/V23E structure is represented as a cyan sphere. β-Strands 1 and 2 are shown as loops for clarity. B

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 2. 1H−15N HSQC spectra recorded for the Δ+NVIAGLA/V23E (A) and Δ+PHS/V23E (B) variants of SNase at 100 mM KCl and 25 °C. Spectra were recorded at pH 5.0 (red), 5.7 (orange), 6.6 (green), and 7.6 (blue) for the Δ+NVIAGLA/V23E variant. Seventeen residues (14−25, 33−34, 59, 62, and 66, highlighted in boxes) displayed line broadening between pH 6.6 and 7.6, which is not observed in the Δ+NVIAGLA protein used as a reference. Spectra were recorded at pH 5.1 (red), 5.8 (orange), 6.6 (green), and 7.5 (blue) for the Δ+PHS/V23E variant.

entering the region of the spectrum characteristic of resonances of charged, surface Glu side chains exposed to water.37 This provides further, clear evidence that in solution, at pH values above the apparent pKa of Glu-23, the opening of the β-barrel allowed the charged Glu-23 to contact bulk water. pKa of Glu-23. The pKa values of Glu residues sequestered in the hydrophobic interior of proteins are usually higher than the normal pKa of 4.5 of Glu in water.20 In a hydrophobic environment the equilibrium between the charged and the neutral form of Glu is shifted in favor of the neutral form, resulting in an anomalous, elevated pKa. ΔΔG°H20 (pH) = ΔΔG°mut − RT ⎛ 1 + e z 2.303(pKaD− pH) ⎞ ⎟⎟ ln⎜⎜ N ⎝ 1 + e z 2.303(pKa − pH) ⎠

Figure 3. Asp/Glu selective, Asn/Gln suppressed, CBCGCO spectra of the Δ+NVIAGLA/V23E variant of SNase at pH 5.0, where Glu-23 is neutral (red), and pH 8.5 where it is charged (blue). The typical chemical shift ranges for surface Asp Cβ-Cγ and Glu Cγ-Cδ cross peaks are highlighted by boxes. The resonances of Glu-23 at both pH values are indicated by an arrow with the same color scheme as above. The Glu-23 cross peak at pH 5.0 was folded into the spectrum in the indirect dimension; the aliased chemical shift of the 13Cγ is 31.4 ppm.

(1)

The apparent pKa of the internal Glu-23 was determined by fitting eq 1 to the pH-dependence of the difference in thermodynamic stability between the proteins with and without Glu-23 (Figure 4 and Table S2). The rightmost term in this expression describes the pH-dependent free energy associated with differences in the pKa of Glu-23 in the denatured state (pKaD), where Glu-23 is in water, and in the native state (pKaN), where it is buried in the hydrophobic interior, at least while the group is in the neutral state. ΔΔG°mut is the pHindependent free energy difference that accounts for insertion of the ionizable residue into the reference protein under conditions where the ionizable moiety is neutral (i.e., below pH 4.5). In the Δ+PHS/V23E variant, Glu-23 has a pKa value of 7.1 ± 0.2.20 The pKa of Glu-23 was remeasured in the Δ+NVIAGLA/V23E variant to correlate directly with the structural data obtained with this protein. In the Δ+NVIAGLA background, the pKa of Glu-23 was 7.5 ± 0.2.

exact chemical origins of this large upfield shift are not known, but they appear to be characteristic of protonated 13Cγ/δ nuclei of Asp/Glu side chains embedded in a hydrophobic environment. Similar chemical shifts were observed for Glu-23 and Asp-66 residues in the Δ+PHS/V23E and Δ+PHS/V66D variants of SNase at pH values below the pKa of Glu-23 and Asp-66, respectively.24,38 Between pH 5.7 and 7.6, the resonance for Glu-23 was not observable in any of the CBCGCO spectra, presumably because of exchange broadening as a result of both titration of the carboxyl side chain itself and the concomitant conformational change in the β1-β2 elements of the protein. Above pH 7.6, the resonance became observable again, albeit shifted significantly in both the Cγ and Cδ dimensions (Cγ = 36.1 ppm, Cδ = 183.9 ppm at pH 8.5), C

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

protein by minimizing exposure of the hydrophobic core to bulk solution. The elimination of internal ionizable groups diminishes the probability of a protein encountering pH regimes in which the ionization of the internal group would lead to the exposure of hydrophobic surfaces that could promote aggregation. It is not obvious why the ionization of Glu-23 led to the particular reorganization of β1-β2 and not some other type of structural response. Internal ionizable groups have been used previously to adversely affect protein folding through the preferential stabilization of folding intermediates.41 One unique aspect of the present study is that the open conformation observed in SNase is not part of its known folding pathway.42 On the contrary, the β1-β3 meander is thought to be one of earliest elements to form during folding and constitutes a scaffold upon which the rest of the protein may fold. The structural reorganization observed above pH 7 suggests that, at least at neutral or basic pH, the folding pathway of SNase is likely altered. Although the elucidation of details of the folding mechanism is beyond the scope of this study, we surmise that early formation of the β1-β3 meander in WT SNase may be favored thermodynamically, but is not necessary for proper folding. Furthermore, because the open state does not exist simply as an intermediate along the folding pathway, it is not merely a low population state observable only as an “invisible” state in NMR relaxation experiments.4,5 Instead, the open form of SNase achieved through the ionization of Glu-23 is the dominant equilibrium species in solution at high pH and the conformational transition triggered by ionization of Glu-23 can be described in molecular detail, with full accounting of the pKa values of the group that drives the transition (i.e., Glu-23). Visual inspection of the structure of SNase suggests several possible alternative pathways for structural reorganization that could facilitate contact between the ionizable moiety of Glu-23 and bulk water (e.g., release of β1 alone or increased water penetration into the protein interior). The probability that any particular conformational state other than the native state is populated under native conditions is governed by the free energy difference between the native state and that alternative state. Owing to the anomalous pKa of Glu-23, as pH is increased, the energy gap between the closed and the open states decreases in the Δ+NVIAGLA/V23E variant relative to any other alternative open state or the fully unfolded protein. This drives the protein to a new lowest energy state where β1β2 are released from the structure. If in the open state Glu-23 is assumed to titrate with the normal pKa of 4.5 for Glu in water, then ΔGunfolding for β1-β2 may be estimated as 1.36|ΔpKa|, where ΔpKa is the difference between the normal pKa of 4.5 and the apparent pKa of 7.5. This ΔGunfolding of 4.1 kcal/mol is substantial and consistent with the disruption of the hydrophobic core of the protein observed in the crystal structure. These results also suggest that a systematic study using ionizable groups at other sites could lead to an unprecedented mapping of the conformations of accessible states onto the entire energy landscape of SNase, not just folding intermediates, with exact descriptions of the energetics and corresponding structures. Our ability to describe with atomic detail the conformational transition coupled to the ionization of Glu-23 constitutes the clearest evidence to date that the pKa values of buried ionizable groups can be governed by the propensity of a protein to undergo conformational transitions into states where the buried group gains access to water. This also provides clear evidence

Figure 4. pH dependence of ΔG°H2O for Δ+NVIAGLA (black triangles) and Δ+PHS (red triangles) and the respective V23E variants, (black squares) and (red squares), respectively. The solid lines are meant only to guide the eye. The difference in thermodynamic stability of the V23E variants in the Δ+NVIAGLA (black circles) and Δ+PHS (red circles), backgrounds, respectively. The dashed lines describe fits to eq 1. Errors are on the order of the symbol size.

■

DISCUSSION The pKa of Glu-23 in SNase is anomalous because it is buried in a highly hydrophobic microenvironment that is neither sufficiently polar nor polarizable to compensate for the loss of interactions with water. For this reason the equilibrium between the neutral and charged forms of the carboxylic group is shifted in favor of the neutral state, resulting in an elevated pKa value. The precise value of the pKa of Glu-23 appears to be governed by the energetics of the transition between the open and closed states. Having observed the presence of an open state of SNase under conditions of pH where Glu-23 is charged, it becomes possible to ascertain that the measured, apparent pKa of 7.5 for Glu-23 reflects a population weighted average of the pKa of Glu-23 in two different conformational states of the protein. In the open state, the pKa can be assumed to be close to 4.5, the normal pKa of Glu in water. Although the pKa of Glu-23 when fully buried in the closed state cannot be measured directlyits ionization triggers the reorganization of β1-β2its value must be considerably greater than the apparent pKa of 7.5. The description of apparent pKa values of ionizable residues as averages in different conformational states was proposed originally by Whitten et al.40 and more recently in the context of constant-pH molecular dynamics (CpHMD) simulations by Roitberg and colleagues.36 Ionizable groups buried in the hydrophobic interior of proteins usually play essential functional roles. The relatively low abundance of internal ionizable groups suggests that when these groups are not needed for function they are eliminated through evolution. In a laboratory setting, it has been possible to introduce Lys, Arg, Asp, and Glu at 25 internal positions in SNase.20−23 All 100 variants of SNase are less stable than the parent protein. This suggests that internal ionizable groups not essential for function are eliminated to minimize any deleterious impact on thermodynamic stability. The demonstration that the ionization of an internal group can trigger structural reorganization, exposing the hydrophobic core of a protein to solvent, suggests that evolution may also eliminate internal ionizable groups to enhance the solubility of the D

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

exhibiting only a slightly perturbed pKa value. This ensures that a mutation that eliminates any single residue that contributes to pH-sensing does not eliminate the potential to switch. The case of V23E illustrates how the anomalous pKa value of an internal ionizable group can be used to engineer artificial pH-sensing domains where a single residue introduced by mutagenesis into an environment that leads to a highly anomalous pKa value can drive substantial conformational rearrangements of the protein.

that the relatively high dielectric constants needed in continuum electrostatics methods to reproduce pKa values of the buried ionizable residues can reflect water penetration, and perhaps more importantly, the conformational change coupled to the ionization reaction.20,21,43 It should be emphasized that when the protein reorganizes in response to the ionization of a buried group, allowing it to gain access to water, the protein dielectric constant needed to reproduce the measured pKa value has no physical meaning; it is merely an ad hoc parameter used to correct a flawed, static model that does not reflect the ability of the protein to undergo conformational reorganization in response to the ionization process.44 The observation of conformational reorganization coupled to the ionization of Glu-23 offers an unparalleled opportunity to test the ability of structure-based electrostatics calculations to reproduce details of conformational reorganization coupled to the ionization of a buried ionizable group. Specifically, these data will be useful to test CpHMD methods designed to compute coupling between changes in conformation and ionization state. These methods are already being used to identify the structural origins of anomalous apparent pKa values of buried ionizable groups. In some cases, the calculations show that apparent pKa values correspond to the pH at which the transition between “open” and “closed” states of proteins take place,36 consistent with the behavior observed in the V23E variant of SNase. In other tests of CpHMD methods with SNase itself, the anomalous, macroscopic pKa values of internal ionizable groups were interpreted as an equilibrium between “open” and “closed” states with microscopic pKa values of the model compound in water and an organic solvent, respectively.32 Although these calculations attempt to capture proton-driven conformational reorganization, they seem to fail to accurately reproduce the structural changes that have been documented experimentally.29 It appears that such calculations are not capable of sampling conformational reorganization properly; by exaggerating the initial relaxation of structures and allowing water penetration to solvate the buried ionizable side chains, these methods miss the larger structural rearrangements of the protein backbone. Indeed, CpHMD methods are not necessary to observe such effects. Even standard MD simulations are sufficient to conclude that the ionization of internal residues can promote water penetration into the protein interior. In general, the results of these simulations are at odds with the types of structural reorganization observed in crystal structures and NMR spectroscopy experiments. In the specific case of Glu-23, accurate simulation of the pKa of Glu-23 will require faithful reproduction of the conformational changes observed in the crystal structure concomitant with ionization of Glu-23, which are fully consistent with the exact locations of conformational change observed in solution with NMR spectroscopy. The NMR data also have information about the time scale (ms) of this reorganization, consistent with the exchange processes that give rise to the observed broadening. Reproducing these experimental observations will challenge the accuracy of both force fields and sampling methodology. The experimentally observed coupling between conformational state and the charge state of Glu-23 implies that the V23E protein effectively acts as a pH-sensing switch that can interconvert between open and closed conformations in response to a change in pH in the physiological range. Naturally occurring pH-sensing switches, such as hemoglobin45 and the hemagglutinin protein of the influenza virus,46 tend to encode the switching potential across many residues, each

■

MATERIALS AND METHODS The V23E variants were engineered using two highly stable variants of SNase, Δ+PHS and Δ+NVIAGLA, using protein purification and molecular cloning techniques described previously.29 The crystal structure of Δ+NVIAGLA/V23E was solved by molecular replacement phasing using Δ+PHS (3BDC37), with residues 14−25 omitted, as a search model and iterative model building was performed using Refmac547 and COOT.48 NMR data were collected using a Bruker Avance II600 equipped with a cryoprobe and spectra were processed and analyzed with NMRPipe49 and Sparky,50 respectively. Thermodynamic stabilities were measured by monitoring the intrinsic fluorescence of Trp-140 and data were collected with an Aviv Automatic Titrating Fluorometer 105 (Lakewood, NJ) as described previously.29

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.6b00843. Table of crystal structure collection and refinement statistics, table of pH dependence of free energy for Δ+NVIAGLA/V23E, substitutions for stabilized forms of staphylococcal nuclease, far-UV spectra, crystal contacts stabilizing open conformation of Δ+NVIAGLA/V23E, and NMR pulse sequence for CBCGCO experiment: Figure S1 (PDF), Figure S2 (PDF), Figure S3 (PDF), Figure S4 (PDF), Table S1 (PDF), Table S2 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: (410)-516-4497 Fax: (410) 516-4118. ORCID

Aaron C. Robinson: 0000-0001-6410-7147 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by NIH Grant GM 061597 to B.G.M.E. NMR spectroscopy experiments were performed in the BioNMR facility at Johns Hopkins University. X-ray data were collected at beamline X25, National Synchrotron Light Source (NSLS), Brookhaven National Laboratory. Use of the NSLS was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-98CH10886.

■

ABBREVIATIONS SNase, staphylococcal nuclease; NMR, nuclear magnetic resonance; HSQC, heteronuclear single quantum coherence E

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

■

(22) Cannon, B. (2008) Thermodynamic Consequences of Substitution of Internal Positions in Proteins with Polar and Ionizable Residues, Johns Hopkins University, Baltimore, MD. (23) Harms, M. J., Schlessman, J. L., Sue, G. R., and García-Moreno E, B. (2011) Arginine Residues at Internal Positions in a Protein Are Always Charged. Proc. Natl. Acad. Sci. U. S. A. 108, 18954−18959. (24) Chimenti, M. S., Khangulov, V. S., Robinson, A. C., Heroux, A., Majumdar, A., Schlessman, J. L., and García-Moreno E, B. (2012) Structural Reorganization Triggered by Charging of Lys Residues in the Hydrophobic Interior of a Protein. Structure 20, 1071−1085. (25) Harms, M. J., Castañeda, C. A., Schlessman, J. L., Sue, G. R., Isom, D. G., Cannon, B. R., and García-Moreno E, B. (2009) The pKa values of acidic and basic residues buried at the same internal location in a protein are governed by different factors. J. Mol. Biol. 389, 34−47. (26) Chimenti, M. S., Castañeda, C. A., Majumdar, A., and GarcíaMoreno E, B. (2011) Structural Origins of High Apparent Dielectric Constants Experienced by Ionizable Groups in the Hydrophobic Core of a Protein. J. Mol. Biol. 405, 361−377. (27) Karp, D. A., Stahley, M. R., and García-Moreno E, B. (2010) Conformational Consequences of Ionization of Lys, Asp, and Glu Buried at Position 66 in Staphylococcal Nuclease. Biochemistry 49, 4138−4146. (28) Richman, D. E., Majumdar, A., and García-Moreno E, B. (2015) Conformational Reorganization Coupled to the Ionization of Internal Lys Residues in Proteins. Biochemistry 54, 5888−5897. (29) García-Moreno E, B., Dwyer, J. J., Gittis, A. G., Lattman, E. E., Spencer, D. S., and Stites, W. E. (1997) Experimental measurement of the effective dielectric in the hydrophobic core of a protein. Biophys. Chem. 64, 211−224. (30) Fillingame, R. H., Angevine, C. M., and Dmitriev, O. Y. (2003) Mechanics of coupling proton movements to c-ring rotation in ATP synthase. FEBS Lett. 555, 29−34. (31) Vorburger, T., Ebneter, J. Z., Wiedenmann, A., Morger, D., Weber, G., Diederichs, K., Dimroth, P., and von Ballmoos, C. (2008) Arginine-induced conformational change in the ca-subunit interface of ATP synthase. FEBS J. 275, 2137−2150. (32) Goh, G. B., Laricheva, E. N., and Brooks, C. L. (2014) Uncovering pH-dependent Transient States of Proteins with Buried Ionizable Residues. J. Am. Chem. Soc. 136, 8496−8499. (33) Shi, C., Wallace, J. A., and Shen, J. K. (2012) Thermodynamic Coupling of Protonation and Conformational Equilibria in Proteins: Theory and Simulation. Biophys. J. 102, 1590−1597. (34) Harms, M. J., Schlessman, J. L., Chimenti, M. S., Sue, G. R., Damjanović, A., and García-Moreno E, B. (2008) A Buried Lysine That Titrates with a Normal pKa: Role of Conformational Flexibility at the Protein-Water Interface as a Determinant of pKa values. Protein Sci. 17, 833−845. (35) Karp, D. A., Gittis, A. G., Stahley, M. R., Fitch, C. A., Stites, W. E., and García-Moreno E, B. (2007) High Apparent Dielectric Constant Inside a Protein Reflects Structural Reorganization Coupled to the Ionization of an Internal Asp. Biophys. J. 92, 2041−2053. (36) Di Russo, N. V., Estrin, D. A., Martí, M. A., and Roitberg, A. E. (2012) pH-Dependent Conformational Changes in Proteins and Their Effect on Experimental pKas: The Case of Nitrophorin 4. PLoS Comput. Biol. 8, e1002761. (37) Castañeda, C. A., Fitch, C. A., Majumdar, A., Khangulov, V., Schlessman, J. L., and García-Moreno E, B. (2009) Molecular determinants of the pKa values of Asp and Glu residues in staphylococcal nuclease. Proteins: Struct., Funct., Genet. 77, 570−588. (38) Robinson, A. C., Castañeda, C. A., Schlessman, J. L., and GarcíaMoreno E, B. (2014) Structural and thermodynamic consequences of burial of an artificial ion pair in the hydrophobic interior of a protein. Proc. Natl. Acad. Sci. U. S. A. 111, 11685−11690. (39) Grzesiek, S., Anglister, J., and Bax, A. (1993) Correlation of Backbone Amide and Aliphatic Side-Chain Resonances in 13C/15NEnriched Proteins by Isotropic Mixing of 13C Magnetization. J. Magn. Reson., Ser. B 101, 114−119. (40) Whitten, S. T., García-Moreno E, B., Hilser, V. J., and DeGrado, W. F. (2005) Local Conformational Fluctuations Can Modulate the

REFERENCES

(1) Anfinsen, C. B. (1973) Principles that govern folding of protein chains. Science 181, 223−230. (2) Rose, G. D., Fleming, P. J., Banavar, J. R., and Maritan, A. (2006) A backbone-based theory of protein folding. Proc. Natl. Acad. Sci. U. S. A. 103, 16623−16633. (3) Bai, Y., Sosnick, T., Mayne, L., and Englander, S. (1995) Protein folding intermediates: native-state hydrogen exchange. Science 269, 192−197. (4) Bouvignies, G., Vallurupalli, P., Hansen, D. F., Correia, B. E., Lange, O., Bah, A., Vernon, R. M., Dahlquist, F. W., Baker, D., and Kay, L. E. (2011) Solution structure of a minor and transiently formed state of a T4 lysozyme mutant. Nature 477, 111−114. (5) Hansen, D. F., Vallurupalli, P., and Kay, L. E. (2008) Using relaxation dispersion NMR spectroscopy to determine structures of excited, invisible protein states. J. Biomol. NMR 41, 113−120. (6) Westerheide, S. D., and Morimoto, R. I. (2005) Heat Shock Response Modulators as Therapeutic Tools for Diseases of Protein Conformation. J. Biol. Chem. 280, 33097−33100. (7) Dul, J. L., Davis, D. P., Williamson, E. K., Stevens, F. J., and Argon, Y. (2001) Hsp70 and Antifibrillogenic Peptides Promote Degradation and Inhibit Intracellular Aggregation of Amyloidogenic Light Chains. J. Cell Biol. 152, 705−716. (8) Neudecker, P., Robustelli, P., Cavalli, A., Walsh, P., Lundström, P., Zarrine-Afsar, A., Sharpe, S., Vendruscolo, M., and Kay, L. E. (2012) Structure of an Intermediate State in Protein Folding and Aggregation. Science 336, 362−366. (9) Smith, D. P., Jones, S., Serpell, L. C., Sunde, M., and Radford, S. E. (2003) A Systematic Investigation into the Effect of Protein Destabilisation on Beta 2-Microglobulin Amyloid Formation. J. Mol. Biol. 330, 943−954. (10) Chiti, F., Taddei, N., Bucciantini, M., White, P., Ramponi, G., and Dobson, C. M. (2000) Mutational analysis of the propensity for amyloid formation by a globular protein. EMBO J. 19, 1441−1449. (11) Baldwin, R. L. (2008) The Search for Folding Intermediates and the Mechanism of Protein Folding. Annu. Rev. Biophys. 37, 1−21. (12) Englander, S. W. (2000) Protein Folding Intermediates and Pathways Studied by Hydrogen Exchange. Annu. Rev. Biophys. Biomol. Struct. 29, 213−238. (13) Sosnick, T. R., and Barrick, D. (2011) The folding of single domain proteins  have we reached a consensus? Curr. Opin. Struct. Biol. 21, 12−24. (14) Ihee, H., Rajagopal, S., Šrajer, V., Pahl, R., Anderson, S., Schmidt, M., Schotte, F., Anfinrud, P. A., Wulff, M., and Moffat, K. (2005) Visualizing reaction pathways in photoactive yellow protein from nanoseconds to seconds. Proc. Natl. Acad. Sci. U. S. A. 102, 7145− 7150. (15) Lanyi, J. K. (2006) Proton transfers in the bacteriorhodopsin photocycle. Biochim. Biophys. Acta, Bioenerg. 1757, 1012−1018. (16) Pisliakov, A. V., Sharma, P. K., Chu, Z. T., Haranczyk, M., and Warshel, A. (2008) Electrostatic basis for the unidirectionality of the primary proton transfer in cytochrome c oxidase. Proc. Natl. Acad. Sci. U. S. A. 105, 7726−7731. (17) von Ballmoos, C., Wiedenmann, A., and Dimroth, P. (2009) Essentials for ATP Synthesis by F1F0 ATP Synthases. Annu. Rev. Biochem. 78, 649−672. (18) Harris, T. K., and Turner, G. J. (2002) Structural Basis of Perturbed pKa Values of Catalytic Groups in Enzyme Active Sites. IUBMB Life 53, 85−98. (19) Gutteridge, A., and Thornton, J. M. (2005) Understanding nature’s catalytic toolkit. Trends Biochem. Sci. 30, 622−629. (20) Isom, D. G., Castañeda, C. A., Cannon, B. R., Velu, P. D., and García-Moreno E, B. (2010) Charges in the hydrophobic interior of proteins. Proc. Natl. Acad. Sci. U. S. A. 107, 16096−16100. (21) Isom, D. G., Castañeda, C. A., Cannon, B. R., and GarcíaMoreno E, B. (2011) Large shifts in pKa values of lysine residues buried inside a protein. Proc. Natl. Acad. Sci. U. S. A. 108, 5260−5265. F

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry Coupling between Proton Binding and Global Structural Transitions in Proteins. Proc. Natl. Acad. Sci. U. S. A. 102, 4282−4287. (41) Zheng, Z., and Sosnick, T. R. (2010) Protein Vivisection Reveals Elusive Intermediates in Folding. J. Mol. Biol. 397, 777−788. (42) Wang, Y., and Shortle, D. (1995) Equilibrium folding pathway of staphylococcal nuclease: identification of the most stable chainchain interactions by NMR and CD spectroscopy. Biochemistry 34, 15895−15905. (43) Dwyer, J. J., Gittis, A. G., Karp, D. A., Lattman, E. E., Spencer, D. S., Stites, W. E., and García-Moreno E, B. (2000) High Apparent Dielectric Constants in the Interior of a Protein Reflect Water Penetration. Biophys. J. 79, 1610−1620. (44) Schutz, C. N., and Warshel, A. (2001) What are the dielectric constants of proteins and how to validate electrostatic models? Proteins: Struct., Funct., Genet. 44, 400−417. (45) Perutz, M. F. (1970) Stereochemistry of Cooperative Effects in Haemoglobin: Haem-Haem Interaction and the Problem of Allostery. Nature 228, 726−734. (46) Wiley, D. C., and Skehel, J. J. (1987) The Structure and Function of the Hemagglutinin Membrane Glycoprotein of Influenza Virus. Annu. Rev. Biochem. 56, 365−394. (47) Bailey, S. (1994) The CCP4 suite - programs for protein crystallography. Acta Crystallogr., Sect. D: Biol. Crystallogr. 50, 760− 763. (48) Emsley, P., and Cowtan, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 60, 2126−2132. (49) Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) NMRPipe: A multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6, 277−293. (50) Goddard, T., and Kneller, D. (2008) SPARKY 3, San Fransisco University of California.

G

DOI: 10.1021/acs.biochem.6b00843 Biochemistry XXXX, XXX, XXX−XXX