Filling small, empty protein cavities: structural and energetic consequences

Most proteins contain small cavities that can be filled by replacing cavity-lining residues by larger ones. Since shortening mutations in hydrophobic cores tend to destabilize proteins, it is expected that cavity-filling mutations may conversely increase protein stability. We have filled three small cavities in apoflavodoxin and determined by NMR and equilibrium unfolding analysis their impact in protein structure and stability. The smallest cavity (14 A3) has been filled, at two different positions, with a variety of residues and, in all cases, the mutant proteins are locally unfolded, their structure and energetics resembling those of an equilibrium intermediate of the thermal unfolding of the wild-type protein. In contrast, two slightly larger cavities of 20 A3 and 21 A3 have been filled with Val to Ile or Val to Leu mutations and the mutants preserve both the native fold and the equilibrium unfolding mechanism. From the known relationship, observed in shortening mutations, between stability changes and the differential hydrophobicity of the exchanged residues and the volume of the cavities, the filling of these apoflavodoxin cavities is expected to stabilize the protein by approximately 1.5 kcal mol(-1). However, both urea and thermal denaturation analysis reveal much more modest stabilizations, ranging from 0.0 kcal mol(-1) to 0.6 kcal mol(-1), which reflects that the accommodation of single extra methyl groups in small cavities requires some rearrangement, necessarily destabilizing, that lowers the expected theoretical stabilization. As the size of these cavities is representative of that of the typical small, empty cavities found in most proteins, it seems unlikely that filling this type of cavities will give rise to large stabilizations.     

Side-chain contributions to membrane protein structure and stability

The molecular forces that stabilize membrane protein structure are poorly understood. To investigate these forces we introduced alanine substitutions at 24 positions in the B helix of bacteriorhodopsin and examined their effects on structure and stability. Although most of the results can be rationalized in terms of the folded structure, there are a number of surprises. (1) We find a remarkably high frequency of stabilizing mutations (17%), indicating that membrane proteins are not highly optimized for stability. (2) Helix B is kinked, with the kink centered around Pro50. The P50A mutation has no effect on stability, however, and a crystal structure reveals that the helix remains bent, indicating that tertiary contacts dominate in the distortion of this helix. (3) We find that the protein is stabilized by about 1kcal/mol for every 38A(2) of surface area buried, which is quite similar to soluble proteins in spite of their dramatically different environments. (4) We find little energetic difference, on average, in the burial of apolar surface or polar surface area, implying that van der Waals packing is the dominant force that drives membrane protein folding.     

Determination of the second virial coefficient of bovine serum albumin under varying pH and ionic strength by composition-gradient multi-angle static light scattering

Composition-gradient multi-angle static light scattering (CG-MALS) is an emerging technique for the determination of intermolecular interactions via the second virial coefficient B₂₂. With CG-MALS, detailed studies of the second virial coefficient can be carried out more accurately and effectively than with traditional methods. In addition, automated mixing, delivery and measurement enable high speed, continuous, fluctuation-free sample delivery and accurate results. Using CG-MALS we measure the second virial coefficient of bovine serum albumin (BSA) in aqueous solutions at various values of pH and ionic strength of a univalent salt (NaCl). The systematic variation of the second virial coefficient as a function of pH and NaCl strength reveals the net charge change and the isoelectric point of BSA under different solution conditions. The magnitude of the second virial coefficient decreases to 1.13 x 10⁻⁵ ml*mol/g² near the isoelectric point of pH 4.6 and 25 mM NaCl. These results illuminate the role of fundamental long-range electrostatic and van der Waals forces in protein-protein interactions, specifically their dependence on pH and ionic strength. Electronic supplementary material The online version of this article (doi:10.1007/s10867-014-9367-7) contains supplementary material, which is available to authorized users.     

Orientation of the water molecules of hydration of human serum albumin

Through contact-angle measurements with a number of liquids, on layers of hydrated human serum albumin (HSA), built on anisotropic ultrafilter membranes, the apolar, Lifshitz-van der Waals surface tension component, as well as the polar, electron-acceptor and electron-donor parameters of the hydrated layers could be determined. From these data, it was found that the degree of orientation of the water molecules of hydration of HSA is 98% in the first layer of hydration and 30% of the second layer. The water molecules of hydration are oriented with the H atoms closest to, and the O atoms farthest from, the protein surface.     

Surface thermodynamic properties and chromatographic and salting-out behavior of IgA and other serum proteins

By means of contact angle determinations with two liquids, on hydrated as well as on dried protein layers, the long-range and the short-range contributions to the protein surface tensions, and from these the protein (G ₁₃₁) and the protein-ligand (G ₁₃₂) free energies of interaction in aqueous media, were determined. For human serum albumin (HSA), human IgG, and human IgA, the differences between G ₁₃₁ ^HYDRATED and G ₁₃₁ ^DRY were connected with the behavior of these proteins in low concentrations of (NH₄)₂SO₄ versus saturated (NH₄)₂SO₄ solutions. By interpolation, intermediate points are found that correlate well with the known salting-out properties of these three proteins. On the basis of the data, it is predicted that the precipitation of IgG by 1/3 saturated (NH₄)₂SO₄ is preventable, or reversible, by the admixture of 15% dimethylsulfoxide; both predictions are confirmed experimentally. From the G ₁₃₂ values found, it is shown that HSA and IgG should attach to phenyl ligands under physiological conditions, but that IgA is so hydrophilic that it only can adhere to phenyl ligands after partial dehydration brought about when admixed with 1 M (NH₄)₂SO₄. Closer analysis of the values obtained for the long-range and short-range components of the surface tensions of HSA, IgG, and IgA allow deeper insight into their functional, chemical, and physicochemical properties.     

Computational redesign of the SHV-1 beta-lactamase/beta-lactamase inhibitor protein interface

β-lactamases are enzymes that catalyze the hydrolysis of β-lactam antibiotics. β-lactamase/β-lactamase inhibitor protein (BLIP) complexes are emerging as a well characterized experimental model system for studying protein-protein interactions. BLIP is a 165 amino acid protein that inhibits several class A β-lactamases with a wide range of affinities: picomolar affinity for K1; nanomolar affinity for TEM-1, SME-1, and BlaI; but only micromolar affinity for SHV-1 β-lactamase. The large differences in affinity coupled with the availability of extensive mutagenesis data and high-resolution crystal structures for the TEM-1/BLIP and SHV-1/BLIP complexes make them attractive systems for the further development of computational design methodology. We used EGAD, a physics-based computational design program, to redesign BLIP in an attempt to increase affinity for SHV-1. Characterization of several of designs and point mutants revealed that in all cases, the mutations stabilize the interface by 10- to 1000-fold relative to wild type BLIP. The calculated changes in binding affinity for the mutants were within a mean absolute error of 0.87 kcal/mol from the experimental values, and comparison of the calculated and experimental values for a set of 30 SHV-1/BLIP complexes yielded a correlation coefficient of 0.77. Structures of the two complexes with the highest affinity, SHV-1/BLIP (E73M) and SHV-1/BLIP (E73M, S130K, S146M), are presented at 1.7 Å resolution. While the predicted structures have much in common with the experimentally determined structures, they do not coincide perfectly; in particular a salt bridge between SHV-1 D104 and BLIP K74 is observed in the experimental structures, but not in the predicted design conformations. This discrepancy highlights the difficulty of modeling salt bridge interactions with a protein design algorithm that approximates side chains as discrete rotamers. Nevertheless, while local structural features of the interface were sometimes miscalculated, EGAD is globally successful in designing complexes with increased affinity.     

Flow-induced conformational changes and phase behavior of aqueous poly-L-lysine solutions

Poly-L -lysine exists as an α-helix at high pH and a random coil at neutral pH. When the α-helix is heated above 27°C, the macromolecule undergoes a conformational transition to a β-sheet. In this study, the stability of the secondary structure of poly-L -lysine in solutions subjected to shear flow, at temperatures below the α-helix to β-sheet transition temperature, were examined using Raman spectroscopy and CD. Solutions initially in the α-helical state showed time-dependent increases in viscosity with shearing, rising as much as an order of magnitude. Visual observation and turbidity measurements showed the formation of a gel-like phase under flow. Laser Raman measurements demonstrated the presence of small amounts of β-sheet structure evidenced by the amide I band at 1666 cm⁻¹. CD measurements indicated that solutions of predominantly α-helical conformation at 20°C transformed into 85% α-helix and 15% β-sheet after being sheared for 20 min. However, on continued shearing the content of β-sheet conformation decreased. The observed phenomena were explained in terms of a “zipping-up” molecular model based on flow enhanced hydrophobic interactions similar to that observed in gel-forming flexible polymers. © 1998 John Wiley & Sons, Inc. Biopoly 45: 239–246, 1998     

Cell damage-induced conformational changes of the pro-apoptotic protein Bak in vivo precede the onset of apoptosis

Investigation of events committing cells to death revealed that a concealed NH2-terminal epitope of the pro-apoptotic protein Bak became exposed in vivo before apoptosis. This occurred after treatment of human Jurkat or CEM-C7A T-lymphoma cells with the mechanistically disparate agents staurosporine, etoposide or dexamethasone. The rapid, up to 10-fold increase in Bak-associated immunofluorescence was measured with epitope-specific monoclonal antibodies using flow cytometry and microscopy. In contrast, using a polyclonal antibody to Bak, immunofluorescence was detected both before and after treatment. There were no differences in Bak protein content nor in subcellular location before or after treatment. Immunofluorescence showed Bcl-xL and Bak were largely associated with mitochondria and in untreated cells they coimmunoprecipitated in the presence of nonioinic detergent. This association was significantly decreased after cell perturbation suggesting that Bcl-xL dissociation from Bak occurred on exposure of Bak's NH2 terminus. Multiple forms of Bak protein were observed by two dimensional electrophoresis but these were unchanged by inducers of apoptosis. This indicated that integration of cellular damage signals did not take place directly on the Bak protein. Release of proteins, including Bcl-xL, from Bak is suggested to be an important event in commitment to death.     

The Role of Solvent in Protein Folding and in Aggregation

We discuss features of the effect of solvent on protein folding andaggregation, highlighting the physics related to the particulate nature and the peculiar structure of the aqueous solvent, and the biological significance of interactions between solvent and proteins. To this purpose we use a generalized energy landscape of extended dimensionality. A closer look at the properties of solvent induced interactions and forces proves useful for understanding the physical grounds of `ad hoc' interactions and for devising realistic ways of accounting for solvent effects. The solvent has long been known to be a crucially important part of biological systems, and times appear mature for it to be adequately accounted for in the protein folding problem. Use of the extended dimensionality energy landscape helpseliciting the possibility of coupling among conformational changes and aggregation, such as proved by experimental data in the literature.     

Regulation of focal adhesion formation and filopodia extension by the cellular prion protein (PrPC)

While the prion protein (PrP) is clearly involved in neuropathology, its physiological roles remain elusive. Here, we demonstrate PrP functions in cell-substrate interaction in Drosophila S2, N2a and HeLa cells. PrP promotes cell spreading and/or filopodia formation when overexpressed, and lamellipodia when downregulated. Moreover, PrP normally accumulates in focal adhesions (FAs), and its downregulation leads to reduced FA numbers, increased FA length, along with Src and focal adhesion kinase (FAK) activation. Furthermore, its overexpression elicits the formation of novel FA-like structures, which require intact reggie/flotillin microdomains. Altogether, PrP modulates process formation and FA dynamics, possibly via signal transduction involving FAK and Src.     

On the relative importance of specific and non-specific approaches to oral microbial adhesion

Abstract In this paper, it is suggested that specificity and non-specificity in (oral) microbial adhesion are different expressions for the same phenomena. It is argued that the same basic, physico-chemical forces are responsible for so-called 'non-specific' and 'specific' binding and that from a physico-chemical point of view the distinction between the two is an artificial one. Non-specific interactions arise from Van der Waals and electrostatic forces and hydrogen bonding, and originate from the entire cell. A specific bond consists of a combination of the same type of Van der Waals and electrostatic forces and hydrogen bonding, now originating from highly localized chemical groups, which together form a stereo-chemical combination. The absence or presence of specific receptor sites on microbial cell surfaces must therefore be reflected in the overall, non-specific surface properties of cells as well. This point is illustrated by showing that glucanbinding lectins on mutans streptococcal strains may determine the pH dependence of the zeta potentials of these cells. When studying microbial adhesion, a non-specific approach may be better suited to explain adhesion to inert substrata, whereas a specific approach may be preferred in case of adhesion to adsorbed protein films. Adhesion is, however, not as important in plaque formation in the human oral cavity as is retention, because low shear force periods. during which adhesion presumably occurs, are followed by high shear force periods, during which adhering cells must withstand these detachment forces. Evidence is provided that such detachment will be through cohesive failure in the pellicle mass, the properties of which are conditioned by the overall, non-specific substratum properties. Therefore, in vivo plaque formation may be more readily explained by a non-specific approach.     

Solution structure of a tobacco lipid transfer protein exhibiting new biophysical and biological features

Plant lipid transfer proteins are small soluble extracellular proteins that are able to bind and transfer a variety of lipids in vitro. Recently, it has been proposed that lipid transfer proteins may play a key role in plant defence mechanisms, especially during the induction of systemic acquired resistance. However, very little is known about the proteins expressed in developing plants and tissues, since almost all the biophysical and structural data available to date on lipid transfer proteins originate from proteins present in storage tissues of monocot cereal seeds. In this paper, we report the structural and functional characteristics of a lipid transfer protein (named LTP1_1) constitutively expressed in young aerial organs of Nicotiana tabacum (common tobacco). The unlabelled and uniformly labelled proteins were produced in the yeast Pichia pastoris, and we determined the three-dimensional (3D) structure of LTP1_1 using nuclear magnetic resonance (NMR) spectroscopy and molecular modeling techniques. The global fold of LTP1_1 is very close to the previously published structures of LTP1 extracted from cereal seeds, including an internal cavity. However, the chemical shift variations of several NMR signals upon lipid binding show that tobacco LTP1_1 is able to bind only one LysoMyristoylPhosphatidylCholine (LMPC), while wheat and maize LTPs can bind either one or two. Titration experiments using intrinsic tyrosine fluorescence confirm this result not only with LMPC but also with two fatty acids. These differences can be explained by the presence in tobacco LTP1_1 of a hydrophobic cluster closing the second possible access to the protein cavity. This result suggests that LTP1 lipid binding properties could be modulated by subtle changes in a conserved global structure. The biological significance of this finding is discussed in the light of the signalling properties of the tobacco LTP1_1-jasmonate complex described elsewhere.     

Odorant binding and conformational changes of a rat odorant-binding protein

Odorant-binding proteins (OBPs) are lipocalins secreted in the nasal mucus of vertebrates, which convey odorants to their neuronal receptors. We compared the binding properties of a recombinant rat OBP (OBP-1F) using a set of six odorants of various chemical structures. We examined the binding properties by both fluorescent probe competition and isothermal titration calorimetry. OBP-1F affinity constants, in the micromolar range, varied by more than one order of magnitude and were roughly correlated to the odorant size. The observed binding stoichiometry was found to be around one odorant per dimer. Using tyrosine differential spectroscopy, the binding of ligand was shown to induce local conformational changes. A three-dimensional structure of OBP-1F, modelled using the known structure of aphrodisin as template, allowed us to suggest the location of the observed structural changes outside of the binding pocket. These results are consistent with one binding site located in one of the two beta-barrels of the OBP-1F dimer and a subtle conformational change correlated with binding of an odorant molecule, which hampers uptake of a second odorant by the other hydrophobic pocket.