Search

The conformational changes of membrane proteins are crucial to their function and usually lead to fluctuations in the electrostatic environment of the protein surface. A very effective way to quantify these changes is by calculating the pK a values of the protein's titratable residues, which can be regarded as electrostatic probes. To achieve this, we need to take advantage of the fast and reliable pK a calculators developed for globular proteins and adapt them to include the explicit effects of membranes. Here, we provide a detailed linear response approximation protocol that uses our own software (PypKa) to calculate reliable pK a values from short MD simulations of membrane proteins.

Gelation is an efficient way to fabricate fibrous protein materials. Briefly, it is an aggregation process where protein molecules assembly from a random structure into an organized structure such as nanofibrillar networks. According to their mechanisms, the fibrous proteins gelation can be classified into physical gelation and chemical gelation. The physical gelation is formed by the conformational transformation of fibroin proteins, which can be triggered by temperature, concentration, pH, or shear force. On the other hand, the chemical gelation is to cross-link fibrous proteins through chemical and/or enzymatic reactions. In this chapter, we summarize the protocols for preparing fibrous protein hydrogels, including both physical and chemical methods. The mechanisms of these gelation methods are also highlighted.

Fibrous proteins are promising bioinks for three-dimensional printing techniques to fabricate sophisticated structures that find applications in both biomedical engineering and materials science. The critical point of manufacturing these fibrous protein inks is to adjust the cross-linking and rheology properties of proteins that matching the requirements of various printing techniques. In recent years, 3D printing techniques such as extrusion-based printing, droplet-based printing, and light-assisted printing techniques have widely been applied to build sophisticated fibrous protein architectures. In this regard, a series of fibrous protein-based bioinks have been developed, such as bioinks prepared from silk fibroin, collagen, fibrin, gelatin, and recombinant spider silk. In this chapter, we present the protocols to make various fibrous protein inks, as well as how to use these bioinks to print 3D structures via different printing techniques.

Molecular dynamics (MD) simulations performed with force fields that include explicit electronic polarization are becoming more prevalent in the field. The increasing emergence of these simulations is a result of continual refinement against a range of theoretical and empirical target data, optimization of software algorithms for higher performance, and availability of graphical processing unit hardware to further accelerate the simulations. Polarizable MD simulations are likely to be most impactful in biomolecular systems in which heterogeneous environments or unique microenvironments exist that would lead to inaccuracies in simulations performed with fixed-charge, nonpolarizable force fields. The further adoption of polarizable MD simulations will benefit from tutorial material that specifically addresses preparing and analyzing their unique features. In this chapter, we introduce common protocols for preparing routine biomolecular systems containing proteins, including both a globular protein in aqueous solvent and a transmembrane model peptide in a phospholipid bilayer. Details and example input files are provided for preparation of the simulation system using CHARMM, performing the simulations with OpenMM, and analyzing interesting dipole moment properties in CHARMM.

Natural silk protein fibers have shown a great attraction to the researchers due to the extraordinary mechanical property, biocompatibility, and functional diversity. Unfortunately, the low yield and unevenness have hampered the scale use of the natural silk fibers. Herein, the appearance of the bioinspired artificial spinning strategy offers an effective way to fabricate silk fibers with controllable structures and functionality. This chapter describes an experimental method to prepare silk protein fibers on a large scale and summarizes the method to investigate the effects of the structure-property relationship of the recombinant protein fibers.

Conformational flexibility of protein structures can play an important role in protein function. The flexibility is often studied using computational methods since experimental characterization can be difficult. Depending on protein system size, computational tools may require large computational resources or significant simplifications in the modeled systems to speed up calculations. In this work, we present the protocols for efficient simulations of flexibility of folded protein structures that use coarse-grained simulation tools of different resolutions: medium, represented by CABS-flex, and low, represented by SUPRASS. We test the protocols using a set of 140 globular proteins and compare the results with structure fluctuations observed in MD simulations, ENM modeling, and NMR ensembles. As demonstrated, CABS-flex predictions show high correlation to experimental and MD simulation data, while SURPASS is less accurate but promising in terms of future developments.

Transmembrane proteins include membrane channels, pores, and receptors and, as such, comprise an important part of the proteome, yet our knowledge about them is much less complete than about soluble, globular proteins. An important aspect of transmembrane protein structure is their exact position within the lipid bilayer, a feature hard to investigate experimentally at the atomic level. Here we describe MemBlob, a novel approach utilizing difference electron density maps obtained by cryo-EM studies of transmembrane proteins. The idea behind is that the nonprotein part of such maps carries information on the exact localization of the membrane mimetics used in the experiment and can be used to extract the positional information of the protein within the membrane. MemBlob uses a structural model of the protein and an experimental electron density map to provide an estimation of the surface residues interacting with the membrane.

The detailed knowledge of azurin structure and the new possibil-ities offered by site-directed mutagenesis make it a convenient model for studies on the stability of small globular proteins. In particular the importance of a very stable hydrophobic core for maintaining the native, biologically active conformation appears evident. The peculiar location of the single tryptophan, just at heart of this core, has two important consequences. First, the spectroscopic features of this tryptophan are similar to those of indole in non-aqueous solutions and very low tem-peratures, even though, as demonstrated by the anisotropy decay, it has a considerable freedom of rotation. Second, it represents a very useful, built-in probe not only of the native-denatured transition, but also of subtler modifications of the structure which may preceed its collapse toward a disordered state.

The dityrosine bond (DT) is an oxidative covalent cross-link between two tyrosines. DT cross-linking is increasingly identified as a marker of oxidative stress, aging and disease, and has been detected in diverse pathologies. While DT cross- linked proteins have been documented, the consequences of the DT link on the structure and function of the so modified proteins are yet to be understood. With this in view, we have studied the properties of intermolecular DT-dimers of four proteins of diverse functions, namely the enzyme ribonuclease A, the signal protein calmodulin, and the eye lens proteins alpha- and gamma B-crystallins. We find that DT is formed through radical reactions and type I photosensitization (including •OH, O2•– and OONO–), but not by 1O2 and NO2– (which modify his, trp and met more readily). Tyr residues on the surface of the protein make DT bonds (intra- and intermolecular) most readily and preferentially. The conformation of each of these DT-dimers, monitored by spectroscopy, is seen not to be significantly altered in comparison to that of the parent monomer, but the structural stability of the DT cross-linked molecule is lower than that of the parent native monomer. The DT-dimer is denatured at a lower temperature, and at lower concentrations of urea or guanidinium chloride. The effect of DT-cross-linking on the biological activities of these proteins was next studied. The enzymatic activity of the DT-dimer of ribonuclease A is not lost but lowered. DT-dimerization of lens alpha-crystallin did not significantly affect the chaperone-like ability; it inhibits the self-aggregation and precipitation of target proteins just as well as the parent, unmodified alpha-crystallin does. DT-dimerization of gamma B-crystallin is however seen to lead to more ready aggregation and precipitation, a point of interest in cataract. In the case of calmodulin, we could generate both intermolecular and intramolecular DT cross-linking, and study both the DT-dimer and DT-monomer. The DT-dimer binds smooth muscle light chain kinase and also Ca2+, but less efficiently and over a broad concentration range than the native monomer. The intramolecular DT-monomer is weaker in all these respects, presumably since it is structurally more constrained. These results suggest that DT cross-linking of globular proteins weakens their structural stability and compromises (though does not abolish) their biological activity, both of which are pathologically relevant. The intramolecular DT cross-link would appear to lead to more severe structural and functional consequences.

A subtle interplay between polar and non-polar interactions governs protein stability in aqueous solutions (Branden and Tooze, 1991). It also plays a decisive role in determining the interfacial property of proteins. For example, many small globular proteins such as a-lactalbumin, lysozyme and others, can not pack all their hydrophobic residues inside the protein core: these residues are often found exposed to the aqueous solution (Haynes and Norde, 1995). This exposure is energetically costly as the water molecules in the vicinity of protein hydrophobic residues lose some of their entropy relative to the bulk water. The free energy of the system will decrease when the exposed hydrophobic residue binds to another hydrophobic entity, or when it makes contact with a hydrophobic surface.

In recent years, NMR has been applied with great success to determine precise three- dimensional solution structures of compact globular proteins and small, inflexible cyclic peptides. NMR also holds promise for providing structural information on flexible biomolecules, although a quantitative description of structure is complicated by the inevitable population-weighted averaging of the key NMR parameters. Here we describe simple procedures for calculation of the dominant structure in the conformational ensemble of a linear peptide and for characterizing the domain motions and overall structural preferences of multidomain proteins.

The proposal by Pauling and his coworkers (1951) of an atomic model for the structure of the alpha helix stimulated research in several areas of protein chemistry. It excited chemists as few discoveries have before or since, giving impetus to structural modeling efforts that resulted in the structure of DNA 2 years later, and in a whole new field of structural biology within two decades. Pauling’s feat pointed out the importance of understanding the conformation of the peptide group itself, rather than building models based on idealized helical structures. Working on the same problem, Bragg et al. (1950) failed to produce a structure of comparable elegance because they were unaware the peptide bond was planar (Crick, 1988). The alpha helix could be identified in the diffraction patterns from crystals of the globular proteins myoglobin and hemoglobin, as well as in the classical “α” patterns from fibrous proteins like keratin and synthetic polypeptides, poly(γ-l-glutamate) being the first to show the α pattern (Elliott, 1967).

Circular dichroism (CD) spectra are used extensively for studying the conformation of proteins and polypeptides. The correlation of certain spectral features with well-defined peptide conformations has been used to develop computational procedures for conformational analysis (Perczel et al., 1992; Johnson, 1992; Venyaminov et al., 1993). While these procedures yield reasonable estimates of the fractions of a helix, 13 strand and sheet as well as various types of bends present in a test polypeptide, significant fractions are often ascribed “random” or “other” conformations. In these computations, conformations related to structures rich in imino peptide bonds are seldom taken into account. The imino-rich scleroprotein type I collagen and other members of the collagen gene family account for over one-third of the total protein content in the vertebrate body. The conformation of collagen is related to the polyproline II helix. There is growing evidence that many globular proteins may contain small domains with collagenlike structure (Ananthanarayanan et al., 1987; Adzhubei and Sternberg, 1993, 1994). Collagen, polyproline II, and related synthetic polypeptides exhibit CD spectra that appear to be similar to the spectra of many globular proteins in the so-called random coil conformation arising from the collapse of stabilizing interactions.

Circular dichroism (CD) is one of the most sensitive physical techniques for determining structures and monitoring structural changes of biomolecules. It can directly interpret the changes of protein secondary structure, even though the method is empirical. The far-ultraviolet (far-UV) CD specra (below 250 nm) of proteins are extremely sensitive toward protein structure, and the near-UV spectra reflect the contributions of aromatic side chains, disulfide bonds, and induced CD bands of prosthetic groups. Together these measurements provide information about the overall structure of a protein molecule as well as its local conformation around the aromatic and prosthetic groups and disulfide linkages. The ease of CD measurements is attractive, but CD, unlike two other powerful techniques—x-ray diffraction of protein crystals and NMR for protein solutions—cannot determine the three-dimensional structure of a protein. In this chapter we will discuss several methods of CD analysis of proteins, which can provide estimates of α helix, β sheet, β turn, and unordered form. These empirical methods utilize a set of reference proteins of known structure from x-ray diffraction studies. Thus, proteins are presumed to have the same structure in the crystalline state and in aqueous solution.

Proteins have always been known as dynamic molecules. With the success of X-ray crystallography in solving protein structures over the past thirty years, the static aspects of proteins have been emphasized, primarily because this technique can only characterize relatively rigid and completely folded proteins. On the other hand, NMR can study partially folded proteins and can characterize well internal motions. Among the most fruitful techniques to characterize internal mobility in proteins are relaxation time measurements, and a large number of studies have focused on that aspect recently (see e.g. Wagner, 1993). Globular proteins must undergo a range of motions in space and time to modulate the stunning array of critical biological processes such as enhance the rate of transcription of DNA, transport electrons, maintain structural integrity, or modulate cellular immune responses (McCammon & Harvey, 1985; Brooks et al., 1988). Understanding such motion inside a protein might thus provide some insight into their behavior. For example, it may allow the elucidation of the potential energy surface on which these proteins move, it may provide some clues on how they fold from a linear chain to three dimensional structure. Finally, dynamics is important in the context of structure refinement. Mobility studies can allow one to detect erroneously too tightly constrained structures, and identify regions that are rigid but appear artifactually disordered in structure calculations due to incomplete data analysis.

Deoxyribonucleic acid (DNA) is the most important molecular structure in biology: its aesthetic double-helical structure has come to symbolize the dramatic progress made in molecular biology over the past fifty years. In spite of its central role in all living systems, remarkably little is known about the detailed structure, dynamics and energetics of DNA. Although Watson and Crick’s model building study predicted the double-helix structure over 40 years ago (Watson and Crick, 1953) and the structure was soon confirmed by fibre diffraction (Langridge et al., 1960), it was not until 1981 that a three-dimensional structure of DNA was obtained from single-crystal diffraction (Drew et al., 1981). Since then, there have been over two hundred A, B and Z form DNA double-helical structures deposited in the Brookhaven Protein Data Bank. All the 51 different crystal structures for B-form DNA, the biologically important state, are of dodecamer or decamer double-helical sequences. 29 of the 31 dodecamers pack in exactly the same way found originally by Drew et al. (1981) and show more disorder than seen in most protein crystals (resolution ranges from 1.9 A to 3 A). In fact, even at very low temperatures (16 K) the X-ray resolution is moderate at 2.7 A (Drew and Dickerson, 1982). Nuclear magnetic resonance (NMR) studies done in solution at room temperature also show DNA to be more mobile than many globular proteins (Patel et al., 1987). NMR determination of DNA structural details in solution is more difficult than it is for proteins: in DNA there are very few closely packed hydrogen atoms to give the needed inter-hydrogen NOE distances. The situation is more even more difficult as the effect of sequence on DNA double helix structure is very subtle.

Inspection of globular proteins shows that the most common protein structures are those that have some advantage in stability. This would cause no wonder if the advantages were not so small and the difference in occurrence was not so great.

The rapidly increasing availability of high-resolution protein structures from X-ray crystallography and multidimensional nuclear magnetic resonance (NMR) opens the possibility of using structural information in the design of proteins and Pharmaceuticals. A prerequisite to structure-based design strategies is understanding which features dictate the energetics of stabilization of that structure. Recently, we have shown that it is possible to predict rather accurately the energetics of folding of a globular protein, if its crystal structure is known, using a semiempirically derived set of structural energetic parameters (Murphy et al., 1992; Murphy and Freire, 1992). This approach also shows promise for predicting the energetics of binding of peptides and proteins (Murphy et al., 1993, 1995). In this chapter we will review the structural energetic analysis and the implications of this analysis regarding the general energetic features of assembly processes that bear upon the rational design of proteins and drugs. A structural energetic analysis can be envisioned as part of an overall scheme of rational protein or drug design as shown schematically in Fig. 1.

The study of the unfolding process of proteins has been a subject of growing interest during recent years. The availability of recombinant cholinesterases in large amounts permits now to use biophysical tools for gaining information about folding and stability of these enzymes.

It is well established that many small, globular proteins undergo a two-state unfolding transition that is closely similar to first-order phase transitions in macroscopic systems. In such transitions the protein exists in just two detectable states, native and denatured, separated by an eneregy barrier; all the intermediate states are unstable and only transiently populated. The two-state model has been extraordinarily successful in accounting for the equilibrium denaturation behavior of proteins. However, tantalizing suggestions have appeared that this model is insufficient. Experimental evidence for residual structure in the denatured state has been accumulating for many years. Recently, for instance, changes in the amount and type of residual structure in large fragments of staphylococccal nuclease (nuclease) have been observed upon mutation or varying solvent conditions (Shortle & Meeker, 1989). Acid denatured nuclease also appears to have residual structure (Nakano & Fink, 1990). The importance and relevance of these observations to the stability and structure of nuclease, and of proteins in general, has been subject to question.

Predicting protein structure using computer simulation is still very much in its infancy. Proteins are far from being easy systems to treat, firstly, because of their size, typically involving many thousands of atoms and, secondly, because of their complex structures which are stabilised by many weak interactions between sites often distant from one another in terms of the primary amino acid sequence. As a result of these features, current simulation techniques, even with the help of the fastest computers presently available, are far from being able to explain the folding pathways which lead from the denatured to the native state of typical globular proteins.

We describe a computational method for constructing hydrophobically clustered compact conformations of polymer chains having hydrophobic and polar monomers. We call the method “hydrophobic zippers.”1,2 We have used hydrophobic zippers to construct conformational ensembles that resemble the compact denatured and molten globule states of proteins. We have also used hydrophobic zippers to attempt to find global minimum (“native”) conformations of simple HP (H: hydrophobic, P: polar) copolymers in lattice models. In addition, we believe that the sequence of serial steps followed by the hydrophobic zippers construction process closely parallels the sequence of kinetic events in the fast stages of the folding processes of globular proteins.