|The peptide bond||Protein Structure||Protein Denaturation|
|Properties of amino acids||Secondary Structure||Gelation|
|Non-covalent Forces||Supersecondary structure||Gelation of globular proteins|
|Covalent Forces||Tertiary structure||Protein-Protein Interaction|
|Protein Stability||Quaternary structure||Protein-Polysaccharide Interactions|
|BACK||Protein structure determination||HERB PAGE|
The primary structure of a segment of a polypeptide chain or of a protein is the amino-acid sequence of the polypeptide chain(s), without regard to spatial arrangement (apart from configuration at the alpha-carbon atom). This definition does not include the positions of disulphide bonds, and is, therefore, not identical with "covalent structure" (IUPAC-IUB, 1970). The commonly occurring amino acids are of 20 different kinds which contain the same dipolar ion group H3N+.CH.COO-. They all have in common a central carbon atom to which are attached a hydrogen atom, an amino group (NH2) and a carboxyl group (COOH). The central carbon atom is called the Calpha-atom and is a chiral centre. All amino acids found in proteins encoded by the genome have the L-configuration at this chiral centre.
This configuration can be remembered as the CORN law. Imagine looking along the H-Calpha bond with the H atom closest to you.
When read clockwise, the groups attached to the Calpha spell the word CORN (Richardson, 1981). There are 20 side chains found in proteins encoded by the genetic machinery of the cell. The side chains confer important properties on a protein such as the ability to bind ligands and catalyse biochemical reactions. They also direct the folding of the nascent polypeptide and stabilise its final conformation. In molecular graphics, atoms can be represented in different ways. For expedience, molecules are often displayed only as lines or vectors between the atoms bonded together covalently. An elegant representation is the ball-and-stick type in which atoms are drawn as coloured spheres and their bonds as rod-like connections. Another useful display is the space-filling representation in which a surface is drawn around the atoms to indicate their van der Waals radii. This surface can be drawn as a series of dots or as a solid entity (Lesk, 1991). Amino acids in proteins(or polypeptides) are joined together by peptide bonds. The sequence of R-groups along the chain is called the primary structure.
The Peptide bond
Pauling et al. (1951) analysed the geometry and dimensions of the peptide bonds in the crystal structures of molecules containing either one or a few peptide bonds. Their results are summarised in the diagram below where the consensus bond lengths are shown in Angstrom units. Bond angles in degrees are also shown for the peptide N and C atoms. It should be noted that the C-N bond length of the peptide is 10% shorter than that found in usual C-N amine bonds (Schulz and Schirmer, 1990; Creighton, 1993).
This is because the peptide bond has some double bond character (40%) due to resonance which occurs with amides. The two canonical structures are:
As a consequence of this resonance all peptide bonds in protein structures are found to be almost planar, i.e. atoms Calpha(i), C(i), O(i), N(i+1) and Calpha(i+1) are approximately co-planar. This rigidity of the peptide bond reduces the degrees of freedom of the polypeptide during folding. The peptide bond nearly always has the trans configuration since it is more favourable than cis, which is sometimes found to occur with proline (Pro) residues (Schulz and Schirmer, 1990).
As can be seen from the previous page, steric hindrance between the functional groups attached to the Calpha atoms will be greater in the cis configuration. However, for proline residues, the cyclic nature of the side chain means that both cis and trans configurations have more equivalent energies. Thus proline is found in the cis configuration more frequently than the other amino acids.
Properties of amino acids
The sequence and properties of side chains determine all that is unique about a particular protein, including its biological function and its specific three-dimensional structure.
Histidine (His) is the only side chain that titrates near physiological pH, making it especially useful for enzymatic reactions.
Lysine (Lys) and arginine (Arg) are normally positively charged and aspartate (Asp) and glutamate (Glu) are negatively charged. These charges are very seldom buried in protein interiors except when they are serving some special purpose, as in the activity and activation of chymotrypsin (Blow et al., 1969; Wright, 1973).
Asparagine (Asn) and glutamine (Gln) have interesting hydrogen-bonding properties, since they resemble the backbone peptides. The hydrophobic residues provide a very strong driving force for folding, through the indirect effect of their ceasing to disrupt the water structure once they are buried (Kauzmann, 1959). They also, however, affect the structure in a highly specific manner because their varied sizes and shapes fit together in very efficient packing (Lee and Richards, 1971).
Proline has stronger stereochemical constraints than any other residue, with only one instead of two variable backbone angles, and it lacks the normal backbone NH for hydrogen bonding. It is both disruptive to regular secondary structure and also good at forming turns in the polypeptide chain, so that in spite of its hydrophobicity it is usually found at the edge of the protein (Richardson, 1981).
Glycine (Gly) has three different unique capabilities. As the smallest side group (only a hydrogen), it is often found where main chains approach each other very closely. In addition Gly can assume conformations normally restricted by close contacts of the beta-carbon and finally it is more flexible than other residues, thus contributing to parts of the backbone that need to move or hinge (Richardson, 1981).
Serine (Ser) and threonine (Thr) carry aliphatic hydroxyl groups capable of forming hydrogen bonds with suitable donor or acceptor groups, such as the imino nitrogen or the carbonyl oxygen of the main polypeptide chain. Serine reacts with phosphate by an ester bond, forms part of the catalytic site of many hydrolytic enzymes (Dickerson and Geis, 1969) and contributes to the lining of ion channels. Serine, threonine, and asparagine are also the binding sites of carbohydrates that are attached to the surface of many proteins. Carbohydrates bound to serine and threonine form O-glycosidic bonds and those linked to asparagine form N-glycosidic bonds (Perutz, 1992).
Cysteine (Cys) carries the highly reactive sulphydryl group. This does not ionise at physiological pH nor form hydrogen bonds of significant strength, but two cysteines placed some distance apart along a polypeptide chain, or forming part of different chains, can be joined by oxidation to form the disulphide bridge of cystine which plays an important part in stabilizing protein structures. Disulphide bonds increase the conformational stability mainly by constraining the unfolded conformations of the protein and thereby decreasing their conformational entropy (Pace, 1990). Cysteines also bind zinc, copper, and iron ions. The sulphur atom in methionine is unreactive and generally serves no function other than imposing a special configuration on the aliphatic sidechain, but in cytochrome c it forms the link between the protein and the heme iron (Olson, 1992).
Protein structure determination
In terms of the accuracy of protein structure determinations, all of the bond lengths are invariant. Bond angles are also essentially invariant, except perhaps for , the backbone N-Calpha-C angle. The alpha-carbon is tetrahedral, which would give 110°, but there are indications from accurately refined protein structures (Deisenhofer and Steigemann, 1975; Watenpaugh et al., 1979) that can sometimes stretch to larger values in order to accommodate other strains in the structure. The dihedral angle at the peptide is very close to 180° (producing a trans, planar peptide with the neighbouring alpha-carbons and the N, H, C, and O between them all lying in one plane). The remaining dihedral angles are the source of essentially all the interesting variability in protein conformation. The backbone dihedral angles are and in sequence order on either side of the alpha-carbon, so that is the dihedral angle around the N-Calpha bond and around the Calpha-C bond. The side chain dihedral angles are 1, 2, etc. An extremely useful device for studying protein conformation is the Ramachandran plot (Ramachandran et al., 1963) which plots and . The values of and that are possible are constrained geometrically due to steric clashes between non-neighboring atoms. The permitted values of and are indicated on a two-dimensional map of the - plane (Branden and Tooze, 1991).
The secondary structure of a segment of polypeptide chain is the local spatial arrangement of its main-chain atoms without regard to the conformation of its side chains or to its relationship with other segments (IUPAC-IUB, 1970). There are three common secondary structures in proteins, namely alpha helices, beta sheets and turns. That which cannot be classified as one of the standard three classes is usually grouped into a category called "other" or "random coil". This last designation is unfortunate as no portion of protein three dimensional structure is truly random and it is not a coil either. Regular secondary structure conformations in segments of a polypeptide chain occur when all the bond angles in that polypeptide segment are equal to each other, and all the bond angles are equal. The rotational angles for and bonds for common regular secondary structures are shown in the table below.
|Right-handed alpha helix[3.613helix]||-57||-47||3.6||5.4||13|
|Parallel beta strand||-119||113||2.0||6.4|
|Antiparallel beta strand||-139||135||2.0||6.8|
The alpha-helix and beta-structure conformations for polypeptide chains are generally the most thermodynamically stable of the regular secondary structures. However, particular amino acid sequences of a primary structure in a protein may support regular conformations of the polypeptide chain other than alpha-helical or beta-structure. Thus, whereas alpha-helical or beta-structure are found most commonly, the actual conformation is dependent on the particular physical properties generated by the sequence present in the polypeptide chain and the solution conditions in which the protein is dissolved. In addition, in most proteins there are significant regions of unordered structure in which the and angles are not equal. A large proportion of (85%) of helices are distorted in some way i.e. radius of curvature greater than 90Å and deviation of axis from straight line is equal to or greater than 0.25Å. These may be due to a number of reasons:
Besides the alpha-helix, beta-sheets are another major structural element in globular proteins containing 20-28% of all residues (Kabsch and Sander, 1983; Creighton, 1993). The basic unit of a beta-sheet is a beta strand (which can be thought of as a helix with n=2 residues/turn) with approximate backbone dihedral angles phi = -120 and psi = +120 producing a translation of 3.2 to 3.4 Å / residue for residues in antiparallel and parallel strands, respectively. The beta strand is then like the alpha-helix, a repeating secondary structure. However, since there are no intra-segment hydrogen bonds and van der Waals interactions between atoms of neighbouring residues are not significant due to the extended nature of the chain, this extended conformation is only stable as part of a beta-sheet where contributions from hydrogen bonds and van der Waals interactions between aligned strands exert a stabilizing influence. The beta-sheet is sometimes called the beta pleated sheet since sequential neighbouring Calpha atoms are alternately above and below the plane of the sheet giving a pleated appearance. beta-sheets are found in two forms designated as "Antiparallel" or "Parallel" based on the relative directions of two interacting beta strand (as shown below).
Hydrogen bond patterns in beta-sheets. A four-stranded beta-sheet is drawn schematically which contains three antiparallel and one parallel strand. Hydrogen bonds are indicated with red lines (antiparallel strands) and blue lines (parallel strands) connecting the hydrogen and receptor oxygen.
Like alpha-helices, beta-sheets have the potential for amphiphilicity with one face polar and the other apolar. However, unlike alpha-helices which are composed of residues from a continuous polypeptide segment (i.e., hydrogen bonds between CO of residue I and NH of residue I+3), beta-sheets are formed from strands that are very often from distant portions of the polypeptide sequence. Hydrogen bonds in beta-sheets are on average 0.1 Å shorter than those found in alpha-helices (Baker and Hubbard, 1984). The classical beta-sheets originally proposed are planar but most sheets observed in globular proteins are twisted (0 to 30 degrees per residue).
Antiparallel beta-sheets are more often twisted than parallel sheets. Another irregularity found in antiparallel beta-sheets is the hydrogen-bonding of two residues from one strand with one residue from another called a beta bulge (as shown above). Bulges are most often found in antiparallel sheets with ~5% of bulges occurring in parallel strands (Richardson, 1981).
Turns are the third of the three "classical" secondary structures that serve to reverse the direction of the polypeptide chain. They are located primarily on the protein surface and accordingly contain polar and charged residues. Antibody recognition, phosphorylation, glycosylation, and hydroxylation sites are found frequently at or adjacent to turns. Turns were first recognised from a theoretical conformational analysis by Venkatachalam (1968). He considered what conformations were available to a system of three linked peptide units (or four successive residues) that could be stabilised by a backbone hydrogen bond between the CO of residue n and the NH of residue n+3. He found three general types, one of which (type III) actually has repeating Turns are the third of the three "classical" secondary structures that serve to reverse the direction of the polypeptide chain. They are located primarily on the protein surface and accordingly contain polar and charged residues. Antibody recognition, phosphorylation, glycosylation, and hydroxylation sites are found frequently at or adjacent to turns. Turns were first recognised from a theoretical conformational analysis by Venkatachalam (1968). He considered what conformations were available to a system of three linked peptide units (or four successive residues) that could be stabilised by a backbone hydrogen bond between the CO of residue n and the NH of residue n+3. He found three general types, one of which (type III) actually has repeating , values of -60deg, -30deg and is identical with the 310-helix. The three types each contain a hydrogen bond between the carbonyl oxygen of residue i and the amide nitrogen of i+3. These three types of turns are designated I, II, and III. Many have speculated on the role of this type of secondary structure in globular proteins. Turns may be viewed as a weak link in the polypeptide chain, allowing the other secondary structures (helix and sheet) to determine the conformational outcome. In contrast (based on the recent experimental finding of "turn-like" structures in short peptides in aqueous solutions (Dyson et al., 1988), turns are considered to be structure-nucleating segments, formed early in the folding process. Type I turns occur 2-3 times more frequently than type II. There are position dependent amino acid preferences for residues in turn conformations. Type I can tolerate all residues in position i to i+3 with the exception of Pro at position i+2. Proline is favoured at position i+1 and Gly is favoured at i+3 in type I and type II turns. The polar sidechains of Asn, Asp, Ser, and Cys often populate position i where they can hydrogen bond to the backbone NH of residue i+2.
Rao and Rossmann (1973) observed that there were structural components comprising a few alpha-helices or beta-strands which were frequently repeated within structures, called "supersecondary structures" (being intermediate to secondary and tertiary structure) and suggested that these structures might be due to convergence. A variety of recurring structures were subsequently recognised such as the "Greek key" (Richardson, 1977). Some of these structural motifs can be associated with a particular function whilst others have no specific biological function alone but are part of larger structural and functional assemblies. The first comprehensive attempt to classify proteins on the basis of structural comparisons was due to Levitt and Chothia (1976) who used four major classifications:
The tertiary structure of a protein molecule, or of a subunit of a protein molecule, is the arrangement of all its atoms in space, without regard to its relationship with neighbouring molecules or subunits (IUPAC-IUB, 1970). As an example of a protein's tertiary structure, the structure of lysozyme is shown below.
The quaternary structure of a protein molecule is the arrangement of its subunits in space and the ensemble of its intersubunit contacts and interactions, without regard to the internal geometry of the subunits (IUPAC-IUB, 1970). The subunits in a quaternary structure must be in noncovalent association. Haemoglobin contains four polypeptide chains (alpha2b2) held together noncovalently in a specific conformation as required for its function.
To be biologically active, proteins must adopt specific folded three-dimensional, tertiary structures. Yet the genetic information for the protein specifies only the primary structure, that is the linear sequence of amino acids in the polypeptide backbone. Many purified proteins can spontaneously refold in vitro after being completely unfolded, so the three-dimensional structure must be determined by the primary structure (Anfinsen, 1973). Different conformations of a protein differ only in the angle of rotation about the bonds of the backbone and amino acid side-chains. It may, therefore, appear surprising that a protein folds into a single unique conformation from all the possible rotational conformations available around single bonds in the primary structure of a protein. The question arises as to when a protein folds up to its native conformation, does this structure represent a local or a global energy minimum? When a protein folds it samples a number of conformations. Does the structure which results from the folding depend on its stability or on the energy barriers encountered by the polypeptide? The polypeptide whilst folding may become trapped in the local energy well and cannot fold to the global energy minimum (kinetic hypothesis of protein folding, Wetlaufer, 1973; Wetlaufer and Ristow, 1973).
The native structure of the protein may correspond to a metastable state with a very long lifetime. If proteins are only metastable, their structures must be grossly different from the most stable one. The polypeptide may adopt a structure corresponding to the global minimum. This means the final structure does not depend on the size of the energy barriers (thermodynamic hypothesis of protein folding, Epstein et al., 1963). Anfinson's work on ribonuclease provided some evidence for the thermodynamic hypothesis (Haber and Anfinsen, 1962). The initial stages of folding were considered to be nearly random. However, if the rest of the folding pathway was a random search, it would not be feasible for any protein to try out all of its conformations on a practical time scale. For example, if each residue of a 100 residue polypeptide had only three conformations, the total number of conformations would be 3100 = 5 x 1047. Since conformational changes occur on the timescale of 10-13 seconds, the time required by the 100 residue protein to search all conformations would be 5x1047x10-13 » 1037 years. Nevertheless, proteins are observed to fold in 10-1 - 103 seconds both in-vivo and in-vitro (Creighton, 1993). The conclusion, therefore, was that proteins do not fold by sampling all possible conformations randomly until the one with the lowest free energy is encountered.
This led to the idea of a biased random search resulting in faster folding since a proportion of the conformations would be sterically disallowed. A framework model proposed (Baldwin, 1989; Kim and Baldwin, 1990) that the folded structure was formed by packing together pre-existing individual elements of secondary structure which had significant stability in the unfolded protein. Another mechanism postulated the unfolded polypeptide chain to undergo rapid hydrophobic collapse under refolding conditions (Dill, 1985), perhaps to something approximating the molten globule state. Simply constraining the polypeptide chain to be compact might greatly increase the probability of the final folded conformation being encountered (Gregoret and Cohen, 1991). It became evident through some experiments that the equilibrium of unfolding of proteins does not always follow a simple two-state model in which only the native and fully unfolded states are significantly populated (Wong and Tanford, 1973). An intermediate compact structure known as the molten globule which is different from the native structure and whose formation is determined mainly by non-specific interactions of amino acid residues with their environment was presented. Specific interactions could direct the folding pathway by stabilizing folded conformations. The best studied example is bovine pancreatic trypsin inhibitor (Creighton, 1978). For this protein it was shown that formation of a disulphide bridge stabilizes secondary structure elements, and the protein refolds by a specific pathway of disulphide bond formation and rearrangement (Directed folding model). Since noncovalent forces act on the primary structure to cause a protein to fold into a unique conformational structure and then stabilize the native structure against denaturation processes, it is of importance to understand the properties of these forces.
Non-covalent forces are weak forces of bonding strength of 1-7kcal mol-1 (4-29kJ mol-1) as compared to the strength of covalent bonds which have a bonding strength of at least 50 kcal mol-1. The non-covalent bonding forces are just higher than the average kinetic energy of molecules at 37°C (0.6 kcal mol-1). Apart from their involvement in the stabilization of molecules, they contribute to the ability of molecules to undergo changes in conformation and interact with each other. Since a major part of this project (Friedli, 1996)involves a study of interactions, an explanation of the forces involved is crucial for an understanding of the mechanisms at the molecular level. The major forces and interactions are:
Ion - dipole attractions depend on;
A polar water molecule can induce a dipole in non-polar O2. For the dipole to be induced depends on the atom's or molecule's polarizability. As the molecular mass of a molecule increases, either there is an increase in the number of valence electrons or the valence electrons are less tightly held. Therefore, the ease of polarization of the valence electron cloud generally increases with mass. Since a dipole is more readily induced as the polarizability increases, the strengths of dipole - induced dipole interactions generally increase with mass. Also, since the solubility of substances such as CO2 or O2 depends on the strength of the dipole - induced dipole interaction, the solubility of non-polar substances in polar solvents generally increases with mass (Kotz and Purcell, 1991).
The weakest of all intermolecular forces is between two induced dipoles. Such forces are often called London forces or dispersion forces. When atoms or molecules approach each other, each experiences the electric field provided by the other. This electric field distorts the charge distribution (Webster, 1990). The attraction of a molecule to its own distorted charge distribution creates an attractive force between the molecules even when they are a long distance apart. This force acts to bind the approaching atoms or molecules. Fritz London (1930) gave an approximate result for the attractive energy resulting from the interaction between two induced dipoles. The interaction energy, ELondon/J, which is attractive for all inter-atomic, or inter-molecular distances, R/m, is expressed by the formula:
where IA/J and IB/J are the first ionization energies for A and B, and and are their polarizability volumes. ELondon is called the London dispersion energy. The London energy will be larger in magnitude as R decreases but the key feature is that it creates an attraction even at large R values. For this reason it is usually classified as a long-range interaction. The London dispersion force, however, is only one component of van der Waals forces. A second component, the John Lennard-Jones potential (LJ potential), U(R)/J, is defined by:
The parameter represents the binding energy of a van der Waals molecule at its equilibrium geometry. The parameter is the value of R, the inter-atomic or inter-molecular distance, when U(R) = 0, other than at . From equation (2), the term R-6, approximates the London dispersion energy which is attractive and this is counteracted by a repulsive term having an R-12 dependence. An alternative expression which approximates the energy, U(R)/kJmol-1, of van der Waals molecule is :
The van der Waals force is of great importance in biopolymer structure. This force has an attractive term dependent on the 6th. power of the distance between two interacting atoms and a repulsive term dependent on the 12th. power of the distance between them. The attractive component is due to the induction of complementary partial charges or dipoles in the electron density of adjacent atoms when the electron orbitals of the two atoms approach to a close distance whereas the repulsive component of the van der Waals force predominates at closer distances, when the electro orbitals of the adjacent atoms begin to overlap. This type of repulsion is commonly called steric hindrance. The distance of maximum favourable interaction between two atoms is known as the van der Waals contact distance, which is equal to the sum of the van der Waals radii for the two atoms. While a van der Waals - London dispersion interaction between any two atoms in a protein is usually less than 1 kcal mol-1, the total number of these weak interactions in a protein molecule is in the thousands. Thus the sum of the attractive and repulsive van der Waals - London dispersion forces are extremely important to protein folding and stability. The van der Waals contact distances of 2.8-4.1 Å are longer than hydrogen-bond distances of 2.6-3.1 Å, and at least twice as long as normal covalent bond distances of 1.0-1.6 Å between C, H, N, and O atoms. Although the latter bonds are shorter than the van der Waals contact distance, a repulsive van der Waals force must be overcome in forming hydrogen bonds and covalent bonds between atoms.
The fundamental law of electrostatics namely Coulomb's law, expresses the inverse square law of force between two electric charges q1 q2 separated by a distance R in a vacuum in the form:
where is the permittivity of free space (vacuum). If q1 and q2 have the same sign, the force is a repulsion; if they are of opposite sign, the force is an attraction. In the presence of a material medium surrounding both charges, the force is reduced by a factor , the relative permittivity (or dielectric constant) of the medium. The work done in bringing two charges together from infinite separation to a distance R in a medium of permittivity is, therefore, given by:
and measures the electrical free energy of the system relative to that at infinite separation. Electrostatic interactions between charged groups are of importance to particular protein structures, in the binding of charged ligands and substrates to proteins, interaction between basic and acidic proteins (Friedli, PhD thesis chapter 8), and repulsion between charges of the same sign as between SWP and sodium alginate (Friedli et al., 1995). The strength of the electrostatic force (E) is directly dependent on the charge (q) for each ion, and is inversely dependent on the dielectric constant () of the solvent and the distance between the charges (R). Water has a high dielectric constant ( =80), and ionic charge interactions in water are relatively weak in comparison to electrostatic interactions in the interior of a protein, where the dielectric constant ( = 2-40) is approximately a factor of 1 : 40 to 1 : 2 that of water. Consequently, the strength of an electrostatic interaction in the interior of a protein, where the dielectric constant is low, may be of significant energy. However, most charged groups of proteins are on the surface of the protein where they do not strongly interact with other charged groups from the protein or other biopolymers due to the high dielectric constant of the water solvent, but are stabilized by hydrogen bonding and polar interactions to the water. Electrostatic interactions in water are less than those in other solvents because of water's high dielectric constant, which results from the tendency of the large dipoles of water molecules to align with any electric field. The dielectric constant of pure water at 25°C is 78.5 and it decreases at higher temperatures because thermal motion overcomes the orienting effects of the water dipoles (Creighton, 1993). This effect of temperature explains some of the findings in (Friedli, PhD thesis, chapter 5). When small diffusible ions such as Na+ and Cl- are present in water, the apparent dielectric constant of the solution increases because the ions tend to concentrate in the vicinity of charges of the opposite sign. Since the present project considered biopolymer interactions, most of the experiments were performed with water in order to avoid interference from salts.
The strong inclination of water molecules to form hydrogen bonds with each other influences their interactions with non-polar molecules that are incapable of forming hydrogen bonds (e.g., alkanes, hydrocarbons inert atoms etc.). When water molecules come in contact with such a molecule they are faced with an apparent dilemma that whichever way the water molecules face, it would appear that one or more of the four charges per molecule (ST2 model of water, named after Stillinger and Rahman, 1974) will have to point towards the inert solute molecule and thus be lost to hydrogen bond formation. Clearly the best configuration would have the least number of tetrahedral charges pointing towards the unaccommodating species so that the other charges can point towards the water phase and, therefore, participate in hydrogen bond attachments much as before. There are many options to salvage lost hydrogen bonds. If the non-polar solute molecule is not too large, it is possible for water molecules to pack around it without giving up any of their hydrogen-bonding sites, thus forming clathrate 'cages' around a dissolved non-polar solute molecule. Such structures are not rigid but labile, and their hydrogen bonds are not stronger than in pure water, but the water molecules forming these cages are more ordered than in the bulk liquid.
It is also clear that the sizes and shapes of non-polar solute molecules are fairly critical in determining the water structure adopted around them. This is often referred to as hydrophobic solvation or hydrophobic hydration. At present there is no simple theory of such solute-solvent interactions. However, both theoretical and experimental studies indicate that the re-orientation, or re-structuring, of water around non-polar solutes or surfaces is entropically very unfavourable, since it disrupts the existing water structure and imposes a new and more ordered structure on the surrounding water molecules. This immiscibility of inert substances with water, and the mainly entropic nature of this incompatibility is known as the hydrophobic effect (Kauzmann, 1959; Tanford, 1980), and such substances, e.g., hydrocarbons and fluorocarbons, are known as hydrophobic substances. Similarly, hydrophobic surfaces are not 'wetted' by water but when water comes into contact with such surfaces it rolls up into small lenses and subtends a large contact angle on them.
Closely related to the hydrophobic effect is the hydrophobic interaction, which describes the unusually strong attraction between hydrophobic molecules and surfaces in water which are often stronger than their attraction in free space. For example, the van der Waals interaction energy between two contacting methane molecules in free space is -2.5 x 10-21 J, while in water it is -14 x 10-21 J. Because of its strength it was originally believed that some sort of 'hydrophobic bond' was responsible for this interaction. But at present it is known that there is no bond associated with this mainly entropic phenomenon, which arises primarily from the rearrangements of hydrogen bond configurations in the overlapping solvation zones as two hydrophobic species come together, and which is also much of a longer range than any typical bond. To date there have been very few direct measurements of the hydrophobic interaction between dissolved non-polar molecules, mainly because they are so insoluble. Tucker et al. (1981) reported values of -8.4 and -11.3 kJ mol-1 for the free energies of dimerization of benzene-benzene and cyclohexane-cyclohexane, respectively, and Ben Naim et al. (1973) deduced a value of about -8.5kJ mol-1 for two methane molecules. There is as yet no satisfactory theory of the hydrophobic interaction, though a number of promising theoretical approaches have been proposed (Dashevsky and Sarkisov, 1974; Pratt and Chandler, 1977; Marcelja et al., 1977; Pangali et al., 1979; Nicholson and Parsonage, 1982). Israelachvili and Pashley (1982) measured the hydrophobic force law between two macroscopic curved surfaces in water and found that in the range 0-10nm, the force decayed exponentially with distance with a decay length of about 1nm. Based on these findings Israelachvili and Pashley proposed that for small solute molecules, the hydrophobic free energy of dimerization increased in proportion with their diameter s according to:
where is in nanometres.
The hydrophobic interaction plays a central role in;
As mentioned earlier, the van der Waals force between similar particles in a medium is always attractive, so that if only van der Waals forces were operating, we might expect all dissolved particles to stick together (coagulate) immediately and precipitate out of solution as a mass of solid material. This normally does not happen, because particles suspended in water and any liquid of high dielectric constant are usually charged and can be prevented from coalescing by repulsive electrostatic forces. Other repulsive forces that can prevent coalescence are solvation and steric forces. The charging of a surface can come about by the ionisation or dissociation of surface groups (e.g., the dissociation of protons from surface carboxylic groups [ -COOH --> -COO- + H+], which leaves behind a negatively charged surface). The surface charge is balanced by an equal but oppositely charged region of counterions, some of which are bound, usually transiently, to the surface within the so-called Stern or Helmholtz layer, while others form an atmosphere of ions in rapid thermal motion close to the surface, known as the diffuse electric double layer (Figure 1.5).
The electric potential in the solution falls off exponentially with distance from the surface.
where is the potential at the potential determining surface and the potential at a distance x from the surface in the electrolyte solution. The quantity k is called the Debye-Hückel parameter. The quantity 1/k is referred to as the thickness of the double layer (Sennett and Olivier, 1965; Shaw, 1986; Ross and Morrison, 1988; Everett, 1989; Hunter, 1993; Atkins, 1994). As two charged surfaces come together, their double layers overlap and as a rough approximation, the electrical potentials arising from the two surfaces are additive. This implies an increase in the electrical contribution to the free energy of the system. A dispersion represents a state of higher free energy than that corresponding to the material in bulk. Passage to a state of lower free energy will, therefore, tend to occur spontaneously unless there is a substantial energy barrier preventing the elimination of the dispersed state (Everett, 1989). In the presence of such a barrier the system will be metastable and may remain in that state for a long time. On the other hand, if conditions are adjusted so that the energy barrier becomes negligibly small, or disappears altogether, then the dispersion becomes unstable and aggregate. The energy necessary to carry the system over the energy barrier comes from Brownian motion of the particles, which results from the random bombardment of the surface of the particles by molecules of the medium. Instability will ensue if the ratio of the energy barrier height to kT is reduced. This may arise in various ways. In principle, if the absolute height remained constant, then instability could be induced by an increase in temperature. The barrier height is also influenced by concentration and ionic strength (Hunter, 1993; Israelachvili, 1995). The double layer repulsion depends on the ionic strength of the medium as follows:
When two or more atoms come together to form a molecule, as when two hydrogen atoms and one oxygen atom combine to form a water molecule, the forces that tightly bind the atoms together within the molecule are called covalent forces, and the inter-atomic bonds formed are called covalent bonds. In a covalent bond electrons are shared between two or more atoms so that the discrete nature of the atoms is lost. Depending on the position an atom (or element) occupies in the periodic table, it can participate in a certain number of covalent bonds with other atoms. This number or stoichiometry is known as the atomic valency. A further characteristic of covalent bonds is their directionality, that is, they are directed or oriented at well-defined angles relative to each other. Covalent forces are short range, that is, they operate over very short distances of the order of inter-atomic separations (0.1 - 0.2 nm) and tend to decrease in strength with increasing bond length. There are two types of covalent bonds in proteins, the peptide bond (section 1.1.2) and the disulphide bond.
Disulphide bonds occur between the sulphurs of two cysteine side chains. They predominantly occur in extracellular proteins and are part of the primary structure. Inside the cell the sulphydryl is maintained in a reduced state by glutathione, but extracellularly, in the presence of oxygen, thiols are unstable relative to S-S bridges (Fahey et al., 1977). Although the disulphides are part of the primary structure, it has been shown that some native S-S bridges are only formed once the secondary and even tertiary structure of the protein has been achieved (Creighton, 1978). In extracellular proteins of known sequence, which contain disulphides, there is rarely more than one free -SH group. Ovalbumin, with one S-S and four cysteines, was the only extracellular exception found (Nisbet et al., 1981). Sulphydryls are very reactive in an extracellular environment and readily oxidise to form disulphides. Consequently if a cysteine is external this may lead to disastrous polymerisation or make folding more complex. When the disulphide conformation of all the proteins in the Protein Data Bank (PDB) were examined in detail, it was found that they could be grouped into two major categories (Richardson, 1981):
A detailed analysis was done on the number of residues between half-cystines (Thornton, 1981) and it was found that the most frequent separation is 10 to 14 residues. The shortest connection found was two residues and connections longer than 150 residues are rare. A study was carried out to find out how disulphides fit into the three-dimensional structure of a protein (Sternberg and Thornton, 1976). The disulphides were divided into two groups: local disulphides, with half-cystines separated by less than 45 residues and non-local disulphides with a separation greater than 45 residues. Thirty-three out of the 34 local disulphides investigated had one of the structures shown in Figure 1.7. The most common being the coil-coil (cc), followed by c-bb-c. No helix-helix (aa) local disulphides were observed, although (aa) disulphides were found between non-local half-cystines.
A left-handed spiral disulphide from hen egg white lysozyme (left) and a right-handed hook disulphide from carboxypeptidase A (right). [Richardson, 1981]
Local disulphide "single loop" topologies. c=coil, beta=strand, alpha=helix. (Thornton, 1981)
Protein denaturation has been defined in several ways, for example as a change in solubility (Mirsky, 1941) or by simultaneous changes in chemical, physical and biological properties (Neurath et al., 1944; Langmuir, 1938) under some standard reference set of conditions (Timasheff and Gibbs, 1957). These changes in physical, and to a lesser extent chemical properties are manifestations of configurational changes taking place in the polypeptide chains. The denaturation process presumably involves an unfolding or at least an alteration in the nature of the folded structure ( Foster & Samsa, 1951). Most denaturation changes consist of changes in secondary bonds: ion-dipole, hydrogen and Van der Waals, and in the rotational positions about single bonds which are controlled by the secondary bond structure (Lumry and Eyring, 1954). The term denaturation denotes the response of the native protein to heat, acid, alkali, and a variety of other chemical and physical agents which cause marked changes in the protein structure. Rice et al. (1958) suggested denaturation to mean a class of reactions which lead to changes in the structure of the macromolecule with no change in molecular weight. Timasheff and Gibbs (1957), pointed out that the approaches used to define the concept of denaturation can be classified into two types:
The special importance attached to an understanding of the denaturation process is due to the fact that denaturation is usually a prerequisite for gelation. Since this project investigationed the elucidation of the gelation process it was considered that the control of denaturation could lead to a specific gelation properties. Evidence suggests that the various individual non-covalent bonds do not act independently but that there is a cooperative action of particular groups of 'bonds' or contacts in stabilizing various segments of the structure or even the total conformation (Tanford, 1968). Thus conformational transitions are found to pass through a few intermediate states or take place by an all-or-none type of mechanism between two states with no intermediates occurring in substantial concentration (Creighton, 1994).
The denaturation process can be achieved by any one of the following methods: increasing temperature, changing pH, using denaturants (i.e. urea, guanidine hydrochloride, beta-mercaptoethanol, dithiothreitol), inorganic salts (i.e lithium bromide, potassium thiocyanate, sodium iodide), organic solvents and (i.e. formamide, dimethylformamide, dichloro- and trichloroacetic acids and their salts), detergents (i.e. sodium dodecyl sulphate), high pressure and ultrasonic homogenisation.
The temperatures at which various proteins unfold vary enormously. Most proteins unfold at elevated temperatures, and some unfold at very low temperatures. Many proteins unfold at temperatures only a few degrees higher than those at which they function. Others are stable to much higher temperatures such as the gluten proteins. The driving force for denaturation is the increase in entropy that accompanies the transition of a single conformation into an ensemble of random ones. With increasing temperature the contribution of this entropy increases and becomes more significant, and at some temperature it overcomes the energy effect ( the protein is heat denatured). It is interesting to consider possible intermediate structures. The early unlocking of the tertiary structure deletes a large number of the bonds holding the structure together but increases the randomness only insignificantly. The later stages of denaturation lead to larger increases in entropy. Thus, the intermediate states are relatively unstable, and heat denaturation is often an all-or-none phenomenon. The unfolding of the protein exposes the buried non-polar amino acid residues. Their intermolecular clustering leads to aggregation of the denatured protein. Consequently, heat denaturation is essentially irreversible.
In chemical denaturation the secondary bonds holding the protein segments together are disrupted by some chemicals capable of forming equally strong or stronger bonds with the groups holding the conformation together. For disrupting the hydrogen bonds, urea or guanidine hydrochloride are used. At high concentrations of these substances (e.g, 8M urea or 5M guanidine hydrochloride) many proteins adopt a highly unfolded conformation in solution. Proteins of multiple subunits are likely to be separated into their constituent polypeptide chains. Other proteins aggregate upon denaturation in urea or guanidine hydrochloride which is frequently due to the formation of disulphide bridges between sulphydryl groups made accessible by the unfolding of the polypeptide chains. Such reactions may be inhibited by the addition of iodoacetate (Friedli, PhD thesis, chapter 3 and 5). Under these conditions, the denatured molecules remain in solution and may revert into native molecules if the denaturing agent is slowly dialysed away. Powerful detergents like SDS disrupt both hydrophobic and hydrogen bonds and effectively solvate the denatured molecule. beta-Mercaptoethanol and dithiothreitol (DTT) disrupt disulphide bonds and can be used in conjunction with urea or SDS to fully solubilise protein molecules.
Protein-protein interactions occur widely. These can either involve specific binding or non-specific interactions. Numerous examples of specific binding can be observed in biological systems:
Protein-protein interactions are generally favoured under conditions which reduce the net charge on the molecules, i.e. pH values near the isoelectric point. High ionic strength tends to reduce electrostatic repulsion between proteins due to the shielding of ionizable groups by mobile ions. Protein-protein association involves the specific complementary recognition of two macromolecules to form a stable assembly (Jones and Thornton, 1995). Fundamental to the stabilization of protein association is the hydrophobic interaction (Chothia and Janin, 1975). The term hydrophobic interaction is used to describe the gain in free energy which occurs when non-polar residues of proteins associate in an aqueous environment (Kauzmann, 1959). The process of folding and protein-protein aggregation reduces the surface area in contact with water. When the protein-solvent interaction is attractive, the protein can reduce its total energy by surrounding itself with solvent molecules, conversely, when the interaction is repulsive, the solvent is excluded (Tanaka, 1981). The aggregation of protein subunits buries the hydrophobic residues of the proteins, and hence minimizes the number of thermodynamically unfavourable solute-solvent interactions as found when SWP is hydrated in distilled water at room temperature (Friedli, PhD thesis, chapter 3).
Most of the proteins in food systems are denatured to varying degrees depending upon the type of processing used. The functionality required dictates the type and concentrations of ingredients and their environment. Protein-protein or protein-polysaccharide interactions under these circumstances are not specific, like in biological systems, but rather, depend mainly on physico-chemical forces. Proteins spontaneously aggregate when hydrated, therefore, molecular interactions are best studied in dilute systems. As their concentrations increase, their behaviour is better explained with colloid chemistry. Physical functions associated with proteins in a food system typically includes hydration and water binding which affect viscosity and gelation; modification of surface and interfacial activity which control emulsification and foaming ability and chemical reactivity leading to altered states of cohesion/adhesion and a potential for texturization (Fligner and Mangino, 1991).
A particular functional property may be a manifestation of a specific component of the food protein used or due to interactions involving the biopolymers in the system. Schoen,(1977) pointed out that one major objective of protein functionality research, as it relates to foods, is to understand how proteins interact with each other and with other components in mixed systems. Protein interactions under certain circumstances lead to gelation.
Gels may be defined by their ability to immobilize liquid, macromolecular structure, textural or rheological properties (Kinsella, 1976). The Collins English Dictionary defines a gel as a semi-rigid jelly-like colloid in which liquid is dispersed in a solid ( Latin, gelare, meaning 'to freeze' ). Bungenberg de Jong (1949), defined a gel as a system of solid character in which the colloidal particles somehow constitute a coherent structure. Hermans (1949), defined a gel with three propositions : (a) they are coherent colloid systems of at least two components; (b) they exhibit mechanical properties characteristic of a solid; (c) both the dispersed component and the dispersion medium extend themselves continuously throughout the whole system. Flory (1974), classified gels based on their structure into four types as follows :
Biopolymer gels differ from synthetic polymer gels in a number of ways:
And on a shorter timescale the behaviour of non-covalently cross-linked gels approximates to that expected for a permanent network (Morris, 1983).
Tanaka (1981), defined a gel as a form of matter intermediate between a solid and a liquid which consists of polymers, or long chain molecules, cross-linked to create a tangled network immersed in a liquid medium. The liquid prevents the polymer network from collapsing into a compact mass, whilst the network prevents the liquid from flowing away. Tanaka identified three forces which interact to either expand or shrink polymer networks. The forces are; rubber elasticity, the polymer-polymer affinity and the hydrogen-ion pressure. The sum of these three forces was called the osmotic pressure of the gel, because it determines whether the gel tends to take up fluid or to expel it. The rubber elasticity is the elasticity of the individual polymer strands (i.e. the resistance the strands offer to either stretching or compression). The polymer-polymer affinity is the interaction between the polymer strands and the solvent. Such interactions can be either attractive or repulsive, depending on the electrical properties of the molecules. The hydrogen-ion pressure, is associated with the ionization of the polymer network, which releases an abundance of positively charged hydrogen ions (H+) into the gel fluid. The hydrogen-ion give rise to pressure in the gel. A strand can be represented by a chain of rigid, jointed segments, each of which is in constant thermal motion. If the chain is stretched almost taut, the random movements of the segments give rise to a tension that pulls the ends of the chain inwards. If the chain is compressed into a compact ball, the force is directed outward. At an intermediate length of the chain the average force is zero. The rubber elasticity is proportional to the absolute temperature because thermal agitation is the ultimate root of the force. The polymer-polymer affinity decides if two polymers will aggregate and thereby exclude the solvent from the space between them or if there is a greater force of attraction between the polymer and the solvent than a polymer and another polymer.
A morphological classification of gels was proposed by Russo (1987) as :
Burchard and Ross-Murphy (1988), proposed a phenomenological definition, stating that : all gels possess at least one property which can stand as the operational definition of a gel. They possess a plateau in the real part of the complex modulus extending over an appreciable window of frequencies - i.e. they are, or can be coaxed under appropriate conditions to be, viscoelastic solids.
Almdal et al., (1993) also suggested a phenomenological definition of a gel based on two criteria :
Most of the theories discussed so far are related to single component gels. The present project was carried out on mixed binary systems, therefore, a brief discussion on two component gels proposed by Morris (1985) would be appropriate. In a two component gel, the polymer that forms the network structure was labelled as 'active' and the other which is merely contained within the network structure as 'non-active'. A gel is called type I if it is made up of both active and non-active polymers, and type II, if both polymers are active. Complex formation between the active and non-active polymers in type I gels is unlikely. The non-active polymers will tend to concentrate the active polymers promoting intra- and inter-molecular interactions between them. The network structure in type II gels will depend on the relative values of the two active components and their degree of phase separation prior to gelation. The network formed were classified as:
Gelation of globular proteins
Several factors affect gelation, such as protein type, protein concentration, temperature, ionic strength, type of ion and pH (Mulvihill and Kinsella, 1987). Probably the two most important factors in gelation are the protein concentration and heating temperature. If either, or both the temperature (Dunkerley and Hayes, 1980) and protein concentration (Ross-Murphy, 1991) are too low, gelation will not occur. Once these factors are above their critical values, gel strength increases and gelation time decreases with increasing temperature (Schmidt and Illingworth, 1978; Dunkerley and Hayes, 1980; Dunkerley and Zadow, 1984) and concentration (Ross-Murphy, 1991; Plock et al., 1992). Ferry (1948) suggested a two-step mechanism of gelation which involves: (1) an initiation step involving unfolding or dissociation of the protein molecules, followed by (2) an aggregation step in which association or aggregation reactions occur, resulting in gel formation under appropriate conditions. For the formation of a highly ordered gel, it is essential that the aggregation step proceeds at a slower rate than the unfolding step (Hermansson, 1978, 1979). Richardson and Ross-Murphy (1981) concluded from their gelation experiments, using 9% BSA solutions at pH 6.5 and at various temperatures, that the unfolding of BSA is rate-determining at temperatures below 57°C and that aggregation is rate-determining above that temperature. This finding is consistent with the views of Ferry (1948) that unfolding precedes aggregation and provides the driving force for protein-protein interaction.
There are a number of ways in which two or more proteins can interact with each other which will affect the properties of the gel formed. After mixing, proteins may be qualitatively considered incompatible, semicompatible, or compatible, depending on whether two immiscible phases are formed, partial mixing takes place at the molecular level, or a single thermodynamically stable phase is formed (Manson and Sperling, 1976). Composite or multicomponent gels are produced from mixtures of two or more gelling agents, or a single gelling and nongelling components. A second protein capable of gel formation may act as a non-gelling component if it is present in the mixture at a concentration below its critical concentration for gel formation (Ziegler and Foegeding, 1990). The second protein component may behave like a filler, interspersed throughout the primary gel network. The gel may be single phase with the filler remaining soluble (Figure1.9A), or two phase, where thermodynamic incompatibility causes phase separation to occur, with the filler existing as dispersed particles of liquid or as a secondary gel network (Figure 1.9B). The single phase system was labelled by Tolstoguzov (1986) as type I and the phase separated system as type II filled gel. In Figure 1.9C, the non-gelling component associates with the primary network in a random fashion via nonspecific interactions which may reduce the flexibility of the primary network chains and add rigidity to the gel. The two protein components may co-polymerise to form a single, heterogenous network (Figure 1.5D). An example of this type is the polymerisation which occurs in BSA-ovalbumin gels (Clark et al., 1982). Figure 1.9E is an example of the interpenetrating polymer network where both components form separate continuous network throughout the system.
Types of mixed gels (Ziegler and Foegeding, 1990)
Protein-polysaccharide interactions in food systems often plays a role in determining the functional properties of these systems (Stainsby, 1980). Understanding the mechanisms involved in the interactions between proteins and polysaccharides and the way in which these interactions are affected during processing is important when these components are added into foods to improve their functional properties (Ledward, 1979; Stainsby, 1980). Although there is evidence to indicate that the major forces responsible for these interactions are electrostatic in nature (Imeson et al., 1977), other interactions such as hydrogen, hydrophobic or covalent bonds may also be significant in the stabilization of the interaction.
Proteins and polysaccharide combinations have been used to stabilize emulsions (Dickinson and Euston, 1991; Dickinson and Galazka, 1991) through electroststic interactions and when covalently linked (Kato et al., 1990; Dickinson and Galazka, 1991; Dickinson and Galazka, 1992). The formation of soluble protein-polysaccharide complex in the pH range that would lead to protein precipitation has been utilised in the preparation of fruit flavoured milk beverages in which fruity flavour is best expressed at pH 4.5-5.0. CMC has been found to be particularly effective in keeping milk proteins in solution. In addition to holding milk proteins in solution, excess polysaccharide may resolubilize a precipitated complex at low pH. This behaviour is explained as a result of two reactions (Hidalgo and Hansen, 1969): a primary ionic reaction leading to the formation of insoluble complex, followed by a 'peptization' reaction, involving redistribution of protein molecules on the polysaccharide, giving rise to increased hydration and thus solubilization. The protein-polysaccharide interaction also inhibits precipitation of some water soluble proteins following denaturation (Hidalgo and Hansen, 1969).
Georges-Louis Friedli, PgDip., MSc.,CH(t), MACN, PhD.
Last updated September 1996
Blow, D. M., Birktoff, J. J. and Hartley, B. S. (1969). Role of a Buried Acid Group in the Mechanism of Action of Chymotrypsin. Nature 221; 337-340
Branden, C. and Tooze, J. (1991). Introduction to Protein Structure. Garland Pub. New York.
Clark, A. H., Richardson, R. K., Robinson, G., Ross-Murphy, S. B. and Weaver, A. C. (1982). Structure and mechanical properties of agar/BSA co-gels. Prog. Food Nutr. Sci. 6; 149-160.
Dickerson and Geis, 1969)
Dickinson and Euston, 1991;
Dickinson and Galazka, 1991;
Dickinson and Galazka, 1992
Deisenhofer and Steigemann, 1975;
Dunkerley and Hayes, 1980;
Dunkerley and Zadow, 1984)
Hidalgo, J. and Hansen, P. M. T. (1969). Interaction between food stabilizers and beta-lactoglobulin. J. Agric. Food Chem. 17; 1089-1092.
Imeson et al., 1977
Kato et al., 1990;
Lee and Richards, 1971
Manson and Sperling, 1976
Pauling et al. (1951
Plock et al., 1992)
Ramachandran et al., 1963
Richardson and Ross-Murphy (1981)
Schmidt, R. H. and Illingworth, B. L. (1978). Gelation properties of whey protein and blended protein systems. Food Product Development 12(10); 60-64.
Schulz, G. E. and Schirmer, R. H. (1990). Principles of Protein Structure. In 'Springer Advanced Texts in Chemistry'. (Ed. C. R. Cantor), Springer-Verlag, New York.
Stainsby, G. (1980). Proteinaceous gelling systems and their complexes with polysaccharides. Food Chemistry 6; 3-14.
Tolstoguzov, V. B. (1986). Functional properties of proteins-polysaccharide mixtures. In Functional properties of Food Macromolecules. (Eds. J. R. Mitchell and D. A. Ledward)p.385. Elsevier. Amsterdam.
Watenpaugh, K. D., Sieker, L. C. and Jensen, L. (1979). The Structure of Rubredoxin at 1.2Å Resolution. J. Mol. Biol. 131; 509-522.
Wright, H. T. (1973). Comparison of the Crystal Structures of Chymotrypsinogen-A and alpha-Chymotrypsin. J. Mol. Biol. 79; 1-23.
Ziegler, G. R. and Foegeding, E. A. (1990). The Gelation of Proteins. Adv. Food Nutri. Res. 34; 203-298.