Sec. I - PROTEIN STRUCTURE

Karp ch 2. and these notes

Topics: weak interactions, structure, stability, and relevant concepts of energetics.

The functions that a cell can perform are almost totally determined by which proteins it is capable of synthesizing because proteins are the molecular machinery of cells and, in multicellular organisms, of the extracellular environment. Throughout this course we will have the opportunity to talk about many such functions and the proteins responsible for them, whether they behave as enzymes, as membrane transporters of many kinds, as receptors of chemical signals, as signals (e. g. insulin), and a very large etc. We will see many examples of the precise relationship between protein function and its structure, and will hopefully become convinced that this structure determines the function.

A. Weak Interactions (see pages 36-39)

• Weak Bonds - various types of electrostatic interactions that hold atoms together with bond energies of less than 30 KJ/mol. (7.170 Kcal/mol). Include:
• Hydrogen bonds (H-bonds)
• Ion-ion interactions
• Dative bonds
• Van der Waals (Ion-dipole, dipole-dipole, London dispersion)

Whereas electrostatic forces between two ions vary inversely with the square of the distance between the charges, forces that involve dipoles and induced dipoles vary inversely to the fourth, fifth, sixth or seventh power of the distance, so they rapidly loose effectiveness as the distance increases. The presence of water weakens all electrostatic bonds due to its high dielectric constant of 80.

H-bonds consist of the electrostatic attraction between the partial positive charge in the H of an OH- or an NH- group and the partial or full negative charge of a nearby O or N atom. When the three atoms involved are aligned, the strength of the bond is maximal because the distance between the two electronegative atoms is the longest for a given distance between the H and the acceptor atoms.

• Hydrophobic Interactions - tendency of hydrophobic surfaces to interact with each other rather than with water because of the strong organization they introduce in the structure of liquid water. If a protein molecule has two regions with hydrophobic surfaces, the association of those two surfaces has a negative delta G because, in delta G = delta H-Tdelta S, the delta S of the process has a relatively large positive value due to the large degree of order lost by the water molecules squeezed away from the hydrophobic surfaces of A as they join in the formation of the dimer (see Quaternary Structure below for more details).

B. Structure (See pages 51-66)

Proteins are linear arrays of amino acid molecules interconnected by peptide bonds. These are amido links between the carboxyl group of one amino acid and the amino group of the next one in the chain.

Levels of Structural Organization

• Primary - amino acid composition and sequence

• Secondary - local arrangement of segments of the peptide backbone

• Tertiary - 3-dimensional shape of the entire chain determined by the folding pattern

• Quaternary - number and relative position of the various subunits in a multimeric protein.

X-ray diffraction studies (X-ray crystallography) is the technique that has contributed most to our understanding of the structure of biological (and non-biological as well) molecules.

I. Secondary structure

• The peptide bond. X-ray diffraction studies on crystals of amino acids and of dipeptides by Pauling and Corey (1930's) led to the following characterization:

• Short C-N bond distance of the peptide bond (1.32 A vs. 1.46 A for normal C-N bonds), 120 degree angles formed by the bonds formed by carbon and nitrogen, and co-planarity of the six atoms around it could be justified by a partial double bond characteristic of the peptide bond, which is independently predictable by resonance and by atomic orbital theories.
• Main secondary structures. Additional studies by Pauling and Corey (1951) on X-ray diffraction of fibrous proteins, model building taking into account the properties of the peptide bond, and theoretical calculations of the stability (minimum G content) of probable structures, led to characterization of the alpha-helix first, and of the beta-sheet later as follows:
• alpha-helix, found abundantly in alpha-keratins such as hair, wool, etc. (See Fig 2.30)
• The backbone, formed by the amino N, the alpha-C and the carboxyl C of each consecutive residue, is coiled forming a right handed helix with a 5.4A pitch, a 1.5 A rise per residue and 3.6 amino acid residues per turn. The side chains ( R groups) project away from the helix.
• Each carboxyl O is H- bonded to the amino group of the fourth residue below, which is almost perfectly alined for this purpose. This arrangement determines that each residue will form two H-bonds, except for the first and last four which will form only one. This large number of H-bonds at 25 KJ/mol each represents a significantly large free energy of stabilization.
• The right handedness of the helix contributes some dextrorotatory power to solutions of proteins which contain alpha-helices. This will reduce the levorotatory power of the l-amino acids that form natural proteins.
• Some amino acids (ala, glu, leu and met) are good alpha-helix formers, whereas others (pro, gly, tyr and ser) are poor.

• beta-sheet, found abundantly in beta-keratins such as silk, spider webs, etc. (See Fig 2.31)
• The backbone is more stretched and follows an almost fully extended zig-zag shape with the alpha-C's at the vertices of the zig-zag.
• Sheets that remind of corrugated paper are formed by various segments of the chain arranged in either a parallel or antiparallel relationship to each other. Each strand forms H-bonds with the two adjacent strands between the carboxyl O and amino groups which are well alined. Each amino acid residue in an internal strand will form two H-bonds.
• The amino acid side chains stick perpendicular to the plane of the sheet both above and below.

• Other secondary structures. A small number of loops ( hairpin, omega and others) have been characterized and are frequently found joining other types of secondary structures. In globular proteins, these loops are mostly found at the surface. Other types of helices are rarely found in proteins. Simple arrangements of several secondary structure elements with a well defined geometry are found repeated in different proteins with almost no modification. These so called motifs are quite frequent and sometimes are associated with a specific protein function (see fig.2.38).

II. Tertiary Structure 

A polypeptide chain folds in a specific way to form compact globular structures. A protein may consist of one or more such structures which, if more than one, are referred to as domains (see Fig 2.36 ). Domains may contain various motifs. They have the property of independently folding into a specific and stable configuration. Frequently, different domains in a protein are associated with different functions, and a given domain may be found as part of different proteins in which it performs the same function.

The folding patterns of proteins with either single or multiple domains are characteristic and determined to a large extent by hydrophobic interactions between their amino acid side chains. These folding patterns can be further stabilized by a variety of weak bonds and, in some proteins, covalent disulfide bonds (-S-S-) formed by oxidation of the -SH groups of two cysteine residues that could be remote in terms of primary structure, but proximal due to the folding of the chain (see Fig 2.43).

In order to yield enough information for the accurate description of the spatial distribution of atoms in the molecules of a compound, the material subjected to X-ray diffraction must be a crystal as close to perfection as possible. The crystallization of proteins, even from highly purified solutions, is difficult and, in many cases, not yet possible. For this reason, the detailed structures of only about 1000 proteins have been determined to date. Pioneers in these attempts were Max Perutz (hemoglobin) and John Kendrew (myoglobin), both of whom worked during the 1940's and 1950's in the X-ray diffraction laboratory established earlier by Bragg in Cambridge at that time under the direction of J. D. Bernal.. (Bernal, along with Dorothy Crowfoot, had demonstrated in the mid 1930's that X-ray diffraction could be used to elucidate the structure of proteins). Clearly, the findings of Pauling and Corey in the late forties and early fifties provided a fundamental contribution to the work of Kendrew and Perutz. It was not coincidental that, also in the early 1950's, two other young men called Watson and Crick were working on DNA in the same laboratory.

III. Quaternary Structure

Some proteins consist of more than one polypeptide chain or subunit. These are also called multimeric proteins, and may be formed by several identical polypeptides or by different ones. Quaternary structure of a protein is the number, structures and relative positions of all its subunits (see Fig 2.40). The formation and stability of these sometimes large structures with many subunits depends on weak interactions.

Self-assembly. In a similar fashion, proteins sometimes form very large assemblies as collagen fibers, viral coats, contractile fibers in muscle cells, etc. The assemblage of these structures is, in many instances, spontaneous, i.e., not dependant on the expenditure metabolically derived energy by the cell. Click on self assembly for an explanation of the energetics of this process.

C. Chemical Modification of Amino Acid Residues

Certain proteins are chemically modified after the polypeptide synthesis has been completed and this modification is essential in order to achieve the proper functional capacity. Those proteins destined to be inserted in the plasma membrane or to be secreted by the cell undergo glycosylation (addition of sugar residues through glycosidic linkages) of amino acid side chains that have either a hydroxyl group (serine, threonine) or an amido group (asparagine). This change happens in the ER and Golgi, as will be seen later in the course. The phosphorylation of specific side chains with hydroxyl groups (serine, threonine, tyrosine) in some proteins accomplishes a major change in their functional capacity. This property is the basis for the mode of regulation of cellular activities by hormones, neurotransmitters and other chemical signals through the activation or inactivation of protein kinases and phosphatases, as will also be seen later in the course.

D. Stability of Native Protein Structures (See pages 64-65)

Proteins have long been known to undergo drastic changes in physical and biological properties when treated by some physical or chemical agents (heat, acids, bases, high concentrations of urea, etc.). A consequence of these treatments is a loss of the normal biological function of the protein, which is said to be denatured. Sometimes these changes could be reversed by removing their cause. On the basis of the knowledge of protein structure that existed by 1960, it was clear that, certainly for those changes in the protein's function that are reversible, these changes must correspond to structural changes due to altered weak interactions. In other words, some aspects of quaternary, tertiary and secondary structure.

In the early 1960's, Christian Anfinsen conducted experiments with ribonuclease, which is a small protein that has four disulfide bonds (see Fig. 2.43).

• Experiment 1. Denature ribonuclease with 8 M urea (which disrupts hydrogen bonds and hydrophobic interactions) and beta-mercaptoethanol( which reduces disulfide bonds to two -SH groups). Various physical measurements indicate that the secondary and tertiary structures of the enzyme have been greatly destroyed and that the protein is close to the state of random coil. At this point, all the enzymatic activity has been lost. Then, remove the urea by dialysis. Even though a great deal of the physical properties are then restored, the enzymatic activity is not. Subsequent removal of the beta-mercaptoethanol (BME) and exposure to oxygen, which will oxidize the -SH groups, recovers almost 100% of the activity. This strongly suggests that the disulfide bonds are essential to bring together the parts of the protein involved in catalysis.
• Experiment 2. From the denatured enzyme, remove the BME and expose to air while still in the presence of urea. This results in the oxidation of the -SH groups to form -S-S- bonds. Now remove the urea. A measurement of activity yields less than one percent of the original activity, which indicates that less than 1% of the molecules of the enzyme have recovered the native structure. This state of the enzyme was called "scrambled" enzyme. The difference between this experiment and the previous one is only the order in which the denaturing agents were removed, but their results are totally different. Still, physical measurements indicate that much secondary structure and some sort of tertiary structure have been restored. The presence of urea (which destroys hydrogen bonds and hydrophobic interactions and converts the chain into a random coil) while the -S-S- bonds are formed should be expected to allow all sorts of conformations of the chain to coexist. This leads to the conclusion that the eight -SH groups will be able to form -S-S- bonds following all possible combinations, (N = 112), based on which -SH groups happen to be adjacent at the time of their oxidation. These randomly achieved combinations will be found in equal amounts, but only one of them is the one found in the native conformation. Consequently, when urea is removed from the solution only 1/112 of the enzyme molecules have the right -S-S- bonds and, since these covalent bonds are quite stable, only they will have the freedom to fold so as to achieve the native conformation.
• Experiment 3. To the scrambled enzyme solution add catalytic amounts of BME so as to maintain a small concentration of this reducing agent in the presence of oxygen. The BME will slowly reduce the -S-S- bonds while oxygen will oxidize the -SH groups. In other words, a continuous process of breakage and formation of these bonds will ensue during which sections of the polypeptide chain will become, for brief periods of time, free of the restrictions imposed by the -S-S- bonds, and, consequently, free to modify their folding pattern. The result of this treatment is that, in time (10-12 hours), 100% of the enzymatic activity is recovered. Obviously, the 99% of the molecules whose -S-S- bonds were "scrambled" were capable of recovering their native conformation. Since this process is spontaneous, it must be concluded that the native conformation is stable, i.e., its G content is at a minimum.

The denatured enzyme obtained in experiment 1 retained only one characteristic of the native protein: its primary structure. Still, without any additional information it was able to spontaneously recover its native conformation. Consequently, it must be concluded that secondary and tertiary structures are determined by primary structure, i.e., that, for a given primary structure, the free energy of the native conformation is at a minimum and, therefore, having a high degree of stability. This type of experiment is successful with small proteins and with isolated small domains of large proteins, but not with large proteins or domains. Furthermore, this behavior in vitro is not the same as in vivo. The folding of proteins being synthesized in the cell is a complex process usually requiring the help of a group of other proteins collectively referred to as chaperones (see pgs 75-78).

BACK TO CELL PHYSIOLOGY