Proteins perform many of the functional roles necessary for living systems

• The function of most proteins is based upon the unique conformation that their polypeptide chain adopts in solution (i.e. the "native" or "folded" conformation)
• The folded conformation is determined by the primary sequence and the interaction of amino acid side chains with the solvent.
• The pH, ionic composition and concentration, and the solvent dielectric, can influence the electrostatic interactions that stabilize the folded conformation

Non-covalent interactions stabilizing protein structures:

• Hydrogen bonds. Structural studies of folded proteins indicate that hydrogen bonding groups almost always find an appropriate partner. Hydrogen bonding groups include the main chain carbonyl and amide groups as well as polar side chains. Polar groups exposed on the surface of proteins often have water as their hydrogen bonding partner. Polar groups within the core region usually form hydrogen bonds with other groups within the protein. Most proteins include at least some buried solvent groups, and these hydrogen bond with side chains and/or main chain groups in the interior of the protein. Mutational studies have shown that elimination of a hydrogen bonding partner is often destabilizing to a protein structure, and hydrogen bonds typically contribute about 12kJ/mol in stabilization energy.
• Hydrophobic interactions. A primary driving force for protein folding involves the removal of non-polar side chains from solvent exposure. This is accomplished by sequestering them within the core region. Related to this, interior packing (i.e. van der Waals forces) is optimized by appropriate choice of non-polar side chains in the primary sequence. Exposing non-polar groups to solvent is entropically costly (due to water clathrate structure), thus, folding of the polypeptide chain so as to sequester nonpolar sidechains within the core region is "hydrophobically driven". Some non-polar groups are still found on the surface of folded proteins.

This is a "slice" through the center of a small protein known as fibroblast growth factor. Notice that the core region is rich in non-polar amino acids (white atoms are carbons). Polar groups (red oxygen and blue nitrogen) are rich on the surface, and H-bond with solvent. Some polar groups can be found in the interior, and some non-polar groups can be found on the surface, however.

• Electrostatic interactions. These interactions between oppositely charged ionic side chains are also known as "salt-bridges". The main chain amino and carboxyl terminal are fully ionized at physiological pH, as are the side chains Asp (-), Glu (-), Lys (+), and Arg (+). Histidine can also be charged (+) at pH <=6.0. Opposite charge attraction is modulated by the dielectric constant of the environment. Charge groups on the surface experience a dielectric constant of 78.5 (that of water) and are therefore weaker than those in the hydrophobic core (with a dielectric of ~4). Thus, any buried electrostatic interactions are quite strong. The presence of ions in solution can also screen electrostatic charges and weaken them.
• Van der Waals interactions. Well packed hydrophobic cores of proteins represent optimized van der Waals interactions between non-polar residues. Although individually weak, numerous neighbor interactions in such central cores can contribute a significant stabilization to the native structure.

 This is a "slice" through the center of a small protein known as fibroblast growth factor. Notice that the core packing amino acids (which are primarily hydrophobic) pack closely together with minimal cavity spaces. This maximizes the strength of the van der Waals interactions.

Understanding how proteins fold up is a major challenge in modern biochemistry ("The Protein Folding Problem")

• While the peptide bond is rigid, there is a single bond between the C
_a(1) and C atoms and the N and C(2) which are free to rotate. Thus, while the C_a (1) - C (O) - N (H) - C_a (2) atoms are all planar, this planar peptide bond has two degrees of rotational freedom.
• These two rotation angles are known by the greek letters
F (phi) and Y (psi). Y is the rotation angle about the C_a (1) - C bond and F is the rotation angle about the N - C_a (2) bond. These are also referred to as "main chain" angles.

• It is these two degrees of rotational freedom that allows polypeptides to fold up into unique conformations

The overall three-dimensional structure of a protein is a consequence of the unique F (phi) and Y (psi) angles that each peptide bond adopts

Some angles for F (phi) and Y (psi) are prohibited due to steric clashes with other atoms in the adjacent peptide bond:

Here are some VRML files to highlight these steric problems:

• F
(phi) = 0° and Y (psi) = 0°. Steric clash with R1 carbonyl and R3 amide.
• F
(phi) = 180° and Y (psi) = 0°. Steric clash with R2 amide and R3 amide.
• F
(phi) = 0° and Y (psi) = 180°. Steric clash with R1 carbonyl and R2 carbonyl.
• F
(phi) = 180° and Y (psi) = 180°. One possible main chain configuration with no bad steric clashes.

Ramachandran and coworkers came up with a diagram that shows the allowed regions for main chain F (phi) and Y (psi) angles. It is known as a Ramachandran Plot (dark areas are "low energy", or favored regions for particular combinations of F (phi) and Y (psi) angles):

a-helix and b-sheet secondary structure and the Ramachandran plot

a-helices and b-sheets are types of secondary structure that have characteristic values for main chain F (phi) and Y (psi) angles that are repeated throughout a region of the polypeptide chain.

a-helix

• The
a-helix has typical main chain angles of F (phi) = -60° and Y (psi) = -47°. This does not result in a steric clash of main chain groups (here is a vrml file with a representative structure)
• This regular arrangement of main chain angles results in the polypeptide backbone folding up into a
right-handed helix. All the main chain carbonyl oxygens point "down" the helix, and all the main chain amide nitrogens point "up" the helix.
• Note that since each peptide bond has a dipole, and that since all the peptide bonds in an
a-helix are pointing in the same direction, the peptide dipoles are aligned in an a-helix and reinforce one another, stabilizing the helical structure, and these combined dipole interactions are referred to as the helical macro-dipole
• The
a-helix exhibits an intra-chain hydrogen bonding arrangement between the main chain carbonyl oxygen of group (i) and the main chain amide of group (i+4)
• The are
~3.6 residues per turn in the a-helix. The "average" helix in a protein comprises about 10 residues, or about 3 turns of the helix

Click here for a vrml file of an ideal a-helix

• The amino acid proline will disrupt (put a kink into) an
a-helix . Since it is an imino acid, the main chain amide is unavailable to participate in the intrachain hydrogen bond. The main chain carbonyl partner of the intrachain hydrogen bond will point out of the helix to fulfill its hydrogen bonding requirement
• There are less-common type of helices in proteins with 3.0 residues per turn (3₁₀ helix) and 4.4 residues per turn (4.4₁₆ helix). The subscript in this nomenclature indicates the number of atoms comprising one turn. The typical
a-helix would therefore also be known as a 3.6₁₃ helix (13 atoms in one turn, 3.6 residues in one turn)

b-sheet

• Typical main chain angles for
b-sheets are F (phi) = -120° and Y (psi) = 120°. This does not result in a steric clash of main chain groups (here is a vrml file with a representative structure)
• These values for the main chain angles results in an
alternating up-down orientation to consecutive side chain functional groups
• Two
polypeptide chains adopting these main chain angles can hydrogen bond together in either a parallel, or anti-parallel orientation (resulting in a parallel b-sheet or an anti-parallel b-sheet ). Note that this requires two strands to get together, therefore, the H-bonds are inter-strand (rather than intra-strand as with the a-helix )

Click here for a vrml file of a parallel b-sheet

Click here for a vrml file of an anti-parallel b-sheet

Turns are extremely important structures in globular proteins. Without turn structures, the polypeptide secondary structures would simply continue and you would have a fibrous protein. Thus, globular proteins can be described as segments of secondary structure that are interrupted by turns, and this allows a globular tertiary structure to form.

• A common type of turn is the
b-turn.
• It achieves a 180° turn with three amino acids
• The main chain amide of the first residue hydrogen bonds with the main chain carbonyl of the third residue, and can connect the ends of an anti-parallel
b-sheet
• Residues such as proline and glycine occur often in turns. They can be readily accommodated in tight turns, and tend to destabilize other types of secondary structure (particularly
a-helix )

Click here for a vrml file of a b-turn

The characteristic main chain F (phi) and Y (psi) angles for the different types of secondary structure puts them in characteristic locations in the Ramachandran plot:

An example of a Ramachandran plot of a real protein (2,5-diketo-D-gluconate reductase)

This protein (2,5-DKG for short) is an enzyme involved in carbohydrate biosynthesis. It contains about 270 amino acids, and if we were to plot the location of each amino acid's F (phi) and Y (psi) angles on a Ramachandran plot we would find the following:

• The plot suggests that the protein contains both
a-helices and b-sheets, but it looks like there must be more a-helical secondary structure overall in comparison to b-sheets. It also looks like there may be several b-turns in the protein as well.

The actual tertiary structure of the protein (as diagrammed using a main-chain "ribbon" drawing) looks like this:

• The structure is consistent with our expectations from the Ramachandran plot - lots of
a-helix, some b-sheet and some turn regions.

Send to Friend Bookmark