TUTORIAL ON PEPTIDE AND PROTEIN STRUCTURE

III. TERTIARY STRUCTURE

                                                Copyright: J. E. Wampler, 1996

Tertiary Structure is the folding of the total chain, the combination of the elements of secondary structure linked by turns and loops. Its stability is determined by non-bonding interactions & the disulfide bond.

This section covers:

Interactions in proteins.
The hydrophobic effect and protein structure.
Characteristics of the folded protein.
Folding motifs.

The interactions in proteins that maintain structure cover the spectrum of chemical interactions from the covalent bonds of the amide backbone and the disulfide bridge through very polar charge-charge interactions to a variety of weaker, short distance interactions.

Covalent bonds have been discussed in Section I of this tutorial under amide bonds and crosslinks.

In vacuo, two point charges interact as described by Coulomb's law:

Interaction Potential, V = q1 q2/(Er)

where q1 and q2 are the two charges, E is the dielectric constant of the medium separating them and r is the separation distance.

For a protein in solution, the interactions of charges are more complex than for two charged points in a vacuum because the dielectric constant varies from water (E~80) to the protein interior (E ~ 4). Here the interaction energy is described by the Poisson-Boltzmann equation.

If two oppositely charged atoms are free to move under the force of their attraction, they are drawn into "contact" as defined by the sharply rising repulsion that occurs as their electron clouds start to overlap. As a first approximation, this closest approach distance can be viewed as the contact of hard spheres with radii determined by the atomic number and electronic configuration of the atoms, i. e. the atomic or ionic radii. These radii are often referred to as van der Waal's radii or non-bonded radii.

As the non-bonded interaction between atoms and groups involves less than full formal charge and involves polarization contributions, the distance dependence falls of more quickly than the 1/r dependence of Coulomb's law. In these more complicated cases, where the charges can not be represented by single point locations, the interactions are also less isotropic, falling off not just as a function of distance, but also as a function of orientation:

Distance and Angle Dependence of Non-bonded Interactions.

 With fixed magnitude charges:
     POINT CHARGE with POINT CHARGE  1/r
     POINT CHARGE with DIPOLE        Cos (angle) x 1/r2
     DIPOLE with DIPOLE              F(angle)* x 1/r3
     POINT CHARGE with QUADRAPOLE   ~1/r3
     DIPOLE with QUADRAPOLE         ~1/r4
     QUADRAPOLE with QUADRAPOLE     ~1/r5

  With polarizable charge centers:
     POINT CHARGE  w/  POLARIZABLE DIPOLE ~ 1/r4
     INDUCED DIPOLE-DIPOLE                ~ 1/r6
     INDUCED DIPOLE-OCTUPOLE              ~ 1/r8
     INDUCED QUADRAPOLE-QUADRAPOLE        ~ 1/r10


* where F(angle) is a function of the cosines and sines of the angles 
between the dipole moments and the separation vector.

Some specific interactions:

Hydrogen-bonding:

Hydrogen bound to Oxygen, Nitrogen and Sulfur has a smaller van der Waal's radius and more partial charge than hydrogen on carbon making the interaction between a charge or dipole with such "donar groups" stronger and more orientation dependent.
Aromatic-Aromatic Interactions:

The aromatic amino acid sidechains of phenylalanine, tyrosine and tryptophane prefer interactions with interplane angles of around 90o. The interaction has a quadrupole-quadrupole character (distance dependence ~1/r5) with dependence on the interplane angle.

Aromatic-Polar/Charged-Group Interactions:

The charge separation of aromatics with ring hydrogens having partial positive charges and the pi electron system being partially negative, leads to interactions with O, S and N groups in proteins which are relatively strong (distance dependence ~1/r3 to 1/r4) and orientation dependent.
The database of know structure can also give us information about the potential interactions within a protein that help maintain its conformation. J. Singh & J. M. Thornton published an " Atlas of Protein Side-Chain Interactions" (Vols. I & II, IRL Press, Oxford, 1992) which is now available on-line in an updated version.

A similar " Atlas of Side-Chain and Main-Chain Hydrogen Bonding" is also available, published by I. McDonald and J. M. Thornton.

Return to the top.

The so-called "Hydrophobic Effect" is the name given by Tanford to the entropic driving force for self-association of non-polar groups in an aequeous environment. The weak interactions of non-polar groups for each other and with water are approximately equal, but considerable entropy is lost if each hydrophobic group is "caged" by a layer of hydrogen bonded water molecules. Self-association of the non-polar groups, frees this water to the bulk solvent. A number of studies indicate that proteins fold to limit the solvent accessible surface of non-polar groups. The data below shows how the non-polar amino acid side chains tend to be buried in protein structure.

The anomalies in this plot are easily explained. Proline tends to play a role in loops and turns (see Section II) and cysteine forms disulfide bridges (see Section II).

Return to the top.

The tertiary structure of proteins is characterized by tightly folded structure with polar groups on the surface and non-polar groups buried.

Natural variation from species-to-species tends to favor changes in surface (and therefore polar) groups.

Structure is determined globally and redundantly. Upto 30% of the amino acids in some proteins have been changed to alanine with little change in the folded structure.

Return to the top.

There are a number of ways to represent the folding of a protein and the arrangement of secondary structure elements within the tertiary structure. While these simplifications don't show the sidechain and mainchain interactions that hold the structures together, they do reveal the folding pattern. Examination of such diagrams reveals recurring structural patterns in protein folding.

In this case, alpha helicies are red circles, columns and twists. Beta sheets are green squares, arrows and ribbons with arrows.

At the simplest level, proteins can be classified by their content of secondary structure. Typically, the focus is on alpha-helicies and beta- sheets. However, there are some proteins where turns and disulfide bonds seem to be more important considerations.



Consider wheat germ agglutinin, where 16 turns (in cyan) and 16 disulfide bonds (in brown) are the predominant structural contributions. While a large part of the structure has no classified secondary structure (in black), there are still four structurally homologous domains related by symmetry.



Some proteins are made up of mostly alpha helicies:

Both marine bloodworm hemoglobin (left) and E. coli cytochrome B562 (right) are composed of mostly alpha helicies. The 4 helix bundle of the cytochrome is a common motif.

Some are mostly beta sheet:

The green alga plastocyanin (left) and sea snake neurotoxin (right) are mostly beta sheets.

Of course, many proteins are a mix of alpha helicies and beta sheets.

Two simple proteins with a mix of 2o components: ribonuclease T1 (left) and pancreatic trypsin inhibitor (right).

The beta sheet core of carbonic anhydrase (left) and H-RAS P-21 protein (right) is often referred to as a beta saddle and it is found in many structures.

The beta barrel found in the triose phosphate isomerase (left) and xylose isomerase (right) structures is another very common motif.

Return to the top.

Go back to Introduction.
Go back to Part II.
Continue to Part IV.