12/12/24
Conforma(onal Analysis
• The physical, chemical and biological properties of a molecule depend critically upon the three-
dimensional structures, or conformations, that it can adopt.
• Conformational analysis is the study of the conformations of a molecule and their influence on its
properties.
• The conformations of a molecule are traditionally defined as those arrangements of its atoms in
space that can be interconverted purely by rotation about single bonds.
64
64
Conformational Analysis
Conformational Search
• A key component of the conformational analysis is the conformational search, the objective of
which is to identify the ‘preferred’ conformations of a molecule, those conformations that
determine its behavior. This usually requires us to locate conformations that are at minimum
points on the energy surface.
• Energy minimization methods therefore play a crucial role in conformational analysis.
• An important feature of methods for performing energy minimization is that they move to the
minimum point that is closest to the starting structure. For this reason, it is necessary to have a
separate algorithm which generates the initial starting structures for subsequent minimization.
• Conformational search methods can be conveniently divided into the following categories:
Ø Systematic search algorithms
Ø Model-building methods
Ø Random approaches
Ø Distance geometry
Ø Molecular dynamics
65
65
1
12/12/24
Conformational Analysis
Conformational Search Methods
• Systema(c Search Methods:
• As the name suggests, a systemaIc search explores the conformaIonal
space by making regular and predictable changes to the conformaIon.
• The simplest form of systemaIc search is called ‘a grid search’. In grid
search, first all rotatable bonds in the molecule are idenIfied. The bond
length and angles remain fixed throughout the calculaIon. Each of these
bonds is then systemaIcally rotated through 360° using a fixed increment.
Every conformaIon so generated is subjected to energy minimizaIon to
derive the associated minimum energy conformaIon. The search stops
when all possible combinaIons of torsion angles have been generated and
minimized.
• A major drawback of the grid search is that the number of structures to
be generated and minimized increases in an exceptional fashion with the
number of rotatable bonds, a phenomenon known as the combinatorial
explosion.
66
66
Conformational Analysis
Conforma3onal Search Methods
• Model-building methods:
• One way to at least partially alleviate the combinatorial explosion that accompanies a systematic
search is to use larger ‘building blocks’, or molecular fragments, to construct the conformations.
• Fragment- or model-building approaches to conformational analysis construct conformations of a
molecule by joining together three-dimensional structures of molecular fragments.
• This approach would be expected to be more efficient than the normal systematic search
because there are usually many fewer combinations of fragment conformations than
combinations of torsion angle values.
67
67
2
12/12/24
Conformational Analysis
Conformational Search Methods
• Random approaches:
• A random search is, in many ways the opposite of a systematic search. A random search can
move from one region of the energy surface to a completely unconnected region in a single step.
• A random search can explore conformational space by changing either the atomic Cartesian
coordinates or the torsion angles of rotatable bonds.
• Distance geometry:
• One way to describe the conformation of a molecule other than by coordinates is in terms of the
distances between all pairs of atoms. Distance geometry explores conformational space by
randomly generating many distance matrices, which are then converted into conformations in
Cartesian space.
68
68
Conforma(onal Analysis
Conformational Search Methods
• Molecular dynamics
• MD is widely used for exploring conformational space. During such a simulation, the system is
able to overcome energy barriers and so explore different regions of the conformational
space.
• A common strategy is to perform the simulation at a very high, physically unrealistic
temperature. The additional kinetic energy enhances the ability of the system to explore the
energy surface.
69
69
3
12/12/24
Protein Structure Prediction
Different Levels of Protein Structure
70
70
Protein Structure Prediction
Protein Architecture
• Proteins consist of amino acids linked by pepIde bonds.
• Each amino acid consists of:
• a central carbon atom
• an amino group
• a carboxyl group and
• a side chain
• Differences in side chains disInguish the various amino acids.
• Amino acid side chains have different sizes, shapes, hydrogen-
bonding capabiliIes and charge distribuIons, which enable
proteins to display the vast array of biological funcIons required
by living systems.
71
71
4
12/12/24
Protein Structure Predic(on
Protein Architecture
• The information for 3D structures is coded in the protein sequence.
72
72
Protein Structure Prediction
• The biological function of a protein or peptide is often intimately dependent upon the
conformation(s) that the molecule can adopt.
• In contrast to most synthetic polymers where the individual molecules can adopt very different
conformations, a protein usually exists in a single native state. These native states are found
under conditions typically found in living cells. Proteins can be unfolded (or denatured) using
high-temperature, acidic or basic pH or certain non-aqueous solvents. However, this unfolding is
often reversible and so proteins can be folded back to their native structure in the laboratory.
• X-ray crystallography and NMR are the methods most widely used to provide detailed information
about protein structures. Unfortunately, the rate at which new protein sequences are being
determined far exceeds the rate at which protein structures are determined experimentally.
• The general difficulties in obtaining protein structures using experimental techniques means that
there is considerable interest in theoretical methods for predicting the three-dimensional
structure of proteins from the amino acid sequence à protein folding problem
73
73
5
12/12/24
Protein Structure Predic(on
• If we ignore the small variations in bond angles and bond lengths then the conformation of an
amino acid residue in a protein or peptide can be classified according to the torsion angles of its
rotatable bonds.
• There are three backbone torsion angles, labelled f, y and w.
74
74
Protein Structure Prediction
• The conformations of the side chains are characterized by the torsion angles, c1, c2, etc.
• Subsequent investigations have revealed that the side-chain conformations are often correlated
with the backbone structure; for certain conformations of the backbone only particular side-chain
structures are possible.
• As more protein structures became available, it was observed that some contained more than
one distinct region, with each region often having separate function. Each of these regions is
usually known as a domain, a domain being defined as a polypeptide chain that can fold
independently into a stable three-dimensional structure.
75
75
6
12/12/24
Protein Structure Predic(on
Protein Folding
What stabilizes a given fold ?
76
76
Protein Structure Prediction
Protein Folding
• Driving forces for the protein structure stabilization:
77
77
7
12/12/24
Protein Structure Prediction
Protein Folding
• Driving forces for the protein structure stabilization:
78
78
Protein Structure Prediction
The Hydrophobic Effect and Protein Folding
• When you put a nonpolar molecule in water, the water molecules around that nonpolar surface
become much more ordered (organized) than in the bulk liquid. These water molecules,
therefore, become “ice-like” -- highly organized -- in an effort to maximize their hydrogen bonds
(and minimize the unfavorable effects of disrupting the hydrogen bonds to the water molecules in
order to interact with the nonpolar surface).
• When two non-polar groups associate, then the water molecules surrounding the nonpolar side
chains are released into bulk water, which, like ice melIng, results in increased disorder
(ΔSsurroundings >> 0).
79
79
8
12/12/24
Protein Structure Prediction
The Hydrophobic Effect and Protein Folding
• Water soluble globular proteins usually have an interior composed almost entirely of non-polar,
hydrophobic amino acids such as phenylalanine, tryptophan, valine and leucine with polar and
charged amino acids such as lysine and arginine located on the surface of the molecule.
• This packing of hydrophobic residues is a consequence of the hydrophobic effect, which is the
most important factor that contributes to protein stability. The molecular basis for the
hydrophobic effect considered to be the entropy change of the solvent.
80
80
Protein Structure Prediction
The Hydrophobic Effect and Protein Folding
• The contribu,ons to the overall free energy of folding due to the packing of the
non-polar amino acids:
1. The enthalpy change to remove the non-polar amino acids from water is posiIve due to
dipole/induced dipole electrostaIc interacIons between the polar water molecules and
the hydrocarbon side chains. When packed there are only (weaker) dispersion interacIons
between the side chains à unfavorable
2. The entropy change associated with packing the amino acids is negaIve because the
unfolded state is less ordered than the packed state à unfavorable
3. The enthalpy change of the water is negaIve as the disrupIon to the hydrogen bonding
network is less for the folded protein à favorable
4. The entropy change of the water is posiIve. (When the protein brings the hydrophobic side
chains together, then the water molecules surrounding the nonpolar side chains are
released into bulk water, which results in increased disorder (discussed in the previous
slide). The area of the non-polar interface is much larger for the unfolded protein and so
the entropy change of the water is large (and posiIve) à highly favorable
81
81
9
12/12/24
Protein Structure Predic(on
The Hydrophobic Effect and Protein Folding
• Of these four contributions, the two enthalpy terms are believed to be small, with the entropy
change associated with the ordering of the solvent water molecules being the dominant term.
82
82
10
