The third level of protein conformation is known as tertiary structure

1

• The third level of protein conformation is known
  as tertiary structure
• Tertiary structure depends on the chemical
  properties of the side chain on amino acids.
• That is polar, hyrophobic and charged side
  chains.
• Some of the forces that are thought to be
  involved are

2

• Van der Waals forces, electrostatic forces,
  hydrogen bonding. Disulfide bonds in proteins
  that have Cysteine residues.
• Chaperonins are proteins which help other
  proteins to fold correctly. Note, not all
  proteins require chaperonins to fold correctly.

3

• The hydrophobic effect is considered to be the
  driving force in proteins folding into a compact
  globular form.
• The native structure of most proteins includes a
  hydrophobic core consisting of hydrophobic
  residues
• While the outside of the protein has polar
  residues exposed to water.

4

• The process of folding into a compact form with
  the hydrophobic residues sequestered from water
  is known as hydrophobic collapse.
• A notable exception is the integral proteins that
  span cell membranes.
• These proteins have a hydrophobic coil embedded
  within the lipid layer of the cell membrane.
  They are not surrounded by polar residues.

5

• The significance of isolating hydrophobic
  residues is best demonstrate in sickle cell
  anemia.
• The hemoglobin protein is a tetramer consisting
  of two alpha-globin chains and two beta-globin
  chains.

6

• The mutation of glutamic acid to valine in the
  hemoglobin protein causes a hydrophilic region of
  the protein to become hydrophobic. In turn, the
  hydrophobic effect, drives this region into the
  hydrophobic core of the protein.
• Consequently, the beta-globin chain of hemoglobin
  stick to each other and distort the shape of the
  red blood cells.

7

• The sulfhydrl (-SH) groups on cysteine can be
  oxidixed to form covalent disulfide bonds.
  Typically the cysteine residues are located far
  from each other in the protein and folding of the
  protein brings them into close proximity.
• Experiments using ribonuclease reveal that
  disulfide bonds are important in stabilizing this
  protein.

8

• The number of possible conformations of a protein
  make it impossible to determine if the native
  structure of a protein is most stable.
• We assume that this is probably the case that the
  native structure of a protein is most stable.
• In the 1960s, the Levinthal paradox was
  advanced. It states proteins must fold into
  intermediate structures of increasing stability
  to reach the native structure.

9

• The possibility of a protein folding directly
  into the native structure randomly is highly
  unlikely. It was predicted that this would take
  years to occur.

10

• Computer algorithms have been developed to
  determine the tertiary structure of proteins.
• To simplify the problem, the alpha carbons of
  each amino acid are restricted to positions in a
  two dimensional or three dimensional lattice.
• This reduces the number of conformations of a
  protein.

11

• The H-P (hydrophobic-polar) lattice model
  represents each amino acid as either hydrophobic
  or polar. No other possibilities are allowed
  such as charged.
• The amino acids are circles of fixed radius so
  the size of the amino acids is ignored.
• Hydrophobic amino acids are colored black and
  polar amino acids are clear.

12

• The N terminal amino acid is at position (0,0) in
  the grid.
• Each H-H contact is scored -1. H-H contact
  between contiguous amino acids are ignored.
• So in the example, the score is -3.
• We are searching for a protein that has the
  maximum number of hydrophobic interactions.

13
(No Transcript)
14

• The previous protein can also be represented in a
  three dimensional matrix.

15

• An important consideration, in the H-P lattice
  model is the direction of the amino acids.
• Typically, the absolute direction representation
  is used for a 2D model.
• The choices are up (U), right (R), left (L), and
  down (D). While 3D includes the same as 2D as
  well as back and forward.

16

• For the 2D protein previously shown it would be
  R,R,D,L,D,L,U,L,U,U,R.
• There is a relative direction representation that
  includes the turns taken at each amino acid.
• A problem with these representations is that two
  amino acids can be placed at the same position,
  creating steric collision or bumps.

17

• A possible solution is to assign high energy to
  any configuration with bumps. Typically programs
  search for configurations with the lowest
  possible energy.
• Off-lattice models allow proteins to attain more
  realistic configurations. These models allow phi
  and psi rotations.

18

• These models work well for small polypeptides but
  larger polypeptides the results are mixed.
• These models may include alpha carbons only, all
  backbone atoms, or even all backbone and
  side-chain atoms.
• The side chains can be rigid, semi-flexible or
  fully flexible.

19

• For rigid side chains, the conformation of side
  chains from x-ray crystallography are observed.
• Semi-flexible side chains uses an empirical
  method. Data is taken from x-ray crystallography
  and grouped into similar shape groups.
• The average conformation of each group is a
  rotamer.
• Each side chain is allowed to adopt any of the
  most common rotamers
• This in turn allows different possible
  conformations to be analyzed.

20

• Some Off-Lattice models incorporate energy
  functions. Such as hydrogen bonding,
  electrostatic forces, disulfide bonds, van der
  Waals forces, solvent interactions and
  hyrdrophobic interactions.
• The energy from these forces is computed and
  added together to determine the most stable
  protein with least amount of energy.

21

• Unfortunately, these energy based models do not
  correctly predict protein conformations.
• The most likely reason is lack of understanding
  about the forces involved in protein folding.
• A more successful procedure is comparative
  modeling.

22

• This technique compares the target protein with
  proteins whose conformation is known.
• Comparative modeling uses the following steps
• 1. identify a set of protein structures related
  to the target protein (based on sequence data
  only)
• 2. align target sequence with template sequence
• 3. construct the model
• 4. model loops
• 5. model side chains
• 6. evaluate the model