Protein Folding

Protein Folding

1. Significance

   • Part of the central dogma
     • DNA --> RNA --> Sequence --> Fold --> Function
       • Interpretation: Sequence defines fold
         • Example - a small protein of 101 amino acids
           • In theory: 9^100 possible conformations
             • BUT in practice there are not this many. Each amino acid contributes specific chemical properties to the protein
               • This defines the way regions of the protein interact with one another and therefore the final fold
         • Because we know a lot of protein sequences and the corresponding native fold, it is possible to predict a protein's fold based on its sequence
   • The native fold is the structure of a protein that allows it to fulfil its function
     • i.e. it is not functional without this
   • Anfinsen, 1961
     • The amino acid sequence of a polypeptide chain contains all the information required to fold the chain into its native, 3D structure
     • Ribonuclease denatured with 8M urea and beta-mercaptoethanol
       • Always folds back into its original conformation when allowed to reoxidise in air
     • Small proteins
       • Those which are larger are often more complicated and may require chaperones etc

2. Kinetics and stability

   • Stability of denatured and native states
     • Due to the sum of all non-covalent interactions
       • Between atoms of the protein
       • Between protein and solvent
       • delta-G[D-N] is the difference in the sum of two very large nunbers
         • Free energy of unfolding
           • In considering factors that contribute to the stability of proteins we have to consider both the denatured state D and the native state N.
         • deltaGD - deltaGN
     • Can determine the change in free energy on mutation
       • m values
       • Modify key parts of the protein and see what happens to the free energy
   • Folding is thermodynamically controlled
     • Native state is thermodynamically stable
       • The process is reversible
         • Need an equilibrium which can be disrupted
           • Can only determine stability in equilibrium conditions
             • Many proteins can't be denatured reversibly!
               • Proteins fold and unfold co-operatively
                 • The only two states at equilibrium are D and N
                   • How are these populations measured and monitored?
             • Add different things to try and change the equilibrium conditions
               • HOW do we change the populations of D and N?
     • Entropy/enthalpy
   • Folding must occur rapidly
     • Levinthal's paradox (1968)
       • Folding not simply by protein trying every conformation until it finds the most stable one!
         • Would take too long
         • Unlikely to find the same conformation every time
       • Small protein of 101 residues
         • Each peptide bond, 9 main conformations
           • Amount to the rotation positions around each of the two bonds adjacent to the peptide bond itself
           • 100 peptide bonds
             • 9^100 possible conformations
               • Around 1090
               • Ignores conformations due to side chains
     • Time taken to switch conformation around 10^-13 sec
       • If protein started folding at the time of the earth's origin it would have explored around 1.4x10^30 conformations
         • Only a tiny proportion of those possible!
           • => Folding must occur via intermediates in one or several defined pathways
             • Cuts down the number of conformations that proteins need to search to find the native state
               • Folding funnels
   • There can be more than one energetically stable state for a protein
     • Amyloid
       • Misfolded protein which is very stable
         • Aggregation diseases

3. Hierarchical folding mechanisms

   • Uncouple the formation of secondary and tertiary structure
     • Simplify the search for the native state
   • Framework model
     • Sequential folding one residue at a time
   • Classical nucleation model
     • Central piece forms and the rest of the protein folds around it
   • Hydrophobic collapse model
     • All hydrophobic residues cluster in the centre of the protein
       • Protection from the external aqueous environment
     • The rest of the protein folds in around the hydrophobic residues
   • These models require intermediates!
     • Reduces the conformational space a protein has to search through
       • Solves the Levinthal paradox!
   • Alternative view: no intermediates
     • Jackson and Fersht, 1991
       • Chymotrypsin inhibitor 2
         • Stable intermediates not essential for fast and efficient folding of a protein
           • Extremely short-lived
     • New model: condensation/nucleation
       • Something happens in the centre of the molecule which causes the rest of the residues to fold in

4. Biased

   • Levinthal assumed unbiased search
   • Chemical properties of amino acids result in inherent preferences
     • Folding is more like a funnel than a flat filter
       • The native state is that which is most energetically favourable due to various intramolecular interactions
   • "Energy landscapes"
     • Certain pathways require more energy than others
       • A bit like a mountain range!
       • Accounts for the possibility of misfolding
       • Without the energy required to get out of "dips" could get stuck in incorrect pathways
         • This would lead to misfolding
         • Kinetic traps

5. Probes

   • Fluorescent amino acids
     • Use spectrophotometry to track the fluorophore
       • i.e. is it buried inside the protein or exposed?
         • This is a good indicator of whether or not the protein has folded
     • Try to observe fluorescence in a different direction to transmission
   • In vitro assays
     • Cuvette
   • In vivo assays
     • Under the microscope
   • Trp shifts when its environment changes
     • Easier to understand folding in proteins rich in these residues
   • Tyr changes intensity

6. Unfolding by denaturants

   • How do they work?
     • Preferential binding model
     • Preferential solvation model
     • Increase solubility
       • Polar side chains
       • Non-polar side chains
       • Protein backbone
       • Stabilise denatured state more than native state
         • Larger surface area
   • Common denaturants
     • GuSCN
       • GuHCl
         • Urea
           • Most common
   • Other denaturants
     • Temperature
     • Pressure
     • pH
     • Force
     • Salt

7. Experimental problems

   • Need to characterise all states along the way
     • Also conversion mechanism between them
     • Difficult
       • Dynamic
       • Heterogeneous
       • Transient
   • Measuring D-state
     • NMR
     • SAXS
   • Measuring N-state
     • Crystallography
     • NMR
   • Options
     • Trap semi-stable states
       • Hydrogen/deuterium exchange quenched flow
       • Sometimes partly folded states can be observed by equilibrium methods
         • May not be relevant to the pathway!
         • "molten globule" NMR
         • Native state hydrogen exchange
         • Peptide studies
     • Stopped flow
       • Pneumatic drive activation - two small volumes of solutions driven from high performance syringes through high efficiency mixer
         • Resultant mixture passes through measurement flow cell and into stopping syringe
           • Just prior to stopping steady state flow is achieved
             • Solution fills the stopping syringe and the plunger hits a block
               • Flow is stopped instantaneously
                 • Kinetics of the reaction can be measured in the cell

8. Representing protein folding

   • Chevron plot
     • Protein folding kinetic data in varying denaturant concentrations
       • Can depict protein folding or unfolding
         • Folding in one direction and unfolding in the other
     • Cm is denaturation midpoint
       • Either side are "limbs"
         • Linear limbs
           • 2-state model
             • m-values
             • Does NOT mean there are no intermediates!!!
               • The intermediate states are not significantly populated
                 • "kinetically invisible"
         • Non-linear limbs
           • Non-2-state model
             • Off-pathway intermediate?
             • Artifact?
             • Because it goes through more stages
     • V-shaped kinetic curve
       • Proteins denatured: all parts of protein are more soluble in denaturant
         • deltaG of transfer to denatured state is almost linear with [denaturant]
         • TS is the intermediate structure between denatured and native states
           • Stabilised by denaturant with respect to native structure
           • Destabilised with respect to denatured state
         • Denaturant lowers activation energy of denaturation
           • Raises activation energy of folding
             • Stabilisation energy linear with [denaturant]
   • Phi value analysis
     • Combine kinetic and equilibrium data to produce a ratio
       • Phi value
       • Analysis of a protein's TS for folding
         • Change intermediates by introducing specific mutations
           • Choosing mutations
             • Properties
               • Non-disruptive
               • Conservative
               • Should not change structure
               • Should not change the polar nature of the side chain
             • Good examples
               • Ile --> Val --> Ala
               • Phe --> Leu
               • Thr --> Ser
               • Asp --> Asn
             • Poor examples
               • Leu --> Val
               • Trp --> Anything
                 • Very specific folding
                 • Provides a fluorophore
             • Risky
               • Pro --> ?
               • ? --> Gly
               • Gly --> ?
             • Must be significantly destabilising to get reliable phi value
             • Should make a large number of mutants in order to get a realistic picture
       • Obtained by normalising either (deltadelta)G[int-N] from unfolding kinetics
         • OR (deltadelta)G[D-int] from refolding kinetics
         • OR (deltadelta)G[D-N] from equilibrium data