Protein Folding

FUNdamentals I

10:00 – 11:00

8/22/08 (Friday)

Protein Folding

Slide 1

Today we are going to wrap up proteins by covering PROTEIN FOLDING, which answers the question “How do you go from a one-dimensional primary structure to a 3-D molecule that has all the functions that a protein should have?”
This transition is from a random coil to a unique, 3-D structure.
Going from a random coil (a statistical structure) where all kinds of possibilities are possible transforming to a solid, unique, functional protein.
Right now there are no definitive answers to this question, but there are several considerations that we will discuss.

Slide 2

Fifty years ago work was done in this regard by working on a protein called ribonuclease, which has a function of cleaving nucleic acid chains into various parts.
Proteases have the ability to cleave proteins. Ribonuclease has the ability to cleave nucleic acids.
RNAase is a single chain of 124 residues.
Its tertiary structure is stabilized by four disulfide bonds
The structure has to be formed and then the disulfide bonds have to be correctly placed.
The tertiary structure decides where the disulfide bonds will be because this cysteine and this cysteine are far apart (red cysteines on diagram) in the primary structure, so when the tertiary structure is formed the cysteines can combine and form these disulfide bonds.
The following was not in lecture, but may help clarify cystine vs. cysteine
Cysteine – an amino acid containing a thiol group (S-H), which will be ½ of a disulfide bond…In other words, in contains a sulfur atom.
Cystine - composed of 2 cysteine amino acids joined by a disulfide bond.
One of the disulfide bonds is from cysteine, which are close together, but the others are from amino acids that are quite far apart.
You have a built-in way to determine whether the structure has properly folded when you see where the disulfide bonds have to be.
Using the RNAase, they started to denature it. They took the disulfide bonds in the protein and reduced them with betamercaptoethanol, which is an ethanol that has a mercaptan group instead of the alchohol. Sulfhydral groups are always seeking to exchange with each other.

Slide 3

If you swamp a protein in solution with betamercaptoethanol, then you will destroy the disulfide bonds. Also, if you 8 molar Urea, you will cause the protein to dissociate and flop around. Urea is called a chaotropic solvent, that is, it generates chaos. It’s a solvent that has great hydrogen bonding capacities and it causes all hydrogen bonds within the protein to be ruptured so that the protein can hydrogen bond with urea and consequentially the protein is denatured.

Slide 4

You’ve destroyed the tertiary structure. You’ve destroyed the chemical bonds that maintain the tertiary structure. Now can we get the structure back?
We need to look for restoration of activity.
In order to do this you have to allow oxidation, because reduction is what split the disulfide bonds, so you have to reoxidize the sulfhydryl groups, allow them to come together, and at that point you have some general refolding and some reforming of the disulfide bonds, but not all correct folding. That was apparent because amino acid #84 was paired with #95. #84 should be with #26, and #95 should be with #40.
The colors should be the same.
You can refold a little, but not properly
Then they put a trace of mercaptoethanol, which will start to perturb the disulfide bonds that have been formed and destroy them and allow the molecule to refold in the proper way and get the proper disulfide bond formation.
The probability of this happening in a correct way going from totally scrambled to the native ribonuclease is about 1%. You will only have 1% of the activity of a ribonuclease recovered after totally denaturing it and then allowing it to come back together.
(Side note: The past 3 slides were done in an experiment in the 1950’s by a research group that I simply could not understand on the audio. Just know that this was all one big experiment that Dr. Miller sounded really intense about.)

Slide 5

The previous experiment gave rise to these conclusions on this slide.
Native conformation of a protein is the state with the lowest Gibbs free energy, that is, it is a situation that has given up some energy; it is a spontaneous process, as we will see with some protein folding. This is a process that will give off free energy and reach a stable state.
Proteins follow unique paths to attain a native state because, as we will see in the next couple of slides, if all possible conformations of the protein were examined for just a few picoseconds, it might take millions of years for the protein to come back to its native state. It really has to follow a discrete path.
The most exciting conclusion was the primary structure itself possesses sufficient information for the folding of the polypeptide chain. All you have to have is a primary structure and the appropriate conditions and it will refold in its proper fashion.

Slide 6

Levintal posed this paradox: “Take a small protein with 100 amino acids, and each amino acid can assume three positions (by now you would realize that is a relatively conservative estimate of the positions…once you have a polypeptide and you can allow rotation around bonds, there can be thousands of positions an amino acid might have. When the R-group is involved with planes), then the total number of structures that are available 3100, or 5 x 1047”
If you examined each structure for 1 x 10-13 s, then it would take a total search of 5 x 1034 s, and that is 1.6 x 1027 years, which is a longer period than the entire universe.
Obviously proteins fold in a definitive way. Not all possible structures are examined.

Slide 7

The general principles of thermodynamics: the free energy of a reaction will equal the enthalpy of the reaction minus a factor which counts for the temperature and the entropy (disorder of the system)

ΔG = ΔH – TΔS

For a spontaneous process, the free energy must drop, that is, the free energy of this system has to go downward if it is going to be spontaneous.
Imagine I have a jug of gasoline and I put a spark to it, and we have a little explosion and we go to carbon dioxide and water (gasoline when burned eventually produces carbon dioxide and water)…the hydrocarbons in gasoline are oxidized and the result is a spontaneous process. We go from compounds of high energy gasoline to water. Water is a very dull material and does not have the capacity to give a burst of high energy, nor will carbon dioxide. Again, high energy state to a low energy state by giving off energy, which is the definition of a spontaneous process.
This is what a protein does when it folds.

Slide 8

If ΔG is minus (as it is in protein folding), bonds formation and/or an increase in disorder must predominate. In other words, you must have the formation of new bonds or an increase in disorder so that this ΔG is minus.
When a random coil breaks bonds with an aqueous solvent, that takes energy. The enthalpy there is positive, you are breaking bonds.
We have our random coil surrounded by water, and it has an association with the water, and those associations must be broken, enthalpy has to be positive, but it forms new bonds internally when the molecule has gone to its globular arrangement.
New bonds have been formed and the formation of new bonds is a minus for enthalpy.
The enthalpy consideration on protein folding essentially is “a wash.” The enthalpy term is minimal because you’ve had to break bonds, but you formed new ones when you have gone into globular proteins.
That means that the disorder term, the entropy, must be the important factor here (ΔS). A positive ΔS is the deciding factor. It is positive because on the folding of the molecules, you have released large numbers of water molecules, which had to form clathrate structures around portions of this particular protein in the unfolded state. When it folds, you reduce the contact with water and you take away the hydrophobic bonds and side chains which were ordering the water. You have freed water molecules, and created a great deal of disorder in that water simply by folding the protein. That is the factor that predominates on the folding of a globular protein, and we call this an entropy driven process.
Disorder is the favored conformation of the universe, and it is constantly increasing in disorder, and it is the normal course of events. Disorder will eventually overtake all of our ordering efforts (2nd Law of Thermodynamics).
We will get more into thermodynamics next week…
Main Point: The folding of a globular protein is entropy driven.

Slide 9

The data that supports this is that if you take a protein and denature it, the heat capacity of the solution that is in increases. The heat capacity is the amount of energy that you have to put into a substance to raise its temperature by one degree.
When the protein unfolds, it has ordered all of this water around the unfolded protein to a great extent, and you will have pockets of ice as far as ordering is concerned. The heat capacity of water, which is partially ice, rises because you have to use some of the energy to break ice bonds (the hydrogen bonds between the ice molecules) before the temperature actually will rise. The heat capacity of the solution rises.
On refolding the protein, the heat capacity goes down because you don’t have as much ice left in the solution. So it is easier to heat water one degree in which there are no ice cubes than it is to heat water which has a few ice cubes in it.
Also, if you do something to decrease the hydrophilic properties of the water, in other words, if you start to take a folded protein and you put it in water and you start to add alcohol of any kind to that particular solution, you will decrease the water capabilities and send the water toward a more hydrophobic type substance. The water can be made more hydrophobic.
Therefore, the distinction between an aqueous outside and a hydrophobic inside starts to breakdown and the protein will unfold because there is no real division between water area and non-water area. You are changing the composition of your water area. You are causing the water to become more like the internal.
The more hydrocarbon properties of the alcohol, the faster it will occur. You get to a point where you can’t add any more alcohol because alcohol above the four-carbon stage will not dissolve in alcohol.

Slide 10

Switching gears now to fibrous proteins…
There are no real tendencies for this so called “spherical globular collapse” in fibrous proteins.
The fiber molecules, as we saw with myosin and collagen as our primary examples and actin which can form fibers to an extent, don’t have this collapse as you had with globular proteins.
In this case, it is not an absolute situation.
There is some hydrophobic bonding when myosin is formed and maybe a little hydrophobic binding when collagen triple helixes form, but for the most part those molecules are made of amino acids which large alterations of dihedral bond angles are difficult.
What really drives this particular process is that the foldings of the proteins are usually enthalpy driven. By that, I mean it is driven by the fact that new bonds are formed when a collagen triple helix takes place. In other words, all of those hydrogen bonds that need to be formed to stabilize a collagen triple helix are essentially what is driving the formation of that helix.
The hydrogen bonds are set up so that they are pretty strong hydrogen bonds. They are not the hydrogen bonds that were available when the chains were in water because hydrogen bonds involving water can be skewed.
They cannot be exactly direct. In the collagen helix, they are straight on, and that position is maintained, so the formation of those bonds is quite critical for the assumption of that type of tertiary structure.
Main Point: We would generally consider that enthalpy (the formation of new chemical bonds) would essentially be the driving force for the formation of the tertiary structure in fibrous proteins, as opposed to the entropy driven process of the globular proteins.

Slide 11

This is the most critical and interesting aspect of the folding process
Note that the ΔG is negative for protein folding, but it is only -5 to -15 kcal/mole
If you were to ignite a mole of methane gas, you would get a ΔG on the order of -3,000 kcal/mole. In other words, this is a very weak reaction. The value of your – ΔG is a sign of how robust the action is.
If there is a large – ΔG, the reaction is spontaneous (giving off a lot of energy) and is going rapidly, precisely and clearly to an end product. It is a robust reaction.
This however is a very weak reaction. You would not heat your home on the energy given off by protein folding.
Take a protein that has 100 amino acids, where the ΔG is -10; the stabilization of amino acid is only 0.1 kcal/mole. This is less than the energy of random motion! Your proteins are fragile (see previous lectures). Their stability is very minor no matter how they are stabilized.
The fact that it exists at all is due to the cooperative endeavors of all the amino acids, but each amino acid contributes only 0.1 kcal/mole to the stabilization of that material. Again, that is a product of a weak reaction. A weak reaction gives rise to weak products.
The significance of that is that evolution has favored flexibility.
You’ve seen this whereby a protein to be functional, like histadine, cannot be a highly structured, solidified entity. It has to have some movement. The same with collagen molecules, as they are twisting and flopping back and forth.
Enzymes that are looking for substrates that are preparing to approach a substrate will actually change shape in order to make the active site compatible with the shape and complexity of the substrate.
The conclusion is that native proteins are on the borderline of denaturation.
If you take a well structured, collagen molecule and you treat it with an enzyme like collagenase, which cleaves the molecule like a knife ¼ the distance from the C-terminus, and ¾ the distance from the N-terminus. When you knock down the cooperativity offered by the three parallel chains in collagen, the fragments that you recover (1/4 and 3/4 fragments) simply unravel. So the whole thing is held together by the cooperative action of all the hydrogen bonds throughout the molecule. You cut the molecule in two, and you destroy that cooperativity and you now have random coil polypeptide chains. So we don’t have very stable molecules.
The other thing to think about in this regard is that if the final functional structure is so weak in a slow, unforceful manner, the must be a lot of mistakes along the way as the protein starts to fold. It gets almost there, but not completely. As a matter of fact, one of our major wastes that we have as living organisms is that we only successfully complete (fully functional) about 50% of all the proteins that we make.

Slide 12

How do you get to the final stage? There is a definite set of paths, as discussed earlier in the lecture.
This slide is a one of the cytochrome proteins.
It has a molten globule state, that is, a state where it is almost in the final form, but not completely. In many proteins you can capture these intermediate states as they are beginning to refold….the molten globule state is one of those famous intermediate states.
In this intermediate state, the protein is somewhat larger (left photo), than in its final form (right photo).
Size Exclusion Chromatography (SEC) is how you detect this.
Put a set of proteins on a column and those that come through quickly are the really big molecules and those that come through slow are the small molecules because they have an entry into the beads of the molecules. Larger molecules are excluded from the beads in the column.

Slide 13

The molecular chaperones are here to do exactly what most have done in the past. They are substances that protect other substances from getting into trouble or having problems.
Molecular chaperones do this for proteins. Proteins really need care from the very moment of conception. This is a situation that goes back to the fragility and uncertainty of proteins.
These chaperone are:
to protect the nascent proteins from concentrated protein matrix in the cell,
to accelerate slow steps,
and to ensure correct folding.
Chaperone proteins were first identified as “heat-shock proteins,” that is, proteins that would be synthesized in the cell when the cell was perturbed by physical factors such as the presence of poisons in solution or even heat.
It was soon discovered that when they tried to put the gene for human growth factor into a E. coli cell to be stimulated to make human growth factor, but the human growth factor occurred in the E. coli in clumps, which made it difficult to harvest. These are called “inclusion bodies,” that is essentially where the early mammalian proteins were found when one tried to have a bacterial cell make them. The reason for that is that the bacteria do not have all of these chaperone proteins that protect the molecules from clumping together and precipitating within the cell. We have these though.

Slide 14