The Problem with Proteins

Proteins are one of the basic building blocks of life. They are the workhorses that accomplish most of the physiology of every cell, including DNA replication, cell division, energy production, metabolism, and degradation of cellular waste, to name a few examples. There are multiple thousands of different proteins in each cell, and each one is distinct from every other by virtue of its shape.

Wait, what do you mean, a protein’s shape? Well, proteins are much like keys. They have a very specific 3-dimensional structure, which must be maintained for them to do their jobs. A protein shaped for one job will not work for another job, just as a key for one lock will not open a different one. And if a protein gets warped out of shape, it will no longer work properly, much as a bent key will no longer open its lock. We know how keys get their shape – go to any home improvement store and you can see the key-making machine. But how do proteins get their shape?

Would it surprise you to learn that this simple sounding question is actually one of the biggest unsolved questions plaguing biologists today? How proteins achieve their final shape is referred in science circles to the problem of protein folding. And it is a thorny problem indeed.

To understand why this is such a tough question, let’s talk first about how proteins are made. Imagine that you have lots of different beads that you can string together to make a necklace. These beads are connected together end-to-end, and each necklace that you make will look different from each other depending on which beads you choose. Proteins are the same way. The beads are called amino acids; there are 20 standard amino acids in your cells that can be strung together to make thousands of different proteins. The standard amino acids are all different from each other by several biochemical criteria. Some are large, others are small. Some have positive charges, and others have no charge at all. Some are really simple, and others are very complex. Whatever their characteristics, these amino acids are strung together by a “machine” in the cell called the ribosome. Once the amino acids are connected together, though, the process of making the protein is only halfway done .

Now that the protein is in its raw form, it must fold itself into the final 3-dimensional shape. The final shape is enormously complex, usually comprised of different substructures within the overall shape. The final shape directly depends on the linear string of amino acids. But the big question is this: how does the protein know how to do this? How does it know how to take this end-to-end string of different amino acids and twist, turn, fold and bond to itself in such a way as to form the final, complex structure?

Here’s what we know about protein folding:

1. We know that the folding of a protein will depend largely on its amino acids. Let me give you a few examples. Amino acids can be hydrophobic (“water-fearing”) or polar (“water-loving”). Hydrophobic amino acids do not like contact with water, while polar ones prefer a watery environment. Hydrophobic amino acids, therefore, like to be tucked away into the inside of the protein, where there is no water. Polar amino acids like to be on the surface of the protein, where they will encounter water a lot. Another example depends on the size of the amino acids. Some are extremely bulky, and simply cannot fit into tight pockets. They must be folded in a place with lots of room. Thus the structure of a protein is limited by the physical properties of its amino acids.

2. We know that there are several kinds of folds, or substructures, that proteins can adopt in their overall structure. To understand this, imagine the engine of a car. It has an overall structure. But the individual parts of the engine (the transmission, the cooling system, the fuel injection) have their own shapes, as a subset of that larger structure. The same is true with proteins. Proteins can have some part of themselves fold into an alpha helix (kind of like a slinky), while others fold into a beta sheet (like an accordion fan). These smaller bits then fold together in larger structures to form the overall protein shape.

3. We know the structures of many different kinds of proteins. These have been determined experimentally (and this is actually a problem that I will talk about below).

4. We know that proteins can fold extremely quickly (sometimes as fast as a millionth of a second).


That sounds great, right? Yes, but there are also lots of things we don't know about protein folding. Here are some examples:

1. We cannot predict what the final shape will be of a given stretch of amino acids. So far, the only way we can know the structure of a protein is to solve it biophysically. This is done mostly with x-ray crystallography. But this is difficult, expensive, and doesn’t always work. As a result, we do not know the structure of the majority of the innumerable proteins in the world.

2. We do not know what steps most proteins take to form the final shape. It is a multi-step process, and requires the help of lots of other cellular machinery. But most of those steps are a mystery.

3. We don’t know why some proteins fail to fold properly. In fact, sometimes they misfold so badly that the only thing that cell can do is destroy them and start over.

4. We don’t know how to take a predetermined shape and artificially produce a protein to fold into that shape. This is really the converse of problem number 1. We cannot predict a structure based on certain amino acids, nor can we predict amino acids to give a certain structure.

5. We don't know how on earth proteins fold so fast.

The biggest problem facing protein folding experts right now is problem #1. We simply are unable to predict what a protein will look like, even if we know its sequence of amino acids. This is actually a pretty big problem for molecular biologists. We know lots of protein sequences. But we'd be able to conduct much better research if we knew what shape that protein adopted. Again, imagine a protein to be like the engine of a car. If a mechanic needs to fix the engine, simply having a list of the parts that engine contains won't be much help. He needs to know what the engine looks like to have any success.

Okay, I’ve given you the problems we have with protein folding. The good news is that there are lots of scientists working to solve these very problems. And now that I ’ve laid the groundwork for the problem, I’ll come back to this in another post and discuss how some scientists are trying to solve these problems.