Goals

The lab I am with is the Laboratory of Proteomics and Analytical Technologies. Particularly the proteomics part. If the genome is the sum of all DNA and gene interactions, the proteome is the sum of all protein interactions.

DNA is the code that tells the cell (and further, the body or organism) how to make proteins. Proteins are vastly complicated. There are only 20 amino acids, which act as the base units for making protein. A protein is just a string of amino acids folded into a particular way. Proteins can be strung together in interesting ways to make protein machines. All of the action, anything that is dynamic, in a biological system can be traced to proteins. For example, digestive enzymes are type of protein that breaks apart food stuff so that it can be digested better. Lipase is a digestive enzyme that breaks fats down into triglycerides, which is how the body utilizes fats. Lipase is a protein with a special function. By educated estimates, there are thought to be over 50,000 proteins involved in making humans work properly (and improperly).

There are proteins that stay in the cell nucleus (where DNA is located). Some of these help with DNA copying, or in making other proteins. Many of them bind to DNA at certain locations. That is what we are currently studying.

There are about 10,000 employees of the NIH on this base, so my group (less than 10 people) is studying a very particular piece of the pie.

Proteins are very small. You can't see/identify particular proteins in a microscope. The best method we have currently for looking at proteins is very indirect. It is called mass spectrometry. For this method to work, you need to have a sample of proteins. These are given an electrical charge. They are also cut up. These charged pieces are sent through a constant electrical current from one electrode to another that has a sensor. We are then able to tell the size and mass of these pieces by seeing how long it took them to get through the current. By putting them all together, we are then able to see which proteins were in the sample.

While we are a mass spectrometry lab, we sort of out-source the mass spec to the experts. We design and do the experiments that get the proteins ready to be looked at in the mass spec.

So what are we looking at? We are particularly interested in chromatin and the proteins that bind to it. Chromatin is the fibers that DNA is supported on. When I say DNA, you think double-helix, right? The DNA itself is the rungs on that ladder. The actual structure, helical part is chromatin. So, the long term goal is to develop a way to look at a specific part of chromatin, and the proteins that bind to it. Current research generally looks at the protein complex without DNA, but that is likely an important part of the picture. It seems that some proteins identify where to bind because of the DNA sequence at that location, and that part of its ability to bind is based on DNA as well. This can be illustrated by thinking of a house you want to visit. You find the particular house by its address, but you don't know where to enter until you find the door. By looking at chromatin and the proteins without the DNA, it is as though you visit a house (you've followed very detailed directions to get there) and are inside, but someone studying you from above makes the house invisible to see you standing there in what was the kitchen. If you are holding a frying pan, it might be clear as to where you are. If you are simply sitting down, you might be at the kitchen table, in the living room, or even on the toilet.

Looking at protein-binding on chromatin out of context is like this. We don't know whether that is the location that the protein would normally be if the DNA was still intact or whether the interactions are more or less as complex in vivo.

The intermediate goal is to be able to understand some or all protein interactions through a model system. Also, to learn novel things about proteins that we already think we know about. One issue is what is called "background." If we are studying p53 (a human protein that we are actually studying), there are bound to be tons of other stuff that just happened to be chilling around there at the time that we harvested the sample. How much of it is supposed to be there and how much is non-specific? In the past, background has been washed away, but then you also lose some of your sample. And it's hard to know how this affects the way you would find interactions in vivo. I don't know how we are trying to get around that.

Those are some of the goals of the group I am working with. It's hard to know how to actually get there, but basically you change little things until it gets you closer.

If you want to learn a bit more, the wikipedia post on protemics is pretty helpful as is the proteopedia.