Immunofluorescence

A huge portion of the experiments I’m doing involves a technique called immunofluorescence (IF). Therefore, I thought I’d spend some time (several posts, actually) explaining what that is.

IF uses antibodies in order to identify proteins inside the cell that you are studying causing them to fluoresce under particular wavelengths of light. This allows you to see where in the cell a particular protein is found (we would say, where the protein is localized within the cell). In my case, I am looking at proteins that are always found on/in a certain place in the cell (the Endoplasmic Reticulum (ER), actually) and therefore can be used to mark (or “label”) that particular region of the cell.

Antibodies are a type of protein found in the immune systems of higher organisms. An antibody recognizes a specific protein (or part of a protein) and binds tightly to it. In your immune system, this occurs so that your body can identify that protein as foreign (because it’s attached to a virus or bacteria) and therefore target it for destruction. Immunofluorescence takes advantage of the specificity of antibodies for a particular protein in order to identify a protein in the crowded environment of the cell.

How does this all work? Let’s say you are in a helicopter, high above the ground, and need to locate where a particular structure (let’s say the bathroom) is in Millenium Park in Chicago (aka the ER) on the evening of July 4 when everyone and their brother is in the park. You can’t really identify the bathroom from in the helicopter because it is the same gray as the concrete surrounding it. However, you know that Mr. Protein O’Interest (aka my protein) hangs around the bathroom so if you can find Mr. O’Interest, you have found the bathroom. Now Millenium Park is huge and there are millions of people there, milling about (this is true in a cell, too; the volume of the cell is huge with respect to the size of a single protein and there are millions of proteins inside the cell). How do you locate Mr. O’Interest? Well, if you train a large number of A. Body clones (the antibody, of course) to recognize the face of Mr. O’Interest (which is presumably different enough from every other face in the park and therefore will specifically identify him), you can send the clones out into the crowd and when one of them finds Mr. O’Interest, it latches onto him.

This process is the basis of immunofluorescence. I want to find the ER in the cell. To do that, I need to locate a protein that is always at the ER. To find that protein, I dump some antibodies on my sample. The antibodies are specific for my protein and when they find it, they latch onto it (so to speak).

So, how do you get antibodies for your favorite proteins? Well, you expose an animal (often a rabbit, but in my case, the antibody comes from a mouse) to the protein by injecting purified protein into them. The animal’s immune system then creates antibodies to that protein (like how you trained the A. Body clones to recognize Mr. O’Interest) and the antibodies are collected and purified.

Generally, a researcher will make purified protein and then send it off to a company that specializes in making antibodies, who will then send the antibody back to you. This can be a very expensive process, though, and it can go wrong in any number of ways leaving you out a whole lotta money with nothing to show for it. So, what’s a researcher to do?

Fortunately, there are companies who mass produce antibodies for commonly studied proteins. However, these antibodies are usually for mammalian versions of the proteins and almost never recognize yeast versions. So, researchers have come up with a way around this problem.

Remember, I said that antibodies can recognize a particular region of the protein. That region can be really quite small. So, somebody figured out that if you took the antibody recognition region of a specific protein (let’s call it myc) that already had an antibody made for it, then stuck it on the end of your favorite protein.* The antibody would stick to myc which was in turn attached to your protein, effectively giving you an antibody to your protein. Which means that you don’t need a different antibody for different proteins. All you need to do is attach myc to your protein and use the myc antibody. These protein regions are called epitope tags and there are a few commonly used tags. Because so many people use these tags, companies have mass produced antibodies to those tags, thus saving the researcher from having to have a custom antibody made. Very nearly all of the IF I do uses antibodies for epitope tags.

Let’s put this into the context of the Millenium Park example. Let’s assume that you do not just want to find the bathroom in Millenium Park, but also the boathouse in Central Park, the ranger station in Yellowstone Park and the camping grounds in Yosemite Park. Each of these locations has a different Mr. Protein O’Interest hanging around. Now, you can train separate, customized groups of A. Body clones, each recognizing the face of a different Mr. O’Interest (very expensive and time consuming), or you can have each Mr. O’Interest wear the exact same kind of hat (the epitope tag)**–a hat that nobody else in the park would wear–and purchase a single group of hat-recognizing A. Body clones from a company. The company has put the time and energy into training the A. Body clones into recognizing the hat. All you have to do is buy the clones and set them loose on the various parks to locate the hat-wearing Mr. O’Interest and therefore, your structures (bathroom, boathouse, ranger station, campground) of interest.

Now, you are probably wondering how it helps to have an A. Body clone attached to Mr. O’Interest because all we’ve really done is replaced recognizing Mr. O’Interest with recognizing the A. Body clone. For that matter, how is it easier to see Mr. O’Interest in the first place? Isn’t the bathroom bigger than Mr. O’Interest? Wouldn’t a larger object (even if it blends into the surroundings) be easier to see from a helicopter than a much smaller one? This brings us to step 2 of the immunofluorescence process.

To be continued……

—————————-

*This of course raises the question of how one sticks a bit of one protein onto another protein. The answer is that you put the DNA coding for the tag (bit of protein) into the DNA of the protein you want to attach the tag to. Which raises the question, how do you put DNA from one protein into DNA from another protein? Well, it’s complicated. I’ll have to have another series of posts just for that.

**How Mr. Interest gets the hat will be dealt with another time (see note on about tagging a protein).