Selecting selective antibodies

Antibodies are great eh. The ability of the immune system to generate huge amounts of antibodies that are highly selective against a particular invading protein serves several useful functions. Obviously they're great at protecting our bodies from incoming diseases. They're also useful tools for studying proteins in the lab.

If you're new to the world of molecular biology research, such as I am, then it can be difficult to sort through the confusing mass of misinformation. There's so much marketing information out there that what you read may or may not be valid. How does one go about finding the correct antibody to use?

In the past couple of months, I've learnt a few things that I'd like to share to all the other newbies out there.

1. A good place to start is obviously to look in previous publications and see what antibodies other labs have used. Sometimes however, this just adds to the confusion. There is no 'journal of negative results', and no one will write in their results section 'do not use this antibody from this company etc'. In fact, if a publication cites a particular antibody but doesn't show that basic good-quality image, then consider this suspicious...

2. Customer reviews on major companies websites are handy. Although, remember that everyone is human, nothing is perfect, and the reviews won't necessarily all say the same thing.

3. Check what your antibody can be used for! There are many possible applications for antibodies, and an antibody that is good for one application may not be good for another. For example, the extensive processing for Western Blotting may expose more possible epitopes on the target protein that remain inaccessible to the same antibody when used for immunocytochemistry.

4. Check the data sheet, but don't rely on it. Unless you see a specific image or plot in a publication, then you have no reason to believe that antibody will work for your intended application.

5. Finally, and perhaps most importantly, check what part of the target protein your antibody is actually targeting. The choice of target sequence is apparently expert work, and is often protected information. Sometimes the company will give you this information if you ask. If the antibody is raised against the full length protein, then often they themselves don't know exactly which part of the protein the antibody will bind to. This is a situation I found myself in just recently. It was only when I stumbled across a publication from another laboratory, who had painstakingly characterised where the target sequence lay, did I discover that the antibody was not 'specific' as the company had claimed. Rather, the antibody binds to a highly conserved region present in many other proteins in the same protein!

Good to find these things out early on. :)

[Image from Interactive Biology]