OK, before someone says to do a web search…I already did, and have become frustrated (probably because I’m stupid when it comes to web-searches)
I want to know SPECIFICALLY how does one get the antibody to attach to the microtiter plate. What chemical “handles” are involved to get the antibody to chemically attach. I’d like to know what kind of plastic is used in the wells, how they are treated (if at all) and if the antibodies are altered in some way to get them to hook up in the correct way.
The Web search gave me a bajillion hits on HOW to DO an ELISA or how to make a plate in the biological sense (i.e. just put this solution there and let it sit). I haven’t been able to find a CHEMICAL explanation on how to make one. (i.e. the plastic is chemically modified with an NHS-ester, and the terminal end of the heavy chain of the antibody is modified with an amine, then you add the PBS buffered pH 7.8 antibody solution to the wells and let them sit overnight at 37 degrees).
I’m looking for a source that will give me something like that above.
This is learning in action so please bear with me. I think I now understand that the antibodies bind specifically to ANTIGENS that are already attached to the plastic.
see: http://www.sounddiagnosticsinc.com/illustration.htm#fig1
So I think I can understand the attachment of the antibodies. My question, of course, becomes HOW IN THE HELL DO THEY GET THE “ANTIGENS” TO ATTACH!
When I want to do an ELISA, I use poly-lysine coated plates. However, these plates are purchased rather than chemically modified in any way. As an aside, one can use non coated plates, and the antibodies will stick to the plastic, but you will lose some sensitivity. You’ll be able to find ELISA plates sold from any scientific supply company (try http://www.fishersci.com for one of the largest).
So I’m under the impression that these proteins (either antibodies or antigens) are just physisorbed onto the plastic or poly-lysine. In my world we call this “non-specific binding.” Is this a true way to think of it? Do the proteins just stick in no particular, or controlled fashion?
Nonspecific binding doesn’t necessarily imply that the proteins bind in no particular, controlled, orientation with respect to the surface of the plate. If the plate is coated with poly-lysine, the protein molecules will, over time, orient so that their most acidic face is nearest the lysine. Subsequent tight binding, probably arises from the ever-popular (at least among protein chemists) “subtle conformational changes” that maximize all other attractive forces. The most popular commercial ELISA plates are popular precisely because they have a track record of tightly binding antigens in an orientation that the antibodies can get at them. When a person sets about designing a new ELISA assay, they use an empirical approach to find the best procedure for binding antigen to plate.
Well, the best way to do an ELISA which has specificity and sensitivity is a “sandwich ELISA” where two antigen specific antibodies (recognizing different epitopes on the same protein) are used. One antibody is nonspecifically bound to a coated ELISA plate by simply incubating it on the plate overnight. After washing, the antigens are added, and only the specific protein will be in the plate (because of the antibody). Then, another antibody, which is conjugated (peroxidase and biotin being the most common) is used which binds to another part of the protein. There are then commercially available detection systems for the conjugation of choice.
Now, the easiest way to do an ELISA is to buy a kit which is specific to the protein of interest. In that case, a trained chimp (and not a particularly intelligent one) can do an ELISA.
I haven’t had to run a ELISA myself (I’ve done many other immunological techniques though, but glancing at my copies of “Molecular Biomethods Handbook” and “Short Protocols in Molecular Biology”, attaching the antigen to the plate is passive. You need a plate that is treated to bind your antigen; you can buy them (SPinMB says to use Immulon, which you could probably find on the Web for yourself), or make a poly-L-lysine plate as mentioned above, and then you just incubate to get attachment.
You can find this stuff on the Web or in your organization’s library, just don’t blindly follow whatever you read.
One caveat to keep in mind is that not all antibodies bind antigens the same way when using different conditions, such as the buffers in which you incubate the plate.
If you’re trying to detect an antigen that has a commerically available antibody against it, you got it easy. There are plenty of anti-human, rat, mouse, and such antibodies available. It depends on what antigen you want to detect/measure. Something that’s not common might take some effort.
I wasted months trying to detect a chicken hormone using a rat antibody–not too many people/companies spend their time or money making anti-chicken primary antibodies–in a Western by following vendors’ protocols. The rat antibody (available from Parlow, not a company even) only had very low affinity for the antigen (being different species) after using most transfer protocols (such as transfer buffers with methanol, which may have fixed the protein somewhat after SDS-PAGE). I finally used a low molar phosphate transfer buffer that took hours to run but may have allowed the protein to renature and regain some antigenicity with regard to the detecting primary antibody.
You don’t say what antigen you want to detect, but you may have to play around with your system to get the best results. After you make your plates that is. You may want some controls to show that your antigen attachment protocol works. And you’re not wasting time with a detection protocol that can’t detect antigens that ARE there.
Kits are great for a lot of things, such as secondary detection, as kflanaga metions. But sometimes they’re sold by companies just looking for a buck and they don’t live up to the advertising, and then you might need to think some things out for yourself. Specific protein of intrest are the key words.