I'm trying to study the recognition sites of specific antibodies. For that, I have expressed the recombinant protein and different fragment of it. I have human serum with antibodies againts this protein. I want to know which fragments are recognized by the antibodies. I've done IPs but I have two problems when I do a western blot. I have a lot of non specific bands and furthermore I see recognition using negative controls (healthy patients). I've tried to do a western blot with my fragments and use the human serum as primary antibody, but the results are not good. I don't know if someone has had similar problems with the IPs. Any suggestion would be perfect!
Immunoprecipitation (IP) and western blots are very different techniques. IPs are usually done under native conditions, the expressed protein or protein fragments are folded and in aqueous solution, the antibodies recognise them in a binding buffer (i.e. PBS), and then you remove the unbound antibodies as well as the bound antibodies from the solution by affinity, for instance protein A linked to sepharose or magnetic beads. If everything works out well, you should only see one protein on the gel.
Western blots are usually done after running protein samples on denaturing polyacrylamide gels, which means that native epitopes may be lost. When you have a polyclonal antiserum, it can work in both IP and westerns, but if it is monoclonal, then it may work in IP but not in Westerns if the epitope requires native folding.
If I understand your case correctly, you have expressed your target protein and smaller fragments of it in E.coli, and then you try to use blood samples from different humans containing the antibodies as reagent. First of all, human serum will contain many different antibodies regardless of the patient. If your E.coli proteins are not purified (i.e. crude E.coli extracts containing your overproduced protein as well as endogenous E.coli proteins), then the collection of serum antibodies may bind to random E.coli proteins, and the pattern may be different from individual to individual. So the background is variable. However, you know the molecular weight of your protein of interest, and you can always use an E.coli with an empty vector as control extract in the lane next to your test sample. It should have the same background as the E.coli from which you extract your overproduced protein, but the overproduced protein is missing. So even if you have background bands, it doesn't matter because the one you care about is the one that disappears in the empty vector E.coli control extract.
If you have purified your overproduced proteins so they give a single band on a Coomassie stained gel, then disregard the last paragraph. If you see many bands on the western, it is not background, but it probably means that you have degradation products as well as multimers of your purified protein in small quantities on the gel and the western is so sensitive that you see a whole bunch of proteins. If you give me more information about your actual experiment, I may be able to offer more specific advice. Good luck :)
Dear Ana,
Have you tried linking the antibodies to Pierce™ NHS-Activated Agarose for immunoprecipitation? It works nicely for me. http://www.fishersci.com/ecomm/servlet/fsproductdetail_10652_11964073__-1_0
Also, you may want to preabsorb the antisera with the negative control (healthy patients) to eliminate non-specific interactions for IP or WB, or both. Hope that helps.
Thanks for your advices Jürgen.
I've expressed the protein and their different fragments in HEK293T. After, I've purified them using T7 beads (because I've expressed the fragments with T7 tag in N-ter). I've done the IPs with the protein purified but I see the heavy and light chain of IgG and that masks my results because the protein size is similar to the heavy chain. I don't have the same problem with the fragments because their size is less than light chain. But when I've done the IPs with the different fragments using positive and negative serum I see recognition in both of them. The IPs have been done with agarose protein beads. I thought do it with magnetics beads to see if I lose the non specific binding. I think that a possibility is the amount of antibodies is low so I have problem to discriminate between specific and non specific recognition.
In the case of western blot, I agree with you, but I don't know if the antibodies recognize the epitopes only in native conditions. For that I've tried it to see if work or not. I see recognition but the image is really bad. I see the band in black and the membrane in white just with 2 second of exposition. I've tried improve the results increasing the dilution of antibodies but then I lose the recognition.
Thanks again.
If the HC and LC of the antibodies are masking your proteins try to load them without/with decreased reducing agent. The IGG will go up on the gel, hope will stop disturbing your detection.
OK, so you have T7 purified proteins. I agree with Jan, you can boil in SDS without DTT and your antibody will move up to tetrameric stage. Perhaps you can use anti-T7 tag antibodies from rabbit, and use secondary anti-rabbit for the westerns? Anti-rabbit won't recognise the constant domains of the human antibodies.
- Badges
- Science method