The endogenous protein may be folded in such a way that the epitope that the 3 Abs you've tried cannot find (or do not have access to). When you say you "over-expressed" the protein and could then detect it, did you denature the over-expressed protein or was it translated in such a way that it was a truncated form of the protein with improper folding thus exposing the epitope?  Are you running Western blots here? If so, apparently your SDS-denaturing gels allow you to detect the protein while non-denaturing gels do not allow you to detect the protein correct? The 3 antibodies you've tried may assume that you are always working under denaturing conditions...?

I agree that the antibody could not detect the endogenous protein due to its folding which hides the epitope that antibody can recognize. 

I do western blot and every time I heated my sample with 6X sample bufffer for 10 minutes for both endogenous and over-expressed protein with GFP tag. 

Do you have any suggestions that I can do do expose the target epitope of endogenous protein? 

Hi Nguyen,

So, you can "see" the over-expressed protein by fluorescent GFP detection and not by antibody detection correct?  But you cannot detect the endogenous form of the protein at all.

If so, then, the antibodies you tried for the endogenous protein may then rely on it being a tertiary epitope in order for the antibody to recognize the protein. In other words, perhaps your antibody will only recognize the epitope when your endogenous protein is not denatured and still folded naturally.  Do the 3 antibodies you tried have literature that states whether or not the native/endogenous (non-denatured) protein is required for detection?  Denaturing the protein in your 6X [I assume SDS-containing] buffer may destroy the tertiary folded structure that creates the epitope that your antibodies were designed to recognize. You might try finding the endogenous protein using a non-SDS system (non-denaturing) native gel format. Or am I not understanding correctly?

It is also possible that the endogenous levels are very low. You may need to load more protein or expose your film longer to see it? I don't know what protein you are looking for, but is it possible the gene has to be induced? For example some genes are only expressed when cells are treated with Interferon.

@Jack: When I do over-expressed and denatured protein, I can detect the band. It means that the antibody can recognize the denature protein. I dont think we should use the native gel format. 

@Jared: I also think that the endogenous levels are very low. I used up to 200 ug protein to run the gel but still not recognize it. Also expose longer time and higher concentration of antibody. 

I also think that my gene could be only expressed when it is activated by a chemical. I intend to treat with PDGF because my target protein is a RhoGEF which is activated by PDGF. Then we will run the gel to see whether we can detect the endogenous protein or not.

Ah - I see.  It will be good to hear what you find then after platelet-derived growth factor treatment. Good luck Nguyen~!

Hi Nguyen,

As Jack said, i think it's worthy to try native conditions. Had this before and it worked with the native gel (might be more delicate to manipulate, but at least would give you an idea if the problem is in the 3D structure of the epitope). 

Good Luck !

@Jack and Roland: I dont really understand why I should test my protein in native gel. In my thinking, SDS-running gel will destroy the tertiary protein structure, so it will expose all the epitope in the surface so that antibody can recognize it but the native gel could not do like that. Am I right or wrong?

What they are referring to is the fact that an antibody made against a whole protein in its native conformation could bind to amino acids that are close in 3D space, but are not next to each other in the linear sequence. I really don't think that you need to try it to find out if that is the problem. If you are purchasing commercially available antibodies, they should tell you what epitope or at least a range of amino acids that contain the epitope. Sometimes they will say made by injecting peptide x into rabbit, which would mean they didn't use the whole protein and your epitope is not dependent on conformation. 

Yes, from the information that I got when I purchased the product. They synthesize a short sequence of amino acid and inject into the rabbit.

Umm. @ Almost everyone here: Quick note, antibody production is usually done with denatured protein, so it is unlikely that the antibody picks up conformational states (Not impossible, but quite unlikely) @Nguyen are your antibodies polyclonal?if so, you could try to purify your antibody using the recombinant protein, and then try again that MAY help...

You could use native protein, but in most cases, you must specify, I believe it depends on what you are going to use it for...for WB denatured proteins are the best, but for stuff like Confocal native is the way to go

Thanks Akash! i am thinking generate antibody for myself but it takes time so I want to have some advises to detect it because the over-expressed protein can be detected by the antibody from the company. I used 3 different antibody from different company.

hmmm, I have come across some cases in which commercially generated polyclonal antibodies work much better after a clean up, there are a few protocols that involve using something akin to affinity capture, it's an interesting method, and one that has worked in the past. purifying your polyclonal antibody this way serves to concentrate a particular sub-set of antibodies that are much more specific to your protein, thus this may serve to better increase the sensitivity of any immunological techniques you may try (WB, IP etc.)...you could google 'antibody clean up' protocols, that may give you a good direction to head in..

Thanks Wassin: One of my friend also suggested me to use MG132. I also think about IP method. i will try MG132 first then IP method.

Hello Wassim, after too much effort to detect endogenous of my target protein by treating with MG132 or IP, I still not detectedthe band. I dont know what is wrong. 

When I over-expressed my target protein with GFP and used anti-GFP Ab to detect the protein but I could not observe the band either. For the GFP only, I could observe the band.

Yes, they can but they used different cell line to me.  In my lab, we dont want to buy that cell line to test. However, when I do immuno flouresecence, I can observe the localization of my target protein.