Spiked serum samples used to test ELISAs. - (Apr/30/2007 )
I'm currently trying to optimize an ELISA but I'm having a world of trouble 
My problem is when I come to adding serum samples spiked with my intended antigen there is no signal at all but my standard curve (which is diluted in PTS rather than serum) is really linear with very little background. I tried diluting the serum to 1/10 but still got the same result. Has anbody else had this problem and could someone explain what may be going wrong?
Thanks in advance
Nic.
What's the antigen? Could there already be antibodies to it in serum?
There shouldn't be; we did a tissue cross reactivity with this antibody and didn't get anything. The signal's really low in my samples that should be positive. I thought it might be a steric thing but I'm developing a sandwich ELISA so I'm not sure if that's a still a problem. Would diluting the serum even further make a difference?
I was wondering, does anybody know anything about reducing matrix interference?
Thanks.
Is it human serum? There is endogenous peroxidase in human serum, just like there is in tissue, so that can raise the background if you're using HRP as your system of detection.
Otherwise, just dilute out the serum as much as possible. Or spike more antigen into it, if you can. Maybe the antigen is being degraded by serum proteins after you spike it in?
Otherwise, just dilute out the serum as much as possible. Or spike more antigen into it, if you can. Maybe the antigen is being degraded by serum proteins after you spike it in?
Hmm... I'm using rat serum; I should imagine there's peroxidases in there too. How would you go about performing a peroxidase block in serum? Or would you just stick with diluting the serum a lot? I'm not sure how much peroxidases are contributing because i'm not getting a massive signal in general. As a random aside, would ultrafiltration/centrifugation make a difference?
I'm really hoping my protein isn't being degraded by serum proteins, that's kinda my final conclusion after everything else has been ruled out lol.
Thanks for all your help
I read once that freezing the serum degrades the peroxidase, but I never tested it. I used human serum and just ended up diluting it a lot. But then that kind of defeats the purpose of using serum I guess. I agree that there may be an antigen in your serum that binds to your coat antibody preferentially or there is an antibody, like Swanny said, in your serum that is binding your antigen. Centrifugation makes sense.
Mmmm... seems my recombinant protein had precipitated out of solution... I went back to using a crude lysate and got an awsome signal... I'm going to play around with that for a while and see what the detection limits are with the lysate...
Thanks for teh help.
Yes, you need to be very careful with those little buggers. Even kept at -80 they will degrade over time in solution.
Yea I know :s
How would you store your proteins? I was thinking of putting my next batch in glycerol and bunging them in a -20, but I have no idea if that'll make a difference. My lysate has been in the fridge for months and (touch wood) it still works really well. Unfortunately, the recombinant protein comes in batches that I couldn't possibly use up in one month, so I can't store it in the fridge
I have never tested the effect of glycerol over long term storage. I know some people swear by it though. But if you're using it for plate coating, you would need to dialyze it out first and that could be a pain. I also know that the more concentrated you keep the solution the better. Maybe try adding stabilizers like BSA or something?