ELISA with serum - (Feb/14/2007 )

I hope someone can help me figure out this puzzle?

so, recently we're working on a couple of new ELISAs. were using both capture and direct (two unrelated POIs; one method for each). both are found in human serum.

the problem is that during the detection step, the results are opposite to what they should be, for the standard curve of both proteins...absorbance increases as concentration of spiked samples goes down.

this is the basic set-up for the capture; obviously the direct is quite similar:
1. coat the plate with anti-POI, O/N 4C, diluted in carbonate buffer, on medisorp plates
2. wash X5, pbs/tween/1% bsa
3. block, 60min, RT, pbs/tween/5% bsa, on shaker
4. add 'normal' serum spiked with purified protein; dilute down the wells in 20% serum/80% pbs 1:2 for standard curve, 60 min, RT, on shaker
5. wash X5
6. add biotinylated anti-POI diluted in wash buffer, 30min, RT, on shaker
7. wash X5
8. add STV-HRP diluted in wash buffer, 30min, RT, on shaker
9. wash X5
10. add decection reagent (phos-citrate buffer, H2O2, ABTS)
11. after color develops - 2-5 min - add stop and read the plate

does anyone see a big flaw? the detection reagent is good, none of the stuff is contaminated, and the sample proteins are from entirely different sources and different types of proteins, so it shouldn't be a bad reagent...the only thing that I can think of, is that the amount of 'normal' serum is slightly higher in each successive well as the 1:2 dilutions happen. there must be something in the serum that is causing background with the detection reagent...? any ideas? I've done zillions of elisas, but never with serum and maybe I'm missing something important?

I would be really grateful if anyone could help!

Try freeze-thawing your normal serum because all human serum contains endogenous peroxidase which interferes with the OPD or other reagent you are using to react with the HRP. What you're seeing is a lot of background. That should help. If you don't want to do that because it may interfere with the fact that you may want to use patient samples down the road and don't want to freeze/thaw them...try diluting out the serum more than 20/80. Or increasing your coat concentration and/or blocking time. Or decreasing your HRP incubation time helps with background too...I have done as little as 15-20 min in order to cut down on background.

or switch from hrp to ap detection system.

thank you

I tried freezing the serum, and I will also try diluting my standard curve in a couple of different things to see if I can get rid of the background