Protocol Online logo
Top : New Forum Archives (2009-): : ELISA and Immunoassay

HIV p24 ELISA - problem with high background - (May/09/2012 )

Hi there

I'm trying to optimise a direct sandwich ELISA to detect HIV p24 gag protein.
We got matched antibodies and a protocol from Aalto Bio Reagents. The coating antibodies are a set of three pAbs to p24 and the secondary is a anti-p24 mAb conjugated to alkaline phosphatase. We are using a chemiluminescent substrate for detection. Our samples are virus generated from transfected recombinant HIV DNA which we inactivated in 1% EMPIGEN detergent and is diluted in 0.1% EMPIGEN for the ELISA. Our standard curve is generated from purified recombinant p24 gag protein (also from Aalto) prepared in 0.1% EMPIGEN. We use high-binding white opaque plates for the assay.
I've managed to get the assay to work in that we are getting reliable standard curves (R2 above 0.98) but the problem is that the background will fluctuate from one experiment to the next (background measured from wells containing Ab but no p24 standard) so that in some experiments the RLU readings are higher than those of values in the standard curve.
I've tried different blocking reagents for both after coating and with the conjugate antibody, but this has not altered the relative background.

Does anyone have any experience with the Aalto kit or have suggestions as to how I can lower the background?
I am going to try using standard ELISA plates rather than the high bind ones.


I don't understand how your blank which should have exactly the same composition as the standards except it doesn't have the analyte in it, is giving you a higher signal than the standard curve.

If the blank diluent is different from the standard diluent, then you might expect this. Do you add some protein in the standard/sample diluent?

Please provide more info relating to blockers, coating concentrations, curve range

-Ben Lomond-

Hi Ben
That's what we don't understand as well. The background wells contains 0.1% EMPIGEN, which is used to dilute the analyte and p24 for the std curve and no additional protein.

Both sets of Ab were raised in sheep. I coat the plates with 3ug/ml Ab (overnight) and use 5% BSA as a blocking reagent (1hr). I wash the plates with TBS. I mix the conjugate in 20% sheep serum, 0.05% Tween-20 and then wash the plates with TBS, 0.05% Tween-20.
I've tried other blocking agents at both steps - including 5% skim milk, 0.01% casien, 2% and 5% BSA and 20% sheep serum.
The detection reagent requires washing the plates in the reagent buffer (200mM Tris, 10mM MgCl2, pH 9,8) twice before adding the reagent.
Our dynamic range for the curve should be between 8pg/ml to 32ng/ml but we often only manage to get between 1-32ng/ml.


I am not sure why you are getting this issue, but if it helps, this is our homemade HIV p24 ELISA that has worked for years (and many other labs use):

The Hybridoma capture antibody we use is obtained through the NIH AIDS Reagent program. I don't know where your lab is located and whether you have access to a reagent program such as that, but you can at least find out details on the antibody and try to get it elsewhere. In my lab we just produce our own from hybridoma cells, we've optimized the process and it's very easy to produce large volumes of good antibody in a short time. Here is the info on the antibody:

If you also need tips on the antibody production, or even want me to send you a batch of hybridoma cells (depends where your lab is), we can look into that.

This book chapter is also a great resource for ELISA troubleshooting:


@msymeonides. Thanks so much for the help. Our assay seems to be behaving for the moment - we're managing to get std curve values above background.
Your assay looks like a viable alternate for us if we decide to try something else. Would you mind if I contacted you sometime in the future?

Thanks also for the book chapter suggestion - unfortunately our university (University of Cape Town, South Africa) doesn't have access to it.