Normal serum high OD values on Neutrvidin coated ELISA plates! - (Mar/25/2014 )

Hi everyone,

I have been trying to test rhaumatoid arthritis and lupus patient sera alongside normal/healthy patient sera (as controls) for the presence of specific immunoglobulin. I am using NeutrAvidin-coated plates, as from our previous experiments they proved better, and am coating the plate with 8ug/ml biotinilated peptide (in filter sterilized distilled water) then incubate for 2hrs as plate specification sheet advises.

Then I wash the plates with PBS pH7.2 and block with 2% BSA in PBS for 1hr ar 37 degrees C. After 3 washes with PBS/Tween 20 (0.1%), I add patient anti-sera diluted 1:200 in PBS/Tw20 (0.05%)/BSA (2%) for 1hr.

After 3 washes I use two different secondary HRP conjugated antibodies (affinity purified) to compare the reactivity with the antigen-

one is Polyclonal Goat anti-Human IgG F(ab')s/HRP (at 1:1000 dilution) and the other is Polyclonal Rabbit Anti-Human IgA, IgG, IgM Kappa, Lambda/HRP (at 1:8000 dilution).

The concentrations of secondary antibody were optimised according to manufacturer's instructions and they are diluted in PBS/Tw20 (0.1%)/BSA (2%) and incubated for 1hr in the dark. Plates are then washed, TMB is applied for 4mins and reaction stopped with 2N HCl.

My problem is that the normal patient serum keeps giving me high values (as high as diseased sera) even though the reactivity should only be high for RA and lupus patient sera (diseased sera give me high reactivity as expected). Other quality control and blank wells are mostly fine.

Is there anything that you could suggest please? We have tested multiple patient sera before using exactly the same system and have never experienced this kind of problems. I have no clue as to what could have gone wrong this time.. I would be very grateful for any suggestions!

I think non-specific human IgG from the normal serum is sticking to the plate. Try not washing the plate after blocking at all and add the diluted serum directly. You should also try different dilutions of the serum and see if you get the same quantitation. If not, then the signal is probably due to non-specific binding and you need to optimize the blocking step.

Thank you so much for your suggestions, I will try them out tomorrow!

Hi tkf,

I did as advised, I did not wash the plate after blocking and came in straight with the patient sera. I tried three different blocking reagents this time i.e. BSA 2%, Marvel 2% and Casein 2%. BSA again gave me high values for both diseased (RA) patient sera and normals, Marvel and Casein gave me much lower readings than BSA but still diseased and normal sera had pretty much the same values.

Controls were mostly fine although it seems I am still getting background in conjugate-wells only.

I don't know what to do about it. This assay used to work perfectly fine and now I cannot make sense out of it. Could you please advise me on it? I would be very grateful for any help!

Hmm... if you are getting high background in the wells that get only HRP-conjugate, then the blocking is not working. You can go ahead with marvel/casein block and increase the number of washes after the HRP step. Also, you can use a different wash buffer (for ex: use 0.1% NP-40 or Triton X-100 in PBS  or simply increase the tween 20 concentration to 0.2%).

Just in case, I would suggest preparing all reagents fresh from scratch.

Finally, run serial dilutions of your sera and see if normals and diseased sera titrate the same.

Good Luck :)

Hi Margerita,

This sounds pretty frustrating - I've been there for sure.  In the past when your ELISA was working, was this with the exact same peptide?  Reason I ask is some peptides have an ability to bind antibodies despite the antibody epitope not being exactly against the peptide.  To test for this, you could add just to the wells the anti-human HRP detection (which is an antibody) and if you have unconjugated HRP add this to another well.  If the unconjugated HRP does not yield high background this would suggest the antibodies are non-specifically binding to the peptide.  

If you could screen your sample cohort for generic IgG concentrations, I wonder if this would coincide with the signals (ie, a sample with higher IgG concentrations would give higher back ground).

As tkf suggested, prepping fresh reagents may help.  Mycoplasma or other contamination may be at play.