Protocol Online logo
Top : New Forum Archives (2009-): : Immunology

antigen-antibody precipitation curve - precipitin-test, equivalence curve with aHSA & HSA (Apr/01/2009 )

Hi all,

I'm new to immunology but am supporting a class where we are getting the students to run a precipitin test in a 96-well plate and then read the absorbance at 920nm in a microtitre plate reader. The results once plotted is a bell-curve showing that at a given point the anitbody:anitgen ratio will be at it's equivalence point and then no more binding will occur (I know I haven't described that correctly....but you know what I mean!). So far all we are getting is the rise, the equivalence point at about well 5 or 6 and then 7 and 8 are decreasing - but not the full bell-shape.

We have tried varying the conc of our aHSA (Sigma) by resuspending it in 1ml (very conc) 2ml (as recommended by sigma) and 4ml (Very dilute so we max usage since we have large student numbers). The conc of HSA we have tried is as follows: 1.25mg/ml, 1mg/ml 0.75mg/ml, 0.6mg/ml, 0.5mg/ml, 0.2ml/ml, 0.1mg/m, 0.05mg/ml, 0.01mg/ml

This is our current protocol:
Well Number PBS Antigen Antibody
(HSA) (aHSA)
1 (blank) 220 0 0
2 120 100 0
3 115 100 5
4 95 100 25
5 70 100 50
6 45 100 75
7 20 100 100
8 0 100 120

We have tried the differing conc of HSA/aHSA as well as differing volumes - to try and tease out a 9th and 10th well but we just can't fit the volume we need into the wells (seems we need to increase the aHSA to 300ul but we can't fit this in a well, it will only hold ~250ul). We have tried to decrease our conc so that we can try and shift the curve to the left to give it enough wells to go back down again but we are having no luck.

I've attached some results - an example of the best we have gotten is in row E.

Does anyone, somewhere know of a way )a different plate, a different protocol) that may help as I'm running out of ideas (I don't have a large background in immunology).

Any ideas/suggestions would be greatly appreciated!

Attached File


No dose reponse curve will be bell shaped. Should be sigmoidal. You are seeing a plateau and fall off hook-effect.