Gys de Jongh
GysdeJongh at planet.nl
Tue Oct 31 19:06:37 EST 2000
"BJ Allan" <rjallan at interchange.ubc.ca> wrote in message
news:8titve$c4n$1 at nntp.itservices.ubc.ca...
> We have been having a problem with developing an ELISA. Here is the method
> we are using:
> 1) We coat the plate with a goat polyclonal The antibody is in a 50mM
> carbonate/bicarbonate buffer, pH=9.5, and the coating is for 2 hrs @ 37C
> or overnight @ 4C. We are using EIA/RIA plates from CoStar.
> 2) After 4 washes in PBS, we block for 1 1/2 hrs @ 37C with 2% BSA in PBS.
> 3) The blocking buffer is dumped out and we add the antigen diluted in PBS
> + 1% BSA + 0.05% Tween-20. This is incubated for an hour @ 37C.
> 4) Four more washes in PBS and then we add the primary antibody
> (monoclonals from mice) diluted in PBS + 1% BSA + 0.05% Tween-20. Another
> hour long incubation @ 37C.
> 5) Four more PBS washes and then the secondary antibody (GAM-HRP from
> ProMega) in PBS + 1% BSA + 0.05% Tween-20 is added. One hour incubation @
> 37C follows.
> 6) Four final washes in PBS, then we add ProMega's TMB One substrate and
> incubate for up to 30 min. at room temperature.
> 7) The reaction is stopped with 1N HCl and the absorbance read at 450nm.
> The problem is that I was trying to titrate the capture and primary
> antibodies against each other in a chess board pattern, with constant
> antigen and secondary. The primary behaved as expected, the signal dropped
> as the concentration decreased until, with no primary antibody, there was
> no product from the HRP.
> However, the capture antibody behaved exactly backwards, with the response
> increasing and the highest signal with no antibody at all. (The first
> time I wondered if I had the plate upside down.) I have now seen this with
> two different polyclonals, in combination with two different primaries.
> Does anyone have any idea what might be happening here? I'm wondering if
> our blocking isn't working, but that would suggest that the capture Ab is
> actually interfering with antigen binding. Any thoughts or tips would be
i think your system like this :
(?) The signal you will get from this system depends on all the involved
concentrations. It will only be maximal for infinite concentrations of all the
components. In practice this means that you try to drive all the reactions to ,
let's say 95% completion. Except for the X ; that's the one you want to
determine. So what I do is : make 1) , 3) and 4) as large as
possible/affordable. In this situation titrate X down till you reach the point
where either 95% of the signal is left or you reach a concentration of X which
is already lower than you need for your work. In either case you can than
titrate down 4) and next 3) till about 95% of the signal is left or you reach a
antibody dilution which is quite affordable (price) already. You should be
carefull with titrating 1) because this determines the specificity of your
assay. If there is no Goat=>X on the bottom of the elisa plate than may be the
plate will absorb more X , giving a higher signal. However this signal will not
be specific. If you have a mixture of X and another proteine than may be the
plate will only absorb the other proteine and no X at all. If you dissolve X in
buffer than the plate might absorb at lot of X because there is nothing else.
Even more X than the Goat=>X plate. If your Goat=>X is the most expensive one
than I would make 4) and 3) as high as possible. Make 2) at the highest
concentration which is just Ok for your application ; if this elisa could only
detect still higher X than you would change your assay system. In this situation
titrate down 1) till the signal starts to decline.
In general it is not a good idea to use Goat=>MouseHRP for 4) . Goats might have
anti mouse activity as they are kept in the same animal house. (They both have
antihuman keratine antibodies) Try to find swine or horse =>MouseHRP . Try
adding 1-5% normal goat serum in 2) , 3) and 4) if the background , no X , is to
high. Try adding the double concentration tween in the wash : PBS + 0.1% Tween.
Your buffers are the same we use ; the carbonate buffer at pH=9.5 with no other
protein or soap is good for sticking proteins (like Goat=>X) to plastic.
Should work just fine.
More information about the Methods