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

Trying to design an ELISA - (Jan/22/2012 )

Three days ago i'd barely heard of an ELISA, now i've been asked to come up with a protocol for one. I'm really confused about some of the basics and would appreciate any help.

Previously i have purified a particular GST-fusion protein, collected a series of elution fractions which i then ran on an SDS-PAGE gel and the results matched pretty much perfectly with what i was expecting. Also ran a BCA which indicated high protein concentrations.

But now i have to design an ELISA to demonstrate my fusion proteins "superiority" to the fusion proteins that are currently in use. I don't know what exactly i'm trying to prove! From my reading of it, i presume the ELISA will give me a bunch of OD absorbance values which i can slot into a calibration curve to obtain concentrations? Am i simply to demonstrate that my purification protocol resulted in a higher yield - by relating the concs from the ELISA back to the original weight of the cell pellet i began with?

As for designing the ELISA, i think i have an idea of the overall steps but one thing is really confusing me - which samples do i actually include in it? I know that i have to include a standard control, and positive and negative controls - but what about my elution fractions? I'm a complete beginner when it comes to ELISAs, but it seems to me that i am not trying to indicate any kind of trend (i.e. by running all my elution fractions and coming up with a nice curve) - but trying to obtain results that are as high as possible. For some reason i have something in my head about mixing my 3 or 4 strongest samples, but when i think about it i've no idea what the purpose of this would be. Maybe i'm completely misunderstanding it but i would really appreciate if someone would enlighten me here! Do i pool all my samples together and plate a series of dilutions? Do i pool my 3 or 4 strongest samples and plate a series of dilutions? Do i plate each fraction i obtained at the same dilution?

If the purpose of the ELISA was just to prove my protein was present then it would have been clearer but this trying to prove that my purification was better/leads to a higher protein concentration has confused me completely about what i'm trying to do.

Sorry that this post is long and rambling, it's just the more i read about ELISAs the more i realise i don't understand! My general plan (washing/blocking etc steps aside) is:

1. Coat wells with antigen (diluted samples)
2. Apply primary antibody
3. Apply secondary antibody linked to HRP
4. Add substrate TMB
5. Read results from ELISA plate reader

But i can't really think about the later steps until i know what i'm starting with/understand what results i'm looking for.

One last question, one protocol i was reading mentioned the preparation of a citrate-acetate buffer but this doesn't seem to be mentioned anywhere else. Is it just something that is used in the preparation of the TMB solution, or does it have some other purpose in the ELISA?

Anyway - if you read this, thanks, and i'd be hugely grateful for any help in relation to any of those questions, in particular the choice of samples i am assaying...


Quite a number of questions! I am not familiar with GSTs. However, I believe the test you describe will only be able to tell you how your primary antibody binds to both your preparation and to the "current" one. It will have nothing to do with the function of the protein.

If you are developing a test to quantify the amount of GST in your preparation then you will need to develop a sandwich type assay.


The first thing you need to design an experiment is understand what you are trying to get from it. Sorry to state the obvious, but you really need to know what you are trying to do to be able to decide on samples, controls, protocols.... otherwise you risk designing a completely useless experiment.
Isn't there anyone that you can ask why you are doing what you are doing?

Despite this (and I really cant stress enough how important it is that you understand why you are doing what you are doing) I'll try to help somehow from what I understand from your post. Unfortunately I do have more questions than answers for you, but that's only so I can try and understand (and hopefully help you understand) what you are trying to do so that I can maybe help.

The first question that comes to mind is: What antibodies are you using? both primary and secondary. Is it an anti-GST or an anti-"fused protein" (hope this makes sense).

I'm not entirely sure I understand what you mean by "superiority" of your protein (and I take from your post you are not sure either). If all you want to do is demonstrate a better expression/purification method/process, I'm not sure why you need an ELISA, you could probably just do with measuring concentration, or running SDS-PAGE.

Having said this, one reason I can think of to do ELISA is because the "superiority" you want to show is on some immunological sense: specificity, avidity (?) (which is why knowing the antibodies you are planning to use might help). If this is the case, I do agree with the previous post and I think you definitely will need a sandwich assay.

Another question that comes to mind is, what is the difference between this "new" protocol and the "current" one that you have to compare against. Knowing this might also help us understand and help you.

Finally, what is the final purpose of making this protein. Again, this might help us understand why you need to do ELISA.

Hope this makes sense and I haven't confused you more. Once I know the answer to my questions I'll try and help you.

-almost a doctor-

the specificity or avidity will be of theprimary antibodies used in the test and have nothing to do with the specificity or avidity of the protein. Regardless if the primary antibody is mouse, rabbit, goat etc they may all show binding but have nothing to do with activity of the protein nor if it is even functional.


Completely agree with you PAO_ahac. I just really don't understand what Radiofloyd is trying to do so thought of throwing that option in ;)

Actually, if it is functionality he/she is looking for, I'm not sure that's achievable with standard ELISA assay either.

-almost a doctor-