Stability issue - (Feb/10/2012 )

I am developing an ELISA to detect an IgG therapeutic in several matrices.

Capture reagent is a species specific polyclonal anti IgG heavy and light and detection is via a species specific monoclonal anti Fc.

During stability assessments, recovery of the analyte in the sample is 85% after overnight at 4 degrees, 70% overnight at 20 degrees and 60% overnight at 37 degrees. This is consistent from experiment to experiment.

The stability problem is present in serum, plasma, and a non-complex diluent.

The stabilty problem is not removed by inclusion of 10 mM mercaptoethanol, 1 M NaCl, 0.8 M Tris Base (high pH) or broad spectrum protease inhibitor in the sample. 200 mM acetic acid in the sample stops degradation, but does not reverse it.

What is happening at the molecular level?

Interested to know others' thoughts!

When you are testing your analyte after storage is your control remaining constant from day0(t=0) to t = 16 hrs?

If the analyte is stored in the original tissue culture medium is it stable?

With the Elisa you are not testing stability of the analyte only that it is still recognized by the abs in the system. It could be that the analyte is still functional.

Ab could be undergoing conformational change and not recognized by abs in the system.

For storage you could try diluting in commercial stabilizing agents such as those from Surmodics.

Thanks for your comments.

The test matrix is serum or plasma. The analyte is spiked in, and aliquots frozen. Each aliquot is then subjected to the stability treatment and compared to a control aliquot which is thawed immediately before the assay.

The antibody itself is GMP prepared and undergoes stability testing using a process not related to this assay (HPLC etc)

Yes, I agree, the effect is likely not going to affect biological activity and I suspect that the epitope on the Fc to which the monoclonal detection reagent binds is in some way being modified, just want to know what the modification is so that i can attempt to reverse it. Doesn't appear to be proteolytic or oxidative but is pH dependent. I will try a polyclonal anti-Fc as a detection reagent which might avoid the problem.

I'll let you know how this pans out...

Take either the serum or plasma and vary pH....I might also try not storing in serum/plasma and just stay with original matrix and vary condtions to see what is the most optimum then try to apply that to the test matrices.

The samples will be in serum/plasma and stability has to be demonstrated over their accumulated storage period. I can typically work within accumulated 6 hours at 4 degrees on ice which should be well within acceptance. IgGs are typically rock solid, so my concern is whether the 'assay specific' degradation occurs in vivo prior to sample collection and PK assessments end up being skewed due to the apparent degradation. I can halt the degradation after sample collection using 200 mM acetic acid, but I need to find a way to reverse it or develop the assay that is insensitive to it. Reagents are a bit difficult to come by, because the IgG is not from a common species, but I have sourced a polyclonal anti-Fc that has been absorbed against matrix and capture IgG so this may not be fully dependent on the degrading epitope.

I would love to know the nature of the pH dependent degradation!

I'll keep you posted.

Are you using typical commercial sources of the serum/plasma or individuals? Also, many types of tubes used to produce plasma etc...

You are right IgG is very stable. Another question....I am assuming the therapeutic is also monoclonal...could the IgG after thawing aggregate and thus show lower concentrations? High concentrations of purified Mab are known to precipitate.

I am also going to assume eventually you would be measuring the therapeutic in human samples and are trying to get some baseline info.

We've looked at serum and plasma (K2EDTA) using vacuutainer collection and we've seen the same problem in a casein based phosphate buffered diluent. Actually, this is a polyclonal therapeutic that has gone through a non-affinity route of purification. The stability effect has been demonstrated at low to mid ng/mL concentrations, and doesn't seem to be concentration dependent.

I was convinced it was sulphydril oxidation, but the mercaptoethanol would negate that. Maybe it's pH dependent phosphorylation, but I'm not a chemist so not too clear on this.

The other option is simple sticking to plastic due to charge, which is consistent with the pH dependency, but this is typically a rapid process and would not be time dependent, also, there should be a vast excess of endogenous IgG in these samples which should block the sites.

I am hopeful for the poly's performance.

Thanks for your suggestions!

Final thoughts....if your assay has sufficient sensitivity could you not dilute your samples prior to storage or upon thawing/storage? What comes to mind is making 1:2-1:10 dilution in some type of commercial stabilizing solution. Serasub, products from Surmodics come to mind. Many manufacturers have standards/calibrators or controls which are stable for months at 4C. I think it is just a matter of finding the right matrix.

The use of different detection reagents demonstrated that the 'stability effect' was specific for the eptiope to which the anti-Fc monoclonal bound. Ended up with a monoclonal anti-light chain detector. Sensitivity is down somewhat, but the stability effect is gone in serum. I could have just validated the method within the confines of the stability data, but the risk of not detecting the analyte in biological samples with the functionally active molecule present concerns me. Something to bear in mind for future studies.

Thanks for your input!