I am testing various coating concentration of our recombinant antibody in a direct sandwich ELISA. I am testing coating concentrations of 75, 150, and 300 ng/mL with a standard (His-tagged protein) and human serum (containing native protein) diluted 1:10 in PBS. When the coating concentration is increased from 75 ng/mL to 150 ng/mL, the optical density measurements increase by about 14% for the recombinant protein standard, yet for the serum samples the optical density readings double. Moving from a coating concentration of 150 ng/mL to 300 ng/mL, the optical density measurements increase by about 5% for the recombinant protein standard while they increase by 50% for the serum protein.
No, antigens dissolved in serum generally behave the same in ELISA as antigens dissolved in PBS/BSA. FWIW, most people like to dissolve antigens in their blocker solution, which contains both Tween and a protein.
Congratulations! It appears your assay is sensitive, with a low background.
I suspect the problem you observe is due to recombinant antigen in a non-native conformation. This is often caused by the His tag interfering with folding. This effect is especially pronounced when the native protein exhibits quaternary structure, requiring multimerization or binding to a partner, to achieve the structure recognized by your antibody.
An alternative explanation is the presence of antigen fragments competing with your intact antigen. Did you run out your column eluate on SDS-PAGE?
Antibodies characteristically exhibit more or less sensitivity to antigen structure. And this behavior is sometimes modulated by the plate coating process. If you don't care about distinguishing native structure, you might try switching the capture and probe, or swapping in an Ab that is less picky, if available.
Thanks for including your raw data. They bring up a related issue, which is saturation of your spectrometer. Even if someone has "validated" the instrument for OD readings up to 3, you should suspect any data above 2 to be non-linear and noisy.
Finally, human sera are potentially infectious, so I hope you're under BSL2.
Yes, when utilizing natural sera protein as opposed to recombinant protein in an ELISA, the binding kinetics might vary. This is because the post-translational changes that native serum proteins may undergo, such as glycosylation or phosphorylation, might impact how well they bind to the capture and detection antibodies employed in the ELISA.
Recombinant proteins, on the other hand, are frequently produced in a streamlined environment like yeast or E. coli, which may not have post-translational modifications. As a result, recombinant proteins may differ from their native counterparts in terms of binding affinities and kinetics.
It is significant to note that the ELISA's experimental setup and specific antibodies might affect the binding kinetics, making it crucial to verify the test for every distinct protein and sample type.
Esther Okafor is right. That's a great example of how the real-life protein may antigenically differ from the "standard" analyte that you have (1) expressed recombinantly, and then (2) put onto and off of a nickel column.
- Badges
- Science method