I have been watching the responses to Martin Oaks concerning high "background"
in his ELISA. I suspect that the problem is not background, but cross
reactivity. As described, the assay procedure requires addition of human serum
at one stage, and later the addition of HPR-labelled anti-human. Unless the
human serum has been stripped of Ig's, the immunoglobulins are probably
sticking and are there for the reporter antibody to detect. This is likely to
happen even with blocking. Even detergent will not eliminate this problem if
the percentage of human serum is too high.
If detergent is not included in the wash buffer and serum diluent, its addition
might help the problem. A better solution would be to use the humanized
antibody as the capture antibody and the mouse monoclonal to detect the
analyte, followed by a rabbit or goat anti-mouse - HRP that has been human
serum absorbed. Another alternative is the use of an antibody from a different
species on the top half of the sandwich. It is also possible that the human
serum contains sufficient amount of the analyte so that the human serum could
be used as a coating antigen. The analyte could then be detected with the
mouse monoclonal.
Good luck, Rick Schuman
