Competitive immunosorbent assays for biotin using bifunctional unilamellar vesicles.

Competitive immunosorbent assays for the model antigen biotin were performed using both unilamellar vesicles with covalently attached biotin and horseradish peroxidase (HBVs) and commercially available biotin-labeled horseradish peroxidase (B-HRP) as the enzyme-labeled antigen. The assays were performed using anti-biotin antibody (ABA) surface densities ranging from one-tenth to full monolayer coverage. It was found that assays using HBVs strongly depended on the antibody surface density, while assays using B-HRP were relatively insensitive to the antibody surface density. The HBV assay dependence on the ABA surface density was most likely due to multiple point attachment of vesicles to the surface. The lowest detectable antigen concentration (least detectable dose) for vesicles (approximately 10(-9) M) was an order of magnitude lower than the value found for B-HRP (approximately 10(-8) M). The sensitivity (slope of the response vs biotin concentration curve) of assays with B-HRP was comparable to the sensitivity of assays with HBVs at low antibody surface density, probably due to less extensive multipoint attachment. It was also found that assays could be performed with vesicles at antibody surface densities that were at least 5 times lower, in terms of the bulk antibody concentrations used to coat the wells, than antibody surface densities at which B-HRP gave comparable signals (approximately 1 delta A/min).