Preparation and characterization of domoic acid-protein conjugates using small amount of toxin in a reversed micellar medium: application in a competitive enzyme-linked immunosorbent assay.

With the aim of producing novel antibodies to domoic acid (DA), an original, rapid, and simple procedure for preparing minute amount of hapten-protein conjugates was developed. The amide-bond-generating mixed anhydride method of Erlanger was performed using 0.32-0.64 micromol of DA in a reversed micellar medium allowing strong carrier haptenization as determined by spectrophotometric measurement. Bovine serum albumin (BSA) and ovalbumin (OVA) conjugates were, respectively, used for immunization of BALB/c mice and antibody screening by enzyme-linked immunosorbent assay (ELISA). Specific polyclonal antibodies were produced upon multiple injections of (DA)(17)-BSA conjugate administered by three different routes: (i) intraperitoneal (i.p.), (ii) intraperitoneal + subcutaneous (i.p. + s.c.), (iii) footpad (f.p.). The i.p. route induced antisera of higher titer (1:350000) than did the other protocols (approximately 1:72900) and was selected throughout further experiments. Using a competitive ELISA format with a peroxidase immunoconjugate and a chromogenic substrate, no significant cross-reactivity was observed with glutamic acid, aspartic acid and kainic acid (KA), a structural analogue of DA. The sensitivity of this assay could be enhanced by 1 order of magnitude by using a beta-galactosidase immunoconjugate with a fluorogenic substrate while preserving DA specificity. The calculated dissociation constant (K(D)) for the interaction of the antibodies with free DA was 5 x 10(-)(7) M (chromogenic assay) and 5 x 10(-)(8) M (fluorogenic assay). Using the optimized assay the limit of detection (LOD) and the limit of quantitation (LOQ) in the ELISA buffer were 1.4 and 3 ng/mL, respectively. Moreover this assay was found applicable for measuring DA levels in spiked mussel extracts pre-cleaned through a solid-phase extraction column, as a very good correlation (r(2) = 0.96) was observed between the actual amounts of DA added and amounts detected by ELISA. Thus, accurate determinations of DA in clean extracts could be achieved between 2 and 180 ng/mL in spiked samples which corresponds to 0.02-1.8 microg/g of original mussel tissue. Owing to the regulation limits of 20 microg DA/g of shellfish tissue, these extraction and assay procedures should provide a useful complement to the standard HPLC analytical technique currently employed in monitoring DA in shellfish tissue.