An improved high throughput assay for measuring murine antibodies to squalene (SQE) is described. The assay is highly reproducible and sensitive and can detect 80 ng/ml of antibody to SQE. The assay, an ELISA, is similar to our previously described assay in which plates containing PVDF membranes were used [J. Immunol. Methods 245 (2000) 1]. The PVDF plates worked well for detection of murine monoclonal antibodies (mAbs) to SQE, but substantial PVDF plate variation was observed, resulting in significant loss of signal and reproducibility between different lots of plates. In the new assay, the PVDF plates were replaced with Costar round bottom 96-well sterile tissue culture plates. These latter plates, which are not normally used for ELISA assay, gave high absorbances for monoclonal antibodies and anti-SQE serum binding to SQE and low absorbances for solvent-treated wells. Other commercially available polystyrene ELISA plates were unsuitable, in that either the background was high or the absorbance for antibodies binding to SQE was low, or both. This change in plate from PVDF to polystyrene allowed the use of an ELISA plate washer, which dramatically increased the throughput rate over the hand-washed PVDF plates. The improved assay also replaced fetal bovine serum (FBS), which contained SQE in lipoproteins, with fatty acid-free bovine serum albumin (BSA) as the blocker/diluent. Fifteen nanomoles of SQE were selected as the optimal amount of SQE to add to the wells. The binding of monoclonal antibodies and anti-SQE serum was dependent upon both the amount of antibody added to the wells and the amount of SQE added to the wells. Antibody concentration curves were hyperbolic in shape, as seen with most other antibodies. Antibody binding first increased with SQE amount and then reached a plateau around 10 nmol of SQE/well. At high SQE amounts (>75 nmol/well), antibody binding decreased with the amount of SQE added. Using 3H-SQE, the amount of SQE bound to the wells increased linearly, up to 50 nmol of SQE added. Approximately 90% of the added SQE bound to the well. When amounts greater than 100 nmol of SQE were added, the amount of SQE bound to the wells was greatly reduced to approximately 5-10% of the added SQE. The assay was highly reproducible both from lot to lot of plates and from experiment to experiment.