Detect Melamine With ELISA Assays

After the binding phase, unbound material is removed using a microplate washer, and chromogenic HRP substrate is added. The detected HRP enzyme activity is directly proportional to the amount of bound HRP conjugated melamine and subsequently dependent upon the mela-mine concentration within the unknown sample. The reaction is stopped after a set incubation time, and the amount of formed colored dye is measured. When melamine is not present in the sample, high levels of enzyme activity occur, indicated by the amount of colored dye formed and subsequent high absorbance. These levels of activity and absorbance decrease as levels of unlabeled, free melamine increase. Melamine concentration can then be directly determined using calibration curves (as discussed below).

Sample Preparation

Three different milk samples—natural full-fat milk, fat-free milk, and artificial milk produced from natural milk powder—were spiked with pure melamine. The spiked samples were prepared by mixing the melamine stock solution (2 mg/mL dissolved in distilled water) with the different milk products. All three milk types were spiked with 20 and 100 µg/L, and the full-fat and fat-free samples were also spiked with 550 and 1,000 µg/L melamine.

When using the first of the two assay kits, the spiked milk samples were prepared as described in the kit instructions:

• Approximately 1 mL of spiked milk sample was added to a clean test tube;
• Samples were centrifuged at 1,500 g at 10 °C for 10 minutes;
• Aliquots of 200 µL were taken from below the fat layer and transferred to a clean test tube; and
• 800 µL of assay diluent was added to the milk serum and the diluted samples were carefully mixed.

When using the second kit, however, sample preparation was adapted slightly, based upon information provided directly from the manufacturer, to increase the assay sensitivity. If the milk samples had been prepared in accordance with the original instructions, lower melamine sensitivity would have been observed in comparison to the first assay, because the samples were—according to original instructions—diluted much more for the assay:

• Approximately 1 mL of spiked milk was added to a clean test tube;
• Samples were centrifuged at 10,000 g at 10 °C for 5 minutes, which produced three defined layers;
• A 250 µL aliquot was taken from the middle layer and moved to a clean test tube; and
• This aliquot was diluted at a ratio of 1:3 by adding 500 µL of 10% MeOH/ 20mM PBS solution.

click for large version

Table 2. Results of the Melamine ELISA Assays With Spiked Milk Samples Obtained Using the Thermo Scientific Multiskan FC

Both ELISA kits were used according to the manufacturer instructions. Either 100 or 150 µL (kit dependent) of melamine standard or spiked sample was added to the antibody-coated well. Fifty µL of HRP-melamine conjugate was then added to each well, and the plate was mixed and incubated for 30 minutes at room temperature. All unbound samples were removed by washing with distilled water. Washing was repeated four times with 300 µL of water. One hundred µL of HRP-substrate was aliquoted into each well, and the plate was incubated at room temperature for a further 20 or 30 minutes (kit dependent). The reaction was stopped via the addition of 100 µL of stop solution, and the absorbance at 450 nm was measured using one of six different microplate photometers. Calibration curves were generated and used to identify the unknown concentrations. (This curve can be calculated based on either measured absorbances or normalized values, where absorbance of the zero control has been set to 100%.)

Calibration Curves and Assay Sensitivity

Melamine calibration curves generated using the first assay kit were measured with all six microplate photometer models. When calibration curves from the second kit were calculated, only two of the microplate photometers were used. Previous data showed that the reader had no influence over the obtained results. As shown in Table 1 (see p. 46), the limit of detection (LOD) for this assay was calculated based on the calibration data, as well as the data from the zero control samples, using the standard IUPAC 3*SD method.