Enzyme-linked immunosorbent assays (ELISAs) have become the method of choice for the detection of residues of allergenic foods that might contaminate other foods. ELISAs have a major advantage over other allergen assay methods - the detection of protein residues - because allergens are proteins. ELISAs can be targeted to detect specific proteins including allergenic proteins or can be more generally targeted to detect a range of proteins from the allergenic source. FARRP believes that the detection of proteins from the source is a suitable approach because the detection of proteins from a particular food very likely indicates that the allergenic proteins are also present. FARRP has developed numerous ELISAs over the years and some of these ELISAs are licensed to Neogen Corporation.
Other methods also exist for the detection of residues of food allergens or allergenic foods. PCR (polymerase chain reaction) methods detect DNA from a specific food. Mass spectrometry methods can detect specific proteins from a specific food. ELISAs are more convenient than these other techniques.
ELISAs are commercially developed into several different formats. Quantitative ELISAs used 96-well microtiter plates and are used to detect the quantitative levels of residues present in a food, food ingredient, or other sample. ELISAs are also available in swab methods. The allergen swabs are qualitative indicators of the presence of allergenic residues above some level, often 5 ppm. Allergen swabs are frequently used to detect contamination of equipment surfaces. Lateral flow strips or dipsticks are another qualitative ELISA format that can be used to detect the presence of allergenic residues above some level. Lateral flow strips are also used for environmental assessment but can also be used, in some situations for a quick, qualitative indication of the presence of allergen in a food or food ingredient.
Other methods also exist for environmental assessment including ATP and total protein. ATP methods are excellent for the general assessment of the adequacy of sanitation but these methods do not specifically detect allergens and may not correlate with allergen-specific methods. Total protein methods will detect all proteins and not just proteins from the allergenic source. Thus, the protein methods also do not correlate well in some cases with allergen-specific methods.
Many commercial ELISAs are now available from a dozen or more ELISA kit suppliers. However, the design of different kits intended to detect the same analyte can be variable. As noted, several formats are available - quantitative, swab, lateral flow. The target for the ELISA can also be variable. For the detection of milk residues for example, commercial kits can target the detection of total milk, casein or β-lactoglobulin (a whey protein). Sometimes, a specific allergenic protein is the target such as Ara h 1 from peanut. The antibodies can be either polyclonal or monoclonal. The nature of the antigen can also vary as noted for milk but also the protein can be partially denatured (heated). The extraction protocols can differ even between kits intended to detect the same analyte. And, the kit calibrator can differ. For example, a casein kit could use casein in the standard curve and have casein for the calibrator or could alternative use non-fat dry milk in the standard curve and express results as ppm of NFDM even though the kit was detecting casein.
With so many variables, it is not surprising that analysis of the same sample by several different kits could yield variable results. Limited efforts have been made to compare kits, assess the effectiveness of specific kits for specific food industry applications, and to evaluate the impact of processing methods on kit performance. Thus, FARRP has developed a research interest aimed at the evaluation and improvement of analytical methods for the detection of allergenic food residues focusing on ELISAs.
AOAC Community Projects
FARRP participates in the AOAC (Association of Official Analytical Chemists International) community on allergen test methods. The community has excellent international representation from the U.S., Canada, EU, Australia, and Japan; many of the commercial ELISA kit companies also participate. An approach for the harmonization of criteria for the validation of food allergen ELISAs was completed in late 2008 and was recently published.
Abbott, M., S. Hayward, W. Ross, S. B. Godefroy, F. Ulberth, A. Van Hengel, J. Roberts, H. Akiyama, B. Popping, J. M. Yeung, P. Wehling, S. L. Taylor, R. E. Poms, and P. Delhaut. 2010. Validation procedures for quantitative food allergen ELISA methods: community guidance and best practices. J. AOAC Int. 93:442-540.
The development of this harmonized approach represents a major advance because commercial ELISA kits will be developed in a more standardized fashion if this approach is followed. Using this approach, FARRP participated in the assessment of a commercial walnut ELISA kit, a collaboration organized by Health Canada. Final results are not published. FARRP has also agreed to participate with an EU group on the validation of a gliadin/gluten test method.
Institute of Food Research (IFR)/Food Standards Agency Collaboration
FARRP has agreed to participate in a collaborative project headed by the Institute of Food Research (Norwich, U.K.) to assess the use of incurred foods developed as part of the EuroPrevall project as reference materials for allergen method validation; this project is funded by the U.K. Food Standards Agency. The AOAC harmonization protocol relies primarily on spike-and-recovery assessment as opposed to naturally incurred food standards. FARRP continues to assert that naturally incurred food standards have value in assessment and validation efforts because processing can affect extraction efficiency and detection. The FARRP group published a manuscript delineating our experience with and the advantages of naturally incurred food standards in ELISA kit assessment/validation.
Taylor, S. L., J. A. Nordlee, L. M. Niemann, and D. M. Lambrecht. 2009. Allergen immunoassays - considerations for use of naturally incurred standards. Anal. Bioanal. Chem. 395:83-92.
Detection of Allergen Residues in Salad Dressing
As part of our development of the mustard ELISA, FARRP graduate student, Poi-Wah Lee determined that mustard residues could not be detected very well when mustard was incorporated into a salad dressing matrix. Further study revealed that mustard proteins were irretrievably lost from the salad dressing matrix due to lack of solubility in the acidic environment. FARRP then pursued a study on the comparative stability of egg, mustard, milk, and gluten residues in acidic salad dressing. The recovery of mustard and egg from this matrix was low while the recoveries of milk and gluten were less seriously affected.
Lee, P. W., L. M. Niemann, D. M. Lambrecht, J. A. Nordlee, and S. L. Taylor. 2009. Detection of mustard, egg, milk, and gluten in salad dressing using enzyme-linked immunosorbent assays (ELISA). J. Food Sci. 74:T46-T50.
Effects of Processing on the Detection of Milk Residues
Melanie Downs, as part of her recently completed M.S. project with FARRP, conducted research on the effects of processing on the detectability of allergenic residues of milk using several commercial kits. Milk was incorporated into pastry squares followed by boiling, deep-fat frying, baking, and retorting. Various commercial milk kits were used to assess their comparative ability to detect milk residues after processing. Milk residues were reasonably well detected from boiled and retorted pastry squares using ELISAs for total milk (Neogen) and casein (ELISA Systems). However, poor recovery was obtained when the milk analyte was β-lactoglobulin (ELISA Systems kit). In contrast, milk residues were not detected well from fried or baked pastry squares by any of the commercial ELISA kits. Further research has led to the hypothesis that heating under low-moisture conditions (frying and baking) causes a decrease in the solubility and extractability of milk proteins. Of course, if the proteins are not extracted from the food matrix, they will not be detected well by any ELISA method. A manuscript on this research is currently in preparation. Some efforts have been made to develop improved extraction procedures.
FARRP Immunoassay Evaluation and Improvement Publications
- Abbott, M., S. Hayward, W. Ross, S. B. Godefroy, F. Ulberth, A. Van Hengel, J. Roberts, H. Akiyama, B. Popping, J. M. Yeung, P. Wehling, S. L. Taylor, R. E. Poms, and P. Delhaut. 2010. Validation procedures for quantitative food allergen ELISA methods: community guidance and best practices. J. AOAC Int. 93:442-540.
- Lee, P. W., L. M. Niemann, D. M. Lambrecht, J. A. Nordlee, and S. L. Taylor. 2009. Detection of mustard, egg, milk, and gluten in salad dressing using enzyme-linked immunosorbent assays (ELISA). J. Food Sci. 74:T46-T50.
- Taylor, S. L., J. A. Nordlee, L. M. Niemann, and D. M. Lambrecht. 2009. Allergen immunoassays - considerations for use of naturally incurred standards. Anal. Bioanal. Chem. 395:83-92.