Philip E Johnson
Date of this Version
Allergenic peanut proteins are highly resistant to digestion and are detectable by immunoassays after gastrointestinal digestion. The application of liquid chromatography-tandem mass spectrometry (LC-MS/MS) methods for in vivo detection of peptides originating from allergenic food proteins has not been thoroughly studied. The aim of this work was to develop an in vivo detection method for peanut proteins in serum using LC-MS/MS. The method(s) were validated by analyzing subject serum collected after peanut consumption.
Three de-complexation strategies were evaluated including (1) MS acquisition settings (i.e. inclusion, exclusion lists), (2) commercial depletion kits, and (3) organic solvent fractionation by discovery LC-MS/MS. Overall, none of these approaches were successful. No improvements occurred to peanut peptide detection using inclusion and exclusion lists. The commercial depletion kits removed abundant serum proteins, however, they simultaneously depleted peanut proteins. The fractionation method was efficient in reducing sample complexity, but demonstrated variable peanut protein fractionation.
Due to unsuitable de-complexing strategies, we evaluated non-depleted serum by targeted MS, including parallel reaction monitoring (PRM), multiple reaction monitoring (MRM), and MRM cubed (MRM3). We identified 10 peanut peptides, representing the major peanut allergens. The limit of detection (LOD) of the sera-peanut model matrix (10:1 (w/w)) was similar for PRM and MRM, with detection at 1.0 ppm peanut protein (4.0 ppm peanut). The MRM3 method did not provide improvements to LOD.
Following development of typical targeted methods, we re-investigated PRM with increased protein loading (600 µg). Peanut peptides were detected in two subject sera (sera 1, 2) at two different time points (60, 120 minutes, respectively). However, robust method development was unsuccessful, requiring further investigations in methodology.
Lastly, the intermolecular arrangements of peanut seed storage proteins were evaluated by offline size-exclusion chromatography (SEC) with discovery LC-MS/MS. Gaussian modeling was used to determine the native MW of proteins, isoforms, and complexes. The combination of Gaussian modeling and discovery LC-MS/MS of SEC fractions was a highly effective separation and identification tool.
Advisor: Philip E. Johnson