Both non-targeted and targeted volatile profiling of plant samples, e.g. for phenotyping of plant breeding populations, are often performed via headspace-solid-phase microextraction (SPME) coupled to GC-MS. Although HS-SPME is rapid and convenient, it is also well-known to suffer from matrix effects, and best practices recommend the use of well-matched (preferably isotopically labeled) standards. However, because of the large number of volatile targets in profiling studies, researchers commonly use a single surrogate standard, with the assumption that relative matrix effects are sample independent, i.e. that the matrix effect has no Sample × Compound interaction. We have recently demonstrated that this assumption is incorrect; work on an interspecific grape (Vitis sp.) mapping population showed that relative HS-SPME-GC-MS responses of internal standards ranged from 29% RSD to 343% RSD (relative standard deviation). Alternate strategies which attempt to design ad hoc models to correct for matrix effects are not appropriate for plant mapping populations in which the critical matrix components are not well defined. We propose an alternate strategy in which a cocktail of non-native internal standards is spiked into a plant sample prior to HS-SPME-GC-MS analysis. A post-hoc analysis is then performed for each Analyte × Internal Standard combination across the population. The best internal standard for each analyte is defined as the standard that results in the smallest variance. This general approach can be used both to determine the best internal standards as well as to detect the presence of Sample × Compound interactions in HS-SPME matrix effects.

Laboratory kitchen sink: Determining appropriate internal standards for HS-SPME-GC-MS volatile profiling in plant mapping populations using post hoc evaluations

254th National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.