Development of a LC-QTOF-MS based dilute-and-shoot approach for the botanical discrimination of honeys.

Honeys of particular botanical origins can be associated with premium market prices, a trait which also makes them susceptible to fraud. Currently available authenticity testing methods for botanical classification of honeys are either time-consuming or only target a few "known" types of markers. Simple and effective methods are therefore needed to monitor and guarantee the authenticity of honey. In this study, a 'dilute-and-shoot' approach using liquid chromatography (LC) coupled to quadrupole time-of-flight-mass spectrometry (QTOF-MS) was applied to the non-targeted fingerprinting of honeys of different floral origin (buckwheat, clover and blueberry). This work investigated for the first time the impact of different instrumental conditions such as the column type, the mobile phase composition, the chromatographic gradient, and the MS fragmentor voltage (in-source collision-induced dissociation) on the botanical classification of honeys as well as the data quality. Results indicated that the data sets obtained for the various LC-QTOF-MS conditions tested were all suitable to discriminate the three honeys of different floral origin regardless of the mathematical model applied (random forest, partial least squares-discriminant analysis, soft independent modelling by class analogy and linear discriminant analysis). The present study investigated different LC-QTOF-MS conditions in a "dilute and shoot" method for honey analysis, in order to establish a relatively fast, simple and reliable analytical method to record the chemical fingerprints of honey. This approach is suitable for marker discovery and will be used for the future development of advanced predictive models for honey botanical origin.

Targeted and non-targeted screening of flame retardants in rural and urban honey.

Flame retardants (FRs) are often added to commercial products to achieve flammability resistance, but they are not chemically bonded to the materials, so, they can be easily released into the environment during the production and disposal processes. When honeybees travel to collect nectar during the pollination process, they are prone to be contaminated by chemicals in the air. Therefore, honey contamination has been proposed as an indicator of the pollution status in a particular region. To date, the occurrence of flame retardants in urban honey has yet to be explored. In this study, a direct injection method was used, coupled with LC-QTOF-MS, to analyze honey samples. This method was applied to urban (n = 100) and rural (n = 100) honey samples from the Quebec province (Canada), and the levels of flame retardants in urban and rural honey samples were not significantly different. In the targeted approach, two of the target FRs, tris(2-butoxyethyl) phosphate (TBOEP) and triphenyl phosphate (TPHP), were detected and confirmed at an average trace concentration (<1 ng mL-1). Additionally, a non-targeted screening workflow with an in-house-built library was developed and validated to screen for flame retardants in honey. Tris (2-chloropropyl) phosphate (TCIPP) was identified in honey using the non-targeted screening workflow and confirmed using a pure analytical standard, but there are other compounds detected in the non-targeted analysis that have yet to be validated. This study was the first to report FR compounds based on a direct injection method, coupled with a non-targeted screening workflow, at a trace level in a honey matrix. It also showed that a non-targeted workflow was effective to detect and identify unknown compounds present in the honey sample; hence, this provided a novel angle for the occurrence of FRs in air, with honey as a bio-indicator.