Development and validation of a UHPLC-HR-Orbitrap-MS method for the simultaneous determination of boar taint in porcine meat and meat products

The surgical castration of pigs has been widely practiced for centuries. The main reason for this practice is the prevention of boar taint, i.e. an off-odour caused by the accumulation of 5α-androst-16-ene-3-one (AEON), skatole (SK) or 3-methylindole and indole (IND) in adipose tissue. Because of animal welfare issues, a ban on the surgical castration of piglets will be imposed, starting January 2018, which could consequently evoke negative consumer reactions due to the occurrence of boar taint. In this study, two extraction and UHPLC-HRMS analysis methods, valuable for evaluation of consumer acceptance towards boar tainted meat, were developed for the quantification of AEON, SK and IND in different meat products. Sample pretreatment consisted of extraction with methanol and a homogenizing step (cooked ham, minced meat, tenderloin, bacon, cutlets, blade loin, uncooked ham) or a melting step (salami sausage and liver paste). Both methods were validated according to CD 2002/657/EC and ISO17025 guidelines. To evaluate the specificity and selectivity, both blank and fortified samples were analyzed. In the blank samples, low concentrations of IND and AEON were found. A possible explanation could be the endogenous presence of the boar taint compounds in pigs. When fortifying the blank samples, a significant increase in peak area intensity of the chromatographic peaks at their specific retention times could be observed, taking into account a signal-to-noise ratio of at least 3, also no other interfering substances could be found. The linearity of both methods was assessed through 14-point based matrix-matched calibration curves. The obtained regression model showed good linearity (R2 ≥ 0.99) and no lack of fit was observed. Trueness was assessed as recovery by fortifying blank meat matrices, at three spike levels. The recoveries calculated for both extraction methods meet the permitted levels (-20 % to +10 %). To evaluate precision of the methods, repeatability and within-laboratory reproducibility were determined. Repeatability was proved satisfactory for cooked ham and good for all other meat products. However, given the wide variety of meat products included in this study, generally acceptable precision was obtained for both methods. To illustrate the applicability of the newly developed methods, cooked ham and salami sausage samples produced with meat from boar taint positive carcasses were analyzed. The boar taint compounds were detected in both matrices with obvious peak formation and signal-to-noise ratios exceeding 10. However, not surprisingly, reduced boar taint levels were found if compared to the respective neck fat samples