Comparison of calibration strategies for accurate quantitation by isotope dilution mass spectrometry: a case study of ochratoxin A in flour

Abstract

Analysis of low-level organic contaminants in complex matrices is essential for monitoring global food safety. However, balancing sample throughput with complex experimental designs and/or sample clean-up to best reduce matrix effects is a constant challenge. Multiple strategies exist to mitigate these effects, with internal standard-based methods such as isotope dilution mass spectrometry (IDMS) being the most advantageous. Here, multiple internal calibration strategies were investigated for the quantification of ochratoxin A (OTA) in wheat samples by liquid chromatography-mass spectrometry (LC-MS). Internal standard-based quantitation methods such as single (ID1MS), double (ID2MS), and quintuple (ID5MS) isotope dilution mass spectrometry, as well as external standard calibration, were explored and compared. A certified reference material (CRM) of OTA in flour, MYCO-1, was used to evaluate the accuracy of each method. External calibration generated results 18–38% lower than the certified value for MYCO-1, largely due to matrix suppression effects. Concurrently, consistently lower OTA mass fractions were obtained for the wheat samples upon quantitation by external calibration as opposed to ID1MS, ID2MS, and ID5MS. All isotope dilution methods produced results that fell within the expected range for MYCO-1 (3.17–4.93 µg/kg), validating their accuracy. However, an average 6% decrease in the OTA mass fraction was observed from results obtained by ID1MS compared to those by ID2MS and ID5MS. Upon scrutiny, these differences were attributed to an isotopic enrichment bias in the isotopically labelled internal standard [13C6]-OTA that was used for ID1MS, the OTAL-1 CRM. The advantages and limitations of each isotopic method are illustrated.