RESUMO
Many current tandem mass spectrometry (MS) methods for measuring phosphatidylcholines (PtdChos) rely only on precursor ion scanning of the common 184 m/z phosphocholine fragment with positive electrospray ionization (+ESI), and thus measure pools of PtdChos rather than specific isoforms. In this paper, we developed and compared an isotope dilution, tandem MS method capable of quantifying PtdChos based on specific fatty acid side-chains to the traditional 184 m/z method. The method is based on the detection of PtdCho ammonium formate (AmF) adduct as parent ions and fatty acid fragment daughter ions under negative electrospray ionization (-ESI). Accuracy, imprecision, and recovery were below 15 %, with acceptable linearity (R 2 > 0.99) up to 5 µg/mL. We used the method to analyze the distributions of PtdChos with common side-chain combinations among 60 subjects and showed that it was possible for two individuals to have the same PtdCho pool concentration based on detection of the 184 m/z fragment, but up to a fourfold difference in the levels of specific isoforms comprising the pool based on our method. We then compared the results of both methods across 572 patients with mild cognitive impairment (MCI), Alzheimer's disease (AD), or no impairment (NI), which showed that statistically significant associations between specific PtdCho isoforms and AD were masked with the 184 m/z method. Our findings demonstrate the importance of isoform specificity for quantifying PtdChos, and suggest caution when interpreting analytical data based on pools of biomarkers.