A sensitive and simple ultra-high-performance-liquid chromatography-tandem mass spectrometry based method for the quantification of D-amino acids in body fluids

D-Amino acids are increasingly being recognized as important signaling molecules in mammals, including humans. p-Serine and D-aspartate are believed to act as signaling molecules in the central nervous system. Interestingly, several other p-amino acids also occur in human plasma, but very little is currently known regarding their function and origin. Abnormal levels of D-amino acids have been implicated in the pathogenesis of different diseases, including schizophrenia and amyotrophic lateral sclerosis (ALS), indicating that D-amino acid levels hold potential as diagnostic markers. Research into the biological functions of D-amino acids is hindered, however, by the lack of sufficiently sensitive, high-throughput analytical methods. In particular, the interference of large amounts of L-amino acids in biological samples and the low concentrations of D-amino acids are challenging. In this paper, we compared 7 different chiral derivatization agents for the analysis of D-amino acids and show that the chiral reagent (S)-NIFE offers outstanding performance in terms of sensitivity and enantioselectivity. An UPLC-MS/MS based method for the quantification of D-amino acids human biological fluids was then developed using (S)-NIFE. Baseline separation (R-s > 2.45) was achieved for the isomers of all 19 chiral proteinogenic amino acids. The limit of detection was <1 nM for all amino acids except D-alanine (1.98 nM), D-methionine (1.18 nM) and p-asparagine (5.15 nM). For measurements in human plasma, cerebrospinal fluid and urine, the accuracy ranged between 85% and 107%. The intra-assay and inter-assay were both <16% RSD for these three different matrices. Importantly, the method does not suffer from spontaneous racemization during sample preparation and derivatization. Using the described method, D-amino acid levels in human cerebrospinal fluid, plasma and urine were measured. (C) 2011 Elsevier B.V. All rights reserved.