Metabolite Profiling of Human Amniotic Fluid by Hyphenated Nuclear Magnetic Resonance Spectroscopy Gonc ¸ alo Grac ¸ a, † Iola F. Duarte, † Brian J. Goodfellow, † Isabel M. Carreira, ‡ Ana Bela Couceiro, § Maria do Rosa ´ rio Domingues, | Manfred Spraul, ⊥ Li-Hong Tseng, ⊥ and Ana M. Gil* ,† CICECO-Department of Chemistry, and Mass Spectrometry Centre, Department of Chemistry, Campus Universita ´rio de Santiago, University of Aveiro, 3810-193 Aveiro, Portugal, Bruker Biospin GmbH, Silberstreifen, D76287 Rheinstetten, Germany, Cytogenetics Laboratory and Center of Neurosciences and Cellular Biology, Faculty of Medicine, University of Coimbra, 3001-401 Coimbra, Portugal, and Maternidade Bissaya Barreto, Centro Hospitalar de Coimbra, Coimbra, Portugal The metabolic profiling of human amniotic fluid (HAF) is of potential interest for the diagnosis of disorders in the mother or the fetus. In order to build a comprehensive metabolite database for HAF, hyphenated NMR has been used, for the first time, for systematic HAF profiling. Experiments were carried out using reverse-phase (RP) and ion-exchange liquid chromatography (LC), in order to detect less and more polar compounds, respectively. RP-LC conditions achieved good separation of amino acids, some sugars, and xanthines. Subsequent NMR and MS analysis enabled the rapid identification of 30 com- pounds, including 3-methyl-2-oxovalerate and 4-amino- hippurate identified in HAF for the first time, to our knowledge. Under ion-exchange LC conditions, a different set of 30 compounds was detected, including sugars, organic acids, several derivatives of organic acids, and amino acids. In this experiment, five compounds were identified for the first time in HAF: D-xylitol, amino acid derivatives (N-acetylalanine, N-acetylglycine, 2-oxoleu- cine), and isovalerate. The nonendogenous nature of some metabolites (caffeine, paraxanthine, D-xylitol, sorbitol) is discussed. Hyphenated NMR has allowed the rapid detec- tion of ∼60 metabolites in HAF, some of which are not detectable by standard NMR due to low abundance (μM) and signal overlap thus enabling an extended metabolite database to be built for HAF. Proton NMR spectroscopy has become a routine tool for fast, comprehensive characterization of complex mixtures, 1 including biofluids, 2–5 liquid foods, 6–8 and plant extracts. 9–13 Such detailed characterization is particularly useful in metabonomic studies that aim to understand the effects of perturbations on living systems through the measurement of changes in their metabolome. 1 H NMR spectroscopy enables the identification of many tens or hundreds of compounds simultaneously, is noninvasive, and requires little or no sample preparation. The use of autosampling methods for large numbers of samples to enable statistical validation of the data is also common place. However, when complex mixtures are analyzed, preferably without any previous sample simplification like extraction or separation, the issue of spectral complexity is a serious hindrance due to extensive signal overlap. The desired level of compound identification depends on the problem addressed, and often, partial assignment of the spectra for a selection of compounds (or family of compounds) will suffice. Also, in classification studies where reproducible sample differences are detected and validated by a range of increasingly sophisticated chemometrics methods, the full as- signment of these changes may not always be necessary. In the case of biological fluids and tissues, the quest is often to find specific compounds that may be used as markers for some kind of system perturbation, e.g., disease, diet, and medication. Often, the nature of these markers is not known in advance, and hence, further unveiling of spectral overlap may be of crucial importance. The development of hyphenated NMR methods, with the coupling of an HPLC step immediately prior to NMR analysis (LC-NMR) enables compound separation and analysis to be carried out * To whom correspondence should be addressed. Tel.: +351 234 370707. Fax: +351 234 370084. E-mail: agil@ua.pt. † CICECO-Department of Chemistry, University of Aveiro. ‡ University of Coimbra. § Centro Hospitalar de Coimbra. | Mass Spectrometry Centre, Department of Chemistry, University of Aveiro. ⊥ Bruker Biospin GmbH. (1) Fan, T. W. M. Prog. Nucl. Magn. Reson. Spectrosc. 1996, 28, 161–219. (2) Sean, X. P. Biomed. Chromatogr. 2000, 14, 430–441. (3) Corcoran, O.; Spraul, M. Drug Discovery Today 2003, 8, 624–631. (4) Dear, G. J.; Plumb, R. S.; Sweatman, B. C.; Parry, P. S.; Roberts, A. D.; Lindon, J. C.; Nicholson, J. K.; Ismail, I. M. J. Chromatogr., B: Biomed. Sci. Appl. 2000, 748, 295–309. (5) Lindon, J. C.; Nicholson, J. K.; Wilson, I. D. J. Chromatogr., B: Biomed. Sci. Appl. 2000, 748, 233–258. (6) Gil, A. M.; Duarte, I. F.; Godejohann, M.; Braumann, U.; Maraschin, M.; Spraul, M. Anal. Chim. Acta 2003, 488, 35–51. (7) Le Gall, G.; Colquhoun, I. J.; Davis, A. L.; Collins, G. J.; Verhoeyen, M. E. J. Agric. Food Chem. 2003, 51, 2447–2456. (8) Anatoli, P.; Sobolev, A. S. R. L. Magn. Reson. Chem. 2003, 41, 237–245. (9) Jean-Luc, Wolfender.; Emerson F., Queiroz.; Hostettmann, K. Magn. Reson. Chem. 2005, 43, 697–709. (10) Vassiliki, Exarchou.; Manfred, Krucker.; Teris, A.; van, Beek.; Jacques, Vervoort.; Ioannis P., Gerothanassis.; Albert, K. Magn. Reson. Chem. 2005, 43, 681–687. (11) Jean-Luc, Wolfender.; Hostettmann, K. N. K. Phytochem. Anal. 2001, 12, 2–22. (12) Jaroszewski, J. Planta Med. 2005, 691, 700. (13) Jaroszewski, J. Planta Med. 2005, 795, 802. Anal. Chem. 2008, 80, 6085–6092 10.1021/ac800907f CCC: $40.75  2008 American Chemical Society 6085 Analytical Chemistry, Vol. 80, No. 15, August 1, 2008 Published on Web 06/20/2008 Downloaded by PORTUGAL CONSORTIA MASTER on July 8, 2009 Published on June 20, 2008 on http://pubs.acs.org | doi: 10.1021/ac800907f