Study of metabolite composition of eccrine sweat from healthy male and female human subjects by 1 H NMR spectroscopy Mark Harker, a, * Helen Coulson, a Iain Fairweather, a David Taylor, a and Clare A. Daykin b,c a Unilever Research and Development, Quarry Road East, Bebington, Wirral, CH63 3JW, UK b Division of Analytical Sciences, TNO Food and Nutrition Research, Utrechtseweg 48, 360Zeist, 3700-AJ, The Netherlands c Division of Molecular and Cellular Biology, School of Pharmacy, University of Nottingham, University Park, Nottingham, NG7 2RD, UK Received 8 December 2005; accepted 12 April 2006 The aim of the study was to evaluate metabolite variability in human eccrine sweat using a metabonomics based approach. Eccrine sweat is a dilute electrolyte solution whose primary function is to control body temperature via evaporative cooling. Although the composition of sweat is primarily water, previous studies have shown that a diverse array of organic and inorganic compounds are also present. Human eccrine sweat samples from 30 female and 30 male subjects were analysed using high- resolution 1 H nuclear magnetic resonance (NMR) spectroscopy in conjunction with statistical pattern recognition. High-resolution 1 H NMR spectroscopy produced spectra of the sweat samples that readily identified and quantified many different metabolites. The major metabolite classes found to be present were lactate, amino acids and lipids, with lactate being by far the most dominant metabolite found in all samples. Principal Components Analysis, Principal Components-Discriminant Analysis and Partial Least Squares-Discriminant Analysis of the eccrine sweat samples, revealed no significant differences in metabolite composition and concentration between female and male subjects. Also, the variation between subjects did not appear to be correlated with any other clinical information provided by the subjects. Overall, the spectra data set demonstrates the large physiological variability in terms of number of metabolites present and concentrations between subjects i.e. human eccrine sweat samples exhibit a high degree of inter-individual variability. KEY WORDS: gender; metabonomics; principal components analysis. 1. Introduction Systems biology based approaches that combine gene expression, protein expression and metabolic changes to biomedical end-points have become common place in pharmaceutical research and development. However, application of these techniques to better understand human metabolic physiological variation in healthy individuals has yet to be fully exploited. External influ- ences such as xenobiotics, foods and hormones are known to interact with cells and tissues to disturb the ratios, concentrations and fluxes of endogenous bio- chemicals. Under mild stress, one way in which organ- isms attempt to maintain homeostasis and metabolic control is by varying the composition of the inter- and intra-cellular fluids. Consequently, following metabolic perturbation there are characteristic organ-specific and mechanism-specific alterations in biofluid composition. How these changes manifest in the phenotype are potentially dependent on numerous factors related to age, gender and the health status of the individual in question. To investigate these complex biochemical end-points in response to metabolic perturbations, non-selective, but specific ‘‘information-rich’’ analytical approaches are therefore required. High-resolution 1 H NMR spectroscopic analysis together with chemometrics for profiling metabolism has now become common place in the pharmaceutical industry for the characterisation of xenobiotic toxicity and disease diagnosis (Griffin, 2003; Fernie et al., 2004). Various studies have demonstrated the application of metabonomic/metabolomic analysis utilising 1 H NMR spectroscopy in a wide range of biofluids, particularly in blood (for example Nicholson et al., 1983, 1995; Lenz et al., 2003) and urine (for example Yoshikawa et al., 1982; Nicholson et al., 1984; Keun et al., 2002; Daykin et al., 2005). Such an approach to biofluid metabolic profiling and statistical pattern recognition requires no prior knowledge of potential compounds of interest, metabolites being identified based on their correlated variation between treatment groups. In addition, mul- tivariate pattern recognition methods such as principal component analysis (PCA) can identify distinct patterns of metabolites whose variation as a whole is character- istic of a particular physiological condition, as opposed to the identification of a unique biomarker of a specific physiological state (Griffin et al., 2004) Eccrine sweat is a dilute electrolyte solution whose primary function is to control body temperature via * To whom correspondence should be addressed. E-mail: Mark.Harker@Unilever.com Metabolomics, Vol. 2, No. 3, September 2006 (Ó 2006) DOI: 10.1007/s11306-006-0024-4 105 1011-372X/06/0900–0105/0 Ó 2006 Springer ScienceþBusiness Media, Inc.