The Sweat Mediator Lipidome is Affected by Stimulation Technique but not Sampling Location

Abstract

Recent interest in sweat as a non‐invasive biological matrix has led to multiple characterizations of the sweat metabolome and proteome using both pharmacological and physiological sweat stimulation from various anatomical sites. However, no direct comparisons exist between different methods of sweat stimulation or site of collection. This study addresses this knowledge gap by comparing the lipid mediator profile of eccrine sweat: 1) collected from the volar forearm following pilocarpine iontophoresis (pharmacological) or moderate exercise (physiological) stimulation; and 2) from the volar forearm, anterior distal thigh and lower back following pharmacological stimulation. Healthy male subjects (n = 7, age = 27.2 ± 1.8yr, BMI = 25.0 ± 3.2 kg/m2) were sampled, and collected sweat was analyzed for over 150 lipid mediators including oxygenated lipids, endocannabinoids and ceramides/sphingoid bases using liquid chromatography‐tandem mass spectrometry using API 4000‐ and 6500‐QTRAPs (Sciex, Framingham, MA). Pilocarpine iontophoresis was conducted with the Macroduct® sweat collection system (Wescor Inc., Logan, UT) and physiological stimulation was achieved by exercise on a cycle ergometer for 15min at 60–80% of heart rate reserve. The study protocol was reviewed and approved by the University of California‐Davis Institutional Review Board. Lipid mediators, including 41 oxygenated lipids, 17 endocannabinoids, 6 ceramides/sphingoid bases and 5 polyunsaturated fatty acids, were detected in the sweat. Sweat lipid mediators were unaffected by sampling site (p > 0.05, two‐way ANOVA with false discovery rate correction), though only four subjects produced sweat from either the lower back or thigh, but all produced sweat from the forearm. In comparison to exercise‐induced sweat, pilocarpine iontophoresis‐induced sweat contained increased concentrations of 39 lipid mediators (p < 0.05, paired Student's t‐test with false discovery rate correction) including those derived from 5‐ and 12/15‐lipoxygenase, cytochrome P450 and soluble epoxide hydrolase, diacylglycerol lipase and ceramide synthase, all of which are major cutaneous enzymatic pathways. Overall results indicate that pharmacologically inducing sweat enriches the secretion in multiple sweat lipid mediators, while anatomical site of sampling does not, though the volar forearm provides the most consistent source of sweat. Therefore, it would appear that care must be taken when comparing sweat metabolomics data obtained by different stimulation techniques, whereas sweat metabolomics data obtained from different anatomical sites may be more readily compared.Support or Funding InformationThe study described was supported by NIGMS‐NIH Grant Number T32‐GM008799 (KA), and USDA Intramural Project 2032‐51530‐022‐00D and NIH Grant Number U24DK097154‐01 (JWN). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIGMS, NIH or USDA. The USDA is an equal opportunity provider and employer.