A non-invasive method for in vivo skin volatile compounds sampling

• Development of a simple, non-invasive method and device for in vivo skin volatiles sampling.
• Full investigation off actors that influenced the in vivo sampling reproducibility.
• Comparison of headspace sampling and direct contact sampling modes.
• Proposal of effective protocols for membrane storage before and after sampling.
• Application of the proposed method for dietary biomarkers investigation.

The use of volatile organic compounds (VOCs) emanating from human skin presents great potential for skin disease diagnosis. These compounds are emitted at very low concentrations. Thus, the sampling preparation step needs to be implemented before gas chromatography–mass spectrometry (GC–MS) analysis. In this work, a simple, non-invasive headspace sampling method for volatile compounds emanating from human skin is presented, using thin film as the extraction phase format.The proposed method was evaluated in terms of reproducibility, membrane size, extraction mode and storage conditions. First, the in vial sampling showed an intra- and inter-membrane RSD% less than 9.8% and 8.2%, respectively, which demonstrated that this home-made skin volatiles sampling device was highly reproducible with regard to intra-, inter-membrane sampling. The in vivo sampling was influenced not only by the skin metabolic status, but also by environmental conditions. The developed sampling set-up (or “membrane sandwich”) was used to compare two different modes of sampling: headspace and direct sampling. Results demonstrated that headspace sampling had significantly reduced background signal intensity, indicating minimized contamination from the skin surface. In addition, membrane storage conditions both before and after sampling were fully investigated. Membranes stored in dry ice for up to 72 h after collection were tested and showed no or minimal change in volatile profiles.This novel skin volatile compounds sampling approach coupled with gas chromatography–mass spectrometry (GC–MS) can achieve reproducible analysis. This technique was applied to identify the biomarkers of garlic intake and alcohol ingestion. Dimethyl sulphone, allyl methyl sulfide and allyl mercaptan, as metabolites of garlic intake, were detected. In addition, alcohol released from skin was also detected using our “membrane-sandwich” sampling. Using the same approach, we analyzed skin VOCs from upper back, forearm and back thigh regions of the body. Our results show that different body locations share a number of common compounds (27/99). The area with most compounds detected was the upper back skin region, where the density of sebaceous glands is the highest.

Figure optionsDownload as PowerPoint slide