Body odour of monozygotic human twins: a common pattern of odorant carboxylic acids released by a bacterial aminoacylase from axilla secretions contributing to an inherited body odour type

Abstract

It is currently not fully established whether human individuals have a genetically determined, individual-specific body odour. Volatile carboxylic acids are a key class of known human body odorants. They are released from glutamine conjugates secreted in axillary skin by a specific Nalpha-acyl-glutamine-aminoacylase present in skin bacteria. Here, we report a quantitative investigation of these odorant acids in 12 pairs of monozygotic twins. Axilla secretions were sampled twice and treated with the Nalpha-acyl-glutamine-aminoacylase. The released acids were analysed as their methyl esters with comprehensive two-dimensional gas chromatography and time-of-flight mass spectrometry detection. The pattern of the analytes was compared with distance analysis. The distance was lowest between samples of the right and the left axilla taken on the same day from the same individual. It was clearly greater if the same subject was sampled on different days, but this intra-individual distance between samples was only slightly lower than the distance between samples taken from two monozygotic twins. A much greater distance was observed when comparing unrelated individuals. By applying cluster and principal component analyses, a clear clustering of samples taken from one pair of monozygotic twins was also confirmed. In conclusion, the specific pattern of precursors for volatile carboxylic acids is subject to a day-to-day variation, but there is a strong genetic contribution. Therefore, humans have a genetically determined body odour type that is at least partly composed of these odorant acids.

Figure 1

GC×GC–ToF-MS total ion current (TIC) plots for the pooled sample, treated with N-AGA and derivatized with (trimethylsilyl)-diazomethane: (a) three-dimensional plot and (b) colour plot. Labels identify the key analytes and black dots mark peak apexes. The relative abundance of analyte A2 is 100% and the colour gradient goes from 0 to 5% (blue to red).

Figure 2

Individual GC×GC–ToF-MS three-dimensional plots from pairs 4 and 8. Chromatograms for four individuals (a) 4a, (c) 4b, (b) 8a and (d) 8b are shown. GC peak intensities are the sums of the abundances of the masses m/z 117, 103, 114, 96 and 150, which are characteristic for the analyte series A, B, C, E, F and for the analyte Ph, respectively. The structure of these analytes is shown in table 4. The chromatograms are in the region 22–50 min×1.8–4.2 s (¹DײD) and the relative abundance of the analyte E1 is between 36 and 40%.

Figure 3

Distance analysis for the comparisons of any two samples. The boxplots show the distribution of all these comparisons, whether they are made (i) within one subject within the samples from the same sampling day, (ii) within one subject between samples taken on a different sampling day, (iii) between samples originating from two twins and (iv) between samples originating from different pairs, this latter comparison split also into comparisons within or between sexes. (a) The distance as calculated based on equation (2.1) and (b) the distance values transformed to ranks.

Figure 4

Principal component analysis (PCA) based on the correlation matrix of all the samples from male subjects. The numbers indicate the different pairs and the letters a and b indicate the individuals within the pair. Pair 4 is cohabiting. (a) Score plots and (b) loading plots.

Figure 5

Principal component analysis (PCA) based on the correlation matrix of the samples from female subjects. The numbers indicate the different pairs and the letters a and b indicate the individuals within the pair. (a) Score plots and (b) loading plots.

Figure 6

Dendrogram for the average values of the 24 subjects based on the distance matrix calculated with equation (2.1) and with the complete linkage algorithm.