Do Identical Twins Share the Raw Materials of Body Odor?

How much of each person’s body odor reflects differences in the substances their body produces? A study of underarm secretions from 12 pairs of identical twins found that the composition of odor precursors was more similar within twin pairs than between unrelated people.

Interestingly, the proportions of these substances also changed when samples were collected from the same person on different days. The raw materials of body odor showed both characteristics shared by twins and characteristics that varied from day to day.

Looking for a person’s distinctive odor in its chemical components

Kuhn and Natsch’s study was published online in 2008 and appeared in a journal issue in 2009. Its aim was to connect the idea that human body odor has genetic characteristics with an analysis of specific odor-producing substances.

Volatile substances detected on the skin include not only compounds produced within the body but also those originating in perfumes and cosmetics. Nor is every substance an instrument can detect necessarily important to the human nose. The researchers therefore focused on a group of carboxylic acids known to contribute to underarm odor.

When secreted onto the skin, these acids are bound to glutamine in the form of largely odorless “precursors.” Enzymes in skin bacteria break the bond, releasing the odorous acids. The study used this process to compare the composition of the materials secreted by different people.

Sampling both underarms twice, on different days

The researchers recruited 14 pairs of identical twins. Two pairs were excluded because many of the measured substances were below the detection limit in one member of each pair, leaving 12 pairs, or 24 people, for analysis.

Participants wore cotton pads in both underarms while exercising for 45–60 minutes. Samples were collected on two occasions at least one week apart, and the left and right samples were analyzed separately. The intended total was 96 samples: 24 people × two days × two underarms. One sample was lost, so 95 were actually analyzed.

Diet and sampling time were not standardized, and participants maintained their usual habits. The comparison therefore included day-to-day variation as well as individual characteristics.

Why the same enzyme was added to everyone’s samples

After collection, the samples were treated to limit bacterial changes. A sufficient quantity of the same recombinant enzyme, N-AGA, was then added to the extracts. Standardizing the conditions for releasing acids from their precursors was intended to minimize individual differences in the types and activity of skin bacteria and get closer to the composition of the secreted materials themselves.

In a comparison using pooled samples from several people, odor was weak before the enzyme was added, but a strong, typical underarm odor emerged afterward. The researchers focused on substances that increased markedly after enzyme treatment and selected 24 that could be measured across all samples.

They used two-dimensional gas chromatography, which combines two separation steps, together with mass spectrometry. They also converted the acids into methyl esters to make them easier to measure. An important achievement of the paper was to enable person-by-person comparison of trace components that conventional methods had difficulty detecting or separating.

In other words, what was measured here was the composition of substances released from precursors under standardized enzyme treatment. These values should be distinguished from measurements of the entire body odor naturally entering the air from someone’s skin.

Twins’ chemical profiles were closer than those of unrelated people

Because the amount of sweat collected varied, the researchers used the relative proportion of each substance. After excluding one of the 24 components because it showed substantial irregular variation, they calculated how similar the combinations of the remaining 23 were.

Mean “distances,” with smaller values indicating greater similarity, were 0.09 between the left and right underarms of the same person on the same day, 0.15 for the same person on different days, 0.18 between twins, and 0.33 between unrelated people from different pairs. This index was calculated from component proportions; it does not represent perceived odor intensity or a percentage of similarity.

Differences between twins were only slightly larger than differences between samples collected from the same person on different days, and smaller than differences between unrelated people. The overall pattern of greater similarity between twins remained when male and female combinations were considered separately.

When the researchers averaged each person’s multiple samples, 18 of the 24 people were closer to their own twin than to anyone else. For another four people, their twin was the second-closest person. Several statistical methods likewise showed that many twin pairs clustered close together.

Similarity does not eliminate daily changes or individual differences

Not every twin pair clustered perfectly. Some twins differed from each other, and variation between sampling days was clear even within individuals. Analyses of individual components likewise found that some differed mainly between twin pairs, whereas others showed more prominent differences between individuals or sampling days.

The authors regarded the findings as strong evidence that genetic factors contribute to the composition of body-odor precursors. However, the study had no fraternal-twin control group and was not designed to separate genetic effects rigorously from the effects of twins’ shared environments. The paper’s figure that differences between pairs accounted for an average of 51.5% of variance cannot be reinterpreted as the heritability of body odor.

Body odor also involves sulfur-containing compounds, steroids, and other substances outside the scope of this study. Actual odor reflects bacterial activity and living conditions, such as washing, as well as precursor composition. Associations with immune-related HLA genes were discussed as a topic for future research; the study did not identify the genes responsible.

Summary

In this study of 12 pairs of identical twins, the composition of acids released from underarm secretions using a common enzyme was more similar between twins than between unrelated people. Samples collected from the same person on different days also varied, showing that relatively stable shared characteristics coexist with daily changes in the raw materials of body odor.

A major contribution of the study was to make it possible to investigate a person’s distinctive body odor in terms of the combination of precursors they secrete, using a method that separates and measures trace odor components.

The paper lists both authors as affiliated with Givaudan Schweiz AG. No separate funding or conflict-of-interest statement was found in the main text consulted.

Reference

Kuhn F, Natsch A. 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. Journal of the Royal Society Interface. 2009;6(33):377–392. doi:10.1098/rsif.2008.0223. Open-access full paper.