ANALYTICAL SCIENCES AUGUST 2010, VOL.

26 917
2010 © The Japan Society for Analytical Chemistry

Notes

Determination of Fatty Acids in Human Sweat during Fasting Using GC/MS

Yoko NUNOME,*† Takao TSUDA,** and Kuniyuki KITAGAWA***

*Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Furo, Chikusa,
Nagoya 464–8603, Japan
**Pico-Device, Co., Ltd., NAGOYA-IKO-RENKEI Incubator, 2-22-8 Chikusa, Chikusa, Nagoya 464–0858,
Japan
***EcoTopia Science Institute, Nagoya University, Furo, Chikusa, Nagoya 464–8603, Japan

Fatty acids (FAs) are biological molecules that are used as major metabolic fuels, and are concerned in important
metabolic processes. We have performed a non-invasive and technically rapid and simple method for collecting sweat
from humans, followed by GC/MS determination. The sweat was collected from each volunteer (the middle finger) by
spraying 70% ethanol aqueous solution (no harmful solvent) into a 1.5-cm3 plastic vial. Analysis of FAs in sweat showed
that the sweat solution contains lauric acid (C12:0), myristic acid (C14:0), palmitic acid (C16:0), oleic acid (C18:1), and
stearic acid (C18:0). Here, it is demonstrated that FA concentrations for 4 young subjects correlated positively with
percent of body fat (r = 0.78) and that the total FA levels for them increased progressively with increasing fasting time
when a subject fasted throughout the experiment.

(Received March 3, 2010; Accepted June 8, 2010; Published August 10, 2010)

Introduction Experimental
Fatty acids (FAs) are essential biological components in Equipment
metabolism that are used as metabolic fuels, whereas elevated GC/MS analysis was performed on a QP-5050A (Shimadzu
FA levels are commonly associated with obesity: a major risk Corp., Kyoto, Japan). The GC separation of FAs was carried
factor for cardiovascular disease, hyperlipemia, and insulin out on a DB-5MS capillary column (30 m in length, 0.25 mm in
resistance.1,2 Concentrations of FAs in blood are elevated due to i.d., 0.25 μm in film thickness; Agilent J&W) coated with
obesity, since triglycerides (TG) accumulated as excess energy (5%-phenyl)-methylpolysiloxane.
in adipose tissue lead to release as the FAs.1 Fast and starvation The injection volume of the sample solution was 2 mm3. The
also cause an increase in liberation of FAs from TG adipose flow rate of helium flow as the carrier gas was regulated at
tissue, while blood glucose drops to a lower level,1,3 leading to 2.1 cm3 min–1. The injector and detector temperatures were 250
the increase of low-density lipoprotein (LDL) cholesterol levels. and 260° C, respectively. The oven temperature was held at
In recent years, non-invasive analysis has attracted a great deal 30° C for 1 min, then increased to 170°C at 30° C min–1 and to
of attention because of major demands for health care. 180° C at 5°C min–1, and finally to 220° C at 3°
C min–1. The
Advantages for sweat analysis of non-invasive methods include analysis time was 21 min.
facilitated sample collection and no risk factor for infection; The MS detection was performed in a selected ion monitoring
moreover, samples can be collected as many times as needed mode (SIM). In addition to the molecular ion M+, the peak at
with much less stress. Various biological substances had already 73 was commonly detected for all the FAs and the I.S. The
been found in human sweat,4 such as some inorganic ions,5–8 mass spectrometer was operated in an electron impact ionization
amino acids,9,10 and lactic acids.8,11,12 FAs, the important mode (EI) at 70 eV.
compounds in sweat, have also been observed with procedures
by collection of sweat12 and by ingestion of linseed oil.13 Chemicals
However, there are no reports of disease-related analysis of All the FAs (lauric acid (C12:0), myristic acid (C14:0),
sweat for FAs. In this study, we describe a rapid and simple palmitic acid (C16:0), oleic acid (C18:1) and stearic acid
method for collection of FAs for sweat analysis. The levels of (C18:0)), and methyl decanoate (C10:0) used as the internal
FAs in sweat have a correlation with the body mass index (BMI, standard (I.S.), were of analytical grade, and were purchased
kilograms per square meter), and change in the amount of FAs from Wako (Wako Pure Chemical Industries, Ltd., Osaka,
in sweat during fasting. Japan).

Preparation of standard solutions

Six stock standard solutions containing the I.S. were
separately prepared by dissolving fatty acid standards in a 70%
†
To whom correspondence should be addressed. ethanol–water solution (final concentrations 100 μM). Except
E-mail: [email protected] for I.S. the mixed standard solutions were prepared by mixing
918 ANALYTICAL SCIENCES AUGUST 2010, VOL. 26

Fig. 1 Typical chromatogram of a sweat sample with the internal standard (I.S.). Peaks: 1,
methyldecanoate (I.S.); 2, lauric acid; 3, myristic acid; 4, palmitic acid; 5, oleic acid; 6, stearic acid.

Table 1 Subject information and data of 5 FAs

Fatty acid/pmol cm–2 min–1 Total fatty acid/

Subject Age Gender Body fat, %
Lauric, C12:0 Myristic, C14:0 Palmitic, C16:0 Oleic, C18:1 Stearic, C18:0 pmol cm min
–2 –1

1 25 Female 19 6.2 11.4 32.2 1.0 1.1 52.0

2 22 Male 14 41.8 11.1 34.5 1.0 1.8 90.2
3 23 Male 18 31.1 22.7 91.6 2.8 6.5 154.7
4 23 Male 23 32.7 99.8 170.7 55.8 45.3 404.3

the 5 standard solutions in a small vial (micro-centrifuge tube of proportional to the logarithm of the corresponding FA
inner volume 1.5 cm3, Sorenson, West Salt Lake). concentration. The correlation coefficient was 0.98
After evaporating to dryness with a centrifugal evaporator at (0.25 – 1.5 mM) for lauric acid, 0.99 (2.0 – 6.0 mM, n = 5) for
85° C (EYELA, CVE-2000, Tokyo Rikakikai Co., Ltd., Tokyo, myristic acid, 0.99 (3.0 – 7.0 mM, n = 5) for palmitic acid, 0.92
Japan), the residue was dissolved by adding 15 mm3 of I.S. (1.5 – 5.5 mM, n = 5) for oleic acid, and 0.89 (2.0 – 6.0 mM,
Then 2 mm3 aliquot of the standard solution was injected into n = 5) for stearic acid, respectively.
the GC/MS.
Chromatographic analysis of sweat samples
Preparation of sweat sample Figure 1 shows the typical total ion chromatogram of the FAs
After washing the middle finger with tap water for 10 s and in a sweat sample. The 6 peaks identified were assigned to I.S.
wiping the finger with a tissue paper (Kimwipes-200, supplied (1), lauric acid (2), myristic acid (3), palmitic acid (4), oleic
by Kulesia, Tokyo) wetted with 70% an ethanol–water solution, acid (5), and stearic acid (6), respectively, based on retention
the finger was rinsed with distilled water for 10 s. After 15 min, times (6.2, 8.0, 10.3, 14.0, 17.9, and 18.5 min). This indicates
sweat on the palmar surface of the middle finger was collected their presence in the human sweat. The retention times are in
in the small vial (1.5 cm3) after spraying the solution by an proportion to the chain length of the FAs. The FAs were
inhaler (EW618, DC3V, Panasonic Electric Works Co., Ltd., detectable at levels of pmol cm–2 min–1 in the sweat sample of
Japan). The vial sample was dried in the centrifugal evaporator all the subjects. The FA contents in the sweat are approximately
at 85° C, and 15 mm3 of I.S. added as described above. Then consistent with those in the superficial skin.14
2 mm3 of the sample solution was injected into a GC/MS.
Informed consent was obtained after the purpose and methods Sweat FA levels
of the investigation had been fully explained. Sweat samples A sweat sample was collected from the subjects (3 males and
were collected from four healthy volunteers who were 3 males 1 female) while fasting. The subject information and the
aged 22 – 23 years, weighing 53 – 75 kg and 1 female aged obtained results are listed in Table 1. This indicate the presence
25 years, weighing 47 kg. These four healthy subjects have not of FAs in all subjects’ sweats. Except for Subject 2, who has
taken food for 12 h before the first sweat sampling for mat the highest concentration of palmitic acid,12 palmitic acid was
metabolism (test). Moreover, no food has been taken during present at 25.3% of FAs on skin surface.14 Subject 4 is obese
experiment. Sweat samplings were successively performed with 23% body fat body and the total secretion rate of FA is
from the palmar surface of the middle finger. correspondingly about 8 times higher than that of Subject 1.
It can be seen in the primary experiment that there was a
positive correlation between the total FA secretion and the
Results and Discussion percent of body fat, with the correlation coefficient r = 0.78,
particularly for palmitic acid (r = 0.81); moreover one can see
Linearity that there was a weak correlation between the total rate of FA
The linearity of the relationship between GC peak area and secretion and the BMI with r = 0.61.
FA concentrations was confirmed by analyzing standard
samples. The calibration curves were obtained by dilution of Time-course of sweat FA levels
the mixed standard samples to different concentrations. The Sweat samples were taken successively every 1 h from
relative peak area ratio for each FA obtained by GC was Subject 1 (eight samples in total) under a fasting condition, and
ANALYTICAL SCIENCES AUGUST 2010, VOL. 26 919

marker, and possibly responsible for TG. We have shown that

FAs are present in human sweat and particularly that palmitic
acid is at the highest level in all the subjects, except one subject.
The FA levels in sweat were correlated well with body mass
index (r = 0.78). Moreover, we found that the FA concentrations
increased gradually during the experimental fasting of a subject.
This rapid and simple method for the sweat collection provides
non-invasive determinations of FAs in sweat and possibility of
application to diagnosis of serum TG level in adipose tissue.

References

1. P. Arner, Best Practice and Research Clinical Endocrinology

and Metabolism, 2005, 19, 471.
2. W. Insull and G. E. Bartsch, Am. J. Clin. Nutr., 1967, 20,
13.
Fig. 2 Time-course of FAs rate levels in sweat. 3. V. P. Dole, J. Clin. Invest., 1956, 35, 150.
4. Y. Kuno, “Human Perspiration”, 1956, Charles C Thomas,
Publisher, Springfield, Illinois, 233.
5. M. G. Bulmer and G. D. Forwell, J. Physiol., 1956, 132,
115.
collected into the vial by spraying 70% ethanol–water solution 6. M. J. Patterson, S. D. R. Galloway, and M. A. Nimmo,
toward middle of finger with the inhaler. Figure 2 shows the Expt. Physiol., 2000, 85, 869.
time-course of FA secretion rate in sweat over 21 h of fasting. 7. J. Terasawa, K. Mitsuya, A. Ishii, and T. Tsuda, Bunseki
The concentrations of total FAs in sweat increased along with Kagaku, 2001, 50, 813.
increasing time of fasting. In contrast, however, only lauric acid 8. T. Ishii, K. Ishii, M. Shibuzawa, and K. Ishii, Eisei Kagaku,
decreased approximately linearly with the fasting time. The 1989, 35, 460.
result suggested that the FAs in sweat were released from TG of 9. N. Liappis, E. Kochbeck, G. Eckhardt, H. Hahne, K.
adipose tissue as an energy source into sweat through blood Kesseler, and P. Bautzer, Arch. Dermatol. Res., 1980, 269,
during longer fasting times. Thus, the FAs concentration in 311.
sweat is a promising indicator of TG levels in blood. 10. T. Hiraoka, H. Okamoto, Y. Gosyo, T. Tsuda, and A. R.
It has been reported that fasting time is related to increased Timerbaev, Anal. Chim. Acta, 2007, 581, 83.
FA  levels in plasma.1,3 Additionally, a good correlation was 11. N. Fellmann, G. Grizard, and J. Coudert, J. Appl. Physiol.,
obtained between FA levels in sweat and in blood.13 To our 1983, 54, 355.
knowledge, however, the present work gives the first report on 12. M. Higo and S. Kamata, Chem. Lett., 1994, 23, 593.
FAs in sweat which are elevated during fasting. 13. A. Katsuragawa, S. Niwa, S. Yamamoto, and T. Tsuda,
Hakkangaku, 2006, 13, 63.
14. N. Nicolaides, Science, 1974, 186, 19.
Conclusions
FAs in sweat are an important physiologic and metabolic