- Email: [email protected]
Blood concentrations of polycyclic musks in healthy young adults
Chemosphere 59 (2005) 487–492 www.elsevier.com/locate/chemosphere
Blood concentrations of polycyclic musks in healthy young adults H.-P. Hutter a,*, P. Wallner b, H. Moshammer a, W. Hartl c, R. Sattelberger c, G. Lorbeer c, M. Kundi a a
Institute of Environmental Health, Center for Public Health, Medical University of Vienna, Kinderspitalgasse 15, A-1095 Vienna, Austria b Medicine and Environmental Protection, Vienna, Austria c Federal Environmental Agency, Vienna, Austria
Received 31 August 2004; received in revised form 14 January 2005; accepted 19 January 2005
Abstract Knowledge on the concentration of polycyclic musk fragrance compounds in human blood is sparse. This study examined the concentrations of six polycyclic musks in blood samples from healthy volunteers. Blood was taken from hundred healthy students of the Medical University of Vienna. The lipophilic fraction was extracted and after purification analyzed by GC–MS. Study participants also completed a questionnaire on the use of cosmetics, about nutrition and other life-style aspects. Two compounds—galaxolide and tonalide—were identified in higher percentages of the blood plasma samples. Maximum plasma levels over 100 ng/l were also only found for galaxolide (4100 ng/l) and tonalide (800 ng/l). Women showed significantly higher levels than men. In a statistical multivariate approach only use of body lotion and age were predictive of positive galaxolide concentrations. For tonalide no significant predictor could be found. The findings mirror the replacement of nitro musk fragrances by polycyclic musks, mainly galaxolide. The high concentrations of galaxolide in human blood raise concern since few toxicological data are available. 2005 Elsevier Ltd. All rights reserved. Keywords: Polycyclic musks; Galaxolide; Tonalide; Celestolide; Phantolide; Cashmeran; Traesolide; Human blood
1. Introduction Synthetic musks are used as fragrances in numerous products, such as perfumes, soaps, body lotions, cosmet-
*
Corresponding author. Tel.: +43 1 4277 64727; fax: +43 1 4277 9647. E-mail address: [email protected] (H.-P. Hutter).
ics, fabric softeners, laundry detergents, air fresheners, food additives and fish bait (Rimkus, 1999; Schmeiser et al., 2001). The first synthetic musks (dinitro- and trinitrobenzene derivates, so called nitro musks) were introduced into the market in the beginning of the last century (Rimkus, 1999). Since the 1950s polycyclic musk fragrances have also been used. Today, the production of polycyclic musk compounds is increasing, with a corresponding decline in nitro musk production (Rimkus, 1999; Peck and Hornbuckle, 2004).
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.01.070
488
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
Both groups of chemicals are lipophilic and persistent in fat tissue. They bioaccumulate in the aquatic environment (Yamagishi et al., 1981; Eschke et al., 1994; Rimkus and Wolf, 1995; Fromme et al., 1999; Rimkus, 1999; Gatermann et al., 2002) and can be found in human fat tissue and breast milk (Liebl and Ehrenstorfer, 1993; Rimkus et al., 1994; Eschke et al., 1995; Mu¨ller et al., 1996; Rimkus and Wolf, 1996; Ka¨fferlein et al., 1998; Liebl et al., 2000). In contrast to nitro musks, data on the toxicological properties of polycyclic musk compounds are sparse (Mersch-Sundermann et al., 1998; Liebl et al., 2000). In recent years, concerns have been raised about estrogenic activity of polycyclic musks (Bitsch et al., 2002; Schreurs et al., 2002). Only few studies on synthetic polycyclic musks in human fat tissue and breast milk have been published (Eschke et al., 1995; Mu¨ller et al., 1996; Rimkus and Wolf, 1996; Liebl et al., 2000). Data on the concentration of polycyclic musk fragrance compounds in human blood are lacking. To our knowledge, only one study has been published so far (Bauer and Fro¨ssl, 1999). The aim of this study was to determine the levels of polycyclic musks in blood plasma of young adults in Austria and to investigate the influence of gender, body fat, eating habits, use of cosmetics, etc. on the concentrations of these compounds.
2. Materials and methods Blood samples were collected with Vacuette plasma tubes (coated with sodium heparin) from 114 healthy students who attended courses at the Hygiene Institute of the Medical University of Vienna. Blood samples were put immediately after collection into a refrigerator at 20 C. From these samples 11 were used for validation purposes of the assays, 100 samples served as basis for the analysis. Height, weight, and subscapular skinfold thickness of the study participants were measured. Furthermore, subjects were asked if they had normal, dry or oily skin, how often they used body care products, how often they ate fish, etc.
To remove potential residual contamination with musk compounds all glassware was dried at 200 C for 12 h. In order to avoid losses by adsorption at the glass walls, the walls were rinsed with iso-octane before use. After the solvent was discarded, the residue solvent was evaporated under a fume hood. Glassware of the rotary evaporator was also rinsed prior to use. Sample preparation was carried out in a laminar flow chemical hood with a charcoal filter. 2.1.2. Sample preparation The samples (8.5–9 ml) were defrosted and the internal standard (50 ll of a 100 pg/ll solution of deuterated tonalide-D3 in acetone) was added. Precipitation of the proteins was performed using acetonitrile (8 ml) and shaking vigorously. 3 ml n-pentane was added and shaking was continued for 2 min. The emerging coagulate was homogenised with a spatula. Extraction was carried out on a shaking board for 15 min followed by centrifugation (5 min: 2500g). The pentane phase was withdrawn to a new test tube. The extraction was repeated with n-pentane (2 ml); extraction time was 15 min (shaking board), centrifugation followed for 5 min at 2500g. This procedure was repeated twice. The combined pentane phases were concentrated and reconstructed to 1 ml hexane. Clean up was carried out with silica gel, conditioned by isooctane, dichloromethane and n-hexane, each 5 ml. Elution was performed with hexane/ethyl acetate (4:1) (8 ml), dichloromethane (8 ml) and ethyl acetate (10 ml). The elute was collected and concentrated (1 ml). Another clean up step included elution on an aluminium oxide column (coated with sodium sulphate: 0.5 g) with hexane as solvent (10 ml). Therefore 2 g aluminium oxide were treated overnight at 450 C and deactivated with 10% water. Prior to GC–MS analysis hexachlorobenzene (HCB) 13C6 as injection standard was added. With each sample batch, a blank sample was analyzed. For blank samples 8 ml of acetonitrile were spiked with 50 ll of a 100 pg/ll solution of tonalide-D3 in Acetone. The rest of the sample preparation followed the method described above. Average (n = 11) blind values range from 30 ng/l (tonalide) up to 50 ng/l (galaxolide). For celestolide, phantolide, cashmeran, traesolide blind values are below LOD (Limit of Detection).
2.1. Analytical procedure 2.1.1. Preparatory operations Special care and several preparatory steps are necessary to avoid or reduce contamination of laboratory equipment with the musk compound analyts. Potential sources of contamination with musk compounds are laundry detergents, towels, cosmetics used by the staff, household cleaning agents applied in the laboratory but also the medical tubes used for taking blood samples. Different kinds of medical tubes were tested to select those with the lowest impact on blind values.
2.1.3. Analysis The GC–MS analysis was carried out by a Thermoquest Trace GC/MS System with negative chemical ionisation using selected ion monitoring (SIM) mode (Thermo Electron, USA). Ammonia was applied as reagent gas. A fused silica capillary column (60 m · 0.25 mm i.d., 1.4 lm film thickness) of crosslinked DB624 (J&W, USA) was used. The operating conditions were as follows: column temperature, programmed from 110 C to 210 C at 15 C/min, from 210 C to 270 C at 6 C/min, and hold at 270 C for 18 min. Injection tem-
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
perature was 260 C, transfer line temperature 260 C, and source temperature 255 C. The corresponding ions used for SIM recording and quantification are listed in Table 1. An example of chromatographic separation is given in Fig. 1. For quantification of polycyclic musk compounds the use of tonalide-D3 as internal standard compensates errors made during injection, changes in recoveries caused by matrix effects and losses during sample pretreatment.
489
2.1.4. Characteristics Recovery rates: To verify that tonalide-D3 is a suitable internal standard for quantification of all polycyclic musk compounds the recovery rates of single musk compounds and tonalide-D3 were determined from blood samples fortified with the analyts using the method described above (number of samples = 11). For calculation of the recovery rates, HCB 13C6 was used as injection standard (Table 1).
Table 1 Ions used for SIM (selected ion monitoring) recording and quantification, recovery rates and standard deviations for single musk compounds (n = 11), LODs and LOQs for single musk compounds (values referred to a sample volume of 9 ml)
Cashmeran Celestolide Phantolide Galaxolide Traesolide Tonalide HCB 13C6 Tonalide-D3
m/z
Recovery rates (%)
Standard deviation (%)
LOD (ng/l)
LOQ (ng/l)
205.3 243.3 243.3 257.3 257.3 257.3 290.0 260.3
97.7 99.4 97.5 95.1 97.2 96.4 – 94.3
2.7 3.7 3.5 5.0 5.9 5.0 – 3.8
12.0 25.0 12.0 62.0 25.0 31.0 – –
24.0 50.0 24.0 124.0 50.0 62.0 – –
LOD: limit of detection; LOQ: limit of quantification; m/z: mass to charge ratio.
27.38
100 95
Tonalide
90 85 80 75
65 60
30
Celestolide
45
25
18.83
22.64
40 35
Phantolide
50
26.44
24.08
20
Traesolide
Galaxolide
55
Cashmeran
Relative Abundance
70
26.64
15 10 5 0 19
20
21
22
23
24
25
26
27
Time (min)
Fig. 1. SIM chromatogram of six polycyclic musk compounds (m/z 205.3 RT: 18.0–21.0 min; m/z 243.3 RT: 21.0–25.0 min; m/z 257.3 RT: 25.0–28.0 min). m/z: mass to charge ratio.
490
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
Limits of detection, limits of quantification: The limits of detection (LOD) given in Table 1 are calculated for the following conditions: • • • •
signal to noise ratio of 3 to 1 (S/N = 3), sample volume 9 ml, average recovery rates (Table 1), blind values (see Section 2.1.2).
The limits of quantification (LOQ) are twice the limits of detection. 2.2. Statistical procedures Differences in musk compounds concentrations between groups were calculated using non-parametric tests (U test, Kruskal–Wallis test). Stepwise logistic regression analysis with the following variables as potential predictors of a positive plasma concentration (>LOQ) of the different musk compounds was applied: Age, gender, body mass index, subscapular skinfold thickness, skin type, fish consumption, intensity of use of body lotions, perfumes, etc. These analyses were conducted only for musks with a rate of positive plasma concentrations of 10% or more. A p-value below 0.05 was considered significant. 3. Results From the 100 students 55 were female and 45 male. The age ranged from 19 to 43 years with a median of
Table 2 Body mass index (BMI), subscapular skinfold thickness (SST), fish consumption, skin type and use of cosmetics (n = 100) Parameter BMI (Md, [Q1–Q3]) SST (Md, [Q1–Q3]) Fish consumption (at least once a week)
22 [20–23] 15 [12–18] 50%
Skin type: Normal Dry Oily
62% 31% 7%
Use of creams (daily): On face On hands On body
25% 40% 46%
Use of (daily/more than once a day): Body lotion Perfumes Deodorants Shampoo Fragrance oil Spray
17% 54% 70% 21% 3% 3%
Md: median.
23 years. Table 2 shows descriptive data concerning body mass index, subscapular skinfold thickness, fish consumption and use of diverse cosmetic products. Subjects were predominantly slim. Fish was regularly consumed by half of the students, however, predominantly sea fish. ‘‘At least daily use’’ of perfumes and deodorants was reported by more than half of the students. Two polycyclic musk compounds, galaxolide and tonalide, were identified in higher percentages of the blood plasma samples. Galaxolide was detected in 91% of the samples (in 83% of the samples >LOQ), tonalide in 17% of the samples (16% >LOQ). Phantolide was found in one subject and traesolide in two, with levels lower than 100 ng/l. Cashmeran and celestolide were below the detection limit in all blood specimens. Median concentrations of galaxolide were 420 ng/l, and of tonalide below LOD. Maximum plasma levels were 4100 ng/l for galaxolide and 800 ng/l for tonalide. For galaxolide women showed significantly higher levels than men and lower values in the higher age group (26– 43 years) were detected (Table 3). Persons who more often used cream on their face, hands, or body showed significantly higher plasma concentrations of galaxolide, but not of tonalide. Skin type (dry, normal, or oily) had no influence on the concentrations of these two compounds. Significant effects on the rate of positive galaxolide plasma concentrations were found for gender, age, use of body lotion, perfumes and for fish consumption, with increasing rates for females, younger age, frequent use of body lotion and perfumes and high frequency of fish consumption. In the multivariate approach only use of body lotion and age were predictive of positive galaxolide concentrations. For tonalide no significant predictor could be detected. All other compounds had rates of positivity below 10% and were not further analysed.
4. Discussion Today, the use of polycyclic musk fragrances compounds is increasing. However, data on the toxicological properties of these chemicals and their levels in human blood are sparse. In this study especially galaxolide was detected in a high percentage (91%) of the blood samples of young adults in Vienna, Austria. Tonalide was found in 17%. The plasma concentrations of galaxolide were comparable to those from a German human biomonitoring study which was performed in 1999 (Bauer and Fro¨ssl, 1999), whereas in the present investigation levels of tonalide were about 10 times lower. In relation to nitro musks (Bauer and Fro¨ssl, 1999; Eisenhardt et al., 2001; Ka¨fferlein and Angerer, 2001) galaxolide concentrations were comparatively high, with a maximum up to more than 4 lg/l.
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
491
Table 3 Median galaxolide plasma concentration for stratification according to gender, age, use of body lotion, use of perfumes and fish consumption Independent variable
n
Plasma concentration (ng/l) Md [Q1–Q3]
p-value
Gender Male Female
45 55
260 [98–540] 580 [290–885]
0.002
Age 19–25 26–43
67 33
450 [250–805] 340 [130–670]
0.045
Use of body lotion Never 1/month 1/week 2–4/week Daily Several times/day
26 12 13 32 14 3
235 [94–615] 200 [95–650] 420 [330–470] 390 [260–885] 715 [457–887] 1100 [880–1150]
0.02
Use of perfumes Never 1/month 1/week 2–4/week Daily Several times/day
10 10 8 18 50 4
114 [93–397] 245 [175–898] 270 [173–438] 445 [295–620] 440 [263–855] 1500 [858–2575]
0.014
Fish consumption Never 1/month 1/week >1/week
9 41 39 11
330 450 390 630
0.041
[190–390] [159–765] [230–770] [190–930]
p-value from Mann–Whitney (gender and age) and Kruskal–Wallis tests (other independent variables).
Women showed significantly higher levels of polycyclic musk compounds in blood. This may be explained by the facts that women use creams, body lotions and perfumes more often than men and that polycyclic musks can be taken up through the skin. In a multivariate approach only use of body lotion and (younger) age were predictive of positive galaxolide concentrations (higher than the limit of quantification). For tonalide no significant predictor could be detected. In conclusion, the results of this study which investigated the levels of polycyclic musk compounds in the blood plasma of young adults in Austria mirror the replacement of nitro musk fragrances by polycyclic musks, mainly galaxolide. The high concentrations of galaxolide in human blood should encourage toxicological investigations.
Acknowledgements This study was supported by the Austrian Federal Ministry of Social Security, Generations and Consumer
Protection. Especially the assistance of Mrs. Brigitte Piegler is gratefully acknowledged.
References Bauer, K., Fro¨ssl, C., 1999. Blutkonzentrationen von polycyclischen- und Nitromoschusverbindungen bei deutschen Probanden. Umwelt Medizin Gesellschaft 12, 235–237. Bitsch, N., Dudas, C., Korner, W., Failing, K., Biselli, S., Rimkus, G., Brunn, H., 2002. Estrogenic activity of musk fragrances detected by the E-screen assay using human mcf7 cells. Arch. Environ. Contam. Toxicol. 43, 257–264. Eisenhardt, S., Runnebaum, B., Bauer, K., Gerhard, I., 2001. Nitromusk compounds in women with gynecological and endocrine dysfunction. Environ. Res. 87, 123–130. Eschke, H.D., Traud, J., Dibowski, H.J., 1995. Nachweis und quantifizierung von polycyclischen Moschus–Duftstoffen mittels ion-trap GC/MS/MS in Humanfett und Muttermilch. Dt. Lebensm. Rdschau 12, 375–379. Eschke, H.D., Traud, J., Dibowski, H.J., 1994. Untersuchungen zum Vorkommen polycyclischer Moschus–Duftstoffe in verschiedenen Umweltkompartimenten–Nachweis und Analytik mit GC/MS in Oberfla¨chen-, Abwa¨ssern und
492
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
¨ kotox. Fischen (1.Mitteilung). UWSF—Z. Umweltchem. O 6, 183–189. Fromme, H., Otto, T., Pilz, K., Neugebauer, F., 1999. Levels of synthetic musks; bromocyclene and PCBs in eel (Anguilla anguilla) and PCBs in sediment samples from some waters of Berlin/Germany. Chemosphere 39, 1723–1735. Gatermann, R., Biselli, S., Hu¨hnerfuss, H., Rimkus, G.G., Hecker, M., Karbe, L., 2002. Synthetic musks in the environment. Part 1: Species-dependent bioaccumulation of polycyclic and nitro musk fragrances in freshwater fish and mussels. Arch. Environ. Contam. Toxicol. 42, 437–446. Ka¨fferlein, H.U., Angerer, J., 2001. Trends in the musk xylene concentrations in plasma samples from the general population from 1992/1993 to 1998 and the relevance of dermal uptake. Int. Arch. Occup. Environ. Health 74, 470–476. Ka¨fferlein, H.U., Goen, T., Angerer, J., 1998. Musk xylene: Analysis, occurrence, kinetics, and toxicology. Crit. Rev. Toxicol. 28, 431–476. Liebl, B., Ehrenstorfer, S., 1993. Nitro musks in human milk. Chemosphere 27, 2253–2260. Liebl, B., Mayer, R., Ommer, S., Sonnichsen, C., Koletzko, B., 2000. Transition of nitro musks and polycyclic musks into human milk. Adv. Exp. Med. Biol. 478, 289–305. Mersch-Sundermann, V., Kevekordes, S., Jenter, C., 1998. Testing of SOS induction of artificial polycyclic musk fragrances in E. coli PQ37 (SOS chromotest). Toxicol. Lett. 95, 147–154.
Mu¨ller, S., Schmid, P., Schlatter, C., 1996. Occurrence of nitro and non-nitro benzenoid musk compounds in human adipose tissue. Chemosphere 33, 17–28. Peck, A.M., Hornbuckle, K.C., 2004. Synthetic musk fragrances in Lake Michigan. Environ. Sci. Technol. 38, 367– 372. Rimkus, G.G., 1999. Polycyclic musk fragrances in the aquatic environment. Toxicol. Lett. 111, 37–56. Rimkus, G.G., Wolf, M., 1996. Polycyclic musk fragrances in human adipose tissue and milk. Chemosphere 33, 2033– 2043. Rimkus, G.G., Wolf, M., 1995. Nitro musk fragrances in biota from freshwater and marine environment. Chemosphere 30, 641–651. Rimkus, G., Rimkus, B., Wolf, M., 1994. Nitro musks in human adipose tissue and breast milk. Chemosphere 28, 421–432. Schmeiser, H.H., Gminski, R., Mersch-Sundermann, V., 2001. Evaluation of health risks caused by musk ketone. Int. J. Hyg. Environ. Health 2001, 293–299. Schreurs, R.H., Quaedackers, M.E., Seinen, W., van der Burg, B., 2002. Transcriptional activation of estrogen receptor ERalpha and ERbeta by polycyclic musks is cell type dependent. Toxicol. Appl. Pharmacol. 183, 1–9. Yamagishi, T., Miyazaki, T., Horii, S., Akiyama, K., 1981. Identification of musk ketone in freshwater fish collected from the Tama River, Tokyo. Bull. Environ. Contam. Toxicol. 26, 656–662.
Blood concentrations of polycyclic musks in healthy young adults H.-P. Hutter a,*, P. Wallner b, H. Moshammer a, W. Hartl c, R. Sattelberger c, G. Lorbeer c, M. Kundi a a
Institute of Environmental Health, Center for Public Health, Medical University of Vienna, Kinderspitalgasse 15, A-1095 Vienna, Austria b Medicine and Environmental Protection, Vienna, Austria c Federal Environmental Agency, Vienna, Austria
Received 31 August 2004; received in revised form 14 January 2005; accepted 19 January 2005
Abstract Knowledge on the concentration of polycyclic musk fragrance compounds in human blood is sparse. This study examined the concentrations of six polycyclic musks in blood samples from healthy volunteers. Blood was taken from hundred healthy students of the Medical University of Vienna. The lipophilic fraction was extracted and after purification analyzed by GC–MS. Study participants also completed a questionnaire on the use of cosmetics, about nutrition and other life-style aspects. Two compounds—galaxolide and tonalide—were identified in higher percentages of the blood plasma samples. Maximum plasma levels over 100 ng/l were also only found for galaxolide (4100 ng/l) and tonalide (800 ng/l). Women showed significantly higher levels than men. In a statistical multivariate approach only use of body lotion and age were predictive of positive galaxolide concentrations. For tonalide no significant predictor could be found. The findings mirror the replacement of nitro musk fragrances by polycyclic musks, mainly galaxolide. The high concentrations of galaxolide in human blood raise concern since few toxicological data are available. 2005 Elsevier Ltd. All rights reserved. Keywords: Polycyclic musks; Galaxolide; Tonalide; Celestolide; Phantolide; Cashmeran; Traesolide; Human blood
1. Introduction Synthetic musks are used as fragrances in numerous products, such as perfumes, soaps, body lotions, cosmet-
*
Corresponding author. Tel.: +43 1 4277 64727; fax: +43 1 4277 9647. E-mail address: [email protected] (H.-P. Hutter).
ics, fabric softeners, laundry detergents, air fresheners, food additives and fish bait (Rimkus, 1999; Schmeiser et al., 2001). The first synthetic musks (dinitro- and trinitrobenzene derivates, so called nitro musks) were introduced into the market in the beginning of the last century (Rimkus, 1999). Since the 1950s polycyclic musk fragrances have also been used. Today, the production of polycyclic musk compounds is increasing, with a corresponding decline in nitro musk production (Rimkus, 1999; Peck and Hornbuckle, 2004).
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.01.070
488
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
Both groups of chemicals are lipophilic and persistent in fat tissue. They bioaccumulate in the aquatic environment (Yamagishi et al., 1981; Eschke et al., 1994; Rimkus and Wolf, 1995; Fromme et al., 1999; Rimkus, 1999; Gatermann et al., 2002) and can be found in human fat tissue and breast milk (Liebl and Ehrenstorfer, 1993; Rimkus et al., 1994; Eschke et al., 1995; Mu¨ller et al., 1996; Rimkus and Wolf, 1996; Ka¨fferlein et al., 1998; Liebl et al., 2000). In contrast to nitro musks, data on the toxicological properties of polycyclic musk compounds are sparse (Mersch-Sundermann et al., 1998; Liebl et al., 2000). In recent years, concerns have been raised about estrogenic activity of polycyclic musks (Bitsch et al., 2002; Schreurs et al., 2002). Only few studies on synthetic polycyclic musks in human fat tissue and breast milk have been published (Eschke et al., 1995; Mu¨ller et al., 1996; Rimkus and Wolf, 1996; Liebl et al., 2000). Data on the concentration of polycyclic musk fragrance compounds in human blood are lacking. To our knowledge, only one study has been published so far (Bauer and Fro¨ssl, 1999). The aim of this study was to determine the levels of polycyclic musks in blood plasma of young adults in Austria and to investigate the influence of gender, body fat, eating habits, use of cosmetics, etc. on the concentrations of these compounds.
2. Materials and methods Blood samples were collected with Vacuette plasma tubes (coated with sodium heparin) from 114 healthy students who attended courses at the Hygiene Institute of the Medical University of Vienna. Blood samples were put immediately after collection into a refrigerator at 20 C. From these samples 11 were used for validation purposes of the assays, 100 samples served as basis for the analysis. Height, weight, and subscapular skinfold thickness of the study participants were measured. Furthermore, subjects were asked if they had normal, dry or oily skin, how often they used body care products, how often they ate fish, etc.
To remove potential residual contamination with musk compounds all glassware was dried at 200 C for 12 h. In order to avoid losses by adsorption at the glass walls, the walls were rinsed with iso-octane before use. After the solvent was discarded, the residue solvent was evaporated under a fume hood. Glassware of the rotary evaporator was also rinsed prior to use. Sample preparation was carried out in a laminar flow chemical hood with a charcoal filter. 2.1.2. Sample preparation The samples (8.5–9 ml) were defrosted and the internal standard (50 ll of a 100 pg/ll solution of deuterated tonalide-D3 in acetone) was added. Precipitation of the proteins was performed using acetonitrile (8 ml) and shaking vigorously. 3 ml n-pentane was added and shaking was continued for 2 min. The emerging coagulate was homogenised with a spatula. Extraction was carried out on a shaking board for 15 min followed by centrifugation (5 min: 2500g). The pentane phase was withdrawn to a new test tube. The extraction was repeated with n-pentane (2 ml); extraction time was 15 min (shaking board), centrifugation followed for 5 min at 2500g. This procedure was repeated twice. The combined pentane phases were concentrated and reconstructed to 1 ml hexane. Clean up was carried out with silica gel, conditioned by isooctane, dichloromethane and n-hexane, each 5 ml. Elution was performed with hexane/ethyl acetate (4:1) (8 ml), dichloromethane (8 ml) and ethyl acetate (10 ml). The elute was collected and concentrated (1 ml). Another clean up step included elution on an aluminium oxide column (coated with sodium sulphate: 0.5 g) with hexane as solvent (10 ml). Therefore 2 g aluminium oxide were treated overnight at 450 C and deactivated with 10% water. Prior to GC–MS analysis hexachlorobenzene (HCB) 13C6 as injection standard was added. With each sample batch, a blank sample was analyzed. For blank samples 8 ml of acetonitrile were spiked with 50 ll of a 100 pg/ll solution of tonalide-D3 in Acetone. The rest of the sample preparation followed the method described above. Average (n = 11) blind values range from 30 ng/l (tonalide) up to 50 ng/l (galaxolide). For celestolide, phantolide, cashmeran, traesolide blind values are below LOD (Limit of Detection).
2.1. Analytical procedure 2.1.1. Preparatory operations Special care and several preparatory steps are necessary to avoid or reduce contamination of laboratory equipment with the musk compound analyts. Potential sources of contamination with musk compounds are laundry detergents, towels, cosmetics used by the staff, household cleaning agents applied in the laboratory but also the medical tubes used for taking blood samples. Different kinds of medical tubes were tested to select those with the lowest impact on blind values.
2.1.3. Analysis The GC–MS analysis was carried out by a Thermoquest Trace GC/MS System with negative chemical ionisation using selected ion monitoring (SIM) mode (Thermo Electron, USA). Ammonia was applied as reagent gas. A fused silica capillary column (60 m · 0.25 mm i.d., 1.4 lm film thickness) of crosslinked DB624 (J&W, USA) was used. The operating conditions were as follows: column temperature, programmed from 110 C to 210 C at 15 C/min, from 210 C to 270 C at 6 C/min, and hold at 270 C for 18 min. Injection tem-
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
perature was 260 C, transfer line temperature 260 C, and source temperature 255 C. The corresponding ions used for SIM recording and quantification are listed in Table 1. An example of chromatographic separation is given in Fig. 1. For quantification of polycyclic musk compounds the use of tonalide-D3 as internal standard compensates errors made during injection, changes in recoveries caused by matrix effects and losses during sample pretreatment.
489
2.1.4. Characteristics Recovery rates: To verify that tonalide-D3 is a suitable internal standard for quantification of all polycyclic musk compounds the recovery rates of single musk compounds and tonalide-D3 were determined from blood samples fortified with the analyts using the method described above (number of samples = 11). For calculation of the recovery rates, HCB 13C6 was used as injection standard (Table 1).
Table 1 Ions used for SIM (selected ion monitoring) recording and quantification, recovery rates and standard deviations for single musk compounds (n = 11), LODs and LOQs for single musk compounds (values referred to a sample volume of 9 ml)
Cashmeran Celestolide Phantolide Galaxolide Traesolide Tonalide HCB 13C6 Tonalide-D3
m/z
Recovery rates (%)
Standard deviation (%)
LOD (ng/l)
LOQ (ng/l)
205.3 243.3 243.3 257.3 257.3 257.3 290.0 260.3
97.7 99.4 97.5 95.1 97.2 96.4 – 94.3
2.7 3.7 3.5 5.0 5.9 5.0 – 3.8
12.0 25.0 12.0 62.0 25.0 31.0 – –
24.0 50.0 24.0 124.0 50.0 62.0 – –
LOD: limit of detection; LOQ: limit of quantification; m/z: mass to charge ratio.
27.38
100 95
Tonalide
90 85 80 75
65 60
30
Celestolide
45
25
18.83
22.64
40 35
Phantolide
50
26.44
24.08
20
Traesolide
Galaxolide
55
Cashmeran
Relative Abundance
70
26.64
15 10 5 0 19
20
21
22
23
24
25
26
27
Time (min)
Fig. 1. SIM chromatogram of six polycyclic musk compounds (m/z 205.3 RT: 18.0–21.0 min; m/z 243.3 RT: 21.0–25.0 min; m/z 257.3 RT: 25.0–28.0 min). m/z: mass to charge ratio.
490
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
Limits of detection, limits of quantification: The limits of detection (LOD) given in Table 1 are calculated for the following conditions: • • • •
signal to noise ratio of 3 to 1 (S/N = 3), sample volume 9 ml, average recovery rates (Table 1), blind values (see Section 2.1.2).
The limits of quantification (LOQ) are twice the limits of detection. 2.2. Statistical procedures Differences in musk compounds concentrations between groups were calculated using non-parametric tests (U test, Kruskal–Wallis test). Stepwise logistic regression analysis with the following variables as potential predictors of a positive plasma concentration (>LOQ) of the different musk compounds was applied: Age, gender, body mass index, subscapular skinfold thickness, skin type, fish consumption, intensity of use of body lotions, perfumes, etc. These analyses were conducted only for musks with a rate of positive plasma concentrations of 10% or more. A p-value below 0.05 was considered significant. 3. Results From the 100 students 55 were female and 45 male. The age ranged from 19 to 43 years with a median of
Table 2 Body mass index (BMI), subscapular skinfold thickness (SST), fish consumption, skin type and use of cosmetics (n = 100) Parameter BMI (Md, [Q1–Q3]) SST (Md, [Q1–Q3]) Fish consumption (at least once a week)
22 [20–23] 15 [12–18] 50%
Skin type: Normal Dry Oily
62% 31% 7%
Use of creams (daily): On face On hands On body
25% 40% 46%
Use of (daily/more than once a day): Body lotion Perfumes Deodorants Shampoo Fragrance oil Spray
17% 54% 70% 21% 3% 3%
Md: median.
23 years. Table 2 shows descriptive data concerning body mass index, subscapular skinfold thickness, fish consumption and use of diverse cosmetic products. Subjects were predominantly slim. Fish was regularly consumed by half of the students, however, predominantly sea fish. ‘‘At least daily use’’ of perfumes and deodorants was reported by more than half of the students. Two polycyclic musk compounds, galaxolide and tonalide, were identified in higher percentages of the blood plasma samples. Galaxolide was detected in 91% of the samples (in 83% of the samples >LOQ), tonalide in 17% of the samples (16% >LOQ). Phantolide was found in one subject and traesolide in two, with levels lower than 100 ng/l. Cashmeran and celestolide were below the detection limit in all blood specimens. Median concentrations of galaxolide were 420 ng/l, and of tonalide below LOD. Maximum plasma levels were 4100 ng/l for galaxolide and 800 ng/l for tonalide. For galaxolide women showed significantly higher levels than men and lower values in the higher age group (26– 43 years) were detected (Table 3). Persons who more often used cream on their face, hands, or body showed significantly higher plasma concentrations of galaxolide, but not of tonalide. Skin type (dry, normal, or oily) had no influence on the concentrations of these two compounds. Significant effects on the rate of positive galaxolide plasma concentrations were found for gender, age, use of body lotion, perfumes and for fish consumption, with increasing rates for females, younger age, frequent use of body lotion and perfumes and high frequency of fish consumption. In the multivariate approach only use of body lotion and age were predictive of positive galaxolide concentrations. For tonalide no significant predictor could be detected. All other compounds had rates of positivity below 10% and were not further analysed.
4. Discussion Today, the use of polycyclic musk fragrances compounds is increasing. However, data on the toxicological properties of these chemicals and their levels in human blood are sparse. In this study especially galaxolide was detected in a high percentage (91%) of the blood samples of young adults in Vienna, Austria. Tonalide was found in 17%. The plasma concentrations of galaxolide were comparable to those from a German human biomonitoring study which was performed in 1999 (Bauer and Fro¨ssl, 1999), whereas in the present investigation levels of tonalide were about 10 times lower. In relation to nitro musks (Bauer and Fro¨ssl, 1999; Eisenhardt et al., 2001; Ka¨fferlein and Angerer, 2001) galaxolide concentrations were comparatively high, with a maximum up to more than 4 lg/l.
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
491
Table 3 Median galaxolide plasma concentration for stratification according to gender, age, use of body lotion, use of perfumes and fish consumption Independent variable
n
Plasma concentration (ng/l) Md [Q1–Q3]
p-value
Gender Male Female
45 55
260 [98–540] 580 [290–885]
0.002
Age 19–25 26–43
67 33
450 [250–805] 340 [130–670]
0.045
Use of body lotion Never 1/month 1/week 2–4/week Daily Several times/day
26 12 13 32 14 3
235 [94–615] 200 [95–650] 420 [330–470] 390 [260–885] 715 [457–887] 1100 [880–1150]
0.02
Use of perfumes Never 1/month 1/week 2–4/week Daily Several times/day
10 10 8 18 50 4
114 [93–397] 245 [175–898] 270 [173–438] 445 [295–620] 440 [263–855] 1500 [858–2575]
0.014
Fish consumption Never 1/month 1/week >1/week
9 41 39 11
330 450 390 630
0.041
[190–390] [159–765] [230–770] [190–930]
p-value from Mann–Whitney (gender and age) and Kruskal–Wallis tests (other independent variables).
Women showed significantly higher levels of polycyclic musk compounds in blood. This may be explained by the facts that women use creams, body lotions and perfumes more often than men and that polycyclic musks can be taken up through the skin. In a multivariate approach only use of body lotion and (younger) age were predictive of positive galaxolide concentrations (higher than the limit of quantification). For tonalide no significant predictor could be detected. In conclusion, the results of this study which investigated the levels of polycyclic musk compounds in the blood plasma of young adults in Austria mirror the replacement of nitro musk fragrances by polycyclic musks, mainly galaxolide. The high concentrations of galaxolide in human blood should encourage toxicological investigations.
Acknowledgements This study was supported by the Austrian Federal Ministry of Social Security, Generations and Consumer
Protection. Especially the assistance of Mrs. Brigitte Piegler is gratefully acknowledged.
References Bauer, K., Fro¨ssl, C., 1999. Blutkonzentrationen von polycyclischen- und Nitromoschusverbindungen bei deutschen Probanden. Umwelt Medizin Gesellschaft 12, 235–237. Bitsch, N., Dudas, C., Korner, W., Failing, K., Biselli, S., Rimkus, G., Brunn, H., 2002. Estrogenic activity of musk fragrances detected by the E-screen assay using human mcf7 cells. Arch. Environ. Contam. Toxicol. 43, 257–264. Eisenhardt, S., Runnebaum, B., Bauer, K., Gerhard, I., 2001. Nitromusk compounds in women with gynecological and endocrine dysfunction. Environ. Res. 87, 123–130. Eschke, H.D., Traud, J., Dibowski, H.J., 1995. Nachweis und quantifizierung von polycyclischen Moschus–Duftstoffen mittels ion-trap GC/MS/MS in Humanfett und Muttermilch. Dt. Lebensm. Rdschau 12, 375–379. Eschke, H.D., Traud, J., Dibowski, H.J., 1994. Untersuchungen zum Vorkommen polycyclischer Moschus–Duftstoffe in verschiedenen Umweltkompartimenten–Nachweis und Analytik mit GC/MS in Oberfla¨chen-, Abwa¨ssern und
492
H.-P. Hutter et al. / Chemosphere 59 (2005) 487–492
¨ kotox. Fischen (1.Mitteilung). UWSF—Z. Umweltchem. O 6, 183–189. Fromme, H., Otto, T., Pilz, K., Neugebauer, F., 1999. Levels of synthetic musks; bromocyclene and PCBs in eel (Anguilla anguilla) and PCBs in sediment samples from some waters of Berlin/Germany. Chemosphere 39, 1723–1735. Gatermann, R., Biselli, S., Hu¨hnerfuss, H., Rimkus, G.G., Hecker, M., Karbe, L., 2002. Synthetic musks in the environment. Part 1: Species-dependent bioaccumulation of polycyclic and nitro musk fragrances in freshwater fish and mussels. Arch. Environ. Contam. Toxicol. 42, 437–446. Ka¨fferlein, H.U., Angerer, J., 2001. Trends in the musk xylene concentrations in plasma samples from the general population from 1992/1993 to 1998 and the relevance of dermal uptake. Int. Arch. Occup. Environ. Health 74, 470–476. Ka¨fferlein, H.U., Goen, T., Angerer, J., 1998. Musk xylene: Analysis, occurrence, kinetics, and toxicology. Crit. Rev. Toxicol. 28, 431–476. Liebl, B., Ehrenstorfer, S., 1993. Nitro musks in human milk. Chemosphere 27, 2253–2260. Liebl, B., Mayer, R., Ommer, S., Sonnichsen, C., Koletzko, B., 2000. Transition of nitro musks and polycyclic musks into human milk. Adv. Exp. Med. Biol. 478, 289–305. Mersch-Sundermann, V., Kevekordes, S., Jenter, C., 1998. Testing of SOS induction of artificial polycyclic musk fragrances in E. coli PQ37 (SOS chromotest). Toxicol. Lett. 95, 147–154.
Mu¨ller, S., Schmid, P., Schlatter, C., 1996. Occurrence of nitro and non-nitro benzenoid musk compounds in human adipose tissue. Chemosphere 33, 17–28. Peck, A.M., Hornbuckle, K.C., 2004. Synthetic musk fragrances in Lake Michigan. Environ. Sci. Technol. 38, 367– 372. Rimkus, G.G., 1999. Polycyclic musk fragrances in the aquatic environment. Toxicol. Lett. 111, 37–56. Rimkus, G.G., Wolf, M., 1996. Polycyclic musk fragrances in human adipose tissue and milk. Chemosphere 33, 2033– 2043. Rimkus, G.G., Wolf, M., 1995. Nitro musk fragrances in biota from freshwater and marine environment. Chemosphere 30, 641–651. Rimkus, G., Rimkus, B., Wolf, M., 1994. Nitro musks in human adipose tissue and breast milk. Chemosphere 28, 421–432. Schmeiser, H.H., Gminski, R., Mersch-Sundermann, V., 2001. Evaluation of health risks caused by musk ketone. Int. J. Hyg. Environ. Health 2001, 293–299. Schreurs, R.H., Quaedackers, M.E., Seinen, W., van der Burg, B., 2002. Transcriptional activation of estrogen receptor ERalpha and ERbeta by polycyclic musks is cell type dependent. Toxicol. Appl. Pharmacol. 183, 1–9. Yamagishi, T., Miyazaki, T., Horii, S., Akiyama, K., 1981. Identification of musk ketone in freshwater fish collected from the Tama River, Tokyo. Bull. Environ. Contam. Toxicol. 26, 656–662.