- Email: [email protected]
Biological monitoring of low benzene exposure in Italian traffic policemen
Toxicology Letters 181 (2008) 25–30
Contents lists available at ScienceDirect
Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet
Biological monitoring of low benzene exposure in Italian traffic policemen Paola Manini a,b,∗ , Giuseppe De Palma c , Roberta Andreoli a,b , Diana Poli b , Marta Petyx d , Massimo Corradi a , Antonio Mutti a , Pietro Apostoli c a
Laboratory of Industrial Toxicology, Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma, Via Gramsci 14, 43100 Parma, Italy ISPESL-National Institute for Occupational Safety and Prevention, Research Center at the University of Parma, Via Gramsci 14, 43100 Parma, Italy c Department of Experimental and Applied Medicine, Occupational Medicine and Industrial Hygiene, University of Brescia, Piazzale Spedali Civili 1, 25123 Brescia, Italy d ISPESL-National Institute for Occupational Safety and Prevention, Department of Occupational Medicine, Via di Fontana Candida, 00141Monteporzio Catone, Rome, Italy b
a r t i c l e
i n f o
Article history: Received 13 May 2008 Received in revised form 26 June 2008 Accepted 26 June 2008 Available online 4 July 2008 Keywords: Biological monitoring Benzene Traffic policemen S-Phenylmercapturic acid t,t-Muconic acid
a b s t r a c t A comparative evaluation of urinary biomarkers was carried out to characterize benzene exposure in a group of 100 traffic policemen of the city of Parma (Italy). All subjects were monitored once, in two consecutive days characterized by similar climatic conditions but preceded by two windy days. Benzene ambient concentration measured by municipal air monitoring stations was 1 g/m3 (Day 1) and 2 g/m3 (Day 2). Personal exposure to ambient concentrations of benzene, toluene, ethylbenzene and xylene (BTEX) was assessed by using Radiello® passive-diffusive samplers in a subgroup of 24 workers. Benzene metabolites, t,t-muconic acid (t,t-MA) and S-phenylmercapturic acid (S-PMA) were determined by isotopic dilution liquid chromatography–tandem mass spectrometry on spot urine samples collected at the end of the shift. Urinary benzene (U-B) was determined by solid-phase microextraction gas chromatography–mass spectrometry. Airborne benzene concentration expressed as median [and interquartile range] was 6.07 [0.28–9.53] g/m3 , as assessed by personal sampling. Urinary concentrations of biomarkers in the whole group were 41.8 [34.1–89.8] g/g creatinine for t,t-MA, 0.67 [0.23–1.32] g/g creatinine for S-PMA, and 0.16 [0.13–0.26] g/l for U-B. Smokers eliminated significantly higher concentrations of unchanged BTEX and benzene metabolites than non-smokers (p < 0.05). When traffic policemen were distinguished into indoor (n = 31) and outdoor workers, no significant differences were observed for either airborne benzene or urinary biomarkers. Significantly lower concentrations of S-PMA and U-B were determined in samples collected at Day 1 as compared to Day 2 (p < 0.0001 and p = 0.003, respectively) suggesting that these biomarkers are enough sensitive and specific to detect changes in airborne benzene concentration even at few g/m3 . © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Benzene is a class I carcinogenic chemical (International Agency for Research on Cancer, IARC), known to cause bone marrow damage, leukemia and aplastic anemia after long-term exposure to high concentrations (IARC, 1982; WHO, 1993). Although it is still used as solvent in some countries, benzene exposure in developed countries mainly occurs in the chemical and petroleum industry, and in urban environments, as benzene is a constituent of unleaded fuel (<1%, European Directive 98/70/EC) and a product of incomplete gas combustion. In the past decades, preventive actions were aimed at reducing benzene in industrial environments, where airborne
∗ Corresponding author at: Laboratory of Industrial Toxicology, Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma, Via Gramsci 14, 43100 Parma, Italy. Tel.: +39 0521 033060; fax: +39 0521 033076. E-mail address: [email protected] (P. Manini). 0378-4274/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2008.06.865
concentrations are now well below the occupational exposure limits set by the Italian law at 3.2 mg/m3 (or 1 ppm) and close to the recommended air quality parameter of 5–10 g/m3 . In addition, mainstream and side-stream tobacco smoke represents a relevant source of personal benzene intake (Brugnone et al., 1992; Melikian et al., 1999; Ghittori et al., 1999). Several studies pointed out that biomonitoring of workers exposed to benzene in urban environments (traffic policemen, bus and taxi drivers) represents a complex issue. We have recently characterized benzene exposure in the taxi drivers of a Northern Italy town, Parma, by using an integrated approach based on environmental and biological monitoring (Manini et al., 2006). The most sensitive and reliable benzene urinary biomarkers, namely S-phenylmercapturic acid (S-PMA), trans,trans-muconic acid (t,t-MA) and urinary benzene (U-B) were determined by using highly sensitive and selective analytical techniques. S-PMA resulted the most specific marker in agreement with previous studies (Boogard and van Sittert, 1995; Melikian et al., 2002; Qu et
26
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
al., 2003). Other studies, however, came to different conclusions, indicating U-B as the biomarker of choice in the biomonitoring of low benzene exposures (Waidyanatha et al., 2001; Fustinoni et al., 2005). There is a substantial agreement among different studies about the need of using chromatographic techniques coupled with mass spectrometry to obtain reliable results at low exposure levels. The aims of the present study were (i) to characterize exposure to low levels of benzene and other monoaromatic compounds in a group of traffic policemen of the city of Parma (Italy) by environmental and biological monitoring; (ii) to compare the sensitivity of urinary exposure biomarkers measured by using highly sensitive and selective analytical techniques based on mass spectrometry. 2. Subjects and methods 2.1. Subjects One hundred traffic policemen (66 males, 20 smokers, aged 41.4 ± 10.2 years) participated in this study conducted in February 2005. All subjects were monitored once, during two work-shifts (7:00–13:00 or 13:00–19:00), in two consecutive days characterized by different climatic conditions (Day 1—temperatures: max 12.4 ◦ C, min −0.5 ◦ C, air humidity about 36%, wind 25 km/h, no rain. Day 2—temperatures: max 8.0 ◦ C, min 0.4 ◦ C, air humidity about 45%, no wind, no rain. Source: University of Parma, Istituto e Osservatorio Meteorologico) and preceded by two windy days (36 km/h) that lowered airborne concentrations of pollutants, as shown in Fig. 1. Benzene ambient concentration measured by municipal air monitoring stations was 1 g/m3 (Day 1) and 2 g/m3 (Day 2) (source: ARPA, Agenzia Regionale Protezione Ambiente, Emilia Romagna). Information concerning work-shift, smoking habits and possible sorbic acid intake (as the number of candies ingested during the work-shift) was collected by questionnaire. Traffic policemen were classified according to the working task as indoor (n = 31) and outdoor workers (n = 69). All this information is summarized in Table 1. The study protocol was approved by the local Ethical Committee and the subjects participated after giving written, informed consent. 2.2. Chemicals Benzene, toluene, ethylbenzene, o-, m-, p-xylenes (BTEX, purity 99%), the corresponding deuterated standards (BTEX-d6 ) and 13 C6 -benzene used as internal standards (ISs), trans,trans-muconic acid (98%), cotinine (Cot, 98%), cotinine-d3
Fig. 1. The trend of the concentrations of particulate matter 10 (PM10 ) and airborne benzene in the town of Parma in February 2005. Both parameters decreased in the 2 days before the sampling Days 1 and 2. Source: ARPA, Emilia Romagna, 2005 (www.arpa.emr.it).
Table 1 Demographic characteristics of the study group and summary of information collected by questionnaire No. of subjects Age (years) Gender (male/female) No. of current/ex-/never-smokers Indoor/outdoor workers Cigarettes/day (mean ± S.E.M.) No. of candies (0/1–2/3–4)
100 41.4 ± 10.2 66/34 20/0/80 31/69 15.0 ± 1.6 (SD 7.3) 58/31/8
(99%), carbon disulfide (99.9+%), analytical grade formic acid and ammonium hydroxide were purchased by Sigma–Aldrich (Milan, Italy). dl-Phenylmercapturic acid (S-PMA, purity 98%) was supplied by TCI America (Portland, OR, USA). All standards were used without further purification. S-PMA-d5 and t,t-MA-d4 were obtained biosynthetically from rat urine and purified by solid-phase extraction (SPE) and HPLC (Melikian et al., 1999). HPLC-grade water and methanol were from Carlo Erba (Milan, Italy). Stock solutions containing about 1 g/l of BTEX and S-PMA were prepared in methanol; cotinine (1 g/l) was dissolved in water. The solubilization of t,t-MA (about 0.5 mg/l) in 0.1N aqueous sodium hydroxide was achieved by heating and stirring. 2.3. Exposure assessment Personal exposure to ambient concentrations of BTEX was assessed by using Radiello® passive-diffusive samplers (Fondazione S. Maugeri, Padua, Italy) in a subgroup of 24 workers distributed among all four work-shifts. The sampler was worn by the traffic policemen for the whole work-shift. BTEX were desorbed according to the instructions given by the manufacturer, using low-benzene carbon disulfide. Analytical determinations were performed on a HP 5890 gas chromatograph equipped with a HP 7673 autosampler, a split/spliteless injector, and a flame ionization detector (Agilent Technologies, Palo Alto, CA, USA), using the NIOSH method 1501. The limit of detection (LOD) of the whole procedure was 0.5–1 g (as absolute amount) or 0.5–1.5 g/m3 (as airborne concentration for a 12-h sampling) for all BTEX, and the reproducibility, calculated at 5 g/m3 and expressed as % CV, was 4–6% and 5–8% within and between series, respectively. In the case of benzene, the LOD was 1 g or 1 g/m3 , and the %CV was 6% and 8% (within and between series, respectively). 2.4. Biological monitoring Spot urine samples collected at the end of the work-shift were divided into three aliquots and frozen at −20 ◦ C until analyses of unchanged BTEX, benzene metabolites, and cotinine, respectively. Urinary benzene and other unmodified TEX were determined by solid-phase microextraction gas chromatography–mass spectrometry (SPME GC–MS), as previously described (Andreoli et al., 1999). Briefly, 2 ml of urine were immediately transferred into 4.0-ml SPME glass vials containing NaCl (1.0 g), and added with 2 l of a IS mixture [13 C6 -benzene (0.5 g/l), ethylbenzene-d6 (1 g/l), toluene-d6 and p-xylene-d6 (2 g/l)]. Samples were vigorously shaken and stored at −20 ◦ C until analysis. A 75-m Carboxen PDMS fiber (Supelco, Bellefonte, PA, USA) mounted on a Combi/Pal System autosampler (CTC Analytics, Zwingen, Switzerland) was used for headspace SPME sampling, performed at 45 ◦ C for 30 min under stirring conditions. Analyses were carried out on HP 6890 GC coupled with a HP 5973 mass spectrometer, controlled by a HP Chem Station (Agilent Technologies). Benzene metabolites, t,t-MA and S-PMA, were determined by isotopic dilution liquid chromatography–tandem mass spectrometry (LC–MS-MS) using a PE-Sciex API 365 triple–quadrupole mass spectrometer (Applied Biosystems, Thornhill, Canada) equipped with a ionspray interface for pneumatically assisted electrospray (ESI). Before analyses, urine samples were centrifuged at 3000 × g for 10 min, added with a IS mixture containing t,t-MA-d4 and S-PMA-d5 , and acidified with formic acid (0.1 M). A volume of 20 l was then injected on a Supelcosil LC-18-DB column (75 mm × 3.0 mm i.d., 3 m; Supelco). Elution was achieved at a flow-rate of 0.50 ml/min by running a linear gradient starting from 2% (hold 1.5 min) to 80% methanol (in 6.5 min, and then hold 1 min) in 20 mM aqueous formic acid. Analytes were ionized by negative-ion ESI and detection was performed in selected reaction monitoring (SRM) mode following the transitions characteristic of the analytes and ISs, m/z 141 → 97 for t,t-MA, m/z 145 → 101 for t,t-MA-d4 , m/z 238 → 109 for SPMA, and m/z 243 → 114 for S-PMA-d5 . Concentrations of urinary metabolites were expressed as a function of creatinine concentration (g/g creatinine), measured by the method of Jaffe (Henry, 1974). For quantitative analyses, calibrations were performed in the matrix, by spiking pooled urine samples from non-exposed subjects with appropriate standard mixtures. The LOD, calculated as the ratio S/N > 3, was 0.1 g/l for S-PMA and 2.5 g/l for t,t-MA. The coefficient of variation of the method (calculated at 1.0 g/l and 100 g/l for S-PMA and t,t-MA, respectively, and expressed as %CV) was within 1.3% and 2.6% for all intra- and inter-day determinations. Urinary free cotinine (U-cotinine) was determined by isotopic dilution LC–MSMS. Before analyses, urine samples were added with the IS (cotinine-d3 ) and
P. Manini et al. / Toxicology Letters 181 (2008) 25–30 centrifuged at 3000 × g for 10 min. Chromatography was performed on an Atlantis® dC18 column (100 mm × 2.0 mm i.d., 3 m; Waters, Milford, MA, USA) using variable proportions of 10 mM aqueous formic acid (pH 3.75) and methanol. Elution program: 12% methanol, hold for 12 min; from 12% to 70% methanol in 2.5 min (linear gradient); 70% methanol, hold for 1 min. The flow-rate was 0.2 ml/min and the injection volume 30 l. Analytes were ionized in positive-ion mode and the transitions chosen for SRM detection of cotinine and its IS were m/z 177 → 80 and m/z 180 → 101, respectively. The latter transition of the IS was preferred to the most intense m/z 180 → 80 in order to avoid cross-talking phenomena occurring in the collision cell when different SRM events have the same product ions formed from different precursor ions. The LOD was 0.2 g/l (20 l injected), the %CV (at 1 g/l) below 2% for all intra- and inter-day determinations. 2.5. Statistics Statistical analyses were carried out by the SPSS software (version 14.0 for Windows, Chicago, IL). All analytical determinations were above the corresponding limit of detections. Parametric statistical tests were applied to log-transformed values, when necessary to obtain a normal distribution, which was assessed by the one-sample Kolmogorov–Smirnov test. S-PMA and U-TEX followed a log-normal distribution, t,t-MA and U-B did not distributed normally or log-normally. Nonparametric statistical tests were primarily applied and results were then confirmed by parametric tests on log-transformed variables when applicable. Differences between groups were assessed using the Mann–Whitney U-test for independent samples, and the Spearman’s rho () was used to assess the correlation between variables. Multiple linear regression analysis models were used to assess the contribution of age, sex (males encoded as 1, females as 2), the smoking habits as U-cotinine, sorbic acid intake as the number of candies ingested during the workshift, the working place (indoor encoded as 1, outdoor encoded as 2), and the work-shift (Day 1 encoded as 1, Day 2 encoded as 2) to the variability of benzene biomarkers, set as dependent variables. The significance level for all tests was p ≤ 0.05 (two-tailed). Stepwise regression analyses were run using a significance level of 0.05 for entry and 0.10 for removal from the model.
3. Results 3.1. Ambient monitoring Table 2 summarizes airborne benzene concentrations of BTEX expressed as medians [and interquartile range] in the whole group of traffic policemen and in subgroups of workers classified according to the working task as indoor and outdoor workers. All BTEX values were considerably lower (about 3–4 orders of magnitude) than the corresponding occupational exposure limits (OELs), i.e. 3.2 mg/m3 , 188 mg/m3 , 442 mg/m3 and 221 mg/m3 , respectively. Airborne BTEX concentrations were higher at Day 2 compared to Day 1, but the differences were not statistically significant. When traffic policemen were classified according to their working task,
27
significantly higher concentrations of toluene and xylenes were observed for outdoor workers compared to indoor clerks (p < 0.05). No differences were detected between smokers and non-smokers (data not shown). The ambient concentrations of TEX aromatic hydrocarbons but benzene were correlated each other (data not shown). 3.2. Biological monitoring The distribution of biomarkers of exposure expressed as medians [and interquartile range] in the whole group of traffic policemen and in subgroups of workers classified according to the smoking habits is summarized in Table 3. S-PMA and U-TEX followed a lognormal distribution, whereas t,t-MA and U-B did not distributed normally or log-normally. All values are well below the biological limit values for occupationally exposed people and within the reference values for the general unexposed population. Smokers eliminated significantly higher concentrations of unchanged benzene and its metabolites (p < 0.0001) and other unchanged TEX than non-smokers (p < 0.05) as shown in Table 3. When traffic policemen were distinguished into indoor (n = 31) and outdoor workers (n = 69), no significant differences were observed for urinary biomarkers (data not shown). Significantly lower concentrations of S-PMA, U-B and U-T were determined in samples collected at Day 1 compared to Day 2, i.e. 0.24 [0.17–0.41] vs. 1.22 [0.90–1.92] g/g creatinine for SPMA (p < 0.0001), 0.15 [0.12–0.19] vs. 0.20 [0.16–0.36] g/l for U-B (p < 0.0001), and 0.19 [0.16–0.22] vs. 0.24 [0.17–0.30] g/l for U-T (p = 0.003). The differences were even more pronounced when only non-smoking subjects were considered, as illustrated in Fig. 2. In the narrow range examined in this study, airborne BTEX concentrations were not correlated with biological parameters. In the whole study group significant correlations were observed between urinary biomarkers (0.21 < < 0.75, p < 0.05) except between t,tMA and both U-EB and U-X. Table 4 shows Spearman’s correlation matrix among biological markers in subjects classified according to the smoking habits. Among non-smokers, significant correlations were observed between S-PMA and both t,t-MA and U-B (p < 0.01) and between U-TEX (p < 0.01). In the subgroup of smokers, benzene biomarkers were correlated with each other and with U-cotinine (p < 0.01). In addition, the self-reported number of cigarettes/day was highly correlated with U-cotinine ( = 0.59, p < 0.01), and a significant correlation was observed between the
Table 2 Distributions of BTEX airborne concentrations in the whole group of traffic policemen and in subgroups classified according to the working task as indoor and outdoor workers Compound 3
Benzene (g/m ) Toluene (g/m3 ) Ethylbenzene (g/m3 ) Xylenes (g/m3 )
All workers (n = 24)
Indoor workers (n = 5)
Outdoor workers (n = 19)
6.1 [0.3–9.5] 15.4 [4.0–28.3] 4.5 [3.1–7.7] 19.8 [13.6–30.1]
6.2 [1.8–7.4] 3.2 [3.1–8.0] 3.2 [2.5–4.5] 12.1 [10.3–17.0]
5.9 [0.3–12.8] 24.5 [10.0–32.8]* 4.8 [3.1–8.3] 22.4 [17.0–32.3]*
Airborne concentrations of toluene e xylene were significantly higher (p < 0.05) in outdoor traffic policemen compared to indoor workers. All BTEX values are well below the corresponding occupational exposure limits (OELs), e.g. 3.2 mg/m3 , 188 mg/m3 , 442 mg/m3 and 221 mg/m3 . *p < 0.01, Mann–Whitney U-test.
Table 3 Distributions of exposure biomarkers in the whole group of traffic policemen and in subgroups of workers classified according to the smoking habits Biomarker
All workers (n = 100)
Non-smokers (n = 80)
Smokers (n = 20)
U-Benzene (g/l) U-Toluene (g/l) U-Ethylbenzene (g/l) U-Xylenes (g/l) S-PMA (g/g creatinine) t,t-MA (g/g creatinine)
0.16 [0.13–0.26] 0.20 [0.16–0.26] 0.41 [0.32–0.49] 0.37 [0.29–0.45] 0.67 [0.23–1.32] 41.8 [34.1–89.7]
0.16 [0.13–0.19] 0.19 [0.16–0.22] 0.40 [0.32–0.47] 0.35 [0.28–0.44] 0.42 [0.20–1.07] 38.6 [31.7–51.6]
0.79 [0.24–1.92]** 0.27 [0.21–0.38]** 0.46 [0.38–0.61]* 0.43 [0.35–0.62]* 1.43 [0.92–2.13]** 124.7 [84.2–175]**
Values are expressed as median and interquartile range. All values are well below the biological limit values for occupationally exposed people and within the reference values for the general unexposed population. **p < 0.0001, *p < 0.05, Mann–Whitney U-test.
28
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
Fig. 2. Comparison between urinary concentrations of unchanged benzene (U-B) and toluene (U-T), and benzene metabolites S-PMA and t,t-MA in samples collected in different sampling days (Day 1 vs. Day 2) in subgroups of non-smoking (no. of subjects, 46 vs. 34) and smoking (6 vs. 14) traffic policemen. Data are shown as means ± S.E.M. Table 4 Spearman’s correlation coefficients between urinary biomarkers in subgroups of non-smoking (n = 80) and smoking (n = 20) traffic policemen t,t-MA Non-smokers S-PMA 0.242* t,t-MA U-B U-T U-EB U-X Smokers S-PMA t,t-MA U-B U-T U-EB U-X
0.538**
U-B
U-T
U-EB
U-X
U-cotinine
0.310** 0.233
0.121 0.031 0.602**
0.194 −0.012 0.098 0.327**
0.128 −0.027 0.085 0.309** 0.755**
0.215 0.064 −0.295* −0.169 0.064 0.104
0.755** 0.233
0.380 0.125 0.543*
−0.194 −0.128 −0.120 0.245
0.252 0.023 0.219 0.584** 0.588**
0.773** 0.656** 0.538** 0.268 0.017 0.277
**p < 0.01, *p < 0.05, Spearman’s correlation, two-tailed.
number of cigarettes/day and the concentration of S-PMA ( = 0.54, p < 0.05). Stepwise multiple regression models run to assess the role of smoking habits and other covariates on benzene biomarkers substantially confirmed the findings of univariate analyses. The best predictors for S-PMA were the sampling day (ˇ = 0.714, p < 0.0001), U-cotinine (ˇ = 0.368, p < 0.0001), and age (ˇ = −0.133, p = 0.025), the adjusted r2 of the model being 0.680 (p < 0.0001). U-Cotinine (ˇ = 0.652, p < 0.0001) together with the sampling day (ˇ = 0.207, p = 0.007) and the working place (ˇ = −0.203, p = 0.008) accounted for 52.7% of variance of U-B. Finally, only U-cotinine significantly contributed to the variability of t,t-MA (ˇ = 0.658, p < 0.0001, r2 = 0.427). 4. Discussion This study on Italian traffic policemen was conducted using the same integrated approach between ambient and biological monitoring applied in the case of taxi drivers of our town (Manini et al., 2006). Due to the larger number of enrolled subjects (n = 100), we performed personal ambient monitoring on 24 workers dis-
tributed among all work-shifts and representative of all working tasks. Airborne BTEX concentrations detected in this study were super imposable with those observed in the case of taxi drivers and were also comparable to those reported in other field studies, where Italian traffic policemen were monitored (Crebelli et al., 2001; Tomei et al., 2001; Maffei et al., 2005; Bono et al., 2005; Fustinoni et al., 2005). Airborne BTEX concentrations were lower (3–4 orders) than the occupational exposure limits set by the Italian law for the working environment (3200 g/m3 for benzene, DLgs 25/2) and benzene concentrations were close to the recommended air quality parameter of 5–10 g/m3 . Despite airborne concentrations have been significantly decreased in Italy during the past decade, benzene abatement represents a public health concern since it has been estimated that the risk of developing leukemia is about six cases per million among people who experience lifetime exposure to benzene concentrations of 1 g/m3 in air (WHO, 2000). Starting from this risk estimate, we can calculate the additional excess of risk due to occupational exposure in our study group. Assuming that traffic policemen are exposed for 6 h × 240 days/year × 40 years, the unit risk would become about 5.6 × 10−7 at an air benzene concentration of 1 g/m3 . An additional risk of 1.7 × 10−6 cases of leukemia can be estimated in our cohort of worker exposed to an excess of airborne benzene of 3.1 g/m3 as compared to the general population exposed to an annual mean benzene concentration of about 3 g/m3 . Significant differences between indoor and outdoor workers were observed only for toluene and xylene, which are the main aromatic components (<5%) of unleaded fuel after the reduction of benzene content (<1%). As already observed in our previous study on taxi drivers and in other studies (Crebelli et al., 2001; Sørensen et al., 2003), personal ambient monitoring was not useful to distinguish between smokers and non-smokers. Conversely, all benzene biomarkers including t,t-MA and urinary toluene were higher in smokers than in non-smokers and were correlated with both U-cotinine and with the number of cigarettes smoked the day of the sampling. Biomarkers reflect the whole absorbed dose of BTEX whereas ambient monitoring mirrors the uptake from the air only. This is why we did not observe any significant correlation between environmental and biological indices in the narrow range examined.
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
The main novelty of this study consists in the observation that the internal dose of benzene assessed by urinary biomarkers seems to be modified by a change in airborne benzene concentration even when it occurs at few g/m3 level. To the best of our knowledge, this is the first time that this finding has been reported. Therefore, it should not be over interpreted until a clear confirmation would come from further studies on larger populations. The trend of the concentrations of particulate matter 10 (PM10 ) and airborne benzene in the town of Parma measured in February 2005 by the municipal air monitoring stations is illustrated in Fig. 1. The first 2 weeks of the month were characterized by high concentrations of pollutants, particularly PM10 with eight values above the recommended threshold level of 50 g/m3 . Benzene concentrations were high in the first week and decreased in the second one up to the year’s mean value of 3 g/m3 . The sampling Days 1 and 2 (February 15 and 16) were preceded by two windy days that significantly lowered airborne concentrations of PM10 and further reduced airborne benzene levels to 1 g/m3 . Unexpectedly, benzene biomarkers S-PMA and U-B (but not t,t-MA) were significantly lower in workers sampled at Day 1 compared to those sampled at Day 2. When only the subgroup of non-smokers was considered, this difference was particularly marked for S-PMA (p < 0.0001) and was less evident for U-B (p < 0.001), as shown in Fig. 2. This different behaviour is probably due to the relatively long half-life (about 9 h) of S-PMA compared to that of U-B. Particularly, the concentration of S-PMA in urine samples collected at Day 1 was significantly lower than that we usually observe in non-smokers, indicating that nonsmoking traffic policemen have been exposed to reduced benzene concentrations in the hours before and during the work-shift. Conversely, when the subgroup of smokers is considered, none of the biomarkers was able to detect any difference related to work-shifts. This result supports the conclusion that benzene uptake arising from the smoking habits is considerably higher than that from air (Darrall et al., 1998), as already observed in the case of taxi drivers (Manini et al., 2006). In our previous study, benzene uptake arising from cigarette smoking modified background S-PMA and U-B values compared to non-smokers and resulted in a relative increase of S-PMA levels at the end of the shift compared to pre-shift values. In non-smoking subjects, S-PMA was not modified by prolonged exposure (about 11 h) to airborne benzene. For the general population exposed to few g/m3 , i.e. 2–3 g/m3 in the case of our town, and for workers doing their job in the urban environment, we hypothesize that benzene concentrations are at the steady-state in biological fluids and that metabolite excretion could not be modified by repeated exposures to similar daily concentrations. This study shows for the first time that a prolonged exposure (3 days) to reduced airborne benzene (1 g/m3 ) could modify benzene concentrations in biological fluids, leading to reduced excretion of unchanged benzene and its specific metabolite S-PMA. If confirmed on larger groups of subjects, this finding may encourage further benzene abatement in urban air, as there are exposure biomarkers which are enough sensitive and specific to assess benzene exposure and to mirror changes occurring even at few g/m3 in non-smoking subjects. The lack of performance of t,t-MA in distinguishing workers sampled at Day 1 from those sampled at Day 2 could be ascribed to the dietary intake of sorbic acid and sorbate, both widely used as preservatives in food. We tried to estimate sorbic acid intake by questionnaire as the number of ingested candies, without obtaining a significant contribution to the variance of t,t-MA. Actually, the sources for intake of sorbate are various and, besides candies, include bread and pastry, cheeses, yogurt, meat products, smoked and salted fish, fats and oils, salads, fermented and acidified vegetables, fruit products, dressings, jams, mayonnaise, wine and beverages. It has been demonstrated that about 0.23% of
29
ingested sorbic acid is metabolized to t,t-MA with a large interindividual variability (Renner et al., 1999). Assuming a daily intake of about 30 mg, it has been estimated that sorbic acid would account for at least 50% of t,t-MA background urinary excretion in nonoccupationally exposed non-smokers and 25% in smokers (Ruppert et al., 1997). The use of a more complex dietary questionnaire to assess sorbic acid intake or the simultaneous determination of sorbic acid in urine to correct t,t-MA levels (Renner et al., 1999; Negri et al., 2005) would be necessary, but it is out of the scope of the present investigation. The results of this study confirm that t,t-MA lacks the specificity needed for biomonitoring of low benzene exposure. 5. Conclusions Airborne BTEX concentrations were well below occupational exposure limits. Biomarkers of exposure felt within the reference intervals for the unexposed Italian population (BIOLIND.NET, 2008). Smoking habits was the main determinant and all biomarkers were inter-correlated with each other, despite the very low absorbed dose. A decrease of few g/m3 in benzene airborne levels, due to a change of the state of the atmosphere, resulted in significantly lower S-PMA and U-B concentrations in non-smoking traffic policemen. Finally, this study conducted on a larger group confirms our previous conclusion that S-PMA is the most sensitive and reliable benzene biomarker, probably due to its longer half-life and to its lower intra- and inter-individual variability as compared to U-B. Conflict of interest None. Acknowledgement The kind collaboration of the traffic policemen of the city of Parma is gratefully acknowledged. References Andreoli, R., Manini, P., Bergamaschi, E., Brustolin, A., Mutti, A., 1999. Solid-phase microextraction and gas chromatography–mass spectrometry for the determination of monoaromatic hydrocarbons in blood and urine: application to people exposed to air pollutants. Chromatographia 50, 167–172. ARPA (Agenzia Regionale Protezione Ambiente, Emilia Romagna), 2005. Available at http://www.arpa.emr.it/. BIOLIND.NET, 2008. Available at http://www.biolind.net. Bono, R., Traversi, D., Maestri, L., Schilirò, T., Ghittori, S., Baiocchi, C., Gilli, G., 2005. Urban air and tobacco smoke in benzene exposure in a cohort of traffic policemen. Chem. Biol. Interact. 153–154, 239–242. Boogard, P.J., van Sittert, N.J., 1995. Biological monitoring of exposure to benzene: a comparison between S-phenylmercapturic acid, trans,trans-muconic acid, and phenol. Occup. Environ. Med. 52, 611–620. Brugnone, F., Perbellini, L., Maranelli, G., Romeo, L., Lombardini, F., 1992. Reference values for blood benzene in occupationally unexposed general population. Int. Arch. Occup. Environ. Health 61, 513–518. Crebelli, R., Tomei, F., Zijno, A., Ghittori, S., Imbriani, M., Gamberale, D., Martini, A., Carere, A., 2001. Exposure to benzene in urban workers: environmental and biological monitoring of traffic police in Rome. Occup. Environ. Med. 58, 165–171. Darrall, K.G., Figgins, J.A., Brown, R.D., Phillips, G.F., 1998. Determination of benzene and associated volatile compounds in mainstream cigarette smoke. Analyst 123, 1095–1101. Fustinoni, S., Consonni, D., Campo, L., Buratti, M., Colombi, A., Pesatori, A.C., Bonzini, M., Bertazzi, P.A., Foà, V., Garte, S., Farmer, P.B., Levy, L.S., Pala, M., Valerio, F., Fontana, V., Desideri, A., Merlo, D., 2005. Monitoring low benzene exposure: comparative evaluation of urinary biomarkers, influence of cigarette smoking, and genetic polymorphisms. Cancer Epidemiol. Biomarkers Prev. 14, 2237–2244. Ghittori, S., Imbriani, M., Maestri, L., Capodaglio, E., Cavalleri, A., 1999. Determination of S-phenylmercapturic acid in urine as an indicator of exposure to benzene. Toxicol. Lett. 108, 329–334. Henry, R.J., 1974. Clinical Chemistry Principle and Techniques, second ed. Harper & Row, New York. IARC, 1982. Benzene. Some industrial chemicals and dyestuffs IARC monographs on the evaluation of carcinogenic risk to humans. 29, 93–148.
30
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
Maffei, F., Hrelia, P., Angelini, S., Carbone, F., Cantelli Forti, G., Barbieri, A., Sanguinetti, G., Mattioli, S., Violante, F.S., 2005. Effects of environmental benzene: micronucleus frequencies and hematological values in traffic police working in an urban area. Mutat. Res. 583, 1–11. Manini, P., De Palma, G., Andreoli, R., Poli, D., Mozzoni, P., Folesani, G., Mutti, A., Apostoli, P., 2006. Environmental and biological monitoring of benzene exposure in a cohort of Italian taxi drivers. Tox. Lett. 167, 142–151. Melikian, A.A., O’Connor, R., Prahalad, A.K., Hu, P., Li, H., Kagan, M., Thompson, S., 1999. Determination of the urinary benzene metabolites S-phenylmercapturic acid and trans,trans-muconic acid by liquid chromatography–tandem mass spectrometry. Carcinogenesis 20, 719–726. Melikian, A.A., Qu, Q., Shore, R., Li, G., Li, H., Jin, X., Cohen, B., Chen, L., Li, Y., Yin, S., Mu, R., Zhang, X., Wang, Y., 2002. Personal exposure to different levels of benzene and its relationships to the urinary metabolites S-phenylmercapturic acid and trans,trans-muconic acid. J. Chromatogr. A 778, 211–221. Negri, S., Bono, R., Maestri, L., Ghittori, S., Imbriani, M., 2005. High-pressure liquid chromatographic-mass spectrometric determination of sorbic acid in urine: verification of formation of trans,trans-muconic acid. Chem. Biol. Interact. 153–154, 243–246. Qu, Q., Shore, R., Li, G., Jin, X., Chen, L.C., Cohen, B., Melikian, A.A., Eastmond, D., Rappaport, S., Rupa, D., Waidyanatha, S., Yin, S., Yan, H., Meng, M., Winnik, W., Kwok, E.S., Li, Y., Mu, R., Xu, B., Zhang, X., Li, K., 2003. Validation and evaluation
of biomarkers in workers exposed to benzene in China. Res. Resp. Health Eff. Inst. 115, 1–87. Renner, T., Baer-Koetzle, M., Scherer, G., 1999. Determination of sorbic acid in urine by gas chromatography–mass spectrometry. J. Chromatogr. A 847, 127–133. Ruppert, T., Scherer, G., Tricker, A.R., Adlkofer, F., 1997. trans,trans-Muconic acid as a biomarker of non-occupational environmental exposure to benzene. Int. Arch. Occup. Environ. Health 69, 247–251. Sørensen, M., Skov, H., Autrup, H., Hertel, O., Loft, S., 2003. Urban benzene exposure and oxidative DNA damage: influence of genetic polymorphisms in metabolism genes. Sci. Total Environ. 309, 69–80. Tomei, F., Ghittori, S., Imbriani, M., Pavanello, S., Carere, A., Marcon, F., Martini, A., Baccolo, T.P., Tomao, E., Zijno, A., Crebelli, R., 2001. Environmental and biological monitoring of traffic warderns from the city of Rome. Occup. Med. 51, 198–203. Waidyanatha, S., Rothman, N., Fustinoni, S., Smith, M.T., Hayes, R.B., Bechtold, W., Dosemeci, M., Guilan, L., Yin, S., Rappaport, S.M., 2001. Urinary benzene as a biomarker of exposure among occupationally exposed and unexposed subjects. Carcinogenesis 22, 279–286. WHO, 1993. WHO Environmental health criteria 155. Biomarkers and risk assessment, concepts and principles. ICPS, Geneva. WHO Regional Office for Europe, 2000. Air Quality Guidelines for Europe, second ed. Who Regional Publications, pp. 62–66, European series, no. 91.
Contents lists available at ScienceDirect
Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet
Biological monitoring of low benzene exposure in Italian traffic policemen Paola Manini a,b,∗ , Giuseppe De Palma c , Roberta Andreoli a,b , Diana Poli b , Marta Petyx d , Massimo Corradi a , Antonio Mutti a , Pietro Apostoli c a
Laboratory of Industrial Toxicology, Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma, Via Gramsci 14, 43100 Parma, Italy ISPESL-National Institute for Occupational Safety and Prevention, Research Center at the University of Parma, Via Gramsci 14, 43100 Parma, Italy c Department of Experimental and Applied Medicine, Occupational Medicine and Industrial Hygiene, University of Brescia, Piazzale Spedali Civili 1, 25123 Brescia, Italy d ISPESL-National Institute for Occupational Safety and Prevention, Department of Occupational Medicine, Via di Fontana Candida, 00141Monteporzio Catone, Rome, Italy b
a r t i c l e
i n f o
Article history: Received 13 May 2008 Received in revised form 26 June 2008 Accepted 26 June 2008 Available online 4 July 2008 Keywords: Biological monitoring Benzene Traffic policemen S-Phenylmercapturic acid t,t-Muconic acid
a b s t r a c t A comparative evaluation of urinary biomarkers was carried out to characterize benzene exposure in a group of 100 traffic policemen of the city of Parma (Italy). All subjects were monitored once, in two consecutive days characterized by similar climatic conditions but preceded by two windy days. Benzene ambient concentration measured by municipal air monitoring stations was 1 g/m3 (Day 1) and 2 g/m3 (Day 2). Personal exposure to ambient concentrations of benzene, toluene, ethylbenzene and xylene (BTEX) was assessed by using Radiello® passive-diffusive samplers in a subgroup of 24 workers. Benzene metabolites, t,t-muconic acid (t,t-MA) and S-phenylmercapturic acid (S-PMA) were determined by isotopic dilution liquid chromatography–tandem mass spectrometry on spot urine samples collected at the end of the shift. Urinary benzene (U-B) was determined by solid-phase microextraction gas chromatography–mass spectrometry. Airborne benzene concentration expressed as median [and interquartile range] was 6.07 [0.28–9.53] g/m3 , as assessed by personal sampling. Urinary concentrations of biomarkers in the whole group were 41.8 [34.1–89.8] g/g creatinine for t,t-MA, 0.67 [0.23–1.32] g/g creatinine for S-PMA, and 0.16 [0.13–0.26] g/l for U-B. Smokers eliminated significantly higher concentrations of unchanged BTEX and benzene metabolites than non-smokers (p < 0.05). When traffic policemen were distinguished into indoor (n = 31) and outdoor workers, no significant differences were observed for either airborne benzene or urinary biomarkers. Significantly lower concentrations of S-PMA and U-B were determined in samples collected at Day 1 as compared to Day 2 (p < 0.0001 and p = 0.003, respectively) suggesting that these biomarkers are enough sensitive and specific to detect changes in airborne benzene concentration even at few g/m3 . © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Benzene is a class I carcinogenic chemical (International Agency for Research on Cancer, IARC), known to cause bone marrow damage, leukemia and aplastic anemia after long-term exposure to high concentrations (IARC, 1982; WHO, 1993). Although it is still used as solvent in some countries, benzene exposure in developed countries mainly occurs in the chemical and petroleum industry, and in urban environments, as benzene is a constituent of unleaded fuel (<1%, European Directive 98/70/EC) and a product of incomplete gas combustion. In the past decades, preventive actions were aimed at reducing benzene in industrial environments, where airborne
∗ Corresponding author at: Laboratory of Industrial Toxicology, Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma, Via Gramsci 14, 43100 Parma, Italy. Tel.: +39 0521 033060; fax: +39 0521 033076. E-mail address: [email protected] (P. Manini). 0378-4274/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2008.06.865
concentrations are now well below the occupational exposure limits set by the Italian law at 3.2 mg/m3 (or 1 ppm) and close to the recommended air quality parameter of 5–10 g/m3 . In addition, mainstream and side-stream tobacco smoke represents a relevant source of personal benzene intake (Brugnone et al., 1992; Melikian et al., 1999; Ghittori et al., 1999). Several studies pointed out that biomonitoring of workers exposed to benzene in urban environments (traffic policemen, bus and taxi drivers) represents a complex issue. We have recently characterized benzene exposure in the taxi drivers of a Northern Italy town, Parma, by using an integrated approach based on environmental and biological monitoring (Manini et al., 2006). The most sensitive and reliable benzene urinary biomarkers, namely S-phenylmercapturic acid (S-PMA), trans,trans-muconic acid (t,t-MA) and urinary benzene (U-B) were determined by using highly sensitive and selective analytical techniques. S-PMA resulted the most specific marker in agreement with previous studies (Boogard and van Sittert, 1995; Melikian et al., 2002; Qu et
26
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
al., 2003). Other studies, however, came to different conclusions, indicating U-B as the biomarker of choice in the biomonitoring of low benzene exposures (Waidyanatha et al., 2001; Fustinoni et al., 2005). There is a substantial agreement among different studies about the need of using chromatographic techniques coupled with mass spectrometry to obtain reliable results at low exposure levels. The aims of the present study were (i) to characterize exposure to low levels of benzene and other monoaromatic compounds in a group of traffic policemen of the city of Parma (Italy) by environmental and biological monitoring; (ii) to compare the sensitivity of urinary exposure biomarkers measured by using highly sensitive and selective analytical techniques based on mass spectrometry. 2. Subjects and methods 2.1. Subjects One hundred traffic policemen (66 males, 20 smokers, aged 41.4 ± 10.2 years) participated in this study conducted in February 2005. All subjects were monitored once, during two work-shifts (7:00–13:00 or 13:00–19:00), in two consecutive days characterized by different climatic conditions (Day 1—temperatures: max 12.4 ◦ C, min −0.5 ◦ C, air humidity about 36%, wind 25 km/h, no rain. Day 2—temperatures: max 8.0 ◦ C, min 0.4 ◦ C, air humidity about 45%, no wind, no rain. Source: University of Parma, Istituto e Osservatorio Meteorologico) and preceded by two windy days (36 km/h) that lowered airborne concentrations of pollutants, as shown in Fig. 1. Benzene ambient concentration measured by municipal air monitoring stations was 1 g/m3 (Day 1) and 2 g/m3 (Day 2) (source: ARPA, Agenzia Regionale Protezione Ambiente, Emilia Romagna). Information concerning work-shift, smoking habits and possible sorbic acid intake (as the number of candies ingested during the work-shift) was collected by questionnaire. Traffic policemen were classified according to the working task as indoor (n = 31) and outdoor workers (n = 69). All this information is summarized in Table 1. The study protocol was approved by the local Ethical Committee and the subjects participated after giving written, informed consent. 2.2. Chemicals Benzene, toluene, ethylbenzene, o-, m-, p-xylenes (BTEX, purity 99%), the corresponding deuterated standards (BTEX-d6 ) and 13 C6 -benzene used as internal standards (ISs), trans,trans-muconic acid (98%), cotinine (Cot, 98%), cotinine-d3
Fig. 1. The trend of the concentrations of particulate matter 10 (PM10 ) and airborne benzene in the town of Parma in February 2005. Both parameters decreased in the 2 days before the sampling Days 1 and 2. Source: ARPA, Emilia Romagna, 2005 (www.arpa.emr.it).
Table 1 Demographic characteristics of the study group and summary of information collected by questionnaire No. of subjects Age (years) Gender (male/female) No. of current/ex-/never-smokers Indoor/outdoor workers Cigarettes/day (mean ± S.E.M.) No. of candies (0/1–2/3–4)
100 41.4 ± 10.2 66/34 20/0/80 31/69 15.0 ± 1.6 (SD 7.3) 58/31/8
(99%), carbon disulfide (99.9+%), analytical grade formic acid and ammonium hydroxide were purchased by Sigma–Aldrich (Milan, Italy). dl-Phenylmercapturic acid (S-PMA, purity 98%) was supplied by TCI America (Portland, OR, USA). All standards were used without further purification. S-PMA-d5 and t,t-MA-d4 were obtained biosynthetically from rat urine and purified by solid-phase extraction (SPE) and HPLC (Melikian et al., 1999). HPLC-grade water and methanol were from Carlo Erba (Milan, Italy). Stock solutions containing about 1 g/l of BTEX and S-PMA were prepared in methanol; cotinine (1 g/l) was dissolved in water. The solubilization of t,t-MA (about 0.5 mg/l) in 0.1N aqueous sodium hydroxide was achieved by heating and stirring. 2.3. Exposure assessment Personal exposure to ambient concentrations of BTEX was assessed by using Radiello® passive-diffusive samplers (Fondazione S. Maugeri, Padua, Italy) in a subgroup of 24 workers distributed among all four work-shifts. The sampler was worn by the traffic policemen for the whole work-shift. BTEX were desorbed according to the instructions given by the manufacturer, using low-benzene carbon disulfide. Analytical determinations were performed on a HP 5890 gas chromatograph equipped with a HP 7673 autosampler, a split/spliteless injector, and a flame ionization detector (Agilent Technologies, Palo Alto, CA, USA), using the NIOSH method 1501. The limit of detection (LOD) of the whole procedure was 0.5–1 g (as absolute amount) or 0.5–1.5 g/m3 (as airborne concentration for a 12-h sampling) for all BTEX, and the reproducibility, calculated at 5 g/m3 and expressed as % CV, was 4–6% and 5–8% within and between series, respectively. In the case of benzene, the LOD was 1 g or 1 g/m3 , and the %CV was 6% and 8% (within and between series, respectively). 2.4. Biological monitoring Spot urine samples collected at the end of the work-shift were divided into three aliquots and frozen at −20 ◦ C until analyses of unchanged BTEX, benzene metabolites, and cotinine, respectively. Urinary benzene and other unmodified TEX were determined by solid-phase microextraction gas chromatography–mass spectrometry (SPME GC–MS), as previously described (Andreoli et al., 1999). Briefly, 2 ml of urine were immediately transferred into 4.0-ml SPME glass vials containing NaCl (1.0 g), and added with 2 l of a IS mixture [13 C6 -benzene (0.5 g/l), ethylbenzene-d6 (1 g/l), toluene-d6 and p-xylene-d6 (2 g/l)]. Samples were vigorously shaken and stored at −20 ◦ C until analysis. A 75-m Carboxen PDMS fiber (Supelco, Bellefonte, PA, USA) mounted on a Combi/Pal System autosampler (CTC Analytics, Zwingen, Switzerland) was used for headspace SPME sampling, performed at 45 ◦ C for 30 min under stirring conditions. Analyses were carried out on HP 6890 GC coupled with a HP 5973 mass spectrometer, controlled by a HP Chem Station (Agilent Technologies). Benzene metabolites, t,t-MA and S-PMA, were determined by isotopic dilution liquid chromatography–tandem mass spectrometry (LC–MS-MS) using a PE-Sciex API 365 triple–quadrupole mass spectrometer (Applied Biosystems, Thornhill, Canada) equipped with a ionspray interface for pneumatically assisted electrospray (ESI). Before analyses, urine samples were centrifuged at 3000 × g for 10 min, added with a IS mixture containing t,t-MA-d4 and S-PMA-d5 , and acidified with formic acid (0.1 M). A volume of 20 l was then injected on a Supelcosil LC-18-DB column (75 mm × 3.0 mm i.d., 3 m; Supelco). Elution was achieved at a flow-rate of 0.50 ml/min by running a linear gradient starting from 2% (hold 1.5 min) to 80% methanol (in 6.5 min, and then hold 1 min) in 20 mM aqueous formic acid. Analytes were ionized by negative-ion ESI and detection was performed in selected reaction monitoring (SRM) mode following the transitions characteristic of the analytes and ISs, m/z 141 → 97 for t,t-MA, m/z 145 → 101 for t,t-MA-d4 , m/z 238 → 109 for SPMA, and m/z 243 → 114 for S-PMA-d5 . Concentrations of urinary metabolites were expressed as a function of creatinine concentration (g/g creatinine), measured by the method of Jaffe (Henry, 1974). For quantitative analyses, calibrations were performed in the matrix, by spiking pooled urine samples from non-exposed subjects with appropriate standard mixtures. The LOD, calculated as the ratio S/N > 3, was 0.1 g/l for S-PMA and 2.5 g/l for t,t-MA. The coefficient of variation of the method (calculated at 1.0 g/l and 100 g/l for S-PMA and t,t-MA, respectively, and expressed as %CV) was within 1.3% and 2.6% for all intra- and inter-day determinations. Urinary free cotinine (U-cotinine) was determined by isotopic dilution LC–MSMS. Before analyses, urine samples were added with the IS (cotinine-d3 ) and
P. Manini et al. / Toxicology Letters 181 (2008) 25–30 centrifuged at 3000 × g for 10 min. Chromatography was performed on an Atlantis® dC18 column (100 mm × 2.0 mm i.d., 3 m; Waters, Milford, MA, USA) using variable proportions of 10 mM aqueous formic acid (pH 3.75) and methanol. Elution program: 12% methanol, hold for 12 min; from 12% to 70% methanol in 2.5 min (linear gradient); 70% methanol, hold for 1 min. The flow-rate was 0.2 ml/min and the injection volume 30 l. Analytes were ionized in positive-ion mode and the transitions chosen for SRM detection of cotinine and its IS were m/z 177 → 80 and m/z 180 → 101, respectively. The latter transition of the IS was preferred to the most intense m/z 180 → 80 in order to avoid cross-talking phenomena occurring in the collision cell when different SRM events have the same product ions formed from different precursor ions. The LOD was 0.2 g/l (20 l injected), the %CV (at 1 g/l) below 2% for all intra- and inter-day determinations. 2.5. Statistics Statistical analyses were carried out by the SPSS software (version 14.0 for Windows, Chicago, IL). All analytical determinations were above the corresponding limit of detections. Parametric statistical tests were applied to log-transformed values, when necessary to obtain a normal distribution, which was assessed by the one-sample Kolmogorov–Smirnov test. S-PMA and U-TEX followed a log-normal distribution, t,t-MA and U-B did not distributed normally or log-normally. Nonparametric statistical tests were primarily applied and results were then confirmed by parametric tests on log-transformed variables when applicable. Differences between groups were assessed using the Mann–Whitney U-test for independent samples, and the Spearman’s rho () was used to assess the correlation between variables. Multiple linear regression analysis models were used to assess the contribution of age, sex (males encoded as 1, females as 2), the smoking habits as U-cotinine, sorbic acid intake as the number of candies ingested during the workshift, the working place (indoor encoded as 1, outdoor encoded as 2), and the work-shift (Day 1 encoded as 1, Day 2 encoded as 2) to the variability of benzene biomarkers, set as dependent variables. The significance level for all tests was p ≤ 0.05 (two-tailed). Stepwise regression analyses were run using a significance level of 0.05 for entry and 0.10 for removal from the model.
3. Results 3.1. Ambient monitoring Table 2 summarizes airborne benzene concentrations of BTEX expressed as medians [and interquartile range] in the whole group of traffic policemen and in subgroups of workers classified according to the working task as indoor and outdoor workers. All BTEX values were considerably lower (about 3–4 orders of magnitude) than the corresponding occupational exposure limits (OELs), i.e. 3.2 mg/m3 , 188 mg/m3 , 442 mg/m3 and 221 mg/m3 , respectively. Airborne BTEX concentrations were higher at Day 2 compared to Day 1, but the differences were not statistically significant. When traffic policemen were classified according to their working task,
27
significantly higher concentrations of toluene and xylenes were observed for outdoor workers compared to indoor clerks (p < 0.05). No differences were detected between smokers and non-smokers (data not shown). The ambient concentrations of TEX aromatic hydrocarbons but benzene were correlated each other (data not shown). 3.2. Biological monitoring The distribution of biomarkers of exposure expressed as medians [and interquartile range] in the whole group of traffic policemen and in subgroups of workers classified according to the smoking habits is summarized in Table 3. S-PMA and U-TEX followed a lognormal distribution, whereas t,t-MA and U-B did not distributed normally or log-normally. All values are well below the biological limit values for occupationally exposed people and within the reference values for the general unexposed population. Smokers eliminated significantly higher concentrations of unchanged benzene and its metabolites (p < 0.0001) and other unchanged TEX than non-smokers (p < 0.05) as shown in Table 3. When traffic policemen were distinguished into indoor (n = 31) and outdoor workers (n = 69), no significant differences were observed for urinary biomarkers (data not shown). Significantly lower concentrations of S-PMA, U-B and U-T were determined in samples collected at Day 1 compared to Day 2, i.e. 0.24 [0.17–0.41] vs. 1.22 [0.90–1.92] g/g creatinine for SPMA (p < 0.0001), 0.15 [0.12–0.19] vs. 0.20 [0.16–0.36] g/l for U-B (p < 0.0001), and 0.19 [0.16–0.22] vs. 0.24 [0.17–0.30] g/l for U-T (p = 0.003). The differences were even more pronounced when only non-smoking subjects were considered, as illustrated in Fig. 2. In the narrow range examined in this study, airborne BTEX concentrations were not correlated with biological parameters. In the whole study group significant correlations were observed between urinary biomarkers (0.21 < < 0.75, p < 0.05) except between t,tMA and both U-EB and U-X. Table 4 shows Spearman’s correlation matrix among biological markers in subjects classified according to the smoking habits. Among non-smokers, significant correlations were observed between S-PMA and both t,t-MA and U-B (p < 0.01) and between U-TEX (p < 0.01). In the subgroup of smokers, benzene biomarkers were correlated with each other and with U-cotinine (p < 0.01). In addition, the self-reported number of cigarettes/day was highly correlated with U-cotinine ( = 0.59, p < 0.01), and a significant correlation was observed between the
Table 2 Distributions of BTEX airborne concentrations in the whole group of traffic policemen and in subgroups classified according to the working task as indoor and outdoor workers Compound 3
Benzene (g/m ) Toluene (g/m3 ) Ethylbenzene (g/m3 ) Xylenes (g/m3 )
All workers (n = 24)
Indoor workers (n = 5)
Outdoor workers (n = 19)
6.1 [0.3–9.5] 15.4 [4.0–28.3] 4.5 [3.1–7.7] 19.8 [13.6–30.1]
6.2 [1.8–7.4] 3.2 [3.1–8.0] 3.2 [2.5–4.5] 12.1 [10.3–17.0]
5.9 [0.3–12.8] 24.5 [10.0–32.8]* 4.8 [3.1–8.3] 22.4 [17.0–32.3]*
Airborne concentrations of toluene e xylene were significantly higher (p < 0.05) in outdoor traffic policemen compared to indoor workers. All BTEX values are well below the corresponding occupational exposure limits (OELs), e.g. 3.2 mg/m3 , 188 mg/m3 , 442 mg/m3 and 221 mg/m3 . *p < 0.01, Mann–Whitney U-test.
Table 3 Distributions of exposure biomarkers in the whole group of traffic policemen and in subgroups of workers classified according to the smoking habits Biomarker
All workers (n = 100)
Non-smokers (n = 80)
Smokers (n = 20)
U-Benzene (g/l) U-Toluene (g/l) U-Ethylbenzene (g/l) U-Xylenes (g/l) S-PMA (g/g creatinine) t,t-MA (g/g creatinine)
0.16 [0.13–0.26] 0.20 [0.16–0.26] 0.41 [0.32–0.49] 0.37 [0.29–0.45] 0.67 [0.23–1.32] 41.8 [34.1–89.7]
0.16 [0.13–0.19] 0.19 [0.16–0.22] 0.40 [0.32–0.47] 0.35 [0.28–0.44] 0.42 [0.20–1.07] 38.6 [31.7–51.6]
0.79 [0.24–1.92]** 0.27 [0.21–0.38]** 0.46 [0.38–0.61]* 0.43 [0.35–0.62]* 1.43 [0.92–2.13]** 124.7 [84.2–175]**
Values are expressed as median and interquartile range. All values are well below the biological limit values for occupationally exposed people and within the reference values for the general unexposed population. **p < 0.0001, *p < 0.05, Mann–Whitney U-test.
28
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
Fig. 2. Comparison between urinary concentrations of unchanged benzene (U-B) and toluene (U-T), and benzene metabolites S-PMA and t,t-MA in samples collected in different sampling days (Day 1 vs. Day 2) in subgroups of non-smoking (no. of subjects, 46 vs. 34) and smoking (6 vs. 14) traffic policemen. Data are shown as means ± S.E.M. Table 4 Spearman’s correlation coefficients between urinary biomarkers in subgroups of non-smoking (n = 80) and smoking (n = 20) traffic policemen t,t-MA Non-smokers S-PMA 0.242* t,t-MA U-B U-T U-EB U-X Smokers S-PMA t,t-MA U-B U-T U-EB U-X
0.538**
U-B
U-T
U-EB
U-X
U-cotinine
0.310** 0.233
0.121 0.031 0.602**
0.194 −0.012 0.098 0.327**
0.128 −0.027 0.085 0.309** 0.755**
0.215 0.064 −0.295* −0.169 0.064 0.104
0.755** 0.233
0.380 0.125 0.543*
−0.194 −0.128 −0.120 0.245
0.252 0.023 0.219 0.584** 0.588**
0.773** 0.656** 0.538** 0.268 0.017 0.277
**p < 0.01, *p < 0.05, Spearman’s correlation, two-tailed.
number of cigarettes/day and the concentration of S-PMA ( = 0.54, p < 0.05). Stepwise multiple regression models run to assess the role of smoking habits and other covariates on benzene biomarkers substantially confirmed the findings of univariate analyses. The best predictors for S-PMA were the sampling day (ˇ = 0.714, p < 0.0001), U-cotinine (ˇ = 0.368, p < 0.0001), and age (ˇ = −0.133, p = 0.025), the adjusted r2 of the model being 0.680 (p < 0.0001). U-Cotinine (ˇ = 0.652, p < 0.0001) together with the sampling day (ˇ = 0.207, p = 0.007) and the working place (ˇ = −0.203, p = 0.008) accounted for 52.7% of variance of U-B. Finally, only U-cotinine significantly contributed to the variability of t,t-MA (ˇ = 0.658, p < 0.0001, r2 = 0.427). 4. Discussion This study on Italian traffic policemen was conducted using the same integrated approach between ambient and biological monitoring applied in the case of taxi drivers of our town (Manini et al., 2006). Due to the larger number of enrolled subjects (n = 100), we performed personal ambient monitoring on 24 workers dis-
tributed among all work-shifts and representative of all working tasks. Airborne BTEX concentrations detected in this study were super imposable with those observed in the case of taxi drivers and were also comparable to those reported in other field studies, where Italian traffic policemen were monitored (Crebelli et al., 2001; Tomei et al., 2001; Maffei et al., 2005; Bono et al., 2005; Fustinoni et al., 2005). Airborne BTEX concentrations were lower (3–4 orders) than the occupational exposure limits set by the Italian law for the working environment (3200 g/m3 for benzene, DLgs 25/2) and benzene concentrations were close to the recommended air quality parameter of 5–10 g/m3 . Despite airborne concentrations have been significantly decreased in Italy during the past decade, benzene abatement represents a public health concern since it has been estimated that the risk of developing leukemia is about six cases per million among people who experience lifetime exposure to benzene concentrations of 1 g/m3 in air (WHO, 2000). Starting from this risk estimate, we can calculate the additional excess of risk due to occupational exposure in our study group. Assuming that traffic policemen are exposed for 6 h × 240 days/year × 40 years, the unit risk would become about 5.6 × 10−7 at an air benzene concentration of 1 g/m3 . An additional risk of 1.7 × 10−6 cases of leukemia can be estimated in our cohort of worker exposed to an excess of airborne benzene of 3.1 g/m3 as compared to the general population exposed to an annual mean benzene concentration of about 3 g/m3 . Significant differences between indoor and outdoor workers were observed only for toluene and xylene, which are the main aromatic components (<5%) of unleaded fuel after the reduction of benzene content (<1%). As already observed in our previous study on taxi drivers and in other studies (Crebelli et al., 2001; Sørensen et al., 2003), personal ambient monitoring was not useful to distinguish between smokers and non-smokers. Conversely, all benzene biomarkers including t,t-MA and urinary toluene were higher in smokers than in non-smokers and were correlated with both U-cotinine and with the number of cigarettes smoked the day of the sampling. Biomarkers reflect the whole absorbed dose of BTEX whereas ambient monitoring mirrors the uptake from the air only. This is why we did not observe any significant correlation between environmental and biological indices in the narrow range examined.
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
The main novelty of this study consists in the observation that the internal dose of benzene assessed by urinary biomarkers seems to be modified by a change in airborne benzene concentration even when it occurs at few g/m3 level. To the best of our knowledge, this is the first time that this finding has been reported. Therefore, it should not be over interpreted until a clear confirmation would come from further studies on larger populations. The trend of the concentrations of particulate matter 10 (PM10 ) and airborne benzene in the town of Parma measured in February 2005 by the municipal air monitoring stations is illustrated in Fig. 1. The first 2 weeks of the month were characterized by high concentrations of pollutants, particularly PM10 with eight values above the recommended threshold level of 50 g/m3 . Benzene concentrations were high in the first week and decreased in the second one up to the year’s mean value of 3 g/m3 . The sampling Days 1 and 2 (February 15 and 16) were preceded by two windy days that significantly lowered airborne concentrations of PM10 and further reduced airborne benzene levels to 1 g/m3 . Unexpectedly, benzene biomarkers S-PMA and U-B (but not t,t-MA) were significantly lower in workers sampled at Day 1 compared to those sampled at Day 2. When only the subgroup of non-smokers was considered, this difference was particularly marked for S-PMA (p < 0.0001) and was less evident for U-B (p < 0.001), as shown in Fig. 2. This different behaviour is probably due to the relatively long half-life (about 9 h) of S-PMA compared to that of U-B. Particularly, the concentration of S-PMA in urine samples collected at Day 1 was significantly lower than that we usually observe in non-smokers, indicating that nonsmoking traffic policemen have been exposed to reduced benzene concentrations in the hours before and during the work-shift. Conversely, when the subgroup of smokers is considered, none of the biomarkers was able to detect any difference related to work-shifts. This result supports the conclusion that benzene uptake arising from the smoking habits is considerably higher than that from air (Darrall et al., 1998), as already observed in the case of taxi drivers (Manini et al., 2006). In our previous study, benzene uptake arising from cigarette smoking modified background S-PMA and U-B values compared to non-smokers and resulted in a relative increase of S-PMA levels at the end of the shift compared to pre-shift values. In non-smoking subjects, S-PMA was not modified by prolonged exposure (about 11 h) to airborne benzene. For the general population exposed to few g/m3 , i.e. 2–3 g/m3 in the case of our town, and for workers doing their job in the urban environment, we hypothesize that benzene concentrations are at the steady-state in biological fluids and that metabolite excretion could not be modified by repeated exposures to similar daily concentrations. This study shows for the first time that a prolonged exposure (3 days) to reduced airborne benzene (1 g/m3 ) could modify benzene concentrations in biological fluids, leading to reduced excretion of unchanged benzene and its specific metabolite S-PMA. If confirmed on larger groups of subjects, this finding may encourage further benzene abatement in urban air, as there are exposure biomarkers which are enough sensitive and specific to assess benzene exposure and to mirror changes occurring even at few g/m3 in non-smoking subjects. The lack of performance of t,t-MA in distinguishing workers sampled at Day 1 from those sampled at Day 2 could be ascribed to the dietary intake of sorbic acid and sorbate, both widely used as preservatives in food. We tried to estimate sorbic acid intake by questionnaire as the number of ingested candies, without obtaining a significant contribution to the variance of t,t-MA. Actually, the sources for intake of sorbate are various and, besides candies, include bread and pastry, cheeses, yogurt, meat products, smoked and salted fish, fats and oils, salads, fermented and acidified vegetables, fruit products, dressings, jams, mayonnaise, wine and beverages. It has been demonstrated that about 0.23% of
29
ingested sorbic acid is metabolized to t,t-MA with a large interindividual variability (Renner et al., 1999). Assuming a daily intake of about 30 mg, it has been estimated that sorbic acid would account for at least 50% of t,t-MA background urinary excretion in nonoccupationally exposed non-smokers and 25% in smokers (Ruppert et al., 1997). The use of a more complex dietary questionnaire to assess sorbic acid intake or the simultaneous determination of sorbic acid in urine to correct t,t-MA levels (Renner et al., 1999; Negri et al., 2005) would be necessary, but it is out of the scope of the present investigation. The results of this study confirm that t,t-MA lacks the specificity needed for biomonitoring of low benzene exposure. 5. Conclusions Airborne BTEX concentrations were well below occupational exposure limits. Biomarkers of exposure felt within the reference intervals for the unexposed Italian population (BIOLIND.NET, 2008). Smoking habits was the main determinant and all biomarkers were inter-correlated with each other, despite the very low absorbed dose. A decrease of few g/m3 in benzene airborne levels, due to a change of the state of the atmosphere, resulted in significantly lower S-PMA and U-B concentrations in non-smoking traffic policemen. Finally, this study conducted on a larger group confirms our previous conclusion that S-PMA is the most sensitive and reliable benzene biomarker, probably due to its longer half-life and to its lower intra- and inter-individual variability as compared to U-B. Conflict of interest None. Acknowledgement The kind collaboration of the traffic policemen of the city of Parma is gratefully acknowledged. References Andreoli, R., Manini, P., Bergamaschi, E., Brustolin, A., Mutti, A., 1999. Solid-phase microextraction and gas chromatography–mass spectrometry for the determination of monoaromatic hydrocarbons in blood and urine: application to people exposed to air pollutants. Chromatographia 50, 167–172. ARPA (Agenzia Regionale Protezione Ambiente, Emilia Romagna), 2005. Available at http://www.arpa.emr.it/. BIOLIND.NET, 2008. Available at http://www.biolind.net. Bono, R., Traversi, D., Maestri, L., Schilirò, T., Ghittori, S., Baiocchi, C., Gilli, G., 2005. Urban air and tobacco smoke in benzene exposure in a cohort of traffic policemen. Chem. Biol. Interact. 153–154, 239–242. Boogard, P.J., van Sittert, N.J., 1995. Biological monitoring of exposure to benzene: a comparison between S-phenylmercapturic acid, trans,trans-muconic acid, and phenol. Occup. Environ. Med. 52, 611–620. Brugnone, F., Perbellini, L., Maranelli, G., Romeo, L., Lombardini, F., 1992. Reference values for blood benzene in occupationally unexposed general population. Int. Arch. Occup. Environ. Health 61, 513–518. Crebelli, R., Tomei, F., Zijno, A., Ghittori, S., Imbriani, M., Gamberale, D., Martini, A., Carere, A., 2001. Exposure to benzene in urban workers: environmental and biological monitoring of traffic police in Rome. Occup. Environ. Med. 58, 165–171. Darrall, K.G., Figgins, J.A., Brown, R.D., Phillips, G.F., 1998. Determination of benzene and associated volatile compounds in mainstream cigarette smoke. Analyst 123, 1095–1101. Fustinoni, S., Consonni, D., Campo, L., Buratti, M., Colombi, A., Pesatori, A.C., Bonzini, M., Bertazzi, P.A., Foà, V., Garte, S., Farmer, P.B., Levy, L.S., Pala, M., Valerio, F., Fontana, V., Desideri, A., Merlo, D., 2005. Monitoring low benzene exposure: comparative evaluation of urinary biomarkers, influence of cigarette smoking, and genetic polymorphisms. Cancer Epidemiol. Biomarkers Prev. 14, 2237–2244. Ghittori, S., Imbriani, M., Maestri, L., Capodaglio, E., Cavalleri, A., 1999. Determination of S-phenylmercapturic acid in urine as an indicator of exposure to benzene. Toxicol. Lett. 108, 329–334. Henry, R.J., 1974. Clinical Chemistry Principle and Techniques, second ed. Harper & Row, New York. IARC, 1982. Benzene. Some industrial chemicals and dyestuffs IARC monographs on the evaluation of carcinogenic risk to humans. 29, 93–148.
30
P. Manini et al. / Toxicology Letters 181 (2008) 25–30
Maffei, F., Hrelia, P., Angelini, S., Carbone, F., Cantelli Forti, G., Barbieri, A., Sanguinetti, G., Mattioli, S., Violante, F.S., 2005. Effects of environmental benzene: micronucleus frequencies and hematological values in traffic police working in an urban area. Mutat. Res. 583, 1–11. Manini, P., De Palma, G., Andreoli, R., Poli, D., Mozzoni, P., Folesani, G., Mutti, A., Apostoli, P., 2006. Environmental and biological monitoring of benzene exposure in a cohort of Italian taxi drivers. Tox. Lett. 167, 142–151. Melikian, A.A., O’Connor, R., Prahalad, A.K., Hu, P., Li, H., Kagan, M., Thompson, S., 1999. Determination of the urinary benzene metabolites S-phenylmercapturic acid and trans,trans-muconic acid by liquid chromatography–tandem mass spectrometry. Carcinogenesis 20, 719–726. Melikian, A.A., Qu, Q., Shore, R., Li, G., Li, H., Jin, X., Cohen, B., Chen, L., Li, Y., Yin, S., Mu, R., Zhang, X., Wang, Y., 2002. Personal exposure to different levels of benzene and its relationships to the urinary metabolites S-phenylmercapturic acid and trans,trans-muconic acid. J. Chromatogr. A 778, 211–221. Negri, S., Bono, R., Maestri, L., Ghittori, S., Imbriani, M., 2005. High-pressure liquid chromatographic-mass spectrometric determination of sorbic acid in urine: verification of formation of trans,trans-muconic acid. Chem. Biol. Interact. 153–154, 243–246. Qu, Q., Shore, R., Li, G., Jin, X., Chen, L.C., Cohen, B., Melikian, A.A., Eastmond, D., Rappaport, S., Rupa, D., Waidyanatha, S., Yin, S., Yan, H., Meng, M., Winnik, W., Kwok, E.S., Li, Y., Mu, R., Xu, B., Zhang, X., Li, K., 2003. Validation and evaluation
of biomarkers in workers exposed to benzene in China. Res. Resp. Health Eff. Inst. 115, 1–87. Renner, T., Baer-Koetzle, M., Scherer, G., 1999. Determination of sorbic acid in urine by gas chromatography–mass spectrometry. J. Chromatogr. A 847, 127–133. Ruppert, T., Scherer, G., Tricker, A.R., Adlkofer, F., 1997. trans,trans-Muconic acid as a biomarker of non-occupational environmental exposure to benzene. Int. Arch. Occup. Environ. Health 69, 247–251. Sørensen, M., Skov, H., Autrup, H., Hertel, O., Loft, S., 2003. Urban benzene exposure and oxidative DNA damage: influence of genetic polymorphisms in metabolism genes. Sci. Total Environ. 309, 69–80. Tomei, F., Ghittori, S., Imbriani, M., Pavanello, S., Carere, A., Marcon, F., Martini, A., Baccolo, T.P., Tomao, E., Zijno, A., Crebelli, R., 2001. Environmental and biological monitoring of traffic warderns from the city of Rome. Occup. Med. 51, 198–203. Waidyanatha, S., Rothman, N., Fustinoni, S., Smith, M.T., Hayes, R.B., Bechtold, W., Dosemeci, M., Guilan, L., Yin, S., Rappaport, S.M., 2001. Urinary benzene as a biomarker of exposure among occupationally exposed and unexposed subjects. Carcinogenesis 22, 279–286. WHO, 1993. WHO Environmental health criteria 155. Biomarkers and risk assessment, concepts and principles. ICPS, Geneva. WHO Regional Office for Europe, 2000. Air Quality Guidelines for Europe, second ed. Who Regional Publications, pp. 62–66, European series, no. 91.