Environ Sci Pollut Res DOI 10.1007/s11356-015-5918-0
RESEARCH ARTICLE
Urinary 1-hydroxypyrene concentration as an exposure biomarker to polycyclic aromatic hydrocarbons (PAHs) in Mexican women from different hot spot scenarios and health risk assessment Lucia G. Pruneda-Álvarez 1,2 & Francisco J. Pérez-Vázquez 1,2 & Tania Ruíz-Vera 1,2 & Ángeles C. Ochoa-Martínez 1,2 & Sandra T. Orta-García 1,2 & Jorge A. Jiménez-Avalos 1,2 & Iván N. Pérez-Maldonado 1,2,3
Received: 7 September 2015 / Accepted: 2 December 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Recently, in developing countries, polycyclic aromatic hydrocarbons (PAHs) have been considered contaminants of grave concern for women and children. Therefore, the aim of this study was twofold: (1) evaluate exposure assessment to PAHs using urinary 1-hydroxypyrene (1-OHP) as an exposure biomarker and (2) perform a health risk assessment in women from four different high risk scenarios in Mexico. From 2012 to 2013, in a cross-sectional study, we evaluated a total of 184 healthy women from the following scenarios: (A) indoor biomass combustion site (n=50); (B) brick manufacturing site using different materials such as fuel sources (n=70); (C) industrial site (n=44); and (D) high vehicular traffic site (n=20). 1-hydroxypyrene (1-OHP) was quantified using a high-performance liquid chromatography (HPLC) technique. Afterward, a probabilistic health risk
assessment was performed (Monte Carlo analysis). Mean urinary 1-OHP levels found were 0.92±0.92; 0.91±0.83; 0.22± 0.19; and 0.14±0.17 μg/L for scenario A, B, C, and D, respectively. Then, based on the measured urinary 1-OHP levels, the estimated median daily intake doses of pyrene were calculated: 659, 623, 162, and 77.4 ng/kg/day for the women participating in the study living in areas A, B, C, and D, respectively, and finally, the hazard quotient (HQ) was calculated (22±21, 21±20, 5.5±5.5, and 2.6±3.5; for areas A, B, C, and D, respectively), high health risk was noted for the women living in the studied communities. The data shown in this study (exposure levels to PAHs and health risk assessment) made it reasonable to conclude that the exposure levels found have a significant potential for generating adverse effects on human health in the studied scenarios.
Responsible editor: Philippe Garrigues
Keywords 1-OHP . Biomonitoring . Health risk assessment . Indoor air pollution . PAHs . Women
Electronic supplementary material The online version of this article (doi:10.1007/s11356-015-5918-0) contains supplementary material, which is available to authorized users. * Iván N. Pérez-Maldonado
[email protected] 1
Laboratorio de Toxicología Molecular, Centro de Investigación Aplicada en Ambiente y Salud (CIAAS), Coordinación para la Innovación y Aplicación de la Ciencia y la Tecnología (CIACYT), Universidad Autónoma de San Luis Potosí, San Luis Potosí, Mexico
2
Facultad de Medicina, Universidad Autónoma de San Luis Potosí, San Luis Potosí, Mexico
3
Unidad Académica Multidisciplinaria Zona Media, Universidad Autónoma de San Luis Potosí. Rioverde, Avenida Sierra Leona No. 550, Colonia Lomas Segunda Sección, San Luis Potosí 78210, SLP, Mexico
Introduction Air pollution is a crucial environmental and public health concern in developing countries, as it is known as a significant source of several toxic chemical contaminants including polycyclic aromatic hydrocarbons (PAHs). Increased incidence of lung, skin, bladder, and gastrointestinal cancers is associated with chronic exposure to PAHs (Boffetta et al. 1997; Bach et al. 2003; Olsson et al. 2010; Diggs et al. 2011). Moreover, recent studies have demonstrated an association between PAHs exposure and several human diseases such as diabetes, metabolic syndrome, cardiovascular disease, hypertension, among others (Brucker et al. 2013; Brucker et al. 2014; Feng
Environ Sci Pollut Res
et al. 2014; Yang et al. 2014; Hu et al. 2015; Ruiz-Vera et al. 2015a). PAHs are ubiquitous environmental contaminants (that are either man-made or come about naturally) containing two or more conjoined aromatic rings (Jongeneelen 2014; Bansal and Kim 2015; Ma and Harrad 2015). PAHs are usually released as a result of vehicle exhaust, industrial emissions, residential biomass combustion, and tobacco smoke (Rodgman et al. 2000; Chuang and Chang 2007; Bansal and Kim 2015; Ma and Harrad 2015). Therefore, PAHs are widely distributed in the environment and are not naturally degraded easily (Bansal and Kim 2015; Ma and Harrad 2015). Because of their potential to bio-accumulate, their persistence, and toxic effects on human health, PAHs have been of scientific interest for many years and have raised public health concern (Torres-Dosal et al. 2008; Olsson et al. 2010; Diggs et al. 2011; Alegría-Torres et al. 2013; Kim et al. 2013; Rota et al. 2014; Jasso-Pineda et al. 2015; Ruiz-Vera et al. 2015a). In this context, several studies have shown that urinary PAH metabolites (such as 1hydroxypyrene) are good exposure biomarkers to assess recent exposure to PAHs (Li et al. 2008; Li et al. 2015; Alghamdi et al. 2015). For this reason, 1-hydroxypyrene (1OHP) has been used for the assessment of exposure of these compounds (PAHs) in human populations (Barbeau et al. 2015; Kho et al. 2015; Wang et al. 2015). For instance, elevated concentrations of urinary 1-OHP have been found in people living in scenarios with a high risk of contamination by PAHs (high traffic, indoor biomass combustion, industrial sites, among others) (Viau et al. 2000; Nguyen et al. 2014; Choosong et al. 2014; Kamal et al. 2015). In Mexico, potential contamination sources of PAHs exist in different areas, such as those associated with petrochemical industry, agriculture, biomass combustion, major industry, small-scale industry (like brickyards), oil fields, and noncontrolled waste disposal sites, among others. However, studies done in Mexico regarding population exposure to those chemicals in all the scenarios mentioned above are scarce (Pruneda-Álvarez et al. 2012; Perez-Maldonado et al. 2013; Pérez-Maldonado et al. 2014; Ruiz-Vera et al. 2015b; RuizVera et al. 2015a). On the other hand, risk assessment has emerged in recent years as a powerful tool in the analysis of environmental and/ or occupational hazards (Nieuwenhuijsen et al. 2006). According with the Environmental Protection Agency (EPA), the utility of risk assessment is to characterize the nature and magnitude of health risks to humans from chemical contaminants and other stressors that may be present in the environment. Therefore, the aim of this study was twofold: (1) evaluate exposure assessment to PAHs using 1-OHP as an exposure biomarker and (2) perform a health risk assessment in women from four high risk scenarios in Mexico.
Methods Population Sampling sites were selected considering previous knowledge of the activity in each place. Sites included in the study are recognized for their industrial activity, indoor wood combustion, high vehicular traffic, and brick manufacturing using different materials as fuel sources (Fig. 1 and Table 1). From 2012 to 2013, in a cross-sectional study, we evaluated a total of 184 healthy women from the following scenarios: (A) indoor biomass combustion site (n=50); (B) brick manufacturing site using different materials as fuel sources (n=70); (C) industrial site (n=44); and (D) high vehicular traffic site (n= 20) (Table 1). Only female volunteers (aged 22–63 years), living at the sites, were screened to determine whether they were eligible to take part in the study through personal interview, and they were chosen randomly. We identified 221 eligible women and invited them to participate in this work. Of these women, 184 (83 %) agreed to participate. The inclusion criteria were apparently healthy adult women, non-smokers (only women reported to be unexposed according to survey results were included in the study), non pregnant, non-lactating, and non-morbid (not on medication of any kind, not suffering from any kind of disease). Then, after an informed consent was obtained, a questionnaire was completed and urine samples were taken. The questionnaire registered characteristics such as age, weight, height, exposure to smoke tobacco, and type of stove used in their home. Also, with the questionnaire, we determined household and sociodemographic characteristics, occupation of family members, among others. The study was approved by the ethical committee of the School of Medicine, at the Autonomous University of San Luis Potosi. Moreover, the body mass index (BMI) was calculated according to the weight and height recorded of each woman participating in the work. The BMI values were used to classify them as underweight, healthy weight, overweight, and obese according to sex-specific-BMI-for-age growth charts generated by the World Health Organization (WHO) (WHO 2015). Urine collection Urine was collected from each woman in the morning (first morning void upon waking, approximately 7:00 h). The first morning urine was collected in order to have an agreement on all urinary biomarkers assessed (including 1-OHP and creatinine), taking into account that the sample at that time of day is the most concentrated one (Han et al. 2008). Urine samples were collected in sealable polypropylene bottles and kept in a cooler with ice packs during transportation to the laboratory. Once at the laboratory, the samples were stored in a deep
Environ Sci Pollut Res
Fig. 1 Location of studied communities
freezer at −20 °C until analysis. The collected volume of urine sample was 50–100 mL on average, and as a preservative, we added HCl 6 M (100 μL per 10 mL of urine). Prior to their analysis, samples were thawed at room temperature, homogenized, and 10 mL of urine was transferred to a test tube.
Levels of urinary 1-OHP, which has a short biological half-life (18.6 h), was taken as a representative biomarker of exposure in populations exposed to PAHs mixtures (Jongeneelen 2001; Jacob and Seidel 2002), taking into account that this compound is a pyrene metabolite, and in its turn, pyrene is often present in PAHs mixtures. 1-OHP was quantified following the method described by Jongeneelen et al. (1987) and
Kuusimäki et al. (2004). Each sample (10 mL) was mixed with sodium acetate buffer (10 mL, 0.2 mol/L, pH 5.0), then 30 μL β-glucuronidase/arylsulfatase enzyme was added, and the sample was then incubated at 37 °C for 12 h. The analyte was extracted by solid phase using C-18 cartridges eluted in methanol containing 1 % acetic acid. The material was concentrated with nitrogen current to 1 mL. The concentrated material was filtered through a polyvinylidene fluoride filter (13 mm, 0.45 μm, MillexDurapore, Millipore, Bedford, MA, USA), and an aliquot was transferred to silanized vials (Jongeneelen et al. 1987; Kuusimäki et al. 2004). Then, the analysis was performed by HPLC (HP1100, Agilent Technologies) using a fluorescence detector (G1321A). The precolumn was Zorbax SB-C18, and the column was a Zorbax Eclipse XDB-C18. The column temperature was set to 40 °C,
Table 1 Characteristics of sampled sites
n
Characteristics Indigenous agricultural community in the southern of San Luis Potosi state that use biomass combustion as the principal energy source. 21° 16′ 00″ N; 98° 47′ 00″ W Rural community with brickyards as the principal economic activity in Guanajuato state, different materials (tires, wood, clothes, among others) are used as fuel source. 20° 33′ 54″ N; 100° 51′ 48″ W Industrial community (main petro-chemical industry in México) located in Veracruz state 18° 08′ 56″ N; 94° 24′ 41″ W Urban community in the state of San Luis Potosí with heavy vehicular traffic. 22° 08′ 59″ N; 100° 58′ 30″ W
Determination of urinary 1-hydroxypyrene
Scenarios Biomass combustion
A
50
Brickyard industry
B
70
Petrochemistry industry Urban area
C
44
D
20
Environ Sci Pollut Res
the flow was adjusted to 1 mL/min, and the injection volume was 20 μL. The eluent was 88:12 methanol/water and 1 % ascorbic acid. Data were collected and processed using HP ChemStation software. Urinary 1-OHP concentrations were adjusted to urinary creatinine (Supplementary file 1). Urinary creatinine was determined using the Jaffe colorimetric method (Taussky 1954). Under our conditions (Supplementary file 1), the method detection limits were 1.0 nmol/L. Quality control was certified using the standards, IRIS Clin Cal Recipe (Munich, Germany) 50013, 8867, and 50014 (9.1, 15.6, and 32.5 nmol/L 1-OHP), and there was a recovery of 95 %. Health risk assessment Estimated daily intake doses of pyrene for adult women were calculated using the following equation (Lakind and Naiman 2008). D¼
CP WF
where D is the daily intake dose of pyrene (ng/kg-day), C is the urinary 1-OHP concentration (ng/mL) obtained in this study (for each studied scenario), P is the expected urinary output (mL/day), W is the body weight (kg; mean body weight in each studied scenario), and F is the fraction of the ingested dose of pyrene eliminated as 1-OHP in urine. We used a 24-h urinary output value (i.e., P; 576–2124 mL/day, even distribution) for adult females (Brendle 2007; Ruiz-Vera et al. 2015b), since urine volumes were not recorded in this study. Research has shown that humans are exposed to pyrene mainly through dietary ingestion in non-occupational settings (Li et al. 2010). Therefore, we assumed that 4 % of the ingested dose of pyrene was eliminated as 1-OHP in urine based on a human volunteer study (Viau et al. 2000). To avoid an overestimation of the daily intake dose of pyrene and its risks, calculations were carried out in a probabilistic framework. Probability density functions for each parameter (C, P, and W), rather than deterministic point values, were used, and a Monte Carlo analysis was performed. Monte Carlo analysis enables probabilistic approximations based on stochastic outcomes. In contrast to deterministic calculations that do not take into account the variability in measurements and the heterogeneity in a population or exposure parameter, this approach (probabilistic) incorporates the variation that naturally arises due to physiological reasons (e.g. differences in body weight and inhalation/ingestion rates) or spatiotemporal fluctuations in a contaminant concentration across a region. Instead of a single value, a distribution function is used from which data is selected randomly. Modeling of D was carried out with a Monte Carlo simulation which was applied by executing Crystal Ball 4.0® software with an iteration size of 100,000 to establish statistics
(probability, cumulative probability, mean, median, standard deviation, and percentiles). After using the modeled exposure information (D), a hazard quotient (HQ) was calculated in a second round of Monte Carlo simulation. HQ is the ratio of exposure to a certain substance of concern to its corresponding reference dose (RfD). HQ ¼
D RfD
where D is the modeled daily intake dose of pyrene (first Monte Carlo round) and RfD the reference dose for this chemical compound (30 μg/kg/day). The RfD is an estimate (within an order of magnitude) of the daily dose of a substance, a lifetime exposure which would not likely result in a harmful effect (Barnes and Dourson 1988). If the D is lower than the RfD (HQ1), it is likely that the factor in question will induce a morbid response (Usepa 1993). Oral RfD for pyrene was based on non-carcinogenic effects (i.e., renal toxicity) occurring in a subchronic oral bioassay in mice (EPA-IRIS 1993). Statistics The first thing shown in the data was that the exposure biomarker used in this study (1-OHP) was not normally distributed (Shapiro-Wilk test). Afterward, descriptive statistics (median (PC50), mean, standard deviation, minimum, maximum, and percentiles) were calculated for sociodemographic variables and 1-OHP levels. Taking into account that data were not normally distributed, all statistical elaborations were performed using non-parametric methods for 1-OHP data (Kruskal-Wallis one-way analysis of variance, followed by the Mann–Whitney test. Additionally, in order to evaluate the association between the biomarker (1-OHP) and variables assessed such as age, gender, weight, height, BMI, and among others, a spearman correlation test for non-parametric data was performed. The significance level was fixed at 0.05. For all the statistical analysis, we used GraphPad Prism © 5.01 version (GraphPad Software, Inc., CA, USA).
Results and discussion Sociodemographic characteristics The characteristics of the women in each study area are shown in Table 2. The ages ranged from 22 to 63 years. Regarding BMI, an important percentage of women participating in the study was classified as overweight and obese (between 35 and 65 %) in all studied communities (Table 2), and a slight percentage (5 %) was classified as underweight (Table 2). The
Environ Sci Pollut Res Table 2
General characteristics of study population
Variable
A
Scenario B
C
D
n
50
70
44
20
Age (years) min–max
29 (23–58)
38 (26–63)
32 (24–44)
28 (22–55)
Type of used fuel Socioeconomic statusa
Wood Extreme povertyb
Gas LP/wood Moderate poverty
Gas LP Moderate poverty
Gas LP No poverty
Body mass index (kg/m2) Mean (SD)
25.2 (3.5)
26.2 (4.7)
27.3 (4.2)
24.7 (3.7)
Weight status according to BMI (%) Underweight
0
0
0
5
Healthy weight Overweight
48 33
40 35
35 45
60 15
Obese
19
25
20
20
a
(CONEVAL 2015)
b
Extreme poverty widely refers to earning below the international poverty line of $1.25/day (in 2005 prices), set by the World Bank (World Bank 2014)
main source of fuel used in community A was biomass combustion, with people living in conditions of extreme poverty according to the rates of social development obtained from the National Council for Evaluation of Social Development Policy in Mexico (CONEVAL) and the World Bank (CONEVAL 2015), in communities C (moderate poverty) and D (no poverty and no vulnerable) they used Gas LP as an energy source to cook, and in community B (moderate poverty) they used a combination of both (biomass combustion and Gas LP). Only non-smoker (active or passive smoker) women were recruited to participate in this study, and none had occupational exposure to PAHs. It is important to note that when the association analysis was performed, no significant effects of urinary 1OHP (μg/L) levels were found for assessed variables (such as age, sex, height, or BMI) (p>0.05). Urinary 1-OHP levels World Health Organization (WHO) recommends creatinine concentration within the range of 30–300 mg/dL to be considered a representative sample, i.e., not too diluted or too concentrated (WHO 1996). In our research, the creatinine Table 3 Urinary 1-OHP levels in women (μg/L)
concentrations ranged from 32 to 286 mg/dL, and all samples were included in the statistical evaluation of the data. Tables 3 and 4 show the urinary concentrations of 1hydroxypyrene in women assessed, mean 1-OHP levels were 0.92±0.92; 0.91±0.83; 0.22±0.19; and 0.14±0.17 μg/L for scenarios A, B, C, and D, respectively. As is noted, the women with the lowest (p0.05) difference was found between urinary 1-OHP levels from scenarios A and B (Table 3). Similar results were found when the levels were expressed in micromole per mole creatinine (Table 4), significantly (p
