Assessment of estrogenic activity in PM10 air samples with the ERE-CALUX bioassay: Method optimization and implementation at an urban location in Flanders (Belgium)

Assessment of estrogenic activity in PM10 air samples with the ERE-CALUX bioassay: Method optimization and implementation at an urban location in Flanders (Belgium)

Chemosphere 144 (2016) 392e398 Contents lists available at ScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere Assessme...

720KB Sizes 0 Downloads 30 Views

Chemosphere 144 (2016) 392e398

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Assessment of estrogenic activity in PM10 air samples with the ERE-CALUX bioassay: Method optimization and implementation at an urban location in Flanders (Belgium) Kim Croes a, *, Pieterjan Debaillie a, b, Bo Van den Bril c, Jeroen Staelens c, Tara Vandermarken a, Kersten Van Langenhove a, Michael S. Denison d, Martine Leermakers a, Marc Elskens a a

Department of Analytical, Environmental and Geo-Chemistry (AMGC), Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussels, Belgium Department of Chemistry, Ghent University, Krijgslaan, 9000 Ghent, Belgium c Unit Air, Flemish Environment Agency (VMM), Kronenburgstraat 45, 2000 Antwerp, Belgium d Department of Environmental Toxicology, University of California-Davis, Davis, CA 95616, USA b

h i g h l i g h t s  We measured the overall estrogenic activity of PM10 in Flanders.  The human ovarian BG1Luc4E2 bioassay (ERE-CALUX) was used.  Estrogenic activity was low, but quantifiable in more than 70% of the samples.  Acetone/hexane extracts more cytotoxic compounds than ethanol or acetonitrile.  Use of a more polar extraction solvent is advisable.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 11 March 2015 Received in revised form 17 August 2015 Accepted 6 September 2015 Available online xxx

Endocrine disrupting chemicals represent a broad class of compounds, are widespread in the environment and can pose severe health effects. The objective of this study was to investigate the overall estrogen activating potential of PM10 air samples at an urban location with high traffic incidence in Flanders, using a human in vitro cell bioassay. PM10 samples (n ¼ 36) were collected on glass fiber filters every six days between April 2013 and January 2014 using a high-volume sampler. Extraction was executed with a hexane/acetone mixture before analysis using a recombinant estrogen-responsive human ovarian carcinoma (BG1Luc4E2) cell line. In addition, several samples and procedural blanks were extracted with ultra-pure ethanol or acetonitrile to compare extraction efficiencies. Results were expressed as bioanalytical equivalents (BEQs) in femtogram 17b-estradiol equivalent (fg E2-Eq) per cubic meter of air. High fluctuations in estrogenic activity were observed during the entire sampling period, with mean and median BEQs of 50.7 and 35.9 fg E2-Eq m3, respectively. Estrogenic activity was measured in more than 70% of the samples and several sample extracts showed both high BEQs and high cytotoxicity, which could not be related to black carbon, PM10 or heavy metal concentrations. At this moment, it remains unclear which substances cause this toxicity, but comparison of results obtained with different extraction solvents indicated that acetone/hexane extracts contained more compounds that were cytotoxic and suppressive of responses than those extracted using ultra-pure ethanol. Although more research is needed, the use of a more polar extraction solvent seems to be advisable. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Estrogen PM10 CALUX EDC Belgium

1. Introduction * Corresponding author. E-mail address: [email protected] (K. Croes). http://dx.doi.org/10.1016/j.chemosphere.2015.09.020 0045-6535/© 2015 Elsevier Ltd. All rights reserved.

Exogenous substances that act like hormones and disrupt the

K. Croes et al. / Chemosphere 144 (2016) 392e398

physiologic function of endogenous hormones in the endocrine system are called endocrine disruptors (EDs) or endocrine disrupting compounds (EDCs). They exert their effects by binding to hormone receptors, disturbing cell-signaling pathways, directly affecting the central neuro-endocrine system, inhibiting hormone synthesis and/or by their toxic effects on endocrine responsive organs. Continuous exposure to estrogens can pose severe health risks such as increased risk for breast cancer in adulthood, male and female infertility and precocious puberty (Krstevska-Konstantinova et al., 2001; Diamanti-Kandarakis et al., 2009; Ozen et al., 2012). Due to these health effects, there is growing attention on the presence of EDCs in our environment and our exposure to such chemicals. However, since EDCs represent a broad class of molecules such as organochlorine pesticides, industrial chemicals, fuels, plastics and plasticizers and numerous other chemicals, they are essentially ubiquitous. EDCs have been found globally in waste and surface water (Van der Linden et al., 2008; Belhaj et al., 2015), soil and sediment (Houtman et al., 2006), indoor dust and air samples (Kennedy et al., 2010; Vandermarken et al., 2015) and consumer products (Mertl et al., 2014). In the recent years, most studies on EDC pollution have focused on the aquatic environment as part of the Water Framework Directive (WFD 2000/60/EC), which aims at establishing environmental quality standards (EQS) to limit the concentrations of certain chemical substances that pose a significant risk to the environment or to human health, in surface and ground waters. Research on indoor or outdoor air pollution is still more limited and in most cases only a few, well known pollutants (like e.g. PAHs or bisphenol A) are quantified instead of a global endocrine activity. In this study, a novel methodology to measure the overall response of estrogen active compounds in PM10 air samples was developed and the effect of the extraction solvent on estrogenic potency was investigated. To get a better understanding of the effect of mixtures of environmental EDCs on the hormone system, a bioassay technique was used to quantify the potency of the mixture of estrogenic compounds present in PM10 air samples. A case study in an urban area with a high traffic incidence was implemented, since in this type of environment numerous estrogenic compounds (such as PAHs and their metabolites and phenols), but also anti-estrogenic (e.g. PCB 138, PCB 153 and PCB 180 and certain dioxins) and/or (anti) androgenic pollutants can be present in the atmosphere (Rogers and Denison, 2002; Michałowicz and Duda, 2007; Wenger et al., 2009; Emeville et al., 2013; Nov ak et al., 2013). 2. Materials and methods 2.1. Chemicals and standards Dimethylsulfoxide (DMSO) (minimum 99.7%), hexane (minimum 96%, suitable for CALUX), acetonitrile (minimum 99.99%) and acetone (minimum 99.95%) were purchased from Biosolve (The Netherlands). The estrogen standard 17b-estradiol (minimum 98%), ultra-pure ethanol, Dulbecco's Modification of Eagle's Medium high glucose (DMEM without phenol red) and sodium pyruvate (100 mM, sterile-filtered) were obtained from SigmaeAldrich (Belgium). Minimum Essential Medium Alpha (a-MEM), penicillinstreptomycin, Fetal Bovine Serum (FBS), FBS charcoal-stripped, Lglutamine (200 mM), trypsine, trypsine without phenol red and Phosphate-Buffered Saline (PBS) were obtained from Life Science Technology (United Kingdom). Glass fiber filters MG 227/1/60 with a diameter of 150 mm were purchased from Sartorius (Belgium). Luciferine reagent and lysis reagent were obtained from Promega (The Netherlands). The recombinant, estrogen-responsive BG1Luc4E2 ovarian carcinoma cell line was previously described (Rogers and Denison, 2000).

393

2.2. Sample collection Between April 2013 and January 2014, the Flemish Environment Agency (VMM) collected 36 particulate matter samples (PM10) at an urban location (Borgerhout) in Flanders (Belgium) (Supplementary Table S1). The sampling site (stations R801 and R802, X ¼ 154407, Y ¼ 211080) is located near the ring road of the city of Antwerp and this location is representative for an urban background site with meaningful traffic influence. The PM10 samples were collected during a 24-h period with a high-volume sampler (DHA80-SEQ with 16 filter standard field housing, Digitel) with a theoretical flow of 30 m3 h1 (720 m3 in a 24-h sampling period). The sampler was equipped with a PM10 size-selective inlet DPM10/30/00 PM10 pre-separator (Digitel) and samples were taken every six days on pre-heated (4 h at 450  C) glass fiber filters with a diameter of 150 mm (MG 227/1/60 Sartorius). During the sampling period, also eight field blanks and six procedural (laboratory) blanks were included. 2.3. Extraction and CALUX measurement Sample extraction was carried out with an Accelerated Solvent Extractor (ASE 200, Dionex Sunnyvale, CA, USA) equipped with 33 mL stainless-steel extraction cells, using conditions described in Table 1. Hexane/acetone (50/50%, v/v) was used as an extraction solvent, since this solvent mixture was also used for assessment of genotoxicity with the Ames test (Van Den Heuvel et al., 2014). After extraction, all samples were evaporated at 40  C under a flow of pure air and redissolved in 10 mL hexane for storage until in vitro analysis. In addition, five additional samples and four procedural blanks were extracted with either ultra-pure ethanol or acetonitrile to compare extraction efficiencies. Therefore, each sampled filter was divided in four equal parts and each part of the filter was extracted with a specific solvent (ethanol, acetonitrile or hexane/acetone). The extracts were then redissolved in 5 mL hexane and eight sample aliquots ranging between a dilution factor (df) of 10 (18 m3 air equivalent) and 400 (0.45 m3 air equivalent) were used for in vitro analysis. The ERE-CALUX (Estrogen Responsive Element - Chemical Activated LUciferase gene eXpression) assay is a reporter gene mammalian cell bioassay. In this study, CALUX measurements were executed using a recombinant BG1Luc4E2 human ovarian carcinoma cell line that contains a stably transfected estrogenresponsive gene plasmid (pGudLuc7.0) that responds to estrogenic chemicals with the induction of firefly luciferase reporter gene expression (Rogers and Denison, 2000). 17b-Estradiol (E2) was used as a reference standard and cell treatment and luciferase measurement were done as described previously (Vandermarken et al., 2015). Briefly, cells were cultured in a-MEM medium supplemented with 1% penicillin/streptomycin and 10% fetal bovine serum (FBS). At 24e48 h before each measurement, the cells were transferred to DMEM (high glucose) without phenol red,

Table 1 Specifications of the accelerated solvent extraction (ASE). ASE conditions

Extraction with hexane/acetone

Number of heating cycles Heating period Static period Flushing volume Purging time Pressure Temperature

2 6 min 5 min 60% 60 s 1500 psi 100 C

394

K. Croes et al. / Chemosphere 144 (2016) 392e398

supplemented with 1% penicillin/streptomycin, 4.5% charcoal treated fetal calf serum, 2% L-glutamine and 1% sodium pyruvate. After trypsinizing, the cells were counted and diluted to a concentration of 2  105 cells per mL. Each well on the 96-well plate was seeded with 200 mL cell suspension in DMEM. After 24 h incubation (37  C, 5% CO2), the medium was removed and 190 mL of a standard solution or sample extract in DMEM with 1% DMSO (in duplicate) was added to each well on the plate. After 19e22 h incubation, the medium was again removed and the wells were rinsed with 75 mL PBS buffer pH 7.4 and cells in each well of the plate were visually inspected under a microscope in order to evaluate confluence of the cells, cell morphology and occurrence of cell death. Then, 50 mL lysis reagent was added and the plate was shaken for 5 min. After a 10 min incubation period in the luminometer (Glomax, Promega, USA), 50 mL luciferine reagent was injected and the light output was given in Relative Light Units (RLUs) (integration time 5 s, lag time 6 s). 2.4. Heavy metal analysis To investigate if heavy metals were present in the ERE-CALUX extracts, the concentrations of lead, zinc, copper, manganese, nickel, arsenic, cadmium and chromium were determined in six additional the hexane/acetone sample extracts and in one field blank extract. Therefore, 0.5 mL of the extract was mixed with an equal volume of concentrated nitric acid and this mixture was diluted to 10 mL with MilliQ water. The samples were measured by Inductively Coupled Plasma Sector Field Mass Spectrometry (ICPSFMS, Thermo Scientific Element II) using external standard calibration prepared from a multi-element standard solution (Merck XIII) and In (1 ppb) as internal standard. 2.5. PM10 and black carbon mass concentration The mass concentrations of PM10 and black carbon (BC) in ambient air were determined by the Flemish Environment Agency during the same sampling period and at the same location (Borgerhout, Flanders). PM10 was measured according to the European reference method (EN12341), using a low-volume sampler (Leckel SEQ 47/50; 2.3 m3 h1 flow rate) with a size-selective inlet. PM10 was sampled on quartz fiber filters (Pall Tissuquartz 2500 QAT-UP, 47 mm diameter) during 24 h. The PM10 mass was determined with a Sartorius MP5 balance at 20  C and 50% relative humidity by weighing twice before and after the sampling. Black carbon was measured with a multi-angle absorption photometer (MAAP Thermo Scientific 5012; 1 m3 h1 flow rate) and a total suspended particles (TSP) inlet. The BC content of the sampled aerosol was determined by simultaneously measuring the optical absorption and scattering of light (670 nm) by particles collected on a glass fiber filter tape (GF10), in combination with a constant absorption coefficient (Petzold et al., 2002). BC is formed by incomplete combustion of fossil fuels, biofuels and biomass and is the major component of diesel exhaust particles. It is a better indicator of harmful particulate substances from combustion sources (especially traffic) than undifferentiated PM mass (Janssen et al., 2012). 2.6. Statistical data treatment Estrogenic activity from the ERE-CALUX bioassay was reported as bioanalytical equivalents (BEQs) and expressed in femtograms 17b-estradiol equivalent per cubic meter of air. 17b-Estradiol (E2) was used as a reference standard. A four parameter Hill-function and a BoxeCox transformation were used to characterize the estrogenic response of the E2 standard solutions, while the measured luminescence (expressed in Relative Light Units or RLUs) of an

unknown sample was converted into a BEQ by comparison of the slope of the sample doseeresponse curve with the slope of the standard curve (Elskens et al., 2011; Mihale et al., 2013; Mosteller and Tukey, 1977). The Hill function could in the latter case not be used, since the most concentrated sample extracts were often toxic to the cells and thus no full concentrationeresponse curve (with upper plateau up to 100% RLU induction) could be obtained. Accordingly, the BoxeCox transformation with calculation of the slopes could provide then a reliable alternative for the BEQ determination. The Inverse Prediction (IP) method was additionally performed, but only to obtain an indicative value, since with this method it is assumed that each data point behaves as a dilution of E2 (Elskens et al., 2011). In reality, most samples do not reach a maximum RLU equal to that of the E2 standards and the response depends on the dilution factor, yielding high uncertainties on the calculated BEQs. For samples with a BEQ-response below the value of the blank filter, half of the median of all procedural and field blanks was used. Analysis of variance (ANOVA) was used to examine the statistically significant difference between the sampling periods. 3. Results and discussion 3.1. Performance of the BG1Luc4E2 cell line Results in Table 2 and Fig. 1 provide an overview of the performance parameters for the BG1Luc4E2 cell line, calculated using the Hill equation and BoxeCox (BC) transformation (n ¼ 27). Fig. 1 shows the estrogenic response, expressed as the percentage of Relative Light Units (% RLU), in function of the concentration 17bestradiol standard solution. In Fig. 1A, the mean of the % RLUs of duplicate measurements (with SD error bars) are plotted and a Hill fit is used for quantification. Fig. 1B shows the fit of the individual data points after BoxeCox transformation. The EC 20-50-80 values, Hill parameters (y0 and h) and BoxeCox slope (s) and linearization parameter (l) are described in Table 2. The EC 50 (median 292 fg per well or 5.64  1012 M E2 per well) was relatively low compared to other estrogenic cell lines, where values ranged from 3.7 ± 0.6  1012 M E2 to 5.7  1011 M E2 (Matsumoto et al., 2005; Berckmans et al., 2007; Wenger et al., 2009; Mertl et al., 2014), which is an indication of high sensitivity. The coefficient of variation (CV) was also lower in comparison with the MELN cell lines (Berckmans et al., 2007; Witters et al., 2010), while the y0 (background) and h (hill coefficient) values were comparable to the ones obtained in an earlier study, using the same cell line (Vandermarken et al., 2015). Viability of the cell culture under normal circumstances was measured with a Nucleocounter® and was generally higher than 98%, while the mean fold induction (i.e. the factor between the maximum and minimum e background - response) was 12.7. 3.2. Influence of extraction solvents In addition to the acetone/hexane extraction mixture, two other extraction solvents (ultra-pure ethanol and acetonitrile) were tested on baked blank filters and five sampled PM10 filters. Table 3 gives an overview of the BEQs of each sample using the different extraction solvents. Results were calculated using the slope ratio method (after BoxeCox transformation) and the inverse prediction (IP) method. For the latter method, the mean BEQ of all dilution points between the background response (of the blank filter) and the maximum response was calculated. From Table 3 it is clear that samples extracted with the acetone/ hexane mixture generally had a lower BEQ (samples 1, 2 and 3) and more dilution points with cytotoxicity (visual changes in cell

K. Croes et al. / Chemosphere 144 (2016) 392e398

395

Table 2 Performance parameters of the BG1Luc4E2 Cell Line (n ¼ 27). y0: background Hill parameter; h: Hill coefficient; s: slope after BoxeCox transformation; l: linearization parameter.

Median Min-max CV (%)

EC 20 (fg well1)

EC 50 (fg well1)

EC 80 (fg well1)

EC 50 BC (fg well1)

y0 (%)

h

s

l

120 100e170 14

292 235e397 14

735 505e1004 17

296 244e415 16

8.05 5.64e11.8 18

1.56 1.33e2.28 11

106 74.5e159 15

2.72 0.82e3.98 28

Fig. 1. b-estradiol standard calibration curve: A. Hill function, plot of the mean percentage of the RLUs of a duplicate measurement with SD error bars in function of the b-estradiol concentration; B. BoxeCox transformation, fit of the individual data points in function of the b-estradiol concentration after BoxeCox linearization of the data.

morphology and/or occurrence of cell death) or suppressed responses (samples 2 and 4) compared to filters extracted with ethanol and/or acetonitrile. Extraction with acetonitrile gave the lowest cytotoxicity, but also a lower BEQ in some cases (samples 3 and 5). It is possible that not all estrogenic chemicals or more antiestrogenic (i.e. antagonistic) compounds were extracted with acetonitrile. Extraction of two blank filters with each solvent showed a noticeable lower background value (8 and 11 fg E2-Eq m3 for acetonitrile and twice 12 fg E2-Eq m3 for ethanol, for a theoretical flow rate of 720 m3 air) compared to the acetone/hexane extract (16.6 fg E2-Eq m3 for the median of eight field and two procedural blanks) (Fig. 2).

From the three tested solvents, ethanol showed the best assay characteristics with low background values, high sample BEQs and less cell death or suppressed responses. It seems that the use of a more polar extraction solvent is advisable. However, to relate the endocrine potential of PM10 air extracts to health effects, more research is needed, since from the current limited research it is unclear what kind of solvent used for extraction in the laboratory best reflects an exposed organism. BEQ calculation using the slope-ratio and IP method yielded similar results, but lower standard errors (SEs) and coefficients of variation (CVs) were observed with the slope-ratio method. This shows that the slope-ratio method can be used as a valuable

Table 3 Influence of extraction solvent on BEQ response# Suppressed response indicates that the induction for a certain dilution factor is lower than the induction of the more diluted sample extract. df is dilution factor; IP is Inverse Prediction; SE is Standard Error. BEQ and SE were calculated using multiple dilution points, each analyzed in duplicate. Sample

Extraction solvent

BEQ (fg E2-Eq m3) IP

Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

Acetone/hexane %) Acetonitrile Ethanol Acetone/hexane %) Acetonitrile Ethanol Acetone/hexane %) Acetonitrile Ethanol Acetone/hexane %) Acetonitrile Ethanol Acetone/hexane %) Acetonitrile Ethanol

SE

Slope ratio

SE

Df with cell death or suppressed# response

Response: df used for BEQ calculation

Df with blank DMSO level

(1/1, v/v 27.1

5.7 28.2

5.1 10e20

28.6e400

e

46.0 46.0 (1/1, v/v 13.1

9.2 42.4 9.2 47.9 4.6 17.0

5.5 10e20 6.9 10e28.6 9.1 10-100

28.6e400 33.3e400 133-200

e e 400

76.3 7.8 53.5 59.2 11.6 58.8 (1/1, v/v 21.8 6.0 26.2

12.5 10e28.6 12.4 10e28.6 5.9 10e20

33.3e200 33.3e200 28.6e400

400 400 e

18.4 3.0 18.8 61.1 20.3 57.0 (1/1, v/v e e e

2.5 10 9.0 10 e 10e133

20e200 20e400 e

400 e 200e400

67.1 22.7 46.8 49.2 16.4 39.3 (1/1, v/v 73.5 7.7 65.8

8.0 10e28.6 6.7 10e33.3 7.4 10e28.6

33.3e400 50e400 33e400

e e e

23.2 88.5

4.3 10 11.3 10e33.3

20e133 66.7e400

200e400 e

0.9 37.7 2.3 94.1

396

K. Croes et al. / Chemosphere 144 (2016) 392e398

Fig. 2. Concentration-response curves of procedural blank extracts (in duplicate for each solvent), using different extraction solvents. The percentage RLU induction corresponds to the response of the sample extract (mean of duplicate) relative to that of the standard solution yielding the highest response. A theoretical flow rate of 720 m3 was used.

alternative to the inverse prediction method for samples with low activities (not yielding a full concentrationeresponse curve), where the 4-parameter Hill function is not accurate. 3.3. Case study: estrogenic activities in an urban location Figs. 3 and 4 and Table S1 present an overview of the estrogenic activity (BEQ expressed as fg E2-Eq m3) of 36 samples from a Flemish urban location (Borgerhout, Belgium). All results were calculated with the slope-ratio method, after BoxeCox transformation. The dashed line in Fig. 3 represents half of the median value of all procedural/field blanks (8.29 ± 2.72 fg E2-Eq m3). All samples below the blank response (n ¼ 10) were set at this limit for further calculations. From these figures, it is clear that high fluctuations in estrogenic activity were observed in particulate matter samples, although the monthly activities were borderline not significantly different (p ¼ 0.053, ANOVA). It seems thus that there is no clear summer/ winter trend and that episodic events significantly influence the observed mean monthly values. This resulted in a non-normal distribution with mean and median BEQs of 50.7 and 35.9 fg E2Eq m3, respectively. In 10 out of 36 samples, no estrogenic activity was detected, while in 19 samples induction levels ranged from

20 to 50% of the maximum E2 induction, which corresponds with BEQs of 20e60 fg E2-Eq m3. In six samples, collected during the months June, August and January, high estrogenic activity was found. This activity did not correlate to the concentration of black carbon (p ¼ 0.32, n ¼ 36) or PM10 (p ¼ 0.16, n ¼ 35) in the samples (Spearman Rank Correlation Test), but for the samples of August and January, a higher number of dilution points with cell toxicity were found. Cytotoxicity was observed by microscopic visualization of the cell culture and was commonly detected for only the two highest concentration points (36 m3 and 18 m3 PM10). However, for the highest peaks in August and January, cytotoxicity occurred for all dilutions between 36 m3 and 2.5 m3 air equivalent. Since heavy metals can be cytotoxic to the cells without causing estrogenic activity, metal concentrations in six hexane/acetone sample extracts and one field blank were determined with ICP-MS. All measured elements yielded very low concentrations in the extracts (Table 4). Since these metals were also determined in PM10 (on the same day, using quartz filters) by the Flemish Environment Agency, recoveries in the organic extracts could be calculated. For all metals, recoveries between 0.1% and 4% were found with exception of sample 2, where recoveries yielded 5% and 23% for respectively copper and zinc. Several other studies on PM also reported low maximum induction levels and cytotoxic effects of particulate

Fig. 3. Estrogenic activity of individual PM10 samples from a Flemish urban location (Borgerhout, Belgium). The dashed line shows half of the median field blank activity.

K. Croes et al. / Chemosphere 144 (2016) 392e398

397

Fig. 4. Mean total and monthly estrogenic activity between April 2013 and January 2014. The numbers indicate the monthly number of samples analyzed.

matter, but no toxic substances could be identified either (Clemons et al., 1998; Wenger et al., 2009). Few international studies using in vitro bioassays to assess estrogenic activity of particulate matter were found. In a Swiss study, BEQs between 0.08 and 1.25 pg E2-Eq m3 were found in PM1 samples, collected in winter during a smog episode (Wenger et al., 2009). On the other hand, no estrogenic activity was detected in coarse particulate matter (<50 mm) and gas phase samples from the Czech Republic (Nov ak et al., 2009) and in PM10 and gas phase samples from Bosnia and Herzegovina (Nov ak et al., 2013). In both cases, however, significant anti-estrogenic responses were observed in the PM fractions. A Canadian study from 2001 to 2002 calculated the apparent recovery of particulate matter extracts in relation to an E2 standard solution giving maximum induction (Klein et al., 2006). The amount of air (m3 per well) necessary to elicit a 20% or 50% response level ranged between 2.3 m3 and 18 m3 (20% response) and between 6.9 m3 and 13 m3 (50% response). This apparent recovery is higher than in this study, where median sample responses of 3.5 m3 (for EC20) and 8.3 m3 (for EC50) were found, suggesting that the extracts from the Canadian study were less potent compared to the ones in this study. Finally, an Australian study did not observe any significant estrogenic activity in outdoor air, but high potencies (in the range of several pg E2-Eq m3) were measured in indoor office air and suburban homes (Kennedy et al., 2009). This suggests that although estrogenic activity was measured in more than 70% of the samples in this study, the BEQs were generally low compared to those from other studies reporting on indoor environments and outdoor air samples from smog episodes.

Table 4 Heavy metal concentrations in PM10 organic extracts. Sample

1 2 3 4 5 6 Field blank

Metal concentrations in PM 10 extract (ng m3) Pb

Zn

Cu

Ni

As

Mn

Cd

Cr

0.200 0.006 0.005 <0.001 0.002 0.002 0.001

0.938 11.006 0.332 0.185 0.246 0.252 0.092

0.485 0.454 0.020 0.017 0.016 0.062 0.000

<0.006 <0.006 <0.006 <0.006 <0.006 <0.006 0.006

0.058 0.144 0.020 0.019 0.011 0.014 0.000

0.012 0.001 0.001 0.002 0.004 0.002 0.001

0.006 0.008 0.001 0.001 0.001 0.001 0.000

0.008 0.007 0.003 0.022 0.031 <0.002 0.002

4. Strengths and weaknesses In this case study, 36 samples were analyzed over a 10-month period. This allowed us to define an average estrogenic potency for an urban location and to assess fluctuations in the response. This approach has the advantage, compared to molecule specific techniques, that the effect of a mixture of estrogenic chemicals can be measured. However, since no anti-estrogenic effects were studied, it can not be excluded that low estrogenic responses result from high levels of antagonistic compounds. Accordingly, future studies will examine anti-estrogenic activities and androgenic/antiandrogenic activities of the same samples (the latter effects will be investigated using a different recombinant cell line). Parallel, a cytotoxicity test will be performed to determine under which circumstances cell death or dysfunction occurs. This additional information will allow us to further define and interpret the total endocrine potential of PM10 air samples. 5. Conclusions Exposure to particulate matter, and the EDCs it carries, pose significant health risks. The ERE-CALUX bioassay was shown to be a valuable tool for assessing the total estrogen activity and potency of air samples. The BG1Luc4E2 cell line proved to be stable and sensitive with low background responses and CVs. The slope-ratio BEQ calculation method seemed to be reliable, showing low standard errors, and can be used instead of the inverse prediction method and the 4-parameter Hill function for samples that do not reach the maximum E2 induction level. A case study in an urban area in Flanders confirmed estrogenic activity by PM10 in most samples, but generally low values were found if compared to indoor air or samples from smog episodes. No clear seasonal (summer/winter) trends were observed, although high fluctuations in estrogenic activity were found. Furthermore, a limited number of samples (from August 2013 and January 2014) yielded a clearly higher estrogenic response and more cytotoxicity. This event could not be related to measured black carbon or PM10 concentrations. Additional research proved that heavy metals, which can cause cytotoxic effects in cell cultures, were not extracted with the organic solvent. It thus still remains unclear which substances cause toxicity, but comparison of extraction solvents indicated that the used acetone/hexane extraction mixture

398

K. Croes et al. / Chemosphere 144 (2016) 392e398

extracts more compounds that cause cell death and suppressed responses than ultra-pure ethanol. Although more research is needed, it seems that the use of a more polar extraction solvent is advisable. Acknowledgments The BG1LUC4E2 cell line was developed by the University of California-Davis (USA) with funding from a Superfund Research Program grant (ES04699) from the National Institute of Environmental Health Sciences. The authors acknowledge the technicians from the Flemish Environment Agency for collection of the PM10 samples. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.chemosphere.2015.09.020. References Belhaj, D., Baccar, R., Jaabiri, I., Bouzid, J., Kallel, M., Ayadi, H., Zhou, J.L., 2015. Fate of selected estrogenic hormones in an urban sewage treatment plant in Tunisia (North Africa). Sci. Total Environ. 505, 154e160. Berckmans, P., Leppens, H., Vangenechten, C., Witters, H., 2007. Screening of endocrine disrupting chemicals with MELN cells, an ER-transactivation assay combined with cytotoxicity assessment. Toxicol. In Vitro 21, 1262e1267. Clemons, J., Allan, L., Marvin, C., Wu, Z., McCarry, B., Bryant, D., Zacharewski, T., 1998. Evidence of estrogen-and TCDD-like activities in crude and fractionated extracts of PM10 air particulate material using in vitro gene expression assays. Environ. Sci. Technol. 32, 1853e1860. Diamanti-Kandarakis, E., Bourguignon, J.-P., Giudice, L.C., Hauser, R., Prins, G.S., Soto, A.M., Zoeller, R.T., Gore, A.C., 2009. Endocrine-disrupting chemicals: an endocrine society scientific statement. Endocr. Rev. 30, 293e342. Elskens, M., Baston, D., Stumpf, C., Haedrich, J., Keupers, I., Croes, K., Denison, M., Baeyens, W., Goeyens, L., 2011. CALUX measurements: statistical inferences for the dose-response curve. Talanta 85, 1966e1973. , J.-P., Blanchet, P., Emeville, E., Giton, F., Giusti, A., Oliva, A., Fiet, J., Thome Multigner, L., 2013. Persistent organochlorine pollutants with endocrine activity and blood steroid hormone levels in middle-aged men. PloS one 8, e66460. Houtman, C.J., Booij, P., Jover, E., Del Rio, D.P., Swart, K., Van Velzen, M., Vreuls, R., Legler, J., Brouwer, A., Lamoree, M.H., 2006. Estrogenic and dioxin-like compounds in sediment from Zierikzee harbour identified with CALUX assaydirected fractionation combined with one and two dimensional gas chromatography analyses. Chemosphere 65, 2244e2252. Janssen, N., Gerlofs-Nijland, M., Lanki, T., Salonen, R., Cassee, F., Hoek, G., Fischer, P., Brunekreef, B., Krzyzanowski, M., 2012. In: Health Effects of Black Carbon. WHO publications. ISBN 978 92 890 0265 3. Kennedy, K., Macova, M., Bartkow, M.E., Hawker, D.W., Zhao, B., Denison, M.S., Mueller, J.F., 2010. Effect based monitoring of seasonal ambient air exposures in Australia sampled by PUF passive air samplers. Atmos. Pollut. Res. 1, 50e58. Kennedy, K., Macova, M., Leusch, F., Bartkow, M.E., Hawker, D.W., Zhao, B., Denison, M.S., Mueller, J.F., 2009. Assessing indoor air exposures using passive sampling with bioanalytical methods for estrogenicity and aryl hydrocarbon receptor activity. Anal. Bioanal. Chem. 394, 1413e1421. Klein, G.P., Hodge, E.M., Diamond, M.L., Yip, A., Dann, T., Stern, G., Denison, M.S.,

Harper, P.A., 2006. Gas-phase ambient air contaminants exhibit significant dioxin-like and estrogen-like activity in vitro. Environ. Health Perspect. 114, 697. Krstevska-Konstantinova, M., Charlier, C., Craen, M., Du Caju, M., Heinrichs, C., de Beaufort, C., Plomteux, G., Bourguignon, J.P., 2001. Sexual precocity after immigration from developing countries to Belgium: evidence of previous exposure to organochlorine pesticides. Hum. Reprod. 16, 1020e1026. Matsumoto, H., Adachi, S., Suzuki, Y., 2005. Bisphenol a in ambient air particulates responsible for the proliferation of MCF-7 human breast cancer cells and its concentration changes over 6 months. Archives Environ. Contam. Toxicol. 48, 459e466. Mertl, J., Kirchnawy, C., Osorio, V., Grininger, A., Richter, A., Bergmair, J., Pyerin, M., Washüttl, M., Tacker, M., 2014. Characterization of estrogen and androgen activity of food contact materials by different in vitro bioassays (YES, YAS, ERa and AR CALUX) and chromatographic analysis (GC-MS, HPLC-MS). PLoS One 9, e100952. Michałowicz, J., Duda, W., 2007. Phenolsdsources and toxicity. Pol. J. Environ. Stud. 16, 347e362. Mihale, M., Croes, K., Tungaraza, C., Baeyens, W., Van Langenhove, K., 2013. PCDD/F and dioxin-like PCB determinations in Mtoni Estuarine sediments (Tanzania) using the Chemically activated luciferase gene expression (CALUX) bioassay. Environ. Pollut. 2, 1e19. Mosteller, F., Tukey, F.W., 1977. Data analysis and regression: a second course in statistics. Addison-Wesley series in behavioral science. In: Quantiative Methods, vol. xvii, p. 588. k, J., Giesy, J.P., Kl Nova anov a, J., Hilscherov a, K., 2013. In vitro effects of pollutants from particulate and volatile fractions of air samplesdday and night variability. Environ. Sci. Pollut. Res. 20, 6620e6627. k, J., Ja lova , V., Giesy, J.P., Hilscherova , K., 2009. Pollutants in particulate and Nova gaseous fractions of ambient air interfere with multiple signaling pathways in vitro. Environ. Int. 35, 43e49. Ozen, S., Darcan, S., Bayindir, P., Karasulu, E., Simsek, D., Gurler, T., 2012. Effects of pesticides used in agriculture on the development of precocious puberty. Environ. Monit. Assess. 184, 4223e4232. €nlinner, M., 2002. Continuous measurement of atmoPetzold, A., Kramer, H., Scho spheric black carbon using a multi-angle absorption photometer. Environ. Sci. Pollut. Res. 4, 78e82. Rogers, J., Denison, M., 2000. Recombinant cell bioassays for endocrine disruptors: development of a stably transfected human ovarian cell line for the detection of estrogenic and anti-estrogenic chemicals. In Vitro Mol. Toxicol. 13, 67e82. Rogers, J.M., Denison, M.S., 2002. Analysis of the Antiestrogenic activity of 2,3,7,8Tetrachlorodibenzo-p-dioxin in human ovarian carcinoma BG-1 cells. Mol. Pharmacol. 61, 1393e1403. Van Den Heuvel, R., Den Hond, E., Croes, K., Nawrot, T., 2014. Uitvoeren van effectgerichte metingen in de Gentse Kanaalzone. Report (Dutch). Van der Linden, S.C., Heringa, M.B., Man, H.-Y., Sonneveld, E., Puijker, L.M., Brouwer, A., Van der Burg, B., 2008. Detection of multiple hormonal activities in wastewater effluents and surface water, using a panel of steroid receptor CALUX bioassays. Environ. Sci. Technol. 42, 5814e5820. Vandermarken, T., De Galan, S., Croes, K., Van Langenhove, K., Vercammen, J., Sanctorum, H., Denison, M., Goeyens, L., Elskens, M., Baeyens, W., 2015. Characterisation and implementation of the ERE-CALUX bioassay on indoor dust samples of kindergartens to assess estrogenic potencies. J. steroid Biochem. Mol. Biol. http://dx.doi.org/10.1016/j.jsbmb.2015.01.005. Wenger, D., Gerecke, A.C., Heeb, N.V., Schmid, P., Hueglin, C., Naegeli, H., Zenobi, R., 2009. In vitro estrogenicity of ambient particulate matter: contribution of hydroxylated polycyclic aromatic hydrocarbons. J. Appl. Toxicol. 29, 223e232. Witters, H., Freyberger, A., Smits, K., Vangenechten, C., Lofink, W., Weimer, M., Bremer, S., Ahr, P.H.J., Berckmans, P., 2010. The assessment of estrogenic or antiestrogenic activity of chemicals by the human stably transfected estrogen sensitive MELN cell line: results of test performance and transferability. Reprod. Toxicol. 30, 60e72.