Science of the Total Environment 497–498 (2014) 634–641

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Mutagenicity, dioxin-like activity and bioaccumulation of alkylated picene and chrysene derivatives in a German lignite Wiebke Meyer a, Thomas-Benjamin Seiler b, Andreas Christ a, Regine Redelstein b, Wilhelm Püttmann c, Henner Hollert b,d,e,f, Christine Achten a,⁎ a

University of Münster, Institute of Geology and Palaeontology—Applied Geology, Corrensstrasse 24, 48149 Münster, Germany RWTH Aachen University, Institute for Environmental Research, Department of Ecosystem Analysis, Worringerweg 1, 52074 Aachen, Germany J.W.Goethe-University Frankfurt am Main, Institute for Atmospheric and Environmental Sciences, Department of Environmental Analytical Chemistry, Altenhöferallee 1, 60438 Frankfurt/Main, Germany d Key Laboratory of Yangtze River Environment of Education, Ministry of China, College of Environmental Science and Engineering, Tongji University, Shanghai 200092, China e College of Resources and Environmental Science, Chongqing University, Chongqing 400030, China f School of Environment, Nanjing University, China b c

H I G H L I G H T S • • • • •

Chrysene and picene derivatives in a German lignite sample were investigated Compounds were identified using their mass spectra Low dioxin-like activity and mutagenicity were shown The bioaccumulation potential in Lumbriculus variegatus was limited The environmental risk of these compounds is concluded to be comparatively low

a r t i c l e

i n f o

Article history: Received 25 March 2014 Received in revised form 25 July 2014 Accepted 28 July 2014 Available online xxxx Editor: Mark Hanson Keywords: Bioavailability Polycyclic aromatic hydrocarbons EROD assay Ames Fluctuation test Lumbriculus variegatus Effect-directed analysis (EDA)

a b s t r a c t In a former study, a German lignite extract exhibited toxicity to Danio rerio and Caenorhabditis elegans and was shown to have mutagenic and dioxin-like activity. Besides the comparatively low content of known toxic polycyclic aromatic hydrocarbons (PAH), highly intensive peaks of m/z 274 and m/z 324 were observed during the chromatographic analysis. These compounds are assumed to be alkylated chrysenes and picenes (3,3,7trimethyl-1,2,3,4-tetrahydrochrysene, 1,2-(1′-isopropylpropano)-7-methylchrysene and an isomer of the latter, 1,2,9-trimethyl-1,2,3,4-tetrahydropicene and 2,2,9-trimethyl-1,2,3,4-tetrahydropicene). These compounds are intermediates in the diagenetic formation of chrysene and picene from triterpenoids. Due to their general high abundance in lignites and the toxicity observed for the lignite extract, the mechanism-specific toxicity and bioavailability of these compounds were investigated in the present study using the approach of effect-directed analysis. After the separation of the compounds from other PAH, their mutagenic activity (Ames Fluctuation test) and dioxin-like activity (EROD activity) were studied. Both, mutation induction factor (up to 2.9 ± 2.7) and dioxin-like activity (Bio-TEQ of 224 ± 75 pg/g; represents the amount (pg) 2,3,7,8-tetrachlorodibenzo-pdioxin per g coal that would provoke the same toxic effect) were rather low. Bioavailability estimated by the bioaccumulation test with Lumbriculus variegatus was also very limited. Based on the obtained results, the environmental risk of the highly abundant alkylated chrysenes and picenes in lignites is concluded to be low. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Coals contain native polycyclic aromatic hydrocarbons (PAH) that are generated during the diagenetic process of coal formation from higher plant residues. The content and composition of PAH highly

⁎ Corresponding author. Tel.: +49 251 8336170; fax: +49 251 8333933. E-mail address: [email protected] (C. Achten).

http://dx.doi.org/10.1016/j.scitotenv.2014.07.103 0048-9697/© 2014 Elsevier B.V. All rights reserved.

vary and are assumed to depend on coal characteristics such as biological precursor material, geological setting and maturity (Achten and Hofmann, 2009; Laumann et al., 2011; Wang et al., 2010). Maturity is a crucial parameter influencing the PAH content of coals. While the total PAH content of bituminous coals can reach up to 2,500 mg/kg, lower contents were observed in low mature lignites and subbituminous coals and high mature anthracites (Achten and Hofmann, 2009; Wang et al., 2010). Stout and Emsbo-Mattingly (2008) found up to 8.5 mg/kg total PAH (sum of 43 PAH) and 1.2 mg/kg EPA-PAH in

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641

US lignite samples. Wang et al. (2010) investigated three Chinese lignite samples and detected up to 5.9 mg/kg EPA-PAH. In our former studies (Meyer et al., 2013, 2014), we investigated the toxicity and bioavailability of native polycyclic aromatic compounds (PAC) from eight coals of different type. In the sample set, a German lignite from Schöningen was included. Besides the comparatively low content of EPA-PAH, 40 known toxic PAH and total PAH (3.5, 7.1 and 17 mg/kg, respectively), we observed highly intensive peaks of m/z 274 and m/z 324 from the chromatographic analysis of the PAH-containing fraction of the lignite extract (Meyer et al., 2013). These compounds are likely alkylated chrysene and picene derivatives, which were previously found in extracts of immature coals (lignites and sub-bituminous coals) or sediments (Chaffee and Fookes, 1988; Chaffee and Johns, 1983; Chaffee, 1990; Haberer et al., 2006; Hazai et al., 1986; Püttmann, 1988; Schoell et al., 1994; Stout, 1992; Tan and Heit, 1981; Wakeham et al., 1980; Wang and Simoneit, 1991). It is supposed that these compounds are intermediates in the diagenetic transformation of pentacyclic triterpenoids that occur in angiosperms and thus are used as maturity parameters and biomarkers that indicate angiosperm precursor materials in geological samples (Chaffee and Fookes, 1988; Chaffee and Johns, 1983; Chaffee, 1990; Haberer et al., 2006; Hazai et al., 1986; Oros and Simoneit, 2000; Püttmann, 1988; Schoell et al., 1994; Stout, 1992; Tan and Heit, 1981; Wakeham et al., 1980; Wang and Simoneit, 1991). In our recent studies, the investigation of the toxicity of this lignite extract fraction which is containing PAH revealed 100% mortality of Danio rerio embryos after 48 h of exposure and a reproduction inhibition in Caenorhabditis elegans of 10 ± 8% after 96 h of exposure (Meyer et al., 2013). Further investigations of the mechanism-specific toxicity using the Ames Fluctuation test and the induction of 7-ethoxyresorufin-Odeethylase (EROD assay) showed a mutagenic and dioxin-like activity of the extract: The maximum induction factors (mutation induction relative to the negative control) in the Ames Fluctuation test reached up to 6.7 ± 4.8. In the EROD assay, a Bio-TEQ (toxicity equivalent factor; expresses the amount of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the sample that would lead to the same effect) of 16,353 ± 757 pg/g was detected. Regarding the low content of common toxic PAH and the high content (semiquantitative estimation) of the alkylated chrysene and picene derivatives, it is of interest (a) whether these compounds contribute to the toxicity observed for the coal extract fraction in our former studies, and (b) whether they are bioavailable. The obtained results can provide useful information on the environmental relevance of the investigated compounds and will be useful with respect to a possible prioritization of them for the assessment of sediment quality. Thus, the goals of the present study were (1) to further identify the compounds using their mass spectra, (2) to separate them from the

635

other PAH and evaluate their mechanism-specific toxicity and (3) to study their bioavailability using a bioaccumulation test. The recent and present investigations are an example of the effectdirected analysis (EDA), which is a useful tool for the identification of ecotoxicological relevant substances in complex environmental samples (Brack, 2003; Brack and Schirmer, 2003; Hecker and Hollert, 2009). The EDA approach is currently used by the NORMAN network (Network of reference laboratories, research centers and related organisations for monitoring of emerging environmental substances) for the identification and prioritization of contaminants affecting biological systems (Brack et al., 2012). 2. Materials and methods 2.1. Sample processing procedure The lignite sample was obtained from the open lignite mining site in Schöningen (Lower Saxony, Germany). It originates from the Schöningen formation (seam No. 2) of the subhercynian basin which developed in the Lower Eocene in a paralic depositional environment (both marine and limnic influence). The ground (b 200 μm) lignite sample was extracted and fractionated as described by Meyer et al. (2013) and the F2 fraction was analyzed for PAH. For the present study, the F2 fraction of the lignite extract was further fractionated to separate the observed compounds of m/z 274 and m/z 324 (Fig. 1) from the other PAH. These compounds were to be further investigated regarding their mutagenic and dioxin-like activity. The sub-fractionation of the F2 fraction was performed as described by Nocun and Andersson (2012) using high pressure liquid chromatography (HPLC) with a recently developed silver(I)-mercaptopropano silica gel column and a mixture of cyclohexane and tetrahydrofuran (9:1). The silver(I)-mercaptopropano silica gel column is suitable for the separation of PAH based on the number of aromatic rings (Nocun and Andersson, 2012). The separation was carried out on an Agilent 1100 HPLC System (Agilent Technologies) and a homemade silver(I)mercaptopropano silica gel column. The method details are given in Nocun and Andersson (2012). To ensure that separation worked out properly according to ring number, a PAH standard (containing PAH with 2–4 aromatic rings) was additionally run under the same conditions. Four sub-fractions (SF2.1, SF2.2, SF2.3 and SF3.4) were collected after 13.0, 16.5, 20.0 and 25 min and the separation was repeated 25 times. Corresponding fractions were combined resulting in a few milliliters per fraction. The solvent was reduced using a rotary evaporator and following completely evaporated by a nitrogen stream and gentle heating (40 °C), before a solvent exchange from cyclohexane and tetrahydrofuran to dimethyl sulfoxide (DMSO) was performed. Analyses by gas chromatography-mass spectrometry (GC-MS) of the subfractions were performed to check their content of common PAH

Fig. 1. GC-MS (full scan) chromatogram of fraction F2 of the lignite extract; 1: m/z 274; 2, 3, 4, 5: m/z 324.

636

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641

2.3. EROD assay (sub-fraction SF2.4 of the lignite extract)

Fig. 2. Flow diagram describing sample processing procedure.

and compounds of m/z 274 and m/z 324. According to the obtained results, sub-fraction SF2.4 was further used in the Ames Fluctuation test and the EROD assay. The sample processing procedure is illustrated in Fig. 2.

The induction of EROD is widely accepted as a biomarker for the exposure to contaminants with planar, hydrophobic structures (Behrens et al., 2001; Hilscherova et al., 2000; Okey et al., 1994; Schirmer et al., 2000). The contaminants act as agonists to the aryl hydrocarbon receptor (AhR), which mediates the induction of enzymes of the cytochrome P450 complex (CYP). CYP plays an important role in the metabolism of xenobiotics and it has been shown that the metabolic activation of, e.g., PAH is a result of CYP activity (Behrens et al., 2001; Hilscherova et al., 2000; Schirmer et al., 2000). The EROD assay was used to investigate sub-fraction SF2.4. A solvent exchange to dimethyl sulfoxide (DMSO) was performed, resulting in extract concentrations of 10 g extracted coal equivalent per mL DMSO. Extracts were kept at −20 °C in darkness until required. The EROD assay was carried out with RTL-W1 cells (Lee et al., 1993) according to Behrens et al. (1998) with modifications as described by Gustavsson et al. (2004), Wölz et al. (2011) and Heger et al. (2012). For a detailed description of the test procedure, see Meyer et al. (2014) and references cited therein. From the obtained dose–response-relationship and the EC25 of the sample and TCDD induction, a Bio-TEQ was calculated as described by Engwall et al. (1996) with modifications according to Brack et al. (2000). A Bio-TEQ value represents the amount of TCDD [pg/g] in the sample which would induce the same effect and thus this concept is useful for comparing the obtained results with those from other studies. A higher Bio-TEQ value represents higher toxicity, whereas regarding the EC25, in contrast, a lower value represents higher toxicity. The assay was carried out four times independently, each with six replicate measurements. 2.4. Ames Fluctuation test (sub-fraction SF2.4 of the lignite extract)

2.2. PAH analysis by gas chromatography-mass spectrometry (GC-MS) Fraction F2, sub-fractions SF2.1, SF2.2, SF2.3 and SF2.4 and Lumbriculus tissue extracts (resulting from the bioaccumulation test, see below) were analyzed for 40 target PAH as described before (Meyer et al., 2013) using GC-MS. A simultaneous full scan was performed for non-target analytes. For quantification, response factors were calculated from the peak areas of the most intensive ions of the target compounds and the peak areas of internal standards (chrysene-d12 and benzo[ghi]perylened12, respectively). As there is no external standard available for the quantification of compounds of m/z 274 and m/z 324 to create a calibration curve, calibration curves of available external standards (chrysene and benzo[ghi]perylene, respectively) were used to deduce a concentration out of the obtained response factors. Accordingly, this was a mere semiquantitative estimation of the concentrations. Because GC-MS analysis did not reveal any of the target compounds in worm tissue extracts, the extracts were re-analyzed by the far more sensitive technique of gas chromatography coupled with atmospheric pressure laser ionization-mass spectrometry (GC-APLI-MS) to detect accumulated compounds of m/z 274 and m/z 324. GC-APLI-MS is a highly sensitive and selective method for the determination of aromatic compounds (Stader et al., 2013). In APLI, for aromatic compounds baseline noise is reduced due to a highly efficient stepwise two-photon ionization process (Constapel et al., 2005). The detection limits are reduced by a factor up to 3,500 for individual PAH (Stader et al., 2013). The analysis by GC-APLI-MS was carried out using a gas chromatograph (GC 2010 Plus, Shimadzu) coupled with a MaXis 3G ultrahighresolution-quadropule-time-of-flight mass spectrometer (Bruker Daltonics). Ionization was carried out using a multipurpose ion source (iGenTrax) and a KrF excimer laser providing monochromatic light of 248 nm (ALTEX SI laser, ATL Lasertechnik). For the detailed analytical parameters, see Meyer et al. (2014) or Stader et al. (2013).

The mutagenic and promutagenic activity of single compounds or complex environmental samples can be detected by the Ames test (Maron and Ames, 1983). The sub-fraction SF2.4 in DMSO was used for mutagenicity testing. Instead of the original Ames plate incorporation assay (Maron and Ames, 1983), a miniaturized assay in 384-well micro plates (Reifferscheid et al., 2012) was carried out according to the protocol by Wölz et al. (2011) and Heger et al (2012). Salmonella typhimurium tester strains TA 98 (detection of frameshift mutagens) and TA 100 (detection of base-exchange mutations) were used. Both strains were studied with metabolic activation (indicating promutagens) by means of S9 from rat liver, because the assays with metabolic activation showed highest mutation induction during testing of fraction F2 (Meyer et al., 2014). As solvent control (SC), DMSO and as positive control (PC), 2-aminoanthracene were used for both tester strains. For a detailed description of the test procedure, see Meyer et al. (2014) and references therein. From the results, induction factors (IF) were calculated (induction of sample referred to the solvent control) for each extract concentration. For statistical data evaluation, a William's multiple comparison test was performed to find significant mutation induction in the sample compared to the control. Three independent assays were carried out, each with three replicate measurements. 2.5. Bioaccumulation test with Lumbriculus variegatus (coal particles) The freshwater oligochaete Lumbriculus variegatus was chosen for the present investigation because it is a widely accepted organism for studying PAH bioaccumulation (e. g. Harkey et al., 1995; Brunson et al., 1998; Landrum et al., 2002; Hyötyläinen and Oikari, 2004; Kukkonen et al., 2004; Lyytikäinen et al., 2007; Paumen et al., 2008; Mäenpää et al., 2009; Muijs and Jonker, 2011). The bioaccumulation potential of PAH from the lignite sample (ground coal particles) was investigated using the Lumbriculus bioaccumulation test in agreement with OECD guideline 315 (OECD, 2008). For a detailed description of the

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641

637

Fig. 3. Mass spectra of compounds 1–5 detected in fraction F2 of the lignite sample and identification based on comparison to mass spectra in the literature published by Wakeham et al. (1980), Chaffee and Johns (1983), Hazai et al. (1986), Chaffee and Fookes (1988), and Stout (1992); details on the spectra used for comparison are given in Section 3.1.

638

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641

test procedure and extraction of worm tissue see Meyer et al. (2014). Worm tissue extracts were analyzed using GC-MS and GC-APLI-MS. Additionally, for determination of effects resulting from exposure to coal particles, biomass per replicate and mortality were calculated. 3. Results and discussion 3.1. GC-MS analysis of compounds of fraction F2 and sub-fractions SF2.1-4 As a result of GC-MS analysis of fraction F2, five peaks (Fig. 1) were identified by comparing their mass spectra with mass spectra from the literature (Chaffee and Fookes, 1988; Chaffee and Johns, 1983; Hazai et al., 1986; Stout, 1992; Wakeham et al., 1980). Identification was performed by comparing our spectra with published literature mass spectra with the “naked eye”. The relative intensity of the three most intensive ion fragments was used for the comparison. Compound 1 was identified as 3,3,7-trimethyl-1,2,3,4-tetrahydrochrysene (Fig. 3a). Wakeham et al. (1980) published a mass spectrum of this structure with the following relative intensities: m/z 324 (100%), m/z 274 (55%), and m/z 202 (25%), which fits satisfactorily to our results (100%, 68%, 31%). Trimethyltetrahydrochrysenes were found in recent and ancient sediments (Haberer et al., 2006; Tan and Heit, 1981; Wakeham et al., 1980), low mature coals (Schoell et al., 1994; Stout, 1992; Wang and Simoneit, 1991) and low mature coal smoke (Oros and Simoneit, 2000). Trimethyltetrahydrochrysenes are reported to be intermediates in the process of transformation of pentacyclic triterpenoids to chrysene through A-ring cleavage and loss followed by aromatization of rings B–E (Oros and Simoneit, 2000; Stout, 1992; Tan and Heit, 1981; Wakeham et al., 1980). Compound 2 was identified as 1,2-(1′-isopropylpropano)-7methylchrysene (Fig. 3b). Chaffee and Johns (1983) published a mass spectrum with following relative intensities: m/z 281 (100%), m/z 265/ 266 (20%), and m/z 324 (b20%). Chaffee and Fookes (1988) later reported relative intensities of m/z 281 (100%), m/z 324 (30%) and m/z 266 (25%). Our results for this compound were: m/z 281 (100%), m/z 265 (35%), m/z 266 (33%) and m/z 324 (20%). This compound was detected in Australian brown coals, and the compound is proposed to originate from a precursor based on the lupane skeleton (Chaffee and Fookes, 1988; Chaffee and Johns, 1983). Compound 3 and 5 were identified as 1,2,9-trimethyl-1,2,3,4tetrahydropicene (Fig. 3c and d), and an isomer of the latter according to the identical mass spectrum, probably with varying positions of the two vicinal methyl groups at ring E. Chaffee and Fookes (1988) published the following mass spectrum for 1,2,9-trimethyl-1,2,3,4tetrahydropicene: m/z 324 (100%), m/z 309 (75%), and m/z 279 (25%), Hazai et al. (1986) reported m/z 324 (100%), m/z 309 (95%), and m/z 279 (35%) and Stout (1992) reported m/z 324 (100%), m/z 309 (60%) and m/z 279 (b25%). Our values were: m/z 324 (100%), m/z 309 (83%) and m/z 279 (27%) for compound 3 and m/z 324 (100%), m/z 309 (80%) and m/z 279 (27%) for compound 5. 1,2,9-Trimethyl-1,2,3,4tetrahydropicene is suggested to derive from α-amyrin, a pentacyclic triterpenoid of the ursane-type (Stout, 1992; Tan and Heit, 1981; Wakeham et al., 1980). Compound 4 was identified as 2,2,9-trimethyl-1,2,3,4-tetrahydropicene (Fig. 3e) by comparison with mass spectra from Chaffee and Fookes (1988) which showed relative intensities of m/z 324 (100%),

m/z 268 (75%), m/z 252 (15%) and Stout (1992), who reported intensities of m/z 324 (100%), m/z 268 (55%) and m/z 252 (b25%). Our values were: m/z 324 (100%), m/z 268 (91%) and m/z 252 (26%). 2,2,9Trimethyl-1,2,3,4-tetrahydropicene is a proposed transformation product of β-amyrin, which is based on the oleanane skeleton (Stout, 1992; Tan and Heit, 1981; Wakeham et al., 1980). In contrast to the chrysenes (compound 1), the picenes are assumed to be generated after deoxygenation of ring A and subsequent aromatization of ring A–E (Chaffee and Fookes, 1988; Oros and Simoneit, 2000; Stout, 1992; Tan and Heit, 1981; Wakeham et al., 1980). The semiquantitative estimation yielded a summarized content of all five compounds of 520 mg/kg in the lignite sample (Table 1). The highest content of a single compound was observed for 3,3,7trimethyl-1,2,3,4-tetrahydrochrysene (202 mg/kg), while summation of all picene derivatives (compound 3, 4 and 5) resulted in 265 mg/kg. In our former study (Meyer et al., 2013), results showed 3.5 mg/kg EPA-PAH, 7.1 mg/kg of ∑ 40 toxic PAH and 17 mg/kg of total PAH, which does not include the chrysene and picene derivatives. By the semiquantitative estimation of the picene and chrysene derivatives, it can be seen that these compounds represent the overwhelming part of PAH in the lignite sample compared to commonly analyzed and prioritized PAH. The analysis of fractions SF2.1-4 revealed that the separation of 3,3,7-trimethyl-1,2,3,4-tetrahydrochrysene (compound 1) from the other PAH was not possible, but fraction SF2.4 contained no other analyzed PAH beside compounds 2, 3, 4 and 5 (1,2-(1′-isopropylpropano)7-methylchrysene and trimethyl-tetrahydropicenes). As our goal was to investigate, if an extract fraction which does not contain any other PAH besides the alkylated chrysenes and picenes results in any mutagenic or dioxin-like activity, we chose fraction SF2.4 for the following toxicity tests. We assume that the toxicity observed for the whole F2 fraction (Meyer et al., 2013, 2014) includes the effects of the alkylated chrysenes and picenes. Thus, compound 1 was neglected and fraction SF2.4 was used for the toxicity evaluation of compounds 2–5. 3.2. EROD assay (dioxin-like activity) The EROD assay using the sub-fraction SF2.4 (compounds 2–5) showed an increasing induction of EROD activity with increasing extract concentration in the medium (Fig. 4). Using the EC25 of the sample and the TCDD positive control, a Bio-TEQ of 224 ± 75 pg/g was calculated for the lignite sample. Compared to the Bio-TEQ of fraction F2 (16,353 ± 757 pg/g), the Bio-TEQ was much lower. The results suggest that the compounds 2, 3, 4 and 5 are EROD inducers, but with regard to their high occurrence in the sample compared to the very low contents of other PAH, they show only weak activity. It should be further noted

Table 1 Content [mg/kg] of alkylated chrysene and picene derivatives in the lignite sample; semiquantitative estimations. Compound

Proposed structure

Content [mg/kg]

1 2 3 4 5

3,3,7-Trimethyl-1,2,3,4-tetrahydrochrysene 1,2-(1′-Isopropylpropano)-7-methylchrysene 1,2,9-Trimethyl-1,2,3,4-tetrahydropicene 2,2,9-Trimethyl-1,2,3,4-tetrahydropicene Unknown

202 53 157 102 6

Fig. 4. Dose–response relationship of EROD activity (pmol per mg protein and minute) in RTL-W1 cells after exposure to lignite extract fraction SF2.4 (in mg extracted coal equivalent per mL medium); n = 4 (n: independent assays, each with 6 replicate measurements). a Bio-TEQ of 224 ± 75 pg/g was calculated using the EC25 of the sample and the EC25 of the TCDD positive control.

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641

that compounds 2, 3, 4 and 5 were found not only in sub-fraction SF2.4, and hence the Bio-TEQ for these compounds may be slightly underestimated. The obtained Bio-TEQ is comparably low and in the range of soil extracts from low contaminated reference sites (e.g. Wölz et al., 2011). In the literature, only toxicity reports for the final transformation products chrysene and picene, but none for the detected derivatives can be found. Chrysene was reported to induce EROD in RTL-W1 (Behrens et al., 2001; Bols et al., 1999) and other cells (Bosveld et al., 2002; Fent and Bätscher, 2000; Piskorska-Pliszczynska et al., 1986) in vitro, and was shown to act as AhR agonist in the DR CALUX assay (Machala et al., 2001) as well as the yeast recombinant reporter gene assay. Picene was found as an EROD inducer in rat hepatoma cells (Piskorska-Pliszczynska et al., 1986), and as an AhR agonist in the DR CALUX assay (Machala et al., 2001), whereas Murahashi et al. (2007) detected no AhR agonist activity in the yeast recombinant reporter gene assay. The results from the EROD assay show that the induction of EROD is likely for 1,2-(1′-isopropylpropano)-7-methylchrysene, 1,2,9-trimethyl1,2,3,4-tetrahydropicene, 2,2,9-trimethyl-1,2,3,4-tetrahydropicene and compound 5.

3.3. Ames Fluctuation test (mutagenicity) The sub-fraction SF2.4 (compounds 2–5) was tested for mutagenicity in the Ames Fluctuation test with the tester strains TA 98 and TA 100 after metabolic activation (Fig. 5). The results for TA 98 + S9 show no clear increase of reverted wells per plate with increasing extract concentration in the medium. The highest number of reverted wells was detected for the highest extract concentration (4.3 ± 3.2), which represents a maximum induction factor (IFmax) of 2.1 ± 1.6. However, none of the tested extract concentrations led to a mutation induction that was significantly higher than the control induction according to statistical data evaluation (William's multiple comparisons, p b 0.05). Although the results for tester strain TA 100 + S9 also gave no clear dose–response relationship, the two highest extract concentrations caused a significant mutation induction (8.3 ± 0.6 and 10.7 ± 2.3 reverted wells per plate). The corresponding induction factors were 2.1 ± 1.7 and 2.9 ± 2.7, respectively. Thus, a weak promutagenic activity of the sample could be shown, which suggests that the alkylated chrysene and picenes (compounds 2, 3, 4 and 5) in the extract can induce base-pair exchange mutations after metabolic activation. The maximum induction factors of fraction F2 of the lignite extract were 6.7 ± 0.8 in TA 98 + S9 and 2.5 ± 0.8 in TA 100 + S9, respectively (Meyer et al., 2014). The pronounced decrease in mutation induction in TA 98 suggests that mutagenic activity in the parent fraction F2 was more likely exhibited by other PAH (including the known PAH identified before and compound 1).

639

The literature provides no data about the mutagenic activity of the compounds with our proposed structure; there is only information about chrysene and picene. Chrysene was repeatedly found to induce mutagenicity in TA 100 after metabolic activation (Cheung et al., 1993; Florin et al., 1980; Salamone et al., 1979) and Gibson et al. (1978) reported mutagenic activity in TA 98 after non-enzymatic activation by gamma radiation. The reports on picene are ambiguous. Florin et al. (1980) did not detect any mutagenic activity in tester strains TA 98, TA 100, TA 1535 and TA 1537 with and without metabolic activation. Gibson et al. (1978) also found no mutagenicity in TA 98, TA 1535 and TA 1537 and TA 1538 after activation by gamma radiation. In contrast, Platt et al. (1990) found weak mutagenic activity of picene in TA 100 after metabolic activation as well as tumors in mice after treatment with picene. They concluded from these results that picene is a complete carcinogen (tumor initiating and promoting activity) with very weak tumor initiating activity. This is in accordance with earlier findings from Scribner (1973), who detected a tumor initiating activity of picene in mice. However, a more recent study by Flesher et al. (2002a, 2002b), who investigated the carcinogenicity of picene in rats, could not confirm these results. Our results suggest that the alkylated chrysene and picene derivatives that are commonly found in lignites show weak mutagenic activity in the Ames Fluctuation test.

3.4. Bioaccumulation test Besides the toxicity testing of the sub-fraction containing the alkylated chrysenes and picenes, a bioaccumulation test with the ground coal particles was performed to gain knowledge on the bioavailability of these compounds. The analysis of Lumbriculus tissue extracts with common GC-MS did not reveal the presence of the five compounds in the samples. Detection limits during GC-MS analysis for other PAH (40 PAH) range between 0.04 and 0.66 mg/kg (Meyer et al., 2014). In contrast, the analysis using the more sensitive GC-APLI-MS showed traces of compound 1 (3,3,7trimethyl-1,2,3,4-tetrahydrochrysene), 2 (1,2-(1′-isopropylpropano)-7methylchrysene), 3 (1,2,9-trimethyl-1,2,3,4-tetrahydropicene) and 4 (2,2,9-trimethyl-1,2,3,4-tetrahydropicene) in two out of three replicate samples. The accumulated trace compounds could not be precisely quantified by the recently developed method because the quantification method using internal and external standards had not yet been developed and external calibration alone was judged not to be reliable enough. The results from the bioaccumulation test with Lumbriculus suggest that, although a slight bioaccumulation was observed, the bioavailability of the alkylated chrysenes and picenes is very limited and was only detected by using the highly sensitive APLI technique. The conclusion of a minor risk from the investigated compounds is also supported by the evaluation of Lumbriculus biomass and mortality (Meyer et al., 2014). The total

Fig. 5. Number of reverted wells per plate in the Ames Fluctuation test after exposure of S. typhimurium strains TA 98 and TA 100 (with metabolic activation) to lignite extract fraction SF2.4 (different extract concentrations, expressed in mg extracted coal equivalent per mL medium); SC: solvent control, PC: positive control; n = 3 (n: independent assays, each with 3 replicate measurements); * significantly higher number of reverted wells compared to the solvent control (William's multiple comparisons, p b 0.05).

640

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641

biomass per replicate was lower in the lignite sample (0.0509 ± 0.0037 g) compared to the control (0.0803 ± 0.0249 g), but this decrease was not statistically significant (Dunnett-T3 test for inhomogeneous variances, p b 0.05) and no mortality occurred in Lumbriculus. 4. Conclusions From the results of the present study, we conclude that the observed alkylated chrysenes and picenes, which are highly abundant in lignites, pose a minor environmental risk. The EROD assay and the Ames Fluctuation assay indicate that compounds 2 (1,2-(1′-isopropylpropano)-7methylchrysene), 3 (1,2,9-trimethyl-1,2,3,4-tetrahydropicene), 4 (2,2,9trimethyl-1,2,3,4-tetrahydropicene) and 5 show weak dioxin-like activity and mutagenicity, which was shown in a successful application of the EDA approach. The mechanism-specific toxicity of compound 1 (3,3,7trimethyl-1,2,3,4-tetrahydrochrysene) could not be evaluated because separation from other PAH was not successful. However, the results of the bioaccumulation test indicate that all five compounds, including compound 1, have a limited bioaccumulation potential and thus, show limited bioavailability. However, it should be noted that besides the comparatively low effects and limited bioavailability of alkylated chrysenes and picenes, their partially very high abundance in lignites, other low rank coals and sediments should be taken into account. Worldwide, lignite mining increased by 21% between 2000 and 2012, and Germany was the major producer (185 Mt) of lignite in 2012 (BGR, 2013). Lignite is generally extracted by surface mining activities (open-pit mining), which produces large quantities of dust, including coal dust (Ghose and Majee, 2000). Additionally, the alkylated chrysenes and picenes investigated in the present study were also detected in lignite coal smoke (Oros and Simoneit, 2000). All these facts contribute to a high abundance of the investigated compounds in the environment, which can enhance their environmental relevance. As a further result of the present study, we have shown that GCAPLI-MS can be a useful tool to detect extremely low PAH concentrations, which may be overlooked in organisms if other techniques are applied. Acknowledgements We kindly acknowledge Prof. Dr. Jan Andersson and Dr. Margarete Nocun, Institute of Inorganic and Analytical Chemistry, University of Münster, for performing the sub-fractionation of the parent fraction and Kerstin Winkens, Institute for Environmental Research, RWTH Aachen University for performing the EROD assay. The authors would like to express their thanks to Drs. Niels C. Bols and Lucy Lee (University of Waterloo, Canada) for providing RTL-W1 cells. The resources for carrying out this project came from the University of Münster. References Achten C, Hofmann T. Native polycyclic aromatic hydrocarbons (PAH) in coals—a hardly recognized source of environmental contamination. Sci Total Environ 2009;407: 2461–73. Behrens A, Schirmer K, Bols NC, Segner H. Microassay for rapid measurement of 7ethoxyresorufin-O- deethylase activity in intact fish hepatocytes. Mar Environ Res 1998;46:369–73. Behrens A, Schirmer K, Bols NC, Segner H. Polycyclic aromatic hydrocarbons as inducers of cytochrome P4501A enzyme activity in the rainbow trout liver cell line, RTL-W1, and in primary cultures of rainbow trout hepatocytes. Environ Toxicol Chem 2001;20: 632–43. BGR. Energiestudie 2013. Reserven, Ressourcen und Verfügbarkeit von Energierohstoffen, 17; 2013. p. 112 [Hannover]. Bols NC, Schirmer K, Joyce EM, Dixon DG, Greenberg BM, Whyte JJ. Ability of polycyclic aromatic hydrocarbons to induce 7-ethoxyresorufin-O-deethylase activity in a trout liver cell line. Ecotoxicol Environ Saf 1999;44:118–28. Bosveld ATC, de Bie PAF, van den Brink NW, Jongepier H, Klomp AV. In vitro EROD induction equivalency factors for the 10 PAHs generally monitored in risk assessment studies in The Netherlands. Chemosphere 2002;49:75–83.

Brack W. Effect-directed analysis: a promising tool for the identification of organic toxicants in complex mixtures? Anal Bioanal Chem 2003;377:397–407. Brack W, Schirmer K. Effect-directed identification of oxygen and sulfur heterocycles as major polycyclic aromatic cytochrome P4501A-inducers in a contaminated sediment. Environ Sci Technol 2003;37:3062–70. Brack W, Segner H, Möder M, Schüürmann G. Fixed-effect-level toxicity equivalents—a suitable parameter for assessing ethoxyresorufin-O-deethylase induction potency in complex environmental samples. Environ Toxicol Chem 2000;19:2493–501. Brack W, Dulio V, Slobodnik J. The NORMAN network and its activities on emerging environmental substances with a focus on effect-directed analysis of complex environmental contamination. Environ Sci Eur 2012;24:29. Brunson EL, Canfield TJ, Dwyer FJ, Ingersoll CG, Kemble NE. Assessing the bioaccumulation of contaminants from sediments of the Upper Mississippi River using field-collected oligochaetes and laboratory-exposed Lumbriculus variegatus. Arch Environ Contam Toxicol 1998;35:191–201. Chaffee AL. Molecular indicators of diagenesis in lignite diastereomeric configuration of triterpenoid derived aromatic hydrocarbons. Org Geochem 1990;15:485–8. Chaffee AL, Fookes CJR. Polycyclic aromatic hydrocarbons in Australian coals—III. Structural elucidation by proton nuclear magnetic resonance spectroscopy. Org Geochem 1988;12:261–71. Chaffee AL, Johns RB. Polycyclic aromatic hydrocarbons in Australian coals. I. Angularly fused pentacyclic tri- and tetraaromatic components of Victorian brown coal. Geochim Cosmochim Acta 1983;47:2141–55. Cheung YL, Gray TJB, Ioannides C. Mutagenicity of chrysene, its methyl and benzo derivatives, and their interactions with cytochromes P-450 and the Ah-receptor; relevance to their carcinogenic potency. Toxicology 1993;81:69–86. Constapel M, Schellenträger M, Schmitz OJ, Gäb S, Brockmann KJ, Gise R, et al. Atmospheric pressure laser ionization: a novel ionization method for liquid chromatography/ mass spectrometry. Rapid Commun Mass Spectrom 2005;19:326–36. Engwall M, Broman D, Ishaq R, Näf C, Zebühr Y, Brunström B. Toxic potencies of lipophilic extracts from sediments and settling particulate matter (SPM) collected in a PCBcontaminated river system. Environ Toxicol Chem 1996;15:213–22. Fent K, Bätscher R. Cytochrome P4501A induction potencies of polycyclic aromatic hydrocarbons in a fish hepatoma cell line: demonstration of additive interactions. Environ Toxicol Chem 2000;19:2047–58. Flesher JW, Horn J, Lehner AF. Role of the bay- and L-region in the metabolic activation and carcinogenicity of picene and diben[a, h]anthracene. Polycycl Aromat Compd 2002a;22:737–45. Flesher JW, Horn J, Lehner AF. Comparative carcinogenicity of picene and dibenz[a, h]anthracene in the rat. Biochem Biophys Res Commun 2002b;290:275–9. Florin I, Rutberg L, Curvall M, Enzell CR. Screening of tobacco smoke constituents for mutagenicity using the Ames' test. Toxicology 1980;18:219–32. Ghose MK, Majee SR. Assessment of dust generation due to opencast coal mining—an Indian case study. Environ Monit Assess 2000;61:255–63. Gibson TL, Smart VB, Smith LL. Non-enzymic activation of polycyclic aromatic hydrocarbons as mutagens. Mutat Res 1978;49:153–61. Gustavsson LK, Klee N, Olsman H, Hollert H, Engwall M. Fate of Ah receptor agonists during biological treatment of an industrial sludge containing explosives and pharmaceutical residues. Environ Sci Pollut Res 2004;11:379–87. Haberer RM, Mangelsdorf K, Wilkes H, Horsfield B. Occurrence and palaeoenvironmental significance of aromatic hydrocarbon biomarkers in Oligocene sediments from the Mallik 5L-38 Gas Hydrate Production Research Well (Canada). Org Geochem 2006; 37:519–38. Harkey GA, van Hoof PL, Landrum PF. Bioavailability of polycyclic aromatic hydrocarbons from a historically contaminated sediment core. Environ Toxicol Chem 1995;14: 1551–60. Hazai I, Alexander G, Szekely T, Essiger B, Radek D. Investigation of hydrocarbon constituents of a young sub-bituminous coal by gas chromatography-mass spectrometry. J Chromatogr 1986;367:117–33. Hecker M, Hollert H. Effect-directed analysis (EDA) in aquatic ecotoxicology: state of the art and future challenges. Environ. Sci. Pollut. Res. 2009;16:607–13. Heger S, Bluhm K, Agler MT, Maletz S, Schäffer A, Seiler T-B, et al. Biotests for hazard assessment of biofuel fermentation. Energy Environ Sci 2012;5:9778–88. Hilscherova K, Machala M, Kannan K, Blankenship AL, Giesy JP. Cell bioassays for detection of aryl hydrocarbon (AhR) and estrogen receptor (ER) mediated activity in environmental samples. Environ Sci Pollut Res 2000;7:159–71. Hyötyläinen T, Oikari A. Bioaccumulation of PAHs from creosote-contaminated sediment in a laboratory-exposed freshwater oligochaete, Lumbriculus variegatus. Chemosphere 2004;57:159–64. Kukkonen JVK, Landrum PF, Mitra S, Gossiaux DC, Gunnarsson J, Weston D. The role of desorption for describing the bioavailability of select polycyclic aromatic hydrocarbon and polychlorinated biphenyl congeners for seven laboratory-spiked sediments. Environ Toxicol Chem 2004;23:1842–51. Landrum PF, Gedeon ML, Burton GA, Greenberg MS, Rowland CD. Biological responses of Lumbriculus variegatus exposed to fluoranthene-spiked sediment. Arch Environ Contam Toxicol 2002;42:292–302. Laumann S, Micić V, Kruge MA, Achten C, Sachsenhofer RF, Schwarzbauer J, et al. Variations in concentrations and compositions of polycyclic aromatic hydrocarbons (PAHs) in coals related to the coal rank and origin. Environ Pollut 2011;159: 2690–7. Lee LEJ, Clemons JH, Bechtel DG, Caldwell SJ, Han KB, Pasitschniak-Arts M, et al. Development and characterization of a rainbow trout liver cell line expressing cytochrome P450-dependent monooxygenase activity. Cell Biol Toxicol 1993;9:279–94. Lyytikäinen M, Pehkonen S, Akkanen J, Leppänen M, Kukkonen JVK. Bioaccumulation and biotransformation of polycyclic aromatic hydrocarbons during sediment tests with oligochaetes (Lumbriculus variegatus). Environ Toxicol Chem 2007;26:2660–6.

W. Meyer et al. / Science of the Total Environment 497–498 (2014) 634–641 Machala M, Vondrácek J, Bláha L, Ciganek M, Neca J. Aryl hydrocarbon receptor-mediated activity of mutagenic polycyclic aromatic hydrocarbons determined using in vitro reporter gene assay. Mutat Res 2001;497:49–62. Mäenpää K, Leppänen MT, Kukkonen JVK. Sublethal toxicity and biotransformation of pyrene in Lumbriculus variegatus (Oligochaeta). Sci Total Environ 2009;407:2666–72. Maron DM, Ames BN. Revised methods for the Salmonella mutagenicity test. Mutat Res 1983;113:173–215. Meyer W, Seiler T-B, Reininghaus M, Schwarzbauer J, Püttmann W, Hollert H, et al. Limited waterborne acute toxicity of native polycyclic aromatic compounds from coals of different types compared to their total hazard potential. Environ Sci Technol 2013;47:11766–75. Meyer W, Seiler TB, Schwarzbauer J, Püttmann W, Hollert H, Achten C. Polar polycyclic aromatic compounds from different coal types show varying mutagenic potential, EROD induction and bioavailability depending on coal rank. Sci Total Environ 2014; 494–495:320–8. Muijs B, Jonker MTO. Assessing the bioavailability of complex petroleum hydrocarbon mixtures in sediments. Environ Sci Technol 2011;45:3554–61. Murahashi T, Watanabe T, Kanayama M, Kubo T, Hirayama T. Human aryl hydrocarbon receptor ligand activity of 31 non-substituted polycyclic aromatic hydrocarbons as soil contaminants. J Health Sci 2007;53:715–21. Nocun M, Andersson JT. Argentation chromatography for the separation of polycyclic aromatic compounds according to ring number. J Chromatogr A 2012;1219: 47–53. OECD. OECD Guidelines for Testing of chemicals no. 315: bioaccumulation in sedimentdwelling benthic oligochaetes; 2008. Okey AB, Riddick DS, Harperb PA. The Ah receptor: mediator of the toxicity of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Toxicol Lett 1994;70: 1–22. Oros DR, Simoneit BRT. Identification and emission rates of molecular tracers in coal smoke particulate matter. Fuel 2000;79:515–36. Paumen ML, Stol P, Ter Laak TL, Kraak MHS, Van Gestel CAM, Admiraal W. Chronic exposure of the oligochaete Lumbriculus variegatus to polycyclic aromatic compounds (PACs): bioavailability and effects on reproduction. Environ Sci Technol 2008;42: 3434–40. Piskorska-Pliszczynska J, Keys B, Safe S, Newman MS. The cytosolic receptor binding affinities and AHH induction potencies of 29 polynuclear aromatic hydrocarbons. Toxicol Lett 1986;34:67–74. Platt KL, Pfeiffer E, Petrovic P, Friesel H, Beermann D, Hecker E, et al. Comparative tumorigenicity of picene and dibenz[a, h]anthracene in the mouse. Carcinogenesis 1990;11: 1721–6.

641

Püttmann W. Analysis of polycyclic aromatic hydrocarbons in solid sample material using a desorption device coupled to a GC/MS system. Chromatographia 1988;26:171–7. Reifferscheid G, Maes HM, Allner B, Badurova J, Belkin S, Bluhm K, et al. International round-robin study on the Ames Fluctuation test. Environ Mol Mutagen 2012;53: 185–97. Salamone MF, Heddle JA, Katz M. The mutagenic activity of thirty polycyclic aromatic hydrocarbons (PAH) and oxides in urban airborne particulates. Environ Int 1979;2: 37–43. Schirmer K, Chan AGJ, Bols NC. Transitory metabolic disruption and cytotoxicity elicited by benzo[a]pyrene in two cell lines from rainbow trout liver. J Biochem Mol Toxicol 2000;14:262–76. Schoell M, Simoneit BRT, Wang T-G. Organic geochemistry and coal petrology of tertiary brown coal in the Zhoujing mine, Baise Basin, South China—4. Biomarker sources inferred from stable carbon isotope compositions of individual compounds. Org Geochem 1994;21:713–9. Scribner JD. Tumor initiation by apparently noncarcinogenic polycyclic aromatic hydrocarbons. J Natl Cancer Inst 1973;50:1717–9. Stader C, Beer FT, Achten C. Environmental PAH analysis by gas chromatographyatmospheric pressure laser ionization-time-of-flight-mass spectrometry (GC-APLIMS). Anal Bioanal Chem 2013;405:7041–52. Stout SA. Aliphatic and aromatic triterpenoid hydrocarbons in a Tertiary angiospermous lignite. Org Geochem 1992;18:51–66. Stout SA, Emsbo-Mattingly SD. Concentration and character of PAHs and other hydrocarbons in coals of varying rank—implications for environmental studies of soils and sediments containing particulate coal. Org Geochem 2008;39:801–19. Tan YL, Heit M. Biogenic and abiogenic polynuclear aromatic hydrocarbons in sediments from two remote Adirondack lakes. Geochim Cosmochim Acta 1981;45:2267–79. Wakeham SG, Schaffner C, Giger W. Polycyclic aromatic hydrocarbons in Recent lake sediments—II. Compounds derived from biogenic precursors during early diagenesis. Geochim Cosmochim Acta 1980;44:415–29. Wang T-G, Simoneit BRT. Organic geochemistry and coal petrology of Tertiary brown coal in the Zhoujing mine, Baise Basin, South China. Fuel 1991;70:819–29. Wang R, Liu G, Zhang J, Chou C-L, Liu J. Abundances of polycyclic aromatic hydrocarbons (PAHs) in 14 Chinese and American coals and their relation to coal rank and weathering. Energy Fuel 2010;24:6061–6. Wölz J, Schulze T, Lübcke-von Varel U, Fleig M, Reifferscheid G, Brack W, et al. Investigation on soil contamination at recently inundated and non-inundated sites. J Soils Sediments 2011;11:82–92.

Mutagenicity, dioxin-like activity and bioaccumulation of alkylated picene and chrysene derivatives in a German lignite.

In a former study, a German lignite extract exhibited toxicity to Danio rerio and Caenorhabditis elegans and was shown to have mutagenic and dioxin-li...
1MB Sizes 1 Downloads 6 Views