Trophic transfer of pyrene metabolites between aquatic invertebrates

Environmental Pollution 173 (2013) 61e67

Contents lists available at SciVerse ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Trophic transfer of pyrene metabolites between aquatic invertebrates V. Carrasco Navarro*, M.T. Leppänen 1, J.V.K. Kukkonen 2, S. Godoy Olmos Department of Biology, University of Eastern Finland, Joensuu Campus, P.O. Box 111, FI-80101 Joensuu, Finland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 May 2012 Received in revised form 13 September 2012 Accepted 27 September 2012

The trophic transfer of pyrene metabolites was studied using Gammarus setosus as a predator and the invertebrates Lumbriculus variegatus and Chironomus riparius as prey. The results obtained by liquid scintillation counting confirmed that the pyrene metabolites produced by the aquatic invertebrates L. variegatus and C. riparius were transferred to G. setosus through the diet. More detailed analyses by liquid chromatography discovered that two of the metabolites produced by C. riparius appeared in the chromatograms of G. setosus tissue extracts, proving their trophic transfer. These metabolites were not present in chromatograms of G. setosus exclusively exposed to pyrene. The present study supports the trophic transfer of PAH metabolites between benthic macroinvertebrates and common species of an arctic amphipod. As some PAH metabolites are more toxic than the parent compounds, the present study raises concerns about the consequences of their trophic transfer and the fate and effects of PAHs in natural environments. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Dietary uptake Polycyclic aromatic hydrocarbons Oligochaeta Chironomidae Biotransformation

1. Introduction One of the most important groups of Persistent Organic Pollutants (POPs) are the polycyclic aromatic hydrocarbons (PAHs), present ubiquitously in the aquatic environment, mainly due to anthropogenic activities. Although it has been demonstrated that the concentration of some POPs, such as PCBs and total DDTs increase along the trophic chain (biomagnification) (Hoekstra et al., 2003; Nfon et al., 2008), PAHs do not biomagnify (Nfon et al., 2008; Wan et al., 2007), although can be transferred between prey and predator (Filipowicz et al., 2007). The reported greater biotransformation and excretion capabilities of organisms that belong to higher trophic levels (e.g. fish and mammals) explain this fact (Broman et al., 1990). Although the trophic transfer of PAH metabolites in higher trophic levels may be low, it may be important between invertebrates, as excretion of PAHs or their metabolites is not as efficient as in vertebrates (Landrum et al., 1996). Although PAHs provoke their most toxic effects after biotransformation reactions (Stegeman and Lech, 1991), the existing literature regarding the trophic transfer of their metabolites to predators is scarce, with only few studies reported (McElroy and Sisson, 1989; Palmqvist et al., 2006). There is an increasing * Corresponding author. E-mail address: victor.carrasco.navarro@uef.fi (V. Carrasco Navarro). 1 Present address: Finnish Environment Institute, University of Jyväskylä, The Jyväskylä Office, P.O. Box 35, FI-40014 Jyväskylä, Finland. 2 Present address: Department of Biological and Environmental Sciences, University of Jyväskylä, P.O. Box 35, FI-40014, Finland. 0269-7491/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.envpol.2012.09.023

concern about the toxicological risks and impacts that biotransformation products may cause (McElroy et al., 2011; Ng et al., 2011; van Zelm et al., 2010). Some PAHs derivatives are also a threat due to their persistence (defined as the propensity of a certain chemical to remain in the environment, taking into account the different environmental media and the irreversibility of the chemical’s degradation reactions; Webster et al., 1998; Rodan, 2002). For example, fungal phase II metabolites are not mineralized to CO2 (Schmidt et al., 2010) and oxy-PAHs have been found in similar or greater concentrations than PAHs in several environmental matrices (Layshock et al., 2010). One of the most commonly used PAHs in laboratory tests is pyrene, which is included in the US-EPA list of priority pollutants (http://www.epa.gov/waterscience/methods/pollutants.htm). Although pyrene is not as potentially carcinogenic or toxic as some five ring PAH as dibenzo (a, l) pyrene, it is able to cause other types of toxicity in certain species and conditions. For example, at realistic environmental concentration, pyrene exerted toxicity to benthic microalgae (Petersen et al., 2008) toxicity that increased with the presence of UV light. Also, the blood vessel pulse rate decreased with time in L. variegatus in a continuous exposure to pyrene in water, what was suggested as a sign of narcosis (Mäenpää et al., 2009). Clement et al. (2005) described a decrease in survival and growth in the amphipod H. azteca in two different sediments, but at relatively high concentrations. Pyrene is an excellent model compound due to its simple biotransformation pathway. In eukaryotic organisms, the biotransformation of pyrene results in 1-hydroxy-pyrene as its main phase I

62

V. Carrasco Navarro et al. / Environmental Pollution 173 (2013) 61e67

metabolite (Giessing and Johnsen, 2005), processes performed by isoform(s) of the cytochrome P-450 (CYP) family (Ikenaka et al., 2006). A step further in the biotransformation of PAHs is the phase II, which consists in the conjugation of the phase I metabolite(s) with molecules such as glucose, glucuronic acid, sulphate or other more complicated structures (Beach et al., 2010; Carrasco Navarro et al., 2011; Ikenaka et al., 2006, 2007; Stroomberg et al., 2004; Ueda et al., 2011). For the present study, we chose freshwater and marine species that have a wide distribution and are key organisms in their respective food webs. As prey, Lumbriculus variegatus (Oligochaeta) and Chironomus riparius (Diptera: Chironomidae) were used. They are recommended for the testing of chemicals (OECD, 2007, 2010, respectively) and also represent groups that are an important part of the diet for larger invertebrates and small fish (Leversee et al., 1982). As predator, the marine arctic amphipod Gammarus setosus (Amphipoda) is a common organism present in arctic intertidal and subtidal zones on the Svalbard East coast (Olsen et al., 2007) and was chosen as a predator because a model animal with opportunistic feeding behaviour was required for the test design. Other members of the Gammarus genus are also commonly used in ecotoxicological tests (Gaskell et al., 2007). The principal objective of the present study was to assess the trophic transfer of pyrene metabolites produced by L. variegatus and C. riparius to G. setosus. As far as we know, this is the first study that evaluates the trophic transfer of pyrene and its metabolites between aquatic invertebrates. 2. Materials and methods 2.1. Test organisms The marine arctic amphipod Gammarus setosus (Amphipoda) were collected at the Svalbard shoreline in August 2008, transported in coolers and kept at the University of Eastern Finland in 30 L water tanks filled with Artificial Sea Water (ASW); 32 practical salinity units (psu). ASW was made by mixing Instant OceanÒ synthetic Sea Salt (Aquarium Systems, Sarrebourg, France) with deionized tap water. Organisms were kept in the dark at 2
1  C with constant aeration. Change of ASW and feeding (crushed TetraminÒ) was performed twice a week. Lumbriculus variegatus (Oligochaeta) were originally obtained from the US-EPA, Duluth, MI, USA, and reared in artificial freshwater (AFW) at 20
1  C in a 16 h:8 h light:dark cycle at the same location as amphipods. Powdered TetraminÒ was used as food source three days per week and pre soaked paper towels as a substrate. The nonbiting midge Chironomus riparius (Diptera: Chironomidae) were reared as described by Ristola et al. (1999). Eggs laid by adults were collected and placed in a 200 ml beaker filled with AFW until they hatched and developed into 4th instar larvae.

One day prior to the start of the feeding test, G. setosus were added to 400 ml beaker (one organism per beaker), filled with 350 ml of ASW (32 psu) made as described above. Water was not changed during the whole experiment. Aeration was provided constantly and water temperature was 1.5
1  C. In two different treatments, depurated and control (unexposed) worms were fed to two-days-fasted G. setosus (n ¼ 8 and n ¼ 4, respectively) in the dark. The daily ingestion was recorded. Worms that were not ingested were removed from the beakers. The feeding test was designed to last 168 h, but the test was interrupted approximately at 150 h due to the low feeding rate of the organisms during the last days. G. setosus were sampled, rinsed with deionized water, dried on paper towels, weighed and stored at 20  C. In order to determine the pyrene and metabolite body burdens, a pool of depurated worms (115.3
3.72 mg; n ¼ 3) were sampled during the second day of feeding for later extractions and analyses by HPLC. Additionally, four G. setosus were set at the same conditions with a tea infuser containing two pyrene exposed but not depurated worms. This treatment was set in order to evaluate whether the pyrene equivalents leaked from L. variegatus were available to amphipods directly via water. 2.3.2. Feeding of G. setosus with C. riparius C. riparius (Diptera: Chironomidae) were exposed for 24 h to a nominal water concentration of pyrene of 1.5 mg L1 at their fourth instar larval stage. As feeding experiment to G. setosus lasted three days, three separate exposures of 24 h were set in three consecutive days, starting one day prior to the beginning of the test. Constant aeration was provided but no food was added. For further extractions, HPLC and LSC analyses, samples of 1e2 midges (8.05
0.32 mg; n ¼ 5) were sampled per exposure day. Due to the low amount of organisms, the five midge samples were pooled resulting in a single sample for an analysis. Two treatments were established, i.e. G. setosus (n ¼ 8) fed 24 h exposed C. riparius and G. setosus controls (starved; n ¼ 2). The conditions were the same as in the L. variegatus feeding test, but temperature was 4
1  C. The test was started when one C. riparius larva was added and ingested by the amphipod. Another larva was added after 16 h. After following 24 h of starvation, (40 h after the start of the test), four replicates of amphipods fed C. riparius were sampled randomly, rinsed, dried, weighed and stored at 20  C. The remaining four replicates were fed with three larvae, at 1e2 h intervals, to make a total ingestion of five larvae. After approximately 16 h of starvation (timepoint 65 h), all the remaining replicates, fed C. riparius and controls, were sampled, rinsed, dried and stored at 20  C as described above. All the amphipods ingested the C. riparius larvae immediately. Also in feeding experiments with C. riparius, G. setosus (n ¼ 4) were exposed to prey caged in a tea infuser, in order to evaluate whether the pyrene equivalents leaked from the midge and were available to amphipods directly via water. 2.4. Exposure of G. setosus to pyrene in water In order to determine the biotransformation pathway of pyrene by G. setosus, organisms were exposed to pyrene in ASW. These samples belonged to an experiment where the bioaccumulation, toxicokinetics and biotransformation of pyrene were evaluated at two different temperatures using G. setosus as test organism (Carrasco Navarro et al., in preparation). After exposure, organisms were extracted following the same method described in the present study and run on the HPLC. The resulting chromatograms were compared to the chromatograms of prey and G. setosus fed prey in order to discern what peaks were produced by prey or G. setosus and thus evaluate the trophic transfer.

2.2. Chemicals 2.5. Extractions of pyrene and metabolites Pyrene (98%) and radiolabeled pyrene (specific activity 58.7 mCi mmol1) and triethylamine (TEA; 99%) were purchased from SigmaeAldrich, St. Louis, MO, USA; 1-hydroxy-pyrene and its glucuronide conjugate from Dr. Ehrenstorfer (Augsburg, Germany); ammonium acetate (puriss p.a. for HPLC) from Fluka (Buchs, Switzerland). Solvents (hexane, methanol, acetone, ethanol and acetonitrile) were all reagent or High Performance Liquid Chromatography (HPLC) grade. If not specified otherwise, water used in the experiments was purified Milli-Q water (Millipore Co, Billerica, MA, USA). 2.3. Trophic transfer experiment 2.3.1. Feeding of G. setosus with L. variegatus Approximately 1500 L. variegatus (Oligochaeta) were exposed to AFW borne pyrene (20 mg L1) approximately for two months at 20  C. Exposure waters were renewed once a week and feeding (powdered TetraminÒ) was provided twice per week. After the two months exposure, a large group of worms was placed in clean Lake Höytiäinen (62 410 2100 N, 29 230 4900 E) sediment for 96 h and then in clean AFW until their use as a prey. These were called depurated worms. As worms eliminate pyrene faster than metabolites (Leppänen and Kukkonen, 2000), this treatment was used to decrease the parent pyrene body burdens. The whole depuration process was performed similarly as described in Carrasco Navarro et al. (2012).

All organisms were extracted following a method described in Carrasco Navarro et al. (2011, 2012). In a first step, organisms were homogenized with a polystyrene pestle in mixture of hexane and acetone (2.5:2.5 ml) and sonicated for 20 min. Then two more extractions steps with their correspondent sonication (20 min) were performed: in the second step, a mixture of acetone and methanol (2.5:2.5 ml) was used and in the third step, methanol (3 ml) was used. After each of the sonications, the homogenates were centrifuged for 10 min (1200 g) and the three resulting supernatants collected and pooled in a test tube. Fifty ml of nonane was added to the pooled extracts and they were evaporated to near dryness with nitrogen flow. Methanol (500 ml) was added to resuspend the supernatants. The methanol resuspensions were collected, filtered in Spin-XÒ Centrifuge tube filters (Corning Incorporated, Corning, NY, USA) and centrifuged at approximately 540 g for one minute to facilitate the filtration process. This step was repeated twice (final volume w 1500 ml) and resuspensions were transferred to amber HPLC vials and stored at 20  C. The final tissue residues were not further analysed. 2.6. HPLC analyses Extracts were run on the HPLC system, that included the following apparatus: binary pump, degasser, automatic injector, column thermostat, diode array detector (DAD; set at 339 nm), fluorescence detector (FLD; set at 346/384 nm Ex/Em) and

V. Carrasco Navarro et al. / Environmental Pollution 173 (2013) 61e67

63

automatic fraction collector, all belonging to Agilent 1100 series (Agilent Technologies, Santa Clara, CA, USA). For the chromatographic separation, a Zorbax Eclipse XDB-C8, 5 mm,150  4.6 mm (Agilent) and an XBridgeÔ C8, 3.5 mm; 150  4.6 column (Waters) were coupled. Additionally, a Zorbax SB-C8 narrow-bore guard column (5 mm- 2.1  12.5 mm) was attached. Eluents used were ammonium acetate 10 mM with the addition of 0.1% triethylamine (TEA) set at pH 9 and Acetonitrile (ACN). Using 0.7 ml min1 as flow, the gradient started at 20% of ACN and was raised linearly to 55% in 25 min. During the following three min, the flow was increased to 1 ml/min and the gradient was raised to 100% ACN. Finally, these conditions were kept for five min, making a total length of 33 min. The repeatability of this method was evaluated by four daily measures of the area of known standards over five consecutive days and consequent calculation of the peak area relative standard deviation (RSD). RSD is defined as (%) ¼ 100$SD/X, where SD is the standard deviation and X is the average of the area. The RSD for this method is 4.34, 1.16 and 5.18% for pyrene, 1-hydroxy-pyrene and its glucuronide conjugate, respectively. The pyrene and 1-hydroxy-pyrene concentrations in our samples were quantified by the use of linear regressions of external standards. 2.7. Estimation of pyrene and metabolites ingested by G. setosus The number of worms and larvae ingested by G. setosus were recorded and the total mass of prey ingested by each of the replicates was calculated taking into account the 5 mg average estimation for L. variegatus (based on measurements in our cultures) and the 8 mg average estimation for C. riparius (based on the individuals used in the present study). An estimation of pyrene ingested was calculated by multiplying the estimated weight of prey ingested (mg) by the average value of pyrene in prey samples (ng pyrene mg wet wt). An estimation of the ingested metabolites was calculated by subtracting the pyrene values from the total body burden values (calculated based on radioactive tracer) and multiplying by the mg of prey ingested. 2.8. Data fitting and statistical analyses The depuration of total body burdens in L. variegatus was fitted to the first order equation (Carrasco Navarro et al., 2012)
Ca ¼ Cf þ Co  Cf *eðkd*tÞ

(1)

using Sigma Plot 11.0 (SPSS Corporation, Chicago, IL, USA). The equation represents an exponential decay to a minimum and where Ca is the concentration of pyrene equivalents in organisms, Cf is the concentration in organisms after depuration, C0 is the concentration at the start of the depuration (all in mg pyrene equivalents g wet wt1), t is time (h) and kd is the depuration rate constant (h1). Statistical differences in the trophic transfer experiment (p  0.05) were tested with T or ManneWhitney tests, depending on the equal or unequal variances of the data (SPSS v. 17).

3. Results 3.1. Exposure and depuration in L. variegatus The depuration model represented accurately the experimental data (Adj. r2 ¼ 0.91). The predicted body burden values are also close to the experimental values (Table S1). Overall, the organisms excreted the parent pyrene more rapidly than the metabolites (Fig. S1). Therefore, the set up was successful and increased the fraction of metabolites in prey. 3.2. Biotransformation in prey organisms Six peaks not present in controls were found on the HPLCeFLD chromatograms of L. variegatus extracts (Fig. 1A), two of them corresponding to pyrene and 1-hydroxypyrene tR (31.9 and 30.6 min, respectively). None of the other four peaks (L1 to L4, eluting at tR ¼ 13.9, 15.3, 21.75 and 25 min) corresponded with the glucuronide conjugate of 1-hydroxypyrene. Although only L1 and L2 were proven to be pyrene derivative molecules due to their UV spectra (Fig. S2A and B), all these peaks were assumed to be phase II metabolites of pyrene produced by L. variegatus. The HPLCeFLD chromatograms of C. riparius tissue extracts (Fig. 1B) show a small pyrene peak eluting at tR ¼ 31.9 min that corresponds with the tR of standard pyrene. 1-hydroxy-pyrene is

Fig. 1. HPLCeFLD chromatograms of (A) Lumbriculus variegatus exposed to pyrene for approximately two months and depurated in sediment for four days (depurated worms), (B) Chironomus riparius exposed to pyrene for 24 h, (C) Gammarus setosus fed five C. riparius (65 h timepoint) and (D) G. setosus exposed to pyrene in ASW. The peaks of interest are named depending on their origin, L if they were produced by L. variegatus, CR by C. riparius and GS by G. setosus. Additionally, the tR is also indicated. In Fig. 1C only the tR are displayed because the origin of the peaks is discussed in the text.

not found at detectable levels, however a peak that eluted at tR ¼ 28.8 min (CR3) may be a different phase I metabolite produced by this organism. The two main peaks (CR1 and CR2) elute at tR ¼ 17.5 and 20.7 min. Of these, the UV spectrum of CR2 resembles similarity to the UV spectra of pyrene (Fig. S2C), what confirms that it is a derivate. Other peaks of interest, eluting at tR ¼ 11.17, 13.9, 14.3, 19.7, and 20.75 min, are indicated with asterisks (Fig. 1B). 3.3. Trophic transfer experiment 3.3.1. G. setosus fed L. variegatus Only one amphipod of each treatment ingested at least one worm on the sixth day, and that was the main reason to finish the experiment. Assuming the wet wt of a L. variegatus as 5 mg (average weight in our cultures), G. setosus ingested 51
17.2 (n ¼ 5) and 55.8
12.8 (n ¼ 3) mg wet wt of depurated and control worms,

64

V. Carrasco Navarro et al. / Environmental Pollution 173 (2013) 61e67

respectively. Only the amphipods that ingested more than four worms (20 mg) were considered for the calculations. G. setosus fed depurated worms ingested a total of 0.049
0.016 and 0.37
0.13 mg of pyr equivalents of parent pyrene and metabolites respectively. In G. setosus tissues, 0.0083
0.0028 and 0.11
0.04 mg of pyrene equivalents and metabolites were found for the total amount, respectively (Fig. 2A). The mass of metabolites found in G. setosus after the feeding trial were significantly higher than the mass of pyrene ingested (Fig. 2A; ManneWhitney test: U ¼ 1, Z ¼ 2.4, p < 0.016) indicating that, inevitably a transfer of at least part of the metabolites occurred. Even if pyrene is absorbed and biotransformed at a 100% rate, it is impossible to obtain more metabolites directly from the parent pyrene than the amount of pyrene ingested. Most of the metabolites produced by L. variegatus do not appear in the chromatograms of G. setosus fed depurated worms (Fig. S3). Only one peak (tR ¼ 21.75 min), representing a metabolite, is shared in prey and predator, although its presence also in the chromatograms of G. setosus exposed only to pyrene indicates that it is a metabolite of pyrene produced by this organism. This dual presence hampers the evaluation of the trophic transfer of this particular metabolite. Interestingly, there is a peak in the chromatograms of G. setosus fed L. variegatus (Fig. S3B; tR ¼ 23 min), that is not representative of a pyrene metabolite produced by G. setosus (Fig. S3C). However, as it was not detected in the chromatograms of L. variegatus exposed to pyrene (Fig. S3A), it is impossible to prove

the trophic transfer of this metabolite. As the trophic transfer of the metabolites has been proven with LSC, the most plausible explanation is that the phase II metabolites produced by L. variegatus are deconjugated to 1-hydroxy-pyrene in G. setosus’ guts. 1-hydroxypyrene is then absorbed by the amphipod and biotransformed to the suggested phase II metabolites GS1 and GS2. 3.3.2. G. setosus fed C. riparius The quantities of metabolites ingested by G. setosus (0.015 and 0.039 mg of pyrene equivalents for 40 h and 65 h timepoints, respectively) were 7.5 and 7.8 times higher than the ingested quantities of pyrene (0.002 and 0.005 in mg of pyrene equivalents for 40 h and 65 h replicates, respectively; Fig. 2B). Similarly to G. setosus fed depurated worms, the mg of pyrene equivalents of metabolites found in G. setosus fed C. riparius (0.01
0.002 at 40 h and 0.033
0.009 at 65 h samples) exceeded significantly the amount of pyrene ingested in both timepoints (Fig. 2B; ManneWhitney test: U ¼ 0, Z ¼ 2.46, p ¼ 0.03 in both cases). Again, it inevitably indicates that at least some of the metabolites produced by C. riparius were transferred to G. setosus. Analyses of the LCeFLD chromatograms also support these results. Comparison of the metabolic profiles of C. riparius, G. setosus fed C. riparius and G. setosus exposed exclusively to pyrene (Fig. 1BeD, respectively) indicate that there are two metabolites produced by C. riparius (tR ¼ 17.5 and 28.8 min; Fig. 1B) that are not produced by G. setosus (Fig. 1D). However, these two metabolites are present in the chromatograms of G. setosus fed C. riparius (Fig. 1C), indicating that they were inevitably transferred from prey to predators. 3.4. Tea infuser experiments It was assumed that both freshwater organisms died shortly after being introduced in the ASW, so the hypothesis tested was whether the leaked pyrene body burdens of prey were available to G. setosus. The LCeFLD chromatograms of G. setosus exposed to caged L. variegatus (Fig. S4A) showed a peak of pyrene and two other peaks that have been identified as the main phase II metabolites of pyrene in G. setosus (Fig. 1D).Thus, it was concluded that pyrene is the compound that is mainly leaked to water and available for uptake to G. setosus. The leakage of metabolites may be thus irrelevant or under detection limits. In chromatograms of G. setosus exposed to caged C. riparius (Fig. S3B), all the present peaks are very small, almost undetectable, thus the same conclusions were achieved. 4. Discussion 4.1. Bioaccumulation and depuration of pyrene in L. variegatus

Fig. 2. Comparison between the total mass pyrene and metabolites ingested by G. setosus and mass of pyrene and metabolites found in G. setosus. (A) G. setosus fed L. variegatus (B) G. setosus fed C. riparius (40 ad 65 h timepoints). Black bars represent the ingested pyrene, grey the metabolites ingested, striped the pyrene found in the tissues of G. setosus and white the metabolites found in the tissues of G. setosus. In G. setosus fed C. riparius parent pyrene was not found.

After the two month exposure to pyrene the total body burdens were lower compared to the body burdens found after a similar exposure conducted in our laboratory (Carrasco Navarro et al., 2012). This was likely due to the less frequent change of experimental water and feeding performed in the present study. However, at the end of the depuration in clean sediment, the experimental total body burdens were similar between both studies. These similar levels of body burdens after the depuration reflect that there is a part of the body burdens that are eliminated very slowly. Possibly, this indicates that the depuration may be taking place from two much differentiated compartments, what has already been described in literature (Leppänen and Kukkonen, 2000). Unlike in our previous test (Carrasco Navarro et al., 2012), the body burdens of worms while they were in the clean AFW were not followed, as

V. Carrasco Navarro et al. / Environmental Pollution 173 (2013) 61e67

samples of worms were not taken when L. variegatus were removed from clean sediment. This also may influence the calculations involving depurated worms, as in our previous test the modelled body burdens were used in the trophic transfer calculations. In addition, the higher fraction of metabolites at the end of the depuration (Fig. S1) support the faster excretion of parent PAHs compared to the metabolites in invertebrates (Driscoll and McElroy, 1997; Leppänen and Kukkonen, 2000). 4.2. Biotransformation of pyrene in prey organisms The resolution of the peaks found in L. variegatus HPLCeFLD chromatograms (Fig. 1A) is largely improved with the current HPLC conditions, compared to the methods used to run samples from a exposure conducted in our laboratory previously (Carrasco Navarro et al., 2011, 2012). The largest peak obtained with the previous conditions appeared as two peaks with the current method. The peak eluting at tR ¼ 13.9 min (Fig. 1A) is proposed as the glucoseesulphate conjugate of 1-hydroxypyrene, as this was the only peak in L. variegatus chromatograms with enough intensity to be identified with the Q-ToF mass spectrometer (Carrasco Navarro et al., 2011). Regarding C. riparius, its efficiency in biotransforming PAHs (Leversee et al., 1982; Gerould et al., 1983) is supported by the present study, as only low concentrations of pyrene were found (Fig. 2B). The biotransformation of pyrene by C. riparius is more efficient than by L. variegatus and there is only one peak in common between them, eluting at tR ¼ 21.8 min. It is also possible that other phase I metabolites than 1-hydroxypyrene are present in C. riparius extracts (e.g. tR ¼ 28.8 min). Thus, it might also be possible that these minor phase I metabolites are conjugated further to unidentified phase II products. A more detailed investigation with mass spectrometry analyser and NMR is needed in order to identify these biotransformation products. 4.3. Trophic transfer experiments The uptake of PAHs by aquatic organisms through the diet is a relevant exposure route and an important field of research (McElroy et al., 1990; Clements et al., 1994; Costa et al., 2011; Hellou and Leonard, 2004). Additionally, some authors have expressed their concern about the trophic transfer of biotransformation products of PAHs (Driscoll and McElroy, 1997; Leppänen and Kukkonen, 2000; Stegeman and Lech, 1991). However, there are only a few studies that deal with the dietary transfer of PAH metabolites (McElroy and Sisson, 1989; Carrasco Navarro et al., 2012) and, as far as we know, there is only one study about the trophic transfer of PAHs (fluoranthene) metabolites between aquatic invertebrates (Palmqvist et al., 2006). Thus, the present study is, to our knowledge, the only one that reports the trophic transfer of pyrene metabolites between aquatic invertebrates. An important issue about the trophic transfer of PAH metabolites is to know at what extent phase I and/or phase II metabolites contribute to the total metabolites transferred from the gut lumen. In the present study, the trophic transfer of suggested phase II metabolites is reported. This is in accordance with James et al. (1996), who proved the transport of intact BaP phase II metabolites from lumen to systematic circulation of catfish. In similar trophic transfer studies conducted in our laboratory between L. variegatus and using juvenile S. trutta as a predator, a phase II metabolite produced by L. variegatus was found in extracts of the liver/bile of S. trutta (Carrasco Navarro et al., 2012). In the present study, this peak elutes with the current HPLC method at tR 21.75 min (Fig. 1A), and, contrarily as in our previous experiment (Carrasco Navarro et al., 2012), its trophic transfer could not be proved.

65

However, proving the trophic transfer of phase I metabolites is hampered by the fact that phase II reactions of conjugation are rapid, thus the concentrations of phase I metabolites are usually lower than their conjugates. The trophic transfer and bioaccumulation of phase I metabolites of PAHs can be tested by spiking food or dead prey with the desired metabolite(s). Using this method, Beach et al. (2010) demonstrated the dietary uptake of 1-hydroxy-pyrene in the marine whelk Buccinum undatum, which was also found to biotransform 1-hydroxy-pyrene into large quantities of phase II metabolites. Precisely, the existence of phase II metabolites in the tissues of predators implies the transfer of their phase I metabolite(s), due to the mentioned rapid conjugation reactions. In the case of phase I metabolites, biotransformation is a more important phenomenon than bioaccumulation after the trophic transfer occurs (Beach and Hellou, 2011). McElroy et al. (1991) demonstrated a similar dietary uptake between winter flounder fed on BaP and fed on one of its phase I metabolites (BaP 7, 8-dihydrodiol). On the contrary, a mixture of metabolites produced by Nereis virens fed to winter flounder were less prone to be taken up through the diet than the parent BaP, indicating that the uptake differs among metabolites (McElroy et al., 1991). Also it may indicate that metabolites simply spiked to food or prey homogenates are more available than the metabolites produced by prey and fed as such to predators (Palmqvist et al., 2006). Additionally, the trophic transfer of fluoranthene metabolites from Capitella teleta to Nereis virens has been described (Palmqvist et al., 2006) although it was not measured whether phase I, phase II or both kind of metabolites were transferred. It can be argued, however, that part of the metabolites produced by prey could still be in the gut lumen of the amphipods when the extractions were performed. After the feeding of the last L. variegatus or C. riparius, most of the amphipods spent at least 16 h starved. There was only one amphipod that ingested one L. variegatus the last day, in this single case the sampling time was approximately 6 h after the ingestion of the worm. Although as far as we know there is no information on the gut passage time (GPT) of G. setosus, it has been reported to be 1.6 h for 10e15 mm long Gammarus pulex (Gaskell et al., 2007). Assuming that both organisms have similar organism length/gut length ratio, G. setosus would have a maximum GPT of 5.1 h (assuming a 40 mm long organism). This GPT is shorter than the duration of the starvation period after the last feeding on prey. This would suggest that the guts were emptied when the organisms were sampled. Our two additional experiments where G. setosus were exposed to caged prey (and unable to feed on them) helped to confirm that the prey organisms desorbed pyrene to ASW after dying that was available to G. setosus. This desorption could happen also from the amphipods’ faeces, since these were not collected. It could be argued that the origin of the pyrene equivalents in G. setosus fed the freshwater invertebrates originates partly from the prey or amphipods’ faeces. However, analyses of these G. setosus extracts by HPLC showed that the metabolites produced by prey were not present (or are very low), indicating that mainly parent pyrene was available for uptake by G. setosus. Also, it has been reported that a very small loss of PAHs takes place over 60 days from the faeces of a Capitella species due to their compaction (Horng and Taghon, 2001). Therefore, we assume insignificant transfer from Gammarus faeces. Another observation supporting the trophic transfer of the metabolites is that G. setosus ingested C. riparius within minutes and the uningested L. variegatus were removed. Thus, the desorption during this short ingestion time is most likely insignificant. Additionally, the high water solubility of the metabolites and hence their poor accumulation support our conclusions. However, it might also be possible that phase II metabolites are deconjugated back to phase I metabolites, as it happens in water treatment plants

66

V. Carrasco Navarro et al. / Environmental Pollution 173 (2013) 61e67

with estradiol (Andersen et al., 2003). It could be possible that these phase I metabolites are taken up from water by G. setosus, although conditions are not the same as in water treatment plants. Currently, PAHs as parent compounds are included as priority substances under the EU water framework directive (2008/105/EC; http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri¼OJ:L:2008: 348:0084:0097:EN:PDF), and are also identified as priority pollutants by the US-EPA (US-EPA, 2009; http://water.epa.gov/scitech/ methods/cwa/pollutants.cfm). Since PAHs are prone to biotransformation reactions and clear evidence exists that these biotransformation products are transferred between trophic levels (McElroy and Sisson, 1989; Carrasco Navarro et al., 2011; Palmqvist et al., 2006) and can be toxic (Sepic et al., 2003; Lübcke-von Varel et al., 2011; Schuler et al., 2004), it would be wise to include them in the risk assessment of PAHs. Acknowledgements This study was funded by a grant from the Maj & Tor Nessling Foundation (M.T. Leppänen), the Finnish Graduate School in Environmental Science and Technology (EnSTe; V. Carrasco Navarro) and the Academy of Finland (projects 123587 and 214545). The authors thank Marja Noponen and Julia Keronen for their excellent lab assistance. The authors thank Lionel Camus and Iris Jæger for the capture, transport and maintenance of Gammarus setosus. MSc. Heikki Laitala is acknowledged for his assistance in the performance of the G. setosus fed L. variegatus experiment. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2012.09.023. References Andersen, H., Siegrist, H., Halling-Sorensen, B., Ternes, T.A., 2003. Fate of estrogens in a municipal sewage treatment plant. Environmental Science & Technology 37, 4021e4026. Beach, D.G., Hellou, J., 2011. Bioaccumulation and biotransformation of 1-hydroxypyrene by the marine whelk Neptunea lyrata. International Journal of Environmental Analytical Chemistry 91, 1227e1243. Beach, D.G., Quilliam, M.A., Rouleau, C., Croll, R.P., Hellou, J., 2010. Bioaccumulation and biotransformation of pyrene and 1-hydroxypyrene by the marine whelk Buccinum undatum. Environmental Toxicology and Chemistry 29, 779e788. Broman, D., Naf, C., Lundbergh, I., Zebuhr, Y., 1990. An insitu study on the distribution, biotransformation and flux of polycyclic aromatic-hydrocarbons (Pahs) in an aquatic food-chain (Seston, Mytilus edulis L, Somateria mollissima L) from the Baltic e an ecotoxicological perspective. Environmental Toxicology and Chemistry 9, 429e442. Carrasco Navarro, V., Brozinski, J.M., Leppänen, M.T., Honkanen, J.O., Kronberg, L., Kukkonen, J.V.K., 2011. Inhibition of pyrene biotransformation by piperonyl butoxide and identification of two pyrene derivatives in Lumbriculus variegatus (Oligochaeta). Environmental Toxicology and Chemistry 30, 1069e1078. Carrasco Navarro, V., Leppänen, M.T., Honkanen, J.O., Kukkonen, J.V.K., 2012. Trophic transfer of pyrene metabolites and nonextractable fraction from Oligochaete (Lumbriculus variegatus) to juvenile brown trout (Salmo trutta). Chemosphere 88, 55e61. Clement, B., Cauzzi, N., Godde, M., Crozet, K., Chevron, N., 2005. Pyrene toxicity to aquatic pelagic and benthic organisms in single-species and microcosm tests. Polycyclic Aromatic Compounds 25, 271e298. Clements, W.H., Oris, J.T., Wissing, T.E., 1994. Accumulation and food-chain transfer of fluoranthene and benzo[a]pyrene in Chironomus riparius and Lepomis macrochirus. Archives of Environmental Contamination and Toxicology 26, 261e266. Costa, J., Ferreira, M., Rey-Salgueiro, L., Reis-Henriques, M.A., 2011. Comparision of the waterborne and dietary routes of exposure on the effects of benzo(a)pyrene on biotransformation pathways in Nile tilapia (Oreochromis niloticus). Chemosphere 84, 1452e1460. Driscoll, S.B.K., McElroy, A.E., 1997. Elimination of sediment-associated benzo[a] pyrene and its metabolites by polychaete worms exposed to 3-methylcholanthrene. Aquatic Toxicology 39, 77e91. European Union Water framework directive 2008/105/EC, 2008. http://eur-lex. europa.eu/LexUriServ/LexUriServ.do?uri¼OJ: L:2008:348:0084:0097:EN: PDF. Official Journal of the European Union, Strasbourg, France (last accessed 08.05.12.).

Filipowicz, A.B., Weinstein, J.E., Sanger, D.M., 2007. Dietary transfer of fluoranthene from an estuarine Oligochaete (Monopylephorus rubroniveus) to grass shrimp (Palaemonetes pugio): Influence of piperonyl butoxide. Marine Environmental Research 63, 132e145. Gaskell, P.N., Brooks, A.C., Maltby, L., 2007. Variation in the bioaccumulation of a sediment-sorbed hydrophobic compound by benthic macroinvertebrates: patterns and mechanisms. Environmental Science & Technology 41, 1783e1789. Gerould, S., Landrum, P., Giesy, J.P., 1983. Anthracene bioconcentration and biotransformation in chironomids: effects of temperature and concentration. Environmental Pollution Series A, Ecological and Biological 30, 175e188. Giessing, A.M.B., Johnsen, A.R., 2005. Limited microbial degradation of pyrene metabolites from the estuarine polychaete Nereis diversicolor. Chemosphere 61, 1281e1287. Hellou, J., Leonard, J., 2004. Polycyclic aromatic hydrocarbons bioaccumulation and biotransformation products in trout exposed through food pellets. Polycyclic Aromatic Compounds 24, 697e712. Hoekstra, P.F., O’Hara, T.M., Fisk, A.T., Borgs, K., Solomon, K.R., Muir, D.C.G., 2003. Trophic transfer of persistent organochlorine contaminants (OCs) within an Arctic marine food web from the southern Beaufort-Chukchi Seas. Environmental Pollution 124, 509e522. Horng, C.Y., Taghon, G.L., 2001. Loss rates of polycyclic aromatic hydrocarbons from sediment and deposit-feeder fecal pellets. Environmental Toxicology and Chemistry 20, 1465e1473. Ikenaka, Y., Eun, H., Ishizaka, M., Miyabara, Y., 2006. Metabolism of pyrene by aquatic crustacean, Daphnia magna. Aquatic Toxicology 80, 158e165. Ikenaka, Y., Ishizaka, M., Eun, H., Miyabara, Y., 2007. Glucose-sulfate conjugates as a new phase II metabolite formed by aquatic crustaceans. Biochemical and Biophysical Research Communications 360, 490e495. James, M.O., Kleinow, K.M., Tong, Z., Venugopalan, C., 1996. Bioavailability and biotransformation of 3H-benzo[a]pyrene metabolites in in Situ intestinal preparations of uninduced and BNF-induced channel catfish. Marine Environmental Research 42, 309e315. Landrum, P.F., Harkey, G.A., Kukkonen, J., 1996. Evaluation of organic contaminant exposure in aquatic organisms: the significance of bioconcentration and bioaccumulation. In: Newman, M.C., Jogoe, C.H. (Eds.), Ecotoxicology: a Hierarchial Treatment. CRC Press Inc., Boca Raton, FL, pp. 85e131. Layshock, J.A., Wilson, G., Anderson, K.A., 2010. Ketone and quinone-substituted polycyclic aromatic hydrocarbons in mussel tissue, sediment, urban dust, and diesel particulate matrices. Environmental Toxicology and Chemistry 29, 2450e2460. Leppänen, M.T., Kukkonen, J.V.K., 2000. Fate of sediment-associated pyrene and benzo[a]pyrene in the freshwater Oligochaete Lumbriculus variegatus (Müller). Aquatic Toxicology 49, 199e212. Leversee, G.J., Giesy, J.P., Landrum, P.F., Gerould, S., Bowling, J.W., Fannin, T.E., Haddock, J.D., Bartell, S.M., 1982. Kinetics and biotransformation of benzo(a) pyrene in Chironomus riparius. Archives of Environmental Contamination and Toxicology 11, 25e31. Lübcke-von Varel, U., Machala, M., Ciganek, M., Neca, J., Pencikova, K., Palkova, L., Vondracek, J., Löffler, I., Streck, G., Reifferscheid, G., Flückiger-Isler, S., Weiss, J.M., Lamoree, M., Brack, W., 2011. Polar compounds dominate in vitro effects of sediment extracts. Environmental Science & Technology 45, 2384e2390. McElroy, A.E., Sisson, J.D., 1989. Trophic transfer of benzo[a]pyrene metabolites between benthic marine organisms. Marine Environmental Research 28, 265e269. McElroy, A.E., Farrington, J.W., Teal, J.M., 1990. Influence of mode of exposure and the presence of a tubiculous polychaete on the fate of benz[a]anthracene in the Benthos. Environmental Science & Technology 24, 1648e1655. McElroy, A.E., Cahill, J.M., Sisson, J.D., Kleinow, K.M., 1991. Relative bioavailability and DNA adduct formation of benzo[a]pyrene and metabolites in the diet of the winter flounder. Comparative Biochemistry and Physiology Part C: Comparative Pharmacology 100, 29e32. McElroy, A.E., Barron, M.G., Beckvar, N., Driscoll, S.B.K., Meador, J.P., Parkerton, T.F., Preuss, T.G., Steevens, J.A., 2011. A review of the tissue residue approach for organic and organometallic compounds in aquatic organisms. Integrated Environmental Assessment and Management 7, 50e74. Mäenpää, K., Leppänen, M.T., Kukkonen, J.V.K., 2009. Sublethal toxicity and biotransformation of pyrene in Lumbriculus variegatus (Oligochaeta). Science of the Total Environment 407, 2666e2672. Nfon, E., Cousins, I.T., Broman, D., 2008. Biomagnification of organic pollutants in benthic and pelagic marine food chains from the Baltic Sea. Science of the Total Environment 397, 190e204. Ng, C.A., Scheringer, M., Fenner, K., Hungerbuhler, K., 2011. A framework for evaluating the contribution of transformation products to chemical persistence in the environment. Environmental Science & Technology 45, 111e117. Olsen, G.H., Sva, E., Carroll, J., Camus, L., De Coen, W., Smolders, R., Overaas, H., Hylland, K., 2007. Alterations in the energy budget of Arctic benthic species exposed to oil-related compounds. Aquatic Toxicology 83, 85e92. Organization for Economic Cooperation and Development (OECD), 2007. Guidelines for the Testing of Chemicals No 225: Sediment-water Lumbriculus Toxicity Test Using Spiked Sediment. Paris, France. Organization for Economic Cooperation and Development (OECD), 2010. Guidelines for the Testing of Chemicals No 233: Sediment-water Chironomid Life-cycle Toxicity Test Using Spiked Water or Spiked Sediment. Paris, France. Palmqvist, A., Rasmussen, L.J., Forbes, V.E., 2006. Influence of biotransformation on trophic transfer of the PAK fluoranthene. Aquatic Toxicology 80, 309e319.

V. Carrasco Navarro et al. / Environmental Pollution 173 (2013) 61e67 Petersen, D.G., Reichenberg, F., Dahllof, I., 2008. Phototoxicity of pyrene affects benthic algae and bacteria from the arctic. Environmental Science & Technology 42, 1371e1376. Ristola, T., Pellinen, J., Ruokolainen, M., Kostamo, A., Kukkonen, J.V.K., 1999. Effect of sediment type, feeding level, and larval density on growth and development of a midge (Chironomus riparius). Environmental Toxicology and Chemistry 18, 756e764. Rodan, B.D., 2002. Genesis of the global persistent organic pollutant treaty. In: Rodan, B.D. (Ed.), Foundation for Global Action on Persistent Organic Pollutants: a United States Perspective. U.S. Environmental Protection Agency, Office of Research and Development, Washington, DC. NCEA-I-1200, EPA/600/ P-01/003F. Schmidt, S.N., Christensen, J.H., Johnsen, A.R., 2010. Fungal PAH-metabolites resist mineralization by soil microorganisms. Environmental Science & Technology 44, 1677e1682. Schuler, L.J., Landrum, P.F., Lydy, M.J., 2004. Time-dependent toxicity of fluoranthene to freshwater invertebrates and the role of biotransformation on lethal body residues. Environmental Science & Technology 38, 6247e6255. Sepic, E., Bricelj, M., Leskovsek, H., 2003. Toxicity of fluoranthene and its biodegradation metabolites to aquatic organisms. Chemosphere 52, 1125e1133.

67

Stegeman, J.J., Lech, J.J., 1991. Cytochrome-P-450 monooxygenase systems in aquatic species e carcinogen metabolism and biomarkers for carcinogen and pollutant exposure. Environmental Health Perspectives 90, 101e109. Stroomberg, G.J., Zappey, H., Steen, R.J.C.A., van Gestel, C.A.M., Ariese, F., Velthorst, N.H., van Straalen, N.M., 2004. PAH biotransformation in terrestrial invertebrates: a new phase II metabolite in isopods and springtails. Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology 138, 129e137. Ueda, H., Ikenaka, Y., Nakayama, S.M.M., Tanaka-Ueno, T., Ishizuka, M., 2011. PhaseII conjugation ability for PAH metabolism in amphibians: characteristics and inter-species differences. Aquatic Toxicology 105, 337e343. U.S. Environmental Protection Agency, 2009. List of Priority Pollutants. http://water.epa. gov/scitech/methods/cwa/pollutants.cfm. Washington, DC (last accessed 08.05.12.). van Zelm, R., Huijbregts, M.A.J., van de Meent, D., 2010. Transformation products in the life cycle impact assessment of chemicals. Environmental Science & Technology 44, 1004e1009. Wan, Y., Jin, X.H., Hu, J.Y., Jin, F., 2007. Trophic dilution of polycyclic aromatic hydrocarbons (PAHs) in a marine food web from Bohai Bay, North China. Environmental Science & Technology 41, 3109e3114. Webster, E., Mackay, D., Wania, F., 1998. Evaluating environmental persistence. Environmental Toxicology and Chemistry 17, 2148e2158.