- Email: [email protected]
Benzo(a)pyrene shows low toxicity to three species of terrestrial plants, two soil invertebrates, and soil-nitrifying bacteria
ARTICLE IN PRESS
Ecotoxicology and Environmental Safety 66 (2007) 362–368 www.elsevier.com/locate/ecoenv
Benzo(a)pyrene shows low toxicity to three species of terrestrial plants, two soil invertebrates, and soil-nitrifying bacteria Line E. Sverdrupa,b,, Snorre B. Hagenc, Paul Henning Kroghd, Cornelis A.M. van Gestele a
Department of Biology, University of Oslo, P.O. Box 1050 Blindern, NO-0316 Oslo, Norway Jordforsk—Norwegian Centre for Soil and Environmental Research, Frederik A. Dahls vei 20, N-1432 A˚s, Norway c Institute of Biology, University of Tromsø, NO-9037 Tromsø, Norway d National Environmental Research Institute, Vejlsøvej 25, DK-8600 Silkeborg, Denmark e Institute of Ecological Science, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
b
Received 29 September 2005; received in revised form 10 January 2006; accepted 22 January 2006 Available online 13 March 2006
Abstract We examined the toxicity of benzo(a)pyrene (BaP) to several standard test organisms including the seed emergence and early lifestage growth of three terrestrial plants (Trifolium pratense, Lolium perenne, and Brassica alba), the survival and reproduction of enchytraeids (Enchytraeus crypticus), and the nitrifying ability of soil bacteria. To also have a look at possible food-chain effects, we included a two-species reproduction test with predatory mites (Hypoaspis aculeifer) and collembolan (Folsomia fimetaria) prey. No effect or only weak effects even at very high BaP concentrations were observed for all tests. None of the soil invertebrates were affected within the concentration range tested (up to 947 mg kg1). For soil-nitrifying bacteria, significant effects were recorded at 977 mg kg1, leaving a no observable effect concentration (NOEC) of 293 mg kg1. BaP did not affect seed emergence for any of the plants, but the growth of B. alba was significantly reduced at the highest concentration tested (375 mg kg1), leaving a NOEC of 69 mg kg1. Compared to a number of other polycyclic aromatic compounds previously tested in the same soil type, BaP is generally less toxic. r 2006 Elsevier Inc. All rights reserved. Keywords: Toxicity; Effect; BaP; Growth; Reproduction
1. Introduction Polycyclic aromatic hydrocarbons (PAHs) constitute a variable group of compounds, all being made up of two to seven fused aromatic rings. PAHs are released due to both natural and man-made processes, e.g., burning of biomass or fossil fuels, and they are widespread in the environment. Studies of ice cores in Greenland have shown that the atmospheric level of PAHs is now approximately 100 times the level in the period 1500–1799 (Kawamura et al., 1994). More than 90% of the PAH burden in the UK resides in
Corresponding author. Department of Biology, University of Oslo, P.O. Box 1050, Blindern, NO-0316 Oslo, Norway. E-mail address: [email protected] (L.E. Sverdrup).
0147-6513/$ - see front matter r 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ecoenv.2006.01.007
soil (Wild and Jones, 1995), and PAH contamination is a serious problem on a large number of sites worldwide. PAHs have received significant public attention due to the fact that many of these compounds are genotoxic in humans. Benzo(a)pyrene (BaP) is a typical carcinogenic PAH representative, and this substance is metabolically activated prior to the formation of DNA adducts. Although a number of the epoxide and diol epoxide metabolites formed from BaP are mutagenic, the 7,8-diol9,10-epoxide is believed to be the ultimate carcinogen, at least for mammals (Timbrell, 2000). The metabolic pathway for BaP has been found to differ between species, and so does the fingerprint of DNA adducts from BaP exposure (Pfohl-Leszkowicz et al., 1996). In plants, for instance, the 7,8-diol-9,10-epoxide is not formed at all (Pfohl-Leszkowicz et al., 1996). Due to differences in
ARTICLE IN PRESS L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
metabolization, the mode of toxic action for BaP may differ between species. Several studies have reported on the toxicity of BaP to terrestrial organisms. Among these the majority of data are available for invertebrates; isopods (Van Straalen and Verweij, 1991; Van Brummelen and Stuijfzand, 1993; Van Brummelen et al., 1996), earthworms (Achazi et al., 1995; CCME, 1997; Eason et al., 1999), enchytraeids (Achazi et al., 1995), and springtails (Sverdrup et al., 2002a). There are also a few studies which report on effects of BaP to microorganisms (Park et al., 1990; Eschenbach et al., 1991) and plants (Do¨rr, 1970; El-Fouly, 1980; CCME, 1997). There is substantial variation in the reported threshold values for effects of this substance, partly due to the use of different routes of exposure (food vs. soil). The range of reported lowest observable effect concentrations (LOEC values) in the studies mentioned above is from 1 to 423,800 mg kg1, the lowest values being reported by Achazi et al. (1995). In a previous study on PAH toxicity to the springtail Folsomia fimetaria (Sverdrup et al., 2002a), we related toxicity of these compounds to their concentrations in pore water. By assuming that the organisms were exposed only through pore water and that the substances had a narcotic mode of toxic action, we calculated a threshold level for toxic effects of BaP at 0.0097 mmol/L, which corresponds approximately to the reported water solubility of this substance (Sverdrup et al., 2002a). Based on the fact that BaP was not found to be toxic to springtails in this study, even at soil exposure concentrations up to 840 mg kg1, we hypothesized that this substance did indeed have a rather unspecific mode of toxic action in springtails and that the low water solubility of the substance was limiting its toxicity. Whether this is also valid for other soil-living organisms is still to be determined. The large differences in BaP metabolization that have been observed between species (Pfohl-Leszkowicz et al., 1996) and the possibility that some species might accumulate BaP in excess of their equilibrium with pore water concentrations (due to intake via food) makes it interesting to look for species differences in BaP toxicity based on a broader selection of species. The purpose of the present study was therefore to examine the toxicity of BaP to a range of terrestrial species. By doing this, we expected to be able to reveal some differences in sensitivity that might be attributed to e.g., species differences in exposure pathways and/or in the rate and route of biotransformation of BaP. Further, the data should increase the knowledge about BaP ecotoxicity and thus strengthen the database on which ecological risk assessment of BaP is to be performed. The species selected for sublethal toxicity testing included three species of plants, two species of terrestrial invertebrates, and soilnitrifying bacteria. These are the three groups of organisms that are usually recommended for evaluating toxic stress effects in soils (ISO, 2001). To also have a look at possible food-chain effects, we included a two-species (predator–prey) reproduction test with mites feeding on springtails.
363
All tests were conducted using the same soil type (a sandy loam). 2. Materials and methods 2.1. Experimental soil A Danish agricultural soil type (Askov soil) was dried at 80 1C for 24 h and sieved through a 2 mm mesh. The soil had the following particle size distribution: coarse sand (200–2000 mm) 38.4%, fine sand (63–200 mm) 23.6%, coarse silt (20–63 mm) 10.0%, fine silt (2–20 mm) 12.3%, and clay (o2 mm) 13.0%. The organic matter content of the soil was 2.8%, and the total content of organic carbon was 1.6%. The soil pHH2 O and the total cation exchange capacity (CEC) was 6.2 and 8.14 meq.* 100 g1, respectively. The Askov soil was selected because it has previously been used for toxicity testing of several other PAHs with terrestrial organisms (see Sverdrup et al., 2002a–c).
2.2. Seeds and test organisms Seeds of Lolium perenne and Trifolium pratense were purchased from DLF Trifolium (Randers, Denmark), and Brassica alba seeds were purchased from Ove Rasmussens Planteskole (Silkeborg, Denmark). The seeds were inspected visually; undamaged seeds of similar size were selected for the experiments. Enchytraeus crypticus were mass-bred in petri dishes containing a Bacto agar, kept at 2071 1C with a 12/12-h photoperiod, and fed oatmeal. Test individuals of age 7–8 weeks were transferred from the culture disks to a petri dish with water and examined using a microscope. Sexually mature animals of similar size were selected for the experiments. Hypoaspis aculeifer were cultured in plastic cups where the bottom was covered by a regularly dampened charcoal/plaster of Paris mixture, kept at 2071 1C with a 12/12-h photoperiod, and fed adult F. fimetaria, which had been mass-bred at 2071 1C in petri dishes containing a regularly dampened charcoal/plaster of Paris mixture and fed bakers’ yeast. H. aculeifer of age 16–19 days were used for the experiments.
2.3. Sample preparation BaP (495% purity, Sigma–Aldrich, Copenhagen, Denmark) was dissolved in acetone (J.T. Barker; HPLC quality) in a stock solution corresponding to the highest test dosage. For each of the test concentrations used, dilutions from the stock solution were made using acetone. A volume of acetone stock solution in an amount corresponding to 20% of the soil dry weight was added to each concentration batch and mixed thoroughly in with the soil. Pure acetone was used for the control samples. The solvent was evaporated under a fume head for 24 h.
2.4. Terrestrial plant tests Tests were performed according to the method based on OECD Guidelines 208: ‘‘Terrestrial plants, Growth test’’ (OECD, 1984). The test concentrations were 1, 10, 100, and 500 mg kg1 soil dry weight (nominal values), in addition to solvent controls and controls without solvent. Four replicates were used per concentration. Following sample preparation and evaporation of solvent, 250 mL water was added per 1.25 kg dry wt, corresponding to 65% of the water-holding capacity, and mixed thoroughly in with the soil. The soil was then transferred to the test containers. These were transparent plastic cylinders (35 cm high, inner diameter 9.5 cm), closed at the top and bottom with lids during exposure. To allow for transpiration and gaseous exchange, the upper lid was perforated with one hole of diameter 0.8 mm. Five seeds were added to each replicate or test chamber, which was weighed to the nearest 0.1 g and placed in the greenhouse. The exposure period was 19 days, i.e., 14 days of exposure from the moment that 50% of
ARTICLE IN PRESS 364
L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
the seeds had emerged in the control samples. Test containers were weighed twice a week during exposure, and weight loss was replaced with water. At each time of watering, all replicates were rerandomized for the position that they occupied in the greenhouse. The minimum temperature was regulated and set to 15 1C (night temperature). During daytime, the temperature usually increased, sometimes to more than 25 1C. To provide a photoperiod of 16/8 h light/dark, we used cultivation lamps of the type Philips SON/T Plus (400 W), which emit light in the wavelength spectrum of 310–780 nm. The lamps were regulated to supply additional light if the light intensity provided from outside the greenhouse dropped below 5 klux; light was switched off when intensity rose above 25 klux. Three lamps were used to illuminate a test area of approximately 3 m2. The artificial light source gave an approximate photo flux density of 300 mE m2 s1 at the soil surface. At the end of the test, seedlings were counted and inspected for visible signs of injury, cut off, put separately in labeled paper bags, and dried at 40 1C for 24 h. The dry weight of each seedling was then determined to the nearest 0.1 mg. For the effect estimations, the mean weight of seedlings in each replicate was used.
blender after which the supernatant was used for inoculation. To each glass container containing 20 g of soil (with various concentrations of the test substance), 3.5 mL of the supernatant was added and mixed thoroughly in with the soil. The amount of water (inoculum) corresponded to 57% of the water-holding capacity. The inoculated soil (24 g) was then transferred to the test containers. Each concentration was tested in three replicates. The test was conducted at 2071 1C in total darkness, and the duration of the test was 28 days. At the end of the test, 1070.05-g soil samples were taken from each replicate. To each sample, 45 mL 1 M KCl was added, after which the resulting mixture was shaken in a mechanical shaker at 50 rpm for 60 min. Then samples were centrifuged (5000g, 10 min), and the supernatant was filtered through a GC filter. Particle-free extracts were then stored at 18 1C. Samples were thawed and placed at 5 1C the day before analysis of their nitrate content. Nitrate was measured spectrophotometrically using a LaChat QuikChems analyzer (LaChat Instruments, Milwaukee, WI, USA) using a proper amount of samples with known concentrations of NO 2 and NO3 and blank samples. All samples were measured in duplicate, and the mean nitrate content of each replicate was used for the statistical analysis.
2.5. Soil invertebrates
2.7. Chemical analysis
The soil invertebrate tests all had survival and reproduction as endpoints. The E. crypticus test was performed according to a recently published OECD Draft Guideline (OECD, 2000), and the H. aculeifer test was performed according to a procedure described by Krogh and Axelsen (1998). Following the sample preparation, water corresponding to 65% of the water-holding capacity was added and mixed thoroughly with the soil. The soil (24 g for E. crypticus and 60 g for H. aculeifer) was then transferred to the test containers. After 2 h, test organisms were added. For the E. crypticus test, 10 adults were added to each replicate, and for the H. aculeifer test, 10 females and five males along with 100 live adult F. fimetaria (food items) were added to each replicate. The invertebrate tests were conducted in temperature-regulated rooms (2071 1C) under a lighting of about 400–800 lux and a 12/12-h photoperiod. In the invertebrate tests, seven test concentrations were used (1, 3, 10, 30, 100, 300, and 1000 mg kg1 dry wt), in addition to the controls. Four replicates were used per concentration. During incubation, the test containers were closed at the top and bottom with lids. The exposure period was 21 days, and test containers were opened after 7 and 14 days for feeding (only E. crypticus) and, in case of water loss, addition of water. At the end of the test period, the H. aculeifer and F. fimetaria were extracted from the test soil using a controlled temperature gradient extractor based on the principles of MacFadyen (1961) and Petersen (1978). E. crypticus were extracted by dividing each sample over two plastic containers and adding water (approximately eight times the amount of soil). The plastic containers were closed with lids, shaken vigorously for 5 s, and then left to sediment for 24 h, individuals of E. crypticus were then found on the surface of the settled soil particles. Mites (H. aculeifer) and remaining food items (i.e., F. fimetaria) were counted manually under a microscope. Enchytraeids were also counted manually using a microscope. In all tests, juveniles produced during exposure could easily be distinguished from the adult individuals originally added to the test chambers.
Verification of exposure concentrations was performed for one of the invertebrate tests, one of the plant tests, and the soil nitrification test. Soil samples were collected from the 0-, 10-, 100-, and 1000-mg kg1 exposure concentrations for the invertebrate test, the 10- and 1000-mg kg1 concentrations for the nitrification test, and the 1-, 10-, 100-, and 500mg kg1 concentrations for the plant tests at the start of the tests. To 100mg (high BaP content) or 1000-mg (low BaP content) soil samples, an equal amount of sodium sulfate was added. Known amounts of the internal standard (d10-BaP) were added to the soil–sodium sulfate mixture. The soils were extracted ultrasonically six times with a 50:50% mixture of dichloromethane (Merck, p.a.) and acetone (Merck; p.a.). Each extraction lasted 30 min. The combined extracts were concentrated to 1.5 mL by evaporation of solvents. Identification and quantification of compounds were carried out with a gas chromatography–mass spectrometric system consisting of a Varian 3400 Star GC and a Varian Saturn III ion trap MS, using temperatureprogrammable splitless injection and a 30-m XTI-5 fused-silica capillary column coated with 95% methyl-siloxane/5% phenyl-siloxane (Restek Corp., Bellefonte, PA, USA).
2.6. Soil-nitrifying bacteria The test was performed using a procedure similar to the method proposed by the International Standardization Organization (ISO, 1997), using test concentrations of 1, 3, 10, 30, 100, 300, 1000, and 3000 mg kg1. However, the use of a high-temperature-treated soil and the use of a solvent in the sample preparation had probably eliminated the nitrifying bacteria originally present in the Askov soil. Therefore, a new microbial community was added to the soil in the form of an inoculum. The inoculum was prepared using fresh Askov soil (stored at 5 1C for 24 h prior to use) and followed a procedure described by Lindahl and Bakken (1995). Essentially, 20 g soil+180 mL deionized water were mixed with a kitchen
2.8. Statistical analysis of results Toxic effect levels were determined on the basis of measured initial concentrations. No observable effect concentrations (NOEC values) and LOEC values were assessed using ANOVA and Dunnett’s procedure (on a 5% significance level) in SAS/STAT (SAS Institute, 1999), after testing for fulfilment of ANOVA requirements using SAS/LAB (SAS Institute, 1992). Estimation of the concentration causing a 10% reduction of growth, reproduction, or nitrate production (EC10 values) was done by linear interpolation, and the confidence intervals were calculated by the ICp method (Norberg-King, 1993).
3. Results The chemical analysis showed that measured start concentrations were close to nominal (Table 1). The response in the solvent control and the control without solvent were not significantly different for any of the tests, and replicates from the two controls were therefore pooled. All tests could be considered valid according to the validity criteria defined in the respective test guidelines (i.e., with respect to seed emergence in the plant tests (OECD, 1984), survival and reproduction criteria for the enchytraeid tests
ARTICLE IN PRESS L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368 Table 1 Results of the chemical analysis of benzo(a)pyrene in soils used in toxicity tests with different soil organisms
Table 2 Overview of the toxicity data generated for benzo(a)pyrene Test (organism)
Test(s)
Soil invertebrates
Terrestrial plants
Nitrifying bacteria
Nominal concentration (mg kg1 dry wt)
Measured concentration (mg kg1 dry wt)
0 10 100 1000 1 10 100 500 10 1000
o0.1 7.7 106 947 0.73 8.0 86 470 7.7 977
Soil samples were collected at the beginning of the tests.
365
Soil nitrification Brassica alba (mustard) Trifolium pratense (red clover) Lolium perenne (ryegrass) Enchytraeus crypticus (Enchytraeidae) Hypoaspis aculeifer (Acari)
Growth/reproduction/nitrate productiona NOEC
LOEC
293b 86 4470 4470 4947
977 470 — — —
4947
—
This substance did not significantly affect the seed emergence of plants or the survival or reproduction of three invertebrate species. For soilnitrifying bacteria and growth of the plants, the no observable effect concentration (NOEC) and the lowest observable effect concentration (LOEC) are given. a Soil effect concentrations (mg kg1 dry wt) based on measured initial concentrations. b Estimated from the measured concentrations at the 100- and 1000mg kg1 nominal treatment levels.
observed, but there was a trend toward a reduction in growth with increasing BaP concentrations for all three species (Fig. 2). However, only B. alba was significantly (ANOVA; Dunnett’s; Po0:05) affected at the highest concentration tested (LOEC ¼ 470 mg kg1 dry soil). Estimated NOEC values for the plant tests are included in Table 2. For B. alba, we also observed a significant increase in seedling growth at the 10-mg/kg exposure level (Fig. 2). For soil nitrification, the dose–response relationship was nearly linear (Fig. 3), and the nitrification process was significantly (ANOVA; Dunnett’s; Po0:05) inhibited at the two highest test concentrations, leaving a NOEC value of 293 mg kg1 dry soil. Summary statistics for the nitrification test is also included in Table 2. 4. Discussion
Fig. 1. Dose–response relationship for effects of benzo(a)pyrene on the survival (mean values) and reproduction (mean value7SD; n ¼ 4) of (A) Enchytraeus crypticus and (B) Hypoaspis aculeifer.
(OECD, 2000), and water content of test containers in the nitrification test (ISO, 1997)). For the soil invertebrates, no significant effect (ANOVA; Dunnett’s; P40:05) of BaP could be detected with respect to survival or reproduction of the two species (Fig. 1 and Table 2). For the plants, there was no effect on seed emergence and no change in leaf color or form was
Neither the organisms in the single-species tests (i.e., plants, springtails, and soil-nitrifying bacteria) nor the mite predator in the two-species test were particularly sensitive to BaP. The absence of strong effects made it difficult to draw conclusions on the relative sensitivity of the tested species to BaP, but plants and bacteria seem to be slightly more sensitive than soil invertebrates (Table 2). We also note that at the 10-mg/kg exposure level, the most sensitive plant (B. alba) exhibited increased growth, which can possibly be explained as an indication of hormesis, an overcompensation in response to low levels of contaminants (Calabrese, 2005). The results presented here are in accordance with the relatively low toxicity reported for BaP in most previous studies on soil organisms (an overview is given in Table 3). However, there is a study by Achazi et al. (1995) showing quite deviating results, with significant toxicity of BaP to enchytraeids (E. crypticus) and earthworms (Eisenia fetida)
ARTICLE IN PRESS 366
L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
Fig. 2. Dose–response relationship for effects of benzo(a)pyrene on the growth of terrestrial plants (error bars indicate SD; n ¼ 4). Results are shown for (A) Brassica alba (mustard), (B) Lolium perenne (ryegrass); and (C) Trifolium pratense (red clover). The asterisk (*) means significantly different from control (ANOVA, Dunnett’s test, Po0:05).
Fig. 3. Dose–response relationship for effects of benzo(a)pyrene on soil nitrate production (mean value7SD; n ¼ 3).
observed even at soil concentrations of 10 mg kg1. The soil type used in the Achazi study is similar to those used in the present study and in the study of Bleeker et al. (2003) with
regard to sorption capacity (organic matter content of 3.9% vs. 2.8% and 2.3% in the latter studies), and the large difference in observed toxicity is therefore not due to the soil type. As pointed out by Jensen and Sverdrup (2003), the data generated by Achazi et al. (1995) are currently used as a reference for setting soil quality standards for BaP in many countries (see, e.g., Kalf et al. (1997)), despite the fact that they deviate from the remaining data pool on this substance. As neither Bleeker et al. (2003) nor we were able to reproduce the high sensitivity found for E. crypticus in the Achazi study, we suggest that both this and the E. fetida test should be repeated before these data are used for setting soil quality standards for BaP. Based on the data generated by Achazi et al. (1995), Kalf et al. (1997) has previously calculated a ‘‘maximum permissible concentration’’ of 0.26 mg/kg for BaP. This value was calculated in the same way as the predicted no effect concentration (PNEC value), which is used by the European Union (European Commission, 1996). By using the data presented here, a PNEC value of 8.6 mg/kg can be calculated based on the lowest NOEC value (86 mg/kg for B. alba) and an assessment factor of 10 according to the Technical Guidance Document (European Commission, 1996). There may be several reasons that BaP shows a low toxicity in this and previous studies with soil organisms. Some reasons relate to exposure, i.e., that the substance might be strongly sorbed to soil particles, thereby limiting exposure, or that pore water concentrations remain low even at high soil concentrations due to a saturation at BaP solubility. With regard to sorption, the soil that we used has relatively low organic carbon content (1.6%) and is therefore a relatively conservative soil type with regard to sorption capacity for organic contaminants. A number of polycyclic aromatic compounds have previously been tested in this soil type, and they seem to fall into one of two groups; either they have typical threshold values for toxicity in the range of 10–40 mg kg1 (Sverdrup et al., 2002a–c, 2003) or they are not toxic even at soil exposure concentrations of more than 500 mg kg1 (Sverdrup et al., 2002a; Bleeker et al., 2003). The latter group contains the very lipophilic PAHs (log Kow45.5) containing four or more rings. For aquatic organisms, it has long been known that some very hydrophobic organic substances do not elicit acute toxicity to fish (Ko¨nemann and Musch, 1981), and this has been explained by their very low water solubility in combination with a limited bioaccumulation potential (i.e., they are certainly accumulated to high tissue concentrations compared to concentrations in the surrounding water, but the water concentrations are still too low to obtain a critical internal concentration and thereby elicit toxicity). Based on this knowledge and our toxicity test results for springtails (Sverdrup et al., 2002a), we previously suggested that very lipophilic PAHs might not be toxic to soil organisms exposed through soil pore water because of their limited water solubility. The low toxicity of BaP found in this study suggests that the results found for springtails might also be valid for a range of other soil
ARTICLE IN PRESS L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
367
Table 3 Overview of literature toxicity data for effects of benzo(a)pyrene on soil organisms Test organismsa
Endpoint
Test duration (days)
Exposure matrix
LOEC (mg kg1)
Enchytraeus crypticus (Oligochaeta)1 Enchytraeus crypticus (Oligochaeta)1 Enchytraeus crypticus (Oligochaeta)1 Enchytraeus crypticus (Oligochaeta)2 Eisenia fetida (Oligochaeta)1 Eisenia fetida (Oligochaeta)1 Eisenia fetida (Oligochaeta)3 Eisenia fetida (Oligochaeta)4 Oniscus asellus (Isopoda)5 Porcellio scaber (Isopoda)5 Oniscus asellus (Isopoda)6 Oniscus asellus (Isopoda)6 Porcellio scaber (Isopoda)6 Folsomia fimetaria (Collembola)7 Folsomia candida (Collembola)2 Soil bacteria8 Soil bacteria and fungi9 Raphanus sativa (radish)4 Lactuca sativa (lettuce)4 Secale cereale (rye)10
Cocoon fertility Cocoon fertility Reproduction Reproduction Cocoon prod. Survival Growth Survival Growth Growth Growth Reproduction Growth Reproduction Reproduction Respiration Community structure Seed emergence Seed emergence Emergence/growth
4 7 30 28 28 28 28 28 63 63 329 329 112 21 28 10 4100 14 14
Agar Food (oats) Soil, 3.9% OMb Soil, 2.3% OCc Soil, 3.9% OM Soil, 3.9% OM Soil OECD soil Food (490% OM) Food (490% OM) Food (490% OM) Food (490% OM) Food (490% OM) Soil, 2.8% OM Soil, 2.3% OC Soil Two different soils OECD soil OECD soil Soil
1 75 100 4931 1 10 4100 448,000 100 100 4316 31.6d 4316 4840 4931 410 433 423,800 11,900 43.3
a
Superscript numbers refer to the following references: 1Achazi et al. (1995); 2Bleeker et al. (2003); 3Eason et al. (1999); 4CCME, (1997); 5Van Brummelen and Stuijfzand (1993); 6Van Brummelen et al. (1996); 7Sverdrup et al. (2002a); 8Eschenbach et al. (1991); 9Park et al. (1990); 10Do¨rr (1970). b OM ¼ organic matter. c OC ¼ organic carbon. d The effect observed was an increase in reproduction.
living species, and the fact that toxicity to all our test organisms was low also supports the assumption that narcosis is the mode of action for this substance. However, there are several potential aspects of BaP toxicity to soil organisms that are not covered by this study. First, all tests have been performed using a relatively short exposure period (about 3 weeks). Both Van Brummelen et al. (1996) and Saint-Denis et al. (2000) found DNA adducts in soil animals (isopods and earthworms, respectively) exposed to BaP, suggesting that BaP undergoes biochemical activation in these animals. Similarly, DNA adducts have also been found in plants (PfohlLeszkowicz et al., 1996). The accumulated effects from DNA damage are of particular importance for organisms suffering long-term exposure, and the effect of such exposure can probably be fully explored only by using multigeneration tests. Second, the fact that BaP usually occurs in mixtures along with other polycyclic aromatic compounds makes it difficult to consider the toxicity of this substance alone. For substances acting by a narcotic mode of toxic action, the summed effect from exposure to various compounds can be characterized by concentration-addition (toxic units approach). Thus even low pore water concentrations of BaP may contribute to the toxicity of a PAH-contaminated sample, and, for marine sediments, methods to deal with this problem have already been developed (e.g., Swartz et al. (1995)). Third, there might still be organisms that are more sensitive than those already
tested. Finally, there is still a possibility that some species might accumulate BaP in excess of their pore water equilibrium concentrations due to different exposure routes or longer exposure periods. In conclusion, our data indicate that, for many different soil organisms, BaP shows a lower rather than a higher toxicity than less-lipophilic PAHs that have previously been tested in the same soil type. We attribute the low toxicity to the low water solubility of this substance. With regard to further work on BaP, several issues still seem unresolved. These include the potential contribution of BaP to the toxicity of PAH mixtures in soils and the ecological significance of BaP and other genotoxic substances in the environment. Acknowledgments This study was financed by a grant from the Norwegian Research Council and Hustadmarmor. We thank Zdenek Gavor and Elin Jørgensen for excellent laboratory assistance and Axel E. Kelley for performing the chemical analysis. References Achazi, R.K., Chroszcz, G., Du¨ker, C., Henneken, M., Rothe, B., Schaub, K., Steudel, I., 1995. The effect of fluoranthene (Fla), benzo(a)pyrene (BaP) and cadmium (Cd) upon survival rate and life cycle parameters
ARTICLE IN PRESS 368
L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
of two terrestrial annelids in laboratory tests systems. Newslett. Enchytraeidae 4, 7–14. Bleeker, E.A.J., Wiegman, S., Droge, S.T.J., Kraak, M.H.S., Van Gestel, C.A.M., 2003. Towards an improvement of the risk assessment of polycyclic (hetero)aromatic hydrocarbons. Report 2003–04 of the Institute of Ecological Science, Vrije Universiteit, Amsterdam, The Netherlands. Calabrese, E.J., 2005. Paradigm lost, paradigm found: the re-emergence of hormesis as a fundamental dose response model in the toxicological sciences. Environ. Pollut. 138, 378–411. CCME, 1997. Recommended Canadian Soil Quality Criteria. Report from the Canadian Council of Ministers of the Environment (CCME), ISBN: 1-895-925-92-4, pp 1–185. Do¨rr, V.R., 1970. Die Aufnahme von 3,4-Benzpyren durch Pflanzenwurzeln. Landwirtsch. Forsch. 23, 371–379. Eason, C.T., Svendsen, C., O’Halloran, K., Weeks, J.M., 1999. An assessment of the lysosomal neutral red retention test and immune function assay in earthworms (Eisenia andrei) following exposure to chlorpyrifos, benzo-a-pyrene (BaP), and contaminated soil. Pedobiologia 43, 641–645. El-Fouly, M.M., 1980. Effect of low concentrations of 3,4-benzpyrene on growth and N-fractions of seedlings. Landwirtsch. Forsch. 33, 108–117. Eschenbach, A., Gehlen, P., Bierl, R., 1991. Untersuchungen zum Einfluss von Fluoranthen und Benzo(a)pyren auf Bodenmikroorganismen and zum Mikrobiellen Abbau dieser Substanzen. Mitteilungen d. Dt. Bodenkundl. Ges. 63, 91–94. European Commission, 1996. Technical guidance document in support of commission directive 93/67/EEC on risk assessment for new notified substances and commission regulation (EC) no 1488/94 on risk assessment for existing substances, Luxembourg. ISO, 1997. Soil quality—biological methods—determination of nitrogen mineralization and nitrification in soils and the influence of chemicals on these processes. ISO Method 14238. International Organization for Standardization, Geneva, Switzerland. ISO, 2001. Soil quality–guidance on the ecotoxicological characterization of soils and soil materials. ISO/DIS 15799. Draft, March 2001. International Organization for Standardization, Geneva, Switzerland. Jensen, J., Sverdrup, L.E., 2003. Polycyclic aromatic hydrocarbon (PAHs) ecotoxicity data for developing soil quality standards. Rev. Environ. Contam. Toxicol. 179, 71–95. Kalf, D.K., Crommentuijn, T., van de Plassche, E.J., 1997. Environmental quality objectives for 10. Ecotoxicol. Environ. Saf. 36, 89–97. Kawamura, K., Suzuki, I., Fujii, Y., Watanabe, O., 1994. Ice core record of polycyclic aromatic hydrocarbons over the past 400 years. Naturwissenschaften 81, 502–505. Ko¨nemann, H., Musch, A., 1981. Quantitative structure-activity relationships in fish toxicity studies. Part 1: relationship for 50 industrial pollutants. Toxicology 19, 209–221. Krogh, P.H., Axelsen, J.A.A., 1998. Test on the predatory mite Hypoaspis aculeifer preying on the collembolan Folsomia fimetaria. In: Løkke, H., Van Gestel, C.A.M. (Eds.), Handbook of Soil Invertebrate Toxicity Tests. Wiley, Baffins, West Sussex, England, pp. 239–251. Lindahl, V., Bakken, L.R., 1995. Evaluation of methods for extraction of bacteria from soil. FEMS Microbiol. Ecol. 16, 135–142. MacFadyen, A., 1961. Improved funnel-type extractors for soil arthropods. J. Anim. Ecol. 30, 171–184. Norberg-King, T.J., 1993. A Linear Interpolation Method for Sublethal Toxicity: The Inhibition Concentration (ICp) Approach (Version 2.0), July 1993. National Effluent Toxicity Assessment Center, SAS Institute, Duluth, Minnesota.
OECD, 1984. Terrestrial plants, growth test. OECD Guideline 208, Organization for Economic Cooperation and Development, Paris, France. OECD, 2000. Enchytraeidae reproduction test. OECD Guidelines 220, Draft guideline, March 2000, Organization for Economic Cooperation and Development, Paris, France. Park, K.S., Sims, R.C., Dupont, R.R., Doucette, W.J., Matthews, J.E., 1990. Fate of PAH compounds in two soil types: influence of volatilization, abiotic loss and biological activity. Environ. Toxicol. Chem. 9, 187–195. Petersen, H., 1978. Some properties of two high gradient extractors for soil microarthropods. Natura Jutlandica 20, 95–121. Pfohl-Leszkowicz, A., Masfaraud, J.F., Weber-Lotfi, F., Marty, J., Guillemaut, P., Castegnaro, M., 1996. Comparison of animal and plant DNA-adducts after exposure to benzo(a)pyrene. Polycycl. Aromat. Compounds 10, 323–331. Saint-Denis, M., Pfohl-Leszkowicz, A., Narbonne, J.F., Ribera, D., 2000. Dose–response and kinetics of the formation of DNA adducts in the earthworm Eisenia fetida andrei exposed to B(a)P-contaminated artificial soil. Polycycl. Aromat. Compounds 18, 117–127. SAS Institute Inc., 1992. SAS/LABs Software: User’s Guide, Version 6, first ed. SAS Institute Inc., Cary, NC. SAS Institute Inc., 1999. SAS/STATs User’s Guide, Version 8. SAS OnlineDocs, SAS Institute Inc., Cary, NC. Sverdrup, L.E., Nielsen, T., Krogh, P.H., 2002a. Soil ecotoxicity of polycyclic aromatic hydrocarbons in relation to soil sorption, lipophilicity, and water solubility. Environ. Sci. Technol. 36, 2429–2435. Sverdrup, L.E., Ekelund, F., Krogh, P.H., Nielsen, T., Johnsen, K., 2002b. Soil microbial toxicity of eight polycyclic aromatic compounds: effects on nitrification, the genetic diversity of bacteria and the total number of protozoans. Environ. Toxicol. Chem. 21, 1644–1650. Sverdrup, L.E., Krogh, P.H., Nielsen, T., Stenersen, J., 2002c. Relative sensitivity of three terrestrial invertebrate tests to polycyclic aromatic compounds. Environ. Toxicol. Chem. 21, 1927–1933. Sverdrup, L.E., Krogh, P.H., Nielsen, T., Kjær, C., Stenersen, J., 2003. Toxicity of eight polycyclic aromatic compounds to the red clover (Trifolium pratense), ryegrass (Lolium perenne) and mustard (Sinapsis alba). Chemosphere 53, 993–1003. Swartz, R.C., Schults, D.W., Ozretich, R.J., Lamberson, J.O., Cole, F.A., P DeWitt, T.H., Redmond, M.S., Ferraro, S.P., 1995. PAH: A model to predict the toxicity of polynuclear aromatic hydrocarbon mixtures in field-collected sediments. Environ. Toxicol. Chem. 14, 1977–1987. Timbrell, J.A., 2000. Principles of Biochemical Toxicology, third ed. Taylor & Francis Ltd., London, UK. Van Brummelen, T.C., Stuijfzand, S.C., 1993. Effects of benzo(a)pyrene on survival, growth and energy reserves in the terrestrial isopods Oniscus asellus and Porcellio scaber. Sci. Total Environ. Suppl., 921–930. Van Brummelen, T.C., Van Gestel, C.A.M., Verweij, R.A., 1996. Longterm toxicity of five polycyclic aromatic hydrocarbons for the terrestrial isopods Oniscus asellus and Porcellio scaber. Environ. Toxicol. Chem. 15, 1199–1210. Van Straalen, N.M., Verweij, R.A., 1991. Effects of benzo(a)pyrene on food assimilation and growth efficiency in Porcellio scaber (Isopoda). Bull. Environ. Contam. Toxicol. 46, 134–140. Wild, S.R., Jones, K.C., 1995. Polynuclear aromatic hydrocarbons in the United Kingdom environment: a preliminary source inventory and budget. Environ. Pollut. 88, 91–108.
Ecotoxicology and Environmental Safety 66 (2007) 362–368 www.elsevier.com/locate/ecoenv
Benzo(a)pyrene shows low toxicity to three species of terrestrial plants, two soil invertebrates, and soil-nitrifying bacteria Line E. Sverdrupa,b,, Snorre B. Hagenc, Paul Henning Kroghd, Cornelis A.M. van Gestele a
Department of Biology, University of Oslo, P.O. Box 1050 Blindern, NO-0316 Oslo, Norway Jordforsk—Norwegian Centre for Soil and Environmental Research, Frederik A. Dahls vei 20, N-1432 A˚s, Norway c Institute of Biology, University of Tromsø, NO-9037 Tromsø, Norway d National Environmental Research Institute, Vejlsøvej 25, DK-8600 Silkeborg, Denmark e Institute of Ecological Science, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
b
Received 29 September 2005; received in revised form 10 January 2006; accepted 22 January 2006 Available online 13 March 2006
Abstract We examined the toxicity of benzo(a)pyrene (BaP) to several standard test organisms including the seed emergence and early lifestage growth of three terrestrial plants (Trifolium pratense, Lolium perenne, and Brassica alba), the survival and reproduction of enchytraeids (Enchytraeus crypticus), and the nitrifying ability of soil bacteria. To also have a look at possible food-chain effects, we included a two-species reproduction test with predatory mites (Hypoaspis aculeifer) and collembolan (Folsomia fimetaria) prey. No effect or only weak effects even at very high BaP concentrations were observed for all tests. None of the soil invertebrates were affected within the concentration range tested (up to 947 mg kg1). For soil-nitrifying bacteria, significant effects were recorded at 977 mg kg1, leaving a no observable effect concentration (NOEC) of 293 mg kg1. BaP did not affect seed emergence for any of the plants, but the growth of B. alba was significantly reduced at the highest concentration tested (375 mg kg1), leaving a NOEC of 69 mg kg1. Compared to a number of other polycyclic aromatic compounds previously tested in the same soil type, BaP is generally less toxic. r 2006 Elsevier Inc. All rights reserved. Keywords: Toxicity; Effect; BaP; Growth; Reproduction
1. Introduction Polycyclic aromatic hydrocarbons (PAHs) constitute a variable group of compounds, all being made up of two to seven fused aromatic rings. PAHs are released due to both natural and man-made processes, e.g., burning of biomass or fossil fuels, and they are widespread in the environment. Studies of ice cores in Greenland have shown that the atmospheric level of PAHs is now approximately 100 times the level in the period 1500–1799 (Kawamura et al., 1994). More than 90% of the PAH burden in the UK resides in
Corresponding author. Department of Biology, University of Oslo, P.O. Box 1050, Blindern, NO-0316 Oslo, Norway. E-mail address: [email protected] (L.E. Sverdrup).
0147-6513/$ - see front matter r 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ecoenv.2006.01.007
soil (Wild and Jones, 1995), and PAH contamination is a serious problem on a large number of sites worldwide. PAHs have received significant public attention due to the fact that many of these compounds are genotoxic in humans. Benzo(a)pyrene (BaP) is a typical carcinogenic PAH representative, and this substance is metabolically activated prior to the formation of DNA adducts. Although a number of the epoxide and diol epoxide metabolites formed from BaP are mutagenic, the 7,8-diol9,10-epoxide is believed to be the ultimate carcinogen, at least for mammals (Timbrell, 2000). The metabolic pathway for BaP has been found to differ between species, and so does the fingerprint of DNA adducts from BaP exposure (Pfohl-Leszkowicz et al., 1996). In plants, for instance, the 7,8-diol-9,10-epoxide is not formed at all (Pfohl-Leszkowicz et al., 1996). Due to differences in
ARTICLE IN PRESS L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
metabolization, the mode of toxic action for BaP may differ between species. Several studies have reported on the toxicity of BaP to terrestrial organisms. Among these the majority of data are available for invertebrates; isopods (Van Straalen and Verweij, 1991; Van Brummelen and Stuijfzand, 1993; Van Brummelen et al., 1996), earthworms (Achazi et al., 1995; CCME, 1997; Eason et al., 1999), enchytraeids (Achazi et al., 1995), and springtails (Sverdrup et al., 2002a). There are also a few studies which report on effects of BaP to microorganisms (Park et al., 1990; Eschenbach et al., 1991) and plants (Do¨rr, 1970; El-Fouly, 1980; CCME, 1997). There is substantial variation in the reported threshold values for effects of this substance, partly due to the use of different routes of exposure (food vs. soil). The range of reported lowest observable effect concentrations (LOEC values) in the studies mentioned above is from 1 to 423,800 mg kg1, the lowest values being reported by Achazi et al. (1995). In a previous study on PAH toxicity to the springtail Folsomia fimetaria (Sverdrup et al., 2002a), we related toxicity of these compounds to their concentrations in pore water. By assuming that the organisms were exposed only through pore water and that the substances had a narcotic mode of toxic action, we calculated a threshold level for toxic effects of BaP at 0.0097 mmol/L, which corresponds approximately to the reported water solubility of this substance (Sverdrup et al., 2002a). Based on the fact that BaP was not found to be toxic to springtails in this study, even at soil exposure concentrations up to 840 mg kg1, we hypothesized that this substance did indeed have a rather unspecific mode of toxic action in springtails and that the low water solubility of the substance was limiting its toxicity. Whether this is also valid for other soil-living organisms is still to be determined. The large differences in BaP metabolization that have been observed between species (Pfohl-Leszkowicz et al., 1996) and the possibility that some species might accumulate BaP in excess of their equilibrium with pore water concentrations (due to intake via food) makes it interesting to look for species differences in BaP toxicity based on a broader selection of species. The purpose of the present study was therefore to examine the toxicity of BaP to a range of terrestrial species. By doing this, we expected to be able to reveal some differences in sensitivity that might be attributed to e.g., species differences in exposure pathways and/or in the rate and route of biotransformation of BaP. Further, the data should increase the knowledge about BaP ecotoxicity and thus strengthen the database on which ecological risk assessment of BaP is to be performed. The species selected for sublethal toxicity testing included three species of plants, two species of terrestrial invertebrates, and soilnitrifying bacteria. These are the three groups of organisms that are usually recommended for evaluating toxic stress effects in soils (ISO, 2001). To also have a look at possible food-chain effects, we included a two-species (predator–prey) reproduction test with mites feeding on springtails.
363
All tests were conducted using the same soil type (a sandy loam). 2. Materials and methods 2.1. Experimental soil A Danish agricultural soil type (Askov soil) was dried at 80 1C for 24 h and sieved through a 2 mm mesh. The soil had the following particle size distribution: coarse sand (200–2000 mm) 38.4%, fine sand (63–200 mm) 23.6%, coarse silt (20–63 mm) 10.0%, fine silt (2–20 mm) 12.3%, and clay (o2 mm) 13.0%. The organic matter content of the soil was 2.8%, and the total content of organic carbon was 1.6%. The soil pHH2 O and the total cation exchange capacity (CEC) was 6.2 and 8.14 meq.* 100 g1, respectively. The Askov soil was selected because it has previously been used for toxicity testing of several other PAHs with terrestrial organisms (see Sverdrup et al., 2002a–c).
2.2. Seeds and test organisms Seeds of Lolium perenne and Trifolium pratense were purchased from DLF Trifolium (Randers, Denmark), and Brassica alba seeds were purchased from Ove Rasmussens Planteskole (Silkeborg, Denmark). The seeds were inspected visually; undamaged seeds of similar size were selected for the experiments. Enchytraeus crypticus were mass-bred in petri dishes containing a Bacto agar, kept at 2071 1C with a 12/12-h photoperiod, and fed oatmeal. Test individuals of age 7–8 weeks were transferred from the culture disks to a petri dish with water and examined using a microscope. Sexually mature animals of similar size were selected for the experiments. Hypoaspis aculeifer were cultured in plastic cups where the bottom was covered by a regularly dampened charcoal/plaster of Paris mixture, kept at 2071 1C with a 12/12-h photoperiod, and fed adult F. fimetaria, which had been mass-bred at 2071 1C in petri dishes containing a regularly dampened charcoal/plaster of Paris mixture and fed bakers’ yeast. H. aculeifer of age 16–19 days were used for the experiments.
2.3. Sample preparation BaP (495% purity, Sigma–Aldrich, Copenhagen, Denmark) was dissolved in acetone (J.T. Barker; HPLC quality) in a stock solution corresponding to the highest test dosage. For each of the test concentrations used, dilutions from the stock solution were made using acetone. A volume of acetone stock solution in an amount corresponding to 20% of the soil dry weight was added to each concentration batch and mixed thoroughly in with the soil. Pure acetone was used for the control samples. The solvent was evaporated under a fume head for 24 h.
2.4. Terrestrial plant tests Tests were performed according to the method based on OECD Guidelines 208: ‘‘Terrestrial plants, Growth test’’ (OECD, 1984). The test concentrations were 1, 10, 100, and 500 mg kg1 soil dry weight (nominal values), in addition to solvent controls and controls without solvent. Four replicates were used per concentration. Following sample preparation and evaporation of solvent, 250 mL water was added per 1.25 kg dry wt, corresponding to 65% of the water-holding capacity, and mixed thoroughly in with the soil. The soil was then transferred to the test containers. These were transparent plastic cylinders (35 cm high, inner diameter 9.5 cm), closed at the top and bottom with lids during exposure. To allow for transpiration and gaseous exchange, the upper lid was perforated with one hole of diameter 0.8 mm. Five seeds were added to each replicate or test chamber, which was weighed to the nearest 0.1 g and placed in the greenhouse. The exposure period was 19 days, i.e., 14 days of exposure from the moment that 50% of
ARTICLE IN PRESS 364
L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
the seeds had emerged in the control samples. Test containers were weighed twice a week during exposure, and weight loss was replaced with water. At each time of watering, all replicates were rerandomized for the position that they occupied in the greenhouse. The minimum temperature was regulated and set to 15 1C (night temperature). During daytime, the temperature usually increased, sometimes to more than 25 1C. To provide a photoperiod of 16/8 h light/dark, we used cultivation lamps of the type Philips SON/T Plus (400 W), which emit light in the wavelength spectrum of 310–780 nm. The lamps were regulated to supply additional light if the light intensity provided from outside the greenhouse dropped below 5 klux; light was switched off when intensity rose above 25 klux. Three lamps were used to illuminate a test area of approximately 3 m2. The artificial light source gave an approximate photo flux density of 300 mE m2 s1 at the soil surface. At the end of the test, seedlings were counted and inspected for visible signs of injury, cut off, put separately in labeled paper bags, and dried at 40 1C for 24 h. The dry weight of each seedling was then determined to the nearest 0.1 mg. For the effect estimations, the mean weight of seedlings in each replicate was used.
blender after which the supernatant was used for inoculation. To each glass container containing 20 g of soil (with various concentrations of the test substance), 3.5 mL of the supernatant was added and mixed thoroughly in with the soil. The amount of water (inoculum) corresponded to 57% of the water-holding capacity. The inoculated soil (24 g) was then transferred to the test containers. Each concentration was tested in three replicates. The test was conducted at 2071 1C in total darkness, and the duration of the test was 28 days. At the end of the test, 1070.05-g soil samples were taken from each replicate. To each sample, 45 mL 1 M KCl was added, after which the resulting mixture was shaken in a mechanical shaker at 50 rpm for 60 min. Then samples were centrifuged (5000g, 10 min), and the supernatant was filtered through a GC filter. Particle-free extracts were then stored at 18 1C. Samples were thawed and placed at 5 1C the day before analysis of their nitrate content. Nitrate was measured spectrophotometrically using a LaChat QuikChems analyzer (LaChat Instruments, Milwaukee, WI, USA) using a proper amount of samples with known concentrations of NO 2 and NO3 and blank samples. All samples were measured in duplicate, and the mean nitrate content of each replicate was used for the statistical analysis.
2.5. Soil invertebrates
2.7. Chemical analysis
The soil invertebrate tests all had survival and reproduction as endpoints. The E. crypticus test was performed according to a recently published OECD Draft Guideline (OECD, 2000), and the H. aculeifer test was performed according to a procedure described by Krogh and Axelsen (1998). Following the sample preparation, water corresponding to 65% of the water-holding capacity was added and mixed thoroughly with the soil. The soil (24 g for E. crypticus and 60 g for H. aculeifer) was then transferred to the test containers. After 2 h, test organisms were added. For the E. crypticus test, 10 adults were added to each replicate, and for the H. aculeifer test, 10 females and five males along with 100 live adult F. fimetaria (food items) were added to each replicate. The invertebrate tests were conducted in temperature-regulated rooms (2071 1C) under a lighting of about 400–800 lux and a 12/12-h photoperiod. In the invertebrate tests, seven test concentrations were used (1, 3, 10, 30, 100, 300, and 1000 mg kg1 dry wt), in addition to the controls. Four replicates were used per concentration. During incubation, the test containers were closed at the top and bottom with lids. The exposure period was 21 days, and test containers were opened after 7 and 14 days for feeding (only E. crypticus) and, in case of water loss, addition of water. At the end of the test period, the H. aculeifer and F. fimetaria were extracted from the test soil using a controlled temperature gradient extractor based on the principles of MacFadyen (1961) and Petersen (1978). E. crypticus were extracted by dividing each sample over two plastic containers and adding water (approximately eight times the amount of soil). The plastic containers were closed with lids, shaken vigorously for 5 s, and then left to sediment for 24 h, individuals of E. crypticus were then found on the surface of the settled soil particles. Mites (H. aculeifer) and remaining food items (i.e., F. fimetaria) were counted manually under a microscope. Enchytraeids were also counted manually using a microscope. In all tests, juveniles produced during exposure could easily be distinguished from the adult individuals originally added to the test chambers.
Verification of exposure concentrations was performed for one of the invertebrate tests, one of the plant tests, and the soil nitrification test. Soil samples were collected from the 0-, 10-, 100-, and 1000-mg kg1 exposure concentrations for the invertebrate test, the 10- and 1000-mg kg1 concentrations for the nitrification test, and the 1-, 10-, 100-, and 500mg kg1 concentrations for the plant tests at the start of the tests. To 100mg (high BaP content) or 1000-mg (low BaP content) soil samples, an equal amount of sodium sulfate was added. Known amounts of the internal standard (d10-BaP) were added to the soil–sodium sulfate mixture. The soils were extracted ultrasonically six times with a 50:50% mixture of dichloromethane (Merck, p.a.) and acetone (Merck; p.a.). Each extraction lasted 30 min. The combined extracts were concentrated to 1.5 mL by evaporation of solvents. Identification and quantification of compounds were carried out with a gas chromatography–mass spectrometric system consisting of a Varian 3400 Star GC and a Varian Saturn III ion trap MS, using temperatureprogrammable splitless injection and a 30-m XTI-5 fused-silica capillary column coated with 95% methyl-siloxane/5% phenyl-siloxane (Restek Corp., Bellefonte, PA, USA).
2.6. Soil-nitrifying bacteria The test was performed using a procedure similar to the method proposed by the International Standardization Organization (ISO, 1997), using test concentrations of 1, 3, 10, 30, 100, 300, 1000, and 3000 mg kg1. However, the use of a high-temperature-treated soil and the use of a solvent in the sample preparation had probably eliminated the nitrifying bacteria originally present in the Askov soil. Therefore, a new microbial community was added to the soil in the form of an inoculum. The inoculum was prepared using fresh Askov soil (stored at 5 1C for 24 h prior to use) and followed a procedure described by Lindahl and Bakken (1995). Essentially, 20 g soil+180 mL deionized water were mixed with a kitchen
2.8. Statistical analysis of results Toxic effect levels were determined on the basis of measured initial concentrations. No observable effect concentrations (NOEC values) and LOEC values were assessed using ANOVA and Dunnett’s procedure (on a 5% significance level) in SAS/STAT (SAS Institute, 1999), after testing for fulfilment of ANOVA requirements using SAS/LAB (SAS Institute, 1992). Estimation of the concentration causing a 10% reduction of growth, reproduction, or nitrate production (EC10 values) was done by linear interpolation, and the confidence intervals were calculated by the ICp method (Norberg-King, 1993).
3. Results The chemical analysis showed that measured start concentrations were close to nominal (Table 1). The response in the solvent control and the control without solvent were not significantly different for any of the tests, and replicates from the two controls were therefore pooled. All tests could be considered valid according to the validity criteria defined in the respective test guidelines (i.e., with respect to seed emergence in the plant tests (OECD, 1984), survival and reproduction criteria for the enchytraeid tests
ARTICLE IN PRESS L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368 Table 1 Results of the chemical analysis of benzo(a)pyrene in soils used in toxicity tests with different soil organisms
Table 2 Overview of the toxicity data generated for benzo(a)pyrene Test (organism)
Test(s)
Soil invertebrates
Terrestrial plants
Nitrifying bacteria
Nominal concentration (mg kg1 dry wt)
Measured concentration (mg kg1 dry wt)
0 10 100 1000 1 10 100 500 10 1000
o0.1 7.7 106 947 0.73 8.0 86 470 7.7 977
Soil samples were collected at the beginning of the tests.
365
Soil nitrification Brassica alba (mustard) Trifolium pratense (red clover) Lolium perenne (ryegrass) Enchytraeus crypticus (Enchytraeidae) Hypoaspis aculeifer (Acari)
Growth/reproduction/nitrate productiona NOEC
LOEC
293b 86 4470 4470 4947
977 470 — — —
4947
—
This substance did not significantly affect the seed emergence of plants or the survival or reproduction of three invertebrate species. For soilnitrifying bacteria and growth of the plants, the no observable effect concentration (NOEC) and the lowest observable effect concentration (LOEC) are given. a Soil effect concentrations (mg kg1 dry wt) based on measured initial concentrations. b Estimated from the measured concentrations at the 100- and 1000mg kg1 nominal treatment levels.
observed, but there was a trend toward a reduction in growth with increasing BaP concentrations for all three species (Fig. 2). However, only B. alba was significantly (ANOVA; Dunnett’s; Po0:05) affected at the highest concentration tested (LOEC ¼ 470 mg kg1 dry soil). Estimated NOEC values for the plant tests are included in Table 2. For B. alba, we also observed a significant increase in seedling growth at the 10-mg/kg exposure level (Fig. 2). For soil nitrification, the dose–response relationship was nearly linear (Fig. 3), and the nitrification process was significantly (ANOVA; Dunnett’s; Po0:05) inhibited at the two highest test concentrations, leaving a NOEC value of 293 mg kg1 dry soil. Summary statistics for the nitrification test is also included in Table 2. 4. Discussion
Fig. 1. Dose–response relationship for effects of benzo(a)pyrene on the survival (mean values) and reproduction (mean value7SD; n ¼ 4) of (A) Enchytraeus crypticus and (B) Hypoaspis aculeifer.
(OECD, 2000), and water content of test containers in the nitrification test (ISO, 1997)). For the soil invertebrates, no significant effect (ANOVA; Dunnett’s; P40:05) of BaP could be detected with respect to survival or reproduction of the two species (Fig. 1 and Table 2). For the plants, there was no effect on seed emergence and no change in leaf color or form was
Neither the organisms in the single-species tests (i.e., plants, springtails, and soil-nitrifying bacteria) nor the mite predator in the two-species test were particularly sensitive to BaP. The absence of strong effects made it difficult to draw conclusions on the relative sensitivity of the tested species to BaP, but plants and bacteria seem to be slightly more sensitive than soil invertebrates (Table 2). We also note that at the 10-mg/kg exposure level, the most sensitive plant (B. alba) exhibited increased growth, which can possibly be explained as an indication of hormesis, an overcompensation in response to low levels of contaminants (Calabrese, 2005). The results presented here are in accordance with the relatively low toxicity reported for BaP in most previous studies on soil organisms (an overview is given in Table 3). However, there is a study by Achazi et al. (1995) showing quite deviating results, with significant toxicity of BaP to enchytraeids (E. crypticus) and earthworms (Eisenia fetida)
ARTICLE IN PRESS 366
L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
Fig. 2. Dose–response relationship for effects of benzo(a)pyrene on the growth of terrestrial plants (error bars indicate SD; n ¼ 4). Results are shown for (A) Brassica alba (mustard), (B) Lolium perenne (ryegrass); and (C) Trifolium pratense (red clover). The asterisk (*) means significantly different from control (ANOVA, Dunnett’s test, Po0:05).
Fig. 3. Dose–response relationship for effects of benzo(a)pyrene on soil nitrate production (mean value7SD; n ¼ 3).
observed even at soil concentrations of 10 mg kg1. The soil type used in the Achazi study is similar to those used in the present study and in the study of Bleeker et al. (2003) with
regard to sorption capacity (organic matter content of 3.9% vs. 2.8% and 2.3% in the latter studies), and the large difference in observed toxicity is therefore not due to the soil type. As pointed out by Jensen and Sverdrup (2003), the data generated by Achazi et al. (1995) are currently used as a reference for setting soil quality standards for BaP in many countries (see, e.g., Kalf et al. (1997)), despite the fact that they deviate from the remaining data pool on this substance. As neither Bleeker et al. (2003) nor we were able to reproduce the high sensitivity found for E. crypticus in the Achazi study, we suggest that both this and the E. fetida test should be repeated before these data are used for setting soil quality standards for BaP. Based on the data generated by Achazi et al. (1995), Kalf et al. (1997) has previously calculated a ‘‘maximum permissible concentration’’ of 0.26 mg/kg for BaP. This value was calculated in the same way as the predicted no effect concentration (PNEC value), which is used by the European Union (European Commission, 1996). By using the data presented here, a PNEC value of 8.6 mg/kg can be calculated based on the lowest NOEC value (86 mg/kg for B. alba) and an assessment factor of 10 according to the Technical Guidance Document (European Commission, 1996). There may be several reasons that BaP shows a low toxicity in this and previous studies with soil organisms. Some reasons relate to exposure, i.e., that the substance might be strongly sorbed to soil particles, thereby limiting exposure, or that pore water concentrations remain low even at high soil concentrations due to a saturation at BaP solubility. With regard to sorption, the soil that we used has relatively low organic carbon content (1.6%) and is therefore a relatively conservative soil type with regard to sorption capacity for organic contaminants. A number of polycyclic aromatic compounds have previously been tested in this soil type, and they seem to fall into one of two groups; either they have typical threshold values for toxicity in the range of 10–40 mg kg1 (Sverdrup et al., 2002a–c, 2003) or they are not toxic even at soil exposure concentrations of more than 500 mg kg1 (Sverdrup et al., 2002a; Bleeker et al., 2003). The latter group contains the very lipophilic PAHs (log Kow45.5) containing four or more rings. For aquatic organisms, it has long been known that some very hydrophobic organic substances do not elicit acute toxicity to fish (Ko¨nemann and Musch, 1981), and this has been explained by their very low water solubility in combination with a limited bioaccumulation potential (i.e., they are certainly accumulated to high tissue concentrations compared to concentrations in the surrounding water, but the water concentrations are still too low to obtain a critical internal concentration and thereby elicit toxicity). Based on this knowledge and our toxicity test results for springtails (Sverdrup et al., 2002a), we previously suggested that very lipophilic PAHs might not be toxic to soil organisms exposed through soil pore water because of their limited water solubility. The low toxicity of BaP found in this study suggests that the results found for springtails might also be valid for a range of other soil
ARTICLE IN PRESS L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
367
Table 3 Overview of literature toxicity data for effects of benzo(a)pyrene on soil organisms Test organismsa
Endpoint
Test duration (days)
Exposure matrix
LOEC (mg kg1)
Enchytraeus crypticus (Oligochaeta)1 Enchytraeus crypticus (Oligochaeta)1 Enchytraeus crypticus (Oligochaeta)1 Enchytraeus crypticus (Oligochaeta)2 Eisenia fetida (Oligochaeta)1 Eisenia fetida (Oligochaeta)1 Eisenia fetida (Oligochaeta)3 Eisenia fetida (Oligochaeta)4 Oniscus asellus (Isopoda)5 Porcellio scaber (Isopoda)5 Oniscus asellus (Isopoda)6 Oniscus asellus (Isopoda)6 Porcellio scaber (Isopoda)6 Folsomia fimetaria (Collembola)7 Folsomia candida (Collembola)2 Soil bacteria8 Soil bacteria and fungi9 Raphanus sativa (radish)4 Lactuca sativa (lettuce)4 Secale cereale (rye)10
Cocoon fertility Cocoon fertility Reproduction Reproduction Cocoon prod. Survival Growth Survival Growth Growth Growth Reproduction Growth Reproduction Reproduction Respiration Community structure Seed emergence Seed emergence Emergence/growth
4 7 30 28 28 28 28 28 63 63 329 329 112 21 28 10 4100 14 14
Agar Food (oats) Soil, 3.9% OMb Soil, 2.3% OCc Soil, 3.9% OM Soil, 3.9% OM Soil OECD soil Food (490% OM) Food (490% OM) Food (490% OM) Food (490% OM) Food (490% OM) Soil, 2.8% OM Soil, 2.3% OC Soil Two different soils OECD soil OECD soil Soil
1 75 100 4931 1 10 4100 448,000 100 100 4316 31.6d 4316 4840 4931 410 433 423,800 11,900 43.3
a
Superscript numbers refer to the following references: 1Achazi et al. (1995); 2Bleeker et al. (2003); 3Eason et al. (1999); 4CCME, (1997); 5Van Brummelen and Stuijfzand (1993); 6Van Brummelen et al. (1996); 7Sverdrup et al. (2002a); 8Eschenbach et al. (1991); 9Park et al. (1990); 10Do¨rr (1970). b OM ¼ organic matter. c OC ¼ organic carbon. d The effect observed was an increase in reproduction.
living species, and the fact that toxicity to all our test organisms was low also supports the assumption that narcosis is the mode of action for this substance. However, there are several potential aspects of BaP toxicity to soil organisms that are not covered by this study. First, all tests have been performed using a relatively short exposure period (about 3 weeks). Both Van Brummelen et al. (1996) and Saint-Denis et al. (2000) found DNA adducts in soil animals (isopods and earthworms, respectively) exposed to BaP, suggesting that BaP undergoes biochemical activation in these animals. Similarly, DNA adducts have also been found in plants (PfohlLeszkowicz et al., 1996). The accumulated effects from DNA damage are of particular importance for organisms suffering long-term exposure, and the effect of such exposure can probably be fully explored only by using multigeneration tests. Second, the fact that BaP usually occurs in mixtures along with other polycyclic aromatic compounds makes it difficult to consider the toxicity of this substance alone. For substances acting by a narcotic mode of toxic action, the summed effect from exposure to various compounds can be characterized by concentration-addition (toxic units approach). Thus even low pore water concentrations of BaP may contribute to the toxicity of a PAH-contaminated sample, and, for marine sediments, methods to deal with this problem have already been developed (e.g., Swartz et al. (1995)). Third, there might still be organisms that are more sensitive than those already
tested. Finally, there is still a possibility that some species might accumulate BaP in excess of their pore water equilibrium concentrations due to different exposure routes or longer exposure periods. In conclusion, our data indicate that, for many different soil organisms, BaP shows a lower rather than a higher toxicity than less-lipophilic PAHs that have previously been tested in the same soil type. We attribute the low toxicity to the low water solubility of this substance. With regard to further work on BaP, several issues still seem unresolved. These include the potential contribution of BaP to the toxicity of PAH mixtures in soils and the ecological significance of BaP and other genotoxic substances in the environment. Acknowledgments This study was financed by a grant from the Norwegian Research Council and Hustadmarmor. We thank Zdenek Gavor and Elin Jørgensen for excellent laboratory assistance and Axel E. Kelley for performing the chemical analysis. References Achazi, R.K., Chroszcz, G., Du¨ker, C., Henneken, M., Rothe, B., Schaub, K., Steudel, I., 1995. The effect of fluoranthene (Fla), benzo(a)pyrene (BaP) and cadmium (Cd) upon survival rate and life cycle parameters
ARTICLE IN PRESS 368
L.E. Sverdrup et al. / Ecotoxicology and Environmental Safety 66 (2007) 362–368
of two terrestrial annelids in laboratory tests systems. Newslett. Enchytraeidae 4, 7–14. Bleeker, E.A.J., Wiegman, S., Droge, S.T.J., Kraak, M.H.S., Van Gestel, C.A.M., 2003. Towards an improvement of the risk assessment of polycyclic (hetero)aromatic hydrocarbons. Report 2003–04 of the Institute of Ecological Science, Vrije Universiteit, Amsterdam, The Netherlands. Calabrese, E.J., 2005. Paradigm lost, paradigm found: the re-emergence of hormesis as a fundamental dose response model in the toxicological sciences. Environ. Pollut. 138, 378–411. CCME, 1997. Recommended Canadian Soil Quality Criteria. Report from the Canadian Council of Ministers of the Environment (CCME), ISBN: 1-895-925-92-4, pp 1–185. Do¨rr, V.R., 1970. Die Aufnahme von 3,4-Benzpyren durch Pflanzenwurzeln. Landwirtsch. Forsch. 23, 371–379. Eason, C.T., Svendsen, C., O’Halloran, K., Weeks, J.M., 1999. An assessment of the lysosomal neutral red retention test and immune function assay in earthworms (Eisenia andrei) following exposure to chlorpyrifos, benzo-a-pyrene (BaP), and contaminated soil. Pedobiologia 43, 641–645. El-Fouly, M.M., 1980. Effect of low concentrations of 3,4-benzpyrene on growth and N-fractions of seedlings. Landwirtsch. Forsch. 33, 108–117. Eschenbach, A., Gehlen, P., Bierl, R., 1991. Untersuchungen zum Einfluss von Fluoranthen und Benzo(a)pyren auf Bodenmikroorganismen and zum Mikrobiellen Abbau dieser Substanzen. Mitteilungen d. Dt. Bodenkundl. Ges. 63, 91–94. European Commission, 1996. Technical guidance document in support of commission directive 93/67/EEC on risk assessment for new notified substances and commission regulation (EC) no 1488/94 on risk assessment for existing substances, Luxembourg. ISO, 1997. Soil quality—biological methods—determination of nitrogen mineralization and nitrification in soils and the influence of chemicals on these processes. ISO Method 14238. International Organization for Standardization, Geneva, Switzerland. ISO, 2001. Soil quality–guidance on the ecotoxicological characterization of soils and soil materials. ISO/DIS 15799. Draft, March 2001. International Organization for Standardization, Geneva, Switzerland. Jensen, J., Sverdrup, L.E., 2003. Polycyclic aromatic hydrocarbon (PAHs) ecotoxicity data for developing soil quality standards. Rev. Environ. Contam. Toxicol. 179, 71–95. Kalf, D.K., Crommentuijn, T., van de Plassche, E.J., 1997. Environmental quality objectives for 10. Ecotoxicol. Environ. Saf. 36, 89–97. Kawamura, K., Suzuki, I., Fujii, Y., Watanabe, O., 1994. Ice core record of polycyclic aromatic hydrocarbons over the past 400 years. Naturwissenschaften 81, 502–505. Ko¨nemann, H., Musch, A., 1981. Quantitative structure-activity relationships in fish toxicity studies. Part 1: relationship for 50 industrial pollutants. Toxicology 19, 209–221. Krogh, P.H., Axelsen, J.A.A., 1998. Test on the predatory mite Hypoaspis aculeifer preying on the collembolan Folsomia fimetaria. In: Løkke, H., Van Gestel, C.A.M. (Eds.), Handbook of Soil Invertebrate Toxicity Tests. Wiley, Baffins, West Sussex, England, pp. 239–251. Lindahl, V., Bakken, L.R., 1995. Evaluation of methods for extraction of bacteria from soil. FEMS Microbiol. Ecol. 16, 135–142. MacFadyen, A., 1961. Improved funnel-type extractors for soil arthropods. J. Anim. Ecol. 30, 171–184. Norberg-King, T.J., 1993. A Linear Interpolation Method for Sublethal Toxicity: The Inhibition Concentration (ICp) Approach (Version 2.0), July 1993. National Effluent Toxicity Assessment Center, SAS Institute, Duluth, Minnesota.
OECD, 1984. Terrestrial plants, growth test. OECD Guideline 208, Organization for Economic Cooperation and Development, Paris, France. OECD, 2000. Enchytraeidae reproduction test. OECD Guidelines 220, Draft guideline, March 2000, Organization for Economic Cooperation and Development, Paris, France. Park, K.S., Sims, R.C., Dupont, R.R., Doucette, W.J., Matthews, J.E., 1990. Fate of PAH compounds in two soil types: influence of volatilization, abiotic loss and biological activity. Environ. Toxicol. Chem. 9, 187–195. Petersen, H., 1978. Some properties of two high gradient extractors for soil microarthropods. Natura Jutlandica 20, 95–121. Pfohl-Leszkowicz, A., Masfaraud, J.F., Weber-Lotfi, F., Marty, J., Guillemaut, P., Castegnaro, M., 1996. Comparison of animal and plant DNA-adducts after exposure to benzo(a)pyrene. Polycycl. Aromat. Compounds 10, 323–331. Saint-Denis, M., Pfohl-Leszkowicz, A., Narbonne, J.F., Ribera, D., 2000. Dose–response and kinetics of the formation of DNA adducts in the earthworm Eisenia fetida andrei exposed to B(a)P-contaminated artificial soil. Polycycl. Aromat. Compounds 18, 117–127. SAS Institute Inc., 1992. SAS/LABs Software: User’s Guide, Version 6, first ed. SAS Institute Inc., Cary, NC. SAS Institute Inc., 1999. SAS/STATs User’s Guide, Version 8. SAS OnlineDocs, SAS Institute Inc., Cary, NC. Sverdrup, L.E., Nielsen, T., Krogh, P.H., 2002a. Soil ecotoxicity of polycyclic aromatic hydrocarbons in relation to soil sorption, lipophilicity, and water solubility. Environ. Sci. Technol. 36, 2429–2435. Sverdrup, L.E., Ekelund, F., Krogh, P.H., Nielsen, T., Johnsen, K., 2002b. Soil microbial toxicity of eight polycyclic aromatic compounds: effects on nitrification, the genetic diversity of bacteria and the total number of protozoans. Environ. Toxicol. Chem. 21, 1644–1650. Sverdrup, L.E., Krogh, P.H., Nielsen, T., Stenersen, J., 2002c. Relative sensitivity of three terrestrial invertebrate tests to polycyclic aromatic compounds. Environ. Toxicol. Chem. 21, 1927–1933. Sverdrup, L.E., Krogh, P.H., Nielsen, T., Kjær, C., Stenersen, J., 2003. Toxicity of eight polycyclic aromatic compounds to the red clover (Trifolium pratense), ryegrass (Lolium perenne) and mustard (Sinapsis alba). Chemosphere 53, 993–1003. Swartz, R.C., Schults, D.W., Ozretich, R.J., Lamberson, J.O., Cole, F.A., P DeWitt, T.H., Redmond, M.S., Ferraro, S.P., 1995. PAH: A model to predict the toxicity of polynuclear aromatic hydrocarbon mixtures in field-collected sediments. Environ. Toxicol. Chem. 14, 1977–1987. Timbrell, J.A., 2000. Principles of Biochemical Toxicology, third ed. Taylor & Francis Ltd., London, UK. Van Brummelen, T.C., Stuijfzand, S.C., 1993. Effects of benzo(a)pyrene on survival, growth and energy reserves in the terrestrial isopods Oniscus asellus and Porcellio scaber. Sci. Total Environ. Suppl., 921–930. Van Brummelen, T.C., Van Gestel, C.A.M., Verweij, R.A., 1996. Longterm toxicity of five polycyclic aromatic hydrocarbons for the terrestrial isopods Oniscus asellus and Porcellio scaber. Environ. Toxicol. Chem. 15, 1199–1210. Van Straalen, N.M., Verweij, R.A., 1991. Effects of benzo(a)pyrene on food assimilation and growth efficiency in Porcellio scaber (Isopoda). Bull. Environ. Contam. Toxicol. 46, 134–140. Wild, S.R., Jones, K.C., 1995. Polynuclear aromatic hydrocarbons in the United Kingdom environment: a preliminary source inventory and budget. Environ. Pollut. 88, 91–108.