Environ. Sci. Technol. 2001, 35, 3109-3112
Alkylphenolic Compounds in Edible Molluscs of the Adriatic Sea (Italy) F U L V I O F E R R A R A , * ,† F A B I O F A B I E T T I , † MIRELLA DELISE,† ADRIANA PICCIOLI BOCCA,† AND ENZO FUNARI§ Food Department and Environmental Health Department, Istituto Superiore di Sanita`, Viale Regina Elena, 299, 00161 Rome, Italy
This paper reports the first group of results on alkylphenol (APE) contamination of seafood in the Adriatic Sea, in the framework of a national project on the quality of this Sea (PRISMA 2). Nonylphenol (NP), octylphenol (OP), and their ethoxylates (NPE and OPE) were detected in edible molluscs, either filter feeders or predators (clams, mussels, cuttlefishes, and squids), caught from 15 harbors along the Italian coast in the Adriatic Sea in 1997. NP was the compound found always at levels much higher than the other APEs in all the examined species. It reached the maximum concentration of 696 ng/g fresh weight in the squids from the central Adriatic Sea. OP generally occurred at levels 30 times lower than NP. OP was found up to a level of 18.6 ng/g in squids from central Adriatic Sea. OPE was the compound always spotted at the lowest concentrations, up to 0.43 ng/g. NPE was always below the detection limit. The pattern of contamination in the three areas examined was different between bivalve and cephalopod species. No exhaustive risk assessment for marine organisms and human health can be conducted on the basis of these results because data are insufficient. Yet, the occurrence of NP suggests a negligible risk for mussels, which represent the only molluscs for which data are adequate. As to the possible human health implications, the consumption of molluscs of the Adriatic Sea implies APE intakes that are some orders of magnitude lower than those responsible for toxic effects in laboratory animals. Despite these apparently low risks for mussels and human health, the reasons for concern still remain because the levels of alkylphenols found in this study indicate a general contamination of the Adriatic Sea even far from the cost. Furthermore, these levels might represent an unacceptable hazard for other marine organisms. Finally, they contribute to the general environmental estrogen pool.
Introduction Nonylphenol (NP) and octylphenol (OP) are important intermediates in the production of their polyethoxylate surfactants (NPEs and OPEs and in general alkylphenolss APEs). The latter are compounds consisting of alkyl chains attached to a phenol ring and combined with a variable number of ethylene oxides. APEs have a wide variety of * Corresponding author phone: (0039) 06 49902046; fax: (0039) 06 49902377; e-mail:
[email protected]. † Food Department. § Environmental Health Department. 10.1021/es010508h CCC: $20.00 Published on Web 07/06/2001
2001 American Chemical Society
industrial, agricultural, and household applications (1, 2). NP is by far the most important alkylphenol commercially. In the European Union, a NPE production of 109 808 tonnes has been reported in 1994 (3). In the environment and during the aerobic treatment of wastewater, NPEs and OPEs as well as other alkylphenol polyethoxylates decompose into compounds with shorter ethylene oxide groups up to the corresponding alkylphenols (4-6). NP and OP are lipophilic compounds (Kow of 4.0 and 4.6, respectively) (7), which are quite resistant to environmental degradation (8, 9). They are considered ubiquitous contaminants of the aquatic environment (10-13, 9, 14), where they tend to associate with particulate matter and ultimately with sediments. In a study in which the examined organisms were exposed to two concentrations of NP, bioconcentration factors of 90-110, 2740-4120, and 12001300 were spotted in shrimps (Crangon crangon), mussels (Mytilus edulis), and fishes (Gasterosterus aculeatus), respectively (15). APEs are widely recognized as potentially estrogenic compounds (16-18, 13, 9, 19-25, 14). The presence of alkylphenols in the environment has been considered a cause for concern at least for two reasons (26). The first is due to a study in which NP induced in vitro proliferation of MCF-7 cells, a human breast cancer cell line (10). The results of this study gave rise to many speculations on the possible role of environmental estrogens on a variety of human disorders (27). The second reason is the contribution of alkylphenols to the environmental estrogen pool. Yet, the available data on the possible environmental and human health effects associated with the environmental occurrence of NP, OP, and their ethoxylates are limited. NP and OP bind to the estrogen receptor and have estrogenic effects in various in vitro assays (28, 10, 11, 16). They caused an increase of plasma vitellogenin and a significant decrease in the testis growth rate in fish at concentrations of 10 µg/L and 3 µg/L, respectively (17, 29). NP did not affect fertilization or early development after 72 h of exposure at 200 µg/L; sublethal effects, such as decreased byssus strength and change of scope for growth, were observed at 56 µg/L (30). NP is more toxic to aquatic organisms than its ethoxylates (31, 9). In 2 mg/L NP10E, mussel embryos (Mytilus edulis) never developed beyond the blastula stage and in 1 mg/L never beyond the veliger larval stage (32). As to human health implications, epidemiological studies are not available, and toxicological ones are limited. In an oral subchronic toxicity (90 days) study in rats, NP did not cause any effect at 50 mg/kg bw/day (33). In a two-generation reproduction study, OP was administered to rats with the diet up to 2000 ppm (roughly equivalent to at least 100 mg/ kg bw) (2). No effects on reproductive parameters and no estrogen-like effects were observed. The NOAELs for systemic and postnatal toxicity were 200 ppm (roughly equivalent to at least 10 mg/kg bw) and at or above 2000 ppm for reproductive toxicity. NP and NPEs gave negative results in a range of genotoxicity assays. Nevertheless, NP9E was positive in a cell transformation study with mammalian cell lines (34). In the same study, the results of these genotoxicity assays on two OPEs were negative. This paper reports the first group of results on alkylphenol contamination of seafood in the Adriatic Sea, in the framework of a national project on the quality of this Sea (PRISMA 2). NP, OP, and ethoxylates, NPE and OPE, were spotted in four species of edible molluscs, two cephalopods, and two bivalves (three only in the northern Adriatic Sea), caught VOL. 35, NO. 15, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3109
from 15 harbors along the Italian coast in the Adriatic Sea in 1998. These species were chosen because they represent the most important commercial molluscs in Italy. The two cephalopod species are predators with different predacious habits; squids are typically pelagic, whereas cuttlefishes live closer to the bottom. The bivalves are filter feeders; clams colonize sandy bottoms, whereas mussels rocky bottoms. The Adriatic Sea is a semiclosed sea with a mean depth of just 40 m in the north and about 1000 m in the south; it has an estimated water exchange time of around 10 years (35). The three areas, in which this Sea is commonly subdivided, show contamination and ecological differences. The northern Adriatic Sea is characterized by a low depth and salinity, sandy bottoms, and the presence of big rivers whose catchment basins cross big cities and highly industrialized and agricultural areas. Eighty-five percent of the total emission of inner water into the Adriatic Sea occurs in this area. The southern Adriatic Sea has oceanic depths, a very good water exchange with the Ionic Sea, higher salinity, rocky bottoms, and few rivers with smaller catchment basins. In this area anthropic activities have a lower impact on the environment. The central Adriatic Sea has intermediate features. This activity has been partially supported with funds of the Italian Ministry of University, Research and Technology in the framework of a national project on the Adriatic Sea (PRISMA 2).
Materials and Methods Chemicals. All the standards were purchased from the Sigma Aldrich (Germany). The purity degree of OP was 97%. NP, NPE, and OPE were of technical grade purity. All the organic solvents used were of HPLC grade (acetone - Baker, Holland; n-hexane - Carlo Erba Reagenti, Italy; diethyl ether - Carlo Erba Reagenti, Italy; acetonitrile - Merk, Germany). Species, Study Sites, and Sampling. The selected species were as follows: the pelagic cephalopod Loligo vulgaris (squid), the demersal cephalopod Sepia officinalis (cuttlefish), and the three bivalves Chamelea gallina (clams) and Mytilus galloprovincialis (mussel). In the northern Adriatic Sea, another species of clam, Ruditapes decussatus, was sampled together with Chamelea gallina, because it is common in this area and lives in similar habitats. Within each of the three areas of the Adriatic Sea, 3-4 commercial ports were selected for sampling (Figure 1). Approximately 2 kg for each species was sampled from each site at the beginning of the project, i.e., by the end of 1997. Once collected, the samples were frozen on dry ice and transported to the laboratory, where they were stored at -70 °C until sample preparation. The molluscs were bought directly from the boats after having interviewed the fishermen and having been ensured that the species were caught in the area of interest for this study. Sample Preparation. Samples were prepared in order to obtain their edible parts. After eliminating the internal organs and skin, squids and cuttlefishes were homogenized after adding of 20% of ultrapure water. The bivalves were opened and rinsed with distilled water to eliminate sand and other impurities; the adduction muscles were cut, and the mussels were collected on a sieve, where they dripped for 5-6 min. The pool of each area was formed by 100 g of homogenized sample from each port. The pools were then stored at -40 °C until the analyses. Analysis and Instrumentation. Each pool was freezedried first. In an aliquot of 1-1.5 g of freeze-dried sample, the fat was extracted according to a published procedure (36). The resulting fat was dissolved in 2 mL of n-hexane (HPLC grade), and APEs were extracted according a published 3110
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 15, 2001
FIGURE 1. Sampling sites along the Adriatic Italian coast. method (37). Finally, the extracts were analyzed by capillary gas chromatography-mass spectrometry (GC/MS) in selected ion monitoring mode (SIM). The instrument used was a GC/MS HP6890 (EI 70 eV) coupled with the computer data system HP Chemstation. The GC column, a 15 m Rtx-5 (0.25 mm i.d.; 5% biphenyl, 95% dimethylpolysiloxane), was connected directly to the source kept at 280 °C. The sample (1 µL) was injected in splitless mode at a temperature of 300 °C. The temperature program was as follows: 80 °C (3 min)15 °C/min up to 210 °C-5 °C/min up to 300 °C (30 min). Standards were analyzed to determine fragmentation patterns and principal ions that could be used for selected ion monitoring GC/MS analysis. Fragment ions were chosen according to the most abundant ions in each oligomer. These ions were m/z 107, 135 (OP); 135, 179 (OPE); 121, 135, 149 (NP); 107, 179, 193 (NP1E); 107, 223, 237 (NP2E); 135, 267, 281 (NP3E); 89, 135, 311 (NP4E); and 92, 246 (dodecylbenzene). The quantification was made from GC/MS analysis by rationing the analyte peak areas, using a response factor measured from the standard mixture. For quantification using the internal standard (DB), the quantifiable limit of detection was taken as the minimum amount of standard required to give 95% of signal reproducibility or 50 times the baseline variation. Extraction Recoveries and Reproducibility. The recoveries (percentage of standard added to sample recovered during extraction) and reproducibility (relative standard deviation for triplicate analysis) for the extraction step of the method were determined by adding to the aliquot of freeze-dried sample (1-1.5 g) 200 µL of each standard solution OP 148, NP 106.8, OPE 106.8, NPE 1100, and DB 85.6 ppm. The extracts were analyzed by GC/MS in scan mode. The results of triplicate extraction showed an efficiency of recovery of 94, 79, 91, and 71% for OP, OPE, NP, and NPE, respectively.
Results and Discussion Table 1 reports the number of specimens and the morphological features of the animals examined in the analyzed pools. The table shows that cuttlefish and mussel specimens were
TABLE 1. Pool Composition: Number and Morphological Features of the Examined Animals
squid cuttlefish mussel clam
a
north center south north center south north center south northa center south
no. of animals
length (cm) mean ( SD
weight (g) mean ( SD
20 28 26 11 18 13 61
19.1 ( 2.8 15.2 ( 3.0 13.2 ( 1.4 12.4 ( 1.6 12.6 ( 1.5 12.3 ( 1.7 5.8 ( 0.7 nsb 4.7 ( 1.0 3.1 ( 0.8 4.3 ( 0.8 2.4 ( 0.8
217 ( 16 142 ( 15 104 ( 5 243 ( 61 242 ( 78 239 ( 84 9.1 ( 0.4 nsb 7.2 ( 1.1 10.9 ( 6.3 7.9 ( 3.2 4.9 ( 1.6
62 204 241 81
Both Ruditapes decussatus and Chamalea gallina. b ns: not sampled.
TABLE 2. Fat Content of the Examined Molluscs squid cuttlefish mussel clam
a
area
fat contenta (mg/g fw) mean ( SD
north center south north center south north center south north center south
4.3 ( 0.1 4.3 ( 0.5 4.0 ( 0.8 2.2 ( 0.1 2.4 ( 0.1 2.0 ( 0.1 7.4 ( 0.6 nsb 6.7 ( 0.01 1.0 ( 0.2 1.1 ( 0.5 0.8 ( 0.1
Triplicate extraction.
b
ns: not sampled.
TABLE 3. Levels of Alkylphenols (ng/g fw) in Molluscsa
squids cuttlefishes mussels clams
a
Edible part.
area
OP mean ( SD
OPE mean ( SD
NP mean ( SD
north center south north center south north center south north center south
11.5 ( 5.1 18.6 ( 2.9 3.9 ( 1.2b 3.6 ( 0.4 3.8 ( 0.3 3.8 ( 0.4 4.4 ( 0.6 nsd 4.9 ( 2.1 2.7 ( 0.5 2.7 ( 0.5 2.8 ( 0.2
0.18 ( 0.02 0.43 ( 0.19b 0.21 ( 0.09 0.14 ( 0.05 0.12 ( 0.02 0.11 ( 0.03 ndc nsd ndc 0.08 ( 0.09b 0.18( 0.001 0.16 ( 0.02
453 ( 75 696 ( 18b 389 ( 52 67 ( 15b 566 ( 70b 87 ( 13b 254 ( 41 nsd 265 ( 70 243 ( 1 252 ( 6 250 ( 12
b
P < 0.05. c nd: < detection limit. d ns: not sampled.
quite homogeneous in size and weight in the three examined areas, whereas squid and clam samples were more heterogeneous. Yet, the variability observed was within a factor of 2. Table 2 shows that the fat content in each mussel species was substantially homogeneous, regardless of their respective fishing area, size, or weight. The concentrations of the examined APEs are shown in Table 3, with the exception of the examined NPEs, whose occurrence was always below the detection limit. NP was the compound found at the highest levels, generally 30 times higher than those of OP. OPE was the compound always detected at the lowest concentrations. NP and OP reached its maximum levels in squids. Squids showed the highest values of APE contamination (P < 0.05) for all the compounds and in all the three areas
compared with the other mollusc species. Within squids, the mean APE concentrations were significantly higher in the central Adriatic Sea (P < 0.05) compared with the north and south. APE levels in cuttlefishes were significantly lower than in squids (P < 0.05). As in squids, NP concentrations peaked in the central Adriatic Sea (P < 0.05). OP and OPE did not show significantly different levels in the three areas. Unfortunately, mussels from the central area were not available at the time of sampling. We found levels of OP and NP in mussels higher in the south than in the north, but these differences were not significant. APEs concentrations in clams were generally similar in the three areas, with the exception of OPE ones, which were significantly lower in the north. As reported in Table 3, we found values of APEs in clams comparable to those in mussels. These data show that the levels of APEs in the three areas are similar within each bivalve species. This might mean that the contamination along the coastal area of the Adriatic Sea is substantially at the same level. This finding is unexpected if we consider the different geographical and ecological features that characterize the three areas of the Adriatic Sea. Additional studies are necessary to better understand these results. Differently, the data on squids and cuttlefishes seem to indicate a different pattern of contamination in the three areas. Indeed, the highest levels of APEs for both the species have been found in the central Adriatic Sea. The first consideration is that these data do not conflict with those on bivalves, as the fishing areas are different, namely those for squids and cuttlefishes which are in open sea. We cannot exclude that other factors may have contributed to the different pattern of contamination of cephalopods, like the migration feature of these species, their metabolism, food chains, and bioaccumulation processes. The sedimentation pattern of the river suspended particles, to which APEs are mainly associated, may have played a role as well (7, 9). NP appears to be the very important, widespread contaminant; indeed it occurred at levels much higher than the other APEs in all the species examined. The environmental and human health implications associated with these results are difficult to predict because of the insufficient data on ecotoxicity and toxicity for the examined compounds. According to the BCF found in mussels the results of this study would imply a NP water concentration in the range of 60-90 ng/L, which is by far lower than that which can give rise to sublethal effects on mussels. As to the possible human health implications, the consumption of molluscs of the Adriatic Sea implies APE intakes which are some orders of magnitude lower than those responsible for some effects in laboratory animals. Indeed, even assuming a mollusc consumption of 20-40 g/person/ day for the general population (38), it would correspond to a mean intake of 6-13 µg of NP. As for strong social seafood consumers, if they ate 150-270 g of mollusc/day/person, they would take in 48-87 µg of NP (approximately up to 1.4 µg/kg bw). Despite these apparently low risks, the reasons for concern still remain because the levels of APEs found in this study indicate a general contamination of the Adriatic Sea even far from the coast. Furthermore, at the levels found in this study, APE contamination does not seem dangerous for mussels, but it is not known about the potential hazards for other marine organisms. Finally, they contribute to the general environmental estrogen pool.
Acknowledgments We wish to thank Dr. Alberto Mantovani for his efforts and precious advice in reviewing the manuscript. VOL. 35, NO. 15, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3111
Literature Cited (1) Naylor, C. G. Tex. Chem. Color. 1995, 27, 29. (2) Tyl, R. W.; Myers, C. B.; Marr, M. C.; Brine, D. R.; Fail, P. A.; Seely, J. C.; Van Miller, J. P. Regul. Toxicol. Pharmacol. 1999, 30(2 Pt 1), 81. (3) CEFIC Survey of nonylphenol and nonylphenol ethoxylate production, use, life cycle emission and occupational exposure; CEFIC ad hoc Nonylphenol Risk Assessment Task Force: CESIO APE Task Force, 1996. (4) Reinard, M.; Goodman, N. Environ. Sci. Technol. 1982, 16(6), 351. (5) Stephanou, E.; Giger, W. Environ. Sci. Technol. 1982, 16, 800. (6) Giger, W.; Brunner, P. H.; Schaffner, C. Science 1984, 225, 623. (7) Ahel, M.; Giger, W. Chemosphere 1993, 26, 1471. (8) Brunner, P. H.; Capri, S.; Marcomini, A.; Giger, W. Water Res. 1988, 22, 1465. (9) Nimrod, A. C.; Benson, W. H. Crit. Rev. Toxicol. 1996, 26(3), 335. (10) Soto, A. M.; Justicia, H.; Wray, J. W.; Sonnenschein, C. Environ. Health Perspect. 1991, 92, 167. (11) Jobling, S.; Sumpter, J. P. Aquat. Toxicol. 1993, 27, 361. (12) Purdom, C. E.; Hardiman, P. A.; Bye, V. J.; Eno, N. C.; Tyler, C. R.; Sumpter, J. P. Chem. Ecol. 1994, 8, 275. (13) Jobling, S.; Sheahan, D.; Osborne, J. A.; Matthiessen, P.; Sumpter, J. P. Environ. Toxicol. Chem. 1996, 15(2), 194. (14) Vos, J.; Dbing, E.; Greim, H. A.; Ladefoge, O.; Lambre´, C.; Tarazona, J. V.; Brandt, I.; Vethaak, D. Crit. Rev. Toxicol. 2000, 30(1), 71. (15) Ekelunnd, R.; Bergman, A.; Granmo, A.; Bergrren, M. Environ. Pollut. 1990, 64, 107. (16) White, R.; Hoare, S. A.; Sumpter, J. P.; Parker, M. G. Endocrinology 1994, 135(1), 175. (17) Jobling, S.; Sheahan, D. A.; Osborne, J. A.; Matthiessen, P.; Sumpter, J. P. Environ. Toxicol. Chem. 1995, 15, 194. (18) Lancet, T. Lancet 1995, 345(8955), 933. (19) Gray, M. A.; Metcalfe, C. D. Environ. Toxicol. Chem. 1997, 16(5), 1082. (20) Harries, J. E.; Sheahan, D. A.; Jobling, S.; Matthiessen, P.; Neall, P.; Sumpter, J. P.; Tylor, T.; Zaman, N. Environ. Toxicol. Chem. 1997, 16(3), 534.
3112
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 15, 2001
(21) Jobling, S.; Nolan, M.; Tyler, C. R.; Brighty, G.; Sumpter, J. P. Environ. Sci. Technol. 1998, 32, 2498. (22) Gray, M. A.; Niimi, A. J.; Metcalfe, C. D. Environ. Toxicol. Chem. 1999, 18(8), 1835. (23) Lye, C. M.; Frid, C. L. J.; Gill, M. E.; Cooper, D. W.; Jones, D. M. Environ. Sci. Technol. 1999, 33, 1009. (24) Pedersen, S. N.; Christiansen, L. B.; Pedersen, K. L.; Korsgaard, B.; Bjerregaard, P. Sci. Total Environ. 1999, 233(1-3), 89. (25) Laws, S. C.; Carey, S. A.; Ferrell, J. M.; Bodman, G. J.; Cooper, R. L. Toxicol. Sci. 2000, 54(1), 154. (26) Blake, C. A.; Boockfor, F. R. Biol. Reprod. 1997, 57(2), 255. (27) Sharpe, R. M.; Skakkeback, N. E. Lancet 1993, 1(1392-1395). (28) Mueller, G. C.; Kim, U.-H. Endocrinology 1978, 102, 1429. (29) Tyler, C. R.; Jobling, S.; Sumpter, J. P. Crit. Rev. Toxicol. 1998, 28(4), 319. (30) Granmo, A.; Ekelund, R.; Magnusson, K.; Berggren, M. Environ. Pollut. 1989, 59, 115. (31) Naylor, C. G. Soap Cosmet. Chem. Spec. 1992, 68, 27. (32) Swedmark, M.; Braaten, B.; Emanuelsson, E.; Granmo, A. Mar. Biol. 1971, 9, 183. (33) Cunny, H. C.; Mayes, B. A.; Rosica, K. A.; Trutter, J. A.; Van Miller, J. P. Regul. Toxicol. Pharmacol. 1997, 26(2), 172. (34) Long, S.; Warren, A.; Little, J. Carcinogenesis 1982, 3, 553. (35) Mosetti, F. Boll. Oceanol. Teor. Appl. 1983, 1, 143. (36) Jensen, S.; Reutergardh, L.; Jansson, B. In Manual of methods in aquatic environment research. Part 9: analyses of metals and organochlorine in fish; FAO Fisheries Technical Paper no. 212; Food and Agriculture Organization of the United Nations: Rome, Italy, 1983; FIRI/T212, p 21. (37) Wahlberg, C.; Renberg, L.; Wideqvist, U. Chemosphere 1990, 20(1-2), 179. (38) Ferrara, F.; Funari, E. Rapporti ISTISAN 1999, 99/16, 1.
Received for review January 5, 2001. Revised manuscript received May 14, 2001. Accepted May 17, 2001. ES010508H
