Dioxin and PCB Contamination in Chinese Mitten ... - ACS Publications

Feb 4, 2009 - The Chinese mitten crab Eriocheir sinensis is an invasive species in North ... and associated environmental damage of the invasive crab ...
0 downloads 0 Views 792KB Size
Environ. Sci. Technol. 2009, 43, 1624–1629

Dioxin and PCB Contamination in Chinese Mitten Crabs: Human Consumption as a Control Mechanism for an Invasive Species PAUL F. CLARK,† DAVID N. MORTIMER,‡ ROBIN J. LAW,§ JON M. AVERNS,| BILL A. COHEN,⊥ DAVID WOOD,⊥ MARTIN D. ROSE,# ALWYN R. FERNANDES,# AND P H I L I P S . R A I N B O W * ,† Department of Zoology, Natural History Museum, Cromwell Road, London SW7 5BD, U.K., Food Standards Agency (FSA), Aviation House, 125 Kingsway, London WC2B 6NH, U.K., CEFAS Lowestoft Laboratory, Pakefield Road, Lowestoft, Suffolk NR33 0HT, U.K., London Port Health Authority (LPHA), Department of Environmental Services, City of London, P.O. Box 270, Guildhall, London EC2P 2EJ, U.K., Scientific Analysis Laboratories Ltd., Hadfield House, Hadfield Street, Manchester M16 9FE, U.K., and Central Science Laboratory, Sand Hutton, York YO41 1LZ, U.K.

Received October 17, 2008. Revised manuscript received December 23, 2008. Accepted December 31, 2008.

The Chinese mitten crab Eriocheir sinensis is an invasive species in North American and northeastern European rivers and estuaries, especially the Thames, England, with the potential to cause considerable ecological and structural environmental damage. The brown meat of sexually ripe mitten crabs is highly prized in far eastern restaurants, suggesting that harvesting for culinary purposes offers a potential population control mechanism. We have analyzed tissues of Thames and Dutch mitten crabs for potentially toxic dietary contaminants, showing that the brown meat contains raised concentrations of dioxins(polychlorinateddioxinsandpolychlorinateddibenzofurans) and PCBs (polychlorinated biphenyls), organochlorines which are chronic toxins. We have compared estimated daily intakes of these toxins by consumers of meals of mitten crab brown meat against their suggested European Tolerable Daily Intakes (TDI), concluding that a male adult or female beyond childbearing age could consume several portions per week derived from Thames crabs, but fewer from the Dutch crabs. With a caveat that excessive consumption of mitten crab brown meat could lead to exposures of potential concern, particularly in the case of children and women of child-bearing age, it does appear that the harvesting of mitten crabs from the Thames for culinary use need not be discouraged.

* Corresponding author phone: 44 020 7942 5275; fax: 44 020 7942 6126; e-mail: [email protected]. † Natural History Museum. ‡ Food Standards Agency (FSA). § CEFAS Lowestoft Laboratory. | London Port Health Authority (LPHA). ⊥ Scientific Analysis Laboratories Ltd. # Central Science Laboratory. 1624

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 5, 2009

Introduction The Chinese mitten crab Eriocheir sinensis is an introduced invasive species to North American and northeastern European aquatic habitats, spending most of its life in freshwater prior to the migration of reproductively mature adults down to estuaries to breed. The planktonic larvae develop in the estuary before metamorphosed benthic juvenile crabs migrate back upstream (1). The crabs cause considerable environmental damage in freshwater habitats, burrowing destructively into earthworks and disrupting local food chains as large, aggressive, and opportunistic omnivores (1). Although the mitten crab was first recorded in Britain (in the Thames estuary) in 1935, the Thames population only became established, albeit in low numbers, in the 1970s and 1980s (1). In the 1990s, however, the population of mitten crabs in the Thames catchment started to increase dramatically (1), a population explosion that has been maintained into this century (2). In its native home of China and other countries in southeast Asia, the mitten crab is considered a delicacy, and sexually ripe crabs at the time of the yearly autumn migration downstream fetch considerable prices in local restaurants, the gonad tissue being particularly prized. This potential commercial market suggests that harvesting for culinary purposes might offer a cost-effective potential control mechanism to reduce the population numbers and associated environmental damage of the invasive crab in the Thames catchment. Commercial exploitation is particularly attractive for the Thames mitten crabs because they do not appear to harbor the lung fluke Paragonimus westermani (3), a debilitating parasite of man prevalent in Far East populations of mitten crabs (1). We therefore set up a trial to assess the most appropriate method of fishing for the crabs in sufficient numbers to meet possible culinary demand, and under the guidance of the London Port Health Authority who are the food authority of the tidal Thames, analyzed two tissues for potentially dangerous persistent contaminants that might affect suitability for human consumption: the “brown meat” consisting of the hepatopancreas (digestive gland) together with the confluent gonad which dominates the brown meat in sexually mature crabs, and “white meat” consisting of muscle tissue. Mitten crabs were collected at monthly intervals in 2006 from four sites in the Thames estuary (Figure 1a). The tissues were analyzed for the trace metals cadmium, mercury, and lead, and for the organic contaminants benzo[a]pyrene (used as a marker for polyaromatic hydrocarbons (PAH) in food), polychlorinated dioxins and dibenzofurans (PCDD/Fs or “dioxins”), and polychlorinated biphenyls (PCBs). In the light of the results obtained in 2006, we carried out further collections and analyses in 2007 in order to refine our estimates of the likely intake of toxic contaminants from a meal and the frequency of meals of mitten crab brown meat in the diet of the consumer. In order to make this assessment, we needed to measure organochlorine residues in crabs specifically collected from the part of the Thames estuary and at the time of the year most likely to be “fished” in any commercial fishery for mitten crabs, focusing only on the brown meat. We chose therefore to sample mitten crabs at four further sites along the Thames estuary between the entry of the River Lea and Greenhythe (Figure 1b) on four occasions between August and December 2007, compositing male and female crabs. We also took the opportunity to change our analysts in order to confirm that the organochlorine concentrations measured were repeatable be10.1021/es802935a CCC: $40.75

 2009 American Chemical Society

Published on Web 02/04/2009

FIGURE 1. Sites of collection of mitten crabs in the Thames estuary, UK. (a) 2006 collections: 1. Greenhythe; 2. Bow Creek, Blackwall; 3. Lot’s Road Power Station, Chelsea; 4. Richmond; (b) 2007 collections from the mouth of the River Lee to Greenhythe. tween analysts, and that the method of preparation and analysis of the crab tissue had not affected these concentrations. In addition, we responded to the knowledge that mitten crabs are being harvested in Holland, apparently for culinary purposes (4) and being imported into the UK, to set the organochlorine concentrations measured in the Thames mitten crabs in a wider perspective. To this end we analyzed the brown meat of mitten crabs collected from the River Lek near Vianen, Holland, and the River Rhine at Hollands Diep near Dordrecht, Holland, in October 2007.

Experimental Procedures Collection and Sample Preparation (2006). Crabs were collected at monthly intervals during 2006 by potting and fyke netting at four sites in the Thames estuary, namely (1) Greenhythe (51° 27.886′ N, 00° 17.230′ E), (2) Bow Creek, Blackwall (51° 30.422′ N, 00° 00.474′ E), (3) Lot’s Road Power Station, Chelsea (51° 28.519′ N, 00° 10.754′W), (4) Richmond (51° 27.638′ N, 00° 18.898′ W), before being stored frozen at -20 °C. No distinction was made between male and female crabs. Upon thawing of the crabs, two tissuessthe hepatopancreas/gonad and the muscles of the sternal plates, legs, and clawsswere dissected out on a ceramic tile. All dissecting instruments, aluminum foil, and the ceramic tile had been previously well rinsed in n-hexane (extra pure grade, Merck). Samples of the two tissues were pooled separately in aluminum foil for all crabs from the same monthly collection at one site, with the objective of obtaining a minimum sample size of 20 g wet weight for analysis. For some months there were sufficient crabs collected from one or more sites for two replicate pooled samples per site of each tissue, while it was also necessary at times to pool tissues from crabs collected over consecutive months at the same site in order to obtain a minimum amount of tissue for analysis. Pooled tissue samples were refrozen at -20 °C and delivered to Scientific Analysis Laboratories Ltd. for analysis. Sample Analysis (2006). Crab tissues were analyzed for polycyclic aromatic hydrocarbons (PAH), for dioxins and for dioxin-like PCBs by high resolution gas chromatographyhigh resolution mass spectrometry (HRGC-HRMS). Analyses for organic compounds were based on appropriate U.S. Environmental Protection Agency methods (624, 625, 1613, 1668, 8260, and 8270), with isotopically labeled congeners and deuterated PAH added to test samples as appropriate as standards. The trace metals were determined by atomic absorption (Hg) and inductively coupled plasma mass spectrometry (ICP-MS) (Cd and Pb). For Hg analysis, the sample was ashed and Hg vapor trapped on a gold amalgam before purging and analysis. For Cd and Pb the sample was digested with a mixture of hydrochloric and nitric acids before

ICP-MS. All methodology used was accredited by the UK Accreditation Service (UKAS) to the ISO/IEC 17025:2005 Standard. Collection and Sample Preparation (2007). Crabs were collected at monthly intervals from August to November 2007 by fyke netting at four sites in the Thames estuary between the River Lee and Greenhythe, namely (1) Mouth of River Lea (51° 30.430′N, 00° 00.463′E), (2) Thamesmead (51° 30.825′N, 00° 06.204′E), (3) Erith (51° 28.839′N, 00° 11.633′E), (4) Greenhythe (51° 27.886′ N, 00° 17.230′ E), before being stored frozen at -20 °C. No distinction was made between male and female crabs. The crabs from the four sites were combined for each monthly sample: 20 August (18, 20, and 20 from sites 1, 2, and 3), 14 September (16, 16, 15, and 14 from sites 1, 2, 3, and 4), 18 October (15, 14, 13, and 13 from sites 1, 2, 3, and 4), and 17 November (15, 15, 17, and 15 from sites 1, 2, 3, and 4). Dutch crabs were obtained from local Dutch fishermen after collection in fyke nets, 54 from the River Lek near Vianen (52° 00.00′N, 05° 05.60′E) in November, and 60 from the River Rhine at Hollands Diep (51° 42.00′N, 04° 33.00′E) in December and frozen at -20 °C. The frozen crabs were delivered to the Central Science Laboratory where they were thawed and the brown meat dissected out for analysis. Sample Analysis (2007). Each crab tissue sample was fortified with known amounts of surrogate (13C12-labeled) analogues of target analytes and exhaustively extracted using mixed organic solvents. Extracts were purified using adsorption chromatography. Ortho-PCBs, non-ortho-PCBs, and PCDDs/Fs were segregated into two separate fractions. Each fraction was concentrated and further purified before the inclusion of additional surrogate standards. Final determination was by high resolution gas chromatography with either low resolution mass spectrometric detection (ortho-PCBs) or high resolution mass spectrometric detection (non-orthoPCBs and PCDDs/PCDFs). Quality control procedures observed the following requirements. All analytical data met published acceptance criteria (5) for PCDDs and PCDFs and equivalent criteria for PCBs. The method used has been validated and published after peer review (6). Each batch of samples analyzed incorporates one of several reference materials (RMs) (7, 8), for which results are compared with certified or assigned data and laboratory performance (indicative) data. Results for the batch RM all fall within the acceptable range, except for 1,2,3,4,7,8-HxCDD, which is just slightly out of range and results for which are marked as iR if appropriate (Supporting Information (SI) Table S1). Each batch of samples analyzed includes a full reagent blank extract. The contribution from the batch blank was found to be negligible. The analytical performance of the laboratory VOL. 43, NO. 5, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1625

TABLE 1. Total Concentrations of Organochlorine Congeners (WHO-TEQ ng/kg Fresh Weight) in “Brown Meat” (Hepatopancreas/ Gonad) of Chinese Mitten Crabs Eriocheir sinensis from the Thames Estuary and Two Sites in Holland in 2007 Thames

Thames

Thames

Thames

Vianen, Holland

Hollands Diep, Holland

August 2007

September 2007

October 2007

November 2007

October 2007

October 2007

7.68 9.70 7.82 25.2

8.40 8.92 4.15 21.5

32.1 29.0 19.2 80.3

31.5 42.7 68.4 143

dioxins non-ortho-PCBs ortho-PCBs sum of WHO-TEQ

WHO-TEQ ng/kg Fresh Weight 8.98 10.1 9.57 25.6 4.83 7.86 23.4 43.5

in international intercomparison studies (9-11) using essentially the same method, has been adjudged to be acceptable or better. Results are calculated using internationally accepted (12, 13) toxic equivalency factors. Organochlorine Toxicity Assessment. The toxicity of the organochlorines (PCDD/Fs) and PCBs varies between congeners. Because such organochlorines are almost invariably found in complex mixtures, the concept of toxic equivalence factors (TEF) has been developed for risk assessment. Each of the individual congeners is assigned a TEF relative to the most toxic congener 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin (TCDD), which has a TEF of 1. The concentration of each congener is multiplied by its specific TEF to give a toxic equivalence value (commonly referred to as WHO-TEQ); these values can be added to give a total WHO-TEQ for mixtures of dioxins and dioxin-like PCBs, expressing the toxicity as if the mixture was pure TCDD. The 1998 dioxin TEQ scheme (12) has been updated to the 2005 scheme (13) for good reasons. Nevertheless we have used the 1998 scheme here because this is the scheme used in the current legislation used for comparisons made here. Similarly the comparative literature quoted here also uses the 1998 scheme and it is impossible to recalculate such data in terms of the 2005 scheme without access to the original raw data.

Results and Discussion Results for 2006. We were interested in comparing measured concentrations of contaminants in the crabmeat samples against maximum permissible levels for these contaminants in foodstuffs as established in European regulations. Limits were originally established through numerous amendments to Commission Regulation (EC) No. 466/2001, which has subsequently been consolidated into Commission Regulation (EC) 1881/2006 of 19 December 2006. The relevant limits deriving from the latter are referred to below. All Cd concentrations in both tissues were below the European regulatory limit of 0.5 mg kg-1 wet weight for crustaceans, although the regulation excludes the brown meat of crabs, as well as the head and thorax meat of lobsters and similar crustaceans. Cd concentrations were higher in the hepatopancreas/gonad than in the muscle tissue; all but 5 of 26 Cd concentrations in the former tissue were below 0.15 mg kg-1 wet weight, and all but 2 out of 22 Cd concentrations in muscle were below 0.05 mg kg-1 wet weight (SI Table S2). Similarly all mercury concentrations in both tissues were below the European regulatory limit of 0.5 mg kg-1 wet weight in fishery products (which again excludes the brown meat of crab). There were higher mercury concentrations in the muscle (5 out of 25 results above 0.2 mg kg-1 wet weight) than in the hepatopancreas/gonad (1 out of 27 results reaching 0.2 mg kg-1 wet weight) (SI Table S2). In contrast to cadmium and mercury, although none of the lead concentrations in the muscle meat exceeded the European regulatory limit of 0.5 mg kg-1 wet weight for crustaceans, there were a few instances of concentrations in the hepatopancreas/gonad being above this value, with 5 out of 25 samples containing between 0.5 and 0.89 mg Pb kg-1 wet 1626

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 5, 2009

weight (SI Table S2). Crab brown meat is excluded from the regulation because, although it may contain raised concentrations of these trace metals, it is consumed less often and in smaller quantities than crab white meat (muscle). In Chinese mitten crabs from Hong Kong sampled in 1996, mean whole body meat Pb concentrations were around 18 mg kg-1 dry weight (ca. 2.7 mg kg-1 wet weight) (14). In blue crabs (Callinectes sapidus) sampled from bays in Florida during 2003-2004, concentrations of Cd and Hg in the hepatopancreas (brown meat) ranged from 0.01 to 4.6 and 0.02 to 1.1 mg kg-1 wet weight, respectively (15). Pb was not detected in any samples at limits of detection around 0.6 mg kg-1 wet weight, and no Cd was detected in muscle (white meat) samples, the limits of detection for Cd being around 0.16 mgkg-1 wet weight (15). Hg concentrations in muscle tissue ranged from 0.07 to 0.24 mg kg-1 wet weight (15). In crab products from the UK retail market, concentrations of Cd, Hg, and Pb ranged from