- Email: [email protected]
Mercury and Selenium Content in Selected Seafood
(2001) 14, 461}467 doi:10.006/jfca.2001.1003 Available online at http://www.idealibrary.com on
JOURNAL OF FOOD COMPOSITION AND ANALYSIS
ORIGINAL ARTICLE Mercury and Selenium Content in Selected Seafood Maria Plessi, Davide Bertelli, and Agar Monzani Dipartimento di Scienze Farmaceutiche, Modena University, Via Campi 183, 4100 Modena, Italy Received July 31, 2000, and in revised form February 28, 2001
The mercury and selenium contents of fresh seafood were determined, respectively, by means of cold vapor atomic absorption spectroscopy (CVAAS) and hydride-generation atomic absorption spectrometry (HGAAS). All the values obtained were lower than the European Union's legal limit of 0.5 mg/kg fresh food, rising to 1.0 mg/kg for the edible parts of some listed species; in "sh they vary between 0.057 mg/kg in sole and 0.579 mg/kg in sword"sh (included in the category of large "sh, for which the legal limit is 1 mg/kg). The levels of selenium vary between 0.073 mg/kg in perch and 0.743 mg/kg in tuna. In shell"sh the mercury content varies between 0.023 mg/kg in moscardino and 0.150 mg/kg in Mediterranean shrimp, while that of selenium varies between 0.067 mg/kg in spiny lobster and 0.605 mg/kg in mussels. A signi"cant di!erence was found between "sh and shell"sh for mercury, but not for selenium. A large excess of selenium in relation to mercury was observed (the mean molar ratio of Hg/Se is 0.23 for "sh and 0.09 for shell"sh), but no signi"cant correlation was found between the two elements. 2001 Academic Press Key =ords: mercury; selenium; "sh; shell"sh; atomic absorption spectrometry.
INTRODUCTION The principal source of trace elements for humans is diet. Nutrition plays a fundamental role in ensuring the correct intake of essential elements but nutrition also exposes to toxicological doses. Some special nutrients can prevent or decrease exposure to toxic metals. Trace metals perform their biological function correctly when their tissue concentrations are within well-de"ned limits; there are sundry regulatory mechanisms. Mercury, a very toxic metal, is present at trace levels in living organisms in both inorganic and organic forms. Its toxic e!ects have been highlighted by some cases of collective poisoning in people who consumed a lot of "sh (Zook et al., 1976). It is generally accepted that seafood represents one of the major sources of mercury in the human food chain. Marine organisms are able to accumulate this metal and its most toxic organic compounds by "ltering their food from sea water. Accordingly, the provisional tolerable weekly intake (PTWI 0.005 mg/kg b.w.) established by the Joint FAO (Food and Agricultural Organization)/WHO (World Health Organization) Committee (WHO, 1993) has led to regulatory guidelines for the mercury concentrations allowed in seafood being established in several countries. European Commission Decision 93/351 sets the maximum limit for mercury in seafood at 0.5 mg/kg for fresh To whom correspondence and reprint requests should be addressed. Tel.: #39-059-205-5147. Fax: #39-059-205-5131. E-mail: [email protected] 0889}1575/01/050461#07 $35.00/0
2001 Academic Press
462
PLESSI, BERTELLI, AND MONZANI
food, increasing to 1.0 mg/kg for the edible parts of some listed species which, for physiological reasons, concentrate mercury more easily in their tissues (EEC, 1993). Selenium is an essential trace element for humans, as obvious from its biochemical role as part of the active site in selenoproteins, such as glutathione peroxidase. The in#uence of dietary selenium upon the activity of glutathione peroxidase is evident, and minimum selenium requirements have been "xed: the selenium requirement of adults is calculated to be 70 and 55 lg/day for males and females, respectively (RDA, 1989). However, it should be borne in mind that selenium plays an ambivalent role in relation to its concentration, high amounts possibly having toxic e!ects. Seafood, which is very important in the human diet all over the world, represents a main source of protein, but it can accumulate mineral ions, including those that are potentially toxic. Elements like mercury and selenium, which are present in sea water as natural resources but unfortunately also as a result of pollution, are assimilated by marine organisms, particularly by the "lter-feeders such as shell"sh, which accumulate metals in their viscera from the marine environment. Predatory "sh higher up the food chain can also accumulate these substances. Selenium protects the organism against organic and inorganic mercury thanks to the metabolic interaction between the two elements (Seppanen et al., 1998; Goyer, 1997), which is therefore a process of great importance. The aim of this study was therefore to determine the mercury and selenium contents of selected samples of "sh and shell"sh. Mercury and selenium in foods have mostly been determined at the trace level using spectrophotometric techniques (ETAAS, HGAAS, CVAAS) (Tinggi, 1999; Adeloju et al., 1994), which are preferred because of their low detection limits. In this study, mercury and selenium were determined, respectively, by means of cold vapor atomic absorption spectroscopy (CVAAS) and hydride-generation atomic absorption spectroscopy (HGAAS). These analytical techniques require that the samples be "rst digested, and, in order to achieve mineralization without loss of metal. Most mercury compounds being highly volatile, samples were digested using a microwave digestion system, which also reduces risk of contamination (Vermeier et al., 1989). Using the HGAAS technique for selenium determination, the accuracy of analysis depends on the e$ciency of the selenium conversion to Se (IV) (Cobo Fernandez et al., 1993); accordingly, digestion was performed with nitric acid in a thermostatic bath followed by hot treatment with hydrochloric acid. MATERIALS AND METHODS Samples Thirty-nine types of seafood (24 saltwater, 1 freshwater, the perch, and 14 shell"sh), representative of what is commonly consumed in this region, were purchased from a wholesaler in Modena (Italy) selling "sh of various provenance. Three samples of each type were bought at di!erent times. Only the edible parts of the "sh were analyzed; in the case of the large "sh (samples 1}11) only the "llets were used; in that of the medium-sized "sh (samples 12}22), the heads, skin, viscera, scales and tail were removed; the small "sh (samples 23, 24) were analyzed whole; the shell"sh had their shells removed. The edible parts of the samples were homogenized, dried to constant weight and kept in polyethylene bottles at !183C.
MERCURY AND SELENIUM IN SEAFOOD
463
Mercury Determination Three replicate portions (300 mg) of each sample were placed in closed digestion vessels with the addition of 5 mL concentrated nitric acid and 0.5 mL hydrogen peroxide and digested in a microwave oven (Milestone MLS 1200 Mega) following a "ve-step program: 250 W for 1 min; 0 W for 1 min; 250 W for 8 min; 400 W for 5 min; 650 W for 5 min. After digestion, the samples were cooled and the digestion solutions were transferred into 10 mL volumetric #asks and diluted to volume with ultrapure water (Milli-Q plus 185 Reagent water system, Millipore Corp.). A 3 mL portion of the digestion solutions was diluted with hydrochloric acid and placed in the generator vessel of a Perkin-Elmer Mod.3030 atomic absorption spectrometer, equipped with a Perkin-Elmer MHS-10 hydride generator. Fresh sodium tetrahydroborate solution, daily prepared by dissolving 6 g of NaBH and 2 g of NaOH in 200 mL of distilled water, was added and the resultant Hg3 was carried into the quartz cuvette by means of an argon stream. The measurement parameters were as follows: wavelength, 253.6 nm; slit width, 0.7 nm; lamp current, 7 mA. All determinations of mercury were carried out by the standard addition method. Selenium Determination Three replicate portions (300 mg) of each sample were placed in microkjeldahl #asks and mineralized by the addition of 4 mL concentrated nitric acid at 803C for 1 h in a thermostated bath. Three milliliters of 65% nitric acid were then added and heated until digestion was complete. The reduction of Se (VI) to Se (IV) was performed by adding 2 mL 37% hydrochloric acid and heating at 1003C for 10 min. After cooling, the digested solutions were diluted with ultrapure water to 20 mL. A 1 mL aliquot was transferred to the reaction vessel, and 9 mL 1.5% hydrochloric acid was added. The selenium content was determined by hydride-generation atomic absorption spectroscopy (Perkin-Elmer Mod.3030 atomic absorption spectrometer, equipped with a Perkin-Elmer MHS-10 hydride generator) using the standard addition method. The measurement parameters were as follows: wavelength, 196.0 nm; slit width, 2 nm; lamp current, 15 mA. Hydride generation was carried out using a solution of 3% sodium tetrahydroborate in 1% NaOH. Two di!erent standard reference materials for mercury and selenium (BCR-CRM no. 278 Mussel Tissue and BCR-CRM no. 422 Cod Muscle) were analyzed in order to evaluate the accuracy of the analytical methods. The mercury and selenium contents in the seafood samples were compared statistically by One-way ANOVA. RESULTS AND DISCUSSION The calculated analytical detection limits, de"ned as three times the standard deviations of blanks, were 4.6 ng for mercury and 0.94 ng for selenium in the instrumental conditions used for the analysis, corresponding to 0.051 mg Hg/kg of dried sample and 0.06 mg Se/kg of dried sample in seafood. Accuracy, tested by analyzing the standard reference materials, was found to be satisfactory (Table 1). These results con"rm that the procedures are suitable for the determination of mercury and selenium in seafood. Table 2 shows the mercury and selenium mean concentrations and standard deviations in "sh, expressed as mg per kg of edible part (fresh weight). The mercury
464
PLESSI, BERTELLI, AND MONZANI TABLE 1 Analysis of standard reference materials Hg content (lg/g)
BCR-CRM 278 Mussel tissue BCR-CRM 422 Cod muscle Mean and
S.D.
Se content (lg/g)
Found
Certi"ed
Found
Certi"ed
0.178$0.02
0.188$0.007
1.62$0.03
1.66$0.04
0.537$0.037
0.559$0.016
1.69$0.12
1.63$0.07
of eight individual measurements.
concentration varies from 0.057 mg/kg in sole to 0.579 mg/kg in sword"sh, the average being 0.143 mg/kg. All the values are within the EEC guidelines, the highest value (the only one higher than 0.5 mg/kg) being found in sword"sh, for which the limit for edible parts is 1.0 mg/kg. Compared with other reported mercury values in "sh consumed in Italy (Dossena and Palmieri, 1998; Baldini et al., 1994), the levels found in this study are slightly lower. The selenium concentrations vary from 0.073 mg/kg in perch, the only freshwater "sh, to 0.734 mg/kg in tuna, the average value being 0.307 mg/kg fresh weight; these values are consistent with other reported values for "sh consumed in England (MAFF, 1998), Belgium (Guns et al., 1992) and Spain (Diaz-Alarcon et al., 1994). Table 3 shows out the mercury and selenium mean concentrations and standard deviations in di!erent varieties of shell"sh, expressed as mg per kg of edible parts (fresh weight). The mercury concentrations range from 0.023 mg/kg in moscardino to 0.150 mg/kg in Mediterranean shrimp, the average value for mercury being 0.063 mg/kg. The selenium concentration varies from 0.067 mg/kg in spiny lobster to 0.605mg/kg in mussels, the average value (0.328 mg/kg) being consistent with the reported values for "sh, whereas other studies report higher values for shell"sh, (MAFF, 1998; Guns et al., 1992). It can therefore be stated that the average concentrations of selenium, an essential element, in "sh and shell"sh are similar and that they are important to ensure the recommended daily amount. Mercury, which is due to water contamination, is present in greater concentrations in "sh, which accumulate it, rather than in shell"sh. However, mercury and selenium concentrations vary considerably from sample to sample, probably owing to environmental factors and also to certain characteristics of the species (size, alimentary habits, etc.). The results obtained were analyzed by means of ANOVA, which showed a signi"cant di!erence between the concentration of mercury in "sh and that in shell"sh (P(0.05), with higher values registered for "sh, but no di!erence in the selenium concentration. Regression analysis did not reveal any signi"cant correlation between mercury and selenium content. Tables 2 and 3 also show the molar ratio between mercury and selenium, where the di!erence between "sh and shell"sh was at a higher level of signi"cance (P(0.01); the average of molar ratios between mercury and selenium are 0.23 for "sh and 0.09 for shell"sh, the only samples with high values being sword"sh (0.81), gray"sh (0.80) and perch (0.75). A large excess of selenium in relation to mercury was found not only for shell"sh but also for "sh; this result is consistent with the results of a research recently published by Dietz et al. (2000). The favorable ratio between the elements increases the nutritional importance of seafood, and selenium is widely reported as being one of the major detoxifying agents for mercury and methylmercury (Stopford and Goldwater, 1975).
465
MERCURY AND SELENIUM IN SEAFOOD TABLE 2
Mercury and selenium contents in edible portion of "shes: mean and RSD of three independent measurements
1. Angler "sh (¸ophius piscatorius) 2. Cod (Gadus morrhua) Iceland 3. Cod (Gadus morrhua) Norway 4. Gray"sh (Squalus Acanthias) 5. Hake (Merluccius merluccius) 6. Halibut (Hyppoglossus Hyppogl.) 7. Salmon (Salmo salar) Norway 8. Salmon (Salmo salar) Alaska 9. Sword"sh (Xiphias Gladius) 10. Tuna (¹hunnus thynnus) 11. Whithing (Gadus merlangus) 12. Alaskan pollack (¹eragra chalcogramma) 13. Goby (Gobius niger) 14. Gurnard (¹rigla lucerna) 15. Mackerel (Scomber scombrus) 16. Mullet (Mullus surmuletus) 17. Ox-eye bream (Box boops) 18. Perch (Perca -uviatilis) 19. Porgy (Pagrus pagrus) 20. Red"sh (Sebastes marinus) 21. Scorpion "sh (Scorpaena scrofa) 22. Sole (Solea vulgaris) 23. Anchovy (Engraulis encrasicholus) 24. Red band"sh (Cepola rubescens) 25. Sardine (Sardina pilchardus)
Hg mg/kg (RSD)
Hg lM/kg
Se mg/kg (RSD)
Se lM/kg
Hg:Se molar ratio
0.089 (1.1)
0.44
0.173 (2.9)
2.19
0.20
0.159 (0.9)
0.79
0.306 (15.3)
3.87
0.20
0.077 (12.9)
0.38
0.188 (14.8)
2.38
0.16
0.282 (1.1)
1.41
0.139 (18.7)
1.76
0.80
0.086 (4.6)
0.43
0.466 (5.4)
5.9
0.07
0.069 (13.1)
0.34
0.134 (10.2)
1.70
0.20
0.086 (13.9)
0.43
0.195 (15.2)
2.47
0.17
0.117 (1.7)
0.58
0.353 (10.2)
4.47
0.13
0.579 (2.8)
2.89
0.283 (10.1)
3.58
0.81
0.249 (3.2)
1.24
0.734 (4.2)
9.29
0.13
0.187 (3.7)
0.93
0.290 (14.8)
3.67
0.26
0.062 (4.8)
0.31
0.164 (16.4)
2.07
0.15
0.145 (4.1)
0.72
0.279 (6.1)
3.53
0.20
0.138 (0.7)
0.69
0.372 (8.06)
4.71
0.15
0.126 (0.8)
0.63
0.356 (6.2)
4.51
0.14
0.103 (4.8)
0.51
0.426 (5.3)
5.39
0.09
0.104 (8.6)
0.52
0.337 (2.7)
4.77
0.11
0.139 (2.9)
0.69
0.073 (19.1)
0.92
0.75
0.204 (5.9)
1.02
0.286 (4.9)
3.62
0.28
0.096 (7.3)
0.49
0.247 (2.1)
3.13
0.16
0.134 (2.2)
0.67
0.417 (13.2)
5.28
0.13
0.057 (12)
0.28
0.262 (15.3)
3.32
0.08
0.075 (16.1)
0.35
0.225 (12.8)
2.85
0.12
0.058 (12.1)
0.29
0.299 (10.7)
3.79
0.08
0.156 (1.9)
0.78
0.678 (2.2)
8.59
0.09
466
PLESSI, BERTELLI, AND MONZANI TABLE 3
Mercury and selenium contents in edible portion of shell"shes: mean and RSD of three independent measurements
1. Clam (
Hg mg/kg (RSD)
Hg lM/kg
Se mg/kg (RSD)
Se lM/kg
Hg:Se molar ratio
0.055 (12.7)
0.27
0.330 (0.6)
4.18
0.06
0.120 (6.6)
0.60
0.487 (3.3)
6.17
0.10
0.064 (6.3)
0.32
0.454 (3.1)
5.75
0.06
0.074 (2.7)
0.37
0.198 (12.1)
2.51
0.15
0.042 (9.5)
0.20
0.403 (10.3)
5.10
0.04
0.032 (6.3)
0.16
0.257 (6.6)
3.25
0.05
0.150 (1.3)
0.75
0.509 (6.3)
6.44
0.12
0.023 (13.2)
0.11
0.120 (5.0)
1.52
0.07
0.038 (5.1)
0.20
0.605 (5.3)
7.66
0.03
0.068 (8.8)
0.34
0.435 (3.1)
5.51
0.06
0.040 (10.0)
0.20
0.176 (10.2)
2.20
0.09
0.028 (5.5)
0.14
0.067 (10.1)
0.85
0.16
0.064 (1.6)
0.32
0.106 (10.5)
1.34
0.24
0.089 (6.7)
0.44
0.444 (3.1)
5.62
0.08
REFERENCES Adeloju, S. B., Dhindsa, H. S., and Tandon, R. K. (1994). Evaluation of some wet decomposition methods for mercury determination in biological and environmental materials by cold vapour atomic absorption spectroscopy. Anal. Chim. Acta 285, 359}364. Baldini M., Molinaro, M. G., Stacchini, P., Zanasi, F., Comi, R., and Leoni, V. (1994). Valutazione della ingestione settimanale di mercurio con la dieta in Italia Riv. Sci. Alim. 23, 177}182. Cobo Fernandez, M. G., Palacios, M. A., and Camara, C. (1993). Flow-injection continuous-#ow systems for the determination of Se (IV) and Se (VI) by hydride generation atomic absorption spectrometry with on-line prereduction of Se (VI) to Se (IV). Anal. Chim. Acta 283, 386}392. Diaz-Alarcon, J. P., Navarro-Alarcon, M., Lopez-Martinez, M. C., and Lopez-Garcia de la Serrana, H. (1994). Determination of Selenium in Fresh Fish from southeastern Spain for calculation of daily dietary intake. J. Agric. Food Chem. 42, 334}337. Dietz, R., Riget, F., and Born, E. W. (2000). An assessment of selenium to mercury in Greenland marine animals. Sci. ¹otal Environ. 245, 15}24. Dossena, A. and Palmieri, R. (1998). Contenuto di mercurio nel pesce del Mar Ligure di ponente ed esposizione alimentare al metallo con i prodotti della pesca locale. Riv. Sci. Alim. 27, 35}41. European Commission (EEC), (1993). Commission decision 93/351 of May 19, 1993. O+. J. Eur. Communities, L144. Goyer, R. A. (1997). Toxic and essential metal interactions. Ann. Rev. Nutr. 17, 35}50. Guns, M., Hoeyweghen, P-van, Vyncke, W., and Clerck, R.-de. (1992). Selenium assessment and its relation to mercury levels in "sh, shrimp and mussels from Belgian continental shelf waters. Rev. Agric. 45, 731}739.
MERCURY AND SELENIUM IN SEAFOOD
467
Ministry of Agriculture, Fisheries and Food (MAFF) (UK), (1998). Concentration of metals and other elements in marine "sh and shell"sh. Food surveillance information sheet No. 151. RDA from Food and Nutrition Research Council. (1989). Recommended Dietary Allowance, 10th edn. National Academic Press, Washington, DC. Seppanen, K., Laatikainen, R., Salonen, J. T., Kantola, M., Lotjonen, S., Harri, M., Nurminen, L., Kaikkonen, J., and Nyyssonen, K. (1998). Mercury-binding capacity of organic and inorganic selenium in rat blood and liver. Biol. ¹race Elem. Res. 65, 197}210. Stopford, W. and Goldwater, L. J. (1975). Methylmercury in the environment: a review of current understanding. Environ. Health Perspect. 12, 115}118. Tinggi, U. (1999). Determination of selenium in meat products by hydride generation atomic absorption spectrophotometry. J. Assoc. O+. Anal. Chem. Int. 82, 364}367. Vermeier, G., Vandecasteele, C., and Dams, R. (1989). Microwave dissolution for the determination of mercury in biological samples. Anal. Chim. Acta 220, 257}261. WHO (1993). Evaluation of certain food additives and contaminants. Forty-"rst Report of the Joint FAO/WHO expert committee on food additives. =HO ¹echnical Report Series No. 837. Zook E. G., Powell, J. J., Hackley, B. M., and Emerson, J. A. (1976). National marine "sheries service preliminary survey of selected seafood for mercury, lead, cadmium, chromium and arsenic content. J. Agric. Food Chem. 24, 47}53.
JOURNAL OF FOOD COMPOSITION AND ANALYSIS
ORIGINAL ARTICLE Mercury and Selenium Content in Selected Seafood Maria Plessi, Davide Bertelli, and Agar Monzani Dipartimento di Scienze Farmaceutiche, Modena University, Via Campi 183, 4100 Modena, Italy Received July 31, 2000, and in revised form February 28, 2001
The mercury and selenium contents of fresh seafood were determined, respectively, by means of cold vapor atomic absorption spectroscopy (CVAAS) and hydride-generation atomic absorption spectrometry (HGAAS). All the values obtained were lower than the European Union's legal limit of 0.5 mg/kg fresh food, rising to 1.0 mg/kg for the edible parts of some listed species; in "sh they vary between 0.057 mg/kg in sole and 0.579 mg/kg in sword"sh (included in the category of large "sh, for which the legal limit is 1 mg/kg). The levels of selenium vary between 0.073 mg/kg in perch and 0.743 mg/kg in tuna. In shell"sh the mercury content varies between 0.023 mg/kg in moscardino and 0.150 mg/kg in Mediterranean shrimp, while that of selenium varies between 0.067 mg/kg in spiny lobster and 0.605 mg/kg in mussels. A signi"cant di!erence was found between "sh and shell"sh for mercury, but not for selenium. A large excess of selenium in relation to mercury was observed (the mean molar ratio of Hg/Se is 0.23 for "sh and 0.09 for shell"sh), but no signi"cant correlation was found between the two elements. 2001 Academic Press Key =ords: mercury; selenium; "sh; shell"sh; atomic absorption spectrometry.
INTRODUCTION The principal source of trace elements for humans is diet. Nutrition plays a fundamental role in ensuring the correct intake of essential elements but nutrition also exposes to toxicological doses. Some special nutrients can prevent or decrease exposure to toxic metals. Trace metals perform their biological function correctly when their tissue concentrations are within well-de"ned limits; there are sundry regulatory mechanisms. Mercury, a very toxic metal, is present at trace levels in living organisms in both inorganic and organic forms. Its toxic e!ects have been highlighted by some cases of collective poisoning in people who consumed a lot of "sh (Zook et al., 1976). It is generally accepted that seafood represents one of the major sources of mercury in the human food chain. Marine organisms are able to accumulate this metal and its most toxic organic compounds by "ltering their food from sea water. Accordingly, the provisional tolerable weekly intake (PTWI 0.005 mg/kg b.w.) established by the Joint FAO (Food and Agricultural Organization)/WHO (World Health Organization) Committee (WHO, 1993) has led to regulatory guidelines for the mercury concentrations allowed in seafood being established in several countries. European Commission Decision 93/351 sets the maximum limit for mercury in seafood at 0.5 mg/kg for fresh To whom correspondence and reprint requests should be addressed. Tel.: #39-059-205-5147. Fax: #39-059-205-5131. E-mail: [email protected] 0889}1575/01/050461#07 $35.00/0
2001 Academic Press
462
PLESSI, BERTELLI, AND MONZANI
food, increasing to 1.0 mg/kg for the edible parts of some listed species which, for physiological reasons, concentrate mercury more easily in their tissues (EEC, 1993). Selenium is an essential trace element for humans, as obvious from its biochemical role as part of the active site in selenoproteins, such as glutathione peroxidase. The in#uence of dietary selenium upon the activity of glutathione peroxidase is evident, and minimum selenium requirements have been "xed: the selenium requirement of adults is calculated to be 70 and 55 lg/day for males and females, respectively (RDA, 1989). However, it should be borne in mind that selenium plays an ambivalent role in relation to its concentration, high amounts possibly having toxic e!ects. Seafood, which is very important in the human diet all over the world, represents a main source of protein, but it can accumulate mineral ions, including those that are potentially toxic. Elements like mercury and selenium, which are present in sea water as natural resources but unfortunately also as a result of pollution, are assimilated by marine organisms, particularly by the "lter-feeders such as shell"sh, which accumulate metals in their viscera from the marine environment. Predatory "sh higher up the food chain can also accumulate these substances. Selenium protects the organism against organic and inorganic mercury thanks to the metabolic interaction between the two elements (Seppanen et al., 1998; Goyer, 1997), which is therefore a process of great importance. The aim of this study was therefore to determine the mercury and selenium contents of selected samples of "sh and shell"sh. Mercury and selenium in foods have mostly been determined at the trace level using spectrophotometric techniques (ETAAS, HGAAS, CVAAS) (Tinggi, 1999; Adeloju et al., 1994), which are preferred because of their low detection limits. In this study, mercury and selenium were determined, respectively, by means of cold vapor atomic absorption spectroscopy (CVAAS) and hydride-generation atomic absorption spectroscopy (HGAAS). These analytical techniques require that the samples be "rst digested, and, in order to achieve mineralization without loss of metal. Most mercury compounds being highly volatile, samples were digested using a microwave digestion system, which also reduces risk of contamination (Vermeier et al., 1989). Using the HGAAS technique for selenium determination, the accuracy of analysis depends on the e$ciency of the selenium conversion to Se (IV) (Cobo Fernandez et al., 1993); accordingly, digestion was performed with nitric acid in a thermostatic bath followed by hot treatment with hydrochloric acid. MATERIALS AND METHODS Samples Thirty-nine types of seafood (24 saltwater, 1 freshwater, the perch, and 14 shell"sh), representative of what is commonly consumed in this region, were purchased from a wholesaler in Modena (Italy) selling "sh of various provenance. Three samples of each type were bought at di!erent times. Only the edible parts of the "sh were analyzed; in the case of the large "sh (samples 1}11) only the "llets were used; in that of the medium-sized "sh (samples 12}22), the heads, skin, viscera, scales and tail were removed; the small "sh (samples 23, 24) were analyzed whole; the shell"sh had their shells removed. The edible parts of the samples were homogenized, dried to constant weight and kept in polyethylene bottles at !183C.
MERCURY AND SELENIUM IN SEAFOOD
463
Mercury Determination Three replicate portions (300 mg) of each sample were placed in closed digestion vessels with the addition of 5 mL concentrated nitric acid and 0.5 mL hydrogen peroxide and digested in a microwave oven (Milestone MLS 1200 Mega) following a "ve-step program: 250 W for 1 min; 0 W for 1 min; 250 W for 8 min; 400 W for 5 min; 650 W for 5 min. After digestion, the samples were cooled and the digestion solutions were transferred into 10 mL volumetric #asks and diluted to volume with ultrapure water (Milli-Q plus 185 Reagent water system, Millipore Corp.). A 3 mL portion of the digestion solutions was diluted with hydrochloric acid and placed in the generator vessel of a Perkin-Elmer Mod.3030 atomic absorption spectrometer, equipped with a Perkin-Elmer MHS-10 hydride generator. Fresh sodium tetrahydroborate solution, daily prepared by dissolving 6 g of NaBH and 2 g of NaOH in 200 mL of distilled water, was added and the resultant Hg3 was carried into the quartz cuvette by means of an argon stream. The measurement parameters were as follows: wavelength, 253.6 nm; slit width, 0.7 nm; lamp current, 7 mA. All determinations of mercury were carried out by the standard addition method. Selenium Determination Three replicate portions (300 mg) of each sample were placed in microkjeldahl #asks and mineralized by the addition of 4 mL concentrated nitric acid at 803C for 1 h in a thermostated bath. Three milliliters of 65% nitric acid were then added and heated until digestion was complete. The reduction of Se (VI) to Se (IV) was performed by adding 2 mL 37% hydrochloric acid and heating at 1003C for 10 min. After cooling, the digested solutions were diluted with ultrapure water to 20 mL. A 1 mL aliquot was transferred to the reaction vessel, and 9 mL 1.5% hydrochloric acid was added. The selenium content was determined by hydride-generation atomic absorption spectroscopy (Perkin-Elmer Mod.3030 atomic absorption spectrometer, equipped with a Perkin-Elmer MHS-10 hydride generator) using the standard addition method. The measurement parameters were as follows: wavelength, 196.0 nm; slit width, 2 nm; lamp current, 15 mA. Hydride generation was carried out using a solution of 3% sodium tetrahydroborate in 1% NaOH. Two di!erent standard reference materials for mercury and selenium (BCR-CRM no. 278 Mussel Tissue and BCR-CRM no. 422 Cod Muscle) were analyzed in order to evaluate the accuracy of the analytical methods. The mercury and selenium contents in the seafood samples were compared statistically by One-way ANOVA. RESULTS AND DISCUSSION The calculated analytical detection limits, de"ned as three times the standard deviations of blanks, were 4.6 ng for mercury and 0.94 ng for selenium in the instrumental conditions used for the analysis, corresponding to 0.051 mg Hg/kg of dried sample and 0.06 mg Se/kg of dried sample in seafood. Accuracy, tested by analyzing the standard reference materials, was found to be satisfactory (Table 1). These results con"rm that the procedures are suitable for the determination of mercury and selenium in seafood. Table 2 shows the mercury and selenium mean concentrations and standard deviations in "sh, expressed as mg per kg of edible part (fresh weight). The mercury
464
PLESSI, BERTELLI, AND MONZANI TABLE 1 Analysis of standard reference materials Hg content (lg/g)
BCR-CRM 278 Mussel tissue BCR-CRM 422 Cod muscle Mean and
S.D.
Se content (lg/g)
Found
Certi"ed
Found
Certi"ed
0.178$0.02
0.188$0.007
1.62$0.03
1.66$0.04
0.537$0.037
0.559$0.016
1.69$0.12
1.63$0.07
of eight individual measurements.
concentration varies from 0.057 mg/kg in sole to 0.579 mg/kg in sword"sh, the average being 0.143 mg/kg. All the values are within the EEC guidelines, the highest value (the only one higher than 0.5 mg/kg) being found in sword"sh, for which the limit for edible parts is 1.0 mg/kg. Compared with other reported mercury values in "sh consumed in Italy (Dossena and Palmieri, 1998; Baldini et al., 1994), the levels found in this study are slightly lower. The selenium concentrations vary from 0.073 mg/kg in perch, the only freshwater "sh, to 0.734 mg/kg in tuna, the average value being 0.307 mg/kg fresh weight; these values are consistent with other reported values for "sh consumed in England (MAFF, 1998), Belgium (Guns et al., 1992) and Spain (Diaz-Alarcon et al., 1994). Table 3 shows out the mercury and selenium mean concentrations and standard deviations in di!erent varieties of shell"sh, expressed as mg per kg of edible parts (fresh weight). The mercury concentrations range from 0.023 mg/kg in moscardino to 0.150 mg/kg in Mediterranean shrimp, the average value for mercury being 0.063 mg/kg. The selenium concentration varies from 0.067 mg/kg in spiny lobster to 0.605mg/kg in mussels, the average value (0.328 mg/kg) being consistent with the reported values for "sh, whereas other studies report higher values for shell"sh, (MAFF, 1998; Guns et al., 1992). It can therefore be stated that the average concentrations of selenium, an essential element, in "sh and shell"sh are similar and that they are important to ensure the recommended daily amount. Mercury, which is due to water contamination, is present in greater concentrations in "sh, which accumulate it, rather than in shell"sh. However, mercury and selenium concentrations vary considerably from sample to sample, probably owing to environmental factors and also to certain characteristics of the species (size, alimentary habits, etc.). The results obtained were analyzed by means of ANOVA, which showed a signi"cant di!erence between the concentration of mercury in "sh and that in shell"sh (P(0.05), with higher values registered for "sh, but no di!erence in the selenium concentration. Regression analysis did not reveal any signi"cant correlation between mercury and selenium content. Tables 2 and 3 also show the molar ratio between mercury and selenium, where the di!erence between "sh and shell"sh was at a higher level of signi"cance (P(0.01); the average of molar ratios between mercury and selenium are 0.23 for "sh and 0.09 for shell"sh, the only samples with high values being sword"sh (0.81), gray"sh (0.80) and perch (0.75). A large excess of selenium in relation to mercury was found not only for shell"sh but also for "sh; this result is consistent with the results of a research recently published by Dietz et al. (2000). The favorable ratio between the elements increases the nutritional importance of seafood, and selenium is widely reported as being one of the major detoxifying agents for mercury and methylmercury (Stopford and Goldwater, 1975).
465
MERCURY AND SELENIUM IN SEAFOOD TABLE 2
Mercury and selenium contents in edible portion of "shes: mean and RSD of three independent measurements
1. Angler "sh (¸ophius piscatorius) 2. Cod (Gadus morrhua) Iceland 3. Cod (Gadus morrhua) Norway 4. Gray"sh (Squalus Acanthias) 5. Hake (Merluccius merluccius) 6. Halibut (Hyppoglossus Hyppogl.) 7. Salmon (Salmo salar) Norway 8. Salmon (Salmo salar) Alaska 9. Sword"sh (Xiphias Gladius) 10. Tuna (¹hunnus thynnus) 11. Whithing (Gadus merlangus) 12. Alaskan pollack (¹eragra chalcogramma) 13. Goby (Gobius niger) 14. Gurnard (¹rigla lucerna) 15. Mackerel (Scomber scombrus) 16. Mullet (Mullus surmuletus) 17. Ox-eye bream (Box boops) 18. Perch (Perca -uviatilis) 19. Porgy (Pagrus pagrus) 20. Red"sh (Sebastes marinus) 21. Scorpion "sh (Scorpaena scrofa) 22. Sole (Solea vulgaris) 23. Anchovy (Engraulis encrasicholus) 24. Red band"sh (Cepola rubescens) 25. Sardine (Sardina pilchardus)
Hg mg/kg (RSD)
Hg lM/kg
Se mg/kg (RSD)
Se lM/kg
Hg:Se molar ratio
0.089 (1.1)
0.44
0.173 (2.9)
2.19
0.20
0.159 (0.9)
0.79
0.306 (15.3)
3.87
0.20
0.077 (12.9)
0.38
0.188 (14.8)
2.38
0.16
0.282 (1.1)
1.41
0.139 (18.7)
1.76
0.80
0.086 (4.6)
0.43
0.466 (5.4)
5.9
0.07
0.069 (13.1)
0.34
0.134 (10.2)
1.70
0.20
0.086 (13.9)
0.43
0.195 (15.2)
2.47
0.17
0.117 (1.7)
0.58
0.353 (10.2)
4.47
0.13
0.579 (2.8)
2.89
0.283 (10.1)
3.58
0.81
0.249 (3.2)
1.24
0.734 (4.2)
9.29
0.13
0.187 (3.7)
0.93
0.290 (14.8)
3.67
0.26
0.062 (4.8)
0.31
0.164 (16.4)
2.07
0.15
0.145 (4.1)
0.72
0.279 (6.1)
3.53
0.20
0.138 (0.7)
0.69
0.372 (8.06)
4.71
0.15
0.126 (0.8)
0.63
0.356 (6.2)
4.51
0.14
0.103 (4.8)
0.51
0.426 (5.3)
5.39
0.09
0.104 (8.6)
0.52
0.337 (2.7)
4.77
0.11
0.139 (2.9)
0.69
0.073 (19.1)
0.92
0.75
0.204 (5.9)
1.02
0.286 (4.9)
3.62
0.28
0.096 (7.3)
0.49
0.247 (2.1)
3.13
0.16
0.134 (2.2)
0.67
0.417 (13.2)
5.28
0.13
0.057 (12)
0.28
0.262 (15.3)
3.32
0.08
0.075 (16.1)
0.35
0.225 (12.8)
2.85
0.12
0.058 (12.1)
0.29
0.299 (10.7)
3.79
0.08
0.156 (1.9)
0.78
0.678 (2.2)
8.59
0.09
466
PLESSI, BERTELLI, AND MONZANI TABLE 3
Mercury and selenium contents in edible portion of shell"shes: mean and RSD of three independent measurements
1. Clam (
Hg mg/kg (RSD)
Hg lM/kg
Se mg/kg (RSD)
Se lM/kg
Hg:Se molar ratio
0.055 (12.7)
0.27
0.330 (0.6)
4.18
0.06
0.120 (6.6)
0.60
0.487 (3.3)
6.17
0.10
0.064 (6.3)
0.32
0.454 (3.1)
5.75
0.06
0.074 (2.7)
0.37
0.198 (12.1)
2.51
0.15
0.042 (9.5)
0.20
0.403 (10.3)
5.10
0.04
0.032 (6.3)
0.16
0.257 (6.6)
3.25
0.05
0.150 (1.3)
0.75
0.509 (6.3)
6.44
0.12
0.023 (13.2)
0.11
0.120 (5.0)
1.52
0.07
0.038 (5.1)
0.20
0.605 (5.3)
7.66
0.03
0.068 (8.8)
0.34
0.435 (3.1)
5.51
0.06
0.040 (10.0)
0.20
0.176 (10.2)
2.20
0.09
0.028 (5.5)
0.14
0.067 (10.1)
0.85
0.16
0.064 (1.6)
0.32
0.106 (10.5)
1.34
0.24
0.089 (6.7)
0.44
0.444 (3.1)
5.62
0.08
REFERENCES Adeloju, S. B., Dhindsa, H. S., and Tandon, R. K. (1994). Evaluation of some wet decomposition methods for mercury determination in biological and environmental materials by cold vapour atomic absorption spectroscopy. Anal. Chim. Acta 285, 359}364. Baldini M., Molinaro, M. G., Stacchini, P., Zanasi, F., Comi, R., and Leoni, V. (1994). Valutazione della ingestione settimanale di mercurio con la dieta in Italia Riv. Sci. Alim. 23, 177}182. Cobo Fernandez, M. G., Palacios, M. A., and Camara, C. (1993). Flow-injection continuous-#ow systems for the determination of Se (IV) and Se (VI) by hydride generation atomic absorption spectrometry with on-line prereduction of Se (VI) to Se (IV). Anal. Chim. Acta 283, 386}392. Diaz-Alarcon, J. P., Navarro-Alarcon, M., Lopez-Martinez, M. C., and Lopez-Garcia de la Serrana, H. (1994). Determination of Selenium in Fresh Fish from southeastern Spain for calculation of daily dietary intake. J. Agric. Food Chem. 42, 334}337. Dietz, R., Riget, F., and Born, E. W. (2000). An assessment of selenium to mercury in Greenland marine animals. Sci. ¹otal Environ. 245, 15}24. Dossena, A. and Palmieri, R. (1998). Contenuto di mercurio nel pesce del Mar Ligure di ponente ed esposizione alimentare al metallo con i prodotti della pesca locale. Riv. Sci. Alim. 27, 35}41. European Commission (EEC), (1993). Commission decision 93/351 of May 19, 1993. O+. J. Eur. Communities, L144. Goyer, R. A. (1997). Toxic and essential metal interactions. Ann. Rev. Nutr. 17, 35}50. Guns, M., Hoeyweghen, P-van, Vyncke, W., and Clerck, R.-de. (1992). Selenium assessment and its relation to mercury levels in "sh, shrimp and mussels from Belgian continental shelf waters. Rev. Agric. 45, 731}739.
MERCURY AND SELENIUM IN SEAFOOD
467
Ministry of Agriculture, Fisheries and Food (MAFF) (UK), (1998). Concentration of metals and other elements in marine "sh and shell"sh. Food surveillance information sheet No. 151. RDA from Food and Nutrition Research Council. (1989). Recommended Dietary Allowance, 10th edn. National Academic Press, Washington, DC. Seppanen, K., Laatikainen, R., Salonen, J. T., Kantola, M., Lotjonen, S., Harri, M., Nurminen, L., Kaikkonen, J., and Nyyssonen, K. (1998). Mercury-binding capacity of organic and inorganic selenium in rat blood and liver. Biol. ¹race Elem. Res. 65, 197}210. Stopford, W. and Goldwater, L. J. (1975). Methylmercury in the environment: a review of current understanding. Environ. Health Perspect. 12, 115}118. Tinggi, U. (1999). Determination of selenium in meat products by hydride generation atomic absorption spectrophotometry. J. Assoc. O+. Anal. Chem. Int. 82, 364}367. Vermeier, G., Vandecasteele, C., and Dams, R. (1989). Microwave dissolution for the determination of mercury in biological samples. Anal. Chim. Acta 220, 257}261. WHO (1993). Evaluation of certain food additives and contaminants. Forty-"rst Report of the Joint FAO/WHO expert committee on food additives. =HO ¹echnical Report Series No. 837. Zook E. G., Powell, J. J., Hackley, B. M., and Emerson, J. A. (1976). National marine "sheries service preliminary survey of selected seafood for mercury, lead, cadmium, chromium and arsenic content. J. Agric. Food Chem. 24, 47}53.