arsenic in two fish species from Lake Tanganyika, Burundi, provide values that ... The data illustrates that Lake Tanganyika is still a non-polluted area at the time ...
the Science o f the Total Environment in{o the E ~ a m c n l
ELSEVIER
a n d i u R¢llt~nship with Mtn
The Science of the Total Environment 144 (1994) 103-115
Copper, zinc, manganese, iron, lead, cadmium, mercury and arsenic in fish from Lake Tanganyika, Burundi E. S i n d a y i g a y a a, R . V a n C a u w e n b e r g h b, H . R o b b e r e c h t *b, H . D e e l s t r a b "Faculty of Sciences, University of Burundi, P.O. Box 2700, Bujumbura, Burundi, Ajrica hDepartment of Pharmaceutical Sciences, University of Antwerp, Universiteit~plein 1, 2610 Wilrijk, Belgium (Received 10 September 1992; accepted 28 October 1992)
Abstract
Atomic absorption spectrometric determination of copper, zinc, manganese, iron, lead, cadmium, mercury and arsenic in two fish species from Lake Tanganyika, Burundi, provide values that are lower than most literature data, especially for cadmium, arsenic and mercury. Tissue analysis of Lates stapersii reveals that liver accumulates the highest amount of most elements. The data illustrates that Lake Tanganyika is still a non-polluted area at the time of analysis. Key words. Heavy metals; Arsenic; Mercury; Fish; Africa
1. Introduction
The freshwater Lake Tanganyika in northeastern Burundi is the second deepest lake in the world (1470 m). Fish from this lake serve as the main source of protein for the population of the four surrounding countries: Burundi, Tanzania, Zaire and Zambia (Fig. 1). Fish can accumulate significant amounts of trace elements and heavy metals from both water and food (Vukasinovic, 1988; Hodson, 1988). Bioaccumulation of trace elements is influenced by many variables: properties of the element itself; its chemical form (species); the co-presence of
* Corresponding author. Elsevier Science B.V. SSDI 0048-9697(92)03618-C
other metals; fish species and other characteristics, such as age, site, body size and feeding habits; season and temperature of the water; and factors that influence the metabolic rate of fish (Brooks and Rumsey, 1974; Wahbeh and Madasnek, 1987; Vukasinovic, 1988; Paulsson and Lundbergh, 1989). When consumed in large amounts, these essential trace elements may reach toxic levels (Tsoukali-Papadopoulou et al., 1989; S0rensen, 1991; Nriagu, 1992). Also, further changes in the organoleptic quality of food may occur, e.g. rancidity, caused mainly by the role of the metals as catalysts in many degradation processes. In a previous paper we have analysed the essential trace elements copper, zinc and selenium in different types of fish in central Africa (Bene-
104 •
E. Sindayigaya et al. / Sci. Total Environ. 144 (1994) 103-115 29~
¢10 °
fish species (of which only the most recent are Marcovecchio et al., 1988a, 1988b; Pfeiffer et al., 1989; Tariq et al., 1991; Gonz~lez et al., 1991; Barghigiani et al., 1991; Giordano et al., 1991; Leah et al., 1991, 1992; Barghigiani and de Ranieri, 1992; Navarro et al., 1992). The tissue of greatest concern to humans is skeletal muscle. Methylation of up to 100% of the total mercury in this tissue probably plays a minor role in the half-life of mercury in muscle, estimated to be 2-3 years (SCrensen, 1991). This is also the major reason why, unlike other elements, highest total mercury concentrations were found in fish muscle (Marcovecchio et al., 1988a, 1988b). Various destruction procedures and determination techniques are available (see Hight and Corcoran, 1988; Holak, 1989; Thibaut and Cossa, 1989; Taylor et al., 1992). In this study only total mercury is determined. Different parts of Stolothrissa tanganyikae and Lates stapersii, the two most important fish species in the diet of the local population, have been analysed using atomic absorption spectrometry after wet acid destruction. Trace elements found are compared with literature data on metal content in fish muscle.
2. Experimental
Fig. 1. Lake Tanganyika,Central Africa.
mariya et ai., 1991). We extended our research to more elements (copper, zinc, manganese,-'iron, lead, cadmium, mercury and arsenic) to establish natural background levels of these elements in a non-polluted basin in central Africa (Lake Tanganyika, Burundi). Mercury is one of the most hazardous to fish, primarily because of the tenacity with which it binds to sulphydryl groups. This explains the numerous reports on mercury determination in
2.1. Sample preparation Fish samples were collected at various sites on the shore of Lake Tanganyika. Some fish samples were bought directly from local fish markets. Muscle from the mid-dorsal region of the Lates stapersii species was removed, while for the Stolothrissa tanganyikae, because of the small size and because of consumption habits, whole-body samples were taken for analysis. The latter species was first sundried on the beach sand after capturing until nearly dry (about 16% of the water remained). Storing and pre-treatment of the samples were as described by Benemariya et al. (1991). Wet acid digestion procedures varied, depending on the element analysed. For copper, zinc, manganese, lead, cadmium and mercury, dried sample (1 g) and 3 ml of a nitric-perchloric acid mixture (3.5:1.5, v/v) were placed in a digestion tube fitted with a 400-mm air-
E. Sindayigaya et al./ Sci. Total Environ. 144 (1994) 103-115
105
Table 1 Graphite furnace operating conditions for trace element determination Operating conditions
Elements
Wavelength (nm) Drying (°C/s) Charring (°C/s) Atomization (°C/s)
Pb
Cd
Hg
As
283.3 130/I 0 500/45 2300/5
228.8 130/10 250/65 2100/5
253.6 130/10 200/30 2000/6
193.7 130/10 600/40 2200/6
pyrrolidine dithiocarbamate (APDC) extraction in methylisobutylketone solution (1 ml) was carried out using 1 ml of the digestion solution with following pre-treatment. One milliliter of the digestion solution was neutralised with NHaOH after the addition of' a bromophenol blue indicator; 0.1 ml HCI (32%), 0.25 ml Na2S203 (2%) and 1 ml bidistilled water were sequentially added. For arsenic determination a similar extraction procedure was used, including the addition of 1 ml KI (10%).
cooled condenser. A temperature-controlled and time-controlled heated digestion unit was used (Tecator, H6ganfis, Sweden) as follows: after 1 hour at room temperature the solution was gently heated to 140°C (6 h) until digestion was complete. After cooling, the solution was quantitatively transferred to a standard flask and diluted to 25 ml with bidistilled water of which the trace element content was carefully checked. For arsenic determination the digestion procedure was as follows: dried sample (1 g) and 3 ml nitric acid were placed in the digestion tubes at room temperature overnight; after addition of 1 ml sulphuric acid the temperature was gradually heated to 160°C and maintained for 6 h. The condenser was removed to reduce the volume to about 1 ml. After addition of 1 ml of concentrated HC1 the air condenser was replaced and the mixture was heated at 100°C for 15 min to convert As(VI) to As(Ill). Cooling and diluting resulted in a volume of 25 ml. For total mercury determination an ammonium
2.2. Techniques of analysis A flame atomic absorption spectrometer (Perkin-Elmer 2280) was used for the determination of copper, zinc, manganese and iron. Concentrations were calculated using external calibration and linear regression. Cadmium, lead, mercury and arsenic were determined by flameless atomic absorption spectroscopy with a Perkin-Elmer 4000 unit equipped with a PE A500 furnace. The temperature pro-
Table 2 Results of the analysis (/~g • g-l) of some standard reference materials Element
Material
Our results (mean 4- S.D.)
Certified levels (mean -4- S.D.)
Cu Zn Mn Fe Pb Cd Hg As
NBS NBS NBS NBS NBS NBS NBS NBS
62 830 16 186 0.46 3.1 0.056 0.38
63 852 17 185 0.48 3.5 0.057 0.41
1566 1566 1566 1566 1566 1566 1566 1568
+ + + 4. 4. 4. 4+
2 16 1 8 0.02 0.02 0.02 0.03
± + ± 4. 4444-
3 14 1 34 0.04 0.4 0.015 0.05
E. Sindayigaya et al. / Sci. Total Environ. 144 (1994) 103-115
106
Table 3 Trace element content (/zg. g - t ; m e a n ± S.D.) of Stolothrissa tanganyikae (whole body) a n d Lates stapersii (muscle) Element
n
Cu Zn Mn Fe Pb Cd Hg As
Concentration
50 50 50 50 50 50 50 50
Stolothrissa tanganyikae
Lates stapersii
3.2 ± 134 ± 17 ± 200 ± 0.04 ± 0.27 ± 0.06 4. < 0.05
1.7 ± 21 ± 5.0 ± 35 ± 0.01 ± 0.03 ± 0.04 ± < 0.05
0.4 18 12 52 0.02 0.11 0.03
0.2 5 2.0 2.0 0.001 0.01 0.02
concentration levels for several elements have also been published for other fish (Lowe et al., 1985; Jaffar and Ashraf, 1988; Ramelow et al., 1989; Giordano et al., 1991; Ashraf and Jaffar, 1991; Tariq et al., 1991). Higher trace element levels in the Stolothrissa tanganyikae could probably be due to the fact that this fish was consumed as a whole. Okoye (1991) showed that fish head, eaten as a local delicacy in Nigeria, contributed significantly to total heavy metals in consumed fish. For cadmium a concentration as much as ten times higher was found for the Stolothrissa species. Mercury was not detected above the 0.05/~g/g detection limit. Since fish may absorb dissolved heavy metals from the surrounding water, these trace elements may become accumulated in various tissues and
gramme used for the graphite tube for the different elements is summarised in Table 1. The method of standard addition was used for calibration. The accuracy of the method for all elements was checked by analysis of oyster tissue (NBS 1566) and rice flour (NBS 1568), and shown to be excellent (Table 2). 3. Results and discussion Table 3 presents the concentrations of the various elements in the two fish species analysed. All concentrations are expressed in #g/g, dry weight. Concentration levels found in Stolothrissa tanganyikae are up to five times higher than those in the Lates stapersii muscle. A similar observation was made by Benemariya et al. (1991) for selenium levels in the same species. Species differences in
Table 4 Trace element content (p.g . g - l ; m e a n ± S.D.) of different o r g a n s of Lates stapersii (n = 3) Element
Concentration Heart
Cu Zn Mn Fe Pb Cd As Hg
2.9 ± 15.6 ± 1.3 ± 5.1 40.04 ± 0.06 ±
