Biomagnificação do mercúrio na cadeia de um quelônio de água doce (Chelus fimbriata) e seu perfil genotóxico em ambientes fluviais do médio rio Negro, Amazonas, Brasil

Abstract

Turtles can be used as biological models in toxicological studies due to their unique ecological and life-history attributes, including their wide geographic distribution, the diversity of their habitats and the variety of trophic levels they occupy. Besides this, turtles are long-lived organisms which can be exposed to long-term contamination. The objective of the present study was to investigate the bioaccumulation of total mercury, the genotoxic profile and the biomagnification of Mercury in the food chain of the turtle Chelus fimbriata in fluvial environments of the middle Negro River, Amazonas, Brazil. Biological samples (muscle, carapace and claw) were collected from 25 individuals of Chelus fimbriata and the principal aquatic plants (phytoplankton, periphyton, emergent macrophytes and shrubs) and consumers in its food chain during the months of February and March of 2013 for the analysis of mercury, stable isotopes and, in the case of turtles, genetic defects. Some limnological variables associated with mercury dynamics (pH, DO, temperature, conductivity and DOC) were also measured. Total mercury in all samples was determined, following high temperature digestion, by cold vapor atomic absorption spectroscopy. Stable isotope ratios for carbon and nitrogen in all biological samples were determined, after drying at 60ºC for 48 hours and grinding, by isotope ratio mass spectroscopy. To evaluate genetic defects in turtles, two blood smears for each animal, were prepared and stained with Giemsa in pH 5,6 phosphate buffer. 2000 blood cells from each animal were examined with a compound microscope and the frequency of nuclear anomalies was quantified and expressed as anomalies/1000 cells. A significant difference was found in HgT levels between sample types in turtles (ANOVA F2,70=172 p<0,001), with the highest average concentrations occurring in carapace (3677 ng/g) and claws (3787 ng/g) and the lowest concentration occurring in muscle (406 ng/g) (wet weight). The average values of δ13C for phytoplankton, periphytic algae, shrubs and emergent macrophytes were -32,99 ‰; -34,33‰; -30,70‰ and -30,15‰ and the average values of δ15N were 5,08‰, 7,33‰, 8‰ and 7,29‰, respectively. The average values of δ13C e δ15N in Chelus fimbriata were 11,9‰ and -31,7‰, respectively. On average, the turtles analyzed were 2 trophic levels above plants and derived most of their energy from food chains beginning with phytoplankton and periphytic algae. A significant positive relationship was encountered between body length and HgT in turtles (F = 21,17; r2 = 0,467; p < 0,001). The average frequency of micronuclei encountered in turtle blood was 1,21±0,65/1000 cells. No significant relationship was found between the frequency of nuclear anomalies and HgT in turtle muscle. A significant positive linear relationship was encountered between Log10HgT e δ15N for the entire food chain of Chelus fimbriata (r2 = 0,8009; p < 0,001), indicating a strong and consistent biomagnification of mercury through this system. The equation was: log10Hgt (ng/g) = 0,247+0,2008 δ15N. The average concentration of HgT in the muscle of Chelus fimbriata was below the maximum level recommended by the WHO and the Brazilian Health Ministry for food fish.