3.1 Results
3.1.1 THg content
The average carapace length for females was 74.42 cm (n=14) and 47.18 cm (n=16) for males. The present results show that although the female carapaces were larger than those of the males, higher THg concentrations were not observed in the female animals. The muscle tissue samples showed the highest concentrations of THg (0.6929mg.kg-1) as compared to maximum values in the skin of (0.5457mg.kg-1) and in the brain of (0.0610mg.kg-1). The mean THg concentrations found in the skin (n=28), muscle (n=19) and brain (n=2) were, respectively, 0.1045 mg.kg-1, 0.1092 mg.kg-1, and 0.0601mg.kg-1. The results are shown in Table 1.
Regarding the tissues, the muscle should be considered the most important, because of its higher accumulation of THg in P. expansaas compared to the skin, but the present research found no significant difference (P<0.05). However, a notable difference was detected in the tissue concentrations, considering a minimum of "not detected" in the muscle and skin and a maximum of 0.6929mg.kg-1 in the muscle. The average values found in the male samples demonstrated greater THg bioaccumulation in the male muscle than in the female muscle, and a significant difference (P<0.05) was observed between the female muscle samples and the samples of the other tissues.
A positive, gender independent correlation was observed (R2=0.87) between the skin and muscle concentrations. Also, the correlation was positive (R² = 0.50) between tissues (muscle and skin) for males, but no such correlation was observed for females (R² = 0.43). The present research found no correlation between carapace length and THg concentrations in the muscle or skin for males (R² = 0.011) or females (R² = 0.15), possibly due to the high variation between the tissue concentrations in the animal diet and the diversity of origin.
3.1.2 Human health risk assessment
Amongst those interviewed, 18% (15/91) consumed chelonian meat, especially tracajá (P. unifilis) and the Giant South American river turtles (P. expansa). However, only 8% (7/91) reported consuming the eggs. Regarding the consumption data used to calculate the HQ, 8/91 ate chelonian meat at least once a year, 5/91 ate it once a month, and 2/91 once a week during through the egg-laying months (September to December). The people interviewed confirmed that these animals and eggs were caught in Vila Balbina, Presidente Figueiredo, Amazon.
The Hazard Quotient (HQ) was calculated for the different consumption intervals (annually, monthly, and weekly) with respect to the Thigh concentrations in male plus female P. expansa, and the mean THg values for both genders separately. Table 2 shows the HQ results obtained for the Human risk assessment of P. expansa muscle.
3.2 Discussion
3.2.1 THg content
The total mercury content in the different turtle tissues is related to Hg kinetics. Researchers have indicated that the total mercury content in freshwater fish is related to the species and habitat evaluated (Yoshimoto et al., 2016, Zang et al., 2020). Brazilian studies have reported a THg content in Amazon fish muscle of over 0.500 mg kg−1, above the limit recommended by World Health Organization (Custódio et al., 2020, Queiroz et al., 2018, WHO, 2003). Although there are few studies with Amazonian freshwater chelonians, it is essential to consider the influence of the environment on the contents of trace elements.
The P. Unifilis caught in the lower River Xingu accumulated lower concentrations of THg throughout its life when compared to other species of Amazon turtles. Thus, it appears that the Amazon basin rivers have mercury in a bioavailable form that can be accumulated by organisms. Recent studies have shown that P. unifilis and Podocnemis expansa are predominantly herbivorous, but may consume insects, crustaceans, and molluscs (Balensiefer and Vogt, 2006, Vogt, 2008). The present authors observed that some studies indicate a difference in feeding preference between genders, with the females consuming more fruits and seeds and the males more stalks and shoots (Teran et al., 1995). Thus, the authors postulated that the difference in Hg concentrations between males and females could be explained by the absence of difference in trophic levels between species gender (Lara et al., 2012). Frossard et al. (2021) identified a maternal transfer of metals (essential or not) to the eggs in tests with udines nesting eggs, triggering genotoxic effects in the hatchlings. Research carried out by Schneider et al. (2011) evaluated non-lethal and non-invasive ways of collecting samples from Amazonian chelonian species for Hg monitoring. They found no significant differences between the carapace and internal organs of P. erythrocephala and P. sextuberculata. In the present study, the THg concentrations in the turtle skin and muscle indicated that anthropic actions might impact animals in different environments.
Different THg concentrations may be observed between males and females due to changes in eating habits (Schneider et al., 2009) and the elimination of Hg via the eggs during the reproductive period (Jagoe et al., 2003, Frossard et al., 2021). However, in contrast, the present study diverged from this observation and the THG concentration for males did not significantly differ from that of the females. Schneider et al. (2010) found 0.0624 mg.kg-1 THg in P. expansa muscle samples, 0.432 mg.kg-1 THg in C. fimbriatus muscle and Schneider et al. (2009) found 0.033 mg.kg-1 THg in P. erythrocephala muscle. These animals present distinct eating behaviors between the genders and life periods. Small amounts of fish and crab were observed in studies that analyzed the chelonian gastrointestinal content, but their primary food source is seeds and fruits, especially those from riverside trees and flooded forests. However, the literature also reported that juvenile turtles prefer to eat fish and other meat.
3.2.2 THg bioaccumulation
Toxic metals entering the river body would be absorbed in sediments and migrate to the sediment and biota through a biological and chemical process (Eagles-Smith et al., 2016, Windsor et al., 2019), so researchers began to pay more attention to pollution by toxic metals (Kronberg et al., 2016, Pignati et al., 2018, Yi et al., 2008).
The bioaccumulation of toxic metals by reptiles results from surface contact with the water and food chain (Schneider et al., 2011, Zupo et al., 2019), and the concentrations depend on the environmental levels of trace metals in the habitat, since they are absorbed and stored in the tissues (Yi, Yang, & Zhang, 2011). Researchers have shown that the mercury concentrations in the muscles of freshwater animals varied widely depending on the capture location (Eagles-Smith et al., 2016, Zang et al., 2020).
Captive breeding possibly contributed to reduced mercury concentrations when compared to the data reported by Schneider et al. (2014). The researchers evaluated the THg concentrations in the muscles and keratin samples from the carapaces of free-living P. expansa in the Amazonas region and found THg concentrations in the muscle of 0.1 mg.kg-1 and in the carapaces of 2.8 mg.kg-1. However, the diet and food consumption were human-controlled, thus interfering with the Hg kinetics.
In addition, the muscle THg concentration is related to eating habits, which interferes with the biomagnification of this trace element. On carrying out research on the THg concentrations in Amazon chelonian muscles from different species, Scheneider et al. (2010) found concentrations of 0.106 mg.kg-1 in P. expansa, corroborating the results of the present research.
C. fimbriatus, a predatory carnivorous animal, is adapted for collecting fish and other animals in the aquatic environment (Lemell et al., 2019) via a well-developed suction system and head movements in the direction of the prey (Lemell et al., 2010). In contrast, when raised in captivity, young P. expansa prefer meat and fish (Ojast et al., 1971), which may explain the higher average THg concentrations found in C. fimbriatus specimens as compared to P. expansa.
Different Amazonian reptiles have been used to monitor THg bioaccumulation. Eggins et al. (2015) evaluated this trace metal in various C. crocodilus tissues, this being an alligator from the River Purus (Amazonas state), and found a THg concentration of 0.389 mg.kg-1. Correia et al. (2014) evaluated THg in alligators from the Mamirauá reserve (Tefé, Amazonas State), and found THg concentrations of 0.407 mg.kg-1, these animals showing predatory eating behavior. Thus, the THg concentrations represent the biomagnification principle, which is associated with the consumption of fish, chelonians, small mammals, and birds.
3.2.3 Human Health Risk Assessment
Although the established PTWI limits (WHO 1990, WHO 2017) suggest secure mercury intake values, researchers and global health organizations highlight that this metal is dangerous at any concentration due to its bioaccumulation characteristics, and thus there is no healthy mercury level for humans (Manavi and Mazumder, 2018, US EPA, 1997). Bioavailability can be affected by controlling the biotic and abiotic factors. The most critical tool that can be used to reduce mercury in the environment and in animals is the prevention of pollution, which is considered a long-term goal (Burger and Gochfeld, 2011). Balbina village (Vila Balbina), in Amazon State, is highly influenced by mining and damming, both anthropogenic activities on the River Uatumã causing environmental impact and interfering with the Hg biological cycle (Kehrig et al, 1998).
The maximum tolerable limit of 0.5 mg MeHg.kg-1 for fish (except for predatory fish) (FAO, 2016) is established for seafood based on a weekly consumption. The Brazilian regulatory agencies also set the acceptable THg level for non-predatory fish, crustaceans, mollusks, cephalopods, and bivalve mollusks at 0.5 mg.kg-1, and at 1.0 mg.kg-1 for predatory fish (Brasil, 2019). Although the agencies did not determine a THg limit for chelonians, considering the limit established for fish, the THg concentration found in P. expansa muscle (0.1045 mg.kg-1) did not exceed the legislation values, although two turtles presented THg concentrations above 0.5 mg.kg-1.
A Hazard Quotient (HQ) above 1 (HQ> 1) represents a potential health risk to consumers (US EPA, 2009). To determine the HQ for the consumption of P. expansa, it is crucial to consider the weekly consumption of the male categories. When considering the mean THg concentration in males and females for monthly or annual consumption, the HQ was <1 in all categories. Research carried out by Correia et al. (2014) evaluated the HQ for alligator consumption by riverside populations and determined a value of 7.19, considering this quotient due to the higher Hg concentrations in these animals as compared to P. expansa, since they show predatory feeding behavior and are piscivores.
One of the characteristics of the consumption of turtles in the Amazon area is its seasonality, with the hunting of turtles and other animals for consumption being more intense in preserved and isolated areas (Isaac et al., 2015). In the present study, 18% (15/91) of the chelonian meat consumers interviewed considered it rare and occasional, only occurring in months with greater availability. Turtle consumption is taboo among the Amazonian population due to preservation practices of these species and constant supervision by environmental agencies (Silva, 2007). Although in this research, only a small percentage of individuals reported consuming turtle meat and eggs, the true consumption might be higher. Female specimens are more consumed since they are easier to capture, their meat is more appreciated, and the yield is higher (Miorando et al, 2015).
There is little information available on the risk evaluation associated with chelonian consumption. The authors highlighted that total mercury concentrations in river fish from Malaysia and Ghana were influenced by the habitat, feeding habits, species, and body length, and the increased HQ was due to the nearby gold mining area (Doke and Gohlke, 2018). Brazilian researchers have reported HQ values ranging from 1.5 to 28.5 in the River Amazon and concluded that the elevated risk was associated with artisanal gold mining areas, where people also catch fish (Castilhos et al, 2015). They suggested population risks at almost the sampling points due to consumption. It appears there is considerable contamination of living beings that share the same environment with turtles due to their consumption without considering food safety (Manavi and Mazumder, 2018). Hence, determining the allowable limit for the consumption of these chelonians (daily, weekly or monthly) is crucial to preserve public health.