Trace Metals Distribution in Tissues of 10 Different Shark Species from the Eastern Mediterranean Sea

: As long-living apex predators, sharks tend to bioaccumulate trace metals through their diet. The distribution of Al, As, Cd, Co, Cr, Cs, Cu, Fe, Mn, Ni, Pb, V, Zn and Hg in different tissues (muscle, liver, heart, gills and gonads) of large-size (58–390 cm) sharks, some of which rare, of the eastern Mediterranean Sea was studied. Trace metals analyses in samples originating from ten different Chondrichthyes species were performed by inductively coupled plasma–mass spectrometry (ICP-MS) and Cold Vapor Atomic Absorption Spectrometry (CVAAS) for Hg. Data on trace metal levels are for the first time reported herewith for the species O. ferox and H. nakamurai . Higher median concentrations of trace metals were generally determined in the liver. The concentrations of Hg, Cs and As in the muscle increased proportionally with body length. Statistically significant differences between sexes were recorded for Hg, Cr, Ni and As ( p = 0.015) in the muscle tissues of P. glauca . Muscle tissue Hg concentrations exceeded the EU maximum limit (1 µ g g − 1 wet weight) in 67% of the individuals sampled, with the highest concentrations detected in O. ferox and S. zygaena , whereas regarding Pb (limit 0.30 µ g g − 1 ww), the corresponding percentage was 15%. Arsenic concentrations were also of concern in almost all shark tissues examined.


Introduction
The eastern Mediterranean Sea is an oligotrophic sea, where Aegean waters, which differ from Ionian and Levantine in nutrient and oxygen concentrations [1], constitute a great habitat for Chondrichthyes species, with only a few being endemic.The Mediterranean Sea has been subject to fishing pressures, particularly near the coastal zone, where shark populations have declined dramatically over the last two centuries [2], as a by-catch of commercial Mediterranean fisheries, representing 1-2% of total landings [3].
Overexploitation of fishing species, principally near costal zones, leads top predators such as sharks towards searching for food at higher levels, resulting in the bioconcentration of metals in their tissues [4].As apex predators, Chondrichthyes are susceptible to bioaccumulation of contaminants such as metals.Metal toxicity may potentially affect growing embryos, having a further impact on shark populations, particularly those which are limited to coastal distributions [5].Anatomy and life cycle features of Chondrichthyes can potentially affect the intake and retention of certain metals.Since non-essential metals are not subject to regulatory mechanisms, their concentrations in organisms can reveal the degree of marine pollution [6].
The increase in demand for seafood, combined with a lack of cheap alternatives, has led developing countries, albeit considering shark meat as a low-quality product, to commercialize it under brands which consumers are familiar with and relate to animals.Hence, several buyers are often unaware of what they consume labelled under these brands.In addition, governments finance the use of shark meat in public school meals as seafood [7].
Due to the growing consumption of sharks and shark products, information on food safety is required, particularly in relation to toxic elements [8].Most studies, however, focus on mercury concentrations [9,10], although studies with toxic and essential metals attract a similar interest [11,12].Regarding Chondrichthyes, available relevant data still remain limited [6,[13][14][15][16]] and more species need to be studied.
The principal aim of the present study was the evaluation of Al, As, Cd, Co, Cr, Cs, Cu, Fe, Mn, Ni, Pb, V, Zn and Hg concentrations in the tissues of 10 species of large mainly pelagic Chondrichthyes, of differing ecology and biology, harvested in the eastern Mediterranean Sea (Greece).Pertinent available data referring to the species Prionace glauca, Isurus oxyrinchus, Sphyrna zygaena, Heptranchias perlo and Hexanchus griseus are limited, whereas regarding the species which are relatively rare in the Mediterranean, Mobula mobular, Alopias supercilliosus and Oxynotus centrina, scarce data are available.For Odontaspis ferox and Hexanchus nakamurai no relevant data are available.The results obtained were compared to those of previous studies and to the limits imposed by European legislation.

Sample Collection and Pretreatment
A total of 33 individual large sharks, consisting of 10 different large sized species, were collected from the eastern Mediterranean Sea, specifically from the Aegean, Ionian and Libyan Seas, between July 2015 and August 2016 (Table 1, Figure S1 Supplemental Material).All specimens were incidentally caught by commercial trawlers or long-liners in Greece and samples were kindly offered by fishermen.Species identification was carried out by Compagno catalogue [17] and morphological parts were recorded individually.
The species identified were the oceanic M. mobular (n = 2), an epipelagic, endemic and endangered species of the Mediterranean Sea considered as rare [3], the pelagic and oceanic species regularly observed in Mediterranean fisheries I. oxyrinchus (n = 1) and P. glauca (n = 14) [18] and the semi-pelagic A. superciliosus (n = 2) distributed in temperate and tropical seas worldwide, with its presence in the Mediterranean considered as uncommon by its sparse records [19].Also examined were O. ferox (n = 2), both a demersal and pelagic species occurring "infrequently but regularly" in Mediterranean fisheries and considered as globally vulnerable; S. zygaena (n = 1), a pelagic and semi oceanic species regularly seen in tropical and warm temperate waters, although not so commonly in Mediterranean waters due to its population's decline in biomass [2]; and H. griseus (n = 5), the presence of which in the Mediterranean Sea is well documented, although it is more common in the western basin [18].The demersal or benthopelagic species identified were O. centrina (n = 1), an uncommon to relatively rare bottom shark for Mediterranean waters [20,21]; H. perlo (n = 4), a bathypelagic species of temperate and warm temperate seas [22], with its presence being well documented in both western and eastern waters and considered as rare in the Mediterranean Sea due to lack of data [19]; and H. nakamurai (n = 1), considered as rare in the Mediterranean Sea [19], although data for this species are insufficient due to misidentification with H. griseus.
For each specimen obtained, the location and date of capture, total length (LT) and gender were recorded (Table 1).In the case of M. mobular, length was recorded as the width of its disc (WD).Three unsexed specimens were characterized due to the lack of pelvic fins.Muscle samples were obtained from all 33 individual sharks and due to the lack of internal organs, 27 gills, 13 livers, 15 hearts and 5 gonad tissues were collected.Throughout the treatment of the samples, ceramic tools were utilized, thoroughly cleaned prior to their use and between each sample processing with ethanol 95% (Merck, Darmstadt, Germany), HNO 3 10% (Merck) and MilliQ ultra-pure water of 18.2 MΩ.cm (Millipore, Bedford, MA, USA).
Samples were weighed, placed into clean plastic vials, sealed and maintained frozen at −20 • C. Tissue samples were subsequently lyophilized in a Virtis Freeze Dryer System (The VIRTIS Company, New York, NY, USA) for 48 h, ground with a pestle and mortar and homogenized.Wet weights to dry weights were calculated by measuring the moisture loss in each tissue type.The average wet to dry weight ratio was 4.2:1 for muscle tissue, 4.6:1 for gill tissue, 4.8:1 for gonad tissue, 5.4:1 for heart tissue and 2.1:1 for liver tissue.

Trace Metals Determination
For the determination of Al, As, Cd, Co, Cr, Cs, Cu, Fe, Mn, Ni, Pb, V and Zn, 0.1-0.2g tissue samples were digested with the addition of 5 mL HNO 3 65% supra pure and 1 mL H 2 O 2 30% supra pure in a microwave digestion system (Anton Paar Multiwave GO Plus, Graz, Austria) [23].Trace metals determination of the digested samples was performed by inductively coupled plasma-mass spectrometry (ICP-MS), employing a Thermo Scientific ICAP Qc (Waltham, MA, USA) instrument.Measurements were carried out in a single collision cell mode, with kinetic energy determination (KED) using pure He.Matrix-induced signal suppressions and instrumental drift were corrected by internal standardization ( 45 Sc, 103 Rh).
For Hg determination, the tissue samples were digested with HNO 3 65% supra pure (Merck), according to the procedure of Meador et al. [24], slightly modified.Mercury measurements were performed by cold vapor atomic absorption spectrometry (CVAAS), employing a Varian SpectrAA 200 with VGA-77 instrument (Varian, Mulgrave, Australia).
Quality assurance of the determinations was provided by the analysis of a certified reference material (CRM) (IAEA-436; tuna fish flesh homogenized, IAEA Reference Materials), with the recoveries of the trace metals measured being in the range 100 ± 10%.Within each batch of samples analyzed, a procedural blank was included, in which the detected analytes concentrations were lower than the respective LODs, with the exception of Zn, where the concentrations were corrected.
All metal concentrations were converted to µg g −1 wet weight (ww) for comparison to those of other studies and to EU legislation limits.The equation used for the conversion to wet weight is Concentration µg•g −1 ww = Concentration µg•g −1 dw x ((100-Moisture loss %)/100).

Statistical Analysis
The Kolmogorov-Smirnov and Shapiro-Wilk tests were used in order to evaluate the normality of the data.Both tests provided p values lower than 0.05 and the null hypothesis (that data are normally distributed) was rejected.Therefore, the Mann-Whitney statistical comparison of values among two groups and Kruskal-Wallis test for more than two groups were used.In the Mann-Whitney and Kruskal-Wallis tests, the null hypothesis is that the medians of two or more groups, respectively, are equal.The tests were two-sided and p values lower than 0.05 were considered statistically significant.Principal component analysis (PCA) was used in order to obtain an overview of the potential relationships existing between the metals and the different tissues studied.For the statistical treatment of the data, values lower than the LOD were assigned the value of LOD divided by two (the percentage of metal concentrations under LOD was less than 10% per tissue).

Trace Metals Concentrations
Mean concentrations exceeding 10 µg g −1 ww were determined for As, Fe and Zn in muscle tissues (M), for Fe in gills (B), for Fe and Zn in hearts (H), for As and Fe in gonads (G) and for Fe, Hg, Zn and As in livers (L).Concentrations ranging between 1 and 10 µg g −1 ww were obtained for Hg and Al in muscle tissues, for Zn, As, Al and Hg in gills, for As, Hg, Cu and Al in hearts, for Zn in gonads and for Al and Cu in livers.For the rest of the trace metals examined in the five different tissues, concentrations lower than 1 µg g −1 ww were measured (Figure 1, Table S1 Supplemental Material).
Among the five different tissues examined, statistically significant differences were recorded regarding the median concentrations of Al (Kruskal-Wallis test, p = 0.003), V (0.003), Cu (0.005),As (0.010) and Mn, Fe, Co, Zn, Cd, Cs (0.001).The case is different for Hg, Cr, Ni and Pb (Kruskal-Wallis test, p > 0.05).Higher concentrations of most metals were generally determined in the liver tissue, especially of H. griseus and H. perlo, for which more samples were available (Figure 1).Metals are absorbed in blood and then distributed to target organs with high metal-accumulating capacity, such as the liver.The liver is a lipid store accumulating lipophilic toxic metals, considered the main target organ of toxic and essential metals, also having a detoxification role [26,27].The high levels of MT (metalothionein) contained in the liver indicate binding of metals, reducing their availability for participation in metabolic interactions [28,29].
Data on metal levels are for the first time reported herewith for the species O. ferox and H. nakamurai, whereas for O. centrina and A. supercilliosus, available data are scarce [30,31] (Table S1).The M. mobular (WD = 285 cm) sample, a species for which data are rare for the Mediterranean [32], which remains severely understudied (especially concerning metal evaluations) [33], exhibited the highest concentrations of Fe, Zn, As and Cs in the muscle tissue and of V in the gills.These values are significantly higher compared to those of other species and of the other examined specimen of M. mobular (WD = 203 cm).For H. perlo (TL = 100 cm) the highest values of Cr and Ni were exhibited in the liver.The case was similar also for Co and Cd in H. perlo (TL = 111 cm) and for Hg in H. griseus (TL = 238 cm).Regarding P. glauca (TL = 129 cm, 266 cm), the highest values of Cu and Pb were measured in the muscles and gills, respectively.In the gills of O. ferox, the highest value of Mn was found.Trace metal concentrations measured in the gills of H. griseus (TL = 247 cm) were lower than the limits of detection, excluding Hg, V, As, Cd and Cs.High Pb concentrations were observed only in individual samples of P. glauca gills and H. griseus muscle.

Trace Metals Concentrations
Mean concentrations exceeding 10 µg g −1 ww were determined for As, Fe and Zn in muscle tissues (M), for Fe in gills (B), for Fe and Zn in hearts (H), for As and Fe in gonads (G) and for Fe, Hg, Zn and As in livers (L).Concentrations ranging between 1 and 10 µg g −1 ww were obtained for Hg and Al in muscle tissues, for Zn, As, Al and Hg in gills, for As, Hg, Cu and Al in hearts, for Zn in gonads and for Al and Cu in livers.For the rest of the trace metals examined in the five different tissues, concentrations lower than 1 µg g −1 ww were measured (Figure 1, Table S1 Supplemental Material).
Among the five different tissues examined, statistically significant differences were recorded regarding the median concentrations of Al (Kruskal-Wallis test, p = 0.003), V (0.003), Cu (0.005),As (0.010) and Mn, Fe, Co, Zn, Cd, Cs (0.001).The case is different for Hg, Cr, Ni and Pb (Kruskal-Wallis test, p > 0.05).Higher concentrations of most metals were generally determined in the liver tissue, especially of H. griseus and H. perlo, for which more samples were available (Figure 1).Metals are absorbed in blood and then distributed to target organs with high metal-accumulating capacity, such as the liver.The liver is a lipid store accumulating lipophilic toxic metals, considered the main target organ of toxic and essential metals, also having a detoxification role [26,27].The high levels of MT (metalothionein) contained in the liver indicate binding of metals, reducing their availability for participation in metabolic interactions [28,29].Data on metal levels are for the first time reported herewith for the species O. ferox and H. nakamurai, whereas for O. centrina and A. supercilliosus, available data are scarce [30,31] (Table S1).The M. mobular (WD = 285 cm) sample, a species for which data are rare for the Mediterranean [32], which remains severely understudied (especially concerning metal evaluations) [33], exhibited the highest concentrations of Fe, Zn, As and Cs in the muscle tissue and of V in the gills.These values are significantly higher compared to those of other species and of the other examined specimen of M. mobular (WD = 203 cm).For H. perlo (TL = 100 cm) the highest values of Cr and Ni were exhibited in the liver.The case was similar also for Co and Cd in H. perlo (TL = 111 cm) and for Hg in H. griseus (TL = 238 In the principal component analysis (PCA) performed for trace metals, in each tissue individually studied, the value of the Keiser-Meyer-Olkin (KMO) statistical criterion was calculated equal to 0.835 in the muscle tissue and <0.5 for the other tissues.In the PCA for trace elements in the muscle tissue, two principal components (PCs) with eigenvalues > 1 were extracted, where PC1 explained up to 67.1% of the total variance and PC2 explained up to 9.1% (Table S2).According to the component plots obtained (Figure 2), a strong correlation of all metals in muscles was derived, excluding Hg, Pb, Cu and Cr.For the rest of the tissues investigated (liver, heart, gills), a scatter plot is presented, showcasing these appearing even higher in gonads.On the basis of the aforementioned strong correlation, Fishes 2024, 9, 77 the muscle appears to be the most appropriate tissue for the evaluation of metal levels and generally of their bioconcentration in sharks.This stems from the fact that the muscle is not an active metabolic tissue, such as the gills or liver, capable of accumulating metals, excluding Hg [34].
of the tissues investigated (liver, heart, gills), a scatter plot is presented, showcasing these appearing even higher in gonads.On the basis of the aforementioned strong correlation, the muscle appears to be the most appropriate tissue for the evaluation of metal levels and generally of their bioconcentration in sharks.This stems from the fact that the muscle is not an active metabolic tissue, such as the gills or liver, capable of accumulating metals, excluding Hg [34].

Muscle Gill Heart Liver
Fishes 2024, 9, x FOR PEER REVIEW 8 o Regarding Hg, quite high concentrations in muscles were detected in O. ferox (m concentration 13 µg g −1 ww; n = 2) and S. zygaena (9.1 µg g −1 ww; n = 1), whereas in other species examined, the mean concentrations were lower than 4 µg g −1 ww (Table High Hg concentrations may reflect a diet of large carnivorous teleost fishes and ot shark species [35], which are a great proportion of O. ferox's [36] and S. zygaena's [37] d which may explain the differences characterizing the specific species compared to others studied.High Hg levels in the muscle of S. zygaena are linked with interspec biological mechanisms enhancing metal retention in muscles [16].Regard metallothionein detoxification efficiency in liver, these metalloproteins seem to be poo involved in Hg detoxification, since the total Hg bound to the heat-stable fraction was v low, as also previously reported for dolphins [38].However, most of the spe investigated exhibited higher Hg levels in the liver in comparison to the muscles (liv Regarding Hg, quite high concentrations in muscles were detected in O. ferox (mean concentration 13 µg g −1 ww; n = 2) and S. zygaena (9.1 µg g −1 ww; n = 1), whereas in all other species examined, the mean concentrations were lower than 4 µg g −1 ww (Table S1).High Hg concentrations may reflect a diet of large carnivorous teleost fishes and other shark species [35], which are a great proportion of O. ferox's [36] and S. zygaena's [37] diet, which may explain the differences characterizing the specific species compared to the others studied.High Hg levels in the muscle of S. zygaena are linked with interspecific biological mechanisms enhancing metal retention in muscles [16].Regarding metallothionein detoxification efficiency in liver, these metalloproteins seem to be poorly involved in Hg detoxification, since the total Hg bound to the heat-stable fraction was very low, as also previously reported for dolphins [38].However, most of the species investigated exhibited higher Hg levels in the liver in comparison to the muscles (liver > muscle > gills > heart > gonads), with H. griseus having the highest value, presumably attributed either to recent exposure to Hg, accompanied by subsequent accumulation in the liver, or to strong binding of Hg to MT [39].Mercury is highly species specific and the differences characterizing only H. perlo and not H. griseus or P. glauca may be attributed to prey preferences [40], local environmental conditions [41], habitat, the capability to induce metal-binding proteins such as MT [42] and physiological differences in metal digestion and metabolic capacity of methylmercury (MeHg) [43].Hg concentrations may increase with the preference of the benthic diet to the pelagic one and at higher levels of the trophic web [44].Low hepatic Hg may be linked to Se excess, in comparison with Hg, which is related to Hg detoxification in marine organisms [29,45].Hauser-Davis et al. (2021) [29] calculated Se:Hg molar ratios of 12:1 in liver and 1.5:1 in the muscles of blue sharks, indicating excess Se in relation to Hg in both tissues, denoting another possible Hg detoxification pathway, in addition to MT metal binding.This is in contrast with other studies, e.g., of Alves et al. (2023), where all blue sharks sampled presented negative Se/Hg ratios in the muscles.This difference may be due to differences in sex, age and geographic distribution of the assessed blue shark specimens [46].According to Kazama et al. (2020) [47], the detoxification of MeHg in the liver of blue shark is highly developed, explaining the lower Hg concentration.

Gonads
Variations in metal concentrations can be observed among species, even of the same genus, being attributed mainly to interspecific regulation mechanisms [48].Among the different shark species studied, statistically significant differences in metal concentrations in muscle tissues were recorded for Hg, Cd, Co and Cs (Kruskal-Wallis test, p = 0.014, 0.025, 0.028, 0.039, respectively).The differentiation characterizing Co concentrations among species may be attributed either to ineffective metabolic regulation of basic metals, the concentrations of which are affected by environmental factors, or to different physiological requirements of the species despite the effective regulation [49].
The trace metals V, Fe, Cd and Cs exhibited liver differentiation in P. glauca, H. perlo and H. griseus (Kruskal-Wallis test, p < 0.05).Food ingestion does not represent a V source and in accordance to relatively low V concentrations detected, it does not appear to have any prominent ecological or biological role [49].
Iron is essential to living organisms as a constituent of many enzymes, electron transfer complexes and oxygen carriers.Higher Fe values were detected in the liver and gills, partially attributed to blood irrigation.The blood of Chondrichthyes, compared to other marine organisms, has hemoglobin, an Fe-containing molecule.Gills receive a great amount of blood for oxygen transportation.The liver has an important role in blood purification processes.In addition, exposure to Fe can be enforced by nutrient absorption from dietary paths via bioaugmentation, engendering high Fe levels in local sharks [50].
Cephalopods are considered as the main factor of Cd transfer to apex predators.High bioavailability of Cd in their digestive gland cells suggests a high prospect of trophic transport of the metal to their predators [51].Increases in Cd concentrations induce the synthesis of metallothioneins in the liver, with Cd successively binding to MT along with Zn and Cu [52].
Cs is ingested from fishes through the water phase and is transported to upper levels of the food web [53].Indications of its biomagnification in the marine food web have been reported [54], albeit no biological function is known for Cs so far.However, it has been reported to act as proportional to plasma K + [54].
Manganese is an essential element participating in many metabolic activities.Absorption from the water column constitutes an insufficient explanation of its presence in internal tissues [55]; nevertheless, diet contribution, together with Mn absorption from the environment, may explain the increased values detected [56].Copper provokes a protective response by inducing MT synthesis in the liver [57].It can be toxic under specific circumstances, causing swimming alterations and even mortality [50].
The concentrations measured for Ni, as well as Cr, despite their essential role, were relatively low.Chromium differentiations occurred only among H. perlo's tissues, whereas hepatic Ni variations were found also in H. perlo.Mehrim [58] suggests that Cr exerts a positive effect on fish health, since it inhibits pests growth.Elevated Ni concentrations in ion-regulating organs such as gills do not jeopardize their function but indicate Ni absorption by Chondrichthyes.Ni is responsible for distractions in osmoregulation, with increases in plasma K + and urea [57].Concentrations of As may vary among species due to endogenous factors or target-specific diets, with some species accumulating As more than others [15].Regarding Pb, its levels in Chondrichthyans tissues remain quite low.Gills are the organs usually affected by Pb [50].

Correlation with Gender
Statistical analyses between gender (F, females; M, males) and trace metal concentrations were conducted using only P. glauca data due to the insufficient sampling size of the rest of the species examined.Statistically significant differences of median values between genders (F: n = 6; M: n = 5) and metal concentrations were found for Hg, Cr, Ni and As (Mann-Whitney, p = 0.015) for muscle tissues.A comparison of median concentrations indicated higher trace metal values in female specimens.The differentiation of metal concentrations in Chondrichthyes due to gender can be attributed to diet preference, season change in diet or level of maturity, especially in cases of mature pregnant females [59].Some species can transfer metals during spawning or deliver them directly to the fetus [60].In accordance with the findings of the present work, significantly higher concentrations of Hg and As were detected in females than males.Females have a higher degree of growth, being larger than males (sometimes up to 40%), exhibiting racial dimorphism that may reflect a higher degree of pollutants bioaccumulation [12].

Correlation with Size
The regression analysis for all sizes of species combined exhibited a positive correlation in the muscles between total length (TL) and Hg concentrations (R = 0.464, p = 0.007, n = 33) and in the gills between TL and Fe (R = 0.436, p = 0.023, n = 27).For a further explanation of the fitted model, regression analysis was conducted individually in P. glauca, with the majority of samples in the muscles (n = 14) and gills (n = 9).Regression analysis exhibited a moderately strong relationship between total length and concentrations of Hg (R = 0.816, p < 0.001), As (R = 0.718, p = 0.004) and Cs (R = 0.551, p = 0.041) in the muscles and of As in gills (R = 0.837, p = 0.003) (Figure S2).Riesgo et al. (2023) [61] also reported a positive correlation of Hg with size in muscle samples of P. glauca.A positive correlation between muscle tissue and size reveals metal bioaccumulation [62], while a negative one presents a small representation of all length values [63].Additionally, a negative correlation in muscle tissue is linked with different absorption and more efficient secretion rates [64].A lack of essential metals and size correlation suggests that requirements of essential elements do not vary with size, consequently nor with age [49].

Comparison with Literature
The metal concentrations detected in the present work regarding H. perlo, H. griseus, P. glauca, M. mobular, I. oxyrinchus and S. zygaena, for which several samples were examined, are generally within the range of values reported for the same species worldwide (Table 2).However, the Hg levels in sharks from the eastern Mediterranean appear to be higher than those detected in the same species from other areas, such as the Atlantic Ocean [29,65], the Pacific Ocean [66,67] and western Italian coasts [68].Correspondingly, the As levels were higher in sharks from the Mediterranean than in those from other regions [11,35,69] but differed even among sharks originating from different Mediterranean regions [14,15,70].Overexploitation has affected the life strategies of Chondrichthyes, which in order to adjust are forced to enter the reproduction age earlier, facing the consequence of a maximum size reduction.According to Ferretti et al. [2], a gradual decrease in size of Mediterranean Chondrichthyes over time is observed, with the average size of Mediterranean sharks being the smallest compared to that recorded in other regions worldwide [19].Consequently, an individual species from the Mediterranean is expected to be older and characterized by a higher metal content compared to identical species of the same size from other geographical regions.

Legislation Limits
According to the European legislation's [87] maximum levels of contaminants in foodstuffs, 67% of the raw muscle shark tissue samples examined had concentrations exceeding the maximum limit for Hg (1 µg g −1 ww) (Figure 3) and 15% for Pb (limit 0.30 µg g −1 ww), whereas were 0% for Cd (limit 0.050 µg g −1 ww).Although As limits are not included in the European legislation, the As concentrations determined in all shark tissues were considerable in comparison to the rest of the trace metals examined.Nevertheless, even in cases where legislative limits are not exceeded, shark meat consumption should be still considered with caution.
Mediterranean sharks being the smallest compared to that recorded in other regions worldwide [19].Consequently, an individual species from the Mediterranean is expected to be older and characterized by a higher metal content compared to identical species of the same size from other geographical regions.

Legislation Limits
According to the European legislation's [87] maximum levels of contaminants in foodstuffs, 67% of the raw muscle shark tissue samples examined had concentrations exceeding the maximum limit for Hg (1 µg g −1 ww) (Figure 3) and 15% for Pb (limit 0.30 µg g −1 ww), whereas were 0% for Cd (limit 0.050 µg g −1 ww).Although As limits are not included in the European legislation, the As concentrations determined in all shark tissues were considerable in comparison to the rest of the trace metals examined.Nevertheless, even in cases where legislative limits are not exceeded, shark meat consumption should be still considered with caution.
The combined exposure of Hg with metals such as Al, Cu, Pb, Cd and Mn may act synergistically with other factors [8], causing such dysfunctions in metabolic activities and biological processes that a single metal, even in high doses, could not engender [88].In order to quantify the limits of consumption, the hypothesis was made that the dose of consumption is equal to contaminant absorption and that it is not affected by thermal treatment [89].However, Matos et al. [90] demonstrated differences in metal concentrations following culinary treatments, with total Hg concentrations in the muscle tissue of P. glauca following the decreasing order of grilled (3.57± 0.96) > steamed (3.12 ± 1.37) > raw (2.25 ± 0.71).Kalogeropoulos et al. [91] found that the concentrations of metals in heat-treated fish and shellfish followed a similar sequence of fried > baked > raw.The combined exposure of Hg with metals such as Al, Cu, Pb, Cd and Mn may act synergistically with other factors [8], causing such dysfunctions in metabolic activities and biological processes that a single metal, even in high doses, could not engender [88].In order to quantify the limits of consumption, the hypothesis was made that the dose of consumption is equal to contaminant absorption and that it is not affected by thermal treatment [89].However, Matos et al. [90] demonstrated differences in metal concentrations following culinary treatments, with total Hg concentrations in the muscle tissue of P. glauca following the decreasing order of grilled (3.57± 0.96) > steamed (3.12 ± 1.37) > raw (2.25 ± 0.71).Kalogeropoulos et al. [91] found that the concentrations of metals in heat-treated fish and shellfish followed a similar sequence of fried > baked > raw.

Conclusions
The results of the present study contribute to the metal concentration database of several Chondrichthyes species, for which information is limited due to their either scarce or uncommon occurrence.A wide variation in metal concentrations among tissues, species and genders was highlighted, together with positive correlations between Hg concentrations and total length in muscle tissues.Regarding P. glauca, correlations among Hg, As and Cs in the muscles suggest a possible bioaccumulation of metals with age.The food chain is the main route of uptake for most metals, which are either bioaccumulated or biomagnified, whereas their soluble phase may affect the metal content of their tissues.Mediterranean Chondrichthyes are exposed to higher Hg levels, compared to those originating from other marine areas, due to natural geochemical processes and the occurrence of anthropogenic sources.Indicatively, in >50% of the individual samples examined, the Hg concentrations measured in the present work exceeded the European legislation maximum limit.

Figure 2 .
Figure 2. Principal component analysis (PCA) of trace metals in shark tissues examined.

Figure 2 .
Figure 2. Principal component analysis (PCA) of trace metals in shark tissues examined.

Figure 3 .Figure 3 .
Figure 3. Mean Hg concentrations (µg g −1 ww) in muscle tissue of the species examined.The dotted line indicates the European legislation maximum limit set at 1.0 µg g −1 ww.

Table 1 .
Shark species collected, classified per family.

Table 2 .
Comparison of trace elements concentrations (µg g −1 ww) measured in different tissues of H. griseus, H. perlo and P. glauca, with available literature.