3.1. Level of Potentially Toxic Inorganic Elements in Marine Sediments
The concentrations of the inorganic elements (As, Cd, Cr, Cu, Hg, Ni, Pb and V) in sediments from the investigated area are shown in
Table 4. The study, involving only two sediments, revealed that both have high concentrations in V (14.89 mg kg
−1 and 27.17 mg kg
−1), As (10.76 mg kg
−1 and 13.44 mg kg
−1) and Ni (5.37 mg kg
−1 and 11.11 mg kg
−1) of apparent industrial origin.
The highest concentrations of potentially toxic elements in the sediments have been found in the internal area of the Port due to a lower dispersion capacity of the sedimented material and a greater contribution of municipal sewage containing such pollutants. In fact, as data shows, the levels of potentially toxic inorganic elements had a different distribution profile in the two samples tested and the sediment SED 2, characterized by a dark-gray color and the persistent odor of hydrocarbons and hydrogen sulfide, showed higher concentrations of potentially toxic elements. The analytical data indicate that the levels of Pb, Cr, Ni, V and Cu in SED 2 were approximately two times higher than those of SED 1 and these results evidence a progressive increase in concentration of potentially toxic species from the outermost zone to the innermost one. In addition, the levels of Pb and Ni in SED 1 were comparable with previous studies in the Strait of Sicily [
39]. In both sampling sites, the Cd levels were significantly exceeded (about 10 times) with respect to the Coppola [
39] investigation, indicating anthropogenic contamination of sediments in the area.
To evaluate the pollution and assess the sediment quality, the classification of the Environmental Quality Status (EQS) of sediments was held according to the indications of the European directive [
40], implemented with DL 152/2006 and DM 260/2010 [
41,
42]. Comparing the results with the level established by the Ministerial Decrees [
42] we highlight that Cd in sediments shows the most serious contamination among all the potentially toxic inorganic contaminants considered, exceeding about two time the limits in both samples. Contamination degree of As in sediment was also high, especially in the SED 2 sample in which the legal limits were exceeded. Nevertheless, these levels were lower than those reported for other strongly anthropized port areas of Sicily [
43] and comparable with those recently observed in the Strait of Sicily [
39]. Most likely, the discovery of potentially toxic elements in the sediments is possibly due to impacts deriving from specific anthropic stresses such as the presence of large industrial settlements that persist on the Strait of Sicily. However, the environmental outcome and the spread of potentially toxic elements in the sediment is linked to the performance of multiple competing variables such as: settling times of individual metals, effects of migration currents along the low sandy coast of the Strait of Sicily, sediment accumulation times linked to the sediment granulometry, preferential bioaccumulation processes for some metals (e.g., lead and cadmium) in living organisms. Probably, the presence of mercury in the sediments of the study area is attributable to the activities linked to the Chlorine-Soda plant in Gela, now closed, which for years poured mercurial residues into the sea. The low content in the analyzed sediments can be explained considering that the mercury possibly stratified in deeper layers than those collected. The high vanadium content is most likely not only due to its natural origin and could also be attributed to the petroleum activity of Gela. Generally, we could assume that in this marina area a correlation exists among the concentrations of potentially toxic elements that were present in sediments in higher levels. It may be inferred that such a relationship could be more significant in locations which are more affected by anthropogenic sources, in this case petroleum derivatives and antifouling paints used for boats in the Gulf of Gela.
3.2. Level of Potentially Toxic Inorganic Elements in Sea Urchins
The results on the considered potentially toxic elements in
P. Lividus gonads collected from Gela are shown in
Table 5. The data were compared with control samples from areas not contaminated by industrial activities such as “Marina di Ragusa” and “Piraino”. All the data regarding the concentrations of inorganic elements in fish samples are expressed in mg kg
−1 of wet weight, as indicated by the Commission Regulation N°1881/2006 [
44].
The 30 sea urchin samples were analyzed and grouped by geographic area, each group consisting of 10 samples. As the data show, the samples from Gela were the most contaminated and the average concentration values of potentially toxic elements in the sea urchins from this area were in the order: Pb> As> Cr> Ni> Hg≥ Cd. Moreover, the values evidenced that sea urchins from Gela had the highest mean concentrations of As (6.17 mg kg−1), Cu (2.31 mg kg−1) and Cr (2.29 mg kg−1), compared to samples from “Marina di Ragusa” and “Piraino”. On the contrary, the Cd content (0.02 mg kg−1) was about 60 times lower than that found in samples from “Marina di Ragusa” in which V (0.94 mg kg−1) and the highest mean levels of Ni (1.18 mg kg−1) were determined.
The scientific literature indicates that
P. Lividus is an organism suitable to cover the role of biological-biochemical indicator due to its wide distribution, easy collection, alimentary and filtering habits, and good tolerance to polluting substances. It has been used in several local pollution studies as a bioindicator of heavy metal contamination in the marine environment [
29,
30,
31,
45,
46,
47,
48].
Besides being compatible with the results of the survey carried out in this study on marine sediments, the inorganic contaminants detected in the sea urchin gonads and, in particular, the levels of Ni, V, As and Hg have a possible anthropogenic origin and could be linked to the activities of the Petrochemical plant. The samples taken at “Piraino” were least contaminated and this is consistent with the absence of industrial anthropogenic activities; instead, contrary to what we expected, the samples from “Marina di Ragusa” were not absolutely pristine and this could probably be due to the high vessel traffic, especially during the sampling.
The concentrations determined for Pb and Hg in urchins from Gela were higher as compared to some data in the literature [
45,
47,
48], however, our values were comparable to those determined by Salvo et al. [
29] in other urchin samples from Gela. The Cd levels in Gela samples were on average about 10 times lower than the values reported by Warnau et al. [
48] but comparable with other data [
47]. The Cr content, on average, was comparable to an earlier report [
48] and to values determined by Salvo et al. [
29]. Cu concentrations in urchins from Gela were much higher than those reported by Warnau et al. [
48] but comparable with other data from the Mediterranean area [
29,
47]. The Ni and V results in
P. Lividus from Gela were comparable with previous data determined by Salvo et al. [
29] which, furthermore, are the only data reported in the literature for the Mediterranean area.
European food legislation does not set any limits for heavy metals in sea urchins; however, after comparing the content of Pb, Cd and Hg with the limits established by the European Community for bivalve mollusks [
44,
49], it emerged that Pb in sea urchin samples from Gela was 15 times higher than the limits established for fish safety criterion (1.5 mg kg
−1). On the other hand, according to the regulation, the investigated samples from Gela were characterized by mean concentration of Hg and Cd (0.03 and 0.02 mg kg
−1, respectively), well within the maximum content set at 0.5 and 1.0 mg kg
−1 for mercury and cadmium, respectively.
3.3. Level of Potentially Toxic Inorganic Elements in Fishes
The concentration of potentially toxic elements in fish samples allow us to provide double information: one of environmental nature linked to the behavior of some species as bio-indicators, and another one linked to the food safety of the product intended for human consumption. Therefore, to rationalize the results of the research on fish samples, the data were divided into two Tables.
In
Table 6, the analytical results regarding the muscles of the collected mullets (
Mullus barbatus) are grouped together. This species has a high economic and commercial interest for the Gela city. It is a benthic and territorial fish and is indicated by FAO/UNEP [
50] as a species suitable for monitoring heavy metal pollution. The samples of mullets have been divided into two groups according to their size (Mullet 1 and Mullet 2) since the dimensions are recognized to be very important in the processes of absorption, distribution and elimination of pollutants [
51]. The comparative analysis of potentially toxic elements between the two mullet groups confirms the importance of the dimensions in the metals accumulation: larger specimens (Mullet 2) show higher concentrations than the smallest ones (Mullet 1). This confirms that bioaccumulation is higher as the size increases, or in other words, old and larger fish generally have higher pollutant levels than young and small specimens [
3,
52,
53].
The average concentration values (
Table 6) of the inorganic elements detected in the mullets were in the following order: As > Cd > Pb > Ni > Cr > V > Hg > Cu.
The mean concentration of inorganic elements detected in
Mullus b. from Gela were compared with the data obtained in previous studies carried out on the same species, with similar dimensions and in uncontaminated areas of the Mediterranean Sea [
53,
54,
55,
56], as well as with the law limits available, currently only for Cd, Hg and Pb [
44,
49,
57].
The Cd concentrations in mullets from Gela were much higher than those reported in other works [
53,
56], indicating a possible effect of anthropogenic input on fishes. Thus, cadmium released from human activities such as fuel combustion, application of phosphate fertilizers or sewage sludge could accumulate in fishes. Because it is a non-degradable and cumulative pollutant, cadmium generally accumulates mainly in fish entrails (intestines, liver and kidneys), and from here it spreads later to fish muscles [
58]. Concerning the toxicological value, the average Cd level (0.43 mg kg
−1) in the samples from Gela was about 8.6 times higher than the legal limits established by the Commission Regulation N.488/2014 [
49], amending the Regulation N. 1881/2006 [
44] and fixing the limits for muscle meat of fish at 0.05 mg kg
−1.
The Pb levels (average value 0.36 mg kg
−1) were twice as high in the Gela samples compared to the data in the literature that refers to uncontaminated areas; however, this value does not differ much from the legal limit (0.30 mg kg
−1) [
44]. Among the potentially toxic elements that negatively impact the health status of marine environment, Hg deserve particular attention because its concentration (average value 0.07 mg kg
−1) remains lower than the samples from Capo Passero [
53] and below the legal limits fixed for muscle meat of fish (1.0 mg kg
−1) [
57].
Among the other micro-constituents for which no specific law limits exist, As deserves particular consideration as the concentration of this element in mullets was higher (average value 8.91 mg kg−1) than the other potentially toxic elements; moreover, the value was not comparable with the literature because this element has never been considered before. However, the current As levels suggest that the high concentrations of As, found also in marine sediments of the Gela area (12.09 mg kg−1), are probably of industrial origin and could influence the final As concentration in mullet.
Furthermore, Cr deserves more attention because, it is an essential element for the human body as it could be involved in the blood glucose metabolism [
59,
60]; however, when hexavalent and at high concentrations, chromium enters the bloodstream and can produce toxic, mutagenic and carcinogenic effects in biological systems [
61,
62]. Its presence in the environment is mainly linked to anthropogenic sources [
53]. The comparative analysis of Cr concentrations shows that the levels are fairly homogeneous. Particularly, the results of the Gela samples are in agreement with those obtained by other authors in the Sicilian areas of Catania and Capo Passero [
53].
As
Table 6 shows, to the best of our knowledge, the V, Ni and Cu contaminations that we determined for mullets are not comparable with the data in the literature regarding the Mediterranean Sea. However, their levels in mullet muscles indicate that the possibility of inorganic element enrichment could arise in an area where there have been anthropogenic activities for many years. Particularly, as suggested by Antoniadis et al. [
63], multi-element contamination cases are complex, and the enrichment of many elements is often highly related with the occurrence of other elements with similar chemical behavior. Therefore, multi-element contaminations cause more severe effects cumulatively than the enrichment of any given element alone.
Finally,
Table 7 shows the results of the study carried out on fillets samples of other fish species caught in the Gulf of Gela: 2 sea hen (
Chelidonichthys lucernus) and 1 ray (
Raja species). The samples of Gela were compared with fishes of other species coming from areas (Marina di Ragusa and Patti) not subject to industrial contamination: 2 streenbras (
Lithognathus mormyrus), 3 sea bass (
Dicentrarchus labrax) and 3 sea bream (
Sparus aurata). Considering the variability of potentially toxic elements accumulation among different fish species and in order to make the data comparable, the objective of the study was to examine the same species both in the survey area and in the control area, unfortunately this was not possible as for urchins (
Paracentrotus lividus) or mullets (
Mullus barbutus) for which we considered the data available in the literature. In spite of these difficulties, all of them are fishes that live and feed on or near the bottom of seas; moreover, the fish samples had similar lengths.
The survey on these last fish samples showed that Hg levels (range: 0.02–0.08 mg kg
−1) were always below the limits established by the regulation (1.0 mg kg
−1) [
44,
57] while Cd concentrations (range: 0.37–0.40 mg kg
−1) were consistently far higher than the legal limits (0.050 mg kg
−1) [
49]. In particular, the two species caught in Gela (sea hen and ray) showed Cd levels up to eight times higher (0.39 and 0.40 mg kg
−1, respectively), while the Pb contents (0.31 and 0.33 mg kg
−1, respectively) were slightly above the legal limits (0.30 mg kg
−1) [
44]. What is particularly surprising is that these values were not very different from those determined for the specimens fished in Patti and Ragusa, areas which are far from direct anthropogenic impacts and are believed to be unpolluted and pristine. This result could be explained by the fact that in the Gulf of Patti there is a nature reserve (code ITA030012) in which a colony of yellow-legged gulls reside. Thus, in this area, a bird mediated contamination of trace elements (especially for As, Cd, Pb, Hg, Ni, V, Cr, Cu, and Zn) exists, as studies on a Mediterranean coastal system affected by gull guano revealed [
64,
65]. Regarding the samples from Ragusa, we suppose that this area, although it does not present industrial settlements, is affected by sea currents that convey inorganic contaminants in that direction from the nearby Augusta Bay where an important commercial, industrial and tourist port and a fundamental naval base of the Italian Navy is situated.
The evaluation of the data (
Table 7) concerning the contents of potentially toxic element in fishes shows that in rays (
Raja species), a benthic macro-invertebrate fish that could be considered as a bio-indicator of marine pollution, As levels (17.98 mg kg
−1) are far higher than those found in other species. These results are related to the behavior of this species, to live in close connection with sediment on the sea bed and to the alimentary habit of drawing nourishment directly from marine sediments, which in this area are contaminated. However, the health impact of this result in the edible tissues of fish is difficult to interpret in terms of the ability to induce adverse effects. In fact, while the arsenic dissolved in water is mostly inorganic, in fishes, arsenic is present predominantly in the organic forms of arsenobetaine and arsenocholine, which are virtually non-toxic [
66].
As mentioned before, there isn’t currently a law that cites the safety limits for As, Ni, V, Cr and Cu. Particularly, the Ni and V contents do not present significant differences among the fishes of Gela and the control; Ni was always present in higher concentrations than V, and this distribution pattern reflects the trend observed in mullets (
Table 6). Therefore, we can assume that the inter-species metal capture capabilities may be a result of a similar metal accumulation. Among the other micro-constituents, for which no specific law limits currently exist, higher concentrations were observed for Cr and the data did not show significant differences among the samples from different collection areas (range between 0.26 and 0.29 mg kg
−1). Moreover, the levels were not very different from those determined for mullets. Conversely, the distribution pattern of Cu in the present study shows a different metal accumulation: the samples from Gela had the lowest content (range: n.d–0.07 mg kg
−1) while the highest were in sea bass (
Dicentrarchus labrax) and steenbras (
Lithognathus mormyrus) (0.22 and 0.21 mg kg
−1, respectively). The different abilities of different fishes to store traces of potential toxic elements could be responsible for the differences in the Cu concentrations. However, the different Cu content could also be due to the presence of munitions—brass or bronze objects on the bottom of the sea in Patti or Ragusa—so that this element may be bio-available as a contaminant.
The comparative analysis of metal pollution among the fish species in
Table 6 and
Table 7 evidence that fishes from Gela showed a similar trend in potentially toxic element contamination, characterized by high Cd levels, Hg contents lower than the law limits and Pb concentrations slightly higher than the legal limits.
3.4. Intake Evaluation
The data determined for sea urchins, mullets, sea hens and rays has been applied to assess the health risk for the consumers. Therefore, to predict the human intake we considered the recommended dietary allowance (RDA) for Cu and Cr [
67]; the maximum residue limit (MRL), the tolerable weekly intake (TWI) and the provisional tolerable weekly intake (PTWI) values, established by the Europe Food Safety Agency [
68,
69,
70,
71,
72], were considered for Hg, Cd, Pb, and As. Moreover, since the EFSA CONTAM Panel determined that PTWI is no longer appropriate, in order to make these values more health protective, we determined the benchmark dose of 1% extra risk (BMDL
01) [
68,
70] for As and Pb. To evaluate the chronic dietary exposure to Ni, we calculated the tolerable daily intake (TDI) based on the EFSA experts’ safe level because there are no legal maximum levels for nickel in food [
73]. For vanadium, EFSA failed to establish a tolerable upper intake level because of difficulties in identifying a pivotal study to be used as a point of departure. All these parameters are reported in
Table 8.
In order to evaluate the risk associated with the consumption of fish from the study area, the dietary intake was assessed for an adult (60 kg for an Italian) assuming a daily ingestion of 77.8 gr muscle meat of fish [
37] and a weekly ingestion of 30 gr sea urchins; the results are reported in
Table 8. As mentioned above, MRL levels for potentially toxic elements in sea urchins have not been established yet; then, we considered the maximum reside levels fixed for bivalve mollusks for Pb [
44], Hg [
44,
57] and Cd [
49].
The data analysis showed that the Hg levels were less than the MRL (0.50 mg kg−1) in all the species. On the contrary, with the exception of sea urchins (MRL value 1.0 mg kg−1), Cd concentrations in all the other samples exceeded the MRL value (0.10 mg kg−1). Relating to Pb, all the fish samples were just above the specific MRL (0.3 mg kg−1) while the sea urchins presented levels that were significantly greater than the limit in the European legislation (1.5 mg kg−1).
The comparative analysis of pollution among the species analyzed evidenced that all the samples do not exceed the PTWI value for Hg, although a different trend in metal contamination can be evidenced. In fact, Hg reaches the highest residual levels in mullets, rays and sea hen that live on the seabed and accumulate contaminants from sediments (particularly ray), or have carnivorous feeding habits.
Compared to the safety standards, the results of the present study show that: Cd content in sea urchins (0.45%) remain within safety margins of TWI while in all the fish samples this value was exceeded (143.41–157.21%); conversely, PTWI and the extra-safety margin of BMDL
01 are within the safety limits for Pb in mullets (13.22% and 31.47%, respectively), sea hen (12.09% and 28.79%, respectively) and rays (11.18% and 26.62%, respectively) while BMDL
01 in sea urchins is borderline (98.89%); Ni (1.79–15.93%) never exceeded the fixed TDI limit while the contents of Cu (n.d–0.93%) and Cr (22.86–60.49%) remain within the RDA margins. A separate comment should be made for the arsenic content. As the data show, As surpassed both PTWI and BMDL
01 values in all the analyzed samples, reaching significantly high values for rays (PTWI = 1087.75% and BMDL
01 range = 291.36–7769.63%). However, according to the EFSA opinion [
68], data from seafood and fish, which are the major sources of arsenic in animal feed materials, did not indicate arsenic levels of concern because the toxicological risks mainly refer to inorganic arsenic, while the predominant forms in fish are the less toxic organic forms. EFSA concluded that food derived from fishes contributes only insignificantly to human exposure as the carry-over of arsenic in its inorganic form into edible tissue is low. Because the current data produced many results for total arsenic, but relatively few for inorganic arsenic, the CONTAM Panel considered the possibility of using a conversion factor which might provide an estimate of inorganic arsenic content from the total arsenic data. A problem with this sort of approach is that the relative proportion of inorganic arsenic and the ratio may vary depending on the seafood type. Therefore, because of the observed variability in the reported inorganic arsenic levels, it was not realistic to apply specific conversion factors to the total arsenic data. In our opinion, even if arsenic in fish products is mostly present as an organic compound, the contribution to the diet provided by these foods could be quite high, also because other food sources may contribute to the increase in quantities ingested up to the maximum level allowed by European regulations. The risk would be greater in the case of high consumption of fish, not uncommon in Italy, where some people, such as fishmongers, fishermen or large consumers of fishery products ingest quantities of fish, crustaceans and mollusks in much larger quantities than the average person.
3.5. Human Health Risks
The potential human health risks for fish consumption by the Gela population are summarized in
Table 9 where the results of the target hazard quotient (THQ) and cumulative target hazard quotient (CTHQ) of potentially toxic elements are summarized with the corresponding oral reference dose (RfD
o) established by the United States Environmental Protection Agency (USEPA).
Unfortunately, USEPA has no consensus RfD
o for inorganic lead and compounds, so it was not possible to calculate lead-THQ as for other chemicals. USEPA considers lead to be a special case because of the difficulty in identifying the classic “threshold” needed to develop an RfD
o [
38]. For mercury we used the mercury chloride parameter, as indicated by USEPA (personally contacted), although fishes efficiently absorb methyl mercury, and expel it very slowly [
74]. The oral RfD
o toxicity value for vanadium used, is derived from the oral RfD
o for vanadium pentoxide by factoring out the molecular weight of the oxide ion. The oral RfD
o toxicity value for chromium was evaluated considering that hexavalent chromium (Cr VI) is toxic and also that chromium trivalent (Cr III) exists. Therefore, the THQ was calculated for both Cr(VI) and Cr(III); moreover, according to the USEPA indications, it was assumed that the Cr(VI) to Cr(III) ratio is 1:6. The chromium THQ was also valuated for this value, knowing that this assumption may overestimate or underestimate the risk calculated. In order to determine the appropriate RfD
o for THQ, it was assumed that all arsenic ions were inorganic.
The estimated THQ for the detected inorganic elements did not pose unacceptable risks except for arsenic; furthermore, the contribution of inorganic lead could not be considered in the calculation, as indicated by USEPA after direct contact. As
Table 9 shows, THQ values for Cd, Cr(III)
tot, Cr(VI)
tot, Cu, Hg, Ni and V were less than 1 for all fish specimens, indicating that there is no health risk from these elements over a life time of exposure. Among these elements the maximum target hazard quotient is for Cd in mullets (0.56146), ray (0.51867) and sea hen (0.51218), followed by Hg. For this element, the THQ values were almost comparable among the fish species (0.30688 for mullets and sea hen, 0.30256 for rays) and definitely higher than the value determined for sea urchins (0.00667). It is important to note that THQ values for As in all the specimens tested were above 1, indicating significant adverse health risk for non-carcinogenic effects. In particular, the resulting THQ value was very high for rays (77.69627), followed by mullets (38.50668) and sea hen (22.12546). These results indicate that through the ingestion of this element from all these fish species, people could experience significant health risk. As THQ deals with individual metals only, and since generally, food items contain more than one inorganic element, as detected in our samples, the cumulative risk effect was calculated as CTHQ. Like THQ, it should also not exceed 1; however, as to be expected from the THQ values, the calculated CTHQ showed unacceptable risk for habitual fish consumers. It is important to evidence that As contributed to more than 95.5 % of the non-cancer effect to the CTHQ. So, in order not to overestimate the CTHQ, we recalculated this parameter assuming that the content of inorganic arsenic was 1%. However, also in this case, as shown in
Table 9, the threshold value is abundantly exceeded for as many as three fish species (ray, mullet and sea hen). Nevertheless, in relation to this hypothetical risk, as already stated above and according to EFSA’s opinion [
68], the content of arsenic in fish is considered unimportant because the toxicological risks mainly refer to inorganic arsenic, while the predominant forms in fish are organic ones, arsenobetaine and arsenocholine, virtually considered non-toxic. Nevertheless, in our opinion, even if arsenic in fish products is mostly present as an organic compound, the contribution to dietary intake provided by these foods could be quite high, considering that other food sources can contribute to the increase of the quantities ingested up to the maximum level permitted by FAO/WHO [
75].