Introduction

Metals pollution is recognized all over as a public health hazard since these contaminants are ubiquitous in marine ecosystems (Lozano et al. 2010) and cannot be degraded or destroyed due to their persistence. Chemical contamination is a wide-ranging issue mostly in wetlands and coastal lagoons (Almeida and Soares 2012), sensitive environments with unique ecological characteristics determined by the interaction between land and sea (Castro et al. 1999; Cataudella et al. 2015). Coastal lagoons trap inorganic contaminants from anthropogenic activities in the flow of pollutants from land to sea (Usero et al. 2005); trace elements ultimately accumulate in the estuarine environments and potentially affect the health of these aquatic ecosystems (Kljaković-Gašpić et al. 2010). Shellfish could constitute a persistent risk for human consumption since they ingest metals directly from water or from particulate (van der Oost et al. 2003). Bivalve mollusks submerge themselves in sediments, lie on the seafloor or adhere to rocks, fewer can swim short distances. All these features make them excellent bioindicators of water quality on a local/regional scale (Laffon et al. 2006).

Elements such as copper (Cu), iron (Fe), manganese (Mn), selenium (Se), and zinc (Zn) have specific biological and physiological cellular roles at low concentrations and are considered essential for living beings (Oehlenschläger 2012; FAO 2014). Nevertheless, at high concentrations, these trace elements can exert hazard to aquatic organisms and represent a food safety concern (EFSA 2014a, b, c, 2015a, b, c). The nonessential elements Cd, Hg, and Pb are instead toxic even at low levels of exposure and have not recognized metabolic functions (Tchounwou et al. 2012). Synergy between essential and nonessential elements has been reported, moreover a high total metal content in mollusks could exert a potential hazard for human health (Has-Schön et al. 2006; Castro-Gonzalez and Mendez-Armenta 2008). Bioindicators were extensively used to monitor the presence of chemical in aquatic environments (Rainbow et al. 2004). Bivalves are ideal bioindicators among invertebrates due to their habits and ability to bioaccumulate and concentrate contaminants (Farrington et al. 2016). They have been used in biomonitoring of coastal aquatic ecosystems at different spatial-temporal scales (Chase et al. 2001; Emmanouil et al. 2008; Waykar and Deshmukh 2012; Zuykov et al. 2013; Sericano et al. 2014; González-Fernández et al. 2015; Farrington et al. 2016; Capolupo et al. 2017). The use of bivalves to estimate environmental quality is widespread, and the concentration of trace elements in these aquatic organisms has been investigated worldwide since shellfish are an important source of protein in the diet of coastal population (Pan and Wang 2012; Chen et al. 2014; Chandurvelan et al. 2015; Araújo et al. 2016; Conte et al. 2016; Baltas et al. 2017; Bilgin and Uluturhan-Suzer 2017; Cunha et al. 2017; Wang and Lu 2017; Bonnefille et al. 2018). Among bivalves, mussels are powerful filter feeders: each individual filter around 15 m3 of water every year. This means that they can accumulate toxic substances dissolved in sea water or that is associated with material in suspension by factors of 102–105 (Baltas et al. 2017). Investigation performed in the central-western Mediterranean Sea highlighted the use of mussels as the most suitable organisms to check the state of water contamination (Kljaković-Gašpić et al. 2007; Cardellicchio et al. 2008; Andral et al. 2011; Spada et al. 2012; Stanković and Jović 2012). In particular, the Mediterranean mussel Mytilus galloprovincialis, owing to its relative high frequency of consumption, provides the more complete information for the evaluation of human exposure to trace elements (Cirillo et al. 2010; Stanković et al. 2011).

The Calich Lagoon, a typical brackish area of Sardinia (Western Mediterranean Sea, Italy), was the subject of several studies covering different topics, from shellfish farming (Chessa et al. 2005; Pais et al. 2006, 2007; Cannas et al. 2011;) to fish fauna (Chessa et al. 2007) to fish fauna (Chessa et al. 2007), from ecotoxicology (Ielmini et al. 2014) to the ecology of different planktonic components (Pulina et al. 2017, 2018; Bazzoni et al. 2018; Satta et al. 2020), bacteria and biotoxins in shellfish (Sedda et al. 2016; Bazzoni et al. 2019).

The aims of the present study were:

  • To evaluate the bioaccumulation of 21 trace elements in samples of M. galloprovincialis collected from the natural beds of the Calich Lagoon.

  • To assess the seasonal accumulation of the analyzed trace elements and to evaluate the role of M. galloprovincialis as bioindicators of environmental quality.

  • To verify the compliance with the legal limits established by the in-force regulations for Cd, Hg, and Pb in shellfish.

  • To investigate the presence of inorganic arsenic in this bivalve.

Materials and methods

Study site

The Calich Lagoon is located along the north western coast (40° 35′ 47.5″ N; 8° 17′ 59.9″ E) of Sardinia (Italy) (Fig. 1). It is situated between the city of Alghero and the village of Fertilia and belongs to the Regional Natural Park of Porto Conte. The lagoon extends for 92 ha with a depth varying between about 0.5 and 1.5 m. A natural channel (60-m wide and 2-m deep) connects the lagoon to the sea; at the output of which are located a medium size shipyard and a tourist harbor, very popular during the dry period. The lagoon receives freshwater from two natural fluvial tributaries (Rio Calvia and Rio Barca) and one artificial canal (Oruni channel) which drains 70% of the catchment (Pulina et al. 2017). Water from a wastewater treatment plant enters in the Oruni channel, while waters from municipal sewage treatment plant enter in the Rio Barca. The Rio Calvia flows downstream of the drinking water treatment plant of the city of Alghero receiving water from areas of agricultural use (Fenza et al. 2014). Therefore, the lagoon is highly eutrophic (Bazzoni et al. 2018). Natural beds of bivalve mollusks of commercial interest (e.g., grooved carpet shell (Ruditapes decussatus), olive green cockle (Cerastoderma glaucum), and Mediterranean mussel (Mytilus galloprovincialis)) are present, but despite the abundance of highly valuable commercial species of bivalve mollusks, the Calich Lagoon has not yet been classified for shellfish production.

Fig. 1
figure 1

The study area (black circle “ ”) and the sampling location (red asterisk “*”)

Full size image

Sampling

To evaluate the effects of intensive inputs of human activities, and those deriving from the agricultural and livestock sectors, the study was conducted in three different seasons of 2017: spring, summer, and autumn (i.e., March, June, and September, respectively). Wild specimens of Mediterranean mussel (Mytilus galloprovincialis) were manually sampled at roughly 1-m depth from a single mollusks sampling point (MS) in a natural bed (Fig. 1). A total of 30 adult mussels per month were sampled (i.e., 90 in total) with comparable weight were selected (mean ± SD of total length 52.9 ± 9.5 mm and wet weight 5.0 ± 2.7 g). After sampling, mussels were washed with clean lagoon waters and immediately transported to the laboratory in insulated coolers. The samples were cleaned and measured, and the soft part of the shellfish was removed from the shell, washed with distilled water, and sent frozen to the chemical laboratory to be analyzed.

Detection of trace elements

A Direct Mercury Analyser (DMA80, Milestone, Shelton, CT, USA), based on thermal decomposition, amalgamation, and atomic absorption, was utilized for mercury (Hg) detection following a previous detailed protocol (Esposito et al. 2018). The limit of quantitation of the method (LOQ) was 0.010 mg kg−1.

The other 20 elements silver (Ag), aluminium (Al), arsenic (As), beryllium (Be), bismuth (Bi), cadmium (Cd), cobalt (Co), chrome (Cr), copper (Cu), iron (Fe), gallium (Ga), indium (In), manganese (Mn), nickel (Ni), lead (Pb), selenium (Se), tin (Sn), thallium (Tl), uranium (U), and zinc (Zn) were quantified by inductively coupled plasma mass spectrometry (ICP-MS Xseries II, Thermo Scientific, Bremen, Germany) according a method already described (Squadrone et al. 2016). Samples (1.0 g) were first homogenized then subjected to microwave acid (HNO3 and H2O2) mineralization process (oven ETHOS 1 from Milestone, Shelton, CT, USA). Recovery was checked by using the CRM–SRM 1566b from NIST and was in the range 82–104%. The limit of quantitation of the method (LOQ) was 0.010 mg kg−1.

Arsenic speciation analysis

Arsenic speciation was performed by high performance liquid chromatography coupled to an inductively coupled plasma mass spectrometer (HPLC-ICP-MS); ICP-MS Xseries II, Thermo Scientific, Bremen, Germany and HPLC Spectra System MCS 1000, Thermo Scientific, Bremen, Germany). Samples (about 0.30 g) were added with a solution at 1% v/v HNO3 and 1% v/v H2O2 briefly vortexed and stored overnight at room temperature. Extraction was performed in an ultrasonic bath at 55 °C for 2 h. After centrifugation, samples supernatants were filtered through a 0.20-μm mesh filter and diluted 1:1 with ultrapure water then analyzed to detect the following arsenic species: DMA (dimethylarsinic acid), MMA (monomethylarsonic acid), AB (arsenobetaine), and iAs (inorganic arsenic, sum of arsenite, and arsenate). The limit of quantitation of the method (LOQ) was 0.020 mg kg−1.

Results were reported on wet weight basis according to the European Regulation (1881/2006 UE and amendments).

Statistical analysis

R software (version 1.0.153 - RStudio, Inc.) was employed. The concentrations of trace elements in Mytilus galloprovincialis collected in the Calich Lagoon in relation to season were compared and analyzed with a linear regression model. The model was defined as:

$$ y=\mathrm{season}+e $$

where “y” is the trace elements concentration expressed as mg kgˉ1 of wet weight (w. w.); “season” is the effect of the month and “e” is the error term. Moreover, a multiple pairwise-comparison between the means of groups was compared with the Tukey HSD (Tukey honest significant differences). The results were considered statistically significant when P < 0.05 for all the tests performed.

Results and discussion

Metal concentrations are shown in Table 1 and were also graphically represented in Fig. 2 and Fig. 3. The highest values were found for Al (mean 32 mg kg−1 w. w.), Fe (mean 32 mg kg−1 w. w.), and Zn (mean 25 mg kg−1 w. w.). Be, Bi, Ga, In, Tl, and U are elements scarcely or not investigated in mussels, but we found values < LOQ in all the shellfish samples, according to another investigation reporting very low concentrations for some of these elements in Mytilus spp. (Rodríguez-Hernández et al. 2019). As speciation results was reported in Table 2, while the differences among the sampling periods (ANOVA and the Tukey test) in Table 3. A high significant difference (P < 0.001) between the three sampling seasons was found only for Ag and Hg (Table 3).

Table 1 Trace elements in mussels from Calich Lagoon (mg Kg−1 w.w)
Full size table
Fig. 2
figure 2

Bar plot diagrams of nonessential trace elements for the different sampling periods in the Calich Lagoon (mean ± SD; mg·kgˉ1 wet weight)

Full size image
Fig. 3
figure 3

Bar plot diagrams of essential trace elements for the different sampling periods in the Calich Lagoon (mean ± SD; mg·kgˉ1 w. w.)

Full size image
Table 2 Arsenic speciation in Mediterranean mussel Mytilus galloprovincialis from the Calich Lagoon (mg·kgˉ1 w. w)
Full size table
Table 3 Differences among the sampling periods from ANOVA and the Tukey test
Full size table

In Table 4 a comparison of results with literature data was shown.

Table 4 Trace elements concentrations in Mediterranean mussels (Mytilus galloprovincialis) samples from different countries (mean or range, mg kgˉ1)
Full size table

Nonessential trace elements

Maximum limits (ML) for Cd, Hg, and Pb in shellfish are set by European Regulations 1881/2006 UE and amendments (1.0, 0.5, 1.5 mg kg−1 w.w. for Cd, Hg, and Pb, respectively). The values for Cd, Hg, and Pb were found lower than the ML (Table 1; Fig. 1). Cd concentrations (Table 4) were lower than those registered in the genus Mytilus from Italy (Cubadda et al. 2006; Lafabrie et al. 2007; Desideri et al. 2010; Bille et al. 2015); Montenegro (Stanković et al. 2011; Perošević et al. 2018); Croatia (Bogdanović et al. 2014; Kanduč et al. 2018); France (Gueguen et al. 2011; Richir and Gobert 2014); Spain (Olmedo et al. 2013); Portugal (Anacleto et al. 2015); Ireland (Bennion et al. 2019); Norway (Schøyen et al. 2017); Morocco (Azizi et al. 2018; Mejdoub et al. 2018); South Africa (Firth et al. 2019); China (He and Wang 2013; Li and Gao 2014; Lu and Wang 2018); India (Sivaperumal et al. 2007); Trinidad and Tobago (Balgobin and Singh 2018); and New Zealand (Whyte et al. 2009), but similar to those recorded in the Italian Tyrrhenian coast (Papetti and Rossi 2009); in the Sardinian coast (Piras et al. 2013); in the Gulf of La Spezia, Italy (Squadrone et al. 2016).

No significant seasonal variation of Cd bioaccumulation was registered (Table 3).

Hg concentrations (0.011–0.021 mg kg−1 w. w.) were in the range or even lower of the levels reported in Italy (Cubadda et al. 2006; Lafabrie et al. 2007; Desideri et al. 2010; Brambilla et al. 2013; Piras et al. 2013; Bille et al. 2015; Squadrone et al. 2016); Montenegro (Stanković et al. 2011; Perošević et al. 2018); France (Gueguen et al. 2011); South Africa (Firth et al. 2019); China (Liu et al. 2013; Li and Gao 2014); and New Zealand (Whyte et al. 2009).

A study performed on M. galloprovincialis from Italy reported the presence of Hg in the toxic organic form of methyl mercury (MeHg) ranging from 17 to 49% to total mercury (Di Leo et al. 2010). Since the maximum concentration found in this investigation was 0.021 mg kg−1, MeHg should range from 0.0036 to 0.010 mg kg−1 and should not constitute any risk for consumers.

In this study, a high significant temporal variation in bioaccumulation of Hg was reported (P < 0.001) (Table 3). In fact, in summer Hg was found about twice the values of spring and autumn (0.021 mg kg−1 w. w.).

The concentrations of Pb recovered in the Calich Lagoon (range 0.071–0.20 mg kg−1w. w.) were in line with previous Italian findings (Cubadda et al. 2006). No significant seasonal variation was recovered for this element (Table 3). Higher levels of Pb were instead described in samples of M. galloprovincialis from the Moroccan coasts (Azizi et al. 2018) and in M. edulis samples collected in Ireland (Bennion et al. 2019).

Aluminium showed the highest values between the nonessential element (mean 32 mg kg−1w. w.) and different concentrations in the three seasons, with the highest found in spring (Table 1). The registered values were higher than those reported in South Africa (Firth et al. 2019) but lower than those described in previous studies from Italy (Squadrone et al. 2016) and Montenegro (Perošević et al. 2018). Interestingly, Richir and coauthors (Richir and Gobert 2014) reported similar values from the nearby Corsica island (Table 4). It is well known that Al has a low bioavailability, but as result of intensive industrialization acid rains may lead to increased Al concentration in natural waters. Dissolved aluminum ions were then absorbed by marine or freshwater biota entering the food chain. It is well known that the major source of human exposure to Al is via diet (EFSA 2008) and after ingestion this element reaches all tissues and accumulates, especially in bone. Al carrier is the Fe-binding protein transferrin; Al can enter the brain and reach the placenta and fetus (EFSA 2008).

The TWI recommended by the European Food Safety Authority (EFSA) for Al is 1 mg Al kg−1 b. w./w. (EFSA 2008), that would be exceeded if consuming 310 g of mussels per day.

Then, the levels of aluminium in mussels should be routinely monitored since its presence could be related to the downstream flow, in one of the natural fluvial tributaries of the Calich Lagoon (Rio Calvia), of the drinking water treatment plant of the city of Alghero, where Al-based flocculants are used and could be influence the seasonal variation.

The inorganic forms of As are highly toxic but bivalve are known to contain mostly total arsenic (tAs), in the organic forms arsenobetaine; dimethylarsinic acid; monomethylarsonic acid, and a low percentage of iAs (EFSA 2009, 2014b). In the present study, tAs was found at values ranging from 1.2 mg kg−1 w. w. in spring to 1.7 mg kg−1 w. w. in summer (Table 1). Similar concentrations were observed by He and Wang (2013) and Firth et al. (2019) in M. galloprovincialis samples and by Olmedo et al. (2013) and Li and Gao (2014) in M. edulis samples. Low values of tAs have been also reported in clams from the Calich Lagoon (Esposito et al. 2018). High levels from moderate contaminated environments have been described by Richir and Gobert (2014) in the Corsica Island and by Kanduč et al. (2018) in Croatia. As reported in Table 2, the level of iAs in M. galloprovincialis was low and corresponded to 0.012 mg kg−1 w. w. (iAs 0.80% of tAs).

Silver and tin were found at trace level in the Calich Lagoon (Table 1), however significant seasonal variation has been highlighted in both elements (Table 3) since they were found higher in spring than in the other two seasons. The levels of Ag and Sn were in the range of concentrations found in the nearby Corsica Island (Richir and Gobert 2014) and in the Gulf of La Spezia (Squadrone et al. 2016), and do not pose any concern.

Essential trace elements

Essential trace elements levels are shown in Fig. 3 and Table 1. Cobalt was recorded in mussels from 0.042 mg kg−1 w. w. in spring to 0.067 mg kg−1 w. w. in summer (Table 1), levels comparable to those recorded in previous studies (Table 4) in Italy (Squadrone et al. 2016); France (Richir et al. 2014); Croatia (Kanduč et al. 2018); and Montenegro (Perošević et al. 2018). The high levels of Cr were found in spring (0.11 mg kg−1 w. w.) according to studies on the same species from Italy (Lafabrie et al. 2007; Squadrone et al. 2016); France (Richir and Gobert 2014); Montenegro (Perošević et al. 2018); South Africa (Firth et al. 2019); and China (He and Wang 2013). High levels of Cr have been instead registered in Mytilus galloprovincialis from Morocco (Azizi et al. 2018). The Cu values were quite different in the three seasons (Table 1) with the highest value in spring (mean 3.6 mg kg−1 w. w.) and the lowest value in summer (mean 1.4 mg kg−1 w. w.). These concentrations are comparable to those found in M. galloprovincialis from Croatia (Kanduč et al. 2018) and China (He and Wang 2013). Fe concentrations (Table 1) were lower than those reported in the same species from Italy (Squadrone et al. 2016); France (Richir and Gobert 2014); Montenegro (Perošević et al. 2018); and Morocco (Azizi et al. 2018) while Mn levels were in the range of published studies carried out in the Mediterranean Sea (Bogdanović et al. 2014; Richir and Gobert 2014; Squadrone et al. 2016; Kanduč et al. 2018; Perošević et al. 2018). Ni and Se values concentrations (Table 1) were homogeneous in the examined seasons, but Ni values were lower to those found from Mediterranean (Richir and Gobert 2014; Squadrone et al. 2016; Azizi et al. 2018; Kanduč et al. 2018; Mejdoub et al. 2018; Perošević et al. 2018).

Selenium values were comparable to concentrations registered in M. galloprovincialis from the Galician Coast (Olmedo et al. 2013), the Corsica Island (Richir and Gobert 2014), and the gulf of La Spezia, Italy (Squadrone et al. 2016), but lower to those found in clams from the Calich Lagoon (Esposito et al. 2018). Zn concentrations were found in the range 21–29 mg kg−1 w. w. (Table 1), according to other findings in M. galloprovincialis from different locations such as Italy, Montenegro, and South Africa (Squadrone et al. 2016; Perošević et al. 2018; Firth et al. 2019). Remarkably high levels of zinc in M. galloprovincialis (range 162–361 mg kg−1 w. w.) have been recently reported in two Moroccan studies (Azizi et al. 2018; Mejdoub et al. 2018).

It is well known that the levels of trace elements in mussels register fluctuations depending on the period of the year in which they are collected (Claisse 1992). In the period in which each species reaches sexual maturity, the bivalve pulp increases the mass of the organism, causing a relative decrease in metals concentrations. The highest metals concentrations are then usually recorded in winter and spring and the lowest in summer and autumn (Piras et al. 2013). According to this phenomenon in this investigation, the highest levels of Ag, Al, Cd, Fe, and Sn were recovered in spring; in addition, seasonal variation of physicochemical parameters such as temperature, salinity, dissolved oxygen, nutrients, and food availability are known to influence metals body burden.

Conclusions

The study provides useful information on the distribution of trace elements in native Mediterranean mussels (Mytilus galloprovincialis) from the Calich Lagoon (North Western Mediterranean Sea, Italy) that could constitute a baseline for future monitoring plans. The concentrations of metals in mussels did not pose concern for human consumption, but even if Cd, Hg, and Pb levels did not exceed the maximum limits set by European regulations, the high significant temporal variation registered for mercury suggests to constantly monitor this element in the different seasons. Moreover, although the recommended TWI was not exceeded, the levels of aluminium in mussels should be routinely monitored, since the presence of anthropogenic activities that could increase Al level in this valuable but fragile lagoon ecosystem.