Effect of steaming on chemical composition of Mediterranean mussel (Mytilus galloprovincialis): Evaluation of potential risk associated with human consumption

Abstract Steaming process is the most popular method for cooking mussels worldwide. The effect of this cooking process on some toxic (Cd, Ni, Pb), essential (Cr, Cu, Fe, Mn, Zn) elements, minerals (Na, K, Ca, Mg), total lipids, and fatty acid profiles in the Mediterranean mussels (Mytilus galloprovincialis) harvested from the Black Sea was studied. Different approaches to assess the benefits and risks (n‐6/n‐3, PUFA/SFA, AI, TI, h/H, EDI, THQ, HI, TR, and HQEFA) were employed. In general, steaming process significantly modified some essential elements and minerals concentrations as well as the fatty acid profiles. Compared to the raw samples, this culinary practice resulted in an increased concentration of Na, Mg, Zn, and saturated fatty acids and a decrease of polyunsaturated fatty acids. Significant changes in the lipid quality indices (PUFA/SFA, AI, TI, and h/H) from the raw samples were observed. No effect on the DHA content was found. However, the significant increase in the absolute content of EPA + DHA indicates that steaming does not compromise the nutritional quality of mussels. Target hazard quotients (THQs) and hazard index (HI) from elemental intake were below 1, indicating that the steamed M. galloprovincialis pose no hazard for the consumers. The target risk (TR) values for Pb, Cr, and Ni were calculated, evaluated, and showed acceptable or negligible levels. In addition, the benefit–risk ratio indicated that the steamed M. galloprovincialis are safe for human consumption.

polyunsaturated fatty acids, and essential macro-and microelements. A few studies characterized this species as a beneficial food which could provide the well-balanced chemical composition and through their consumption could prevent various nonchronic diseases (Merdzhanova et al., 2019;Őzden et al., 2010;Peycheva et al., 2021a). On the other hand, as filter feeding organisms they can accumulate different pathogens and contaminants. The raw mussels can deteriorate quickly and they must undergo heat treatment. Moreover, when cooking this species, it is important to regulate the temperature and time of cooking to better protect its beneficial quality. One of the suitable heat treatment methods for delicate tissues is steaming. This mild and fat-free method can keep the food moisture and biologically active components. The cooking procedures can affect the chemical composition and cause a decrease in nutritional quality of mussel tissues. Some of the major changes which reduce the lipid quality are related to oxidation of longchain polyunsaturated fatty acid (PUFA) compared to saturated fatty acid (SFA) (Biandolino et al., 2021). According to Barbosa et al. (2018), the steaming process can significantly increase the content of microelements such as Cu, Pb, and Cd, and this effect is species related.
Consequently, the study of the nutritional quality changes of steamed mussel tissues is more appropriate than the raw ones. Currently the benefit-risk evaluation assumed from the toxic and essential elements contents and omega-3 PUFA levels are mostly based on the analysis of raw mussel tissues (Barbosa et al., 2018;Ghribi et al., 2021). The studies concerning this topic are mainly focused on various fish species and only few data (Barbosa et al., 2018;Biandolino et al., 2021) are available on mollusks, especially M. galloprovincialis as well as the benefit-risk ratio. However, to the best of our knowledge, no studies concerning the influence of steaming on toxic, essential, and mineral elements and fatty acid content of Black Sea seafood (including mollusk) have been carried out.
Hence, in this article, we had tried to study (1) the toxic, essential, and mineral elements (Cd, Cr, Cu, Fe, Mn, Ni, Pb, Zn, Ca, Mg, Na, and K) concentration and fatty acid composition of raw and steamed M. galloprovincialis farmed in Bulgarian part of the Black Sea; (2) the impact of steamed and raw M. galloprovincialis on human health assessed by the most commonly used risk indices (n-6/n-3, PUFA/SFA, AI, TI, h/H, EDI, THQ, HI, and TR) based on the average concentration of trace elements and n-3 LC-PUFAs contents that could be reached via consumption of mussels; (3) the benefit-risk ratio for human health based on trace elements and n-3 LC-PUFAs contents in mussels (HQEFA).

| Collection of the samples and preparation
The samples were obtained from the local mariculture farm within the northern part of Bulgarian shore of Black Sea during summer of 2021. Approximately 2-4 kg of bivalves with comparable shell length were collected, placed into bags, and brought to the laboratory in ice boxes. Two hundred specimens of each species were taken randomly for the determination of sample mean. The average length of the mussels was 4.49 ± 0.62 cm. The samples were washed with cold distilled water and were randomly divided into two groups: (1) raw and (2) steamed. Each group comprised 60 mussels with 20 individuals for each replicate (n = 3). The raw samples were washed with Milli-Q water, brushed, shucked and soft tissues were removed with a Teflon knife, and stored in polyethylene bags at −20°C until analysis. The rest of the mussels were placed in a steam cooker and were steam-cooked at 90°C for 10 min in core. After the steaming, cooked samples were placed on a filter paper to absorb the excess moisture and then the flesh was removed using a Teflon knife.
Moisture content was determined on raw and steamed mussels by drying the sample in an oven at 105°C for 3 hr (AOAC, 2000).

| Chemical analysis
Each tissue of the raw and steamed sample (around 1 g wet weight) was weighed, placed in Teflon digestion vessels. Acid wet digestions using 8 cm 3 HNO 3 (65% w/v) and 2 cm 3 H 2 O 2 (30%w/v) were performed using a microwave closed vessel digestion system MARS 6 (CEM Corporation, USA) subject to three-stage program, maximum temperature of 210°C, ramp 15 min, pressure 800 psi, and maximum power 1050 W. The digested mollusk samples were cooled to 30°C, diluted to 25 cm 3 with Milli-Q water, and stored in polyethylene bottles until analysis.
The concentrations of Cd, Cr, Cu, Fe, Ni, Pb, Mn, Zn, K, Ca, Mg, and Na in the samples were determined using ICP-OES Spectrometer (Optima 8000; Perkin Elmer) with plasma gas flow-10 L/min, auxiliary gas flow-0.7 L/min, nebulizer gas flow-0.2 L/min, peristatic pump flow rate-1.5 ml/min, processing peak-area/high, spray chamber-cyclonic glass, nebulizer-concentric glass, MEINHARD ® , injector-alumina 2.0 mm i.d., background correction-1 or 2 point, manually, and axial or radial plasma view. The accuracy of the procedure for the determination of trace metals in mollusk was tested using ERM-CE 278k (Mussel Tissues from European Commission, Joint Research Center, Belgium) certified reference material. The CRM was digested and analyzed in the same way as the analytical samples. The recovery values were between 86% and 103% for the individual elements.

| Estimated daily intake, target hazard quotient, and target risk calculations for toxic metals
For the calculation of EDI, the following formula was used: where EDI is the estimated daily intake (mg toxic/essential element/kg body weight/day), FIR-average daily consumption of mollusks (kg/person), M c -average concentration of toxic/essential metal (mg/kg), and EDI = M c x IR BW a B w -body weight (kg). Mussel consumption rate in Bulgaria was 0.8 g/ capita/day in 2013 (FAO, 2020) and 60 kg is the average body weight used for adults (Zhelyazkov et al., 2018). Table (USEPA, 2020) was used to calculate the THQ by the ratio of exposed toxic/essential element concentration to the reference dose concentration and gives information about the long-term noncarcinogenic exposure probabilities.
Values of THQ below 1 show no harmful effect for human health.
Hazard index (HI) is used to estimate the combined effect of contaminants and it was calculated using the formula (USEPA, 2020): The acceptable values of TR are 1 × 10 −6 according to USEPA (2020).

| Fatty acid analysis and nutritional quality indices
Total lipids (TL) were extracted from M. galloprovincialis using Bligh and Dyer procedure (Bligh & Dyer, 1959). Three replicates of 20 mussels each (raw and steamed) were homogenized with a laboratory blender (Isolab Laborgeräte GmbH Co., Eschau, Germany). Tissue homogenates (3 g) were extracted with chloroform/methanol (1:2 v/v), chloroform/ methanol (1:1 v/v), and chloroform, followed by constant mixing for 30 min after each extraction step. NaCl solution in H 2 O (0.9% w/v) was added to the pooled extracts for phase separation. After centrifugation (3500g, 15 min), the bottom organic layer was collected with a Pasteur pipette, filtered through Na 2 SO 4 , and the solvent evaporated to dryness by rotary evaporator at 40°C. The dry residues were weighed and the amounts of total lipids were determined gravimetrically.
The fatty acids (FAs) of M. galloprovincialis were determined as fatty acid methyl esters (FAMEs) after direct transmethylation with 2% sulfuric acid in methanol (Christie, 1993). FAMEs were analyzed by gas chromatography using a Thermo Fisher Scientific FOCUS chromatograph equipped with a TRACE TR-5MS capillary column (30 m × 0.25 mm × 0.25 µm) and a PolarisQ ion trap mass spectrometer. The oven temperature was programmed from an initial oven temperature of 40°C for 4 min, followed by a rate of 20°C/ min from 40°C to 150°C and raised from 150°C to 235°C at a rate of 5°C/min, and then from 235°C to 280°C a rate of 10°C/min for 5 min. The carrier gas used was helium with a flow rate of 1 ml/min. of the total amount of fatty acid. The results for EPA and DHA were calculated as mg/100g w.w. using the corresponding conversion factors (XFA) for mollusks, proposed by Weihrauch et al. (1977).
Three nutritional quality indices were applied to assess the nutritional potential of raw and steamed mussels: • atherogenicity index (AI) (Ulbricht and Southgate, 1991): • thrombogenicity index (TI) (Ulbricht and Southgate, 1991): • hypocholesterolemic to hypercholesterolemic ratio (h/H) (Santos-Silva and Bessa, 2002)): Gladyshev et al. (2009) proposed a formula for the benefit-risk ratio of the consumption of marine organisms based on the content of LC-PUFA and toxic/essential elements. It is estimated through calculation using the following equation:

| Hazard quotient for benefit-risk ratio
where R EFA is the recommended daily dose of essential fatty acids (EFA) for a person (mg/day), C element is the concentration of the essential/ toxic element (mg/kg), C is the content of EFA (EPA + DHA) in a given bivalve (mg/g), RfD is the reference dose (mg/kg/d), and B w is the average adult body weight (70 kg). A value of HQ EFA less than 1 means the health benefit from bivalve consumption and HQ EFA more than 1 means the risk (Gladyshev et al., 2009

| Statistical analysis
All analyses were performed in triplicate and the results were expressed as mean ±standard deviation (SD). The results for toxic, essential, and mineral elements were stated as mg/kg w.w. and for individual fatty acids as the percentage of the total fatty acids in the total lipids. T test was used to compare the results for heavy metals, minerals, and fatty acid composition. Differences at p ≤.05 were considered significant (Graph Pad Prism 6).

| The levels of toxic and essential elements and mineral content in raw and steamed samples
The mean concentrations (mg/kg w.w.) of toxic elements in the studied edible portion of M. galloprovincialis are presented in Table 1.
As it can be seen from  Çelik & Oehlenschläger, 2007;Jureša & Blanuša, 2003;Keskin et al., 2007;Őzden et al., 2010;Stanković et al., 2011). Table 2 shows the dietary intake of essential metals based on the estimated daily intake for both steamed and raw samples through consumption of 100 g of steamed/ raw mussels and the percent coverage of the recommended tolerable metal intakes.   Mn is an essential element for humans and the data obtained from the current study are within the range stated in the literature (Çullaj et al., 2006;Gustily & Zhang, 2002;Locatelli, 2003;Ramsak & Scancar, 2012;Scancar et al., 2007).
The FAO (1983) set a limit daily human intake for Zn 30 mg/ kg. The established maximum level for zinc in Bulgarian legislation above which bivalves' consumption is not permitted is 200 mg/kg (Anonymous, 2004). The values between 3.20 ± 0.017 mg/kg for boiled and 9.68 ± 0.22 mg/kg for raw Oncorhynchus mykiss obtained from Antalya, Turkey has been reported (Gokoglu et al., 2004) and between 40 ± 11 μg/g for raw samples of M. galloprovincialis and 104 ± 45 μg/g for pan-fried samples of the same species (Kalogeropoulos et al., 2012). According to the data, the analyzed samples are within the health legislation levels.

| Dietary intake of essential elements. Effect of cooking on mussels' trace elements content
According to the data presented, the consumption of 100 g of cooked mussels by a normal 60 kg person covers a slightly significant fraction of the RDA or TDI of Cu and Mn (Table 2). Nevertheless, the cooked mussels show as a good source of Fe, with the level of 0.47 g meal reaching almost 28.18% of the required DRI, Cr (23.40%), and Zn (19.62%).
The concentration of toxic and essential elements and minerals in steamed mussels was higher than those of raw samples. The data in the literature are contradictory regarding the correlation between trace element content in raw and cooked sea shellfish.
The type of cooking (baking, grilling, frying, microwave heating, steaming, sous vide) as well as the size of marine species affect the level of toxic and essential elements before and after the thermal processing. Gokoglu et al. (2004) found that Na, K, P, Mg, Zn, and Mn contents of boiled fish decreased and the authors stated that this cooking process is not appropriate for fish preparation.
When the samples were grilled, the levels of Mg, Zn, and Mn decreased, while that of potassium increases (Gokoglu et al., 2004).

| Effect of cooking on total lipid content and fatty acid composition and nutrition quality
Upon cooking, significant changes in the moisture content, the total lipid content, and the percentages of fatty acids are observed (Table 3). Total lipid content increased significantly by 25.9% on a wet weight basis. The raw and steamed mussels showed higher moisture and lower lipid content than those reported by other authors (Biandolino et al., , 2021Kalogeropoulos et al., 2012;Merdzhanova et al., 2019;Prato et al., 2019Prato et al., , 2020. The raw M. galloprovincialis presented a favorable fatty acid composition with the predominance of PUFA (66.6%), followed by SFA (25.3%) and MUFA (8.1%). These data are in agreement with those reported previously for the same species from the Black Sea (Merdzhanova et al., 2019;Panayotova et al., 2021;Peycheva et al., 2021aPeycheva et al., , 2021bStancheva et al., 2017;Stratev et al., 2017). However, other authors reported that SFA was the most abundant fatty acid group in raw M. galloprovincialis from the Mediterranean Sea (Biandolino et al., , 2021Prato et al., 2019Prato et al., , 2020. As observed for the raw mussels, the PUFA fraction was the highest, followed by SFA and MUFA. Thermal treatment can cause changes in the lipid quality of products by destruction of vitamins, pigments, and fatty acids (Sampels, 2015). Steaming TA B L E 2 Essential trace element content (mean ±standard deviation) and nutritional contribution of raw and steamed Mytilus galloprovincialis from Black Sea (Bulgaria) process seemed to cause a significant increase in SFA and MUFA contents, mainly due to the changes in C14:0, C16:0, C16:1, C18:1, and C24:1 levels. On the other hand, a significant decrease in the total PUFA content was induced during the cooking process. For n-6 PUFA, the steaming resulted in a significant increase in the C18:2n-6 levels, but had no overall effect on the total content of n-6 PUFA ( Table 3). The effects of the cooking process on n-3 PUFA mainly involved a significant loss of DPA and thus a significant decrease in total n-3 PUFA, while EPA appeared to increase. No significant differences were found for the DHA levels alone and the total amount of n-3 PUFA.
Lipid quality indices of raw and steamed M. galloprovincialis are given in Table 4.
Considering the changes in n-6 and n-3 PUFA, the n-6/n-3 ratio exhibited no significant changes during the cooking process.
The n-6/n-3 ratio is used as an important index for the nutritional quality of marine lipids. Maintaining a low n-6/n-3 ratio (<1) in the everyday diet plays a crucial role in the prevention of dietinduced diseases (Simopoulos, 2013). PUFA/SFA ratio is used to describe the quality of dietary lipids. In our study, PUFA/SFA ratio decreased significantly by 27%: from 2.63 to 1.91. According to the Department of Health (2004), the recommended minimum value of PUFA/SFA ratio is 0.45 (National Institutes of Health).
Although the steaming process induced a significant reduction in PUFA content, cooked mussels seem to preserve a well-balanced The sum of EPA+DHA is one of the most important nutritional quality index of seafood lipids. These two fatty acids play major roles in a number of metabolic processes in the human organism (Méndez et al., 2017). According to the published data, the EPA + DHA content of mussels lipids is usually higher than that of other shellfish species (Tan et al., 2020). In this study, no significant changes were observed in the DHA expressed as a percentage of the total fatty acids. However, in order to provide  (EFSA, 2015b;FAO/WHO, 2003). In order to meet the daily requirements, a healthy individual needs to consume only 46-92 g raw or 36-72 g steamed mussels. The EPA+DHA content of M. galloprovincialis in this study is much higher than many other mollusk species (Tan et al., 2020).

| Estimated target hazard quotient, hazard index and target cancer risk
Target hazard quotient (THQ ), hazard index (HI), and target cancer risk (TR) are calculated using the formulas described in Section 2.3 and presented in Table 5. THQ is a coefficient which assesses the risk associated with the intake of contaminated M. galloprovincialis. The values below 1 (THQ < 1) reveal a lower level of exposure, which is associated with a daily exposure at this level is not likely to cause any harmful effects for human health during a lifetime in population (Bogdanović et al., 2014). According to our study, there were no THQ or HI values exceeding the value of 1 through the consumption of steamed M. galloprovincialis from Black Sea (Bulgaria) and it may be concluding that all tested elements in the steamed samples indicate that no health risk is present according to regulation or literature data (Jović & Stanković, 2014;Kalogeropoulos et al., 2012;Nekhoroshkov et al., 2021;Peycheva et al., 2021aPeycheva et al., , 2021b. TR values are calculated for intake of those heavy metals, which are considered cancerogenic (Cr, Ni, and Pb) according to IARC (2012). Based on US EPA methods, cancer risk lower than 10 −6 is considered to be negligible, >10 −4 is considered unacceptable, and in the range from 10 −4 to 10 −6 is considered acceptable (US EPA, 1989;USEPA, 2010). The results of this study showed that the carcinogenic risk for Cr, Ni, and Pb were acceptable or lower than the negligible level.

| Benefit-risk ratio of the consumption of marine organisms based on the content of LC-PUFA and toxic/essential elements
Values of benefit-risk hazard quotients (HQ EFA ) based on the content of LC-PUFA and toxic/essential elements are given in Table 6. A value of HQ EFA below 1 indicates there is no risk for people to consume M. galloprovincialis (Gladyshev et al., 2009).

| CON CLUS ION
Thermal treatment method used for M. galloprovincialis culinary preparation affected all elements contents and fatty acid composition.
Compared to the raw samples, steaming resulted in an increased concentration of Na, Mg, Zn, and saturated fatty acids and a decrease of polyunsaturated fatty acids. No effect on the DHA content was found.
However, the significant increase in the absolute content of EPA+DHA indicates that steaming does not compromise the nutritional quality of mussels and people need to consume only 36-72 g of steamed mussels to meet the daily requirements. Both raw and steamed mussels exhibited beneficial antiatherogenic, antithrombogenic, and hypocholesterolemic indices. As all THQ, HI, and TR values were below 1, which lead to the conclusion that consumption of steamed M. galloprovincialis from the Bulgarian Black Sea did not offer any harm risk for the consumer's health concerning the analyzed toxic elements. Based on the benefit-risk ratio index (HQEFA), it can be concluded that steaming process did not significantly affect the nutritional quality of M. galloprovincialis. The benefits of n-3 LC-PUFAs intake provided by the regular consumption of steamed mussels outweigh the risk posed by the content of toxic and essential elements.

ACKNOWLEDGMENTS
This study is financially supported by the National Science Fund of Bulgaria: Project # КП-06-OПP03/11 from December 18, 2018 and by the Ministero della Salute Ricerca Corrente.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare that are relevant to the content of this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated during and/or analyzed in the current study are available from the corresponding author on reasonable request.