Accepted Manuscript Trace element occurrence in the Pacific oyster Crassostrea gigas from coastal marine ecosystems in Italy E.A.V. Burioli, S. Squadrone, C. Stella, C. Foglini, M.C. Abete, M. Prearo PII:
S0045-6535(17)31325-5
DOI:
10.1016/j.chemosphere.2017.08.102
Reference:
CHEM 19800
To appear in:
ECSN
Received Date: 11 May 2017 Revised Date:
17 July 2017
Accepted Date: 18 August 2017
Please cite this article as: Burioli, E.A.V., Squadrone, S., Stella, C., Foglini, C., Abete, M.C., Prearo, M., Trace element occurrence in the Pacific oyster Crassostrea gigas from coastal marine ecosystems in Italy, Chemosphere (2017), doi: 10.1016/j.chemosphere.2017.08.102. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 1
Trace element occurrence in the Pacific oyster Crassostrea gigas from coastal marine
2
ecosystems in Italy
3 E.A.V. Burioli*, S. Squadrone, C. Stella, C. Foglini, M.C. Abete, M. Prearo
RI PT
4
Istituto Zooprofilattico Sperimentale del Piemonte, Liguria e Valle d’Aosta, Turin, Italy.
6
*Corresponding author:
[email protected] 7
Abstract
8
The Pacific oyster is one of the world’s most widespread bivalves and a suitable species for
9
biomonitoring trace elements in marine environments thanks to its bioaccumulation ability. As it is
10
also an edible mollusc, concentrations of harmful elements in its tissues must be monitored. For
11
these purposes, 464 wild individuals were collected from 12 sites along the Italian coasts. The
12
concentration of fourteen trace elements (Al, As, Cd, Cr, Cu, Fe, Hg, Mn, Ni, Pb, Se, Sn, Tl, and
13
Zn) in their tissues was quantified. Among the three heavy metals, cadmium, lead, and mercury,
14
none exceeded the maximum limit for in food set by European Union regulations but Cd in one
15
sample from the Varano Lagoon resulted extremely close to this value. Contamination by Hg of the
16
northern Adriatic and Orbetello Lagoons was also observed. Moreover, there was a positive
17
association between the lagoon’s environmental conditions and the bioaccumulation of this element
18
in oysters. Despite the ban instituted 15 years ago on the use of Sn in antifouling paints, this
19
element is still present in several marine environments, as demonstrated in the oysters sampled from
20
harbour areas. Samples collected from harbours also showed very high concentrations of Cu and Zn
21
due to the ability of oysters to accumulate these elements, which have replaced Sn in antifouling
22
paints. Analysis of the samples from most sites indicated a low risk of human exposure to harmful
23
elements through oyster consumption; nonetheless, chemical sanitary controls should focus
24
primarily on Cd, Cu, and Zn.
AC C
EP
TE D
M AN U
SC
5
25 1
ACCEPTED MANUSCRIPT 26 Keywords Bioaccumulation; Bivalve Molluscs; Mediterranean Sea; Human Consumption;
28
Biomonitoring; Public Health
29
1. Introduction
30
Increased awareness of the far-reaching impact of pollution on the environment and human health
31
has led to the development and implementation of action programs for the protection of marine
32
environments in the last decades. An international framework for addressing threats to marine
33
ecosystems is provided by the Global Programme of Action for the Protection of the Marine
34
Environment from Land-based Activities (GPA), which was launched in 1995 and is coordinated in
35
partnership with the United Nations Environment Programme (UNEP).
36
Protecting vulnerable marine areas like coastal lagoons and estuaries is essential for preserving their
37
unique biodiversity and developing their potential economic value for aquaculture activities such as
38
bivalve farming. However, these environments are often contaminated with a variety of chemical
39
substances, including toxic metals, amongst which cadmium (Cd), lead (Pb) or mercury (Hg), and
40
metalloids like arsenic (As) pose a potential threat to ecological and public health. Naturally present
41
in marine environments, these elements originate from both geogenic processes and anthropogenic
42
activities such as agriculture, mining, and industry (Bradl, 2002; He et al., 2005). Through
43
contaminated river flows, direct waste discharges, antifouling paints on ship hulls, and atmospheric
44
deposition, these elements ultimately reach and pollute coastal marine areas.
45
Approximately 30 trace elements influence an organism’s biology in varied ways and can be
46
classified as essentials and non-essentials (Fraústo da Silva and Williams, 1993; Tamás and
47
Martinoia, 2005). Some trace elements, including cobalt (Co), copper (Cu), chromium (Cr), iron
48
(Fe), manganese (Mn), molybdenum (Mo), nickel (Ni), selenium (Se), and zinc (Zn), are essential
49
for animal and human biologic functioning (Mertz, 1981; WHO/FAO/IAEA, 1996, Nordberg and
50
Nordberg, 2016). The essentiality of arsenic at ultra-trace concentrations remains debated (Gropper
AC C
EP
TE D
M AN U
SC
RI PT
27
ACCEPTED MANUSCRIPT et al., 2008). The excessive intake of some essential elements like Cu (Tchounwou et al., 2008) or
52
Cr (Patlolla et al., 2009) can lead to cell damage, while exposure to other trace elements like Cd,
53
Pb, and Hg, which have no biological functions, can induce serious adverse effects on human and
54
animal health (Tchounwou et al., 2001; 2004a; 2004b; 2014; Sutton et al., 2002) even at low levels
55
of exposure. For these reasons, As, Cd, Hg, and Pb have been included in the World Health
56
Organisation (2010) list of the top ten chemicals of major public health concern.
57
Regular seafood consumption is widely considered beneficial for health (Daviglus et al., 2002;
58
EFSA, 2014); however, shellfish are a source of exposure to inorganic contaminants (particularly
59
As, Cd, Cu, Fe, Hg, Pb, Se, and Zn) that, above certain concentrations, may pose a risk for
60
consumer health (FAO, 2005). Recognizing this problem, European Regulation EC/1881/2006 and
61
amendments set maximum limits for Cd, Pb, and Hg in food for human and animal consumption.
62
However, these elements can also affect marine biota. At the concentrations commonly found in
63
water bodies, many have toxic effects on ecological receptors and at high concentrations, they can
64
reduce or eliminate sensitive species and impact an ecosystem biodiversity (Peterson, 1986). For
65
instance, the early embryo-larval stages of marine invertebrates have been shown to be highly
66
sensitive to micropollutants and to heavy metals in particular (Martin et al., 1981; Beiras and His,
67
1995).
68
Aquatic organisms, being exposed to trace elements from the environment, are often used for
69
pollution monitoring studies (Chiarelli and Roccheri, 2014). In fact, they are a useful tool for
70
assessing water quality and marine ecosystem health in vulnerable areas such as coastal marine
71
regions where anthropogenic activities are concentrated (Rainbow, 1995). Biomonitoring has
72
proven to be a valid method to assess environmental contamination by toxic elements and to
73
determine the concentration of certain toxic metals and other trace elements in marine organisms
74
where they continuously accumulate and concentrate in tissues (Shahidul-Islam and Tanaka, 2004)
75
in a time-dependant manner (Rainbow, 1995). Furthermore, the levels of nonessential and essential
AC C
EP
TE D
M AN U
SC
RI PT
51
3
ACCEPTED MANUSCRIPT elements in aquatic biota depend on many other factors, including geographic location, species,
77
size, feeding patterns, element form and concentration in a given environment, solubility,
78
temperature, salinity, oxygen, organic content in water and sediments, and pH (Luoma, 1983;
79
Stanković et al., 2012). Shellfish can accumulate inorganic elements at concentrations several
80
orders of magnitude higher than the aquatic medium, particularly in mantle, gills and palps (Rodney
81
et al., 2007), and from their filtration feeding habits. Furthermore, being benthic animals, they are
82
more exposed to solid and dissolved matter stored in contaminated sediments. While all marine
83
invertebrates are known to accumulate trace elements in their tissues, their bioconcentration ability
84
is species dependent (Mok et al., 2014).
85
Biomonitoring investigations in the Mediterranean Sea have primarily focused on Mytilus
86
galloprovincialis as a “sentinel” organism (Viarengo et al., 2007). In order to improve our
87
understanding of bioaccumulation mechanisms and trace element cycles in the environment, a wider
88
range of bioindicators needs to be identified (Conti and Cecchetti, 2003). Numerous studies
89
worldwide have successfully used oysters as bioindicator, particularly species of the genus
90
Crassostrea such as C. gigas (Beliaeff et al., 1998; Shulkin et al., 2003; Baudrimont et al., 2005),
91
C. rhizophorae and C. virginica (Rojas de Astudillo et al., 2005). The Pacific oyster, C. gigas,
92
provides an excellent subcosmopolitan “sentinel” species. It is available all year round, abundant in
93
numerous marine areas, sessile, can be easily collected, and is extremely resistant to environmental
94
stress (Rainbow and Phillips, 1993; Boening, 1999) from high metal exposure as compared to
95
mussels (Funes et al., 2006). Nevertheless, the Pacific oyster also raises public health concerns as it
96
is one of the most popular bivalve molluscs (http://www.fao.org/figis/servlet/) and diet is one of the
97
main routes of exposure to toxic elements for humans. Moreover, few studies to date have been
98
performed on C. gigas in the Mediterranean (Ochoa et al., 2013; Squadrone et al., 2016).
99
Though the natural beds of C. gigas are quite common along the Italian coast (Burioli et al., 2016),
100
commercial Pacific oyster farming is practiced only in three regions, mostly in Sardinia and, to a
AC C
EP
TE D
M AN U
SC
RI PT
76
ACCEPTED MANUSCRIPT lesser extent, Veneto and Liguria. Oyster farming has attracted increasing interest due to recent
102
awareness of the need for aquaculture diversification. This will require developing new areas, both
103
in coastal lagoons and open waters, for oyster farming in the near future. An important part of this
104
evaluation process involves the collection of predictive information on trace element
105
bioaccumulation that takes onto account both human health and zootechnical aspects.
106
With this study, we analysed the trace element occurrence of Al, As, Cd, Cr, Cu, Fe, Mn, Hg, Ni, Pb,
107
Se Sn, Tl, and Zn in C. gigas. The sampling areas were 12 sites in five geographical regions with
108
different environment features and contamination levels. Considering the growing global concern
109
about the effects of pollution on the marine environment, our work contributes to the knowledge of
110
trace element occurrence in different marine ecosystems. It provides an assessment of the risk
111
associated with oyster consumption and an evaluation of the suitability of the sampling sites as
112
potential areas for the further development of mollusc farming activities.
M AN U
SC
RI PT
101
TE D
113 2. Materials and Methods
115
2.1. Sampling
116
Between May and June 2014, 464 wild individuals of the Pacific oyster, C. gigas, were collected
117
along the Italian coasts. The sampling took place during the pre-spawning period. Twelve natural
118
oyster beds were sampled and 30 to 60 individuals (≥80 mm in length) were collected from each
119
site (Table 1) at a depth of about 0.5 m below the low-water line. A minimum length was
120
established in order to collect only those specimens that had resided in the sampling sites for at least
121
two years. The specimens were kept alive, immediately placed in a refrigerated box, and processed
122
within 24 h. Each individual was externally cleaned with deionized water, shucked, drained, and
123
pooled (pools consisted of 5 or 10 whole specimens, based on their singular weight, in order to have
124
at least 50 g in each pooled sample) (Table 1), homogenized, and frozen at -20°C until analysis.
125
The sampling sites were representative of three different types of marine environment: lagoon, open
AC C
EP
114
5
ACCEPTED MANUSCRIPT waters (off-shore and gulfs), and harbour. Eleven sites were located in the Adriatic and one in the
127
Tyrrhenian Sea (Figure 1). The sampling sites along the Adriatic coast are described from north to
128
south. Muggia and Monfalcone are located in the Gulf of Trieste, a semi-enclosed basin in the
129
northern Adriatic. The main freshwater inputs into the Gulf of Trieste are the Isonzo and Timavo
130
Rivers. Until 1995, the Isonzo was the primary source of Hg pollution originating from extensive
131
mining operations in the upper River drainage basin. The northern and northwest part of the
132
Adriatic is characterized by a complex system of transitional environments that includes, amongst
133
others, the Marano, Venezia, and Caleri Lagoons. The Marano Lagoon is characterized by shallow
134
waters, relatively limited water exchanges with the open sea and several freshwater inputs from the
135
Aussa and Corno Rivers, both of which flow through an industrial zone with factories that, for
136
many years, manufactured cellulose, chlor-alkali, and artificial textile fibres, the Stella River, and
137
various irrigation canals. The Caorle sampling site is located in open waters at about 5 km from the
138
coastline. Chemical plants, oil refineries, and shipyards surround the Venice Lagoon, which receives
139
freshwater input from the numerous rivers flowing through the highly populated industrial and
140
intensive agricultural regions of the Po Plain. Besides altering the salinity and trophic conditions of
141
the Lagoon, this input is also an important conduit for the delivery of contaminants to its ecosystem.
142
Cavallino-Treporti is located in the less anthropized eastern part of the Lagoon, opposite the city of
143
Chioggia. The petrochemical plants clustered within the Marghera industrial zone, where a chlor-
144
alkali plant was operated from 1951 to 1988, dominate the central part of the Venice Lagoon. In
145
Caleri, as in other parts of the northern lagoons, the water temperature is high and the oxygen
146
concentration low in summer, partly due to the shallow water depth but it is mostly influenced by
147
open sea environmental conditions because there is no direct entrance of river water at the sampling
148
site and good exchanges with the open sea. Porto Garibaldi is near the Po Delta, where sediments
149
contain high levels of contaminants from the numerous industrial plants located along the river
150
(Guerzoni et al., 1984). Salinity is affected by variations in riverine inputs, with a decreasing impact
AC C
EP
TE D
M AN U
SC
RI PT
126
ACCEPTED MANUSCRIPT towards south. Sampling was conducted at the mouth of the canal that connects the Comacchio
152
Lagoon with the open sea. In Giulianova the sampling was carried out in the central part of the
153
harbour, an enclosed area for both recreational and commercial fishing boats. The Cervia sampling
154
site is located in open water, less than 0.5 km from the coast. Finally, Varano Lake communicates
155
weakly with the open sea via two artificial canals, one of which is located in Capoiale, where
156
sampling was conducted. The area has little industry; the basin receives wastewaters mainly from
157
agricultural drainage water. The salinity is less than 30‰ and stable all year round because of the
158
presence of underwater springs (Spagnoli et al., 2002). The only sampling site on the Tyrrhenian
159
coast is the Orbetello Lagoon. This water body has two communicating basins known as West and
160
East, with a1 m average depth, and communicates with the open sea via two canals. The sources of
161
anthropogenic contamination are now mainly agriculture wastewaters, but until 1991, a chemical
162
factory manufacturing granular fertilizers released waste metals into the western part of the basin. A
163
large contribution to the metal contamination of the lagoon also comes from geogenic processes: the
164
presence of large cinnabar deposits in the catchment basin of the Albegna River, which flows into
165
the lagoon (Grassi and Netti, 2000). Sampling was conducted in the Western lagoon.
SC
M AN U
TE D
EP
166
RI PT
151
2.2. Detection of trace elements
168
Each sample was divided into two aliquots. The first was used for total Hg quantification with a
169
direct mercury analyser (DMA-80 Analyser, Milestone, Shelton, CT, USA). The second aliquot was
170
used for detecting the other elements (Al, As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Se, Sn, Tl, and Zn) by
171
inductively coupled plasma-mass spectrometry (ICP-MS Xseries II, Thermo Scientific, Bremen,
172
Germany) after wet digestion through high-quality grade (Suprapur® Merk, Darmstadt, Germany)
173
mineral acids and oxidants (HNO3 and H2O2) following previously described protocols (Squadrone
174
et al., 2016). The analytical methods were validated according to UNI CEI EN ISO/IEC 17025
175
(General Requirements for the Competence of Testing and Calibration Laboratories). The limit of
AC C
167
7
ACCEPTED MANUSCRIPT 176
quantitation was 0.010 mg Kg-1 for all the analysed elements and all concentrations are given on a
177
wet weight basis.
178 2.3. Data analysis
180
The range, mean, median and standard deviation of metal concentrations of each pool are reported
181
in Table 2. For each element, ANOVA and Bonferroni’s test were applied to evaluate whether there
182
were significant differences between the groups of samples from the different areas, and to assess
183
the differences between three different types of marine environment (open waters, lagoon and
184
harbour). The harbour of Porto Garibaldi was excluded from the final analysis because it is directly
185
connected to the Comacchio Lagoon and not an enclosed area. Results with a p-value Fe (80.00) > Cu (76.95) > Al (38.45) > Mn (9.50) > As (3.5) > Se (0.89)
209
> Cd (0.23) > Pb (0.195) > Ni (0.20) > Cr (0.195) > Sn (0.06) > Hg (0.02), and Tl (0.6; P < 0.05). Interestingly, Cd and Hg appeared to be negatively correlated (r >0.5; P < 0.05).
AC C
270
EP
265
271
3.4. Maximum number for safe consumption
272
The number of oysters required to exceed the safety limits are reported in Table 4. The Sn
273
concentrations were very low and this element was excluded from the analysis.
274 275
4. Discussion 11
ACCEPTED MANUSCRIPT 4.1. Trace metals in relation to environmental conditions
277
The Adriatic, especially its northern area, is strongly influenced by the freshwater input from large
278
rivers, acting on salinity and trophic condition, in particular. Then, the geographic distribution of
279
contaminants of inland origin is influenced by both the main anticlocking current and the
280
anemometric conditions (Figure 1). A large data set shows that a positive gradient of Hg
281
concentration in sediments from the southern to the northern Adriatic exists (Ferrara and Maserti,
282
1992), and the results of the present study in oyster soft tissues support the existence of this
283
phenomenon. However, we only partially observed the regional dynamics of dispersion Covelli et
284
al. (2001) reported for the Gulf of Trieste. Isonzo River was once considered as the primary source
285
of Hg pollution in the Gulf of Trieste (Horvat et al., 1999), with higher concentrations near the river
286
mouth and an evident transport of mercury in a northern, especially in a northwestern, direction
287
respect to a southern direction. However, we detected higher Hg levels in the samples collected
288
from Muggia than Monfalcone. This difference may have been due to an additional and unknown
289
pollution source in the southern part of the Gulf. Our observation of higher Hg concentrations in the
290
samples from Cavallino-Treporti and Marano are shared by the findings of mercury contamination
291
reported by Piani et al., 2005. The inputs of Hg in the Marano and Venice Lagoons, derived from
292
chlor-alkali discharges from industrial plants, resulted in comparable levels of contamination. Part
293
of this contamination may also be associated with the inputs from the Isonzo River through a
294
westward long-range transport of fine particles with organically bound Hg forms (Biester et al.,
295
2000). A similar situation was observed in the Orbetello Lagoon, where regional natural
296
contamination (Romano et al., 2015) is reflected by the Hg bioaccumulation in oysters. Indeed,
297
according to Lacerda and Gonçalves (2001) and Han et al. (2007), wetlands and coastal lagoons are
298
important areas for Hg methylation, which improves the bioavailability of this element. This
299
explains the significant higher Hg concentration observed in lagoons. The Hg levels in the samples
300
from the Varano Lagoon suggest that Hg contamination in this region is absent. Interestingly,
AC C
EP
TE D
M AN U
SC
RI PT
276
ACCEPTED MANUSCRIPT although the study conducted by Fabbri et al. (2001) found the Po River mouth and the Ravenna
302
Lagoon to be highly polluted areas, the Hg levels we detected in the samples from the Caleri
303
Lagoon, Porto Garibaldi and Cervia were all very low.
304
The cadmium uptake depends on its chemical speciation, cadmium ion Cd2+ being more
305
bioavailable, and it is mainly influenced by water salinity (De Lisle et al., 1988; De Wolf et al.,
306
2004) and by the presence of organic ligands in water (Blust et al., 1995). We observed that
307
individuals collected from lagoons contained, on average, three times less cadmium than those from
308
open waters, indicating a negative correlation between Cd and Hg. Several environmental factors
309
differentiate the Orbetello Lagoon from the northern Adriatic lagoons; nonetheless, the comparable
310
levels of Cd detected in these enclosed areas may be imputable to common characteristics induced
311
by high trophic conditions and the presence of shallow waters. Interestingly, the Cd content in
312
individuals from the Varano Lagoon was similar to or slightly higher than those collected from open
313
waters. A study carried out during the summer of 2000 (Spagnoli et al., 2002) reported moderate Cd
314
concentrations in the sediments of the lagoon. The reason for the high levels we detected is still
315
unclear. Several causative hypotheses may be advanced: an unknown source of contamination or
316
some peculiar environmental conditions with respect to other lagoons or effluents from agriculture.
317
As regards this last factor, a high correlation has been demonstrated between the concentrations of
318
Cd and plant nutrients such as phosphates and nitrates in waters (Bewers et al., 1987).
319
In marine environments, lead exhibits a high affinity towards organic substances and clay minerals,
320
resulting in a preponderance of the sediment-bound form. Low Pb concentrations are usually
321
present in shellfish because of the limited bioavailability of both dissolved and dietary lead to
322
marine animals (Amiard et al., 1985). The median Pb levels we detected are comparable to the
323
value of 0.131 mg.kg-1 in Pacific oysters reported in a previous study carried out in the Ligurian Sea
324
(Squadrone et al., 2016), but the concentrations in the samples from Giulianova, and even more so
325
in those from Muggia, were significantly higher than the median value. Differently from the high
AC C
EP
TE D
M AN U
SC
RI PT
301
13
ACCEPTED MANUSCRIPT concentration of Pb in sediments in the Varano Lagoon observed by Spagnoli et al. (2002) between
327
1999 and 2000, we did not detect high Pb levels in the oysters collected from this site. The positive
328
correlation between Pb, Cu, and Sn, and between Sn and Zn could be related to the harbour
329
environment. In fact, the high concentration of Cu in harbours is known to result from the use of
330
antifouling paints for vessel hulls, while the presence of Pb probably derives from fuel discharges.
331
Although regulations on the use of Sn in antifouling paints went into effect in the EU in 1991 and
332
its use was finally prohibited with European Regulation 782/2003, the Sn levels in the samples from
333
Giulianova harbour were much higher than in those collected from the other sites. This port is an
334
enclosed basin with scarce water exchange, which could explain the high concentrations we
335
observed. In contrast, the harbour of Porto Garibaldi is a canal connecting the Comacchio Lagoon to
336
the open sea and it did not show this port-related pollution. Although located on the southern side of
337
the Gulf of Trieste, the Muggia sampling site shares similar characteristics with the harbour of
338
Giulianova, and the samples from Muggia contained high concentrations of the three elements Cu,
339
Pb, and Sn.
340
Arsenic is widely distributed throughout the marine environment and is characterized by a complex
341
cycle. Because its accumulation in animals seems to be promoted by a benthonic lifestyle and a
342
close relationship with sediments (Wu et al., 2014), high As levels may be detected in bivalves in a
343
contaminated environment. The median As concentration (3.5 mg kg-1 w.w.) we found was twice as
344
high as the value Squadrone et al. (2016) reported in C. gigas from the Ligurian Sea, but it was very
345
close to the concentration observed in European flat oysters from the Croatian coasts of the Adriatic
346
(Bilandžić et al., 2016). The high As concentration found in the Orbetello Lagoon was consistent
347
with the results obtained by Romano et al. (2015). As pointed out for Hg, the main source of As
348
contamination is probably related to the basin’s geochemical characteristics, but a contribution from
349
the industrial and agricultural chemical wastes could not be excluded.
350
Cu, Fe, and Zn are three elements essential to life, and the high ability of Pacific oyster to
AC C
EP
TE D
M AN U
SC
RI PT
326
ACCEPTED MANUSCRIPT concentrate these metals (Funes et al., 2006, Pan and Wang 2009) was noted in the present study,
352
with high concentrations detected in samples from all sites. These bivalves also accumulated high
353
quantities of Al, a non-essential element. Cu can be considered a port-related contaminant, and
354
CuO2, the main biocidal compound of antifouling paints, is often associated with Zn to improve its
355
efficiency (Guardiola et al., 2012). The source of the high concentrations of Fe and Zn found in the
356
samples from Cavallino-Treporti could also be related to the pollution from the Marghera industrial
357
area, such as observed in sediments by Bellucci et al. (2002).
358
The other essential elements, Cr, Ni, and Se, were present in low concentrations. The Muggia site is
359
characterized by multifactorial pollution with Hg and port-related metals (As, Cr, Ni, Se, and Zn) at
360
the same time. This difference could be explained by the long-standing presence of industry, port
361
activities, and oil unloading around Trieste. Moreover, because the Bay of Muggia has a reduced
362
hydrodynamism, the high concentrations of Cu, Pb, and Zn in the samples from this site reflected
363
the storage of these elements in the bay sediments. However, the lack of previous data on metal
364
concentrations in oysters precluded detailed assessment of the trend of contamination in this area.
365
Unexpectedly, the Mn concentration in the samples from Muggia was considerably lower than the
366
levels detected in the samples from the other sites.
367
Generally, the values of the trace element levels measured during the present study are partially
368
shared by previous investigations, for instance, those conducted in the Ligurian Sea (Squadrone et
369
al., 2016) or in the native range of the species such as the Korea Strait (Mok et al., 2015). Except
370
for Al, Pb, Cr, and Hg, the trace element levels reported by Squadrone et al. (2016) were lower than
371
those we detected. This difference may have been due to the different sampling seasons and the
372
physiological status of oysters. In the present study, the sampling took place during the pre-
373
spawning period (spring), when the concentration of trace elements detectable in mollusc tissues is
374
supposed to be highest (Najdek and Sapunar, 1987; Claisse, 1992), if compared to late summer and
375
early autumn, when minimal concentrations are observed. Although bivalves and oysters in
AC C
EP
TE D
M AN U
SC
RI PT
351
15
ACCEPTED MANUSCRIPT particular, are good bioindicators, the seasonal fluctuation of trace element concentrations linked to
377
reproductive cycles is a drawback to their use because it complicates the comparison of data.
378
However, because triploid specimens of Pacific oyster are available from hatcheries, and because
379
they are characterized by the inhibition of gametogenesis, they are less subject to seasonal changes
380
than diploid individuals of other bivalves, making them more suitable for biomonitoring campaigns.
RI PT
376
381
4.2. Human exposure to harmful elements and environmental suitability for oyster farming
383
Oysters are edible marine organisms of commercial value worldwide; therefore, it is of primary
384
importance to assess whether there is a risk associated with their consumption due to trace element
385
concentrations in their tissues. Several studies (Eisler et al., 1972) have found that they have a
386
greater capacity to accumulate Cd, but also Cu, Fe, and Zn, than other bivalve species such as clams
387
(Baudrimont et al., 2005), mussels (Shulkin et al., 2003; Squadrone et al., 2016) or scallops (Mok et
388
al., 2014). Szefer et al. (1999) and Soto-Jiménez et al. (2001) demonstrated that the concentrations
389
of these four elements detected in the soft tissues of oysters and in the sediments are related by a
390
biota-sediment accumulation factor (BSAF) (Arnot and Gobas, 2006) of 60, 48, 8, and 5,
391
respectively, while Dominik et al. (2014) reported a BSAF of 0.36 for Hg.
392
With regard to European Regulations, the concentrations of Hg and Pb in all the 72 pooled samples
393
from sites along the Italian coasts were below the current maximum limits set by the European
394
Commission to safeguard human health. The only exception was the Cd level in one pooled sample
395
from Capoiale-Varano, very close to the legal limit of 1 mg kg-1 w.w. It is a non-essential metal,
396
highly toxic for humans, and probably carcinogen (IARC, 1993; Bernard A., 2008). Cd affects
397
typically the kidneys, by involving the proximal tubular cells that may lead to renal failure, and
398
bones by inducing osteoporosis (Jarups et al., 1998; EFSA, 2012). However, N resulted reasonably
399
high. Obviously, this value is estimated without taking into account additional intakes of cadmium
400
from the diet. High Cd levels could also influence the zootechnical performances in aquatic
AC C
EP
TE D
M AN U
SC
382
ACCEPTED MANUSCRIPT animals, interfering with biologic mechanisms, as observed in oysters (Sokolova et al., 2004),
402
mussels (Viarengo et al., 1994), sponges (Müller et al., 1998), sea urchins (Roccheri and Matranga,
403
2010), crustaceans (Verriopoulos and Moriatou-Apostolopoulou, 1981), and fish (Voyer et al.,
404
1979). In invertebrates, cadmium induces changes in tissue organization, immune response, and
405
gene expression (Chiarelli and Roccheri, 2014). It has been shown in C. virginica that cadmium
406
induces haemocyte apoptosis (Sokolova et al., 2004).
407
In the samples collected during this study, the Hg levels were low, also in the specimens from
408
polluted areas, probably due to their low-trophic level. Mercury has highly toxic effects on the
409
environment and human health, especially its methylated form methylmercury (CH3Hg) which
410
tends to bio-accumulate and bio-magnify in biota throughout the aquatic food chain (May et al.,
411
1987). The toxicity of mercury and mercury compounds for marine fauna has been demonstrated by
412
several studies (Liu et al., 2011; Longo et al., 2013; Singaram et al., 2013).
413
The sampling site of Muggia showed high concentrations of lead. If humans and animals ingest lead
414
over tolerable limits, it can damage the nervous system (Nava-Ruiz et al., 2012, WHO, 2015) and it
415
can easily cross the placental barrier (Goyer, 1995). There is no current PTI for Pb, however. Lead
416
causes also irregular early development in invertebrates (Tellis et al., 2014) and biochemical and
417
histological toxic effects (Hariharan et al., 2014) that may penalise the growing performances
418
during farming.
419
Arsenic chemical forms, including both inorganic and organic compounds, have extremely variable
420
toxicological effects. The non-toxic organic compounds, such as methylated arsenicals,
421
arsenocholine, arsenobetaine, arseno-sugars, and arseno-lipids (Gebel, 2001), are the predominant
422
forms present in marine animal tissues (Phillips, 1990; Francesconi et al., 1998). These compounds
423
are thought to be the final products of detoxification processes and are easily eliminated by animals
424
and humans (Kaise et al., 1985). Some exceptions aside (Phillips and Depledge, 1986; Fattorini et
425
al., 2004), these organic forms generally present in the majority of seafood such as bivalves
AC C
EP
TE D
M AN U
SC
RI PT
401
17
ACCEPTED MANUSCRIPT contribute to the low toxicity of these foodstuffs. However, the toxicity of soluble inorganic arsenic
427
has repercussions on marine biota, though the sensitivity depends on species and other
428
environmental factors (Sanders and Vermersch, 1982; Phillips, 1990).
429
Tin and aluminium are non-essential elements for humans. The Sn levels we detected in the samples
430
from all the sampling sites were too low to represent a danger for human health. Nonetheless, a
431
trace presence of Sn in water