(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the author's institution and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited.

In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit:

http://www.elsevier.com/authorsrights Author's Personal Copy

Toxicon 159 (2019) 32–37

Contents lists available at ScienceDirect

Toxicon

journal homepage: www.elsevier.com/locate/toxicon

Comparison of amnesic, paralytic and lipophilic toxins profiles in (Acanthocardia tuberculata) and smooth ( chione) from the T central Adriatic Sea (Croatia)

∗ Ivana Ujević, Romana Roje-Busatto , Daria Ezgeta-Balić

Institute of Oceanography and Fisheries, Šetalište Ivana Meštrovića 63, 21000 Split, Croatia

ARTICLE INFO ABSTRACT

Keywords: Searching for Amnesic (ASP), Paralytic (PSP) and Lipophilic (LT) toxins in is of great importance for Acanthocardia tuberculata consumer protection. Studies are usually focused on the most aquacultured species, the . But, there are a number of potentially commercially important shellfish species as rough cockle Acanthocardia tuberculata Lipophilic toxins (Linnaeus, 1758) and smooth clam Callista chione (Linnaeus, 1758) which are common in the Croatian Adriatic PSP Sea. Investigation of marine biotoxins accumulation in these two species of shellfish from the Adriatic Sea has ASP not been conducted up to now. In order to detect the potential marine biotoxin profile of A. tuberculata and C. Adriatic sea chione wild populations, samples were taken monthly during one-year survey from the estuarine area in the central Adriatic Sea. HPLC-FLD with pre-column oxidation and HPLC-UV-DAD methods were employed for PSP and ASP toxins determination, respectively, while LTs were determined by LC-MS/MS. This research had re- vealed the differences in the accumulation of ASP, PSP and LT toxins between the two studied species, as Acanthocardia tuberculata showed more diverse profile with higher concentrations of analysed toxins. Both in- vestigated shellfish species had shown levels of these biotoxins under the legal limits set by the European Commission.

1. Introduction the Mediterranean coasts of Spain and Morocco (Márquez, 1993; Taleb et al., 2001). Along the Croatian coast, PSP toxicity was sporadically Rough cockle (Acanthocardia tuberculata, Linnaeus, 1758) and the recorded and until today it was detected only in eleven samples of smooth clam (Callista chione, Linnaeus, 1758) are filter feeding shellfish (Mytilus galloprovincialis) collected throughout that naturally inhabit mud, sand or gravel bottoms of the winter 2009 from the southern Istria Peninsula coast (northern Adriatic and the Northeast Atlantic Ocean. On the northern Sea) and in the Mali Ston Bay (southern Adriatic Sea). Saxitoxin was the coast of Morocco A. tuberculata is being intensively exploited and even dominant PSP toxin type extracted from all contaminated leading to overexploitation of their natural banks (Rharrass et al., within the wide range of concentrations, resulting in three total toxi- − 2016). Callista chione is of commercial interest in France (Charles et al., cities above the maximum permitted levels of 800 μg STX eq. kg 1, 1999) and Greece (Metaxatos, 2004) while in Spain, Catalan coast, is according to EU legislation (Ujević et al., 2012). It is interesting to note under experimental studies to determine its suitable broodstock con- that concomitantly to this toxicity in mussels sampled near the southern ditions (Delgado et al., 2016). In the Croatian part of the Adriatic Sea A. Istria Peninsula coast there was also an ascidian (Microcosmus vulgaris) tuberculata and C. chione are common and in terms of bivalve commu- PSP toxicity recorded at the western coast but below the maximum nity biomass and abundance they are among the dominant species on permitted levels (EC 853/2004). Other investigated species: European soft sediments (Peharda et al., 2010), but there is still no data about flat (), Mediterranean () and commercial exploitation and overfishing of these species. Some pre- proteus scallop (Flexopecten proteus) during the same period on these vious studies have been conducted on the biology and accumulation of sampling stations exhibited no PSP toxin presence (Roje-Busatto and biotoxins in these species (Vale and Sampayo, 2002; Sagou et al., 2005; Ujević, 2014). Vale and Taleb, 2005; Takati et al., 2007). Acanthocardia tuberculata is In the Mediterranean coast of Morocco low levels of domoic acid known to often display Paralytic Shellfish Poisoning (PSP) toxicity in (DA), the main toxin responsible for the Amnesic Shellfish Poisoning

∗ Corresponding author. . E-mail address: [email protected] (R. Roje-Busatto). https://doi.org/10.1016/j.toxicon.2018.12.008 Received 17 August 2018; Received in revised form 3 December 2018; Accepted 21 December 2018 Available online 16 January 2019 0041-0101/ © 2019 Elsevier Ltd. All rights reserved. Author's Personal Copy

ć I. Ujevi et al. Toxicon 159 (2019) 32–37

(ASP), were found in sweet sampled in 2007 (Rijal Leblad et al., autosampler. Separation of toxin oxidation products was carried out on 2013). ASP toxicity was recorded for the first time in the eastern part of reversed-phase C18 Restek column, Pinnacle II 250 × 4.6 mm and 5 μm the Adriatic Sea during 2006–2008 in four shellfish species: mussel (M. particle size, protected by a Pinnacle II C18 guard cartridge, galloprovincialis), oyster (O. edulis) and (P. jacobaeus and F. 20 × 4.0 mm (Varian, SAD; chromatographic conditions have been proteus), but in only few samples and in concentrations that did not explained in detail in Ujević et al., 2012). exceed the legal limits set by the European Commission. The highest PSP toxins were determined by the pre-column HPLC-FLD method concentrations were recorded in mussels from February through March on peroxide and periodate oxidised whole tissue Solid Phase Extraction 2006 during a Pseudo-nitzschia spp. bloom (Ujević et al., 2010). cleaned extracts and identified by comparing to chromatograms of NRC In comparison to other two aforementioned toxicities, frequent li- standard solutions (Halifax, Canada). Toxicity was calculated by EFSA pophilic shellfish toxicity (LT) events were recorded from 2006 through (2009) Specific Toxicity Factors. Injections of four toxin standard so- 2011 at farming areas along the eastern Adriatic coast by Mouse lutions mixtures were made in order to obtain five points calibration − Bioassay (MBA), but only a few samples were confirmed positive by curves within the ranges of 0.16–0.96 μgmL 1 for saxitoxin (STX), liquid chromatography tandem-mass spectrometry (LC-MS/MS) decarbamoylsaxitoxin (dcSTX), neosaxitoxin (NEO) and gonyautoxins 1 − method (Arapov et al., 2015; Ninčević-Gladan et al., 2011). Most of and 4 (GTX-1,4); 0.016–0.096 μgmL 1 for gonyautoxin 5 (GTX-5) and − those analysed MBA positive samples contained only gymnodimine and N-sulfocarbamoylgonyautoxin 1 and 2 (C-1,2); 0.08–0.48 μgmL 1 for − spirolide toxins confirmed by LC-MS/MS analyses as gymnodimines decarbamoylneosaxitoxin (dcNEO); 0.096–0.576 μgmL 1 for gonyau- − cause very rapid mouse death when extracted by dichloromethane toxins 2 and 3 (GTX-2,3) and 0.064–0.384 μgmL 1 for dec- which explains numerous positive MBA results from that period (Ujević arbamoylgonyautoxins 2 and 3 (dcGTX-2,3). Spiking the mussel tissue et al., 2015). homogenates with known concentrations of PSP toxins showed re- − Aims of this study were to reveal biotoxins profiles in natural covery of 75%–102%. Detection limits (LOD) were below 10 μgkg 1 shellfish populations of A. tuberculata and C. chione in the area with for all the toxin types. Limits of Detection (LOD) were: 1.07 for STX, minor anthropogenic impact, to identify potential differences in accu- 1.91 for dcSTX, 1.78 for NEO, 6.81 for GTX-1,4, 0.39 for GTX-5, 4.22 − mulation and imply possible seasonal variations of marine biotoxins for C-1,2, 2.09 for dcNEO, 2.14 for GTX-2,3 and 1.07 μgkg 1 for that pose potential risk for human consumption. dcGTX-2,3. Limits of Quantification (LOQ) were: 3.41 for STX, 6.08 for dcSTX, 11.31 for NEO, 21.69 for GTX-1,4, 1.24 for GTX-5, 13.42 for C- μ −1 2. Materials and methods 1,2, 6.65 for dcNEO, 6.83 for GTX-2,3 and 12.86 gkg for dcGTX- 2,3. 2.1. Area of investigation and sampling fi Biotoxin profiles of natural populations of A. tuberculata and C. 2.3. Analysis of Amnesic shell sh poisoning toxins (ASP) chione were monitored monthly from March 2009 through February fi 2010 with the purpose of ASP, PSP and LT marine biotoxins preliminary Preparation of homogenised whole body shell sh samples and high- assessment. Thirty similar sized individuals (15 of each shellfish spe- performance liquid chromatography with ultraviolet diode array de- cies) were collected monthly from 1.5 to 3 m deep sampling area in the tection (HPLC-UV-DAD) analysis employed for determination of am- Cetina River estuary (central part of the eastern Adriatic Sea, Croatia; nesic toxins, DA and epi-DA (domoic and epi-domoic acid, respectively) ć Fig. 1) by SCUBA diving and kept frozen until analysis. Temperature was the same as described in Ujevi et al. (2010) and according to the and salinity were measured monthly with YSI Pro hydrographic probe protocol proposed by Quilliam et al. (1995). The HPLC system, in ad- fi above the sea bed at a depth of 2 m (Peharda et al., 2012). dition to modules speci ed above, consisted of a Varian ProSTAR 310 UV/Vis Detector and 335 Photodiode Array Detector. The column used was a Pinnacle II C18, 250 × 4.6 mm (Restek), with a C18 Guard Car- fi 2.2. Analysis of Paralytic Shell sh Poisoning toxins (PSP) tridge (20 × 4 mm) and temperature set to 40 °C. Domoic acid ab- sorption maximum, UV wavelength of 242 nm, was used for detection Paralytic toxins were extracted from the whole body homogenate of and DAD detector for the confirmation of domoic acid presence in the fi ffi each shell sh species following the AOAC O cial Method 2005.06 and samples. Certified Reference Material ASP-Mus-c (National Research samples cleaned with SPE C18 and COOH cartridges (according to Council of Canada, Halifax, NS, Canada) showed recovery of 96%–99%, − − Lawrence et al., 2005). The extracts were analysed using the high- while LOD and LOQ for domoic acid were 0.10 μgg 1 and 0.34 μgg 1, performance liquid chromatography pre-chromatographic oxidation respectively. method with fluorescence detection (HPLC-FLD), employing a Varian ProStar HPLC analytical system coupled with 363 fluorescence detector (Ex 340 nm, Em 390 nm), 230 solvent delivery module and 410

Fig. 1. Sampling area in the Cetina River estuary (43°26′N, 16°41′E), central Adriatic Sea.

33 Author's Personal Copy

ć I. Ujevi et al. Toxicon 159 (2019) 32–37

2.4. Analysis of lipophilic toxins (LT)

For the analysis of lipophilic group of toxins: okadaic acid (OA), dinophysistoxin-1 and 2 (DTX-1,2), pectenotoxin-1 and 2 (PTX-1,2), yessotoxins (YTXs), azaspiracids (AZAs), gymnodimine (GYM) and spirolide (SPX), homogenised whole body shellfish tissue was extracted (EURLMB- Harmonised SOP, 2015) prior to injection into the LC system. LC-MS/MS analysis was performed using a triple quadrupole mass spectrometer (Agilent Technologies 6410) equipped with an electrospray ionisation source. Chromatographic separation was per- formed using a 5 μm Poroshell C18, 50 × 2.1 mm Agilent column kept at 30 °C. Multiple Reaction Monitoring (MRM) experiments were car- ried out in positive and negative modes. The hydrolysis procedure as proposed by EURLMB (2015) was applied to investigate total content of OA and DTXs toxins by converting acylated esters to their parent form. EURLMB-Vigo Proficiency Testing samples showed recovery of 93%–110%. Limits of Detection (LOD) were: 1.77 for AZA1, 2.34 for AZA2, 8.50 for AZA3, 2.20 for OA, 4.19 for DTX1, 3.64 for DTX2, 3.05 − for PTX2, 4.79 for YTX, 8.38 for hYTX, 1.16 for GYM and 0.70 μgkg 1 for SPX. Limits of Quantification (LOQ) were: 5.68 for AZA1, 7.49 for AZA2, 8.50 for AZA3, 7.05 for OA, 13.40 for DTX1, 11.66 for DTX2, 9.75 for PTX2, 15.34 for YTX, 26.83 for hYTX, 7.54 for GYM and − 6.56 μgkg 1 for SPX. Fig. 3. Examples of chromatograms obtained by chromatographic HPLC-UV- DAD analysis of: A) rough cockle (Acanthocardia tuberculata) extract containing − 0.77 μgg 1 domoic acid (DA; Retention Time is 11.877 min); B) smooth clam − 3. Results (Callista chione) extract containing 0.28 μgg 1 domoic acid (DA; Retention Time is 11.948 min). 3.1. ASP toxin level

Amnesic Shellfish Poisoning toxins, DA and epi-DA were determined by HPLC-UV-DAD method. Low concentrations of DA, close to detection limit, were determined in A. tuberculata samples from April through − July 2009 with maximum of 0.77 μgg 1 recorded in May and − minimum concentration of 0.17 μgg 1 in June 2009 (Fig. 2), while C. chione had only one episode of DA toxicity in April 2009 with con- − centration of 0.28 μgg 1 (Figs. 2 and 3). All the results were below the − maximum permitted level (20 mg kg 1) set by the by EU legislation.

Fig. 4. Example of chromatogram obtained by HPLC-FLD with pre-column fi 3.2. PSP toxin pro le periodate oxidation (AOAC, 2005.06) of the rough cockle (Acanthocardia tu- berculata) C-18 whole tissue extract sampled in March 2009, containing sax- Pre-column HPLC-FLD analysis was conducted to detect the poten- itoxin (STX; Retention Time is 10.655 min). tial presence of STX, NEO, GTX2,3, GTX1,4, dcSTX, dcNEO, dcGTX2,3, fi − GTX 5 and C1,2. Analysis of A. tuberculata has identi ed STX (Fig. 4)as (10.82 μgkg 1) in November 2009, but it also contained other toxins the dominant toxin type and revealed its continuous presence during μ −1 μ −1 − belonging to this group (108.82 gkg of GTX5, 12.41 gkg of GTX – μ 1 − the whole investigated period within the range 10.82 98.04 gkg 2,3, and 4.21 μgkg 1 of dcSTX). (Fig. 5). At the same time, C. chione exhibited low levels of PSP con- − − tamination by GTX2,3 (33.92 μgkg 1) and STX (26.59 μgkg 1) in only two samples, from April and May 2009. The highest concentration of − STX in A. tuberculata was found in April (98.04 μgkg 1) and the lowest

Fig. 2. Domoic acid (DA) concentrations found in rough cockle (Acanthocardia tuberculata) and smooth clam (Callista chione) whole tissue samples collected at Cetina River estuary in the central Adriatic Sea (Croatia) from March 2009 through February 2010. Error bars represent ± 1SD (n = 15).

34 Author's Personal Copy

ć I. Ujevi et al. Toxicon 159 (2019) 32–37

Fig. 5. Paralytic Shellfish Poisoning toxins found in rough cockle (Acanthocardia tuberculata) and smooth clam (Callista chione) whole tissue samples collected at Cetina River estuary in the central Adriatic Sea (Croatia) from March 2009 through February 2010. All unmarked bars correspond to saxitoxin (STX). Error bars represent ± 1SD (n = 15).

3.3. LT toxin profile DA levels. These concentrations are within the lower levels found in − mussels from the southern Portugal (0.5–5.3 μgg 1; Vale and Sampayo, Positive ion mode LC/MS/MS analysis was applied for detection of 2002). Accumulation of DA in A. tuberculata and C. chione soft tissues is AZAs, PTX1, PTX2, SPX and GYM, while negative mode was used for a result of Pseudo-nitzschia spp. activity that is commonly present in the YTXs, OA, DTX1, DTX2, as well as acylated esters of OA and DTXs Adriatic Sea throughout the whole year (Bosak et al., 2009; Ujević detection. Results revealed that both shellfish species contained GYM et al., 2010). It has been noted that the abundance of toxic diatom and SPX (Figs. 6–8) toxins in very low concentrations. In A. tuberculata Pseudo-nitzschia calliantha (Lundholm, Moestrup and Hasle, 2003) from samples, the mean value of GYM concentration was 5.61, with the southern Adriatic is in negative correlation to seawater temperature − − 2.65–15.77 μgkg 1 range, while in C. chione tissue it was 3.06 μgkg 1 and in positive to nutrient content, principally nitrogen (Caroppo et al., − (1.17–6.14 μgkg 1 range). Ranges of SPX (13-desmethyl-spirolide C) in 2005). Ujević et al. (2010) also argue negative temperature to Pseudo- − A. tuberculata and C. chione tissues were 0.98–5.90 μgkg 1 and nitzschia spp. abundance correlation, where high abundance − − 0.65–2.14 μgkg 1, respectively. (> 1.0 × 106 cell L 1) of this species was noted in Šibenik area (river Krka estuary, central Adriatic Sea) during February 2006, when mussel − 4. Discussion M. galloprovincialis showed DA concentration of 6.55 μgg 1. These authors stated that this harvesting area may be considered ASP safe if − This is the first study that tackles the biotoxins presence in these two abundance of Pseudo-nitzschia spp. does not exceed 1.0 × 105 cells L 1. shellfish species, A. tuberculata and C. chione from the Adriatic Sea. Nonetheless, potentially dangerous DA concentrations can be accumu- During the investigated period DA was confirmed in A. tuberculata lated in some shellfish species even when Pseudo-nitzschia spp. is found tissue samples from April to July 2009 with concentration range from in low abundance as James at al. (2005) had measured levels of DA in − − 0.17 to 0.77 μgg 1, while C. chione samples exhibited 0.28 μgg 1 DA soft tissue of that were high above the legal limit during in samples from April 2009. However, all DA values were below the the continuously low Pseudo-nitzschia spp. abundance. − legal limit of 20 μgg 1. It was noted that in Cetina River estuary A. Saxitoxin concentrations in A. tuberculata samples from May 2009 tuberculata had the lowest CI (Condition Index) sometime between April and from November 2009 through January 2010 were lower than − − and May, and in general, it was lower during the summer months than 40 μgkg 1, while the maximum of almost 100 μgkg 1 was detected in during the rest of the year (Peharda et al., 2012). The low CI could be April 2009. Only in the sample from April toxins other than STX were one of the reasons for high levels of DA found in A. tuberculata in May, determined in A. tuberculata: dcSTX, GTX2,3 and GTX5; while low le- i.e. lower quantity of shellfish tissue (due to the loss of gametes after vels of GTX2,3 and STX were found in only two samples of C. chione, spawning) with the same rate of toxin accumulation results in higher those from April and May 2009 (Fig. 5). These findings coincide with

Fig. 6. Example of chromatogram obtained by LC-MS/MS analysis of the rough cockle (Acanthocardia tuberculata) whole tissue extract sampled in October 2009, containing gymnodimine and spirolide toxins (Retention Times are 9.036 and 10.043 min for GYM and SPX, respectively).

35 Author's Personal Copy

ć I. Ujevi et al. Toxicon 159 (2019) 32–37

Fig. 7. Gymnodimine concentrations found in rough cockle (Acanthocardia tuberculata) and smooth clam (Callista chione) whole tissue samples collected at Cetina River estuary in the central Adriatic Sea (Croatia) from March 2009 through February 2010. Error bars represent ± 1SD (n = 15). the PSP toxins presence in mussels M. galloprovincialis and ascidians M. collected in 2003. Elaborate study of Ujević et al. from 2015 revealed vulgaris from the northern Adriatic Sea during the same period, with the these toxins to be continuously present in mussel samples collected exception of GTX5 that was absent in mussels and C 1,2 that was ad- from January through June 2011 at mussel farms in Mali Ston Bay ditionally present in ascidian (Ujević et al., 2012; Roje-Busatto and (southern Adriatic Sea). Majority of these samples exhibited low con- − Ujević, 2014). In the Algarve, southern Portugal, A. tuberculata was centrations of GYM and SPX, within the range 5–15 μgkg 1, which is in dominated by STX with smaller amounts of dcSTX (Vale and Sampayo, accordance with the results obtained in the present study for A. tu- 2002). As A. tuberculata has specific metabolism it collects and pre- berculata and C. chione. Concentrations of SPX were lower than those of serves PSP toxins longer than do other bivalves (including C. chione) GYM in both shellfish; nevertheless, the annual mean value of SPX was even when potentially toxic phytoplankton is not present or is present higher in A. tuberculata than in C. chione samples (2.8-fold), just as was at low abundance (Taleb et al., 2001; Vale and Sampayo, 2002; Sagou the case with GYM (1.8-fold higher). Ranges of SPX in A. tuberculata − − et al., 2005; Takati et al., 2007). Results obtained from this study seem and C. chione tissues were 0.98–5.90 μgkg 1 and 0.65–2.14 μgkg 1, to confirm these statements. Sagou et al. (2005) argue that these two respectively. The highest SPX and GYM concentrations in A. tuberculata species accumulate and retain PSP toxins in a different way as they have been detected in wormer part of the year, while C. chione had the undergo different metabolic pathways. Acanthocardia tuberculata highest concentrations of SPX in samples collected in April and GYM in showed higher and more constant toxicity than C. chione, and the April and November. No SPX was detected (< LOD) in November highest toxin concentrations were found in their foot and hepatopan- through February period. Ciminiello et al. (2010) have revealed SPXs creas, respectively. These authors demonstrated that during the rela- (13-desMethyl spirolide C and 27-hydroxy-13,19-didesMethylspirolide tively long depuration period the amount of STX and dcSTX increase at C) in mussel M. galloprovincialis, during the high abundance of A. os- − the expense of other PSP toxin types. This statement may explain the tenfeldii (1.5 × 105 cell L 1) in the Adriatic Sea, while low levels − more diverse PSP profile in the present study that was noted at peak (2–60 μgkg 1) of SPXs were found in mussels and from Cata- toxicity in April, and its subsequent lower, yet persistent STX con- lonia, Spain (García-Altares et al., 2014). Nevertheless, Van Wagoner tamination in A. tuberculata; but also the short period of toxicity found et al. (2011), besides SPX, have isolated novel gymnodimine congener in C. chione as it, according to Sagou et al. (2005), accumulates PSP 12-methylgymnodimine from the A. peruvianum and for the first time toxins in the digestive gland and depurates much faster. To be noted these two cyclic imine were found within the same dinoflagellate. These that all the results in this study were below the maximum permitted results infer that in the Adriatic Sea shellfish species, other than mus- − level (800 μg STX eq. kg 1) set by EU legislation. sels, may accumulate GYM and SPX in low concentrations throughout The presence of cyclic imines in the Mediterranean Sea was con- the whole year. SPX and GYM are the toxins that were not reported to firmed in 2002, when GYM-A was found for the first time in clams from cause the human poisoning, but their accumulation in Mediterranean Tunisia (Biré et al., 2002). Ciminello et al. (2006) have established that shellfish is existent and calls for further research. GYM and SPX are present in the northern Adriatic Sea and they also confirmed the first occurrence of possible spirolide producer Alexan- drium ostenfeldii reported for Emilia Romagna region for samples

Fig. 8. Spirolide concentrations found in rough cockle (Acanthocardia tuberculata) and smooth clam (Callista chione) whole tissue samples collected at Cetina River estuary in the central Adriatic Sea (Croatia) from March 2009 through February 2010. Error bars represent ± 1SD (n = 15).

36 Author's Personal Copy

ć I. Ujevi et al. Toxicon 159 (2019) 32–37

5. Conclusions Delgado, M., Pérez-Larruscain, J., Ignasi Gairín, J., 2016. Broodstock conditioning and gonadal development of the smooth clam Callista chione (Linnaeus, 1758) (: – fi fi ) on the Catalan coast (NE Spain). Acta Adriat. 57 (1), 93 106. This is the rst study that reveals marine biotoxins pro les in wild European Food Safety Authority (EFSA), 2009. Scientific opinion of the panel on con- populations of rough cockle (Acanthocardia tuberculata, Linnaeus, 1758) taminants in the food chain on a request from the european commission on marine and the smooth clam (Callista chione, Linnaeus, 1758) in the eastern biotoxins in shellfish – saxitoxin group. EFSA J 1019, 1–76. European Reference Laboratory for Marine Biotoxins (EURLMB), 2015. EU Harmonised Adriatic Sea (Croatia). Acanthocardia tuberculata was dominated by the Standard Operating Procedure for Determination of Lipophilic Marine Biotoxins in saxitoxin, gymnodimine and spirolide but also showed occasional do- Molluscs by LC-MS/MS. Version 5, Jan 2015. http://www.aecosan.msssi.gob.es/ moic acid presence in low concentrations while the C. chione was CRLMB/docs/docs/metodos_analiticos_de_desarrollo/EU-Harmonised-SOP-LIPO- mainly PSP and ASP toxins free and gymnodimine and spirolide were LCMSMS_Version5.pdf, Accessed date: 19 July 2018. European Regulation (EC) No 853/2004 of the European Parliament and of the Council, found in even lower concentrations than in A. tuberculata. All measured 29 April. biotoxin levels were below the maximum permitted levels set by EU García-Altares, M., Casanova, A., Bane, V., Diogène, J., Furey, A., De la Iglesia, P., 2014. fi fi legislation. Con rmation of pinnatoxins and spirolides in shell sh and passive samplers from Catalonia (Spain) by liquid chromatography coupled with triple quadrupole and high-resolution hybrid tandem mass spectrometry. Mar. Drugs 12 (6), 3706–3732. Ethical statement for toxicon James, K.J., Gillman, M., Amandi, M.F., López-Rivera, A., Puente, P.F., Lehane, M., Mitrović, S., Furey, A., 2005. Amnesic shellfish poisoning toxin in bivalve molluscs in Ireland. Toxicon 46, 852–858. I testify on behalf of all co-authors that our article submitted to Lawrence, J.F., Niedzwiadek, B., Menard, C., 2005. Quantitative determination of pa- Toxicon: ralytic shellfish poisoning toxins in shellfish using prechromatographic oxidation and liquid chromatography with fluorescence detection: collaborative study. J. AOAC Int. 88, 1714–1732. 1) this material has not been published in whole or in part elsewhere; Márquez, I., 1993. Presencia de PSP en el corruco (Acanthocardia tuberculata) en el litoral 2) the manuscript is not currently being considered for publication in de la provincia marítima de Malaga y distritos maritimos de la linea y Algeciras. In: another journal; J., Mariño, J.C., Maneiro (Eds.), III Reunión Iberica sobre Fitoplancton Tóxico y Biotoxinas. Consejería de Agricultura y Pesca, Xunta de Galicia, pp. 27–33. 3) all authors have been personally and actively involved in sub- Metaxatos, A., 2004. Population dynamics of the venerid bivalve Callista chione (L.) in a stantive work leading to the manuscript, and will hold themselves coastal area of the eastern Mediterranean. J. Sea Res. 52, 293–305. jointly and individually responsible for its content. Ninčević-Gladan, Ž., Ujević, I., Milandri, A., Marasović, I., Ceredi, A., Pigozzi, S., Arapov, J., Skejić, S., 2011. Lipophilic toxin profile in Mytilus galloprovincialis during episodes of diarrhetic shellfish poisoning (DSP) in the NE Adriatic Sea in 2006. Molecules 16, Conflicts of interest 888–899. Peharda, M., Ezgeta-Balić, D., Vrgoč, N., Isajlović, I., Bogner, D., 2010. Description of The authors declare no conflict of interest. bivalve community structure in the Croatian part of the Adriatic Sea - hydraulic dredge survey. Acta Adriat. 51 (2), 141–157. Peharda, M., Ezgeta-Bali ć, D., Radman, M., Sinjkević, N., Vrgoč, N., Isajlović, I., 2012. Acknowledgments Age, growth and population structure of Acanthocardia tuberculata (Bivalvia: cardi- idae) in the eastern Adriatic Sea. Sci. Mar. 76, 59–66. Quilliam, M.A., Xie, M., Hardstaff, W.R., 1995. Rapid extraction and cleanup for liquid This study was supported by the Ministry of Science and Education chromatographic determination of domoic acid in unsalted food. J. AOAC Int. 78, of the Republic of Croatia as a part of Multiannual Financing intended 543–554. for institutions. Authors are grateful to Mario Zokić for help with Rharrass, A., Talbaoui, M., Gaspar, M., Kabine, M., Rharbi, N., 2016. Gametogenic cycle of the rough cockle Acanthocardia tuberculata (Mollusca: Bivalvia) in the M’diq Bay sample collection. (SW Mediterranean Sea). Sci. Mar. 80 (3), 359–368. Rijal Leblad, B., Lundholm, N., Goux, D., Véron, B., Sagou, R., Taleb, H., Nhhala, H., Er- Transparency document Raioui, H., 2013. Pseudo-nitzschia Peragallo (Bacillariophyceae) diversity and do- moic acid accumulation in tuberculate cockles and sweet clams in M’diq Bay, Morocco. Acta Bot. Croat. 72 (1), 35–47. Transparency document related to this article can be found online at Roje-Busatto, R., Ujević, I., 2014. PSP toxins profile in ascidian Microcosmus vulgaris https://doi.org/10.1016/j.toxicon.2018.12.008. (Heller, 1877) after human poisoning in Croatia (Adriatic Sea). Toxicon 79, 28–36. Sagou, R., Amanhir, R., Taleb, H., Vale, P., Blaghen, M., Lotfi, M., 2005. Comparative study on differential accumulation of PSP toxin between cockle (Acanthocardia tu- References berculatum) and sweet clam (Callista chione). Toxicon 46 (6), 312–618. Takati, N., Mountassif, D., Taleb, H., Lee, K., Blaghen, M., 2007. Purification and partial characterization of paralytic shellfish poison-binding protein from Acanthocardia tu- Arapov, J., Ujević, I., Ninčević Gladan, Ž., Skejić, S., Ceredi, A., Milandri, A., Pigozzi, S., berculatum. Toxicon 50, 311–321. Riccardi, E., Vilar-González, A., Luisa Rodríguez-Velasco, M., Nazlić, N., Marasović, Taleb, H., Vale, P., Jaime, E., Blaghen, M., 2001. Study of paralytic shellfish poisoning I., 2015. Shellfish lipophilic toxin profile and toxic phytoplankton species along toxin profile in shellfish from the Mediterranean shore of Morocco. Toxicon 39 (12), eastern Adriatic coast. Fresenius Environ. Bull. 24, 4801–4806. 1855–1861. Biré, R., Krys, S., Frémy, J.-M., Dragacci, S., Stirling, D., Kharrat, R., 2002. First evidence Ujević, I., Ninčević-Gladan, Z., Roje, R., Skejić, S., Arapov, J., Marasović, I., 2010. Domoic on occurrence of gymnodimine in clams from Tunisia. J. Nat. Toxins 11, 269–275. acid e a new toxin in the Croatian Adriatic shellfish toxin profile. Molecules 15 (10), Bosak, S., Burić, Z., Djakovac, T., Viličić, D., 2009. Seasonal distribution of plankton 6835–6849. diatoms in Lim Bay, northeastern Adriatic Sea. Acta Bot. Croat. 68, 351–365. Ujević, I., Roje, R., Ninčević-Gladan, Ž., Marasović, I., 2012. First report of paralytic Caroppo, C., Congestri, R., Bracchini, L., Albertano, P., 2005. On the presence of Pseudo– shellfish poisoning (PSP) in mussels (Mytilus galloprovincialis) from eastern Adriatic nitzschia calliantha Lundholm, moestrup et Hasle and Pseudo-nitzschia delicatissima Sea (Croatia). Food Control 25, 285–291. (Cleve) Heiden in the Southern Adriatic Sea (Mediterranean Sea, Italy). J. Plankton Ujević, I., Vuletić, N., Lušić, J., Nazlić, N., Kušpilić, G., 2015. Bioaccumulation of trace Res. 27, 163–774. metals in mussel (Mytilus galloprovincialis) from Mali Ston Bay during DSP toxicity Charles, F., Amouroux, J.M., Grémare, A., 1999. Comparative study of the utilization of episodes. Molecules 20 (7), 13031–13040. bacteria and microalgae by the suspension-feeding bivalve: Callista chione. J. Mar. Vale, P., Sampayo, M.A.M., 2002. Evaluation of marine biotoxin's accumulation by Biol. Assoc. U. K. 79, 577–584. Acanthocardia tuberculatum from Algarve, Portugal. Toxicon 40, 511–517. Ciminiello, P., Dell'Aversano, C., Fattorusso, E., Magno, S., Tartaglione, L., Cangini, M., Vale, P., Taleb, H., 2005. Assessment of the quantitative determination of paralytic Pompei, M., Guerrini, F., Boni, L., Pistocchi, R., 2006. Toxin profile of Alexandrium shellfish poisoning toxins by pre-column derivatization and elimination of interfering ostenfeldii (Dinophyceae) from the northern Adriatic Sea revealed by liquid chro- compounds by solid phase extraction. Food Addit. Contam. 22 (9), 838–846. matography-mass spectrometry. Toxicon 47 (5), 597–604. Van Wagoner, R.M., Misner, I., Tomas, C.R., Wright, J.L.C., 2011. Occurrence of 12- Ciminiello, P., Dell'Aversano, C., Fattorusso, E., Forino, M., Tartaglione, L., Boschetti, L., methylgymnodimine in a spirolide-producing dinoflagellate Alexandrium peruvianum Rubini, S., Cangini, M., Pigozzi, S., Poletti, R., 2010. Complex toxin profile of Mytilus and the biogenetic implications. Tetrahedron Lett. 52, 4243–4246. galloprovincialis from the Adriatic Sea revealed by LC-MS. Toxicon 55 (2–3), 280–288.

37