A Year of Bio-Monitoring (2021): Presence of Algae of the Genus Alexandrium, Dinophysis, Prorocentrum and Non-Compliance for Paralytic Toxins and Lipophilic Toxins in Bivalve Mollusks Bred in Sardinia (W Mediterranean Sea)

Abstract

Bivalve mollusk production represents the principal aquaculture activity in Sardinia (40°03′ N, 9°05′ E). In 2021, 859 water samples and 1270 mollusk samples were analyzed. The species Alexandrium minutum caused the accumulation of Paralytic Shellfish Toxins (PST) in three samples of bivalve mollusks. Dinophysis acuminata complex caused the accumulation of lipophilic toxins (LTs) belonging to the okadaic acid group (OAs) in 18 samples of bivalve mollusks. The research of paralytic shellfish toxins (PSTs) in shellfish samples has been carried out with LC-FLD, as mentioned in the AOAC 2005 Official Method 2005.06. The determination of LTs was carried out by LC-MS/MS analysis. DTX2, belonging to the group of OA toxins, was detected for the first time in Sardinia, in mussels sampled in Tortolì. The presence of Dinophysis and Prorocentrum species was correlated with the accumulation of the OA toxin group in bivalve mollusks, showing a certain repeatability at certain times of the year in the areas included in the study. The results of the present study can help to plan and organize more effective bio-monitoring sampling strategies.

1. Introduction

The European shellfish supply chain is of primary importance, with an estimated production of almost 625,895 tons of bivalves, reaching a value of EUR 1.24 billion in 2017 [1]. In particular, the European production amounted to 464,240 tons of mussels, 45,505 tons of clams and 99,857 tons of oysters. Shellfish farming in Sardinia is mainly based on small production enterprises: the traditional shellfish production was based on the collection of bivalve mollusks from natural beds; nowadays, the current production is based on intensive farming and harvesting [2]. Mussels (Mytilus galloprovincialis) are the main bivalve species bred in Sardinia, producing over 13,000 tons per year [3].

As they are filter-feeding animals, the production of bivalve mollusks is strongly influenced by the presence of phytoplankton, which represents one of the main sources of nourishment. At the same time, phytoplankton may represent a possible hazard, leading to the accumulation of various biotoxins.

The presence and amount of phytoplankton are affected by several environmental conditions [4], such as irradiance, temperature, turbulence, inorganic and organic nutrients, vitamins, and oligoelements [5]. The scientific community recently defined harmful algal bloom (HAB), stating that they can exist even if the microalgal densities do not exceed 100,000 cells/L. This is because worrying levels of marine biotoxins can accumulate in fish products even in low concentrations of toxic phytoplankton [6]. According to previous studies, HABs have increased in frequency, intensity, and geographical distribution in the last few decades, also affecting Mediterranean waters. HABs also have a negative impact on human health and the aquaculture industry [6,7,8].

Among the harmful algal species (HAS), the dinoflagellate group is well known [9].

It includes species of the Alexandrium genus, able to produce toxic compounds named paralytic shellfish toxins (PSTs), including saxitoxin (STX) and gonyautoxins (GTX) and other analogues that may have a great impact on human health [10]. STX and its approximately 60 analogues [11] are neurotoxic alkaloids responsible for paralytic shellfish poisoning (PSP), a human intoxication syndrome, resulting from the ingestion of bivalve mollusks contaminated with PSTs [12]. The main symptoms include numbness, loss of balance, tingling of feet, face, and hands, nausea, muscular paralysis and respiratory distress, sometimes leading to death through respiratory paralysis [12,13,14]. PSTs not only represent a potential hazard to public health but also result in economic losses due to the ban on bivalve mollusk harvesting. The legal limit regarding the PSTs reported in the EC Reg. 853/2004 is 800 µg of equivalents of saxitoxin dihydrochloride per kilogram of the edible part (µg STX 2-HCl eq/kg e.p.) [15].

The most recent findings of PSTs in Sardinia were linked to the genus Alexandrium and were reported in 2018–2019 [16]. However, the concurrent presence of PSTs with A. minutum and A. pacificum has already been described in the Olbia Gulf in previous years [17]. The presence of STX associated with the presence of Alexandrium spp. has been described in Sardinia since the early 2000s [18,19]. Since then, several PST-positive samples have been detected, but no cases of seafood poisoning in humans due to the consumption of bivalve mollusks have been reported.

Among the lipophilic marine toxins (LTs), the okadaic acid group (OAs), which includes okadaic acid, Dinophysis toxins and their esters (OA, DTX1, DTX2 and their acyl fatty ester derivatives, DTX3), are responsible for diarrhetic shellfish poisoning (DSP), causing an intestinal symptom complex including nausea, vomiting, diarrhea, and abdominal pain after the consumption of contaminated seafood [20,21]. These toxins inhibit the serine/threonine protein phosphatases [21]. The HAS involved in the production of OAs toxins are of the Dinophysis genus [22], among which are D. fortii, D. caudata, D. sacculus, D. acuminata, D. acuta, and some species of the genus Prorocentrum [21] such as P. lima and P. mexicanum. The first episodes of OA discovery beyond the legal limits in Sardinia were reported in mussels from the Cagliari area in 2002–2003 [23]. Since 2016, the detection of bivalve mollusks in Sardinia not complying with the legal limits of 160 µg OA-eq kg p.e. (EC Reg. 853/2004) has been described as a recurring phenomenon [16].

In accordance with the EC Reg. 853/2004 [16], the investigated LTs include OA, DTXs, yessotoxins (YTXs) and azaspiracids (AZAs) [24]. The Sardinian Regional Monitoring Sampling Plan [25] has been in force since 1988 and was prepared and coordinated by the Department of Hygiene and Health of the Sardinia region in collaboration with experts at the Experimental Zooprophylactic Institute of Sardinia and the veterinary services of the Local Health Units of Sardinia. The goal of the present study is to investigate the annual composition and abundance of HAS and the toxins produced (PSP and LTs) in bivalve mollusks during 2021, in order to protect the seafood consumers’ health and the economic sustainability of regional aquaculture.

2. Materials and Methods

2.1. Study Site

Sardinia is a Western Mediterranean Island (40°03′ N, 9°05′ E) (Figure 1) that extends to an area of around 24,000 km² with 1800 km of coastline. In 2021, the bivalve mollusk production areas were 19. Mollusk farms are located both in marine and transitional waters along the coastline.

2.2. Collection of Samples

This research was carried out from January to December 2021 to investigate the presence and abundance of potentially toxic microalgae. Samples of water and several mollusk species were collected from the 19 sampling stations with a weekly sampling frequency included in the present study.

A total of 859 water and 1270 mollusks samples were collected in agreement with the Sardinian Regional Monitoring Sampling Plan [25]. The weekly sampling for both water and mollusks sometimes occurred simultaneously, sometimes a few days apart.

2.3. Water Samples

Samples were taken utilizing a clean polyethylene bottle (PE) at a depth of about 1 m from the surface. Water samples were analyzed for the HAS count by using Utermöhl’s method [26], as reported in the EU reference method UNI EN ISO 15204:2006 [27]. After shaking the samples as gently as possible 100 times, Lugol’s iodine was used to fix them. The samples were settled in a counting chamber of a volume of 25 cm³ for almost 12 h and analyzed under an inverted microscope at 200× magnification (Olympus I73). The microalgal concentration was expressed as the number of cells per liter (cells/L) with a detection limit of 120 cells/L, and a level of significance of 0.05 (EN 15204:2006) [27].

An inverted Nikon microscope (Eclipse Ti-U; objectives: 20 × CFI Plan Apo, 40 × CFI Plan Apo and 100 × CFI Plan Flour Oil), powered by a Canon DS-F12 camera and an ultraviolet (UV) 100 W mercury lamp, was employed to analyze the morphology of the fixed cells (×200, ×400 and ×1000 magnifications).

The identification of Alexandrium minutum was based on the morphology of the plates, according to the method described by Balech (1995) [28]. Cells were stained with Calcofluor White (Fluorescent Brightner 28, Sigma, Steinheim, Germany) and observed with UV epifluorescence (UV filter arrangement for 330–380 nm excitation and 420 nm emission wavelengths) (Fritz and Triemer, 1985) [29].

2.4. Mollusk Samples

A total of 1270 mollusk samples of the following species were analyzed: Mytilus galloprovincialis, Ruditapes decussatus, Crassostrea gigas, Cerastoderma glaucum, Solen marginatus and Venus verrucosa (

Table 1. Shellfish samples analyzed in 2021.

Bivalve Mollusk Species	Number of Analyzed Samples
Mytilus galloprovincialis	620
Ruditapes decussatus	311
Crassostrea gigas	215
Venus verrucosa	57
Cerastoderma glaucum	43
Solen marginatus	24
Total samples	1270

). The preparation of the samples followed the following steps: first of all, raw samples were thoroughly cleaned with fresh water on the outside of the shellfish. The mollusks were opened by cutting the adductor posterior muscle. The inside part of each bivalve mollusk was rinsed with fresh water to remove sand and foreign material. After the removal from the shell, the tissues were drained in a sieve to remove water. For a representative sampling, at least 100–150 g of pooled tissues were homogenized in a Waring blender at room temperature and divided into sub-samples stored at −20 °C. Sub-samples of these homogenates were used for the determination of PSTs and LTs [30,31].

2.5. Reagents and Standards

Water for the toxin analysis was prepared with a Milli-Q water purification system (Millipore, Burlington, Massachusetts, (MA) USA). Sample preparation of SPE reagents and pre-COX HPLC-FLD reagents were HPLC grade: acetonitrile and methanol (VWR International, USA).

Ammonium hydroxide (30%), glacial acetic acid and ammonium formate (reagent grade) were purchased from Sigma-Aldrich Corporation (St. Louis, MO, USA). Solvents for UHPLC-MS/MS and HPLC-FLD analysis were LC-MS grade (VWR International, USA).

Certified reference materials (CRM) of paralytic marine toxins, including STX, NEO, dcSTX, dcNEO, dcGTX2,3, GTX1,4, GTX2,3, GTX5, GTX6 and C1,2, and those for lipophilic marine toxins including OA, DTX1, DTX2, YTX, homo-YTX, AZA1, AZA2, AZA3, CRM-OA-Mussel (CRM OA-Mus) and CRM-AZA-Mussel (CRM-AZA-Mus) were purchased from National Research Council-Institute for Marine Biosciences (Halifax, NS, Canada).

2.6. Determination of Algal Biotoxins

2.6.1. PSTs

The detection of PSTs in shellfish samples was based on the AOAC 2005 Official Method 2005.06 [30] employing LC-FLD.

Shellfish Homogenate Extraction and Sample Extract Processing

Homogenates of mollusk tissue (5.0 ± 0.1) g were extracted twice with 3 mL of 1% acetic acid solution. The first extraction was performed in a boiling water bath for 5 min.

After cooling to room temperature, the extracts were centrifuged at 4000× g for 10 min. Both supernatants were combined, and the final volume was adjusted to 10 mL with water.

The extracts were cleaned-up using SPE C18 cartridges (Supelco, Sigma-Aldrich, St Louis, MI (USA)) (Tables S1 and S2). Further fractionation of C18 extracts by SPE-COOH (Bakerbond, J.T. Baker, Deventer, Holland) was needed for the full identification and quantification of all PST toxins. This produced three fractions (1 to 3) which needed oxidation with periodate and peroxide oxidants before the LC-FLD analysis.

LC-FLD Analysis

The LC-FLD analyses were conducted on an Agilent UHPLC Infinity 1290 II and 1260 Series spectra fluorescence detector. Chromatographic separations were carried out at 35 °C and 1 mL/min using an RP C18 LC column Supelcosil (150 × 4.6 (i.d.) mm, 5 µm Supelco, Inc., St. Louis, MO, USA) equipped with a Security Guard C18 column (Supelco, Inc.). The detection wavelengths were set at 340 nm for excitation and 395 nm for emission.

Chromatographic elution was selected with mobile phase A (0.1 M ammonium formate to pH 6 ± 0.1) and mobile phase B (0.1 M ammonium formate in 5% acetonitrile to pH 6 ± 0.1) under the following conditions: 0–5% B in the first 0.25 min, from 5–10% B in 4 min, from 10–90% B in 5 min, then return to 100% A before the next injection. The duration of the total run was 14 min.

The quantification limits of this analytical method were: 25 μg STX 2HCl equivalents/kg for STX; 19 μg for dcSTX; 59 μg for NEO; 135 μg for dcNEO; 45 µg for GTX1.4; 23 µg for GTX2.3; 2 µg for GTX5; 5 µg for GTX6; 30 µg for dcGTX2.3; 5.6 µg for C1.2 and 4.5 µg for C3.4 respectively.

2.6.2. LTs

LTs (OA, DTXs, YTXs, AZAs) were identified and quantified by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) in accordance with the EU official protocol described by EC Regulation 15/2011 [31].

Extraction and Extract Processing

The mollusk tissue extracts were prepared by adding 9 mL of 100% MeOH to 2 g of the mollusk homogenates. All samples were homogenized for 3 min via vortex. Afterwards, they were centrifuged (2000× g) for 10 min at 20 °C. A second extraction was performed by adding 9 mL of 100% MeOH to the solid residue and mixing it for 1 min with an Ultra-Turrax. After centrifugation under the same conditions, the two supernatants were combined and brought to a volume of 20 mL with MeOH. Before the injection into LC-MS/MS system, the sample extracts were passed through a 0.2 μm dry methanol-compatible syringe filter. The hydrolysis of lipophilic toxins in mollusks was performed according to the AECOSAN EURLMB Lipophilic Toxins Versions 5, 2015: 125 μL of NaOH; 2.5 M were added to 1 mL of sample extract and homogenized for 30 sec with a vortex. Extracts were heated at 76 °C for 40 min. After cooling, 125 μL of HCl 2.5 M were added and homogenized in a vortex for 30 s. Hydrolyzed extracts were passed through a 0.2 μm dry methanol-compatible syringe filter before being injected into LC-MS/MS system.

LC-MS/MS Analysis

Chromatographic separations were conducted on a Flexar F-15 system (Perkin Elmer, USA) at 40 °C and 0.4 mL/min on an RP column X BRIDGE BEH C18 (50 × 2.1 mm × 2.5 µm, Waters) equipped with a guard column (5 × 2.1 mm × 2.5 µm, Waters). Basic chromatographic elution was selected with mobile phases 100% water (A) and 90% acetonitrile (B), both containing 0.05% ammonium hydroxide (pH 11) under the following conditions: 10% B in the first 0.5 min, from 10–45% B in 3.5 min, from 45–80% in 1 min and hold at 80% for 2 min, from 80–100% B in 1 min, then return to 10% B before the next injection. The duration of the total run was 13.8 min.

MS detection was conducted with a mass spectrometer 4500 QTRAP through an IonDriveTM TurboV electrospray ion source (AB SCIEX, USA) in both negative and positive electrospray ionization (ESI). LTs analysis was performed by multiple reaction monitoring (MRM). Two MRM channels were applied to each toxin to allow quantification, as well as identification, of the specific toxin. Instrument parameters used in this analytical method were reported in Table S3. Additional parameters were: temperature (TEM) 500 °C, ion spray voltage (IS) 4500 (−) V and 5200 (+) V, respectively, in negative and positive ionization modes, curtain gas (CUR) 25, collision gas (CAD) medium, ion source gas 1 (GS1) at 55 Pa and ion source gas 2 (GS2) at 60 Pa, entrance potential (EP) 10 (+) V and 11 (−) V, collision cell exit potential (CXP) 16 (+) V and 7 (−) V, CEM set as 1800 V and dwell-time 60 ms for transition.

The quantification limits of this analytical method were: 60 μg equivalents/kg mollusk’s tissue for the OA toxin group, 45 μg equivalents/kg mollusk’s tissue for the AZA toxin group and 0.20 mg equivalents/kg mollusk’s tissue for the YTX toxin group.

The recovery rate for each toxin was calculated as follows: for toxins belonging to the OA and AZA groups, certified CRM mussels were used, while for the YTX and Homo-YTX, a contaminated extract was prepared to spike an uncontaminated shellfish homogenate with a CRM-solution.

The calculated recovery rates were: 96% for OA, 75% for DTX2, 85% for DTX1, 85% for AZA1, 92% for AZA2, 95% for AZA3, 87% for YTX and 95% for H-YTX.

3. Results

3.1. Harmful Algae

A. minutum (Figure 2) was present in 1.5% of the samples between February and March, with a maximum algal abundance of 1120 cells/L in Torregrande (Oristano) in March. Alexandrium spp. were detected between September and December, with a peak concentration of 760 cells/L. No accumulation of PSTs has been found in this period.

The detected A. minutum were homogeneous with regard to the dimensions, with an average length of 20.3 µm and 20.5 µm width. On the other hand, a certain variety of the D. acuminata complex with very different dimensions, with a length range ranging from 43.05 µm to 48.83 µm and a width range from 19 to µm 29.83 µm was documented. The average length and width were 45.70 µm and 26.26 µm, respectively.

Among the dinoflagellates group, the D. acuminata complex (Figure 3) was present in almost all production areas, with a maximum algal abundance of 17,640 cells/L in May in Corru s’Ittiri (Oristano). The D. acuminata complex was present in 15% of the samples. D. sacculus was detected only one time (in May) in Corru s’Ittiri with a concentration of 600 cells/L. Among the Prorocentraceae, P. lima reached a concentration of about 1000 cells/L in August in Torrevecchia (Oristano). P. lima was present in 3.7% of samples, mainly between May to June. P. mexicanum was present in 2.5% of samples, and it was mostly present in August and September, with a peak of 760 cells/L recorded in September in Feraxi (Cagliari).

3.2. PSP Toxins

Non-compliant PSP samples (>800 µg eq STX 2-HCl/kg edible part (e.p.) were detected only three times (Table S1). Two samples were collected in winter in the Olbia area: C. gigas in Porto Pozzo in January and M. galloprovincialis in Cugnana in December. This late sample showed the highest value of PST recorded in 2021 (

Table 2. Samples that exceeded the limits of PSP.

Sampling Site	Water Sampling Date	Water Temperature T°	A. minutum cell/L	Bivalve Mollusks Sampling Date	Species	µg STX 2-HCl eq/kg e.p.
Porto Pozzo	27 January 2021	13.2	Non detected	27 January 2021	C. gigas	865 ± 243
Cugnana Degomitili	15 December 2021	14.7	760	16 December 2021	M. galloprovincialis	1368
Capo San Marco	9 February 2021	13	680	4 February 2021	M. galloprovincialis	853 ± 280

. Moreover, the presence of PSP over the legal limit occurred in conjunction with the presence of A. minutum.

The third sample (M. galloprovincialis) was found in February in the Oristano area, precisely at Capo San Marco, again with the contemporary presence of A. minutum. In all three episodes, the water temperature ranged between 13 °C and 14 °C.

The chromatographic profile of the sample (Capo San Marco-9 February 2021) without oxidation for naturally fluorescent co-extractives is reported in Figure S1. The PST MIX I (NEO and GTX1.4 (oxidized to periodate)) is reported in Figure S2, while GTX1.4 and GTX2.3 after periodate oxidation and C18 and COOH purification are shown in Figures S3 and S4, respectively. Figure S5 illustrates MIX II (STX, GTX5, GTX2.3; dcSTX, dcGTX2.3 (oxidized to periodate)) while GTX2.3 profiles with peroxide oxidation of C18 and COOH purifications are reported in Figures S6 and S7, respectively.

3.3. Lipophilic Toxins

A total of 18 samples (1.41%) (

Table 3. Presence of algae belonging to the genus Dinophysis and Prorocentrum and samples with the concentration of OAs exceeding the limits of 160 μg OA eq/kg e.p.

Water Sampling Date	Sampling Site	Water Temperature T °C	HAS Cell/L	Cell/L	Bivalve Mollusks Sampling Date	Species	μg eqOA/kg e.p
5 May 2021	Torre Vecchia 1	18	D. acuminata	800	12 May 2021	R.decussatus	321 (OA)
5 May 2021	Corru s’Ittiri 1	20	D. acuminata	17,640	12 May 2021	R.decussatus	393 (OA)
5 May 2021	Corru s’Ittiri 2	20	D. acuminata	11,600	12 May 2021	R.decussatus	403 (OA)
12 May 2021	Terza peschiera	19	D. acuminata	120	12 May 2021	R.decussatus	323 (OA)
19 May 2021	Torre Vecchia	20	D. acuminata; P. lima	280; 960	19 May 2021	R.decussatus	275 ± 103 (OA)
19 May 2021	Terza peschiera 2	22	D. acuminata	680	19 May 2021	R.decussatus	221 ± 83 (OA)
19 May 2021	Corru s’Ittiri 1	22	D. acuminata; D. sacculus	10,520; 600	19 May 2021	R.decussatus	294 ± 110 (OA)
19 May 2021	Corru s’Ittiri 2	22	D. acuminata	11,600	19 May 2021	R.decussatus	248 ± 93 (OA)
					21 January 2021	M.galloprovincialis	400 (OA)
-	Feraxi 1				25 January 2021	M.galloprovincialis	316 (OA)
-	Feraxi 1				9 February 2021	M.galloprovincialis	245 ± 97 (OA)
17 February 2021	Feraxi 1	9.5	P. lima	240	17 February 2021	M.galloprovincialis	171 ± 71 (OA)
-	Tortoli				13 December 2021	M.galloprovincialis	222 ± 83 *
7 April 2021	Su Petrosu	16	D. acuminata	360	7 April 2021	R.decussatus	243 ± 96 (OA)
3 May 2021	Su Petrosu	20.7	D. acuminata	1440	3 May 2021	R.decussatus	193 ± 79 (OA)
Non detected	Su Petrosu				10 May 2021	R.decussatus	166 ± 70 (OA)
17 May 2021	Su Petrosu		D. acuminata	1000	19 May 2021	R.decussatus	293 ± 109 (OA)
-	Seno Cocciani 1				22 September 2021	M.galloprovincialis	197 ± 80 (OA)

* 222 ± 83 μg eqAO/kg e.p, of which 106 μg eqAO/kg e.p DTX2 and 116 μg eqAO/kg e.p OA. -: Samples not obtained.

) showed OA toxin group values over the legal limit of 160 μg OA eq/kg e.p. One positive sample of mussels was found in September (the Olbia area), while four positive samples of Ruditapes decussatus were detected in Orosei (East Sardinia) between March to May. Eight positive samples of Ruditapes decussatus were detected in May in the Oristano area (West Sardinia) (Table 3). The last five positive samples were detected in the Cagliari area (South Sardinia) in the winter months (January and February), in Feraxi (January and February) and in December in Tortolì (South-east Sardinia) (Table 3). The water temperature recorded in correlation with positive samples for OA group toxins ranged between 9.5 °C and 11.6 °C in Feraxi in winter, while in Oristano and Nuoro, it was around 20 °C in May. The toxicity was generally attributed to OA. In 2021, the DTX2 toxin was detected and quantified for the first time in Sardinia. In particular, the DTX2 concentration was 106 μg eq OA/kg.e p. in a mussel sample collected in December 2021 in Tortolì. On this occasion, the sum of OA and DTX2 determined the positivity of the mussel sample (223 ± 83 µg eq OA/kg e.p.) (Figure 4). The HAS results detected by the same positivity for the OA toxin group highlighted the presence of D. acuminata complex on most occasions. The positivity for the OA toxin group was determined by the presence of P. lima (only once) in February. The contemporary presence of D. acuminata complex and P. lima has been found only in one of the eight positive samples from Oristano. Another sample showed the contemporary presence of the D. acuminata complex and D. sacculus. It should be noted that only toxins belonging to the OAs have been detected in the analyzed samples. The other LTs (YTXs and AZAs) have been investigated but never detected.

4. Discussion

The present study reported the presence of PST in bivalve mollusks harvested in Sardinia in 2021.

Altogether, three shellfish samples with values of PST over the legal limits have been found. The positivity for PST in mussels with the presence of A. minutum was highlighted only in two out of the three episodes. No relationship between the toxic algae and PST has been found in the third positive sample, probably due to the undetectable presence of algae-producing PSP in the previous and the following months. Previous authors [23,24,25,26,27,28,29,30,31,32] already described a discrepancy between the cell abundance, the level of toxins and the temporal gap between the presence of Alexandrium spp. and the accumulation of PST in bivalve mollusks. Consequently, we assume that the algae were present before phytoplankton sampling, but due to the timing of sampling, we were unable to find them.

Positive samples were detected in the Olbia and Oristano areas mainly between December and March (winter months). Most PSP-positive samples were mussels. Mussels have shown different rates of toxin uptake accumulation; they can accumulate more PST than oysters and clams and therefore present greater toxicity [33].

Episodes of OA presence in bivalve mollusks harvested in Sardinia are well known.

All the OA-positive samples recovered in the present study were preceded by the presence of D. acuminata, D. sacculus and P. lima. The simultaneous presence of D. acuminata complex and P. lima was extremely rare. The LC-MS/MS analysis allowed us the quantification of OA and DTX 2. According to EU Reg. 627/2019 [34], since January 2021, the LC-MS/MS method has replaced the Mouse test. To the best of our knowledge, this is the first report of DTX2 detected in Sardinia utilizing the LC-MS/MS method.

5. Conclusions

Phytoplankton control carried out in bivalve mollusk production areas in 2021 revealed the presence of potentially toxic algal species throughout the year. The cell density values recorded for the genus Dinophysis and Prorocentrum were found to be relevant and resulted in an accumulation of the OA toxin group exceeding the EU limit of 160 μg OA eq/kg e.p in bivalve mollusks. The species potentially producing PST, belonging to the genus Alexandrium, determined the accumulation of PST beyond the legal limits of 800 µg eq STX 2-HCl/kg edible parts (e.p) in bivalve mollusks harvested in the production areas of Olbia and Oristano during Spring and Winter. The presence of toxic algae was almost always pointed out prior to or simultaneously with the accumulation of toxins in bivalve mollusks.

The percentage of non-compliant samples, for both PSTs and LTs, was quite low. However, this resulted in a consequent closure of shellfish farms with economic losses for the food business operators. Once again, there was a clear relationship between the presence of the OA toxins group and the Dinophysis species. This study showed a certain repeatability of toxin accumulation at certain times of the year in the production areas included in the present study.