Microscopic planktonic algae are critical food for filter-feeding shellfish including oysters, mussels, scallops and clams. In most cases, intensive proliferation of phytoplankton cells (so-called algal blooms; up to millions of cells per liter) is beneficial for aquaculture and fishery harvesting. However, in some situations, algal blooms can cause severe damage to aquaculture and can have an adverse impact on human health. Among the 5000 species of extant marine phytoplankton according to Sournia et al.
], there are about 80 species that have the capacity to produce potent toxins [2
], which can, through the food web, have a negative impact on human health and cause a variety of gastrointestinal and neurological illnesses. There are several types of toxicity, which are divided by the symptoms they cause in sea mammals and humans. In Croatian waters, only toxins associated with Diarrheic Shellfish Poisoning (DSP) have been identified to date in concentrations that can impact on human health [3
The first record was during the summer of 1989 on the north-west coast of the Adriatic Sea when the presence of dinoflagellates genera Dinophysis
resulted in shellfish intoxication with Diarrheic Shellfish Poisoning (DSP) [5
]. Subsequent DSP episodes in Croatian waters have been reported. The occurrence of DSP toxicity in the middle Adriatic was first registered in the summer of 1993 in Kaštela Bay [6
] and has been repeatedly observed [7
]. The National monitoring program of shellfish breeding areas has revealed that most of the DSP events have occurred in the northern Adriatic [3
The DSP toxins are all heat-stable polyether and lipophilic compounds that have been isolated from various species of shellfish and dinoflagellates [10
]. Originally they were comprised of three groups of polyether toxins because they often co-occur and their toxins are coextracted in the same lipophilic fraction from plankton cells and shellfish. The first group, comprising acidic toxins, includes okadaic acid (OA) and its derivatives named dinophysistoxins (DTXs). The second group, comprising neutral toxins, consists of polyether-lactones of the pectenotoxin group (PTXs), while the third group includes a sulfated compound called yessotoxin (YTX), a brevetoxin-type polyether, and its derivative 45-hydroxyyessotoxin (45-OH-YTX) [10
]. Nowadays, it is known that these three groups of toxins have different biological effects. YTXs are nondiarrheagenic, and compared to OA show a much lower potency for the inhibition of protein phosphatase 2A. For this reason, it has been proposed that YTXs should not be included in the list of DSP toxins.
DSP toxin profiles in Croatian waters have shown that OA was only occasionally the main toxin causing DSP toxicity events [3
]. In most cases where DSP toxicity in shellfish was detected by mouse bioassay, OA was not present in sufficient concentrations to account for the recorded toxicity [6
]. These findings imply that an unidentified DSP compound might have contributed to the toxic effect. Since YTX occurrence in shellfish from the middle Adriatic has been reported [12
] as well as the presence of the dinoflagelate Lingulodinium polyedrum
that is known to produce yessotoxin, we hypothesized that YTX could be the toxin responsible for most of the DSP toxicity events in Croatian waters. With the aim of investigating whether YTX is responsible for most of the DSP events in Croatian waters we combined a modified Yasumoto’s method, which allows us to extract YTX among other DSP toxins [13
] with LC-MS/MS analysis of these lipophilic toxins.
2. Results and Discussion
Among 453 mussels and seawater samples analyzed in 2007, 10 samples were DSP positive (Tables 1
). Investigations have suggested the presence of DSP toxins other than OA in shellfish from Croatian waters [3
]. Results obtained in the period when the method that discriminated YTX from others DSP toxins [14
] was used revealed that most of the samples were positive for YTX, except samples from Lim Bay (LB 1) (Table 2
). The ELISA method identified the presence of YTXs in mussels (Tables 1
). The DSP toxin profiles showed the presence of OA in three samples and YTXs in four samples (Table 3
), out of the nine samples that were analyzed by LC-MS/MS. In two of the samples that tested positive for YTX using the modified Yasumoto’s method, this toxin was not found and could be due to the presence of YTX analogs, including metabolites in the shellfish, which were not analyzed for using the LC-MS/MS method.
At the station in Medulin Bay (MB), modified Yasumoto’s method revealed the presence of YTX and this was confirmed by a LC-MS/MS toxin profile that showed a high concentration of YTX analogs. Hydroxylated and carboxylated YTX derivatives are largely a result of the metabolism of YTX in shellfish after ingestion [15
]. The most abundant of the YTXs in shellfish from station MB was carboxy-homo YTX, contributing to 64% of the total identified YTXs. According to Ciminiello et al.
], the profiles of YTXs in Mytilus galloprovincialis
collected at different times in the Adriatic Sea revealed YTX and homo YTX as the most abundant. High abundances of carboxy YTX have been reported in Mytilus edulis
collected in Flødevigen Bay (Norway) when the total levels of YTXs were decreasing and were close to the lowest levels for that year, while YTX and 45-hydroxy YTX were predominant during and shortly after a bloom of Protoceratium reticulatum
]. These findings imply that after absorption from the alga, YTX was rapidly oxidized to 45-hydroxy YTX and more slowly to carboxy YTX. In the remaining three samples from stations SB3, IP3 and IP2, YTXs were determined in much lower concentrations and mostly in the form of YTX and homo YTX.
Results obtained with these two methods confirmed that YTX could be the major toxin in shellfish from Croatian waters and the cause of positive DSP mouse bioassays. The response of the modified Yasumoto’s MBA method was higher for YTXs than the measured values of known YTX analogs by LC-MS and could be a result of the presence of other YTXs analogs. Since its discovery, close to 90 other analogs have been identified [17
]. The response from the ELISA method was more sensitive than from LC-MS/MS. Indeed the greater sensitivity of the ELISA method and its greater response to YTXs in both mussels and phytoplankton has already been reported [15
]. Those authors attributed this higher sensitivity to antibodies in ELISA to other YTX analogs whereas the LC-MS method only quantifies those YTX congeners that have standards associated with them.
The biological origins of YTX are from the dinoflagellates Protoceratium reticulatum
(Claparhde and Lachmann) Buetschli, Lingulodinium polyedrum (Stein) Dodge, and Gonyaulax spinifera (Claparede and Lachmann) Diesing. The phytoplankton community structure pattern revealed L. polyedrum
as a potential YTX-producing species. At station MB1, where the MBA test revealed the presence of YTX, and this was confirmed by LC-MS/MS analyses, L polyedrum
was present in the phytoplankton community (Figure 1
). YTX presence in mussel was recorded in non bloom conditions, although the continuous presence of L. polyedrum
in abundances from 80 to 1440 cells L−1
four months prior to toxicity events was recorded. It is possible that L. polyedrum
cells accumulated in mussels until they manifested as a toxicity event, and may be ascribed to carboxy YTX as the major analog in the mussel, due to metabolism after ingestion. In support of this are findings of Aasen et al
], who noted that after absorption from the algae YTX was rapidly oxidized to 45-hydroxy YTX and more slowly to carboxy YTX by blue mussel. The depuration rate for carboxy YTX was considerably slower than for YTX or 45-hydroxy YTX [21
]. From seawater samples collected at four other stations where YTXs were recorded in mussels, L. polyedrum
was also present in the phytoplankton assemblages. At station IP3 the abundance of L. polyedrum
in May was up to 12,100 cells m−2
. At station MSB4, L. polyedrum
abundance ranged from 1,100 to 2200 cells m−2
in the period from the end of the August to the first half of September when MBA for YTX was positive. In addition, at this station in September, G. spinifera
had been recorded in abundance of 1100 cells m−2
. The YTXs profile in the producing organisms revealed that G. spinifera
cultures produce unspecified YTXs identified by ELISA analysis [23
], which could explain why YTX at this station was not determined by LC-MS/MS while MBA [13
] and ELISA detected it. At station IP2, the first occurrence of L. polyedrum
was in May with an abundance of 30,000 cells m−2
and again its appearance was recorded during shellfish toxicity at an abundance of 4400 cells m−2
. L. polyedrum
is a bloom forming species, which can produce intensive blooms with more than a million cells per liter, although for the cases reported here, relatively low concentrations of this species were noted during the entire investigation period. Remarkable amounts of YTXs associated with low concentrations of YTX producers in the Adriatic Sea [24
] have been previously reported and attributed to the toxins’ accumulation.
At stations where OA was determined as the major toxin in shellfish, Dinophysis species were abundant in the phytoplankton community. In Mali Ston Bay at station MSB5, a potential toxin producer was D. caudata, which was present in the community from June to December, in an abundance ranging from 3300 to 75,480 cells m−2. In Lim Bay at station LB1, shellfish toxicity caused by OA presence occurred after an intensive bloom of D. fortii with an abundance of 31,080 cells L−1.
Knowledge of the toxicity of the various YTXs to humans is very limited and, although YTX is toxic by intraperitoneal exposure in the mouse bioassay (0.1–0.8 mg/kg), it appears to be of low toxicity when administered orally [25
]. Comparison of results obtained by three different methods (MBA, ELISA, LC-MS/MS) showed good agreement, although the ELISA method appears to be the most sensitive. Due to the high sensitivity of ELISA and the relatively low toxicity of YTXs for humans when taken orally, further research is needed to determine regulatory levels when ELISA is employed, with the goal of protecting public health while minimizing the risk of unnecessary closures of shell fisheries.