Figure 2 displays the selected ion channel m/z 874.5 → 821.4, 876.5 → 823.4, 892.5 → 839.4, 894.5 → 823.4 chromatograms of standards of PTX-12a/b, PTX-2, PTX-11 and PTX-2 SA, respectively. The retention time (r.t.) of standard materials of PTX-12a, PTX-12b (Figure 2A) and PTX-11 (Figure 2C) were 5.81 min, 7.32 min and 5.94 min, respectively. The retention time of the standards were used in the qualitative analysis of PTXs in seawater. The calibrations linear range of standards of PTX-2 (Figure 2B, r.t. = 6.52 min) and PTX-2 SA (Figure 2D, r.t. = 5.79 min) was from 0 to 40 ng/mL. If the concentration of PTX-2 or PTX-2 SA in the disc eluents was higher than 40 ng/mL it was diluted to a suitable level with 80% methanol before being re-injected into the LC-MSMS system.

Figure 3 displays the selected ion channel m/z 874.5 → 821.4 (A), 876.5 → 823.4 (B), 892.5 → 839.4 (C), 894.5 → 839.4 (D) chromatograms of PTX-12a/b, PTX-2, PTX-12 SA, PTX-2 SA and 7-epi-PTX-2 SA, respectively, in extracts of seawater on August 1, 2006. The levels of PTXs in seawater were investigated from July 28 to August 29, 2006. Figure 4 shows the variation of PTX levels during the evaluation period.

PTX-12a/b and PTX-11 were not found in any of the adsorption resin disc extracts (Figure 3A, 3C). The result showed that PTX-2, PTX-2 SA and 7-epi-PTX-2 SA are present in the Yellow Sea (Figure 3B, 3D). 7-epi-PTX-2 SA is the result of interconversion of PTX-2 SA, and the latter is more thermodynamically stable [13]. Thus, the concentration of 7-epi-PTX-2 SA in seawater was calibrated by comparing with the standard of PTX-2 SA. The total amount of PTX-2 seco acid was expressed by PTX-2 SA plus 7-epi-PTX-2 SA. In July the levels of PTX-2 (88.0 ± 6.8 ng/g of resin, n = 3) and PTX-2 SA plus 7-epi-PTX-2 SA (49.5 ± 7.2 ng/g of resin, n = 3) in seawater were relatively high. On August 1 the highest levels of PTX-2 (107.3 ± 10.0 ng/g of resin, n = 3) and PTX-2 SA plus 7-epi-PTX-2 SA (50.2 ± 5.7 ng/g of resin, n = 3) were found. From 1 August to 9 August the levels of PTXs decreased relatively quickly. From 9 August to the end of August the variation in PTX levels was small. On 29 August, the levels of PTX-2 and PTX-2 plus 7-epi-PTX-2 SA were 40.0 ± 7.9 ng/g, n = 3, of resin and 18.5 ± 1.0 ng/g, n = 3, of resin, respectively; which is about the one-third of the highest levels (Figure 4). During the evaluation period the range of PTX-2 concentration in seawater was from 40 to 107 ng/g of resin. Compared with the concentration of PTX-2 (100–1,500 ng/g of resin) in seawater at a monitoring location at Wedge Point, Queen Charlotte Sound, New Zealand [23], the levels of PTXs in seawater from the Yellow Sea were relatively lower.

Pectenotoxins (PTXs) are a group of lipophilic algal toxins. It is important to study the geographic distribution of marine toxins. PTXs have been found in algae or shellfish in Japan, Ireland, New Zealand, Norway, Italy, Spain, Portugal [4–6,13,15,24]. Along the coast of the Yellow Sea China, they have never been reported. The SPATT technique was first described by MacKenzie et al. as a technique to monitor the occurrence of toxic algal blooms and shellfish contamination events [23]. This technique is an effective method for the investigation of lipophilic marine toxins in seawater [19,20,23]. It is simple, economical and efficient. The monitoring results can show the toxins produced by algae, avoid derivatives occurring in shellfish [21].

The Diaion HP-20 resin has been identified as a suitable resin for effectively accumulating lipophilic toxins from seawater [21]. Furthermore, the HP-20 resin has been used to adsorb a large amount of OA and DTX-2 from seawater in Spain and Norway [19,20]. In agreement with other experiments [21,23,25], the HP-20 resin was selected as the absorbent in the investigation of PTX profiles in the Yellow Sea. The HP-20 is a porous polystyrene adsorbent resin which, by its relatively large pore sizes (260 Å), makes it suitable for recovery of algal toxins that can be eluted with common organic solvents. Experiments showed HP-20 could also accumulate OA and DTX-1 linearly up to 72 h contact time [21].

In our experiment, the SPATT discs which contained HP-20 resin were immerged in seawater for a period of 4 days. In the field trials of toxin monitoring with SPATT techniques, the discs were generally replaced weekly in New Zealand [23]. The immerged time of SPATT discs in seawater might affect the amount of toxins adsorption onto the HP-20 resin, but to date there are no agreed guidelines for immerged time of the discs in seawater. Though the SPATT technique has showed its advantage as a tool used in the investigation of lipophilic toxins in seawater, it still needs to be developed. One of the problems waiting to be solved is to derive the absolute concentration of lipophilic toxins in seawater. Many studies which used SPATT technique reported the relative concentration of lipophilic toxins in seawater using the unit ng toxin/SPATT bag [23] or μg/g of resin [21].

It was revealed that PTXs originated from D. fortii, D. acuta, D. acuminate in Japan, New Zealand, Norway, Italy, Spain and Portugal [4,6,13,15,24]. Living toxic algae can produce PTXs which can be released into seawater. OA, DTXs and PTXs are always found together in New Zealand Dinophysis spp. [6]. Zhou et al. [26] reported the distribution of DSP along the Chinese coast, but few studies on PTXs have been reported in China. In the present study PTX-2, PTX-2 SA and 7-epi-PTX-2 SA were detected in all extracts of SPATT discs from 28 July to 29 August, 2006 (Figure 4). Toxic algae were not investigated in parallel with this study, but, from May to July 2006, another Chinese research group revealed the existence of D. acuminata in this bay (H. Xianqin, Yellow Sea Fishery Research Institute, personal communication). D. acuminata may be the source of PTXs. Up to now PTX-12a/b has only reported from D. acuta, D. Norvegica and D. rotundata from the Scandinavian Sea [13]. PTX-11 is found mainly in D. acuta from New Zealand [27]. PTX-12a/b and PTX-11and their original algal sources were not found in coastal waters of the Yellow Sea. Further investigation into the algal source and the seasonal variation of PTXs, in this bay is recommended.

HP-20 (Diaion HP-20, 260Å, Mitsubishi Chemical Corporation) was selected as the absorbent resin. Methanol, acetonitrile were of HPLC grade, formic acid and ammonium formate were of analytical-regent grade. Both were obtained from Rathburn Chemical (Walkerburn, UK). Water used was deionized (MilliQ). Standards of PTX-2 [28], PTX-2 SA [27] were obtained from NRC Canada. Standard material PTX-12a/b [29] and PTX-11 were purified in NVI. The discs which were produced in NVI (8 cm in diameter) contained the absorbent resin (3.0 g HP-20) between two layers of plankton net (40 μm in mesh diameter). Totally 27 discs were prepared. Before being immerged in seawater, the discs were activated by placing the frame into a beaker filled with methanol for 15 min, then transferred into a beaker filled with water. The water was renewed three times at 5 minute intervals [19].

SPATT techniques were used to adsorb toxins from seawater [19–23]. Each time, three discs were attached to a scallop basket that was 6 m beneath sea level at the pointed sampling location (N36°12.428′, E120°17.826′). The discs were replaced by new ones at four day intervals. Extraction of the toxins from the resin was performed following the method from Fux et al. [21] with modifications by Rundberget et al. [19]. The discs were rinsed with distilled water and kept in the already marked plastic bags or stored at 4 °C until further preparation. The resin was quantitatively transferred into an empty column cartridge fitted with a nylon frit (Varian) and washed free of salts with 30–50 mL of deionized water. Excess water was drawn out of the column prior to application of 10 mL of methanol. After addition of methanol, the column material was stirred and left static for 15 min before the column was eluted slowly into a rotary flask. This was repeated with another 10 mL of methanol. After elution, an additional 3 mL of methanol was pushed through the column to wash out the remaining dead volume of solvent. The methanol fraction was evaporated to dryness, and redissolved in 1.0 mL 80% methanol. All samples were diluted to appropriate concentrations with 80% MeOH. The 80% methanol extraction was filtered through a 0.22 μm Costar Spin-X^® centrifuge filter and transferred into auto-sampler vials. A 10 μL aliquot of the extract was analyzed on the LC-MS-MS system.

For the determination of PTXs in the extracts of seawater, a HPLC-MS-MS method following Quilliam was adopted [28,30]. LC-MS-MS spectrometry was performed using a Waters Alliance 2695 LC-system coupled to a Micromass QuattroUltima triple quadrupole mass spectrometer. Data collection was performed using MassLynx v4.0 software. The MS system was equipped with an electrospray ion source (ESI) operating in positive ionization mode. Injection aliquots were separated on a Waters XTerra C18 column (50 × 2.1 mm, 3.5 μm) connected to a C₁₈ pre-column (10 × 2.1 mm, 3.5 μm). The mobile phases consisted of 2 mM ammonium formate and 50 mM formic acid in 95% acetonitrile (mobile phase A) and 2 mM ammonium formate and 50 mM formic acid in water (mobile phase B). A gradient was run at 0.4 mL/min from 35% A to 100% A for 11 min, followed by an isocratic period for 5.5 min at 100% A. The gradient was returned to 35% A over 0.5 min and left to equilibrate at 35% A for 3 min before the next injection [28,30]. The ion source was operated at a temperature of 100 °C, using a capillary voltage of 3.5 kV, a cone voltage of 80 V and a cone gas flow of 98 L/hr. The desolvation temperature was set at 250 °C and desolvation gas flow at approximately 510 L/hr. Chromatograms were recorded for 13 min (from 2 min to 15 min). MS recording parameters were as follows: for determination of PTX-2, PTX-2 SA, PTX-11, PTX-12, MRM-chromatograms were recorded for the transitions m/z = 876.5 → 823.5 (PTX-2), m/z = 894.5 → 823.5 (PTX-2 SA and 7-epi-PTX-2 SA), m/z = 874.5 → 821.5 (PTX-12a/b), m/z = 892.5 → 839.5 (PTX-11). The concentration of PTX-2 SA and 7-epi-PTX-2 SA were calculated by making a calibration curve for PTX-2 SA standard and comparing peak areas for PTX-2 SA and 7-epi-PTX-2 SA with that for PTX-2 SA. PTX-2 was calculated by making a calibration curve for PTX-2 standard and comparing its peak area with that for PTX-2. PTX-12a, PTX-12b and PTX-11 were determined by comparing the retention time with that for PTX-12a, PTX-12b and PTX-11 reference material.