INTRODUCTION

Domoic acid (DA) is an amnesic toxin produced by the red algae Chondria armata (Kützing) Okamura, 1907 and diatoms from the genera Pseudo-nitzschia H. Peragallo in H. Peragallo & M. Peragallo, 1900 and Nitzschia A.H. Hassall, 1845 [9, 18]. This compound belongs to the class of excitatory amino acids and is known as an agonist of ionotropic glutamate receptors [18]. Exposure to high concentrations of DA that can accumulate in mollusks and other marine animals and, when transmitted through food chains, causes numerous events of poisoning and mortality of fish, birds, and marine mammals that occur mainly off the coast of the United States and Canada, less often in France, Portugal, Vietnam, and other countries [9, 13, 14, 23]. Diatoms of the genus Pseudo-nitzschia, as potential DA producers, are known as one of the dominant groups of toxic phytoplankton in the Far Eastern seas of Russia [17, 21]. However, data on the DA content in plankton and mollusks from the seas of Russia are scarce. In particular, for the Russian waters of the Sea of Japan and the Sea of Okhotsk, DA was recorded in mollusks from the Peter the Great Bay [16, 21] and the coastal waters of Sakhalin Is.; the presence of this toxin was confirmed in laboratory cultures of Pseudo-nitzschia multiseries (Hasle) Hasle, 1995, P. multistriata (Takano) Takano, 1995, and in P. calliantha Lundholm, Moestrup & Hasle, 2003 from the coastal waters of the city of Vladivostok [16, 21]. Moreover, DA was confirmed in P. calliantha cells from the Russian waters of the Black Sea near the city of Sevastopol [6, 10]. For other water areas of the Far Eastern seas of Russia, especially for the coastal waters of the Kamchatka Peninsula, which is the only region where reliable cases of human deaths caused by toxic phytoplankton “blooms” were noted [1–3], no data on DA concentrations in plankton and mollusks are available.

In this work, we applied two methods for determination of the DA concentrations, that is, enzyme immunoassay and high performance liquid chromatography combined with mass spectrometry, in Pseudo-nitzschia cultures and in samples of bivalve mollusks from the Russian waters of the Sea of Japan and the northwestern part of the Pacific Ocean off the coast of Kamchatka.

MATERIALS AND METHODS

Cultures of Microalgae

We used strains of two species of diatoms of the genus Pseudo-nitzschia: P. delicatissima (Cleve) Heiden, 1928 (MBRU_PD-19) and P. pungens (Grunow ex Cleve) G.R. Hasle, 1993 (MBRU_PP-21), isolated from Avacha Bay (53°0′ N, 158°38′ E) off the Pacific coast of Kamchatka in September 2019 and October 2021, respectively, and the strain P. multistriata (MBRU_PMS-21) isolated from the Amursky Bay of the Sea of Japan near Cape Krasny (43°19′ N, 131°54′ E and 43°12′ N, 131°50′ E) in October 2021 during the bloom event caused by this species. The cultures were maintained at the Resource Collection, Marine Biobank Core Shared Research Facility of the National Scientific Center of Marine Biology, Far Eastern Branch, Russian Academy of Sciences (http://marbank.dvo.ru/index.php/ru/). The strains were grown in 400 mL Erlenmeyer flasks IN f/2 culture medium at 18°C and illumination of 3500 lux in each flask. To analyze the total concentration of domoic acid in the cell suspension (cells plus the medium), the culture was thoroughly mixed and a 50-mL sample was taken on the 19th, 26th, 33rd, and the 40th days of maintenance in the culture for the MBRU_PP-21 and MBRU_PMS -21 strains and on 40th day for the MBRU_PD-19 strain (Table 1). Samples for determination of the total concentration of DA were sonicated in a Branson Sonifier 450 (Branson Ultrasonics, United States) with a power of 100 W using a probe 1 cm in diameter, with cooling on ice for 2–4 min in order to destroy the cells; then, the results of this process were checked under a light microscope. If necessary, the procedure was repeated. The sample was filtered through a Millex-GS Syringe Filter Unit membrane filter (mixed cellulose ethers) with a pore size of 0.22 µm (Merck, Germany). Because of the low values, the total concentration of DA in cultures (pg/mL) was measured only by the method of direct competitive enzyme immunoassay.

Table 1. The concentration of domoic acid (pg/mL) in Pseudo-nitzschia cultures from the Russian waters of the Sea of Japan and the Pacific waters of Kamchatka, as determined by ELISA-cELISA
Full size table

Shellfish Samples

Samples of Mytilus trossulus A. Gould, 1850 were collected in the Avacha Bay in September 2021, while Modiolus kurilensis F.R. Bernard, 1983 and Crenomytilus grayanus (Dunker, 1853) were sampled in Amursky Bay near Cape Krasny in October 2021 (Table 2). A portion of soft tissues of each species (50 g in weight) was homogenized in a blender; 16 mL of a 50% methanol solution were added to a 4 g portion of the homogenate. The material was stirred in a Heidolph Reax top vortex shaker (Heidolph, Germany) for 1 min and then centrifuged at 3000 g for 10 min (NF 800R centrifuge; Nuve, Turkey); the supernatant was collected for analysis. The content of DA in mollusks was determined using the direct competitive enzyme immunoassay (cELISA) and high performance liquid chromatography in combination with mass spectrometry (HPLC-MS).

Table 2.   The concentrations of domoic acid (mg/kg) in bivalve mollusk samples from the Russian waters of the Sea of Japan and the Pacific waters of Kamchatka, as determined by the HPLC-MS and ELISA-cELISA methods
Full size table

ELISA

cELISA was performed using an ASP direct cELISA kit (Biosence Laboratories AS, Norway), which passed a full interlaboratory test on an international scale, which resulted in the approval of the AOAC 2006.02 official method [15], as recommended by Commission Regulation (EC) no. 1244 /2007 [11] for the purpose of screening mollusks for their DA content. Sample preparation and analysis were performed according to the manufacturer’s recommendations [12].

Immediately before analysis, an aliquot of samples of mollusks and microalgal cultures was diluted with a buffer solution (10% methanol in phosphate buffer with Tween 20, pH 7.4). Each sample was analyzed in triplicate, which corresponded to three different dilutions of the sample with the buffer solution. For each dilution (replicate), we determined whether the calibration curve fell within the working range of 10–250 pg/mL. In one case (an MBRU_PMS-21 culture sample on the 40th day of cultivation), three values were obtained corresponding to the working range of the calibration curve; the mean value ± standard deviation was calculated based on these values (Table 1). The optical density of the solutions was measured using an El 800G reader (microplate photometer) (BioTek Instruments, United States) with a 450 nm filter. The limit of quantification was 0.010 mg/kg according to the information posted on the supplier’s website (Stylab Ecosystem).

High Performance Liquid Chromatography in Combination with Mass Spectrometry (HPLC-MS)

Samples were prepared for analysis by solid phase extraction (SPE). For this, a 4-mL aliquot of the mollusk extract was evaporated under reduced pressure using an Eppendorf Concentrator plus a centrifuge (Eppendorf, Germany), redissolved in 2 mL of 50% methanol, and purified with SPE. A 100 mg Bond Elut C18 SPE cartridge, 1 mL (Agilent Technologies, United States) was preconditioned with 3 mL of methanol and 3 mL of water. The extract to be analyzed was slowly passed through the cartridge drop by drop (a flow rate of about 1 mL/min), then, the cartridge with the applied sample was washed with 3 mL of a 0.5% formic acid solution. DA was eluted with 1.7 mL of 50% methanol. For HPLC, we used a Poroshell 120 SB-C18 2.1 × 150 mm column, particle size 1.9 μm (Agilent Technologies, United States). Chromatographic separation was performed in the gradient elution mode with the use of a Bruker Elute UHPLC chromatograph (Bruker Daltonics, Germany), which consisted of an Elute Pump HPG 1300 pump unit, an Elute Autosampler UHPLC autosampler, and an Elute Column Oven column thermostat. A 0.1% aqueous solution of formic acid (phase A) and a 0.1% aqueous solution of formic acid in acetonitrile (phase B) were used as the mobile phase at a flow rate of 0.4 mL/min. The gradient program was: 0–1.5 min, 5% phase B; to 10 min, a gradient up to 40% phase B; to 21 min, a gradient up to 100% phase B; to 21–25 min, 100% phase B; 25.1–28 min, 5% of phase B. The column temperature was 40°C, the injection volume was 10 μm. Our choice of this gradient program was due to the additional task of detecting lipophilic toxins in mollusc tissues: okadaic acid, dinophysistoxin-1, and brevetoxin-3. However, no other known toxins besides domoic acid were found in the studied samples.

Mass-spectrometric detection of DA was carried out on a Bruker impact II QTOF instrument (Bruker Daltonics, Germany) with positive electrospray ionization. The ionization source operation parameters were: a capillary voltage of 4500 V, the drying gas was nitrogen at a volume flow of 6 L/min, the pressure was 2.5 bars, and the temperature was 200°C. Mass spectra were recorded in the scanning mode in the range from 50 to 2000 m/z. The mass spectrometer was tuned using a Agilent ESI-L Low Concentration Tuning Mix mixture (Agilent Technologies, United States). HPLC–MS chromatograms were analyzed using the otofControl (version 4.1) and DataAnalysis (version 4.4) (Bruker Daltonics, Germany) programs. DA was detected based on the chromatographic peak with a retention time of 5.1 min in a chromatogram constructed from the current of the [M + H]+ ion at m/z 312.14 ± 0.02.

To confirm the specificity of the method, we analyzed tissue samples of the Pacific horse mussel M. trossulus collected in Amursky Bay in August 2020, which did not contain DA within the detection range of the method. Quantification of DA in mollusk samples was carried out using the external standard method. Calibration standards were prepared by adding DA to the mussel tissue homogenate. The initial (standard) solution of DA, NRC CRM-DA-f (National Research Council, Canada), with a concentration of 101.8 μg/mL [22], was added to weighed portions of homogenate (4 g) in certain quantities in order to obtain DA concentrations of 0.02, 0.04, 0.4, 2, and 4 mg/kg in tissue homogenate. After addition of DA, the homogenate was stirred for 1 min on a Heidolph Reax top vortex shaker, then 16 mL of a 50% methanol solution was added. Samples were further prepared for HPLC-MS analysis as described above for ELISA.

A calibration curve was constructed on the basis of the area of chromatographic peaks in chromatograms obtained from the flow of the domoic acid ion [M + H]+ at m/z 312.14 ± 0.02 (R2 = 0.9905). The resulted calibration dependence was used to calculate the DA concentration in the sample; then, the DA content in the initial sample was determined using the formula C = (С1 × V × 10−6)/m, where C1 is the concentration of domoic acid in the initial sample (ng/mL), V is the volume of the extract (16 mL), and m is the weighed mass of the mollusc tissue used to obtain the extract (0.004 kg).

RESULTS

The concentrations of domoic acid determined in cultures of diatoms by the cELISA method and in mollusk samples by cELISA and HPLC-MS are given in Tables 1 and 2.

DA in Microalgal Cultures

The total DA concentration in the Pseudo-nitzschia multistriata strain from the Amursky Bay varied from 187 to 1958 pg/mL with the maximum on the 26th day of cultivation; the concentration in the P. pungens strain from the Avacha Bay varied from 2263 to 2663 pg/mL, with the maximum on the 33th day. The total concentration in the P. delicatissima strain from the Avacha Bay was 159 pg/mL on the 40th day of cultivation (Table 1).

DA in Mollusks

Chromatograms for the DA ion current in samples of the Gray mussel and the Kuril horse mussel are shown in Fig. 1. The HPLC-MS analysis showed DA as an asymmetric peak with a retention time of 5.1 min (Fig. 1), while a symmetrical peak was present on the chromatograms of the calibration standards. The presence of asymmetric peaks on the chromatograms was probably caused by the presence of the sum of DA with an admixture of DA isomers.

Fig. 1.
figure 1

Chromatograms for the ion current of domoic acid [M + H]+ at m/z 312.14 ± 0.02 of the mollusk extract of the Gray mussel Crenomytilus grayanus (a) and the Kuril horse mussel Modiolus kurilensis (b), obtained by HPLC-MS.

Full size image

Relatively high concentrations of DA were recorded in mollusk samples of the Kuril mussel Modiolus kurilensis (2.92 mg/kg according to HPLC-MS and 2.8 mg/kg according to ELISA) and in the samples of the Gray mussel (0.07 mg/kg according to HPLC-MS and 0.2 mg/kg according to ELISA) collected in Amursky Bay in late October 2021 (Table 2) during the bloom period of P. multistriata. DA was not found in the Pacific mussel from Kamchatkan waters.

DISCUSSION

The presence of domoic acid at relatively low concentrations in cultures of two diatom species (Pseudo-nitzschia pungens and P. delicatissima) was confirmed by ELISA in Kamchatka material for the first time (Tables 1 and 2). The highest total concentration in the cultures we studied was found in the Pseudo-nitzschia pungens strain (2663 pg/mL) isolated from Kamchatkan waters.

The maximum content of the toxin was recorded in the Modiolus kurilensis from the Russian waters of the Sea of Japan, where DA accumulation was observed in late October 2021 (Table 2) due to a water bloom caused by P. multistriata. Although the DA concentrations in the modiolus (2.8–2.92 mg/kg) appeared to be below the approved allowable limit of 20 mg/kg [7], these values were approximately an order of magnitude higher than the estimates (0.1–0.3 mg/kg) found earlier in bivalve samples from the same area [21].

The relatively high concentrations of DA in mollusks allowed us to apply, along with ELISA, the HPLC-MS method for analysis of this toxin in mollusks from the Russian waters of the Sea of Japan for the first time. The similar DA concentrations for two species of mollusks from the Sea of Japan obtained with the use of ELISA and HPLC-MS methods (Table 2) indicate the possibility of using both methods for determination of DA in mollusks from the area of study.

In contrast to ELISA analysis, HPLC-MS analysis showed the possible presence of isodomoic acids in some cases [19]. Since the main DA peak is chromatographically inseparable from the accompanying peaks and isomeric domoic acids have similar mass spectra, our results present data for the sum of DA and its isomers.

Thus, ELISA, as the more sensitive of the two methods, appeared to be optimal for measuring the relatively low content of DA in cultures of diatoms from the coastal waters of Kamchatka. The HPLC-MS method made it possible to perform measurements in samples of two mollusk species from the Russian waters of the Sea of Japan, where the content of this toxin exceeded 0.05 mg/kg.

The method approved by many international and regional (interstate) standards necessary for the application and fulfillment of the requirements of TR TS 021/2011 for the control of DA is HPLC analysis [5], as regulated by GOST EN 14176-2015. However, when assessing relatively low concentrations of amnesiotoxin (less than 5 ng/mL in microalgal cultures and less than 0.05 mg/kg in mollusks), cELISA is an alternative to HPLC, which cannot identify domoic acid in such samples.