Proposed Biotransformation Pathways for New Metabolites of Paralytic

Article pubs.acs.org/JAFC

Proposed Biotransformation Pathways for New Metabolites of Paralytic Shellfish Toxins Based on Field and Experimental Mussel Samples Ling Ding,† Jiangbing Qiu,† and Aifeng Li*,†,‡ †

College of Environmental Science and Engineering, Ocean University of China, Qingdao, Shandong 266100, People’s Republic of China ‡ Key Laboratory of Marine Environment and Ecology, Ocean University of China, Ministry of Education, Qingdao, Shandong 266100, People’s Republic of China ABSTRACT: A seafood poisoning event occurred in Qinhuangdao, China, in April 2016. Subsequently, the causative mussels (Mytilus galloprovincialis) were harvested and analyzed to reveal a high concentration [∼10 758 μg of saxitoxin (STX) equiv kg−1] of paralytic shellfish toxins (PSTs), including gonyautoxin (GTX)1/4 and GTX2/3, as well as new metabolites 11-hydroxy-STX (M2), 11,11-dihydroxy-STX (M4), open-ring 11,11-dihydroxy-STX (M6), 11-hydroxy-neosaxitoxin (NEO) (M8), and 11,11dihydroxy-NEO (M10). To understand the origin and biotransformation pathways of these new metabolites, uncontaminated mussels (M. galloprovincialis) were fed with either of two Alexandrium tamarense strains (ATHK and TIO108) under laboratory conditions. Similar PST metabolites were also detected in mussels from both feeding experiments. Results supposed that 11hydroxy-C2 toxin (M1) and 11,11-dihydroxy-C2 (M3) are transformed from C2, while 11-hydroxy-C4 toxin (M7) and 11,11dihydroxy-C4 (M9) are converted from C4. In addition, the metabolites M2, M4, and M6 appear to be products of GTX2/3, and the metabolites M8 and M10 are likely derived from GTX1/4. KEYWORDS: paralytic shellfish toxins (PSTs), metabolites, biotransformation, Mytilus galloprovincialis, Alexandrium tamarense

■

INTRODUCTION Paralytic shellfish toxins (PSTs) threaten seafood safety and human health as a result of their acute neurotoxicity. There is a high chemical diversity of PSTs based on the basic structure of saxitoxin (STX). The structures of common analogues and new metabolites of PSTs1 are shown in Figure 1. Phytoplankton species of the Alexandrium, Gymnodinium, and Pyrodinium genera have been reported to produce PSTs in marine ecosystems.2 Additionally, resting cysts of these dinoflagellates and some strains of freshwater cyanobacteria have also been demonstrated to synthesize these toxins.3,4 The accumulation, transformation, and elimination of PSTs in bivalve mollusks have been evaluated in many previous studies. For example, oysters,5,6 mussels,7−10 scallops,11,12 and clams13−16 can accumulate PSTs through filter feeding on toxic dinoflagellates, such as Alexandrium tamarense, Alexandrium catenella, Alexandrium minutum, and Gymnodinium catenatum. PSTs can also be accumulated by crustaceans, such as paddle crabs (Ovalipes catharus) and lobsters (Panulirus stimpsoni), via trophic transfer.17,18 As a result, human poisoning events occur frequently as a result of consumption of seafood contaminated by these algal toxins.19,20 Even benthic mollusks have been contaminated by PSTs in the arctic and subarctic Chukchi and Bering seas.21 Therefore, contamination by PSTs is still a clear risk to seafood safety and human health worldwide. Bivalve mollusks exposed frequently to toxic dinoflagellates can develop mechanisms allowing them to exploit these microalgae as a food resource yet avoid serious harm.7 Additionally, the accumulation and elimination rates of PSTs are species-specific. Toxin accumulation and elimination in © 2017 American Chemical Society

mussels are generally faster (on the order of weeks) than those in scallops, whereas removal of toxins by clams is comparatively slow as a result of strong binding of the siphon tissue with highly toxic STX.11,16,22 PST profiles produced by microalgae are usually modified in bivalves as a result of biometabolism. Metabolic transformation pathways, including reduction, hydrolysis, and enzyme-catalyzed reactions, have been reported in previous studies.16,23−32 Some natural reductants in shellfish, such as glutathione and cysteine, can mediate reductive reactions, including elimination of the hydroxyl group at the N-1 site and the O-sulfate group at the C-11 position. Additionally, some enzymes can catalyze the transformation of gonyautoxins-1, -2, -3, and -4 (GTX1−GTX4) and C toxins (C1/2). The conversion products of PSTs in shellfish identified in these studies can also be detected in toxic microalgae, although there are many different transformation pathways for these common PST analogues. Notably, toxicity of biotransformation products can increase or decrease relative to their precursor compounds. Several exclusively shellfish metabolites of PSTs, comprising several new variants of STX (M1−M4), were discovered in mussels contaminated by A. tamarense in 2008.1 These hydroxyl or dihydroxyl derivatives of STX (M2 and M4) and C2 (M1 and M3) at the C-11 site were found only in shellfish but not in microalgae. These compounds were proposed to be metabolic Received: Revised: Accepted: Published: 5494

May 5, 2017 June 14, 2017 June 15, 2017 June 15, 2017 DOI: 10.1021/acs.jafc.7b02101 J. Agric. Food Chem. 2017, 65, 5494−5502

Article

Journal of Agricultural and Food Chemistry

and oceanic bivalves from the Portuguese coast, including mussels, clams, and cockles.33 High levels of these metabolites were also detected in toxic scallops (Patinopecten yessoensis) and clams (Saxidomus purpuratus) collected from the northern coast of China.34 Such findings indicate that these new PST metabolites are widespread and related to the detoxification process in bivalves. It is essential to explore the origin and biotransformation pathways of these metabolites to elucidate the detoxification mechanism of PSTs in mollusks. Moreover, this work will supplement investigations of algal toxin metabolomics in shellfish. In the present study, field mussel samples were analyzed for PSTs and the new metabolites described above following a suspected paralytic shellfish poisoning (PSP) event in China during April 2016. Feeding experiments using uncontaminated mussels were also carried out using two toxigenic A. tamarense strains (ATHK and TIO108), to better understand the origin and biotransformation pathways of these new PST metabolites.

■

MATERIALS AND METHODS

Chemicals. Certified reference material (CRM) calibration solutions for C1/2, GTX1/4, GTX2/3, dcGTX2/3, GTX5, neosaxitoxin (NEO), STX, and dcSTX were purchased from the National Research Council Canada (Halifax, Nova Scotia, Canada, www.nrc.gc. ca/crm). Formic acid, acetic acid (CH3COOH), and ammonium formate were acquired from Fisher Scientific (Fair Lawn, NJ, U.S.A.). Acetonitrile, methanol, and trichloromethane were procured from Merck, Ltd. (Whitehouse Station, NJ, U.S.A.). Deionized water (18 MΩ cm quality or better) was obtained from a Milli-Q water purification system (Millipore, Ltd., Bedford, MA, U.S.A.). All reagents

Figure 1. Structures of common analogues and new metabolites of PSTs.

intermediates or products of detoxification as a result of their lower toxicity as determined from structure−activity relationships. Subsequently, M1 was also identified in several estuarine

Table 1. Acquisition Parameters of SRM Positive Mode Used for PSTs and Their Metabolites toxin

precursor ion (m/z)

STX

300.1 [M + H]+

NEO and M2

316.1 [M + H]+

C3/4

412.1 [M + H − SO3]+

GTX1/4, M3, and M7

412.1 [M + H]+

C1/2

396.1 [M + H − SO3]+

GTX6 (B2), GTX2/3, M1, and M5

396.1 [M + H]+

GTX5 (B1)

380.1 [M + H]+

dcSTX

257.1 [M + H]+

dcNEO

273.1 [M + H]+

dcGTX1/4

369.1 [M + H]+

dcGTX2/3

353.1 [M + H]+

M4 and M8 M6

332.1 [M + H]+ 316.1 [M + H]+

M9

428.1 [M + H]+

M10

348.1 [M + H]+ 5495

product ion (m/z)

fragmentor (V)

collision energy (V)

282.1 204.1 298.1 220.1 332.1 314.1 332.1 314.1 316.1 298.1 316.1 298.1 300.1 282.1 239.1 180.1 197.1 162.1 289.1 271.1 273.1 255.1 314.1 257.1 178.1 348.1 330.1 330.1

130 130 120 120 90 90 90 90 90 90 90 90 100 100 130 130 100 100 90 90 90 90 100 110 110 100 100 100

20 25 20 25 10 20 10 20 10 20 10 20 10 20 15 25 15 25 10 20 10 20 20 20 20 20 20 20 DOI: 10.1021/acs.jafc.7b02101 J. Agric. Food Chem. 2017, 65, 5494−5502

Article


Figure 2. HILIC−MS/MS chromatograms for (A) PSTs in a mixed standard solution and (B) whole soft tissues of naturally contaminated mussels (M. galloprovincialis) from coastal Qinhuangdao, China. Collection of Field-Contaminated Mussels. According to a

and solvents were analytical- or high-performance liquid chromatography (HPLC)-grade.

report by the Chinese news media, nine consumers were poisoned and 5496

DOI: 10.1021/acs.jafc.7b02101 J. Agric. Food Chem. 2017, 65, 5494−5502

Article

Journal of Agricultural and Food Chemistry hospitalized after eating mussels (Mytilus galloprovincialis) originating in Qinhuangdao, China, at the end of April 2016. Cultured mussels (M. galloprovincialis) were collected immediately from an aquaculture area near Qinhuangdao, China. A homogenate of whole soft mussel tissues was sent to the phycotoxin laboratory in Qingdao, China, and stored at −20 °C until extraction and analysis. Batch Culture of A. tamarense. Two PST-producing A. tamarense strains (ATHK and TIO108) were cultured in filtered (0.45 μm, Jinjing, Ltd., China) and sterilized (at 121 °C for 20 min) seawater before enrichment with f/2-Si medium amendments.35 The temperature and irradiance levels were set at 16 °C and 108 μmol m−2 s−1, respectively, with a 12 h light/12 h dark cycle. Cell densities were determined by microscope counts, and cells were collected at the stationary growth phase for mussel-feeding studies. Feeding Experiments with Uncontaminated Mussels. Uncontaminated mussels (M. galloprovincialis, shell length of 5−6 cm) were obtained from a local marine aquaculture zone near Qingdao, China. The outside of these mussels was cleaned immediately upon receipt at the laboratory, and the animals were acclimated for 3 days in filtered seawater with continuous aeration at 16 ± 2 °C in a tank (60 L). Seawater was refreshed every day, and no food was added during this acclimation period. Six individuals were removed randomly to evaluate the background toxin profiles before the mussels were fed with toxic microalgae. A total of 40 individuals were maintained in the culture tank for exposure to the PST-producing algal strains. The initial cell density of strain ATHK was set at 2000 cells mL−1 in the feeding experiment. The water temperature was controlled at 16 ± 2 °C, and the feeding tank was aerated continuously. The microalgae were supplied continuously using a peristaltic pump to provide a consistent cell density of 2000 cells mL−1. These mussels were fed with strain ATHK for 33 days to maximize accumulation of PST. Six individuals were taken randomly from the tank and then dissected into digestive glands and residual soft tissues. The other individuals were collected and used to extract and purify toxins for use in another study. A separate batch of uncontaminated mussels was fed with another A. tamarense strain (TIO108) following the procedure described above. However, the feeding period with this strain was limited to only 7 days to explore initial metabolic conversions over the short term. Toxin Extraction. A portion (1 g) of the homogenized tissues of the field mussel sample was combined with 2 mL of 0.1 M acetic acid, blended using a vortex mixer, and then centrifuged at 8000 × g for 5 min. The residue was re-extracted twice with 1 mL of 0.1 M acetic acid. The combined supernatants were transferred to a volumetric flask (5 mL) and brought up to volume with 0.1 M acetic acid. Matrix solid-phase dispersion (MSPD) was used to extract PSTs from the fresh mussels fed with toxic microalgae in the laboratory to avoid chemical conversion of toxins as much as possible. The C18 sorbent (Agilent Technologies, Bondesil-C18, 40 mm, 2.0 g) was activated with methanol before use. Digestive glands and residual soft tissues were homogenized using a standard household blender. A 0.5 g subsample was blended thoroughly with C18 sorbent (2 g) in a glass mortar. The mixture was transferred and compacted in an empty solidphase extraction (SPE) cartridge (7 mL). PSTs were eluted with a solution of CH3CN/H2O/HCOOH (80:19.9:0.1, v/v/v), and the eluent was collected in a volumetric flask (10 mL). A portion (5 mL) of the extract was partitioned in a glass tube using 5 mL of CHCl3 and 2 mL of H2O with vortex mixing. The CHCl3 layer was discarded, and then 2 mL of fresh CHCl3 was added to partition again. Finally, the aqueous phase was transferred to a separate flask, and the volume was made up to 5 mL using water. All samples were filtered through a 0.22 μm membrane prior to liquid chromatography−tandem mass spectrometry (LC−MS/MS) analysis. Hydrophilic Interaction Liquid Chromatography−Tandem Mass Spectrometry (HILIC−MS/MS) Analysis of Toxins. PSTs in all samples were analyzed by HILIC−MS/MS using Agilent 1290 HPLC coupled to an Agilent 6430 tandem quadrupole mass spectrometer (Palo Alta, CA, U.S.A.) with an electrospray ionization interface. PSTs were separated on a TSKgel Amide-80 HILIC column (250 × 2 mm inner diameter, 5 μm, Tosoh Bioscience, LLC,

Montgomeryville, PA, U.S.A.). A binary mobile phase included “solvent A” and “solvent B”, in which “solvent A” was water containing 2.0 mM ammonium formate and 50 mM formic acid and “solvent B” was 100% acetonitrile. Full parameters of HPLC and MS were described in our previous study.21 Common analogues and several new metabolites of PSTs were monitored in selective reaction monitoring (SRM) mode (Table 1).

■

RESULTS The LC−MS/MS chromatograms for the mixed standards and the field mussel sample collected from Qinhuangdao, China, are shown in Figure 2. Only carbamate toxins GTX1/4 and GTX2/3 were detected as common PST analogues, whereas the new metabolites M2, M4, M6, M8, and M10 reported in previous studies1,34 were also found. Interestingly, no Nsulfocarbamoyl toxins (e.g., C1/2, C3/4, and GTX5/6) nor decarbamoyl derivatives (e.g., dcGTX2/3, dcGTX1/4, dcSTX, and dcNEO) were detected in the field sample. The ATHK strain of A. tamarense produces C1/2, C3/4, GTX1/4, GTX2/3, and GTX5/6, whereas the TIO108 strain of A. tamarense contains only N-sulfocarbamoyl toxins C1/2 and GTX5. The concentrations of various PST analogues in extracts of both microalgal strains are shown in Table 2. A high molar percent of N-sulfocarbamoyl toxins C1−C4 and GTX5/6 (>76%) was present in the ATHK strain, followed by the carbamate toxins GTX1/4 (>23%). The total molar percent of C1/2 was about 85% of the toxin components in the TIO108 strain. HILIC−MS/MS analysis confirmed that there were no significant levels of PSTs in the uncontaminated mussels before the feeding experiments (Figure 3A). A complex PST profile was observed in the mussels fed with strain ATHK for 33 days (Figure 3B). NEO, STX, and metabolites M1, M2, M3, M7, M8, and M9 were detected in the mussels but not in the ATHK cells. As expected, common PST analogues and their metabolites were also detected in the uncontaminated mussels fed with microalgal strain TIO108 (Figure 3C). Except for the toxins originally in the algal cells, only metabolites M1 and M3 appeared in these mussels. Quantitative results of PSTs and metabolites in the field and experimental mussel samples are shown in Figure 4. Concentrations of new metabolites were estimated on the Table 2. PST Composition of the Two A. tamarense Strains Used To Feed Experimental Mussels toxic algae ATHK

TIO108

5497

toxin C1 C2 C3 C4 GTX5 (B1) GTX6 (B2) GTX2 GTX3 GTX1 GTX4 C1 C2 GTX5 (B1)

toxin concentration (nmol/L)

or 5 minmolar percent (%)

3.5 2.7 10 4.0 6.7

6.8 5.2 20 7.6 13

13

24

0.1 0.052 8.9 2.8 6.0 21 4.9

0.26 0.10 18 5.6 19 66 15


Article


Figure 3. continued

5498


Article


Figure 3. HILIC−MS/MS chromatograms for (A) uncontaminated mussels (M. galloprovincialis), (B) mussels (M. galloprovincialis) fed with strain ATHK (A. tamarense), and (C) mussels (M. galloprovincialis) fed with strain TIO108 (A. tamarense).

therefore essential that these new PST metabolites be purified and tested for toxicity, to assess their contribution to total bivalve toxicity and corresponding human health impacts. The toxin profile of the causative mussels was very simple and included no N-sulfocarbamoyl (C1−C4 and GTX5/6) or decarbamoyl (dcGTX1−dcGTX4) toxins. However, in our previous study34 of scallop and clam samples collected from the Bohai Sea, China, more complex toxin profiles of PSTs and metabolites were encountered. Recently, PST composition was investigated in phytoplankton and shellfish samples collected periodically from five aquaculture zones around the Bohai Sea, including the Qinhuangdao, China, location sampled in this study.38 Multiple common PST analogues were detected in 13 of 20 phytoplankton samples, and C1/2 toxins dominated the toxin profile. Interestingly, one phytoplankton sample containing an exceptionally high PST concentration (3.25 nmol L−1) was obtained from Qinhuangdao, China, in June 2014. This sample was dominated by GTX1/4 (80%), with a moderate amount of GTX2/3 (11%) and C1/2 (9%).38 Except for C1/2 toxins, primary toxin components of this phytoplankton sample were identical to the field mussel sample composition determined in the present study. There are likely different PST-producing microalgal taxa in the net-concentrated phytoplankton samples analyzed in the previous study.38 Diverse and highly variable PST analogues were also detected in different shellfish species, with the samples collected from Qinhuangdao, China, showing a high level of PST contamination.38 Notably, several strains of Alexandrium isolated from the Qinhuangdao, China, coast were identified as A. tamarense and found to produce only GTX1/4 and GTX2/3 (Li et al., unpublished data). Findings reported in these studies confirmed a high level of PST molecular diversity for A. tamarense strains originating from the Bohai Sea, China. However, algal toxin profiles from the Qinhuangdao, China, do appear to be dominated by the highly potent carbamate toxins GTX1−GTX4, which is consistent with the predominant

basis of the peak area relative to the response of STX as a result of the lack of reference materials. In the field mussel sample, GTX1/4 predominated the toxin profile and their molar percent was about 61% of the total toxins. In comparison, the molar percent of all metabolites was about 28% (Figure 4A). According to the PST toxicity equivalence factors published by Oshima,36 the toxicity of carbamate toxins (GTX1−GTX4) in the field sample was equivalent to 10 758 μg of STX equiv kg−1. The molar percent of the various metabolites, including M1, M2, M3, M7, M8, and M9, in the mussels fed with strain ATHK was less than 4% of the total detected analogues (Figure 4B), although the exposure time was 33 days. However, the molar percent of M1 and M3 was about 9.4% of the total PST congeners in the mussels fed with strain TIO108 for only 7 days (Figure 4C).

■

DISCUSSION In the past several decades, no PSP incidents related to consumption of bivalve mollusks were reported in Qinhuangdao, China, prior to the event at the end of April 2016. Mussels were collected from a local aquaculture area immediately following this outbreak to identify the toxins in this contaminated seafood. HILIC−MS/MS analysis of these mussels revealed that a high concentration of PSTs was likely responsible for this poisoning event. The total toxicity was calculated on the basis of the concentrations of GTX1/4 and GTX2/3 at about 10 758 μg of STX equiv kg−1 of whole soft tissues. This level is approximately 13 times the regulatory limit in bivalves (800 μg of STX equiv kg−1) adopted by the European Food Safety Authority (EFSA).37 Toxicity was calculated as the average level of whole soft tissues because a homogenate of mussels was shipped to the phycotoxin laboratory in Qingdao, China, for analysis. However, the PST metabolites also present in this material were not included in the toxicity calculation for the contaminated mussel sample as a result of a lack of toxicological data, although their mole percent was about 28% of the total detected analogues. It is 5499


Article


accompanied by trace amounts of M4 and M6 (Figure 4A). These metabolites are likely converted from GTX1/4 and GTX2/3, because only GTX1−GTX4 were detected as common analogues and the A. tamarense isolated from this location also only produced these same carbamate toxins (Li et al., personal communication). To better understand the conversions involved in the formation of these PST metabolites in mussels, uncontaminated mussels were fed with two A. tamarense strains, each showing a different PST profile. A more complex toxin profile was observed in the mussels fed with A. tamarense strain ATHK because the algal cells contained Nsulfocarbamoyl toxins (C1/2, C3/4, and GTX5/6) and carbamate toxins (GTX1/4 and GTX2/3) (Table 2). In addition to these analogues, very small percentages of NEO, STX, and metabolites M1, M2, M3, M7, M8, and M9 appeared in the mussels fed with the ATHK strain after 33 days (Figure 4B). The appearance of carbamate toxins (STX and NEO) was also reported in short-necked clams (Tapes japonica) feeding on A. catenella, which were undetectable in the algal cells.24 The toxin profile of A. catenella is similar to that of A. tamarense strain ATHK used here. This conversion yielding carbamate analogues was hypothesized to result from enzymatic hydrolysis of N-sulfocarbamoyl toxins (GTX5/6) in shellfish.24 In comparison to the toxin profile of the field mussel sample, the metabolites M1, M3, M7, and M9 were different metabolic intermediates or products. We suggest that these metabolites (i.e., M1, M3, M7, and M9) are derived from the Nsulfocarbamoyl toxins (C1/2 and C3/4) originating in the algal cells. On the basis of their chemical structures, M1 and M3 are proposed as metabolites of C1/2, whereas M7 and M9 are possibly metabolic products of C3/4. We infer that M1 is the O-desulfation derivative of C1/2 at the C-11 site and M3 is the oxyhydroxide of M1 at the C-11 position. Moreover, M7 and M9 are the oxyhydroxide derivatives of M1 and M3, respectively, at the N-1 site. Metabolites M2, M4, and M6 may be converted from GTX2/3, and M8 and M10 are potentially derived from GTX1/4 in the field-collected mussels. M2 is the O-desulfation derivative of GTX2/3, and M4 is the oxyhydroxide of M2 at the C-11 site, whereas M6 is the openring conversion of M4 at the C-4 position. Correspondingly, M8 is possibly the O-desulfation conversion of GTX1/4, and M10 is the oxyhydroxide of M8 at the C-11 site. Unfortunately, an Alexandrium strain producing only GTX1−GTX4 is not available in our laboratory to confirm elaboration of the field mussel toxin profile using a feeding experiment. With the aim of examining more specifically the metabolic conversions of N-sulfocarbamoyl toxins, uncontaminated mussels were exposed to a simpler PST profile by feeding with A. tamarense strain TIO108, producing only C1/2 and GTX5 (Table 2). The feeding period was also shortened to 7 days to focus on PST metabolism at the initial exposure stage. As expected, M1 and M3 occurred in the mussels as a result of the metabolism of C1/2 toxins (Figure 4C). Taking into consideration the field mussel sample and both experimental mussel samples, we infer that metabolites M1 and M3 are transformed from the β-epimer toxin C2 and metabolites M7 and M9 may be derived from the β-epimer toxin C4. In the case of M2, M4, and M6, these metabolites likely originate from GTX2/3, whereas the metabolites M8 and M10 may be produced from GTX1/4. The comparative results also suggested that transformation of C1/2 and C3/4 toxins at the C-11 site preceded that of GTX2/3 and GTX1/4, respectively, in mussels. Of course, M1 and M3 are definitely

Figure 4. Concentration of PSTs observed in mussels: (A) naturally contaminated mussels (M. galloprovincialis) from Qinhuangdao, China, (B) mussels (M. galloprovincialis) fed with strain ATHK (A. tamarense), (C) mussels fed with strain TIO108 (A. tamarense). The concentrations of metabolites were estimated from the relative peak area of a STX standard. The molar percentage (%) of each toxin determined by HILIC−MS/MS is given above the bars.

toxins in field mussels associated with the April 2016 PST outbreak. Apart from the common PST analogues (GTX1/4 and GTX2/3), new metabolites M2, M8, and M10 were also detected in field-collected mussels at significant concentrations, 5500


Article


Figure 5. Determined biotransformation pathway of metabolites M1 and M3 from C2 toxin. (5) Murray, S. A.; O’Connor, W. A.; Alvin, A.; Mihali, T. K.; Kalaitzis, J.; Neilan, B. A. Differential accumulation of paralytic shellfish toxins from Alexandrium minutum in the pearl oyster, Pinctada imbricata. Toxicon 2009, 54, 217−223. (6) Xie, W.; Liu, X.; Yang, X.; Zhang, C.; Bian, Z. Accumulation and depuration of paralytic shellfish poisoning toxins in the oyster Ostrea rivularis Gould−Chitosan facilitates the toxin depuration. Food Control 2013, 30, 446−452. (7) Fernández-Reiriz, M. J.; Navarro, J. M.; Contreras, A. M.; Labarta, U. Trophic interactions between the toxic dinoflagellate Alexandrium catenella and Mytilus chilensis: Feeding and digestive behaviour to longterm exposure. Aquat. Toxicol. 2008, 87, 245−251. (8) Kwong, R. W. M.; Wang, W.-X.; Lam, P. K. S.; Yu, P. K. N. The uptake, distribution and elimination of paralytic shellfish toxins in mussels and fish exposed to toxic dinoflagellates. Aquat. Toxicol. 2006, 80, 82−91. (9) Navarro, J. M.; Contreras, A. M.; Chaparro, Ó . R. Short-term feeding response of the mussel Mytilus chilensis exposed to diets containing the toxic dinoflagellate Alexandrium catenella. Rev. Chil. Hist Nat. 2008, 81, 41−49. (10) Navarro, J. M.; Contreras, A. M. An integrative response by Mytilus chilensis to the toxic dinofagellate Alexandrium catenella. Mar. Biol. 2010, 157, 1967−1974. (11) Choi, M. C.; Hsieh, D. P. H.; Lam, P. K. S.; Wang, W. X. Field depuration and biotransformation of paralytic shellfish toxins in scallop Chlamys nobilis and green-lipped mussel Perna viridis. Mar. Biol. 2003, 143, 927−934. (12) Estrada, N.; de Jesús Romero, M.; Campa-Córdova, A.; Luna, A.; Ascencio, F. Effects of the toxic dinoflagellate, Gymnodinium catenatum on hydrolytic and antioxidant enzymes, in tissues of the giant lions-paw scallop Nodipecten subnodosus. Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 2007, 146, 502−510. (13) Li, S.-C.; Wang, W.-X.; Hsieh, D. Feeding and absorption of the toxic dinoflagellate Alexandrium tamarense by two marine bivalves from the South China Sea. Mar. Biol. 2001, 139, 617−624. (14) Li, S. C.; Wang, W. X.; Hsieh, D. P. H. Effects of toxic dinoflagellate Alexandrium tamarense on the energy budgets and growth of two marine bivalves. Mar. Environ. Res. 2002, 53, 145−160. (15) Chou, H. N.; Huang, C. P.; Chen, C. Y. Accumulation and depuration of paralytic shellfish poisoning toxins by laboratory cultured purple clam. Toxicon 2005, 46, 587−590. (16) Samsur, M.; Takatani, T.; Yamaguchi, Y.; Sagara, T.; Noguchi, T.; Arakawa, O. Accumulation and elimination profiles of paralytic shellfish poison in the short-necked clam Tapes japonica fed with the toxic dinoflagellate Gymnodinium catenatum. Shokuhin Eiseigaku Zasshi 2007, 48, 13−18. (17) Jester, R.; Rhodes, L.; Beuzenberg, V. Uptake of paralytic shellfish poisoning and spirolide toxins by paddle crabs (Ovalipes catharus) via a bivalve vector. Harmful Algae 2009, 8, 369−376. (18) Jiang, T. J.; Niu, T.; Xu, Y. X. Transfer and metabolism of paralytic shellfish poisoning from scallop (Chlamys nobilis) to spiny lobster (Panulirus stimpsoni). Toxicon 2006, 48, 988−994.

derived from C2 toxin in mussels (Figure 5), but the biotransformation pathways of other metabolites will require further targeted experimental studies. Additionally, no dcGTX2/3 or dcGTX1/4 toxins were detected in the mussels fed with strain ATHK for 33 days, which demonstrated that conversion of the 11-hydroxysulfate function of N-sulfocarbamoyl toxins (i.e., C2) to metabolites (i.e., M1) may be favored over desulfonation of the N-sulfocarbamoyl group (i.e., dcGTX2/3). Transformation of 11-hydroxysulfate groups to hydroxyl functions is likely catalyzed by sulfotransferases similar to those reported previously in shellfish.29,30 It is anticipated that the findings reported herein will aid in elucidating the PST detoxification mechanism in bivalve mollusks and also contribute to research on the metabolomics of algal toxins in these shellfish.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +86-532-66781935. E-mail: [email protected]. ORCID

Aifeng Li: 0000-0003-4487-1249 Funding

This work was funded by the National Natural Science Foundation of China (41376122). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors sincerely thank Dr. Yubo Liang for helping to collect the field mussel sample, Dr. Haifeng Gu for providing the A. tamarense strain TIO108, Dr. Pearse McCarron for reviewing this manuscript before submission, and Dr. Gregory J. Doucette for polishing this manuscript before resubmission.

■

REFERENCES

(1) Dell’Aversano, C.; Walter, J. A.; Burton, I. W.; Stirling, D. J.; Fattorusso, E.; Quilliam, M. A. Isolation and structure elucidation of new and unusual saxitoxin analogues from mussels. J. Nat. Prod. 2008, 71, 1518−1523. (2) Anderson, D. M.; Alpermann, T. J.; Cembella, A. D.; Collos, Y.; Masseret, E.; Montresor, M. The globally distributed genus Alexandrium: Multifaceted roles in marine ecosystems and impacts on human health. Harmful Algae 2012, 14, 10−35. (3) Persson, A.; Smith, B. C.; Wikfors, G. H.; Quilliam, M. A. Grazing on toxic Alexandrium f undyense resting cysts and vegetative cells by the eastern oyster (Crassostrea virginica). Harmful Algae 2006, 5, 678−684. (4) Wang, D. Z.; Zhang, S. F.; Zhang, Y.; Lin, L. Paralytic shellfish toxin biosynthesis in cyanobacteria and dinoflagellates: A molecular overview. J. Proteomics 2016, 135, 132−140. 5501


Article

Journal of Agricultural and Food Chemistry (19) McLaughlin, J. B.; Fearey, D. A.; Esposito, T. A.; Porter, K. A. Paralytic shellfish poisoning: Southeast Alaska, May−June 2011. Morb. Mortal. Wkly. Rep. 2011, 60, 1554−1556. (20) García, C.; del Carmen Bravo, M.; Lagos, M.; Lagos, N. Paralytic shellfish poisoning: Post-mortem analysis of tissue and body fluid samples from human victims in the Patagonia fjords. Toxicon 2004, 43, 149−158. (21) Li, A.; Chen, H.; Qiu, J.; Lin, H.; Gu, H. Determination of multiple toxins in whelk and clam samples collected from the Chukchi and Bering seas. Toxicon 2016, 109, 84−93. (22) Gacutan, R. Q.; Tabbu, M. Y.; de Castro, T.; Gallego, A. B.; Arafiles, M. B.; Icatlo, F. Detoxification of Pyrodinium-generated paralytic shellfish poisoning toxin in Perna viridis from western Samar, Philippines. In Biology, Epidemiology and Management of Pyrodinium Red Tides; Hallegraeff, G. M., MacLean, J. L., Eds.; International Center for Living Aquatic Resources Management: Manila, Philippines, 1984; pp 80−85. (23) Blanco, J.; Reyero, M. I.; Franco, J. Kinetics of accumulation and transformation of paralytic shellfish toxins in the blue mussel Mytilus galloprovincialis. Toxicon 2003, 42, 777−784. (24) Samsur, M.; Yamaguchi, Y.; Sagara, T.; Takatani, T.; Arakawa, O.; Noguchi, T. Accumulation and depuration profiles of PSP toxins in the short-necked clam Tapes japonica fed with the toxic dinoflagellate Alexandrium catenella. Toxicon 2006, 48, 323−330. (25) Jaime, E.; Gerdts, G.; Luckas, B. In vitro transformation of PSP toxins by different shellfish tissues. Harmful Algae 2007, 6, 308−316. (26) Tian, H.; Gao, C.; Wang, Z.; Sun, P.; Fan, S.; Zhu, M. Comparative study on in vitro transformation of paralytic shellfish poisoning (PSP) toxins in different shellfish tissues. Acta Oceanol. Sin. 2010, 29, 120−126. (27) Artigas, M. L.; Vale, P. J. V.; Gomes, S. S.; Botelho, M. J.; Rodrigues, S. M.; Amorim, A. Profiles of paralytic shellfish poisoning toxins in shellfish from Portugal explained by carbamoylase activity. J. Chromatogr. A 2007, 1160, 99−105. (28) Suzuki, T.; Ichimi, K.; Oshima, Y.; Kamiyama, T. Paralytic shellfish poisoning (PSP) toxin profiles and short-term detoxification kinetics in mussels Mytilus galloprovincialis fed with the toxic dinoflagellate Alexandrium tamarense. Harmful Algae 2003, 2, 201− 206. (29) Lin, H. P.; Cho, Y.; Yashiro, H.; Yamada, T.; Oshima, Y. Purification and characterization of paralytic shellfish toxin transforming enzyme from Mactra chinensis. Toxicon 2004, 44, 657−668. (30) Cho, Y.; Ogawa, N.; Takahashi, M.; Lin, H. P.; Oshima, Y. Purification and characterization of paralytic shellfish toxin-transforming enzyme, sulfocarbamoylase I, from the Japanese bivalve Peronidia venulosa. Biochim. Biophys. Acta, Proteins Proteomics 2008, 1784, 1277−1285. (31) Turner, A. D.; Lewis, A. M.; Hatfield, R. G.; Galloway, A. W.; Higman, W. A. Transformation of paralytic shellfish poisoning toxins in Crassostrea gigas and Pecten maximus reference materials. Toxicon 2012, 60, 1117−1134. (32) Turner, A. D.; Lewis, A. M.; O’Neil, A.; Hatfield, R. G. Transformation of paralytic shellfish poisoning toxins in UK surf clams (Spisula solida) for targeted production of reference materials. Toxicon 2013, 65, 41−58. (33) Vale, P. Metabolites of saxitoxin analogues in bivalves contaminated by Gymnodinium catenatum. Toxicon 2010, 55, 162−165. (34) Li, A.; Ma, J.; Cao, J.; Wang, Q.; Yu, R.; Thomas, K.; Quilliam, M. A. Analysis of paralytic shellfish toxins and their metabolites in shellfish from the North Yellow Sea of China. Food Addit. Contam., Part A 2012, 29, 1455−1464. (35) Guillard, R. R. L.; Ryther, J. H. Studies of marine planktonic diatoms: I. Cyclotellanana Hustdtand, and Detonula confervacea (Cleve) Gran. Can. J. Microbiol. 1962, 8, 229−239. (36) Oshima, Y. Postcolumn derivatization liquid chromatographic method for paralytic shellfish toxins. J. AOAC Int. 1995, 78, 528−532. (37) European Food Safety Authority (EFSA). Marine biotoxins in shellfishSaxitoxin group scientific opinion of the panel on contaminants in the food chain. EFSA J. 2009, 1019, 1−76.

(38) Liu, Y.; Yu, R.-C.; Kong, F.-Z.; Chen, Z.-F.; Dai, L.; Gao, Y.; Zhang, Q.-C.; Wang, Y.-F.; Yan, T.; Zhou, M.-J. Paralytic shellfish toxins in phytoplankton and shellfish samples collected from the Bohai Sea, China. Mar. Pollut. Bull. 2017, 115, 324−331.

5502


Proposed Biotransformation Pathways for New Metabolites of Paralytic

Recommend Documents