Variation in the growth and toxin production of Gymnodinium catenatum under different laboratory conditions

The chain-forming dinoflagellate Gymnodinium catenatum is the only known gymnodinioid dinoflagellate that produces paralytic shellfish toxins (PST). Dense blooms caused by the dinoflagellate have been frequently reported in coastal waters of Fujian, China since 2017. While there is still limited understanding of the major physiological characteristics of G. catenatum isolated from Fujian coastal waters, the growth and toxin production of the G. catenatum strain were examined in batch cultures with different levels of irradiance, temperature, salinity, nitrate, and phosphate conditions. The results indicated that the highest maximum cell density of the strain was achieved at 70 µmol m−2 s−1, with the highest growth rate at 120 µmol m−2 s−1. The strain grew well within the temperature range of 15–30 °C, with maximum growth rate and cell density achieved at 20 °C. The dinoflagellate also showed higher tolerance to salinity variation (20–40), with the highest growth rate at salinity 25. Meanwhile, G. catenatum showed higher demand for nitrogen and phosphorus as indicated by its higher half-saturation constant. A decrease in nitrate and phosphate greatly inhibited the growth of G. catenatum. The toxin profile of the G. catenatum strain was conservative and dominated mainly by the N-sulfcarbamoyl C-toxins (> 95%), indicating its hypotoxicity. The cellular toxicity increased with the algal growth, with the highest cellular toxicity observed at the stationary growth phase. The cellular toxicity of G. catenatum also responded to environmental variations including lower temperature (15 °C), lower salinity (20), nitrate-repletion, and phosphate-depletion conditions which enhanced the cellular toxicity, while irradiance exerted non-significant influence. The present study depicted the physiological characteristics of the particular G. catenatum strain and provided valuable insight on the ecophysiology of G. catenatum in natural coastal waters.


Introduction
Over the last several decades, coastal regions throughout the world have experienced frequent occurrences of harmful algal blooms (HABs). The most common effects of HAB are poisoning events, resulting in mass mortality or illness of marine organisms. The accumulated toxins in seafood may lead to eventual human poisoning (Anderson et al. 2002;Green et al. 2004;Shumway et al. 2018;Usup et al. 1994). Among the algal toxins, paralytic shellfish toxins (PST) are among the most hazardous biotoxins due to their severe toxicity and wide distribution. They are mainly produced by eukaryotic dinoflagellates and prokaryotic cyanobacteria (Hummert et al. 1997;Lefebvre et al. 2008;Usup et al. 2012). Recently, six cases of human poisoning occurred in Qinhuangdao, China, due to the consumption of seafood containing PST during blooms in 2019. Monitoring by the local government indicated PST levels reached 2220 MU Responsible Editor: S.Shumway.
Major environmental factors exert varying degrees of impact on the growth and cellular toxicity of G. catenatum. The algal strains from different geographic regions showed distinct toxin profiles as well. In particular, the optimum temperature for growth of G. catenatum strains from Tasmania was 14.5-20 °C, and 22-25 °C in the strains isolated from coastal waters of southern China (Blackburn et al. 1989;Ye et al. 2018;Zhang 2009). Higher cellular toxicity was detected in the Japanese strain cultured at 18 °C was twice as toxic as that cultured at 25 °C , while there was no such variation in the strain from Mexico (Band-Schmidt et al. 2014). The growth and toxicity characteristics of G. catenatum showed higher intraspecific variability (strain differences). Intraspecific variability in key characteristics such as life-history traits, behavior, nutrition, genetics and toxicity, has been experimentally documented for many toxigenic microalgae, including species of cyanobacteria, dinoflagellates, haptophytes, raphidophytes and diatoms Sinclair et al. 2009;Thessen et al. 2009). Although often overlooked, it is likely of fundamental importance to species survival and evolution. Maintenance of an array of co-existing genotypes within a population might be especially important for commonly haploid dinoflagellates (Tillmann and Hansen 2009). The cause of intraspecific variability is unknown and may arise from environmental variability and epigenetic factors.
In the coastal waters of China, G. catenatum cysts were reported ubiquitously (Gu et al. 2013;Qi et al. 1996), and massive G. catenatum blooms have occurred in the coastal waters of the East China Sea (e.g., Jiangsu and Fujian coasts) (Cheng et al. 2009;Zhang et al. 2020), while its vegetative cells have rarely been observed in the South China Sea and the Yellow Sea (Jiao et al. 2010;Lu and Hodgkiss 2004). In 2017, a large-scale bloom of G. catenatum occurred in Fujian coastal waters and resulted in huge economic losses to local aquaculture ventures and severe PSP events (Chen 2018). The PSTs produced by this strain were dominated by C1/2 and dcGTX2/3 without C3/4 (Liu et al. 2020). The toxin profile was unique compared to those strains from Japan, Australia, South Korea, Mexico, Spain, Portugal, Uruguay, Malaysia, Hong Kong and the Yellow Sea of China (Liu et al. 2020), suggesting it may represent a separate strain. Limited studies have been conducted to explore the physiological responses of this strain to major environmental factors. In the present study, growth, cellular toxicity, and toxin composition of G. catenatum in batch cultures under different environmental conditions were investigated to provide better understanding of the toxin production of the species.

Algal isolation and identification
The dinoflagellate Gymnodinium catenatum (isolate ID# TIO523) was originally generated from cysts collected in Xiamen Bay, Fujian Province, China by Prof. Haifeng Gu's laboratory at the Third Institute of Oceanography, SOA. The monoculture of the strain was further established by pipetting single cells into 24-well polystyrene cell culture plates for laboratory experiments at the Institute of Oceanology, CAS. The stock cultures were maintained in the exponential growth phase by transferring the cells to a fresh medium biweekly and were gently shaken once daily. The cultures were maintained routinely in an f/2 medium supplemented with selenium (10 -9 M H 2 SeO 3 , (Doblin et al. 2000)) at 20 °C, 80 μmol photons m −2 s −1 under a 12-h light/12-h dark cycle.
Genomic DNA was extracted from the vegetative cultures of G. catenatum harvested at the exponential growth phase, using the PowerSoil ™ DNA Isolation Kit (MO BIO Laboratories Inc., USA). Partial D1-D2 of LSU rDNA and ITS regions were amplified using primers D1R and D2C (Scholin et al. 1994) and primers ITS-A and ITS-B (Adachi et al. 1994). The PCR reaction procedure was 5 min at 94 °C, followed by 40 cycles of 1 min at 94 °C, 2 min at 55 °C, 1 min at 72 °C, and a final extension of 7 min at 72 °C. Amplification products were purified using the E.Z.N.A ® Gel Extraction Kit (OMEGA, USA) and sequenced using ABI 3730XL DNA Analyzer (Applied Biosystems, Foster City, CA, USA). The nucleotide sequences of each gene were aligned using the default setting of ClustalW 2.0 (Larkin et al. 2007) and then concatenated for tree construction. Additional sequences of G. catanetum were acquired from the National Center for Biotechnology Information (NCBI) database. Gymnodinium nolleri (GenBank number: FN649409), Gymnodinium microreticulatum (GenBank number: AY036078), Lepidodinium viride (GenBank number: AY464689), Gymnodinium nolleri (GenBank number: AM998535), and Gymnodinium inusitatum (GenBank number: KF234061) were selected as outgroups. A phylogenetic tree was constructed using the Neighbor-Joining (NJ) based on the maximum composite likelihood model (Tamura and Kumar 2004) with 1000 bootstrap replicates, using MEGA 7.0 software (Kumar et al. 2016).

Characterization of algal growth and toxin profile
Triplicate batch cultures (180 mL) were grown in an f/2 + Se medium and maintained under the conditions described above, with an initial cell density of 220 ± 30 cells mL −1 in individual Erlenmeyer flasks (250 mL in volume). Every second day, 1 mL of sub-sample was fixed in Lugol's solution (final concentration 0.5%), loaded on a plankton counting chamber (0.1 mL), and then counted with an inverted microscope (Olympus, IX71, Japan). To assess cellular density, algal cells were collected at the lag phase (Day 2), the exponential phase (Day 8) and the stationary phase (Day 14) post transfer of G. catenatum cells into the batch cultures. The acquired cell densities were used to generate growth curves in which the growth rates (μ, day −1 ) were calculated with the following equation: where C 0 and C t are the cell density at time t 0 and t n , respectively.
In addition, aliquots of 50-100 ml algal cultures were pipetted according to the cell density, filtered onto GF/C membranes (nominal pore size 1.2 μm, Whatman, Kent, U) packaged in tinfoil and stored at − 20 °C until further toxin analysis. (

The effects of major environmental factors on algal growth
Time-course experiments were carried out to determine the effects of major environmental factors on the growth of G. catenatum in batch cultures. The culture conditions varied in irradiance (0, 10, 30, 70, 120, 190 µmol photons m −2 s −1 ), temperature (10, 15, 20, 25, 30 °C), salinity (20, 25, 30, 35, 40), nitrate concentration (0, 25, 50, 100, 200, 800 μmol L −1 ) and phosphate concentration (0, 2, 4, 8, 16, 36 μmol L −1 ). All experiments were conducted in triplicate in culture flasks, with an initial cell density of 300 ± 30 cells mL −1 . To monitor the growth of G. catenatum in the experimental cultures, a 1-ml sub-sample was pipetted from each flask and fixed immediately with Lugol's solution (final concentration 0.5%). The cell densities and growth rates of G. catenatum in the fixed samples were evaluated as described previously. The relationship between growth rate and environmental conditions was fitted to a Monod growth kinetic model using the following equation (Monod 1942): where μ was the calculated growth rate at the exponential growth phase (Eq. 1), μ m was the maximum specific growth rate (day −1 ), S was the irradiance level (µmol m −2 s −1 ) or nutrient concentration (µmol L −1 ), and K μ was the irradiance level or substrate concentration which supports halfmaximum specific growth rate.

The effects of major environmental factors on algal toxin production
Additional time-course experiments were conducted to study further the effects of major environmental factors on toxin production of G. catenatum, based on the preliminary results of algal growth experiments. Cultures were grown under different irradiance (30, 70, 190 µmol photons m −2 s −1 ), temperature (15, 20, 30 °C), salinity (20, 30, 40), nitrate concentration (25, 100, 800 μmol L −1 ) and phosphate concentration (2, 8, 36 μmol L −1 ). All treatments were conducted independently in triplicate cultures, with an initial cell density of 300 ± 30 cells mL −1 . During the stationary phase of experimental cultures, 50 mL of algal culture was harvested by filtration. The filtered membranes (GF/C, nominal pore size 1.2 μm, Whatman, Kent, UK) were frozen immediately and stored (− 20 °C) until further processed for toxin analysis. (

Statistical analysis
One-way analysis of variance (ANOVA) was used to determine the differences in growth rates, cellular toxin concentrations, and toxin composition of the various cultures treated with different levels of irradiance, temperature, salinity, nitrogen, and phosphate concentrations, as well as the toxin concentration and composition at different growth phases. All the data were tested for normality and homogeneity of variance before applying Duncan's post hoc multiple comparison test using the SPSS Statistics software (version 21.0, SPSS, USA) with a significance level of p < 0.05. Growth curves, Monod and Michaelis-Menten curves were fitted with Origin software (OriginPro 9, USA).

Growth and characteristics of G. catenatum
After inoculation into the f/2 + Se medium, G. catenatum entered a lag growth phase until day 4. Then the algae grew exponentially and reached the highest cell density of 12,898 ± 790 cells mL −1 on day 14, with a growth rate of 0.19 ± 0.10 d −1 (Fig. 3A). Compared with the lag phase and stationary phase, a larger proportion of longer chains (8-24 cells) were observed in exponential phase, with the longest chain of 32 cells. Toxin analysis showed that N-sulfocarbamoyl toxins (C1/2) dominated the toxin profile of the algal isolate (> 99%), with trace amounts of GTX3-5, dcneoSTX contributing less than 1%, thus these trace derivatives were referred to collectively as "Others" in the following report of toxin profile (Figs. 2, 3B). The equivalent cellular toxicity of G. catenatum increased along with the growth of experimental cultures, varying from 8.72 ± 0.47 on day 2 to 12.26 ± 0.10 pg STX eq cell −1 on day 14 (p < 0.05). As for toxin profile per cell on a molar basis, the relative contents of C1 toxin increased significantly from 8.96% (day 2) to 16.96% (day 14) (p < 0.05), whereas the percentage of C2 decreased significantly from 91.03 to 82.39% (p < 0.05), the proportion of other trace toxins also increased significantly (p < 0.05) during the sampling period, from 0 (day 2) to 0.65% (day 14) (Fig. 3C).

Effects of environmental factors on growth of G. catenatum
Irradiance significantly affected the growth of G. catenatum. Increments in the cell densities of experimental G.  (Fig. 4A). The maximum growth rate (0.26 ± 0.02 d −1 ) was observed in 120 µmol m −2 s −1 . Additionally, the growth of G. catenatum in experimental cultures exhibited standard Monod-type kinetics within the range of 0-190 µmol m −2 s −1 , with the maximum specific growth rate of 0.28 ± 0.02 d −1 and the half-saturation constant (K μ ) of 24.54 ± 10.32 µmol m −2 s −1 (Fig. 4B).
Temperature also had a significant effect on the growth of G. catenatum under laboratory conditions. Growth occurred in all treatments and showed an exponential increase except at 10 °C, with the maximum cell density (12,898 ± 790 cells mL −1 ) obtained in the 20 °C treatment by day 14 (Fig. 4C). The growth rates of experimental cultures in the temperature treatments varied between 0.05 ± 0.004 and 0.25 ± 0.01 d −1 , with the maximum growth rate obtained in the 20 °C treatment (Fig. 4D). The dinoflagellate exhibited constant growth in the salinity range of 20 ~ 40, and the growth rates varied from 0.13 ± 0.013 to 0.24 ± 0.012 d −1 , with the maximum growth rate in the salinity25 treatment (Fig. 4F). The maximum cell density increased within the salinity range of 20-30 and reached the highest of 5703 ± 232 cells mL −1 , while there was no significant difference (p > 0.05) in the maximum cell densities within the salinity range of 25-35 (Fig. 4E).
The G. catenatum strain was able to grow in different concentrations of nitrate, but showed different growth patterns. Limited growth was observed in the cultures at the initial nitrate concentration of 0 μmol L −1 (cell density < 2000 cells mL −1 ). The algal growth increased with nitrate concentration, with the maximum cell density of 11,638 ± 722 cells mL −1 obtained in 800 μmol L −1 nitrate treatment (Fig. 5A). Growth rates also increased with nitrogen concentrations, within the range of 0.04 ± 0.004-0.26 ± 0.02 d −1 . Additionally, the growth fit standard Monod-type kinetics, with the maximum specific growth rate (μ m ) of 0.29 ± 0.08 d −1 and the half-saturation constant (K μ ) of 53.91 ± 50.49 μmol  (Fig. 5B). Similarly, the algal growth increased with increased phosphate concentrations. Limited growth was observed in the treatment without supplement of phosphate (cell density < 2000 cells mL −1 ), and both the highest maximum cell density (12,898 ± 790 cells mL −1 ) and growth rate (0.28 ± 0.02 d −1 ) were obtained in the 36 μmol L −1 phosphate treatment (Fig. 5C, D). Growth also exhibited standard Monod-type kinetics, with the maximum specific growth rate (μ m ) of 0.38 ± 0.08 d −1 and the half-saturation constant (K μ ) of 9.51 ± 4.36 μmol L −1 (Fig. 5D).

Effects of environmental factors on toxin production of G. catenatum
Irradiance exerted no overt influence on toxin production of G. catenatum in the laboratory treatments. The cellular toxicity varied from 11.43 ± 0.68 to 13.62 ± 1.73 pg STX eq cell −1 in the algal cells cultured under different irradiance levels, but with no significant difference among treatments (p > 0.05). The toxin compositions of the laboratory cultures were all dominated by C2, with no significant difference (p > 0.05) in the proportions of major toxins among treatments (Fig. 6A).

Discussion
The toxic dinoflagellate Gymnodinium catenatum is a cosmopolitan species distributed widely in temperate (12-18 °C) and tropical waters (21-30 °C), and blooms have been associated with outbreaks of paralytic shellfish poisoning (PSP) events in some regions Oh et al. 2010;Ordas et al. 2004;Proena et al. 2001). Nutrients, together with light and temperature, are among the major environmental factors impacting the development of dinoflagellate blooms. In the present study, G. catenatum was very sensitive to variation of environmental factors in laboratory treatments, and the fluctuations in irradiance, temperature, salinity, nitrate, and phosphate concentrations affected not only the growth of G. catenatum, but also its toxin production. The eurythermal and euryhaline species were able to tolerate a wide range of temperature (10-30 °C) and salinity (20-40). Its growth responded positively to irradiance levels, while depleted nutrients limited the growth of G. catenatum. In addition, lower levels of temperature and salinity, as well as limitation of phosphate, exhibited beneficial effects on the toxin production of G. catenatum. Moreover, increases in nitrate concentrations enhanced the cellular toxicity of G. catenatum significantly, suggesting nitrogen might be a key element for its toxin production.

The effects of major environmental factors on growth of G. catenatum
The G. catenatum strain isolated from Fujian, China, grew well in f/2 + Se medium, and the exponential growth phase lasts for 10-15 days, which was similar to the Korean strains (Han et al. 2019). The maximum cell density reached 12,000 cells mL −1 , which was two times higher than those of Hongkong and Guangxi of China (Hu et al. 2012;Ye et al. 2018). The highest growth rate was 0.28 d −1 , which was comparable to the algal strains from Mexico and Tasmania, Australia, but was slightly lower than that of the Japanese strain (0.31 d −1 ) and the strain from Derwent, Australia (0.34 d −1 ) (Band-Schmidt et al. 2014;Doblin et al. 2000;Yamamoto et al. 2010). The differences in the maximum cell densities and growth rates may be related to their geographic origin, as well as the differences of culture conditions. For example, G. catenatum exhibited a much higher growth rate in GSe medium (0.82 d −1 ) attributed to the chelation of trace metals and provision of nutrients Usup et al. 1989). The trace element selenium was also an important factor for growth of G. catenatum, and depressed maximum cell density and growth rate were observed in Fig. 6 Cellular toxicity and toxin composition of cultured G. catenatum in response to different irradiance levels (A), temperatures (B) and salinities (C). Error bars represent standard deviations the dinoflagellate cultured in a Se-deplete medium (Doblin et al. 2000). The strain TIO523 grew well in the temperature range of 10-30 °C, attaining maximum growth rate at 20 °C. Its optimal salinity range (25-35) was similar to the Australian (23-34) and Japanese (20-32) strains (Blackburn et al. 1989;Yamamoto et al. 2010), but was slightly higher than the Korean strains which could tolerate salinity as low as 15 (Han et al. 2019). Such tolerance of salinity changes reflects the natural habitat favorable for the algal bloom. For example, a Venezuelan strain grew better in the salinity range of 33.2 and 39, and the algal blooms were observed in the northeast coastal waters with salinity of 36.5-38 ( Bar-beraSánchez and Gamboamaruez 2001).
The growth of G. catenatum was greatly enhanced in nitrate-sufficient cultures, which was consistent with several other dinoflagellates (Lei and Lu 2011;Liu et al. 2015;Fig. 7 Cellular toxicity and toxin composition of cultured G. catenatum in response to different nitrogen (A) and phosphate concentrations (B). Error bars represent standard deviations Wang and Hsieh 2002). Nitrogen is a component of numerous cellular structures and thought to be the primary limiting macronutrient for algae growth. The maximum cell density and growth rate decreased with nitrate concentrations, probably due to the retarded cell cycle resulting from nitrate stress (Lei and Lu 2011). In the present study, the calculated half-saturation constant (K μ ) of G. catenatum was 53.91 ± 50.49 μmol L −1 , indicating its higher demand for nitrate. The algae evolved physiological functions to adapt nitrate stress, including efficient uptake subsystems (Lei and Lu 2011), and the capability to utilize multiple forms of dissolved nitrogen (unpublished data). As observed in previous studies (Hu et al. 2012;Zhang 2009), phosphate also promoted the growth of G. catenatum. The calculated half-saturation constant (K μ ) of G. catenatum was 9.51 ± 4.36 μmol L −1 , while the phosphate concentration in most marine environments was below 1 μmol L −1 . Thus, to survive the phosphate-deplete conditions, microalgae have evolved mechanisms for utilization of phosphate and counteraction of phosphate stress (Chai et al. 2006;Cheng et al. 2009;Chung et al. 2003;Zhao et al. 2009). Similarly, G. catenatum could also utilize various forms of phosphate substrates (e.g., sodium-β-glycerophosphate, unpublished data). Moreover, algae could store phosphorus intracellularly in phosphorus-replete environments, then utilize it in phosphate-deplete environment (John and Flynn 2000).

Effects of environmental factors on toxic characteristic of G. catenatum
The cellular toxicity and toxin profile of G. catenatum are extraordinarily variable among different populations or strains around the world. The cellular toxicity (7.77 ± 1.31 to 15.26 ± 0.54 pg STX eq cell −1 ) of the present strain was at levels comparable to the strains isolated from Japan, Korea, Brazil and Mexican Pacific (Band-Schmidt et al. 2014;Oh et al. 2010;Park et al. 2010;Proena et al. 2001), but was much lower than the Mexican strain reported in Bahía de Mazatlan (> 101 pg STX eq cell −1 ) ). The observed difference in cellular toxicity among strains might be attributed to the difference in toxin composition, particularly the low-toxic strains that contained extreme higher proportions of N-sulfocarbamoyl C-toxins (Band-Schmidt et al. 2014;Lee et al. 2012;Park et al. 2010). Among the G. catenatum strains isolated in different areas of the East China Sea, C1/C2 toxins were the dominant components (47-70%) in the toxin profiles, with trace or even no C3/4 toxins (Lin et al. 2022;Liu et al. 2020). Similar toxin profiles were also reported in the Singapore and Japanese strains (Holmes et al. 2010;Oh et al. 2010), indicating their closer relationship with those isolates from the East China Sea. Even though the toxin profile was a conservative characteristic within a strain or natural population (Oshima et al. 1993), there was still a noticeable difference in the detailed toxin components of particular dinoflagellates. The present G. catenatum strain (TIO523) produced trace GTX5, while the strain (MEL11) isolated from the same region had a high content of GTX5 (Lin et al. 2022). Such discrepancies may be attributed to the difference in initial isolation methods as the present strain was separated from cysts and the other strain was isolated directly from vegetative cells (Oshima et al. 1992).
Among the environmental factors tested in the present study, temperature appeared to be the prominent factor affecting the variation of toxin production in G. catenatum. The cellular toxicity of G. catenatum was higher in the lower temperature treatments, which was also observed in previous studies (Band-Schmidt et al. 2014;Ye et al. 2018). Similar trends have been found in other PST-producing species, for example, Alexandrium fundyense produced four times higher toxin levels at 5 °C than that at 20 °C (Etheridge and Roesler 2005). Low temperature (< 20 °C) suppressed the cell division process leading to the accumulation of arginine (one precursor of PST) (Wang et al. 2016), thereby accelerating the synthesis of toxins. Yoshida et al. (2002) found the activity of N-sulfotransferase in G. catenatum was affected by temperature. The decreasing percentage of C2, but increasing percentage of C1, in G. catenatum cultured at 30 °C might also result from stimulated enzyme activity (e.g., keto-enol epimerization). The response of toxin synthesis to salinity changes was complicated (Etheridge and Roesler 2005;Flynn et al. 1996;Lim and Ogata 2005). One of the well-accepted explanations is that rapid growth resulted in reduced cellular toxicity in the daughter cells (Koji et al. 2001), which may explain the observed higher toxicity at 20 than that of 30. To adapt higher salinity (40), more nutrients and energy were used to enhance osmoregulation, leading to a decrease in toxin production (Lim and Ogata 2005). In general, the effects of salinity on toxicity were species dependent. A decline of toxicity with elevated salinities has been reported in Alexandrium minutum, which was consistent with our results (Lim and Ogata 2005). In addition, a tight correlation between cell volume and PST quota (Pearson correlation, r = 0.63, p < 0.05) was also observed in Alexandrium ostenfeldii at different salinities (Martens et al. 2016).
Variation in the toxin production of harmful algae cultured in different nutrient conditions has been observed in multiple marine dinoflagellates (Flynn et al. 1996;Lee et al. 2012;Li et al. 2016;Lin et al. 2022;Wang and Hsieh 2002;Zhang 2009). In general, it is well documented that sufficient N and depleted P increase toxin production in toxic dinoflagellates (Lee et al. 2012;Macintyre et al. 1997;Touzet et al. 2007). The current results present a similar pattern in that the cellular toxicity was enhanced with the increasing nitrate or decreasing phosphate concentrations. It is not surprising that a sufficient supply of N appeared to be important for high toxin production in G. catenatum. This is because PSP toxins compose high molecular contents of N, which accounted for 5-10% of the total N in the cell of Alexandrium tamarense (Macintyre et al. 1997). As a result, insufficient N supply will influence the toxin synthesis directly. In contrast to N, P was not contained in PST molecules. Phosphorus stress could cause an increase in the availability of intracellular arginine, a presumed precursor in PST biosynthesis (Anderson et al. 1990;John and Flynn 2000;Taroncher-Oldenburg et al. 1997), thereby cellular toxicity was promoted. Compared with the cellular toxicity, the toxin composition remained constant with nitrate variations rather than with phosphate. In the present study, the ratio of C2:C1 toxins decreased in lower phosphate concentrations. Other strains of G. catenatum indicated that the toxin composition was not affected by nutrient status (Flynn et al. 1996;Oshima et al. 1993). In Alexandrium tamarense, however, the ratio increased slightly under low phosphate conditions (Lee et al. 2012). This reaffirms that the stability of toxin composition is greatly species/strain dependent.
In summary, the G. catenatum strain in the present study presented special growth and toxic characteristics, although it shared similar ribosomal sequences with strains from other regions. Variations in environmental conditions could also affect its growth and toxin production, and should be taken into consideration when assessing the harmful effects of G. catenatum blooms in the field. In addition to the environmental factors in the present study, various other factors, such as biotic factors and nutrient types, have also been shown to affect both the growth and cellular toxicity of other dinoflagellates Xu et al. 2012), which need to be further studied to reveal the comprehensive physiological characteristic of G. catenatum.