Can Increases in Temperature Stimulate Blooms of the Toxic Benthic Dinoflagellate Ostreopsis Ovata?

Harmful Algae 10 (2011) 165–172

Contents lists available at ScienceDirect

Harmful Algae

journal homepage: www.elsevier.com/locate/hal

Can increases in temperature stimulate blooms of the toxic benthic dinoﬂagellate Ostreopsis ovata?

Edna Grane´li a,*, Nayani K. Vidyarathna a, Enzo Funari b, P.R.T. Cumaranatunga c, Raffaeli Scenati b a Marine Ecology Department, Linnaeus University, 39182 Kalmar, Sweden b Environment & Health Dept., National Institute of Health, v.le Regina Elena 299, 00161 Rome, Italy c Dept. of Fisheries & Aquaculture, Faculty of Fisheries & Marine Sciences and Technology, University of Ruhuna, Matara, Sri Lanka

ARTICLE INFO ABSTRACT

Article history: Ostreopsis ovata Fukuyo is an epiphytic, toxic dinoﬂagellate, inhabiting tropical and sub-tropical waters Received 12 February 2010 worldwide and also in certain temperate waters such as the Mediterranean Sea. Toxic blooms of O. ovata Received in revised form 22 April 2010 have been reported in SE Brazil in 1998/99 and 2001/02 and the French–Italian Riviera in 2005 and 2006. Accepted 6 September 2010 These blooms had negative effects on human health and aquatic life. Chemical analyses have indicated that O. ovata cells produce palytoxin, a very strong toxin, only second in toxicity to botulism. Increase in Keywords: water temperature by several degrees has been suggested as the reason for triggering these blooms. Four Ostreopsis ovata laboratory experiments were performed with O. ovata isolated from Tyrrhenian Sea, Italy to determine Benthic dinoﬂagellate the effects of water temperature and co-occurring algae on the cell growth and/or the toxicity of O. ovata. Toxicity Palytoxin The cells were grown under different temperatures ranging from 16 8Cto308C, and cell densities, Temperature growth rates and the cell toxicities were studied. Results indicated high water temperatures (26–30 8C) Climate change increased the growth rate and biomass accumulation of O. ovata. In mixed cultures of O. ovata with other co-occurring algae, biomass decreased due to grazing by ciliates. Cell toxicity on the other hand was highest at lower temperatures, i.e., between 20 and 22 8C. The present study suggests that sea surface temperature increases resulted by global warming could play a crucial role inducing the geographical expansion and biomass accumulation by blooms of O. ovata. ß 2010 Elsevier B.V. All rights reserved.

1. Introduction Climate Change, continued emissions will lead to increase of world average temperature by 1.8–4 8C over the 21st century (IPCC, 2007). Climate-induced changes and other anthropogenic activities Since the oceans are core components of the climate system, these have been attributed to be the major driving forces behind the changes have a particular significance on the structure and function stimulation, distribution and the intensification of harmful algal of the marine ecosystems. The direct and indirect impacts of the blooms on the last decades (Van Dolah, 2000; Anderson et al., increases in greenhouse gas concentrations on the ocean will include 2002; Hallegraeff, 2003; Grane´li, 2004; Glibert et al., 2005; Cloern increasing sea surface temperatures, acidification, changes to the et al., 2005). The ecological and biogeochemical significance of density structure of the upper ocean which will alter vertical mixing blooms depend strongly on the species composition of the growing of waters, intensification/weakening of upwelling winds and community, which is shaped by changing physical dynamics over changes in the timing and volume of freshwater runoff into coastal multiple time scale (Cloern et al., 2005). These physical dynamics marine waters (Moore et al., 2008). may vary from the local winds and heat flux that result in a The processes that select a particular algal species to increase in stratification inside a small bay, to the ocean – basin scale cell numbers ultimately forming a bloom are still unresolved atmospheric processes that produce variability in wind-driven questions (Cloern et al., 2005). It is however, well understood that circulation patterns (Cloern et al., 2005). the growth and toxicity of many toxic dinoflagellates are There is near unanimous scientific consensus that the world may influenced by physical factors such as temperature, salinity, light be entering a period of global warming in response to atmospheric and by the amount of inorganic nutrients available, which play a greenhouse gas accumulation generated due to human activities significant role on the bloom formation (Morton et al., 1992; (Van Dolah, 2000). According to the Intergovernmental Panel on Grane´li and Flynn, 2006). Warmer temperatures may result in expanded ranges of warm water Harmful algal species such as Gambierdiscus toxicus, which * Corresponding author. Tel.: +46 0 480 447307; fax: +46 0 480 447305. was evident by the high abundance and extended distribution of G. E-mail address: [email protected] (E. Grane´li). toxicus in higher latitudes in response to elevated sea surface

1568-9883/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.hal.2010.09.002 166 E. Grane´li et al. / Harmful Algae 10 (2011) 165–172 temperature during the warm phases of ENSO cycle (Hales et al., Tyrrhenian Sea; (2) to determine whether the optimum tempera- 1999; Chateau-Degat et al., 2005). Low temperature has been ture to growth would be the same for the highest production of reported to decrease the growth rate and to increase the toxin toxin and (3) to find whether O. ovata cells produce allelochemicals concentration of PSP (Paralytic Shellfish Poisoning) producing at high temperatures, which allow the species to out-compete the dinoflagellates such as, Alexandrium catenella, A. cohorticula and other co-existing algae. To answer these questions we performed Gymnodinium catenatum (Ogata et al., 1989; Grane´li and Flynn, four laboratory experiments with O. ovata and its co-occurring 2006). Indirect impacts of increased temperature may also favor species under different temperatures. algal blooms, e.g., coral bleaching events free up space for macroalgae to colonize making available more habitats for benthic 2. Materials and methods epiphytic toxic dinoflagellates to explore (Moore et al., 2008). Interactions of toxin producing algal species with non-toxic Four experiments were performed at the Linnaeus University, algal species can be a factor of ultimate importance, on the Sweden. Seawater from the Genova Bay and cultures of O. ovata, structuring of the plankton communities. Specially if the toxins are Kalmar Algae Collection (KAC 85) which was used for the the same as allelopathic substances, which are released into the experiments were shipped to Sweden. O. ovata cells were isolated water, enabling the harmful species to out-compete co-occurring from seaweeds collected from the Tyrrhenian Sea, Italy in 2007 phytoplankton species (Grane´li and Johansson, 2003b; Legrand during a bloom of O. ovata. Prior to commencement of the et al., 2003; Glibert et al., 2005), it will ultimately determine which experiments, O. ovata cells were grown in f/10 media (Guillard and phytoplankton species will be able to co-exist and what their Ryther, 1962), prepared using water brought from Italy (filtered population size will be. Allelochemicals seems to be grazer through Whatman GF/C glass-fibre filters and autoclaved). These deterrents in most of the cases (Grane´li and Johansson, 2003a; cultures were maintained at a salinity of 38 psu, temperature of 2 1 Fistarol et al., 2004). Reductions in grazer abundance due to 25 8C and illuminated under 140 mEcm S by cool-white Osram deterrent substances can also play a key role in the development of tubes and exposed to 16:8 hr light: dark cycle. Algal species found toxic algal blooms (Smetacek, 2001; Guisande et al., 2002; in natural community assemblages were collected from Monte Frango´ pulos et al., 2004). Argentario, Tyrrhenian Sea, Italy and brought to Sweden. Much less information is available in the international scientific literature regarding interactions concerning benthic harmful 2.1. Experiment 1 dinoflagellates. The existing knowledge on species distribution and factors involved in the production of toxins are on the benthic The aims of the experiment 1 were to determine the optimum dinoflagellates producing ciguatera (Morton et al., 1992; Van temperature for growth of O. ovata and effect of temperature on Dolah, 2000; Lewis, 2001; Chateau-Degat et al., 2005; Yasumoto exudation of allelochemicals. et al., 1977). Three treatments were used in this experiment, viz. Oovata Ostreopsis ovata Fukuyo is a toxic benthic dinoflagellate, monocultures, a natural community assemblage and a mixture of inhabiting tropical and sub-tropical waters worldwide. Ostreopsis the two. Polystyrene 8 ml Petri dishes (3.5 cm in diameter) were species are often found in epiphytic association with red and used for the experiment. Stock cultures for controls were brown macroalgae (Faust et al., 1996; Grane´ li et al., 2002). Recent prepared by diluting 100 ml of O. ovata monoculture and occurrence of O. ovata blooms has been also recorded from many 100 ml of natural community assemblage with 150 ml of f/10 temperate regions (Shears and Ross, 2009). O. ovata has attracted medium and the mixed treatment was prepared by mixing attention during the past years due to its unusual blooms and 100 ml of O. ovata monoculture and 100 ml of natural community production of palytoxin, which is one of the most potent known assemblage diluted with 50 ml of f/10 medium to obtain the existing toxin. O. ovata and O. lenticularis have been shown to same initial cell concentrations for each treatment. Five produce palytoxin and its analogues (e.g. ostreosin-D), that can millilitres from each of the three stock solutions were poured cause human fatalities by accumulating higher up in marine food into the petri dishes and tightly closed using paraffin. Each webs (Grane´ li et al., 2002; Rhodes et al., 2002; Ashton et al., treatment was exposed to 6 temperatures (16 8C, 20 8C, 24 8C, 2003; Taniyama et al., 2003; Ciminiello et al., 2006; Monti et al., 26 8C, 28 8C, 30 8C) in triplicates. 2007). Observation of toxic compounds found both in Ostreopsis The Petri dishes were kept floating on water in 13 L cells and shellfish tissues collected in Greek coastal waters polyethylene aquaria exposed to the temperatures given above, (Aligiazaki et al., 2008)alsoinOstreopsis cells and parrot fish, and photoperiod as mentioned above. Temperatures in the collected from Tokushima Prefecture, Japan (Taniyama et al., aquaria were maintained by using aquarium heaters. Whenever 2003) indicate that the toxins produced by Ostreopsis cells can be droplets of water were formed inside the lids of the Petri dishes, transferred to higher trophic levels via food chains. Aerosol duetoevaporation,theyweregentlyshakentoavoidsalinity containing toxins from the bloom of O. ovata in Genova, Italy in increases in the culture media. Cell densities of O. ovata and/or the 2005 and 2006 have caused respiratory problems for more than two most dominant co-occurring algae (Sp. 1 and 2) in each 150 people, who had to be hospitalized (Brescianini et al., 2006; chamber were estimated every second day, by directly counting Durando et al., 2007). In Cabo, Frio, SE Brazil in summer of 1998/ the live cells after placing the Petri dishes on an inverted 99 and 2001/02, sea urchins (Echinometra lucunter)startedto microscope (Olympus CKX 41). Cell sizes of O. ovata,other loose their spines, finally facing death due to necrosis after dominant co-occurring algal species and ciliates were measured. ingestion of O. ovata cells, epiphytic on the seaweed species. The experiment was continued for 13 days. Samples from each During the time of the blooms the water temperatures have been treatment were subjected to epifluorescence microscopy (Olym- around 22–26 8C(Grane´ li et al., 2002; Ciminiello et al., 2006). pusBX50).Thespecificgrowthrate(m) for each culture was Even though the blooms of O. ovata have been recorded at high calculated during the exponential growth phases at each water temperatures and atmospheric pressures and at low water temperature using the following formula. turbulence (Brescianini et al., 2006), no documentation exists on the environmental conditions and factors promoting palytoxin ln N2 ln N1 production in O. ovata cells. m ¼ t t The aims of the present study were: (1) to determine the 2 1 optimum temperature for the growth of O. ovata, isolated from the N1 and N2 are the average values of cell numbers at time t1 and t2. E. Grane´li et al. / Harmful Algae 10 (2011) 165–172 167

2.2. Experiment 2 2.6. Hemolytic procedure

Experiment 2 was performed to determine whether the The tests were carried out in duplicates. A standard hemolytic optimum temperature for the growth of O. ovata is same as the curve based on concentrations of saponin (Sigma S-2149) in an temperature for highest toxicity. isotonic phosphate buffer was used as the reference, and the Monocultures of O. ovata were used for this experiment and hemolytic activity of the cells was determined as saponin same experimental conditions used for experiment 1 were equivalents. The results were expressed as saponin nano-equiva- applied. Samples for analyses of toxin were obtained when cell lent per cell (ng SnE cell1). Hemolytic activity in methanol was densities could be visually detected as brownish filaments inside also determined, following the same procedure, to exclude the the Petri dishes. Cells were collected by concentrating the possible toxic effect of methanol that could be interpreted as the cultured cells on 25-mm glass-fiber filters (Whatman GF/C) and toxin itself. they were placed in 2 ml eppendorf tubes. Cells and the cell free The hemolytic test for cultures in experiment 2 was performed filtrates were kept frozen, to be used for the hemolytic test (details on 5% horse blood using the method adopted by Stolte et al. (2002) are given below). 2.0 ml from each replicate was preserved with and the hemolytic activity was measured with a plate reader (BMG acid Lugol’s solution and cells were counted as in experiment 1. FLUOstar) after an incubation period of 4 h, by measuring the One-way ANOVA test was used to determine whether there is a absorbance at 540 nm. variation in final cell counts in cultures maintained under In experiments 3 and 4 the same method was adopted but with different treatments (i.e. different temperatures) and Tukey’s some modifications according to Eshbach et al. (2001). The horse pairwise comparison test was performed whenever significant blood concentration used was 1% and the absorbance was differences were observed. measured after 4 h at 405 nm instead of 540 nm using the same equipment as in experiment 2.

2.3. Experiment 3 The HE50 of the algae cells was calculated according to Simonsen and Moestrup (1997) using the following equation: Observation of highest toxicity in O. ovata cultures maintained between 16 and 24 8C in the experiment 2 led to perform the HE CK S1 experiment 3 and it was carried out with the objective of 50 50 determining the most critical temperature for the toxicity of O. 1 Where C is the concentration of cells (cells ml ), S is saponin ovata, as the interval for temperatures used in experiment 2 were 1 equivalents (mg ml ), and K50 is the concentration of saponin of 4 8C. Experimental design was similar to experiment 2 but the 1 (mg ml ) causing 50% hemolysis of the horse blood cells. cells were exposed to 17 8C, 18 8C, 19 8C, 20 8C, 21 8C, 22 8C and 23 8C and instead of using Petri dishes, 250 ml Erlenmeyer flasks 2.7. Cell carbon analysis containing 200 ml of cultures were used. The cell growth was determined by manually counting the number of cells in sub- The cell carbon contents of O. ovata were analysed in cells samples drawn from the cultures every second day and the retained on 25 mm precombusted (450 8C, 2 h) Whatman GF/C experiment was continued for 20 days. 60.0 ml of the culture from glass-fiber filters after filtering 30 ml of cultures grown at the same each replicate was harvested at day 14 and the rest were harvested conditions as above, but at 25 8C. The filters were dried at 65 8C for at the end of the experiment for analyses of toxin concentrations in 24 h, and analysed for particulate carbon (POC) with a CHN the cells. The cell densities and the toxin concentrations were elemental analyzer (Fisons Instruments model NA 1500). Cell statistically compared in cultures maintained under different carbon contents of Nitzschia closterium (Sp.1) and Navicula sp. (Sp. temperatures using one-way ANOVA and Tukey’s pairwise 2) were calculated following the stereometric formulas explained comparison test. by Edler (1979).

2.4. Experiment 4 3. Results

Experiment 4 was performed to further clarify the pattern of 3.1. Experiment 1 cell growth vs. cellular toxin levels of O. ovata under different temperatures. Cells were grown at the temperatures where the 3.1.1. Cell densities highest (20 8C and 22 8C) and lowest (30 8C) cellular toxicity levels O. ovata cell densities increased at all the temperatures except were observed in the previous experiments. 16 8C, following a typical growth pattern (Fig. 1(a)). Maximum cell Monocultures of O. ovata in f/10 media were grown at above densities (5.67 0.4 103 cells ml1) were obtained at 26 8Conday9 three temperatures and other experimental conditions were followed by 5.41 0.4 and 5.13 0.4 103 cells ml1,observedonthe maintained as in experiment 3. Except toxin analyses were day 13, at 28 8Cand308C respectively. On the last day of the performed on three occasions representing different growth experiment the highest cell counts were recorded at 30 8C while the stages. The experiment was carried out until few days after lowest were at 16 8C. The highest specific growth rates were recorded at commencing the declining phase of their growth and therefore the 30 8C (0.74 0.1), which was followed by that of 28 8C(0.59 0.1) and final harvest was taken on day 38. The results were statistically 26 8C(0.53 0.1), and the lowest were recorded at 16 8C(0.1 0.1). analyzed using one-way ANOVA and repeated measures ANOVA. In the natural community assemblage and in the mixed treatments, Nitzschia closterium and Navicula sp. (25 mmin 2.5. Extraction of toxic substances length and 7 mm in width) were identified as sp. 1 and 2, the two most dominant co-occurring algal species. In the mixed Hemolytic tests were performed on sub-samples drawn from treatments, O. ovata cells decreased to very low levels from day 2 the cultures on different occasions (depending on the experiment). while the cell densities of diatom N. closterium increased. At the Cells in each sample were filtered onto 25-mm glass-fiber filters end of the experiment, O. ovata cells were not observed in any of (Whatman GF/C) and cells retained on the filters were extracted in the cultures maintained at different temperatures (Fig. 1b). At 1.5 ml methanol for 30 min in the dark. The extracts were used higher temperatures (28–30 8C) no cells of O. ovata were observed immediately for analyses or stored at 20 8C prior to analyses. after day 4. 168[(Fig._1)TD$IG] E. Grane´li et al. / Harmful Algae 10 (2011) 165–172

diatoms (13 pg cell1 and 2.42 pg cell1 for N. closterium and Navicula sp. respectively). Total cell carbon contents (mgC l1)ofN. closterium and Navicula sp. in the mixed treatments were signiﬁcantly lower than that of O. ovata and remained in the same low levels throughout the whole experimental period at all the temperatures except 16 8C. Slight increase in total cell carbon content of N. closterium in the mixed treatments at 16 8C was evident which decreased after day 10. In the mixed treatments at all the temperatures different species of ciliates, in the size range of 90–160 mm (length) and 25–45 mm (width), e.g. Euplotes sp. (95 mminlengthand45mm in width) were observed. Some heliozoans and heterotrophic dinoﬂagellate, Oxyrrhis marina were also present in the mixed treatments.

3.2. Experiment 2

3.2.1. Cell densities In this experiment 2, O. ovata cell densities increased with temperature i.e. the higher the temperature the higher cell densities (Fig. 3). Statistically significant differences were found between the average final cell numbers at different temperatures 1 Fig. 1. Cell densities (mean SD, cells ml )ofO. ovata grown at 16 8C, 20 8C, 24 8C, (ANOVA, F = 10.96, p < 0.05). Cell densities at 30 8C were 26 8C, 28 8C and 30 8C in the monocultures (a) and in the mixed treatments (b). significantly higher than those maintained at 16 8C, 20 8C, 24 8C, and 26 8C (Tukey’s, p < 0.028). Also cell densities at 28 8C were The cell densities of N. closterium in the mixed treatment were significantly higher than those maintained at 16 8C, 20 8C (Tukey’s, significantly higher than that of the control treatment of the p < 0.016). natural algal community. Even though the cell numbers of Navicula sp. in mixed treatments and in control treatments did not show a 3.2.2. Hemolytic activity significant difference, it was fairly high in mixed treatments when Hemolytic activity was found at all temperatures. In contrast to compared to the controls. The cell densities of sp. 1 and 2 were the cell growth, a significantly higher toxicity was observed at higher at lower temperatures than at higher temperatures. 20 8C, which was 18.1 ng SnE cell1. The second highest value was To compensate for the differences in cell sizes in O. ovata, N. found for cells grown at 16 8C and it decreased gradually towards closterium and Navicula sp., two most dominant co-occurring algae the higher temperatures. Thus toxicity of O. ovata grown at lower in the mixed treatment, the biomass accumulation was expressed temperatures produces higher amounts of toxins per cell than in terms of carbon content (Fig. 2). O. ovata cell carbon when grown at high temperatures, where maximum cell numbers (3439.1[(Fig._2)TD$IG] 563.4 pg cell1) was much higher than the other two were observed (Fig. 3).

Fig. 2. Cellular carbon contents (mgl1)ofO. ovata, N. closterium and Navicula sp. grown in mixed treatments exposed to 16 8C, 20 8C, 24 8C, 26 8C, 28 8C and 30 8C. [(Fig._3)TD$IG] E. Grane´li et al. / Harmful Algae 10 (2011) 165–172 169

Fig. 3. Cell densities (mean SD, cells ml1) and toxicity (ng SnE cell1)ofO. ovata grown at 16 8C, 20 8C, 24 8C, 26 8C, 28 8C and 30 8C on the ﬁnal day of the experiment. The constant line represents the initial cell densities (cells ml1).

3.3. Experiment 3 toxin levels in the cells grown at 22 8C were significantly higher than in those grown under other treatments and it was six fold higher 3.3.1. Cell densities compared to the levels observed in the cells grown at 23 8C(ANOVA, As in experiments 1 and 2 higher cell densities were found at F =10.685, p < 0.05). The results from both experiments 2 and 3 increasing temperatures (Fig. 4). A significant difference in final cell showed that the most critical temperature range for high toxicity of O. densities among the treatments was observed (ANOVA, F =15.570, ovata was observed between 20 8Cand228C. p < 0.05). The highest final cell densities were recorded at 23 8C, which was followed by the cultures in 21 8Cand228C. The cell 3.4. Experiment 4 densities were 3.05 0.6, 2.56 0.4 and 2.51 0.1 103 cells ml1 respectively. The lowest cell densities, 1.08 0.2 103 cells ml1, 3.4.1. Cell densities were recorded at 17 8C. As in the previous experiments O. ovata had significantly higher cell densities at 30 8C than at 22 8C and 20 8C(Fig. 6) during the 3.3.2. Hemolytic activity stationary phase of the growth curve (ANOVA, F = 25.424, In all treatments the cells produced hemolytic activity (HA). p < 0.05). The cell numbers in all the treatments did not differ However, HA on day 14, were lower than that of day 20, except for significantly during the declining phase. The maximum cell the cultures growing at 23 8C. The highest toxicities, densities recorded during the experiment were 2.68 0.3 103 11.57 1.9 ng SnE cell1 were recorded at 22 8Cwhilethelowest, cells ml1, which were recorded on 18th day at 30 8C. The cell 1.93 0.8 ng SnE cell1,at238C(Fig. 5). Toxicity increased from 17 8C densities in the cultures grown at 20 8C were comparatively low to 22 8C, with a rapid drop with further increases in temperature. The during the whole experimental period. [(Fig._4)TD$IG] 3.4.2. Hemolytic activity Although hemolytic activity was found for all treatments, there was a variation in activity along O. ovata growth cycle. Even though the difference in toxin concentrations in the cells harvested at different phases of the growth was not statistically significant, comparatively higher toxin concentrations were detected in the cells harvested at day 31, where the cell growth was in the declining phase. Significantly higher cellular toxin levels were detected[(Fig._5)TD$IG] during the declining phase in the cells grown at 20 8C and

Fig. 4. Cell densities (mean SD, cells ml1)ofO. ovata grown at 17 8C, 18 8C, 19 8C, Fig. 5. Hemolytic activity (mean SD, ng SnE cell1)ofO. ovata cells grown in 17 8C, 20 8C, 21 8C, 22 8C, 23 8C. 18 8C, 19 8C, 20 8C, 21 8C, 22 8C, and 23 8C on days 14 and 20 of the experiment. 170[(Fig._6)TD$IG] E. Grane´li et al. / Harmful Algae 10 (2011) 165–172

growth were found to be between 26 and 30 8C during this study. Similar observations have been made by Mangialajo et al. (2008) who recorded the highest abundance of O. ovata in Genova when water temperature exceeded 26 8C. These evidence suggest that there is a pronounced effect of relatively high temperature on the growth of O. ovata. Therefore it can be speculated that the rising in temperatures have stimulated the growth of O. ovata during the bloomsinbothBrazilandItaly. The growth of some other benthic and planktonic dinoflagellates such as Prorocentrum donghaiense (Xu et al., 2010), G. toxicus and Coolia monotis (Morton et al., 1992), A. tamiyavanichii and A. minutum (Lim et al., 2006) and the raphidophyte Heterosigma akashiwo (Fu et al., 2008) have also been shown to be enhanced by high seawater temperatures. The optimum temperature range for P. minimum growth has been recorded between 18 and 26 8C with a Fig. 6. Cell densities (mean SD, cells ml1)ofO. ovata grown under 20 8C, 22 8C and rapid decrease above 26.5 8C and a rapid increase between 13 and 30 8C during the experiment. The arrows indicate the days of sample harvesting (24, 31 18 8C(Grzebyk and Berland, 1996). On the contrary, a different and 38), for toxin analyses. temporal pattern of O. ovata bloom formation has been observed in northern Adriatic Sea where the maximum cell abundance was 22 8C when compared to those grown at of 30 8C (ANOVA, found for decreasing temperatures from 26 8Cto168C(Totti et al., F = 107.093, p < 0.05). The maximum toxin contents per cell 2010). Also the growth of O. heptagona and O. siamensis were found recorded were 5.06 0.2 ng SnE cell1, followed by 5.02 to be optimal at temperatures less than 26 8C(Morton et al., 1992), 0.2 ng SnE cell1 and they were observed in the samples harvested on the other hand, no significant correlation was observed between at day 31, from the cultures in 20 8C and 22 8C respectively (Fig. 7). epiphytic dinoflagellate assemblages and water temperature in the The lowest toxin concentrations were observed in 30 8C at day 24 Mediterranean Sea (Vila et al., 2001). Similarly, in southwestern (1.86 0.7 ng SnE cell1). A slight difference in the cellular toxicity Puerto Rico, the seasonality of dominant epiphytic dinoflagellates levels among treatments was evident at day 24 (ANOVA, F = 10.578, O. lenticularis and G. toxicus were not strongly correlated with p < 0.05) whereas the difference was not significant at day 38. temperature (Ballantine et al., 1988). Thus, the importance of high temperature as a triggering factor for O. ovata cell growth could be 4. Discussion strain specific.

During the present study stimulation of O. ovata cell growth 4.2. Effects of co-occurring algae on the growth of O. ovata and was observed at high temperatures (26–30 8C) while highest vice-versa cellular toxicities were observed at low temperatures (20–22 8C). It has long been argued that many toxic dinoflagellates produce 4.1. Effects of temperature on the growth of O. ovata allelochemicals in order to compete with other co-occurring algae under unfavorable environmental conditions for growth (Grane´li Dinoflagellate species as well as all other phytoplankton and Johansson, 2003b; Grane´li and Hansen, 2006). Several authors groups/species depend on the optimum sea surface temperature have observed that Ostreopsis species are able to produce and to grow at their highest rates (Morton et al., 1992; Hales et al., release hemolytic compounds and polysaccharides (mucilage) into 1999; Chateau-Degat et al., 2005; Grane´ li and Flynn, 2006; the water (Yasumoto et al., 1987; Vila et al., 2001; Ashton et al., Navarro et al., 2006). The results from all the experiments 2003; Taniyama et al., 2003; Lenoir et al., 2004). During the blooms presented here clearly demonstrate that the growth of O. ovata is in Brazil in 1998/99 and 2001/22, microscopic investigations have strongly stimulated under high temperatures and it is consistent revealed that, all seaweeds (most calcarean ones such as Jania sp.), with the previous findings from field observations during the were covered with a 0.5 cm thick layer of mono-specific O. ovata blooms in Genova, Italy (Ciminiello et al., 2006), Brazil (Grane´ li cells in gelatinous mass and no other microalgae were found et al., 2002), Tosa Bay, Japan (Adachi et al., 2008) and on temperate among them (Grane´li et al., 2002; Grane´li, pers. observation). This reefs in New Zealand (Shears and Ross, 2009). Nevertheless, the indicates that either the polysaccharides or the hemolytic present study also shows that O. ovata cangrowatawiderangeof compounds produced by O. ovata cells have been acting as temperatures[(Fig._7)TD$IG] (16–30 8C). The optimum temperatures for O. ovata allelochemicals. However laboratory studies have not yet been conducted on production of allelochemicals and/or their allelopathic effects on other microalgae. During the present study, in the mixed cultures growth of O. ovata was not observed, but N. closterium and Navicula sp. were observed at all the experimental temperatures. However, when transforming the cell number to level of carbon, the results show that O. ovata was not out- competed either by N. closterium or Navicula sp. in mixed treatments. Since O. ovata grew well in monocultures in all the temperatures, the decline of the cells after some days in all the mixed treatments can be explained by the impact of grazing by ciliates that increased quite dramatically during the course of the experiment. These ciliates were present among the microalgae community in the natural assemblages from the beginning of the experiment. The epiflorescence microscopy on the samples from Fig. 7. Hemolytic activity (mean SD, ng SnE cell1)ofO. ovata cells grown under mixed treatments showed that Euplotes sp. contained photosyn- 20 8C, 22 8C and 30 8C in the day 24 and 31 and 38. thetic cells similar to O. ovata cells in size. The ciliates were E. Grane´li et al. / Harmful Algae 10 (2011) 165–172 171 selectively grazing O. ovata cells, which are more size-suitable to stationary phases (Gedaria et al., 2007). Decrease in cellular toxin them than the smaller N. closterium. This might explain why O. content of O. ovata cells after day 31 (declining phase) could be ovata disappeared during the course of the experiment in the probably due to the increased release of toxins to the medium. mixed cultures, as a result of grazing and not due to competition Results of the present study, confirm the findings for other with the other algal species. toxic species (see e.g. Grane´ li and Flynn, 2006), i.e., that the level The ciliate preferences for O. ovata cells as a food item also of toxicity per cell increase at temperatures when the cell growth explain the comparatively high cell densities of N. closterium and is sub-optimal. Even if this is the case, at high temperatures, Navicula sp. in the mixed treatments when compared to the which are favourable for the cell growth and toxicity per cell is controls. At the beginning of the experiment, grazing pressure low, high cell densities can result in same or higher levels of might be high on O. ovata cells, than on other smaller co-occuring toxins, that can be transported through food chain or aerosols, algae. In order to elucidate whether O. ovata toxins and/or with a capacity strong enough to cause considerable negative polysaccharides have allelopathic significance, experiments simi- effects on the human health and the environment. This seems to lar to those conducted by Grane´li and Johansson (2003b) and bethecaseduringthebloomsofO. ovata in Genova, Italy in 2005/ Fistarol et al. (2004) should be carried out. In this case only mono- 06 and in Brazil in 1998/99 and 2001/02. This indicates that if a cultures of the co-occuring algae will be used and grazing by bloom is triggered under lower temperatures (at higher ciliates should be avoided. latitudes), it could result in higher toxin levels and higher impacts on the environment. 4.3. Effects of temperature on the toxicity of O. ovata 5. Conclusions In general changes in toxin content are associated with stress caused under disturbed physiology (Johansson and Grane´li, 1999; The optimum temperatures for growth and toxicity of O. ovata Grane´li and Flynn, 2006) by several factors such as pH (Hwang and were found to be inversely related. High water temperatures (26– Lu, 2000), temperature (Ogata et al., 1989; Ashton et al., 2003), 30 8C) stimulated O. ovata cells growth rate and biomass salinity (Parkhill and Cembella, 1999; Gedaria et al., 2007), accumulation and low toxicities while lower temperatures (20– illumination (Hwang and Lu, 2000), and nutrient deficiency 22 8C) induced higher toxicity per cell and lower cell numbers.- (Grane´li and Johansson, 2003b). Based on the results of the present experiments it can be suggested Present study shows that toxicity was higher in O. ovata cells that increased sea surface temperature, which can result from grown between 16 and 22 8C, which was maximized at 20 8C/22 8C global warming may play a crucial role inducing the geographical where comparatively low O. ovata cell densities were observed. expansion and biomass increase, blooms of O. ovata in future. Contrary to this observation, Ashton et al. (2003) reported, an enhanced toxicity for Ostreopsis lenticularis at 29.5 8C, than at 25 8C. But, as O. ovata and O. lenticularis are two different species, the Acknowledgements environmental conditions that enhance toxicity in them can be completely different from one another. We are most grateful to Paulo Salomon for the help with Similar to the observation made during the present study for O. hemolytic tests and Christina Esplund for the CHN analysis. This ovata, an inverse correlation between temperature and toxin study was supported financially by the funds from SIDA/SAREC concentration have been observed for A. catenella (Navarro et al., (Swedish International Development Cooperation Agency) Marine 2006). Contrary to O. ovata, the optimum growth for Protoceratium Sciences Programme of University of Ruhuna, Sri Lanka, N.K. reticulatum has been reported to be between 16 and 20 8C, while Vidyarathna scholarship and European Commission (Research the highest toxicity has been at 26 8C, at which the growth Directorate General- Environment) through the project, MIDTAL inhibition has been evident (Guerrini et al., 2007). Similar inverse (FP7-ENV 2007-1, Grant Agreement No.: 201724), contract holder correlations between optimum growth conditions and toxicity Edna Grane´li. [SS]. have been found for some other dinoflagellate species, viz. A. cohorticula (Ogata et al., 1989); Pyrodinium bahamense var. References compressum (Gedaria et al., 2007)andHeterosigma akashiwo (Ono et al., 2000). Growth rates and the toxin production rates of A. Adachi, M., Ikegami, T., Yamaguchi, H. 2008. The genus Ostreopsis in the coastal minutum and A. tamiyavanichii have been found to be increased regions in Tosa Bay, southern part of Japan. The 13th International Conference with elevated temperatures (Lim et al., 2006). As it is often on Harmful Algae, Poster Presentation P29, Hong Kong, China 3–7 November, 2008. www.redibal.org/BD_DOCUMENTOS/ICHA_2008_Abstract.pdf. reported, factors stimulating toxin production in one algal Aligiazaki, K., Katikou, P., Nikolaidis, G., Panou, A., 2008. First episode of shellfish species/group may have a different impact on another (Grane´ li contamination by palytoxin-like compounds from Ostreopsis species (Aegean and Flynn, 2006). Sea, Greece). Toxicon 51, 418–427. Anderson, D.M., Glibert, P.M., Burkholder, J.M., 2002. Harmful algal blooms and Results from the experiment 4 (Fig. 7) clearly confirmed that eutrophication: nutrient sources, composition and consequences. Estuaries 25, toxin concentration per cell vary with the growth phase and was 562–584. highest during the declining phase. Although the cell densities Ashton, M., Tosteson, T., Tosteson, C., 2003. The effect of elevated temperature on the toxicity of the laboratory cultured dinoflagellate Ostreopsis lenticularis were not significantly different among different treatments during (Dinophyceae). Rev. Biol.Trop. 51, 1–6. this phase, the cellular toxin levels were twice as higher in the Ballantine, D.L., Tosteson, T.R., Bardales, A.T., 1988. Population dynamics and cultures grown at 20 8C and 22 8C compared to those grown at toxicity of natural populations of benthic dinoflagellates in southwestern 30 8C. Toxin concentrations of O. ovata strains isolated from Puerto Rico. J. Exp. Mar. Biol. Ecol. 119, 201–212. Brescianini, C., Grillo, C., Melchiorre, N., Bertolotto, R., Ferrari, A., Vivaldi, B., Icardi, Tyrrhenian and Adriatic seas have also been found to be high at the G., Gramaccioni, L., Funari, E., Scardala, S., 2006. Ostreopsis ovata algal blooms end of stationary phase than at the end of exponential phase affecting human health in Genova, Italy, 2005 and 2006. Euro Surveill. 11 (36) (Guerrini et al., 2009). An enhanced toxin concentration at the pii = 3040. Chateau-Degat, M.L., Chinain, M., Cerf, N., Gingras, S., Hubert, B., Dewailly, E., 2005. declining stage, compared to the other phases of growth cycle has Seawater temperature, Gambierdiscus spp. Variability and incidence of cigua- also been demonstrated in A. minutum (Hwang and Lu, 2000). tera poisoning in French Polynesia. Harmful Algae 4 (6), 1053–1062. However the toxin content of Pyrodinium bahamense var. Ciminiello, P., Dell’Aversano, C., Fattorusso, E., Forino, M., Magno, G.S., Tartaglione, L., Grillo, C., Melchiorre, N., 2006. The Genoa 2005 outbreak. Determination of compressum during the exponential phase has been the highest putative palytoxin in Mediterranean Ostreopsis ovata by a new liquid chroma- and a rapid decline has been observed at late exponential and tography tandem mass spectrometry method. Anal. Chem. 78, 6153–6159. 172 E. Grane´li et al. / Harmful Algae 10 (2011) 165–172

Cloern, J.E., Schraga, T., Lopez, C., Knowles, N., Labiosa, R., Dugdale, R., 2005. Climate IPCC. 2007. Climate Change 2007: the Physical Science Basis. Contribution of anomalies generate an exceptional dinoflagellate bloom in San Francisco Bay. Working Group I to the Fourth Assessment Report of the Intergovernmental Geophys. Res. Lett. 32, L14608 doi: 14610.11029/12005GL023321. Panel on Climate Change. New York:Cambridge University Press. http:// Durando, P., Ansaldi, F., Oreste, P., Moscatelli, P., Marensi, L., Grillo, C., Gasparini, R., www.ipcc.ch. Icardi, G., 2007. Ostreopsis ovata and human health: epidemiological and clinical Johansson, N., Grane´li, E., 1999. Influence of different nutrient conditions on cell features of respiratory syndrome outbreaks from a two year syndromic sur- density, chemical composition and toxicity of Prymnesium parvum (Hapto- veillance, 2005–06, in north-west Italy. Euro Surveill. 12 (23). phyta) in semi-continuous cultures. J. Exp. Mar. Biol. Ecol. 239, 243–258. Edler, L., 1979. Recommendations on methods for marine biological studies in the Legrand, C., Rengefors, K., Fistarol, G.O., Grane´li, E., 2003. Allelopathy in phyto- Baltic Sea. Phytoplankton and chlorophyll. The Baltic Marine Biologists, Pub- plankton–biochemical, ecological and evolutionary aspects. Phycologia 42 (4), lications No. 5. 406–419. Eshbach, E., Scharsack, J., John, U., Medlin, L., 2001. Improved erythrocyte lysis assay Lenoir, S., Ten-Hage, L., Turquet, J., Quod, J.-P., Bernard, C., Hennion, M.-C., 2004. in microtitre plates for the sensitive detection and efficient measurement of First evidence of palytoxin analogues from an Ostreopsis mascarensis (Dino- haemolytic compounds from ichthyotoxic algae. J. Appl. Toxicol. 21, 513– phyceae) benthic bloom in south-westernIndianOcean.J.Phycol.40,1042– 519. 1051. Faust, M.A., Morton, S.L., Quod, J.P., 1996. Further SEM study of marine dinofla- Lewis, R.J., 2001. The changing face of ciguatera. Toxicon 39 (1), 97–106. gellates: the genus Ostreopsis (Dinophyceae). J. Phycol. 32, 1053–1065. Lim, P.-T., Leaw, C.-P., Usup, G., Kobiyama, A., Koike, K., Ogata, T., 2006. Effect of light Fistarol, G.O., Legrand, C., Selander, E., Hummert, C., Stolte, W., Graneli, E., 2004. and temperature on growth, nitrate uptake, and toxin production of two Allelopathy in Alexandrium spp.: effect on a natural plankton community and on tropical dinoflagellates: Alexandrium tamiyavanichii and Alexandrium minutum. algal monocultures. Aquat. Microb. Ecol. 35, 45–56. J. Phycol. 42, 786–799. Frango´ pulos, M., Guisande, C., deBlas, E., Maneiro, I., 2004. Toxin production and Mangialajo, L., Bertolotto, R., Cattaneo-Vietti, R., Chiantore, M., Grillo, C., Leme´ e, R., competitive abilities under phosphorus limitation of Alexandrium species. Melchiorre, N., Moretto, P., Povero, P., Ruggieri, N., 2008. The toxic benthic Harmful Algae 3 (2), 131–139. dinoflagellate Ostreopsis ovata: quantification of proliferation along the coast- Fu, F.X., Zhang, Y., Warner, M.E., Feng, Y., Sun, J., Hutchins, D.A., 2008. A comparison line of Genoa, Italy. Mar. Pollut. Bull. 56, 1209–1214.

of future increased CO2 and temperature effects on sympatric Heterosigma Monti, M., Minocci, M., Beran, A., Ivesa, L., 2007. First record of Ostreopsis cfr. ovata akashiwo and Prorocentrum minimum. Harmful Algae 7, 76–90. on macroalgae in the Northern Adriatic Sea. Mar. Pollut. Bull. 54, 598–601. Gedaria, A.I., Luckas, B., Reinhardt, K., Azanza, R.V., 2007. Growth response and toxin Moore, S.K., Trainer, V.L., Mantua, N.J., Parker, M.S., Laws, E.A., Backer, L.C., Fleming, concentration of cultured Pyrodinium bahamense var. compressum to varying L.E., 2008. Impacts of climate variability and future climate change on harmful salinity and temperature conditions. Toxicon 50, 518–529. algal blooms and human health. Environ. Health 7 (Suppl 2), S4. Glibert, P.M., Anderson, D.M., Gentien, P., Grane´li, E., Sellner, K.G., 2005. The global, Morton, S.L., Norris, D.R., Bomber, J.W., 1992. Effect of temperature, salinity and complex phenomena of harmful algal blooms. Oceanography 18 (2), 136–147. light-intensity on the growth and seasonality of toxic dinoflagellates associated Grane´li, E., 2004. Eutrophication and Harmful Algal Blooms. In: Wassmann, P., Olli, with ciguatera. J. Exp. Mar. Biol. Ecol. 157, 79–90. K. (Eds.) Drainage Basin Nutrient Inputs and Eutrophication: An Integrated Navarro, J.M., Munoz, M., Contreras, A.M., 2006. Temperature as a factor regulating Approach. Kalmar, http://moritz.botany.ut.ee/olli/eutr/. growth and toxin content in the dinoflagellate Alexandrium catenella. Harmful Grane´li, E., Ferreira, C.E.L., Yasumoto, T., Rodrigues, E.M. and Neves, B., 2002. Sea Algae 5, 762–769. urchins poisoning by the benthic dinoflagellate Ostreopsis ovata on the Brazilian Ogata, T., Kodama, M., Ishimaru, M., 1989. Effect of water temperature and light Coast. In: Steidinger K. A. (ed.). Xth Int. Conf. on Harmful Algae, St. Pete Beach, intensity on growth rate and toxin production of toxic dinoflagellates. In: Florida, p. 113, Book of Abstracts. Okaichi, T., Anderson, D.M., Nemoto, T. (Eds.), Red Tides Biology, Environmental Grane´li, E., Flynn, K., 2006. Chemical and physical factors influencing toxin content. Science and Toxicology. Elsevier, NY, pp. 423–426. In: Grane´li, E., Turner, J.T. (Eds.), Ecology of Harmful Algae, Ecological Studies, Ono, K., Khan, S., Onoue, Y., 2000. Effects of temperature and light intensity on the 189. Springer–Verlag, Berlin Heidelberg, Germany, pp. 229–241. growth and toxicity of Heterosigma akashiwo (Raphidophyceae). Aquacult. Res. Grane´li,E.,Hansen,P.J.,2006.Allelopathyinharmfulalgae:amechanismtocompetefor 31, 427–433. resources? In: Grane´li, E.,Turner, J.T. (Eds.), Ecology of Harmful Algae, Ecological Parkhill, J.P., Cembella, A.D., 1999. Effects of salinity, light and inorganic nitrogen on Studies, 189. Springer–Verlag, Berlin Heidelberg, Germany, pp. 189–201. the growth and toxigenicity of the marine dinoflagellate Alexandrium tamar- Grane´li, E., Johansson, N., 2003a. Effects of the toxic haptophyte Prymnesium parvum ense. J. Plank. Res. 21, 939–955. on the survival and feeding of a ciliate: the influence of different nutrient Rhodes, L., Towers, N., Briggs, L., Munday, R., Adamson, J., 2002. Uptake of palytoxin- conditions. Mar. Ecol. Prog. Ser. 254, 49–56. like compounds by shellfish fed Ostreopsis siamensis (Dinophyceae). New Zeal. J. Grane´li, E., Johansson, N., 2003b. Increase in the production of allelopathic sub- Mar. Fresh. Res. 36, 631–636. stances by Prymnesium parvum cells grown under N- or P-deficient conditions. Shears, N.T., Ross, P.M., 2009. Blooms of benthic dinoflagellates of the genus Harmful Algae 2, 135–145. Ostreopsis; an increasing and ecologically important phenomenon on temperate Grzebyk, D., Berland, B., 1996. Influences of temperature, salinity and irradiance on reefs in New Zealand and worldwide. Harmful Algae 8, 916–925. growth of Prorocentrum minimum (Dinophyceae) from the Mediterranean Sea. J. Simonsen, S., Moestrup, Ø., 1997. Toxicity test in eight species of Chrysochromulina Plankton Res. 18, 1837–1849. (Haptophyta). Can. J. Bot. 75, 129–136. Guerrini, F., Ciminiello, P., Dell’Aversano, C., Tartaglione, L., Fattorusso, E., Boni, L., Smetacek, V., 2001. A watery arms race. Nature 411, 745. Pistocchi, R., 2007. Influence of temperature, salinity and nutrient limitation on Stolte, W., Panosso, R., Gisselson, L.-A˚ ., Grane´ li, E., 2002. Utilization efciency yessotoxin production and release by the dinoflagellate Protocetatium reticu- of nitrogen associated with riverine dissolved organic carbon (>1kDa)by latum in batch-cultures. Harmful Algae 6, 707–717. two toxin-producing phytoplankton species. Aquat. Microb. Ecol. 29, 97–105. Guerrini, F., Pezzolesi, L., Feller, A., Riccardi, M., Ciminiello, P., Dell’Aversano, C., Taniyama, S., Arakawa, O., Terada, M., Nishio, S., Takatani, T., Mahmud, Y., Noguchi, Tartaglione, L., Iacovo, E.D., Fattorusso, E., Forino, M., Pistocchi, R., 2009. T., 2003. Ostreopsis sp., a possible origin of palytoxin (PTX) in parrotfish Scarus Comparative growth and toxin profile of cultured Ostreopsis ovata from the ovifrons. Toxicon 42, 29–33. Tyrrhenian and Adriatic Seas. Toxicon 55, 211–220. Totti, C., Accoroni, S., Cerino, F., Cucchiari, E., Romagnoli, T., 2010. Ostreopsis ovata Guillard, R.R.L., Ryther, J.H., 1962. Studies of marine planktonic diatoms. 1. Cyclo- bloom along the Conero Riviera (northern Adriatic Sea): relationships with tella nana Hustedt and Detonula confervacea (Cleave.). Gran. Can. J. Microbiol. environmental conditions and substrata. Harmful Algae 9, 233–239. 8, 229–239. Van Dolah, F.M., 2000. Marine algal toxins: origins, health effects, and their Guisande, C., Frango´ pulos, M., Maneiro, I., Vergara, A.R., Riveiro, I., 2002. Ecological increased occurrence. Environ. Health Perspect. 108, 133–141. advantages of toxin production by the dinoflagellate Alexandrium minutum Vila, M., Garce´s, E., Maso, M., 2001. Potentially toxic epiphytic dinoflagellate under phosphorus limitation. Mar. Ecol. Prog. Ser. 225, 169–176. assemblages on macroalgae in NW Mediterranean. Aquat. Microb. Ecol. 26, Hales, S., Weinstein, P., Woodward, A., 1999. Ciguatera fish poisoning, El Nino, and 51–60. sea surface temperature. Ecosyst. Health 5, 20–25. Xu, N., Duan, S., Li, A., Zhang, C., Cai, Z., Hu, Z., 2010. Effects of temperature, salinity Hallegraeff, G.M., 2003. Harmful algal blooms: a global overview. In: Hallegraeff, and irradiance on the growth of the harmful dinoflagellate Prorocentrum G.M., Anderson, D.M. (Eds.), Manual on Harmful Marine Microalgae. UNESCO, donghaiense Lu. Harmful Algae 9, 13–17. Paris, pp. 25–49. Yasumoto, T., Nakajima, I., Bagnis, R., Adachi, R., 1977. Finding of a dinoflagellate as a Hwang, D.F., Lu, Y.H., 2000. Influence of environmental and nutritional factors on likely culprit of ciguatera. Bull. Jpn. Soc. Sci. Fish. 43 (8), 1021–1026. growth, toxicity, and toxin prole of dinoflagellate Alexandrium minutum. Tox- Yasumoto, T., Seino, N., Murakami, Y., Murata, M., 1987. Toxins produced by benthic icon 38, 1491–1503. dinoflagellates. Biol. Bull. 172, 128–131.