Temporal and Spatial Distribution of the Toxic Epiphytic Dinoﬂagellate Ostreopsis cf. ovata in the Coastal Waters o ﬀ Jeju Island, Korea

: The temporal and spatial distribution of the toxic epiphytic dinoﬂagellate Ostreopsis cf. ovata was investigated o ﬀ the Jeju coastal waters, Korea, from July 2016 to January 2019. The results showed that the presence of Ostreopsis cf. ovata in 184 macroalgae was 79.3%, and it was more frequently attached to red algae and brown algae than to green algae. The abundance of Ostreopsis cf. ovata as determined by quantitative real-time polymerase chain reactions (qPCR) and microscopic analysis was 4–3204 cells g − 1 , and the maximum abundance observed in September 2018, when the water temperature was 24.4 ◦ C. The abundance was higher in summer and autumn than in spring and winter. Spatially, high abundance was observed in autumn on the northern coast of Jeju Island and, in summer, in the southern and eastern coastal waters. The water temperature of Jeju coastal waters in winter remained higher than 15 ◦ C, and this species could be overwintering in the Jeju waters. Therefore, further monitoring and research are needed to evaluate the proliferation of Ostreopsis cf. ovata , which contains a novel toxin with unidentiﬁed e ﬀ ects on humans.


Introduction
The genus Ostreopsis is one of the toxic epiphytic marine dinoflagellates whose geographical distribution has been expanding from tropical to temperate region [1][2][3]. The genus Ostreopsis produces the palytoxin (PLTX) complex which is one of the most potent biotoxins [4][5][6] and causes human food poisoning by bioconcentration in seafood. Four Ostreopsis species-O. cf. ovata, O. lenticularis, O. mascarenensis, and O. siamensis-produce PLTX-like compounds. Blooms of PLTX-producing algae cause serious problems in aquatic ecosystem and fisheries and are dangerous to human health [7,8]. Recently, global warming has been shown to affect the marine ecosystem structure and promote outbreaks of harmful algal blooms with the consequence production of PLTX caused by Ostreopsis species [9]. Especially, Ostreopsis cf. ovata produces a wide range of PLTX-like compounds and ovatoxins (ovatoxin, a~f) [10][11][12], among which ovatoxin is a non-PLTX compound recently reported as a marine toxin affecting shrimp [13].
In previous decades, Ostreopsis blooms were common in temperate water, mostly in the Mediterranean Sea during summer to autumn and other temperate areas (Table 1). It has also been reported that this genus has high tolerance of a wide range of temperature. [14].
Blooms of Ostreopsis species cause serious human health problems via inhalation of sea water droplets containing Ostreopsis cells or aerosolized toxins [6,15,16]. Ostreopsis can create and proliferate mucus floating clusters. These clusters float on the water surface, and toxins released from the cells

Sample Collection and Treatment
Macroalgae samples were collected by scuba divers from July 2016 to January 2019 within 10 m of water depth. Four different macroalgae species were collected at the four sampling sites (Jeju-north, Gosan-west, Wimi-south, Seongsan-east) off Jeju Island (Figure 1). Environmental factors, such as water temperature and salinity were measured in the surface water using a YSI Professional plus (Xylem, Inc., Rye Brook, USA). The collected macroalgae were morphologically identified using an illustrated book [46]. Epiphytic dinoflagellates attached to the macroalgae were collected as described by Baek [47]. The collected macroalgae were transferred into a 1-L bottle, which as filled with filtered seawater, and then shaken vigorously approximately 200 times to detach the epiphytic dinoflagellates. After filtering the sample through a 100-µm mesh to separate the macroalgae from seawater, the macroalgae were rapidly frozen on dry ice and then transported to the laboratory. A 300-mL sample containing epiphytic dinoflagellates was fixed with formalin (final concentration 1%), and another 300-mL sample was transferred to the laboratory as a live sample for establishing strains of epiphytic dinoflagellates.

Sample Collection and Treatment
Macroalgae samples were collected by scuba divers from July 2016 to January 2019 within 10 m of water depth. Four different macroalgae species were collected at the four sampling sites (Jejunorth, Gosan-west, Wimi-south, Seongsan-east) off Jeju Island (Figure 1). Environmental factors, such as water temperature and salinity were measured in the surface water using a YSI Professional plus (Xylem, Inc., Rye Brook, USA). The collected macroalgae were morphologically identified using an illustrated book [46]. Epiphytic dinoflagellates attached to the macroalgae were collected as described by Baek [47]. The collected macroalgae were transferred into a 1-L bottle, which as filled with filtered seawater, and then shaken vigorously approximately 200 times to detach the epiphytic dinoflagellates. After filtering the sample through a 100-μm mesh to separate the macroalgae from seawater, the macroalgae were rapidly frozen on dry ice and then transported to the laboratory. A 300-mL sample containing epiphytic dinoflagellates was fixed with formalin (final concentration 1%), and another 300-mL sample was transferred to the laboratory as a live sample for establishing strains of epiphytic dinoflagellates. To obtain a DNA sample, 20-50 mL seawater sample obtained by shaking the macroalgae was filtered through a GF/C filter and the filter paper was rapidly frozen on dry ice. Macroalgae were transported to the laboratory and weighed after identification.

Fluorescence Microscopy
The fixed samples were concentrated 5-6 times for microscopic observation. The fixed 300 mL of sample was left still overnight to allow the cells to sink, and the supernatant was removed to adjust the final volume to 50-60 mL. In addition, the frozen macroalgae were weighed fresh after remove excess seawater. A solution of Calcofluor (Sigma-Aldrich, St. Louis, MO, USA) was added to the fixed sample at a final concentration of 10 μg/mL and stained for 1-2 min in the dark. The cells were counted in a Sedgwick-Rafter chamber (SPI Supplies, West Chester, PA, USA) at 100X magnification using an epi-fluorescence microscope (BX 53, Olympus, Tokyo, Japan). The cells were photographed with a DP73 digital camera system (Olympus, Tokyo, Japan).
The final concentration of epiphytic dinoflagellates attached to each macroalgae was calculated by calculating the adhesion density per gram of fresh weight of macroalgae (cells g −1 ), as described by Ishikawa and Kurashima [48]. To obtain a DNA sample, 20-50 mL seawater sample obtained by shaking the macroalgae was filtered through a GF/C filter and the filter paper was rapidly frozen on dry ice. Macroalgae were transported to the laboratory and weighed after identification.

Fluorescence Microscopy
The fixed samples were concentrated 5-6 times for microscopic observation. The fixed 300 mL of sample was left still overnight to allow the cells to sink, and the supernatant was removed to adjust the final volume to 50-60 mL. In addition, the frozen macroalgae were weighed fresh after remove excess seawater. A solution of Calcofluor (Sigma-Aldrich, St. Louis, MO, USA) was added to the fixed sample at a final concentration of 10 µg/mL and stained for 1-2 min in the dark. The cells were counted in a Sedgwick-Rafter chamber (SPI Supplies, West Chester, PA, USA) at 100X magnification using an epi-fluorescence microscope (BX 53, Olympus, Tokyo, Japan). The cells were photographed with a DP73 digital camera system (Olympus, Tokyo, Japan).
The final concentration of epiphytic dinoflagellates attached to each macroalgae was calculated by calculating the adhesion density per gram of fresh weight of macroalgae (cells g −1 ), as described by Ishikawa and Kurashima [48].

Culture, DNA Extraction, and PCR Amplification
To establish a single cell strain of O. cf. ovata, the live sample transferred to the laboratory was placed in a six-well plate, and single cell isolation was performed under a dissecting microscope (SZX10, Olympus, Tokyo, Japan). Clonal cultures of O. cf. ovata were established by two serial single-cell isolations. After sufficient growth, the cells were transferred into a 30-mL flask. Once dense cultures of O. cf. ovata were obtained, the cells were transferred into 200-mL PC bottles containing fresh f/2 seawater media. The DNA sequences of these cells were analyzed when the concentration of O. cf. ovata was more than 10 5 cells ml −1 . The dense culture (1 mL) was centrifuged and the pellet was used for DNA extraction. DNA extraction was carried out with an AccuPrep Genomic DNA Extraction Kit (BIONEER, Daejeon, Korea), and then the SSU-ITS-LSU section of rDNA was amplified using universal eukaryotic primers [49]. The obtained rDNA sequence were confirmed by BLAST (https://blast.ncbi.nlm.nih.gov/ Blast.cgi, NCBI).
To obtain DNA from field samples, 20-50 mL of samples obtained by shaking the macroalgae which collected from the sampling sites in the Jeju coastal waters were filtered with Whatman GF/C (pore size 1.2 µm), and genomic DNA (gDNA) was extracted with an AccuPrep Genomic DNA Extraction Kit according to the manufacturer's protocol.

Culture, DNA Extraction, and PCR Amplification
To establish a single cell strain of O. cf. ovata, the live sample transferred to the laboratory was placed in a six-well plate, and single cell isolation was performed under a dissecting microscope (SZX10, Olympus, Tokyo, Japan). Clonal cultures of O. cf. ovata were established by two serial singlecell isolations. After sufficient growth, the cells were transferred into a 30-mL flask. Once dense cultures of O. cf. ovata were obtained, the cells were transferred into 200-mL PC bottles containing fresh f/2 seawater media. The DNA sequences of these cells were analyzed when the concentration of O. cf. ovata was more than 10 5 cells ml −1 . The dense culture (1 mL) was centrifuged and the pellet was used for DNA extraction. DNA extraction was carried out with an AccuPrep Genomic DNA Extraction Kit (BIONEER, Daejeon, Korea), and then the SSU-ITS-LSU section of rDNA was amplified using universal eukaryotic primers [49]. The obtained rDNA sequence were confirmed by BLAST (https://blast.ncbi.nlm.nih.gov/ Blast.cgi, NCBI).
To obtain DNA from field samples, 20-50 mL of samples obtained by shaking the macroalgae which collected from the sampling sites in the Jeju coastal waters were filtered with Whatman GF/C (pore size 1.2 μm), and genomic DNA (gDNA) was extracted with an AccuPrep Genomic DNA Extraction Kit according to the manufacturer's protocol.  Test species specificity of Ostreopsis cf. ovata by quantitative polymerase chain reaction (PCR). A specificity test for the designed primer and probe was performed by quantitative PCR assay with 12 strains. A specificity test for the designed primer and probe was performed by quantitative PCR assay with 12 strains.

qPCR
The qPCR assay was conducted on a PCRmax Eco 48 real-time PCR system (PCR max, Stone, UK) using O. cf. ovata forward (5 -GGCCATTCCTAAGGACATCA−3 ), O. cf. ovata reverse (5 -TGGCCATATAC AGCATGTTGAC−3 ), and O. cf. ovata probe (5 -ATCATGCATTGTGTGAGTGTGTGATGT−3 ) labeled at the 5 and 3 ends with the fluorescent dyes 6-FAM and BHQ−1. The final volume of the reaction mixtures for qPCR amplification was 20 µL and containing 10 µL of qPCRBIO Probe Mix No-ROX (2X) (PCR Biosystems, London, England), 1 µL of 10 µM each forward/reverse primer, 0.5 µL of 10 µM probe, 4.5 µL of UltraPure™ DNAse/RNAse-Free Distilled Water (Invitrogen, Carlsbad, CA, USA), and 3 µL of DNA template. The reagents were loaded into an Eco 48-well plate on the Eco 48 sample loading dock, and the plate was covered with an Eco 48 Adhesive Plate Seal to block contaminants such as powder, seam, or dust. After centrifuging the plate for 90 s, thermal cycling was carried out under the following conditions: 95 • C for 3 min of initial denaturation followed by 40 cycles of amplification for 10 s at 95 • C and 30 s at 60 • C. The standard curve was constructed by comparing the quantification cycle (Ct) value with the raw fluorescent signal measured by qPCR to known cell concentrations of O. cf. ovata ( Figure 3). Approximately 30,000 O. cf. ovata cells cultured under laboratory conditions were filtered with Whatman (GF/C) filters, and gDNA eluted in a final volume of 100 µL for each sample was extracted using an AccuPrep Genomic DNA Extraction Kit (Bioneer, Seoul, Korea). A serial dilution to prepare samples containing 3000, 1000, 300, 100, 30, and 10 cells was performed by adding sterile distilled water (Invitrogen, Calsbad, USA), and the standard curve was determined by the qPCR method as described above. The Ct value for each field sample was also investigated under thermal cycling conditions, described above, and the relatively quantified cell concentrations of O. cf. ovata were calculated using the standard curve. A template containing sterile distilled water was used as a negative control, and one of the samples used for standard curve construction was measured as a positive control.  (Figure 3). Approximately 30,000 O. cf. ovata cells cultured under laboratory conditions were filtered with Whatman (GF/C) filters, and gDNA eluted in a final volume of 100 μL for each sample was extracted using an AccuPrep Genomic DNA Extraction Kit (Bioneer, Seoul, Korea). A serial dilution to prepare samples containing 3000, 1000, 300, 100, 30, and 10 cells was performed by adding sterile distilled water (Invitrogen, Calsbad, USA), and the standard curve was determined by the qPCR method as described above. The Ct value for each field sample was also investigated under thermal cycling conditions, described above, and the relatively quantified cell concentrations of O. cf. ovata were calculated using the standard curve. A template containing sterile distilled water was used as a negative control, and one of the samples used for standard curve construction was measured as a positive control.

Environmental Conditions
During the sampling period, the water temperature ranged from 11.3 °C in Gosan in January to 26.3 °C in Gosan in September ( Table 2). The average temperatures in summer and autumn were similar (~ 22 °C) and were similar in winter and spring (15-16 °C). Salinity ranged from 22.7 to 36.1, with average value of 31.4. The salinity was relatively lower in summer in the harbor area (Jeju and Wimi), and it was affected by inland discharge after rainfall.

Environmental Conditions
During the sampling period, the water temperature ranged from 11.3 • C in Gosan in January to 26.3 • C in Gosan in September ( Table 2). The average temperatures in summer and autumn were similar (~22 • C) and were similar in winter and spring (15-16 • C). Salinity ranged from 22.7 to 36.1, with average value of 31.4. The salinity was relatively lower in summer in the harbor area (Jeju and Wimi), and it was affected by inland discharge after rainfall.

Collected Macroalgae Species and Attachment Rates
We collected 184 macroalgae samples from the four sampling sites off Jeju coastal waters from July 2016 to January 2019. In total, 26 species of macroalgae were collected, including green algae, red algae, and brown algae. O. cf. ovata was retrieved in 146 out of 184 samples, giving a 79.3% attachment rate; these values were 51% in red algae, 42% in brown algae, and 7% in green algae. The highest attachment rates for red algae were found for Gelidium amansii and Corallina pilulifera, brown algae for Dictyopteris undulata and Ecklonia cava, and green algae for Ulva conglobate and Codium coactum (Figure 4). The lowest attachment rate (33.3%) was observed in March 2017, and O. cf. ovata was found on all macroalgae samples in July 2018.

Collected Macroalgae Species and Attachment Rates
We collected 184 macroalgae samples from the four sampling sites off Jeju coastal waters from July 2016 to January 2019. In total, 26 species of macroalgae were collected, including green algae, red algae, and brown algae. O. cf. ovata was retrieved in 146 out of 184 samples, giving a 79.3% attachment rate; these values were 51% in red algae, 42% in brown algae, and 7% in green algae. The highest attachment rates for red algae were found for Gelidium amansii and Corallina pilulifera, brown algae for Dictyopteris undulata and Ecklonia cava, and green algae for Ulva conglobate and Codium coactum (Figure 4). The lowest attachment rate (33.3%) was observed in March 2017, and O. cf. ovata was found on all macroalgae samples in July 2018.

Morphological Features of Ostreopsis cf. ovata
The cell shape of the O. cf. ovata had a typical oval to tear shape in the apical view and a thecal plate, which was dyed with calcofluor for easy identification an under epifluorescence microscope ( Figure 5). The dorsoventral length and width of these cells were approximately 30-40 μm and 20-30 μm, respectively. The ratio of dorsoventral length to width was 1.2-1.5.

Morphological Features of Ostreopsis cf. ovata
The cell shape of the O. cf. ovata had a typical oval to tear shape in the apical view and a thecal plate, which was dyed with calcofluor for easy identification an under epifluorescence microscope ( Figure 5). The dorsoventral length and width of these cells were approximately 30-40 µm and 20-30 µm, respectively. The ratio of dorsoventral length to width was 1.2-1.5.

Relationship between Microscopic Analysis and qPCR Results
To correctly identify and quantify O. cf. ovata, we combined the results obtained by microscopic analysis with those found by qPCR quantification. The data collected by microscopic analysis possibly included other Ostreopsis species, whereas qPCR method quantifies only the target species. A correspondence and high correlation was found between the results ( Figure 6). The detection limit of microscopic analysis was low as 50 cells g −1 fw macroalgae. In the range of low cell numbers, positive results were found by qPCR, even when no Ostreopsis cells were found by microscopy. Some data obtained by qPCR showed false positive or overestimated when no or low numbers of cells were detected by microscope. We excluded several data showing over 5000 cells g −1 fw macroalgae obtained by qPCR as false positives after quick microscopic observation.

Relationship between Microscopic Analysis and qPCR Results
To correctly identify and quantify O. cf. ovata, we combined the results obtained by microscopic analysis with those found by qPCR quantification. The data collected by microscopic analysis possibly included other Ostreopsis species, whereas qPCR method quantifies only the target species. A correspondence and high correlation was found between the results ( Figure 6). The detection limit of microscopic analysis was low as 50 cells g −1 fw macroalgae. In the range of low cell numbers, positive results were found by qPCR, even when no Ostreopsis cells were found by microscopy. Some data obtained by qPCR showed false positive or overestimated when no or low numbers of cells were detected by microscope. We excluded several data showing over 5000 cells g −1 fw macroalgae obtained by qPCR as false positives after quick microscopic observation.

Relationship between Microscopic Analysis and qPCR Results
To correctly identify and quantify O. cf. ovata, we combined the results obtained by microscopic analysis with those found by qPCR quantification. The data collected by microscopic analysis possibly included other Ostreopsis species, whereas qPCR method quantifies only the target species. A correspondence and high correlation was found between the results ( Figure 6). The detection limit of microscopic analysis was low as 50 cells g −1 fw macroalgae. In the range of low cell numbers, positive results were found by qPCR, even when no Ostreopsis cells were found by microscopy. Some data obtained by qPCR showed false positive or overestimated when no or low numbers of cells were detected by microscope. We excluded several data showing over 5000 cells g −1 fw macroalgae obtained by qPCR as false positives after quick microscopic observation.

Temporal and Spatial Distributions of Ostreopsis cf. ovata
The abundance of the O. cf. ovata during the sampling period ranged from 4 cells g −1 to 3204 cells g −1 (Figure 7). In July 2016, the abundance at all four sites were very low, despite being in the summer season; however, in July 2017, the abundance increased to approximately 3000 cells g −1 and then decreased during winter (December and January). The abundance increased again in summer, with the maximum abundance observed in September 2018 in the Jeju area (3204 cells g −1 ). The abundance of O. cf. ovata fluctuated by season; it increased from summer to autumn and decreased from winter to spring. The abundance of O. cf. ovata during spring ranged from 119 to 319 cells g −1 , with 32-2825 cells g −1 in summer, 4-3204 cells g −1 in autumn, and 24-391 cells g −1 in winter. The averaged abundance was lowest in winter (139.3 cells g −1 ) and highest in autumn (704.4 cells g −1 ), with 183.7 cells g −1 in spring, and 462.2 cells g −1 in summer. The abundance of O. cf. ovata increased after the water temperature reached 20 • C.
The highest abundance was observed at the northern coast of Jeju Island area (Figure 7). In this area, the highest abundance was recorded in September 2018 and ranged 32-3204 cells g −1 . The abundance ranged from 54 to 1.560 cells g −1 in the western coastal area (Gosan), 24-2825 cells g −1 in the eastern coastal area (Seongsan), and 4-504 cells g −1 in the southern coastal area (Wimi). The average abundance in the Wimi area (southern coast) was lowest at 181.5 cells g −1 . The average abundances at Seongsan (east) and Gosan (west) were 402.8 and 481.4 cells g −1 , respectively, and was high at the northern coastal area, at 492.6 cells g −1 . The abundance of O. cf. ovata differed both temporally and spatially; it was high in the west and north in autumn, and east in summer. The southern coastal area (Wimi) showed low abundance throughout the year.

Temporal and Spatial Distributions of Ostreopsis cf. ovata
The abundance of the O. cf. ovata during the sampling period ranged from 4 cells g −1 to 3204 cells g −1 (Figure 7). In July 2016, the abundance at all four sites were very low, despite being in the summer season; however, in July 2017, the abundance increased to approximately 3000 cells g −1 and then decreased during winter (December and January). The abundance increased again in summer, with the maximum abundance observed in September

Discussion
Ostreopsis species around Jeju Island has been reported since 2010 [44,50]. Moreover, Hwang et al. [13,51] reported the novel cytotoxic non-PLTX compounds Ostreol A and B in the strain isolated

Discussion
Ostreopsis species around Jeju Island has been reported since 2010 [44,50]. Moreover, Hwang et al. [13,51] reported the novel cytotoxic non-PLTX compounds Ostreol A and B in the strain isolated from Jeju Island, which have not both been reported in any other strain. Ostreol A showed very high toxicity in an in vitro cytotoxicity test against Artemia, but how it affects humans remains unclear.
We quantified O. cf. ovata over three years in Jeju coastal waters and determined the temporal and spatial distributions. We observed a high abundance in the northern and western coastal areas of Jeju Island, which has popular beaches and is frequently visited by tourists. We found that O. cf. ovata proliferated when the water temperature was >20 • C, when the beaches are open to tourists in summer. Mangialajo et al. [16] reported the proliferation of O. ovata along the urbanized coast of Italy in 2005, which caused human health problems with similar symptoms after exposure to marine aerosols ( Table 2). At that time, the abundance of O. ovata was 2.5 × 10 6 cells g −1 and approximately 200 received local first aid. In this research, the maximum abundance of O. cf. ovata was 3.2 × 10 3 cells g −1 , which is not as high as in the Mediterranean Sea, however O. cf. ovata from Jeju area contains another toxin that has not been reported in the strain from the Mediterranean Sea, the potential toxicity of which remains unclea In this study, various macroalgae species were collected because macroalgae appear to be the preferred substrate for O. cf. ovata. Research by Tindall and Morton showed that the maximum density of epiphytic species attached to macroalgae was generally 10 2 -10 4 cells g −1 [52], and these species tended to attach to Rhodophyta and Phaeophyta at high densities, particularly to Halopteris scoparia, which showed the highest recorded density (5.9-10 5 cells g −1 ) [31]. Our results showed similar tendency to the results of these studies. The three-dimensional flexibility and high surface area of the macroalgae can explain the preference of O. cf. ovata for this substrate rather than coral or sand. Additionally, the detection of epiphytic dinoflagellates in the sampling area follows a clear seasonal pattern [53].
The abundance of O. cf. ovata widely varies by season, increasing from summer to autumn and then decreasing from winter to spring, revealing a relationship with water temperature. The average water temperature in Jeju coastal waters in summer and autumn was 22 • C and in winter and spring was 15 • C. In our laboratory test, O. cf. ovata isolated from Jeju waters showed optimal growth at 15-20 • C and did not survive below 10 • C (unpublished data). The optimal salinity of O. cf. ovata ranged from 30-40, and the salinity range of the study areas was 22.7-36.1, indicating that when the salinity is low, the growth of O. cf. ovata could depressed. In the present study, O. cf. ovata was abundant when the water temperature was 22-25 • C and salinity was 31-34 ( Figure 8). These conditions were the same as the optimal conditions determined in the laboratory growth experiment (unpublished data). In Jeju Island, submarine groundwater discharge in a coastal area was reported [54], which affected the salinity and inorganic nutrient concentration. Jauzein et al. [55] reported that O. cf. ovata has a high uptake ability and availability of nitrate when the nitrate concentration was high. In addition, Lee and Kim [56] suggested that submarine groundwater discharge can lead to the blooming of monospecific dinoflagellates by increasing nitrate concentrations. Therefore, submarine groundwater discharge in the Jeju coastal area, which has a high nutrient concentration, may promote the proliferation of O. cf. ovata but may also possibly play an inhibitory role by lowering salinity.
In conclusion, we found that O. cf. ovata was distributed around the Jeju coastal area and abundant when the water temperature was over 20 • C. This species was abundant in the northern and western coastal areas and survived over the winter in the Jeju coastal waters. The environmental conditions of the Jeju coastal waters may promote the growth of O. cf. ovata. Further monitoring and research are needed to evaluate the proliferation of O. cf. ovata, which contains a novel toxin with unidentified effects on humans.

Conflicts of Interest:
The authors declare no conflict of interest.

Conflicts of Interest:
The authors declare no conflict of interest.