Taxonomy and Physiology of Oxyrrhis marina and Oxyrrhis maritima in Korean Waters

: The genus Oxyrrhis is a heterotrophic dinoﬂagellate found in diverse marine environments. Oxyrrhis spp. have received attention owing to their ecological and industrial importance, high lipid contents, and docosahexaenoic acid formation. To the best of our knowledge, contrary to O. marina , ecophysiological characterization studies on O. maritima have not yet been reported. Therefore, we investigated the taxonomy and ecophysiology of four strains of O. marina from coastal waters and two strains of O. maritima from the littoral tidepool waters of Korea. Based on phylogenetic trees constructed using internal transcribed spacer ribosomal DNA (ITS rDNA) and SSU rDNA of dinoﬂagellates, the clade of all four O. marina strains was divergent from that of the two O. maritima strains. We measured the growth rates of both species at various water temperatures (10–36 ◦ C), salinities (5–90), and light intensities (0–100 µ E · m − 2 · s − 1 ). The lowest ( O. marina and O. maritima : 10 ◦ C) and highest temperatures ( O. marina : <35 ◦ C, O. maritima : >35 ◦ C) revealed that O. maritima has more tolerance to high salinity. This study provides a basis for understanding the ecophysiology of O. marina and O. maritima and their population dynamics in marine ecosystems.


Introduction
Heterotrophic dinoflagellates (HTDs) are of great importance in the marine food web because they coexist as predators and prey for other species [1]. In this study, we focused on HTDs Oxyrrhis marina (O. marina) and Oxyrrhis maritima (O. maritima). There are advantages to using Oxyrrhis as a model for other HTD natural samples, to identify and manipulate in experiments, and hence, to study the model of evolution and exploitation of this species. Oxyrrhis is a widespread, free-living, and ecologically important HTD [2][3][4][5] that has been utilized as an important model organism for a broad range of studies [6][7][8].
The genus Oxyrrhis is commonly found in marine and brackish nearshore waters, tide pools, and salterns. However, it explicitly excluded dinokaryotes in common dinoflagellate classification schemes [9]. Unlike most dinoflagellates, which either have a longitudinal and a transverse flagellum emerging from the sulcus and the cingulum, respectively, or both flagella growing from the apical area (Prorocentrales) [4], Oxyrrhis have both flagella emerging from the ventral side. Therefore, the phylogenetic position of O. marina remains controversial. While some of the morphological and cytological studies support its basal position in the dinoflagellate classification of the higher ranks [10], others infer a highly derived position within the order of Gonyaulacales, and the majority of recent studies support O. marina as an ancestral dinoflagellate [11].
Morphological studies have raised disputes on whether the genus Oxyrrhis contains multiple species (O. marina, O. maritima van Meel 1969, O. phaeocysticola Scherffel, 1900, and O. tentaculifera Conrad, 1939) or only one species (O. marina) [4,[12][13][14][15][16][17] (Table 1). Recently, molecular phylogenetic studies have favored the notion that two sibling species exist in this genus. Lowe et al. suggested that this genus consists of two species, O. marina and O. maritima, and each species consists of two clades [4]. However, the existing data are scattered, despite the increasing number of studies on O. marina [11]. A myriad of ecological, physiological, and genetic studies have been conducted without accurate identification of the species. Thus, we established four strains of O. marina and two strains of O. maritima by isolating and culturing from six different locations in the Korean environmental water samples. The morphological features, phylogeny, and nuclear ribosomal DNA were analyzed based on the six strains of Oxyrrhis spp.
The growth rate of heterotrophic protists is affected by various abiotic factors, such as salinity [18,19], temperature [20], and light intensity [21]. Thus, the physiological study of each species is a critical step in understanding the structure and function of the ocean ecosystem. There have been many studies on the effects of temperature, salinity, and light conditions on the growth rates of O. marina [22][23][24]. However, there have been no studies on the ecophysiological characterizations of O. maritima. Therefore, it is worthwhile to explore this topic. Moreover, the difference between the ecophysiological characterizations of these two Oxyrrhis spp. is poorly understood. Thus, we measured and compared the growth rates of O. marina and O. maritima.
In the present study, we measured the growth rates of O. marina and O. maritima at various water temperatures (10-36 • C), salinities (5-90), and light intensities (0-100 µE·m −2 ·s −1 ). In particular, we tested whether O. marina could grow in darkness when suitable prey was provided. Furthermore, we determined the lowest and highest temperatures and salinities at which O. marina and O. maritima can survive or grow. The results of the present study provide a basis for understanding the ecophysiology of O. marina and O. maritima and their population dynamics in marine ecosystems.

Collection, Isolation, and Culturing of Oxyrrhis spp.
Water samples were collected using a syringe tubing and water sampling bucket from the surface of six different stations at littoral tide pools in Jeju Island and coastal waters in Korea (Shiwha, Gunsan, Masan, and Karorim; Table 2; Figure 1) between May 2001 and May 2013. Water temperature and salinity measured during sampling for O. marina were 13.5-19.7 • C and 27.7-33 • C, respectively, but no physical information was available in the case of O. maritima. The sample was filtered through a 154-µm Nitex mesh screen and placed in six-well tissue culture plates. A clonal culture was established following two serial single-cell isolations, and the dried yeast (Saccharomyces cerevisiae) obtained from Red Star (Lesaffre Yeast Corporation, Milwaukee, WI, USA) were provided as food. As the culture volume increased, cells were transferred to 32-, 250-, and 500-mL polycarbonate (PC) bottles. The bottles were placed on a shelf at 20 • C under continuous illumination of 10 µE·m −2 ·s −1 provided by a cool white fluorescent light source. As the cultures became dense, approximately every 2-3 d, they were transferred to new 500-mL PC bottles containing dried yeast as prey.

Microscopic Observations
The morphologies of O. marina and O. maritima were examined using light microscopy. The length and width of live vegetative flagellated cells were measured using an image analysis system for processing images taken with a compound microscope (BX53 microscope, Olympus, Japan).

DNA Extraction, PCR Amplification, and Sequencing
Three to five cells of O. marina and O. maritima from a clonal culture were transferred to a 0.2-mL PCR tube containing 38.75 µL distilled water. To lyse the cell membranes before PCR analysis, the tube was frozen at −72 • C for 1-3 min and then thawed. The final mixture (50 µL) was vortexed, and PCR was performed in a thermal cycler (Eppendorf AG, Mastercycler ® ep, model 5341, Hamburg, Germany). The final concentrations of PCR products were: 5 µL of 10X F-Star Taq buffer, 1 µL of 10 mM of dNTP mix, 0.25 µL of 5 U/µL BioFACT™ F-Star Taq DNA polymerase (BioFACT Co., Ltd., Daejeon, Korea), and 0.02 µM of each primer (Table S1) and a final volume of 50 µL was made. The PCR conditions were as follows: one activation step at 95 • C for 2 min, followed by 40 cycles at 95 • C for 20 s, the selected annealing temperature for 40 s, and 72 • C for 1 min, and a final elongation step at 72 • C for 5 min. The annealing temperature was adjusted depending on the primers used according to the manufacturer's instructions. Positive and negative controls were used for all the amplification reactions. The purity of the amplicons was checked by performing electrophoresis of 5 µL of PCR products mixed with 1 µL of red fluorescent reagent (Redstar, Daejeon, Korea) using a 0.9% agarose gel at 80 V and observing it under a UV lamp to ensure that a single product was formed. Products containing a single band were then purified using an AccuPrep ® PCR purification kit (Bioneer Corp., Daejeon, Korea) according to the manufacturer's instructions. The purified amplicon was sent to Biomedic Corporation (Gyeonggi, Korea) for sequencing using an ABI PRISM ® 3700 DNA Analyzer (Applied Biosystems, Foster City, CA, USA). Each portion of the target DNA was independently sequenced three times in both directions using primer pairs identical to those used for DNA amplification. These sequences were aligned using the ContigExpress alignment program (InforMax, Frederick, MD, USA) to remove low-quality regions and to assemble the individual sequence reads.

Phylogenetic Analysis
Phylogenetic analyses were performed based on alignments of the partial 3-end small subunit (SSU) and internal transcribed spacer region (ITS1-5.8S-ITS2) sequences. All sequence regions are located in the nuclear genome. Analyses included sequences obtained from GenBank and the present study (Table S2). Multiple sequences were aligned automatically using MEGA-X native implementation of ClustalW [25], and further aligned manually to refine the alignments. Two data sets were analyzed as follows: (1) SSU sequences only, (2) ITS1-5.8S-ITS2 sequences only. SSU sequences were used to investigate the systematic distance between species. Meanwhile, ITS1-5.8S-ITS2 sequences were used to determine intraspecific genetic diversity between Oxyrrhis species. The final alignment of the SSU rDNA contained 19 dinoflagellate species, plus Perkinsus, Apicomplexans, Ciliate, and Stramenopiles sequences that were used as outgroups. The final alignment of the ITS1-5.8S-ITS2 rDNA contained 33 Oxyrrhis sequences, plus Perkinsus, Apicomplexans, Ciliate, and Stramenopiles sequences that were used as outgroups. Two main model-based methods were used to infer phylogenetic relationships: maximum likelihood (ML) and Bayesian inference based on the likelihood function. ML analysis was performed with the RAxML 7.0.3 program [26] using the default GTR + G model. Tree likelihoods were estimated using a heuristic search with 200 random additional sequence replicates and tree bisection and reconnection branch swapping. ML bootstrapping with 1000 replications was also conducted. Bayesian analysis was performed with MrBayes v.3.1.2 [27,28] using the GTR + G + I model according to the model test results. The program was set to operate with a gamma distribution (+G) with five rate categories and by assuming that a certain fraction of sites is evolutionarily invariable(+I), and four Monte Carlo Markov chains (MCMC) starting from a random tree. In total, 2,000,000 generations were calculated with trees sampled every 100 generations and with a prior burn-in of 5000 generations. To determine the actual initial predator densities (cells mL −1 ) at the beginning of the experiment and after incubation, 5 mL aliquots were removed from each bottle and fixed with 5% Lugol's solution, and all O. marina or O. maritima cells and (or >300) prey cells from the triplicate SRCs were enumerated. Before taking subsamples, the conditions of O. marina, O. maritima, and prey were assessed under a dissecting microscope.

Physiology
The specific growth rate of O. marina or O. maritima, µ (d −1 ), was calculated as follows: where H 0 is the initial concentration (cells mL −1 ) of O. marina or O. maritima, and H t is the final concentration after time t (d −1 ).

Morphology
Under a light microscope (LM), the living cell was subovoid, asymmetrical posteriorly; girdle postmedial, not extending to the dorsal surface; sulcus spreading posteroventrally; flagella midventral; tentacular lobe situated between two flagella, dividing the broad undeveloped ventral sulcus ( Figure 2). It appeared colorless, but had pink pigmentation, which was apparent in concentrated cultures. Therefore, this species is called a "pink pig" because of the pink colored water at a high concentration of the organisms [29].

Morphology of O. maritima
As shown in Figure 3, under LM, living cells were observed to be subovoid, posteriorly asymmetrical; girdle postmedial, not extending to the dorsal surface; sulcus spreading posteroventrally; flagella midventral; tentacular lobe present between two flagella, dividing the broad undeveloped ventral sulcus. It appeared colorless but had pink pigmentation, which was apparent in concentrated cultures. Sequence alignments were performed using the MEGA-X native implementation of ClustalW. Sequence divergence was exceptionally high among some strains (Table 3). We compared the genetic divergence (%), except gaps, of the total sequences of the six strains of Oxyrrhis spp.
The difference among the sequences of the four strains of O. marina was 0.1-0.3%, whereas that between the two strains of O. maritima was only 0.1%. However, the difference between the sequences of O. marina and O. maritima was 17.8-22.5%. In the phylogenetic trees based on the ITS1-5.8s-ITS2 rDNA of dinoflagellates, the clade of all four strains of O. marina belonged to lineage I (clades 1 and 2) and was clearly divergent from that of the two strains of O. maritima that belonged to lineage II (clade 4).  Figure 4 shows the phylogenetic trees constructed based on SSU rDNA sequences from nucleotide DNA to identify the phylogenetic position of Oxyrrhis spp. This confirmed that Oxyrrhis spp. were extremely divergent from core dinoflagellates, such as Alexandrium spp., and Amphidinium spp. (Figure 4). Oxyrrhis branched after Perkinsus marinus, which is between the dinoflagellates and Apicomplexans, with high bootstrap support in Bayesian analysis. Moreover, its branches proved that the Oxyrrhis was positioned at the base of dinoflagellate lineages. The branch length of SSU rDNA sequences of Oxyrrhis was approximately eight times as long as that of P. marinus. In all phylogenetic analyses, Oxyrrhis had two highly different species, and the four strains of O. marina and two strains of O. maritima branched together with high support. We used only Stramenopiles as outgroups, but Apicomplexans and Ciliates were apparently divergent with over 0.99 (Bayesian) probability.

Phylogeny
Furthermore, the phylogenetic tree based on ITS rDNA sequences proved the two highly divergent lineages within O. marina morphospecies ( Figure 5). The two Oxyrrhis lineages divided four clades with high bootstrap support as follows: lineages I and II-1.00 (Bayesian)/100 (ML), clades 1 and 2-0.92 (Bayesian)/75 (ML), clades 3 and 4-0.99 (Bayesian)/77 (ML). According to the pairwise distance analysis, the net distance between clades 1 and 2 was 0.219, and this was 0.309 between clades 3 and 4. Moreover, the longest distance between clades was 0.456 between clades 1 and 3. Clades 1, 2, and 3 had a very short distance within each clade, but clade 4 had a slightly longer distance (0.32) than the others.    Figure 6b). At salinities of 5, 10, and 20, the growth rates of O. maritima were significantly greater than those of O. marina (p < 0.05, one-tailed t-test), whereas at a salinity of 90, the growth rate of O. maritima was significantly lower than that of O. marina (p < 0.05, one-tailed t-test). However, at salinities of 33, 50, and 70, the growth rates of O. marina were not significantly different from those of O. maritima (p > 0.05, two-tailed t-test).
0.05, one-tailed t-test). However, at salinities of 33, 50, and 70, the growth rates of O. marina were not significantly different from those of O. maritima (p > 0.05, two-tailed t-test).
At the light intensities of 0-100 µE·m −2 ·s −1 , the growth rates of dried yeast-fed O. marina did not largely change (i.e., 0.58 d −1 at 0 µE·m −2 ·s −1 to 0.59 d −1 at 100 µE·m −2 ·s −1 ). However, with increasing light intensity from 0 to 100 µE·m −2 ·s −1 , the growth rate of O. maritima decreased from 0.85 d −1 at 0 µE·m −2 ·s −1 to 0.59 d −1 at 100 µE·m −2 ·s −1 (Figure 6c). The growth rates of O. marina and O. maritima were not affected by the light intensity (ANOVA test, p > 0.1, Figure 6c). In addition, the growth rate of O. marina was not significantly different from that of O. maritima (p > 0.1, two-tailed t-test).    (Table 1). Some studies have reported that O. maritima could be distinguished from O. marina by its rounder and larger shape [4]. However, in contrast to previous studies, the species collected in this study showed that O. maritima was longer and sharper than O. marina. In the present study, the lengths and ESD of O. maritima were approximately 14% and 45% smaller than that of O. marina, respectively. Most of the studies have analyzed the genetic diversity of Oxyrrhis strains, either by providing only one genetic or morphological information about O. maritima, or based on data from strains collected long ago. The results of this study confirmed that the cell size of O. maritima is generally smaller than that of O. marina by direct analysis of the species collected within 10 years. Additionally, this study attempted to construct integrated data for species classification of Oxyrrhis strains by linking morphological and genetic information.

Phylogeny
The phylogenetic positions of O. marina are great of importance for understanding cell biology and evolutionary diversity. In particular, the relationships between the Apicomplexa and dinoflagellates have a great impact on disease-causing Apicomplexans and fundamental diversity in eukaryotic cell biology [30]. The dinoflagellate lineage includes abstruse organisms such as syndinians, noctilucoids, Pavilucifera, Perkinsus, and Oxyrrhis. These protists are highly related to Oxyrrhis and their phylogenetic positions are still a matter of debate [31]. Thus, our study conducted the phylogenetic analysis of Oxyrrhis using nuclear SSU rDNA with P. marinus, Noctiluca scintillans, the Apicomplexa (Colphodella), and Ciliates. The relationship with Amoebophrya spp. (Syndinians) was also considered (not shown data), but it was excluded since the sequence length was too long to align with others.
In this study, the phylogenetic tree based on nuclear SSU rDNA sequences was divided into Apicomplexa and dinoflagellates, and then branches of Noctiluca, Perkinsus, Oxyrrhis, and other dinoflagellates. The internal node, where Oxyrrhis and other core dinoflagellates were bound together, proved that Oxyrrhis occupies a basal position in the dinoflagellate lineage. Moreover, branches classified as Oxyrrhis JJ with HD, SH with MS, and GS with KRR prove that there is a big difference for each clade in the SSU sequence as well.
However, the bootstrapping number was low and the branch length of Oxyrrhis was irregularly long. This result is from the limit of the nuclear SSU rDNA. Ribosomal DNA has been utilized in myriad of studies since its advantage is the number of taxa or alignment depth, but it has limitations in that the alignment length is limited, and calculations become more difficult with insertions and deletions in the rDNA sequence. Moreover, some studies asserted that O. marina and N. scintillans can invalidate phylogeny because of their long branches [32,33]. To solve these limitations and determine the more accurate phylogenetic location of Oxyrrhis, more sequence information of taxa is needed, and this is not limited to nuclear rDNA, but also mitochondrial DNA or protein DNA information, which should be added to the concatenated alignment of the phylogenetic tree. However, in the case of mitochondrial DNA or protein DNA, information on various taxa is still lacking, and the degree of differences for each matrix is large, and thus, caution should be taken in the analysis.
Genetic differences between Oxyrrhis species can be seen more clearly in the phylogenetic tree based on the nuclear ITS rDNA sequence. According to Lowe et 1 and 2). Moreover, both strains placed highly distinct from O. maritima in clade 3 [35]. This study confirmed that the two Oxyrrhis species are genetically distinct from each other. Furthermore, we suggest that some strains of O. marina (AY566417, UK/ FJ853668, UK/ FJ853677, GRE) mentioned in previous studies should be transferred to O. maritima.
The genetic diversity of O. maritima has been examined in various studies. Lowe et al. reported that O. marina was more widely distributed, from the Atlantic Ocean to the Mediterranean, than O. maritima that was partially distributed and elusive in the EU Ocean coasters [35]. Moreover, from previous reports, the genetic divergence in O. maritima was limited to clade 3 and clade 4 that contained only 2-3 haplotypes. In this study, however, three clades were discovered throughout the Korean coast. This difference in the dominant biological community might be due to the geographic characteristics of South Korea, which is adjacent to three seas with distinct ecological properties. Therefore, it is worth investigating South Korea for the genetic diversity of Oxyrrhis spp. Since O. marina is adaptable to various environments, we cannot explain its genetic diversity without analyzing the environmental conditions. Therefore, it is necessary to collect the strains of O. marina and O. maritima from more diverse environments for analyzing their morphological and genetic characteristics.

Physiology
The environmental factors affecting the growth of O. marina and O. maritima have been measured in the laboratory. O. marina did not grow at temperatures ≥35 • C, whereas O. maritima showed tolerance to high temperatures, which can explain its habitat. O. maritima was collected in tidal waters, which has a significant variation in temperature depending on direct sunlight or weather conditions until high tide.
The effects of salinity on the growth rate examined in this study were also correlated with their habitat. O. marina did not grow at a salinity below 4, whereas O. maritima species could grow at a salinity of 2. As mentioned above, tidal waters are greatly affected by environmental conditions, and heavy rainfalls can dramatically decrease the salinity close to that of freshwater.
The results indicate that O. maritima could survive in a wider range of environmental conditions than O. marina, and their ecological characteristics are distinct. In recent years, studies have investigated changes in species and habitats due to an increase in seawater temperature. The comparison of the optimal growth temperature and ecological characteristics of O. marina and O. maritima in this study can contribute to the accurate distinction between the two species.
Regarding Oxyrrhis strains used in the nuclear ITS sequence-based phylogenetic tree, there are no data on their collection areas and ecological characteristics, and thus it is difficult to make a comprehensive judgment on the species classification. As mentioned, the nuclear genome has several limitations such as sequence length. Thus, it should be supplemented by matching genetic information with ecological and morphological information. In this study, by systematically combining morphological information, ecological information, and genetic information, O. marina and O. maritima were accurately distinguished, and it was confirmed that they show distinct differences in all three aspects.

Conclusions
O. marina is a type species of the genus Oxyrrhis and has been used as a model species for a broad range of ecological, biological, and evolutionary studies. However, the taxonomy and classification of Oxyrrhis spp. are still vague and controversial, and the other species, O. maritima, have not been well investigated to date. For these reasons, we established four different strains of O. marina isolated from the coastal waters off GS, MS, SH, and KRR, Korea, and two strains of O. maritima isolated from littoral tide pool waters in Jeju Island, Korea, and confirmed that O. maritima from JJ and HD were highly divergent from O. marina (clades 1 and 2).
Environmental conditions are an important factor in determining the growth of O. marina and O. maritima. O. marina did not grow at water temperatures ≥ 35 • C, whereas O. maritima did. Thus, the highest temperature for positive growth of O. marina is likely to be approximately 32 • C, whereas that for O. maritima is ≥32 • C. O. marina did not grow at salinities below 4, and as salinity increased from 5 to 50, the growth rate increased continuously but rapidly decreased when salinity was over 70. However, O. maritima could grow at a salinity of 2, and its growth rate increased as salinity increased from 3 to 50, and rapidly decreased as salinity increased over 70. Thus, O. marina and O. maritima may have different ecophysiology, reflecting their habitats; O. maritima inhabits areas of higher salinity and temperatures (tidal pools) than O. marina (water columns).

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/w13152057/s1, Table S1: List of primer pairs used for amplification of SSU rDNA segments, ITS region, and LSU rDNA sequences, Table S2: List of species used in constructing nuclear SSU ribosomal (r) DNA phylogenetic trees. Genbank accession numbers are listed to the right of each species, Table S3: List of species used in constructing nuclear ITS1-5.8S-ITS2 phylogenetic trees. Genbank accession numbers are listed to the right of each species.
Author Contributions: Data analyses, methodology, writing-original draft preparation, M.K.J.; Culture collection, strain setup, culturing and microscopic analysis, T.Y.Y. and S.J.M.; Conceptualizationreview and editing, J.P.; conceptualization, supervision, project administration, writing-review and editing, E.Y.Y. All authors have read and agreed to the published version of the manuscript.
Funding: This research was a part of the project titled "Development of eco-friendly functional strategic materials for improving blood circulation derived from marine microalgae (No. 20210515)" and "Improvement of management strategies on marine disturbing and harmful organisms (No. 20190518)" funded by the Ministry of Oceans and Fisheries, Korea.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.