INTRODUCTION
Dinoflagellates are major components of marine ecosystems and play diverse roles in marine food webs (Brand et al. 2012, Chang and Gall 2013, Jeong et al. 2013, Kang et al. 2023a, 2023b, Ok et al. 2023b, You et al. 2023). They often cause blooms (Holmes et al. 1967, Anderson et al. 2002, Jeong et al. 2013, Intergovernmental Oceanographic Commission of UNESCO 2024). Some dinoflagellates produce noxious and / or toxic stances and cause harmful algal blooms (HABs) that can result in ecological disruption, large-scale faunal mortality, human health risks, and economic loss (Turner and Tester 1997, Turner 2006, 2014, Anderson et al. 2012, Shi et al. 2012). Therefore, predicting the outbreak, persistence, and decline of dinoflagellate blooms is a critical concern for scientists, aquafarmers, officials, and the public. The population dynamics of bloom-forming dinoflagellate species are primarily affected by their growth and mortality rates due to predation (Buskey 1997, Jeong et al. 1999b, 2001, 2015, Lee et al. 2017). In general, the grazing impact by protists, including heterotrophic dinoflagellates, is greater than that by metazoans because the abundance of protist grazers is much higher than that of metazoan grazers (Kim et al. 2013, Yoo et al. 2013, Lee et al. 2017, Lim et al. 2017). Therefore, to understand dinoflagellate bloom dynamics, interactions between bloom-forming dinoflagellate species and their potential protist grazers should be determined.
Many species in the genus Karenia produce neurotoxins and cause HABs (Baden 1989, Botes et al. 2003, Haywood et al. 2004, Van Wagoner et al. 2010, Chang 2011, Brand et al. 2012, Intergovernmental Oceanographic Commission of UNESCO 2024). Several Karenia species have lethal effects on diverse marine organisms, including fish, invertebrates, and mammals (Heil et al. 2001, Shi et al. 2012, Twiner et al. 2012). Many studies have been conducted on the interactions between Karenia species and metazoans, some of which feed on Karenia mikimotoi, Karenia bicuneiformis, and Karenia brevis (Uye and Takamatsu 1990, Tester et al. 2000, Cohen et al. 2007, Ok et al. 2024). They can also affect the survival, growth, movement, and egg production of metazooplankton such as copepods, rotifers, and crustacean nauplii (Prince et al. 2006, Cohen et al. 2007, Kubanek et al. 2007, Chang and Gall 2013, Dang et al. 2015, Ok et al. 2024). There have been few studies on the interactions between Karenia species and heterotrophic protists, including three species of the heterotrophic dinoflagellate genus Gyrodinium, which can feed on K. brevis; the heterotrophic dinoflagellate Noctiluca scintillans; and the tintinnid ciliate Favella ehrenbergii, both of which have been observed to feed on K. mikimotoi (Nakamura et al. 1995, Nakamura 1998, Yoo et al. 2013). Ten Karenia species have been previously described so far (Guiry and Guiry 2024). Therefore, the interactions between other Karenia species and heterotrophic dinoflagellates should be explored.
K. bicuneiformis is known to produce brevetoxins and cause visible dense blooms in Japan and South Africa (Botes et al. 2003, Haywood et al. 2004, Intergovernmental Oceanographic Commission of UNESCO 2024). However, there have been no records of the impact of K. bicuneiformis on humans or animals (Botes et al. 2003). Karenia selliformis causes HABs along the coasts of Chile, Japan, Kuwait, New Zealand, Russia, Tunisia, and the United States (MacKenzie et al. 1996, Heil et al. 2001, Uribe and Ruiz 2001, Naila et al. 2012, Orlova et al. 2022, Intergovernmental Oceanographic Commission of UNESCO 2024, Ohnishi et al. 2024). This species is known to be one of the toxic dinoflagellates that produce gymnodimines (GYMs), brevetoxins, or brevenal, resulting in massive mortalities in fish and shellfish (Miles et al. 2000, Haywood et al. 2004, Mardones et al. 2020, Iwataki et al. 2022). Recently, Ok et al. (2024) reported that the copepod Acartia hongi fed on K. bicuneiformis and survived exposure to K. bicuneiformis. However, K. selliformis is known for its lethal effects on copepods (Ohnishi et al. 2024, Ok et al. 2024). This suggests that, although both species cause blooms, they exert different effects on copepods. Therefore, it is necessary to examine the interactions between these two Karenia species and potential protist grazers.
Heterotrophic dinoflagellates are known to feed on many bloom-forming species, including toxic species (Jeong et al. 2001, 2003, 2010, Stoecker et al. 2002, Tillmann 2004, Clough and Strom 2005, Turner 2006, Jang et al. 2016, You et al. 2020, Kang et al. 2023a). Jeong et al. (2005) estimated that the grazing impact of populations of the heterotrophic dinoflagellate Stoeckeria algicida on the red-tide raphidophyte Heterosigma akashiwo could remove up to 13% of the prey population per minute. Furthermore, during Margalefidinium polykrikoides blooms, populations of the heterotrophic dinoflagellate Polyrkikos kofoidii were estimated to remove 99% of the M. polykrikoides population in one day (Lim et al. 2017). Therefore, feeding by heterotrophic dinoflagellates on a bloom-forming dinoflagellate could significantly affect the population dynamics and interactions between K. bicuneiformis, K. selliformis, and potential heterotrophic dinoflagellate grazers.
In the present study, the feeding behaviors of eight heterotrophic dinoflagellates, Gyrodiniellum shiwhaense, Pfiesteria piscicida, Gyrodinium dominans, Oxyrrhis marina, Gyrodinium moestrupii, N. scintillans, Protoperidinium pellucidum, and Oblea rotunda, were observed after each heterotrophic dinoflagellate was added to each culture of K. bicuneiformis and K. selliformis. Most of these heterotrophic dinoflagellates are known to be present and often abundant in the coastal waters of many countries (Ok et al. 2023a, GBIF.org 2024, Ocean Biodiversity Information System 2024). The growth and ingestion rates of the five heterotrophic dinoflagellates feeding on K. bicuneiformis were measured. In the preliminary test, the growth rate of P. pellucidum on K. bicuneiformis was highest. Therefore, the growth and ingestion rates of P. pellucidum on K. bicuneiformis were measured as a function of prey concentration. The growth and ingestion rates of G. dominans, G. moestrupii, O. marina, and O. rotunda were measured at a single K. bicuneiformis concentration, where the growth rate of P. pellucidum was saturated. Moreover, the survival of G. dominans and G. moestrupii as a function of K. selliformis cell concentration or equivalent culture filtrates was explored. The results of this study provide a basis for understanding the interactions between common heterotrophic dinoflagellates and each of K. bicuneiformis and K. selliformis as well as the ecological roles of K. bicuneiformis and K. selliformis in marine planktonic food webs.
MATERIALS AND METHODS
Preparation of experimental organisms
Clonal cultures of K. bicuneiformis CAWD81 and K. selliformis NIES 4541 were obtained from the Cawthron Institute Culture Collection of Microalgae and the National and Microbial Culture Collection at the National Institute for Environmental Studies, respectively. These cultures were maintained in L1 seawater medium without silicate (hereafter, L1) (Guillard and Hargraves 1993) and incubated at 20°C under illumination of 100 μmol m−2 s−1 using cool white fluorescent light and a 14 : 10 h light and dark cycle. The heterotrophic dinoflagellates G. shiwhaense, G. dominans, O. marina, G. moestrupii, N. scintillans, P. pellucidum, and O. rotunda were isolated from the surface waters off the coast of Geoje, Jeongok, Kunsan, Saemangeum, Dangjin, and Jinhae, Korea, respectively. The culture of P. piscicida was obtained from the National Center for Marine Algae and Microbiota (NCMA) in the United States. Each heterotrophic dinoflagellate species was grown on suitable prey species (Table 1).
Feeding occurrence by potential predators
Experiments 1 and 2 were conducted to investigate the feeding capabilities of the heterotrophic dinoflagellates G. shiwhaense, P. piscicida, G. dominans, O. marina, G. moestrupii, N. scintillans, P. pellucidum, and O. rotunda in K. bicuneiformis and K. selliformis. The two Karenia species and eight heterotrophic dinoflagellates were individually matched (Table 2).
After the cell density of K. bicuneiformis (or K. selliformis) in a culture was determined, a predetermined volume was transferred to wells of a six-well plate (final targeted cell density = 3,000 cells mL−1). In addition, the cell density of the potential predator species culture was determined after prey cells in the culture became undetectable. Subsequently, a predetermined volume was transferred to the wells of a six-well plate. One experimental well (a mixture of K. bicuneiformis or K. selliformis with a potential predator), one prey control well (only K. bicuneiformis or K. selliformis), and one predator control well (only a potential predator) were prepared at 5 mL in a six-well plate. The organisms in the wells of the plates were incubated at 20°C under illumination of 20 μmol m−2 s−1 using a cool white fluorescent light and a 14 : 10 h light and dark cycle. The plates were observed after 0, 2, 6, 24, and 48 h of incubation. Approximately a total of more than 30 potential predator cells were observed under a dissecting microscope at a magnification of 10–63× to determine the ability of the predator to feed on K. bicuneiformis (or K. selliformis). Photographs of the heterotrophic dinoflagellates and Karenia species were captured at 630–1,000× magnification using a digital camera (Zeiss Axiocam 506 and Zeiss Axiocam 820 color; Carl Zeiss Ltd., Göttingen, Germany) attached to an inverted microscope (Zeiss Axiovert 200M and Zeiss Axio Observer 7; Carl Zeiss Ltd; NFEC-2024-12-301531).
Growth and ingestion rates of Protoperidinium pellucidum as a function of Karenia bicuneiformis concentration
In Experiment 1, P. pellucidum fed well on K. bicuneiformis. Therefore, the growth and ingestion rates were measured as a function of K. bicuneiformis concentration (Experiment 3). A culture of K. bicuneiformis (approximately 500 cells mL−1) was transferred to an 800-mL culture flask (Falcon; Corning Inc., New York, USA) with L1. A culture of P. pellucidum was transferred to a 500-mL PC bottle containing Amphidinium carterae as prey and freshly filtered seawater (FSW) and then rotated on a plankton rotating wheel at 0.9 rpm (0.00017 g). When the A. carterae in the P. pellucidum culture became undetectable, a dense culture of P. pellucidum was transferred into a 500-mL PC bottle containing K. bicuneiformis (approximately 16 cells mL−1) and then rotated as described above. A 5-mL aliquot was taken from the predator culture and fixed with Lugol’s solution to 5% every other day. The predator and prey cells in the fixed samples were counted under a light microscope (CX21; Olympus Corporation, Tokyo, Japan). When K. bicuneiformis cells were completely consumed, P. pellucidum were starved for two days to avoid possible residual growth.
Triplicate 42-mL PC experimental bottles containing P. pellucidum and K. bicuneiformis; prey-control bottles containing only K. bicuneiformis; and predator-control bottles containing only P. pellucidum were set up for each predator-prey combination. Predetermined volumes of P. pellucidum or K. bicuneiformis were added to each experimental or control bottle using an autopipette (Table 3). The predator culture was gently filtered using a 0.2-μm disposable syringe filter (DISMIC-25CS type, 25 mm; Advantec, Toyo Roshi Kaisha Ltd., Chiba, Japan) and then added to the prey control bottles in the same volume as the predator culture added to the experimental bottle to create similar conditions in the bottles. The prey culture was also filtered and added to the predator control bottles as described above. All the bottles were filled with FSW and capped. A 5-mL aliquot was pipetted from each bottle and fixed with 5% Lugol’s solution at the start of the experiment. The bottles were refilled with 5 mL of FSW and placed on a plankton-rotating wheel under the conditions described above. The dilution rate of the bottles due to refilling was considered when calculating the growth and ingestion rates. After 72 h of incubation, a 10-mL aliquot was taken from each bottle and fixed with 5% Lugol’s solution. The densities of P. pellucidum and K. bicuneiformis in all fixed samples were determined by counting >200 cells or all cells in triplicate in 1-mL Sedgewick-Rafter chambers (SRCs).
The specific growth rate of P. pellucidum, μ (d−1), was calculated as:
, where C0 and Ct are the concentrations of P. pellucidum at 0 and 72 h, respectively.
Data for the growth rates of P. pellucidum were fitted to a modified Michaelis-Menten equation:
, where μmax is the maximum growth rate (d−1), x is the prey concentration (cells mL−1 or ng C mL−1), x′ is the threshold prey concentration (prey concentration where μ = 0), and KGR is the prey concentration sustaining 1/2 μmax. Data were iteratively fitted to the model using Delta Graph (IBM Corp., Armonk, NY, USA).
Ingestion and clearance rates were calculated using the modified equations of Frost (1972) and Heinbokel (1978). The incubation time for the calculation of the ingestion and clearance rates was the same as that used to calculate the growth rate. Data for ingestion rates (IR, cells predator−1 d−1 or ng C predator−1 d−1) of P. pellucidum were fitted to the Michaelis–Menten equation:
, where Imax is the maximum ingestion rate (cells predator−1 d−1 or ng C predator−1 d−1), x is the prey concentration (cells mL−1 or ng C mL−1), and KIR is the prey concentration that sustains 1/2 Imax. The carbon content per cell of K. bicuneiformis was determined as previously described (Ok et al. 2024).
Growth and ingestion rates of four predators at a single concentration of Karenia bicuneiformis
Experiments 4–7 were designed to determine the growth and ingestion rates of G. dominans, G. moestrupii, O. rotunda, and O. marina on K. bicuneiformis at high prey concentrations. The growth and ingestion rates of four heterotrophic dinoflagellate predators on K. bicuneiformis were quantified at the second- and third-highest prey concentrations in Experiment 3, following the methods described in Experiment 3. The growth and ingestion rates of each heterotrophic dinoflagellate on K. bicuneiformis were selected to compare the rates of each predator at a single high K. bicuneiformis concentration.
Effects of Karenia selliformis cell concentration and equivalent culture filtrate on the survival of Gyrodinium dominans and Gyrodinium moestrupii
Experiments 8 and 9 were conducted to investigate the survival of G. dominans and G. moestrupii as a function of the K. selliformis cell concentration and elapsed incubation time. The experiments included one control and five experimental conditions. The control group consisted of G. dominans (or G. moestrupii) incubated in FSW, while the experimental group exposed G. dominans (or G. moestrupii) to five differential concentrations of K. selliformis in triplicate (Table 3). Predetermined volumes of G. dominans (or G. moestrupii) and K. selliformis were added to 50-mL experimental or control flasks (Falcon; Corning Inc.) using an autopipette. The total volume was 40-mL and was filled with FSW. The flasks were incubated at 20°C under illumination of 20 μmol m−2 s−1 and a 14 : 10 h light and dark cycle. A 5-mL aliquot was taken from each flask and fixed with 5% Lugol’s solution at the beginning of the experiments and after 2, 6, 24, and 48 h of incubation to determine the density of G. dominans (or G. moestrupii) at each incubation time. All or more than 200 cells of G. dominans (or G. moestrupii) in SRCs were enumerated under a light microscope (CX21; Olympus Corporation).
Experiments 10 and 11 were conducted to determine the survival of G. dominans and G. moestrupii as a function of the volume of K. selliformis culture filtrate and elapsed incubation time. The culture of K. selliformis was poured into a syringe with a 0.2-μm disposable syringe filter (DISMIC-25CS type, 25 mm) and gently filtered. All K. selliformis cells were removed, as confirmed from the filtrate using a dissecting microscope. Predetermined volumes of G. dominans (or G. moestrupii) and K. selliformis filtrates were placed in 50-mL culture flasks (Falcon; Corning Inc.) and filled with FSW to the target volume (40 mL). The experiments were set up and proceeded as described in Experiments 8 and 9, except for the use of an equal volume of K. selliformis filtrate.
Survival was calculated as the percentage (%) of surviving cells relative to the total cells of G. dominans (or G. moestrupii) examined after exposure to K. selliformis cells or an equivalent culture filtrate for 2, 6, 24, or 48 h.
To determine the median lethal concentration (LC50), the survival data of G. dominans (or G. moestrupii) were fitted to the following equation from Kim et al. (2016):
If the coefficient of determination for the linear regression was higher, linear regression was applied:
, where x is the initial K. selliformis cell concentration (cells mL−1), and α, β, γ, and δ are variables related to the elapsed time of incubation. The value of SVmax in the Eq. (4) is 100%. Data were fitted to the model using Delta Graph (IBM Corp.).
Statistical analysis
To determine the differences in the growth and ingestion rates of the five heterotrophic dinoflagellates feeding on K. bicuneiformis, the non-parametric Kruskal-Wallis test and Mann-Whitney U test with Bonferroni correction were performed (p < 0.05) (Mann and Whitney 1947, Kruskal and Wallis 1952). A two-tailed independent samples t-test was conducted to evaluate the difference between the growth rate of O. rotunda on K. bicuneiformis at a single high mean prey concentration and that at 0 ng C mL−1. To assess significant differences in the effect of K. selliformis cell concentration or equivalent culture filtrate on the survival of G. dominans and G. moestrupii, one-way ANOVA was performed, followed by a post-hoc Tukey’s honest significant difference test when the data met normality according to the Shapiro-Wilk test and homoscedasticity using Levene’s tests (Tukey 1949). If the data were not normally distributed, the Kruskal-Wallis test was used with the post-hoc Mann-Whitney U test with Bonferroni correction (Mann and Whitney 1947, Kruskal and Wallis 1952). If the variances between the treatment groups were not homogeneous, Welch’s one-way ANOVA and the Games-Howell post-hoc test were performed (Welch 1947, Games and Howell 1976). All statistical analyses were performed using SPSS version 25.0 (IBM-SPSS Corp.).
RESULTS
Feeding occurrence of heterotrophic dinoflagellates on Karenia bicuneiformis and Karenia selli-formis
The heterotrophic dinoflagellates G. dominans, G. moestrupii, O. marina, O. rotunda, and P. pellucidum were capable of feeding on K. bicuneiformis (Table 2). O. rotunda and P. pellucidum towed, enclosed with pallium, and fed on K. bicuneiformis cells (Fig. 1). G. dominans, G. moestrupii, and O. marina swam rapidly around a K. bicuneiformis cell when the prey cell was captured, then engulfed the prey cell. The peduncle feeder G. shiwhaense often attacked K. bicuneiformis cells; however, K. bicuneiformis quickly swam helically and escaped from G. shiwhaense immediately. Cells of the peduncle feeder P. piscicida and the engulfment feeder N. scintillans did not attack K. bicuneiformis.
All cells of the eight heterotrophic dinoflagellates lost their mobility after being incubated with K. selliformis at 3,000 cells mL−1 for 1 h. Moreover, all cells of G. dominans, G. moestrupii, O. marina, and G. shiwhaense burst after 24 h of incubation, whereas P. piscicida, O. rotunda, P. pellucidum, and N. scintillans maintained their cell shapes until 24 h of incubation but lysed after 48 h of incubation (Fig. 2).
Growth and ingestion rates of Protoperidinium pellucidum feeding on Karenia bicuneiformis as a function of prey concentration
The specific growth rate of P. pellucidum feeding on K. bicuneiformis rapidly increased with increasing mean prey concentrations ≤ 95.2 ng C mL−1 (140 cells mL−1), but became saturated at higher concentrations (Fig. 3A). When the data were fitted to Eq. (2), the calculated maximum growth rate (μmax) of P. pellucidum on K. bicuneiformis was 0.19 d−1.
As mean prey concentration increased, the ingestion rate of P. pellucidum feeding on K. bicuneiformis rapidly increased at mean prey concentrations ≤ 399.2 ng C mL−1 (587 cells mL−1), but increased slowly at higher mean prey concentrations (Fig. 3B). When the data were fitted to Eq. (3), the maximum ingestion rate (Imax) of P. pellucidum on K. bicuneiformis was 0.86 ng C predator−1 d−1 (1.26 cells predator−1 d−1).
Growth and ingestion rates of five heterotrophic dinoflagellates on Karenia bicuneiformis at a single prey concentration
At a single high mean prey concentration of 1,139–1,420 ng C mL−1, the growth rates of O. rotunda, G. dominans, O. marina, G. moestrupii, and P. pellucidum on K. bicuneiformis were −0.004, 0.04, 0.06, 0.13, and 0.20 d−1, respectively (Fig. 4A). The growth rates of O. rotunda, G. dominans, G. moestrupii, O. marina, and P. pellucidum on K. bicuneiformis at single high prey concentration were significantly different (Kruskal-Wallis test; H4 = 10.77, p = 0.03). However, the growth rate of O. rotunda feeding on K. bicuneiformis at a single high mean prey concentration was not significantly different from that without K. bicuneiformis (two-tailed t-test, t4 = 1.01, p = 0.37).
At mean prey concentrations of 1,139–1,420 ng C mL−1, the ingestion rate of O. rotunda feeding on K. bicuneiformis was the lowest (0.26 ng C predator−1 d−1), while the ingestion rates of G. dominans, O. marina, G. moestrupii, and P. pellucidum were 0.64, 0.58, 0.45, and 0.76 ng C predator−1 d−1, respectively (Fig. 4B). The ingestion rates of the five heterotrophic dinoflagellates feeding on K. bicuneiformis were significantly different (Welch’s ANOVA, F4, 4.13 = 11.43, p = 0.02).
Effects of cell concentration and equivalent culture filtrate of Karenia selliformis on survival of Gyrodinium dominans
At all initial K. selliformis cell concentrations of 0, 10, 50, 100, 500, and 1,000 cells mL−1, the survival of G. dominans decreased with increasing incubation time (Fig. 5A). At initial K. selliformis concentrations of 0 and 10 cells mL−1, the survival of G. dominans gradually decreased until 48 h of incubation (Fig. 5A). At initial K. selliformis concentrations of ≥50 cells mL−1, the survival of G. dominans rapidly decreased and was almost zero at an elapsed time of 6 h.
After 2, 6, 24, and 48 h of incubation, the survival of G. dominans rapidly decreased with increasing initial K. selliformis cell concentrations (Fig. 5B). After 2 h incubation, the survival of G. dominans rapidly decreased to 8% at K. selliformis cell concentrations of ≤100 cells mL−1 and decreased to 1% at ≥500 cells mL−1 (Fig. 5B). After 6 and 24 h of incubation, the survival of G. dominans was zero at initial K. selliformis concentrations of 100 cells mL−1, while after 48 h, survival reached zero at 50 cells mL−1. Across all incubation times, the survival of G. dominans was significantly affected by the initial K. selliformis cell concentration (Kruskal-Wallis test, H5 = 15.95, p = 0.01 for 2 h; Welch’s ANOVA, F5, 5.29 = 209.96, p < 0.001 for 6 h; Kruskal-Wallis test, H5 = 15.94, p = 0.01 for 24 h; H5 = 15.31, p = 0.01 for 48 h; post-hoc test results) (Table 4). After 2, 6, 24, and 48 h of incubation, the calculated LC50 values for G. dominans were 59, 15, 20, and 20 cells mL−1, respectively (Table 5).
At all filtrate concentrations equivalent to initial K. selliformis cell concentrations of 0, 10, 50, 100, 500, and 1,000 cells mL−1, the survival of G. dominans similarly decreased to 14–19% at the final incubation time (Fig. 5C), indicating no effects of K. selliformis filtrate on the survival of G. dominans.
After 2, 6, 24, and 48 h of incubation, the survival of G. dominans gradually decreased with increasing filtrate concentrations equivalent to the initial K. selliformis cell concentration (Fig. 5D). After 2, 6, 24, and 48 h of incubation, the survival of G. dominans was 63–81% at filtrate concentrations equivalent to initial K. selliformis cell concentrations of 1,000 cells mL−1. After 2 and 6 h of incubation, the survival of G. dominans was significantly affected by filtrate concentrations equivalent to the initial K. selliformis cell concentrations (Kruskal-Wallis test, H5 = 14.28, p = 0.01 for 2 h; H5 = 12.63, p = 0.03 for 6 h; post-hoc test results) (Table 4). However, after 24 and 48 h of incubation, the survival of G. dominans was not significantly affected by filtrate concentrations equivalent to the initial K. selliformis cell concentrations (Kruskal-Wallis test, H5 = 7.39, p > 0.05, for 24 h; one-way ANOVA, F5, 12 = 0.69, p > 0.05, for 48 h). After 2, 6, 24, and 48 h of incubation, the survival of G. dominans with filtrate concentrations equivalent to initial K. selliformis cell concentrations did not decrease below 50%; thus, the LC50 values were not calculated (Table 5).
Effects of cell concentration and equivalent culture filtrate of Karenia selliformis on survival of Gyrodinium moestrupii
At all initial K. selliformis cell concentrations of 0, 10, 50, 100, 500, and 1,000 cells mL−1, the survival of G. moestrupii decreased with increasing incubation time (Fig. 6A). At initial K. selliformis cell concentrations of 0, 10, and 50 cells mL−1, the survival of G. moestrupii slightly decreased after 2 h of incubation. At an initial K. selliformis cell concentration of 100 cells mL−1, survival rapidly decreased at elapsed times of 2 h. At initial K. selliformis cell concentrations of 500 and 1,000 cells mL−1, the survival of G. moestrupii reached zero after 2 h of incubation.
After 2, 6, 24, and 48 h of incubation, the survival of G. moestrupii rapidly decreased with increasing initial K. selliformis cell concentrations (Fig. 6B). After 2, 6, 24, and 48 h of incubation, survival was significantly affected by K. selliformis cell concentration (Kruskal-Wallis test, H5 = 15.49, p = 0.01 for 2 h; H5 = 16.20, p = 0.01 for 6 h; H5 = 16.55, p = 0.01 for 24 h; H5 = 16.33, p = 0.01 for 48 h; post-hoc test results) (Table 4). After 2, 6, 24, and 48 h of incubation, the calculated LC50 values for G. moestrupii with K. selliformis cells were 95, 72, 55, and 60 cells mL−1, respectively (Table 5).
At filtrate concentrations equivalent to initial K. selliformis cell concentrations of 0–100 cells mL−1, the survival of G. moestrupii slightly decreased between 84 and 93% at final incubation time (Fig. 6C). At filtrate concentrations equivalent to the initial K. selliformis cell concentration of 500 cells mL−1, the survival of G. moestrupii decreased to 67% at final incubation time (Fig. 6C). At filtrate concentrations equivalent to initial K. selliformis cell concentrations of 1,000 cells mL−1, survival rapidly decreased to 51% at final incubation time.
After 2, 6, 24, and 48 h of incubation, the survival of G. moestrupii gradually decreased with increasing filtrate concentrations equivalent to the initial K. selliformis cell concentrations (Fig. 6D). After 2, 6, 24, and 48 h of incubation, the survival of G. moestrupii at a filtrate concentration equivalent to initial K. selliformis cell concentration of 1,000 cells mL−1 was 58–73%. After 2, 6, 24, and 48 h of incubation, survival was significantly affected by filtrate concentrations equivalent to the initial K. selliformis cell concentrations (Kruskal-Wallis test, H5 = 12.52, p = 0.03 for 2 h; H5 = 15.67, p = 0.01 for 6 h; H5 = 15.14, p = 0.01 for 24 h; H5 = 12.73, p = 0.03 for 48 h; post-hoc test results) (Table 4). After 2, 6, 24, and 48 h of incubation, the survival of G. moestrupii with filtrate concentrations equivalent to initial K. selliformis cell concentrations did not decrease below 50%, and thus, the LC50 values were not calculated (Table 5).
DISCUSSION
This study is the first to investigate the interactions between the bloom-forming dinoflagellates K. bicuneiformis and K. selliformis and their potential heterotrophic dinoflagellate grazers. The results of the present study clearly show that the heterotrophic dinoflagellates tested responded differently to cell density and equivalent culture filtrates of K. bicuneiformis and K. selliformis; K. bicuneiformis supported the growth of several heterotrophic dinoflagellates, whereas K. selliformis lysed them. Therefore, K. bicuneiformis acts as a suitable prey for heterotrophic dinoflagellates in marine planktonic food webs, whereas K. selliformis acts as a killer. The results of this study expand our knowledge of the interactions between protozooplankton and Karenia species (Table 6).
Ok et al. (2024) demonstrated the differential interaction between K. bicuneiformis and K. selliformis and the calanoid copepod Acartia hongi; A. hongi was killed when exposed to K. selliformis cells but not when exposed to K. bicuneiformis cells. Moreover, there have been reports on the massive mortality of invertebrates and fish during K. selliformis blooms but no report on their mortality during K. bicuneiformis blooms (Arzul et al. 1995, MacKenzie et al. 1996, Clément et al. 2000, Heil et al. 2001, Botes et al. 2003). Thus, the results of the present and previous studies suggest that K. selliformis is capable of killing diverse trophic-leveled organisms, ranging from heterotrophic dinoflagellates to fish, whereas K. bicuneiformis does not exhibit such lethal effects on them.
Engulfment feeders O. marina, G. moestrupii, and G. dominans could feed on K. bicuneiformis, while N. scintillans did not. N. scintillans feeds on prey using mucus that may form at the tip of the tentacle (Kiφrboe and Titelman 1998, You et al. 2022) and feeds on Scrippsiella acuminata, Prorocentrum micans, and K. mikimotoi, which have equivalent spherical diameters of 21.3–22.8 μm, similar to that of K. bicuneiformis (Nakamura 1998, Kiφrboe and Titelman 1998, our unpublished data). The maximum swimming speeds of S. acuminata (348 μm s−1), P. micans (380 μm s−1), and K. mikimotoi (230 μm s−1) are lower than that of K. bicuneiformis (480 μm s−1) (Nakamura 1998, Jeong et al. 1999a, Smayda 2002, Ok et al. 2024). Therefore, N. scintillans may have difficulty catching K. bicuneiformis cells compared to other dinoflagellate prey species. In the present study, the peduncle feeders P. piscicida and G. shiwhaense did not feed on K. bicuneiformis. P. piscicida and G. shiwhaense use a feeding tube to ingest prey materials after capturing a prey cell with a tow filament (Burkholder and Glasgow 1997, Jeong et al. 2011). K. bicuneiformis cells immediately escaped from attacks by G. shiwhaense. Therefore, P. piscicida and G. shiwhaense may have difficulty deploying a tow filament on K. bicuneiformis cells.
At a single high mean prey concentration of K. bicuneiformis, the growth rate of O. rotunda was almost zero, whereas those of G. dominans, G. moestrupii, O. marina, and P. pellucidum were positive. Given their ingestion rates, O. rotunda cells acquired 40% d−1 of their body carbon from K. bicuneiformis, whereas cells of the other four species acquired 158–539% d−1. Therefore, the low ingestion rates of O. rotunda on K. bicuneiformis may be partially responsible for its lack of growth.
The maximum growth rate (μmax) of P. pellucidum feeding on K. bicuneiformis (0.2 d−1) was lower than that on the diatom Ditylum brightwellii (0.7 d−1) and the phototrophic dinoflagellate P. micans (0.4 d−1) (Buskey 1997). The maximum ingestion rate (Imax) of P. pellucidum feeding on K. bicuneiformis (0.86 ng C predator−1 d−1) was considerably lower than that on D. brightwellii (7.10 ng C predator−1 d−1) and P. micans (9.09 ng C predator−1 d−1) (Buskey 1997). The maximum swimming speed of P. micans (380 μm s−1) was lower than that of K. bicuneiformis (480 μm s−1) (Jeong et al. 1999b, Ok et al. 2024). Thus, the higher swimming speed of K. bicuneiformis compared to D. brightwellii and P. micans may be partially responsible for the considerably lower Imax of P. pellucidum on K. bicuneiformis than on D. brightwellii or P. micans, which in turn leads to the lower μmax of P. pellucidum when feeding on K. bicuneiformis.
The calculated growth and ingestion rates of G. moestrupii feeding on K. brevis at a concentration of 1,192 ng C mL−1 were 0.79 d−1 and 1.84 ng C predator−1 d−1 (Yoo et al. 2013), which are higher than those observed when feeding on K. bicuneiformis at the same concentration (0.18 d−1 and 0.67 ng C predator−1 d−1). The highest growth rate of G. moestrupii is 1.6 d−1 when feeding on the mixotrophic dinoflagellate Alexandrium minutum, which forms HABs in the global oceans (Yoo et al. 2013, Intergovernmental Oceanographic Commission of UNESCO 2024). Therefore, G. moestrupii is expected to have lower population densities during or after red tides of K. bicuneiformis compared to those of A. minutum and K. brevis.
K. bicuneiformis does not affect the survival of the calanoid copepod A. hongi; in contrast, it is ingested by A. hongi (Ok et al. 2024). In 2004, a bloom of K. bicuneiformis (previously K. bidigitata) occurred along the coast of Japan, reaching concentrations of 11,261 ng C mL−1 over 16 days (Intergovernmental Oceanographic Commission of UNESCO 2024). At this concentration, the calculated ingestion rates of A. hongi on K. bicuneiformis were 8,062 ng C predator−1 d−1 (Ok et al. 2024), which is 9,374 times higher than that of P. pellucidum on the same prey (0.89 ng C predator−1 d−1). The highest reported field abundances of P. pellucidum, at 10,000 cells L−1, are 588 times higher than that of A. hongi (17 individuals L−1) (Ocean Biodiversity Information System 2024, Zuo et al. 2024). Therefore, when the maximum field abundance and ingestion rate of each grazer of K. bicuneiformis are combined, A. hongi may remove more K. bicuneiformis cells than P. pellucidum. However, A. hongi has been reported to be present in the waters of Korea and Japan, whereas P. pellucidum is found across South America, North America, Europe, East Asia, and Africa (Ocean Biodiversity Information System 2024, Zuo et al. 2024). Therefore, in the waters of countries other than Korea and Japan, P. pellucidum may be an important grazer of K. bicuneiformis.
When exposed to a K. selliformis concentration of 1,000 cells mL−1, the survival rates of G. dominans and G. moestrupii were zero after 2 h of incubation. At the same concentration of K. selliformis, the survival of the copepods Paracalanus parvus, Centropages abdominalis, Acartia tumida, Neocalanus plumchrus, and Eurytemora herdmani was 50% after incubation periods of 3, 6, 12, 108, and 113 h, respectively (Ohnishi et al. 2024). Therefore, during K. selliformis blooms, G. dominans and G. moestrupii may not be found, while the copepod population densities may be reduced.
The filtrate of K. selliformis, which has lethal effects on copepods (Ohnishi et al. 2024), also killed the heterotrophic dinoflagellates G. dominans and G. moestrupii. The dinoflagellate K. selliformis mainly produces GYMs, which cause the rapid death of mice when injected intraperitoneally (Miles et al. 2000, Munday et al. 2004). In addition, K. selliformis isolated from Tunisia secretes extracellular GYMs (Tang et al. 2021). Therefore, the extracellular substances of K. selliformis may have lethal effects on copepods and heterotrophic dinoflagellates such as G. dominans and G. moestrupii. Half-lethality times for the copepod P. parvus, A. tumida, N. plumchrus, and E. herdmani in the filtrate of K. selliformis cultures of 1,000 cells mL−1 were 1.3–3.3 times longer than those in the K. selliformis concentration of 1,000 cells mL−1 (Ohnishi et al. 2024). The survival rates of G. moestrupii and G. dominans in K. selliformis culture filtrates at the elapsed times of 2, 6, 24, and 48 h were generally higher than those in K. selliformis cells. Consequently, both live K. selliformis cells and their extracellular substances have significant lethal effects on the copepods and heterotrophic dinoflagellates; however, the effect of live K. selliformis cells on their survival appears to be more immediate and severe.
The LC50 of G. moestrupii when exposed to K. selliformis cells was 55 and 60 cells mL−1 at 24 and 48 h incubation, respectively. This is higher than the LC50 of G. moestrupii when exposed to Alexandrium pohangense cells for the same exposure times, 11 and 25 cells mL−1, respectively (Kim et al. 2016). Cells of A. pohangense have only been reported in Korea, with the highest recorded field density being 13 cells mL−1 in Yongil Bay, Pohang in 2014 (Lim et al. 2015). K. selliformis has caused red tides in various countries at concentrations of 48–60,000 cells mL−1 (Arzul et al. 1995, MacKenzie et al. 1996, Clément et al. 2000, Heil et al. 2001, Uribe and Ruiz 2001, Naila et al. 2012, Elleuch et al. 2021, Iwataki et al. 2022, Orlova et al. 2022, Boudriga et al. 2023, Intergovernmental Oceanographic Commission of UNESCO 2024). Therefore, in natural environments, the population of G. moestrupii is likely to be more affected by K. selliformis than by A. pohangense.
This study extends our knowledge of the interactions between heterotrophic dinoflagellates and bloom-forming dinoflagellates, K. bicuneiformis and K. selliformis. Furthermore, this study clearly shows that K. bicuneiformis and K. selliformis play different roles in marine ecosystems, which may create differential ecological niches between these two Karenia species. Therefore, it is necessary to explore interactions between heterotrophic dinoflagellates and other Karenia species.
