Sea anemone culture

Polyps of Anthopleura atodai (Yanagi and Daly, 2004) were collected from the intertidal zone of Okinoshima Island, Tateyama, Chiba, Japan, in March 2018. The species was identified based on morphological characteristics and ITS sequences. Exaiptasia diaphana was collected from a culture tank at the Ooarai Aquarium in Ibaraki, Japan, in March 2017. DNA sequences of 12S rDNA matched those of E. diaphana (E. pallida) in Grajales and Rodríguez (2016). Polyps were cultured in plastic Tupperware containers with artificial seawater (ASW) (Red Sea Salt, Red Sea, Israel) in a room kept at 24–27°C and fed daily with the mashed cephalothorax part of the Argentine red shrimp Pleoticus muelleri available in fish markets. For each of these two anemone species, clonal strains were established by selecting individuals that exhibited high asexual reproductive activity. For experiments, an A. atodai clone termed TO18 and an E. diaphana clone OA17 were used.

Table 1.

Algal species, ITS accession number (acc. no.), host origin, and average diameter of algal cells (n = 20).

Fig. 1.

Clonal culture of A. atodai. (A) Five small polyps were produced 2 months after collection in clonal culture. (B) Approximately 60 polyps were produced from a single polyp by asexual production after 9 months.

Bleaching of polyps

The herbicide Kiredar (Agro-Kanesho Co., Ltd., Saitama, Japan) which is similar to herbicides such as Mogeton (Certis Biologicals, MD, USA) available in other countries, or its active ingredient, 2-amino-3-chloro-1,4 naphtoquinone (ACN) (Kanto Chemical Co., Inc., Tokyo, Japan), or quinoclamine as a common name, was dissolved in DMSO or suspended in ASW at 5 mM and added to cultures at a final concentration of 5 µM quinoclamine following preliminary investigations to identify the optimal concentration. The concentration of Kiredar was adjusted based on its 25% w/w quinoclamine content. The treatment cycle was 5-days drug exposure with continuous light exposure at approximately 1500 lx under white LED desk lights and a 2-days rest in ASW in the dark. This cycle was repeated until complete bleaching was achieved. Every 2–3 days, the pedal plane was photographed through the dish bottom using a fluorescent binocular camera (Nikon SMZ18/DS-Ri2 CCD camera), and the intensity of red fluorescence was evaluated using ImageJ. Ejected algal cells during and after the bleaching treatment were observed using a fluorescence microscope (Olympus FSX100).

Inoculation of algal cells into bleached polyps

The species and subclades of the symbiont algae and their hosts are listed in Table 1. Subclades were identified based on ITS sequences. In this study, symbiont algae are indicated by subclades to facilitate comparison with previous literature. Algal cells were collected from the host anemone species for each algal species by mashing host tentacle tissues in ASW. Subclade A3 was obtained by introducing algae from the giant clam Tridacna crocea, which hosts both A3 and C1, into newly formed spats of the coral Acropora tenuis, which incorporates A3 but not C1 at this stage (Yorifuji et al., 2017). Specific incorporation of A3 was confirmed by DNA analyses. Subclade A25 was used immediately after preparation because this species died in seawater within the day after isolation from host tissues. The other species were used the day after preparation to allow flagellum formation. Each bleached polyp was placed in a well of a 6-well multidish with 5 mL of ASW, and algal cells were added to a density of 5 × 10⁵ cells/mL and washed away after incubation for 5 h with swirling of water at 30 rpm using a handmade mixing apparatus (see   Supplementary Figure S1 (zs220099_FigS1.pdf)). The ASW was changed every day without feeding. Fluorescent beads (Fluoresbrite Plain Microspheres 6.0 µm YG, Polysciences, PA, USA) were administered at 1000 particles/mL.

Identification of algal subclades

DNA was extracted from host tissues according to Sargent et al. (1986). PCR for algal genotyping was performed with a Symbiodiniaceae-specific primer set: 5′-CCGGTGAATTATTCG-GACTGACGCAGTGCT-3′ for the forward primer and 5′-TCCTCC-GCTTATTGATATGC-3′ for the reverse primer. For typing of E. diaphana, 12S rDNA was amplified using the primer set 5′-CCG-GTGAATTATTCGGACTGACGCAGTGCT-3′ and 5′-TCCTCCGCT-TATTGATGC-3′ (Lauretta et al., 2014), For typing of A. atodai, a primer set targeting nuclear ITS, 5′-AGGTGAACCTGCG-GAAGGATCA-3′ (Apakupakul et al., 1999) and 5′-CTGTTCCG-CAACTTGGAC-3′, was used for amplification. PCR products were purified using the GENECLEAN II kit (MP Biomedicals, CA, USA) and sequenced using an ABI BigDye v3.1 Cycle Sequencing Kit and a 3730x DNA sequencer. The full-length ITS sequences were subjected to phylogenetic analyses using MEGA7 (Kumar et al., 2016). After aligning of sequences by Muscle, alignments were corrected by hand. Neighbor-joining was used to construct phylogenetic trees.

RNA extraction and reference transcriptome sequencing

One symbiotic and one apo-symbiotic A. atodai used for RNA extraction were cultured under LED light irradiation (7 µmol s^–1m^–2) with an 18 h light:6 h dark cycle and 24 h dark conditions, respectively, in ASW (Instant Ocean Sea Salt, Instant Ocean, VA, USA) at 20°C. Each sample was placed in a microtube with RNAlater RNA Stabilization Solution (ThermoFisher Scientific, MA, USA) for 1–2 hours, blotted with Kimwipes tissue paper, and then placed into a mixture of 8 µL thioglycerol and 400 µL homogenization buffer. The samples in the thioglycerol and homogenization mixture were ground with a Biomasher II (Nippi, Tokyo, Japan) until pieces of tissue were no longer observed. After standing for 10 min, the mixtures were centrifuged for 2 minutes at 12,000 × g and the supernatants were used for RNA extraction using the Maxwell RSC Plant RNA Kit (Promega Corporation, WI, USA) according to the manufacturer's instructions. The quality and quantity of the extracted RNA were measured using an Agilent RNA 6000 Nano Kit on an Agilent Bioanalyzer (Agilent Technologies, CA, USA) and a NanoDrop spectrophotometer (Thermo Fisher Scientific, MA, USA), respectively. Non-directional library preparation and RNA-seq were outsourced to Filgen (Aichi, Japan). Sequencing was performed using an Illumina NovaSeq6000 instrument (paired-end, 150 bp). The raw read data are available from the DDBJ/EMBL/GenBank database under BioProject accession number PRJDB14295.

Sequence assembly and phylogenetic tree construction

All DNA sequencing reads of the two samples (33,609,878 and 34,973,663 reads for symbiotic and apo-symbiotic samples, respectively) were assembled into contigs using the Trinity program (v.2.11.0) (Grabherret et al., 2011) using a ‘--trimmomatic’ option. The longest open reading frames of the assembled contigs with functional gene annotations were predicted using TransDecoder (v.5.5.0) with a ‘--single_best_only’ option (Haas et al., manuscript in prep.  http://transdecoder.github.io). The A. atodai NPC2 gene was translated into protein. The NPC2 protein sequences of other animal species used for constructing an NPC2 phylogenetic tree were obtained from sequences used in a previous study (Ishii et al., 2019). All NPC2 protein sequences were aligned using MAFFT v.7.490 (Katoh et al., 2013) with ‘--localpair --maxiterate 1000’ options. The alignment was trimmed using TrimAl (v.1.4.1) (Capella-Gutiérrez et al., 2009) using the ‘-gt 0.8’ option, and used for construction of a maximum likelihood tree using IQ-TREE (v.2.1.2) with the WAG + I + G4 model selected using the ModelFinder program in IQ-TREE (Nguyen et al., 2015; Kalyaanamoorthy et al., 2017).

Clonal culture of A. atodai

In addition to morphological identification, ITS sequences matched the registered data (accession no. KT852185) for A. atodai(Larson and Daly, 2016). Anthopleura atodai asexually brooded and expelled small polyps as described by Yanagi and Daly (2004). The founder polyp of A. atodai reproduced approximately 60 new polyps during 9 months of culture (Fig. 1), corresponding to more than one polyp per week under culture conditions. Subsequently, as a population including immature polyps, 20 polyps produced 144 new polyps in 45 days, 0.16 polyps per individual per day on average. Feeding with mashed shrimp was empirically found to promote efficient growth of the anemone compared to Artemia nauplii. Daily feeding may promote asexual reproduction, as it reduces the polyp size from 1 cm at capture to 5 mm in diameter after prolonged culture. Newborn polyps took 3 months to grow to 3 mm in diameter, which is a suitable size for experiments, and more than half a year until maturation for asexual reproduction. Sexual traits were not investigated, but there were no signs of sexual reproduction under our culture conditions.

Fig. 2.

ITS phylogeny of Symbiodinium (clade A) and the symbiont alga in A. atodai.

Identification of symbiont algae in A. atodai and other donor hosts

BLAST results using the obtained ITS sequence of the algae hosted by A. atodai revealed high similarity to sequences suggesting that A. atodai hosts algae in the genus Symbiodinium (clade A). Sequence data containing subclade annotations were collected from DDBJ/GenBank and subjected to multiple alignments. Phylogenetic analyses using the neighbor-joining method revealed that this algal sequence corresponded to a novel subclade in the Symbiodinium clade (clade A) (Fig. 2). We designated this novel subclade A25 in this study. The algal sequence obtained from a polyp of Anthopleura uchidai also constituted a novel subclade in the genus Fugacium (clade F) and is dealt with as F7. Since A. uchidai is known to be polymorphic in symbiosis with algae of clades F and A (Miura et al., 2014), we used subclade F7 algae taken from a single polyp. The other algae in Table 1 matched the registered sequences with the subclade annotations. The diameters of the algal cells were measured and are shown in Table 1. The sizes of the algae differed slightly, ranging from approximately 8–10 µm in diameter.

Fig. 3.

Relative intensity of chlorophyll red fluorescence on the pedal plane. Intensity was evaluated for each individual polyp comparing to day 0 value as 100%. (A) A. atodai treated with Kiredar (n = 5). (B) A. atodai treated with pure quinoclamine (n = 5). (C) E. diaphana treated with Kiredar (n = 3). (D) E. diaphana treated with pure quinoclamine (n = 3). Dark rest periods are shown by shadows.

Bleaching of sea anemones using quinoclamine

After trials for determining drug concentrations and light durations for effective bleaching, we identified a suitable condition composed of repeated cycles of 5-days of drug exposure at 5 µM quinoclamine under continuous light followed by 2-days of rest without the drug in the dark. For use of the herbicide Kiredar, the concentration was adjusted to contain 5 µM of the active ingredient quinoclamine. Figure 3 presents the changes in red fluorescence intensity by tracing each single polyp treated in a well of a 6-well multidish. The patterns of fluorescence decrease were similar between polyps treated with Kiredar and pure quinoclamine. Red fluorescence of algae at the pedal disc decreased to less than 20% during the first 5 days of exposure in A. atodai and less than 10% in 2 days in E. diaphana. Fluorescence intensity at the pedal dropped below the detection level after two cycles of treatment in A. atodai and after one cycle in E. diaphana, whereas small numbers of algae remained in the tentacles and body column. Anthopleura atodai maintained no red fluorescence of algae after three cycles of treatment, and E. diaphana after two cycles (Fig. 4). Complete bleaching was confirmed by observing no recovery of algae in the bleached polyps after normal culture for more than 2 weeks. Exposure to 5 µM quinoclamine for 5 days in the dark caused no obvious decrease in the red fluorescence of algae in A. atodai, suggesting that photochemical reactions caused bleaching. Bleached polyps were reared with feeding for more than 2 years, and asexual reproduction was observed as in wild-type symbiotic polyps. Kiredar treatment was efficient for bleaching other sea anemone species; Anthopleura uchidai (Fig. 5), and Mesactinia ganensis (data not shown). A bleached polyp of A. uchidai spawned eggs 8 months after bleaching with Kiredar treatment.

Algal cells ejected from hosts during Kiredar treatment under continuous light for 16 h were observed using light and fluorescence microscopy. Algal cells are distinguished from host cells and debris by their round shape and size. Almost all ejected algal cells retained a brown color. Red fluorescence of chlorophyll was not reduced in algae ejected by A. atodai or E. diaphana after treatment in comparison with freshly prepared algae, but was reduced in algae ejected by a polyp of A. uchidai hosting clade A (see   Supplementary Figure S2 (zs220099_FigS2.pdf)).

Fig. 4.

Fluorescent photographs of wild and bleached polyps. (A) Wild-type A. atodai. (B) Bleached A. atodai. (C) Wild-type E. diaphana. (D) Bleached E. diaphana.

Pilot comparative experiments on host-symbiont specificity

Ingestion step: Two weeks after complete bleaching, the absence of algae in polyps was confirmed by fluorescent binocular observations, and bleached polyps were used for the re-introduction of algae. For each batch, five bleached polyps were exposed to algae for 5 h, swirling in a well of a 6-well multidish. The presence of algae inside the hosts was observed 7 d after inoculation (Table 2). At 100 cells/mL of algal cells, A. atodai did not incorporate any of the six algal species, including the native symbiont A25, whereas E. diaphana incorporated B1 and A3. Anthopleura atodai did not take up any algae even at 10,000 cells/mL, which is an extremely high density compared to natural situations. On the other hand, E. diaphana took up C1, D1a, and F7 at high density. When an aliquot (20 µL) of mashed shrimp was dispersed together with algae at 10,000 cells/mL in each well, A. atodai incorporated all of the algal species. Clade A25 was not taken up by E. diaphana, even when applied with food.

Fig. 5.

A. uchidai bleached using the herbicide Kiredar at 5 µM effective quinoclamine concentration for 2 weeks. (A) An untreated polyp. (B) A bleached polyp 8 months after bleaching. (C) Fluorescent microscopy revealing absence of algae after Kiredar treatment. (D) Spawning of eggs by a bleached polyp 8 months after bleaching. Scale bar = 1 cm.

Table 2.

Ingestion specificity under different inoculation conditions. Five polyps were used for each experimental batch, and fluorescence was observed 7 days after inoculation. Numbers of polyps which incorporated algae are shown.

Propagation step: Polyps that incorporated algal cells were observed to show changes in algal cell numbers for individual polyps until 21 d after inoculation (Fig. 6). Although accurate numbers could not be counted and the results varied by individual, approximate tendencies of algal cell propagation were distinguished between different combinations of hosts and algae. A25 numbers increased very slowly in the native host A. atodai, but did not increase in E. diaphana. B1 increased slowly in A. atodai, but faster than its native symbiont A25, and quickly propagated in its native host, E. diaphana. A3 revealed patterns similar to those observed for B1. C1 increased to and remained at low densities in both host anemones. D1a showed similar patterns to C1 in A. atodai but varied by individual hosts from an increase to a decrease in E. diaphana. F7 increased once and decreased gradually in A. atodai, but only decreased in E. diaphana. After long-term culture for more than 2 months of polyps inoculated with heterologous symbionts, A. atodai harbored B1 at high densities and C1 at low densities (Fig. 7). A3 showed rapid propagation in the initial phase (Fig. 6) but resulted in low densities in the stable phase in E. diaphana (Fig. 7).

Ejection step: To compare the ejection speeds of nonspecific particles after incorporation, fluorescent plastic beads were mixed with food and fed, and their numbers were traced. Although the speed of the decrease varied among individuals, incorporated beads disappeared in several days, and obvious differences were not observed between the two anemone species (Fig. 8).

Fig. 6.

Propagation of symbiont algae after inoculation into bleached polyps. The vertical axis represents numbers of algae observed in the tentacles and oral disc, and the horizontal axis represents days after introduction of algae. The solid lines represent A. atodai and the dashed lines represent E. diaphana for each polyp. Plots are not shown when the numbers of algae became too great to count. Note that the scale of vertical axes in upper panels is approximately twice that of lower panels.

Fig. 7.

Long-term culture of polyps inoculated with heterologous symbionts. (A) 2 months and (B) 1 year after inoculation of clade B1 into A. atodai. (C) 2 months and (D) 1 year after inoculation of clade C1 into A. atodai. (E) 3 months after inoculation of clade A3 into E. diaphana.

Fig. 8.

Decrease of fluorescent beads once incorporated into the oral disc and tentacles. The solid lines represent A. atodai and the dashed lines represent E. diaphana for each polyp (n = 5). The vertical axis represents numbers of beads observed in one individual polyp, and the horizontal axis represents days after introduction of beads.

Reference transcriptome of A. atodai

To facilitate gene-level analyses of the new model sea anemone, reference transcriptome data were generated using a symbiotic and bleached (apo-symbiotic) individual. Similarity searches identified symbiosis-related genes, including those encoding the conserved sterol transporter Niemann-Pick Type C2 (NPC2) (Dani et al., 2014; Hambleton et al., 2019). Phylogenetic analysis of NPC2 protein sequences showed that A. atodai expressed multiple isoforms, forming clusters corresponding to one canonical and three non-canonical variants of the NPC2 gene, each of which was closely related to the counterpart found in Anemonia viridis (Fig. 9).

Establishment of a clonal strain of A. atodai as a new model system

Here, we established a clonal strain of the sea anemone A. atodai. This strain produced an average of 0.16 polyps per individual per day by asexual reproduction in the laboratory environment. This propagation rate meets the requirements of most molecular and cellular biological experiments. In addition, bleached polyps are capable of asexual propagation. The improvement of food supply methods is critical to achieve successful culture of A. atodai. Mashed shrimp as feed enabled stable maintenance and propagation of not only A. atodai, but also several other species in which long-term culture in the laboratory was difficult by feeding with Artemia nauplii. Tiny polyps can also be cultured by suspending mashed shrimp in culture water, enabling the culturing of tiny polyps and physiological experiments using various anemone species.

An alternative focus is the evolution of endosymbiosis within the genus Anthopleura. In the rocky intertidal zone of middle Japan, there are several sympatric species, such as A. atodai, A. asiatica, A. inornata, and A. uchidai hosting different species of Symbiodiniaceae, and A. pacifica and A. fuscoviridis being aposymbiotic. In addition to molecular phylogenetic analyses, comparative experiments regarding host-symbiont specificities and physiological responses will support the elucidation of evolutionary histories of host Anthopleura and symbiont lineages involving the establishment of their mutual association in the temperate coast of Japan. The establishment of methods for the long-term culture and bleaching of various anemone species in this study will provide opportunities for comparative studies surrounding anemone-alga endosymbiosis.

Bleaching of sea anemones by quinoclamine

Quinoclamine is phototoxic to algae and liverworts. This study showed that it also acts on the removal of dinoflagellate algae of Symbiodiniaceae from sea anemone hosts. As the bleaching pace did not differ between Kiredar and quinoclamine exposure (Fig. 3), the observed bleaching is thought to be due to the action of quinoclamine. Quinoclamine has been shown to inhibit chlorophyll synthesis (Koura et al., 1994), but mechanisms of phototoxicity have not been clarified. Even if chlorophyll synthesis can be inhibited by Kiredar treatment in some cases, it is likely that such inhibition was not a major cause of ejecting algal cells, considering the lack of obvious reduction in algal cells ejected by A. atodai (see   Supplementary Figure S2 (zs220099_FigS2.pdf)). Based on the structural features of quinones, it is presumed that quinoclamine removes electrons from the photosystem (Altland et al., 2011), like the herbicide paraquat (Qian et al., 2009), and results in the generation of reactive oxygen species (ROS), which is suggested to be a major cause of bleaching in corals (Lesser, 2011). This also seems to be the case in sea anemone bleaching, considering our findings that bleaching occurred under light, but not dark, conditions.

Although the property elicited by the action of quinoclamine as the cause of algal cell ejection remains to be clarified, this drug is efficient in eliminating Symbiodiniaceae algae from sea anemone hosts without irreversible harm if treated appropriately, as shown by rearing for long periods accompanied by asexual reproduction. Menthol (Matthews et al., 2016) and 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) used for the herbicide diuron (e.g., Xiang et al., 2013) and heat treatment (e.g., Belda-Baillie et al., 2002) have been utilized to bleach Exaiptasia. However, this study revealed the efficiency of quinoclamine, which is commercially available as a cheap herbicide, as an alternative bleaching tool for the bleaching of a variety of sea anemone species, such as A. uchidai (Fig. 6) and M. ganensis, which died before bleaching by heat treatment (data not shown), in addition to A. atodai and E. diaphana (Figs. 4, 5).

Pilot experiments of comparative studies on symbiotic specificity

To assess the suitability of A. atodai for comparative studies on host-symbiont specificity, small-scale pilot experiments were conducted.

Fig. 9.

A phylogenetic tree of NPC2 proteins. White circles indicate NPC2 isoforms identified in this study.

Specificity at the ingestion step differed between A. atodai and E. diaphana (Table 2). Anthopleura atodai ingested algal cells only when the algae were applied together with food, even when they were native symbionts. These results suggested that foods simply stimulate ingestion activity without masking the surface molecules responsible for recognition of algae. Similar stimulation of algal cell ingestion by adding food was observed in larvae of the coral Fungia scutraia (Schwartz et al., 1999). In contrast, E. diaphana incorporated five algal species without food, revealing a high ingestion activity. The lack of uptake of A25 by E. diaphana can be explained by positive rejection. Algal cell size is unlikely to influence ingestion efficiency because the size range of all algal cells tested in this study was approximately 8–10 µm (Table 1), which was shown to be acceptable for Exaiptasia (Aiptasia) (Biquand et al., 2017). Specific molecules on the algal cell surface may affect ingestion efficiency. In case of the coral Acropora, favored clades were incorporated at low density (1 algal cell/mL, Yamashita et al., 2014), on the other hand, unfavored clades were not even at high density (10,000 algal cells/mL, Wolfowicz et al., 2016). Future studies will be conducted to screen untested clades seeking those rejected by the sea anemones and to clarify the molecular mechanisms of algal cell selection at the ingestion step and to examine whether attractive or repellent molecules are presented by algae and the receptor molecules expressed in the host mouth.

Specificity at the propagation step also differed between the two host species (Fig. 5). Algal cell propagation was slow in A. atodai, except in F7, which increased once and then decreased. In A. atodai, even its native symbiont A25 increased slowly, in contrast to the rapid propagation of B1 in E. diaphana. Exaiptasia diaphana allowed swift propagation of A3, but no propagation of F7. The propagation speed may be a result of a balance between the proliferation rate and decreasing speed of algal cells. The decreasing speed can be divided into the speeds of intracellular digestion, exocytosis, and ejection from the host mouth. Quantifying the speed of each process is also a challenge for future studies. We tried to compare the ejection steps using plastic beads, but no obvious differences were detected (Fig. 7). Considering the observation that ejected algal cells were often found at the bottom of E. diaphana culture ware, E. diaphana seems to eject excessive algal cells produced by high-speed growth. However, ejected algal cells were less abundant in A. atodai cultures. In addition to differences in proliferation speed, there may be different ejection speeds specific to algal species in each host, even if differences were not detected in the ejection speed of plastic beads. Digestion of algal cells by host cells is thought to be another possible reason for the decrease in algal cells, especially in A. atodai.

Another aspect of interest is the regulation of algal cell densities in the steady state in the host body, which should also be determined by the balance between proliferation and decrease. Preliminary long-term observations revealed a lower density of algae in some heterologous combinations of host and algal species than in native combinations (Fig. 7; Medrano et al., 2019). The rate of algal cell proliferation and decrease might change from the initial phase of inoculation to the stable phase depending on algal cell densities in the host body and host responses to algae. Specificity at each step concerning the initiation and maintenance of endosymbiosis must be examined in order to understand the total specificity of cnidarian-algal association in dynamic equilibrium. Comparative studies using multiple host-symbiont systems will help to dissect the complex of multiple steps responsible for endosymbiosis and to choose approaches for the identification of molecular mechanisms by focusing on each step.

We present a reference transcriptome analysis to support molecular-level studies in A. atodai. For instance, A. atodai expresses a repertoire of genes encoding multiple NPC2 variants (Fig. 9), which are known to be important for the transport of sterols from symbiotic algal cells to host cells, thus providing a basis for comparative molecular studies on endosymbiosis using current (Dani et al., 2014; Hambleton et al., 2019) and new model cnidarian species. Future comparative omics studies concerning heterologous host-symbiont combinations, with a focus on these symbiosis-related genes, will shed light on the molecular aspects of each step of endosymbiosis that determine host-symbiont specificity, and these studies are now ready to be launched.