Sacoglossan mollusks are suctorial herbivores that live on the cell sap of coenocytic macroalgae (Jensen 1993).As an adaptation for suctorial feeding, these sea slugs have uniserial radula and show narrow specificity for their algal diet (Handeler and Wagele 2007). With few exceptions,algae that have been identified as food for sacoglossan sea slugs belong to the class Ulvophyceae sensu Floyd and O’Kelly (1990). However, food organisms are not known for many sacoglossans yet.

Some sacoglossan mollusks incorporate algal chloroplasts inside their cells, and the incorporated chloroplasts are functional for days to months, depending on the species. Trench et al. (1972) provided conclusive evidence that photosynthetic assimilates are released from the chloroplasts in the sea slug cells. This incorporation and maintenance of foreign chloroplasts is known as kleptoplasty (e.g., Clark et al. 1990). It has long been suspected that genes have been transferred from algae to the sea slugs, because many components of photosystems in active algal plastids are very short-lived, with turnover times especially under bright light condition (Schuster et al. 1988, Vass et al. 1992, Warner et al. 1999). Thus, if the plastids are maintained in an active state for weeks and months, it is logical to assume that the mollusks could produce necessary proteins to keep the photosynthetic activity because the photosynthetic proteins in chloroplasts are turned over quickly (often in less than a day). Recently, accumulating evidences show that algal nuclear genes have been laterally transferred to these sacoglossan species (Rumpho et al. 2001, 2008, Pierce et al. 2007, Schwartz et al. 2010).

Sacoglossan mollusks bind avidly to their food algae. In most cases the sacoglossan mollusks eat less than ten species of algae throughout their life cycle, and the chloroplasts from only one or two species could be sequestered and maintained in the sea slug cells. This feeding specificity is also necessary for the development of successful lateral gene transfer which might render the acquired chloroplast photosynthetically active for long time.

Here, we screened for Korean sacoglossans containing functional chloroplasts using an in vivo chlorophyll fluorescence measuring system (pulse amplitude modu-

[Table1.] Sacoglossans identified in this study

Sacoglossans identified in this study

lation, PAM). The feeding specificity of the sea slugs was tested using 37 species of cultured algae.

The sacoglossans (Table 1, Fig. 1) were collected monthly from Wando and Gangneung, Korea from July 2009 until October 2010. They were identified following Koh (2006) and Internet resources on the subject (Sea slug forum 2010). The animals were kept in IMR medium (Klochkova et al. 2006) in 9 × 5 cm Petri dishes at 15-20°C in 12 : 12 h L : D cycles with 15 ㎛ol photons m^-2 s^-1 provided by cool-white fluorescent bulbs.

The algae used in the feeding experiment, their collection site or source and conditions for laboratory culture are given in Table 2. All algae were grown in IMR medium as described above. While performing feeding experi-

[Fig. 1.] Sacoglossans identified in this study. (A) Elysia atroviridis. (B) Elysia nigrocapitata. (C) Elysia ornata. (D) Ercolania boodleae. (E) Stiliger berghi. (F) Stiliger sp. (G) Placida dendritica.

ments for Elysia atroviridis and E. ornata, each potential algal food was given to sea slugs for a period of 10-14 days. If the algae were completely or partially consumed during that time, other tested algae were added. However, if a particular algal species was not consumed, it was considered non-eatable and was replaced with another algal species. In case of E. nigrocapitata, the longest starvation period exceeded 4 months. In case of Ercolania boodleae, Placida dendritica, Stiliger berghi, and Stiliger sp. the starvation time was 2-7 days, since they do not cope well with longer starvation times. Chlorophyll fluorescence measurements were performed in vivo in the fed and starved sea slugs using Diving-PAM Chlorophyll a Fluorometer (Walz, Effeltrich, Germany).

Genomic DNA was isolated using a genomic DNA isolation kit following the manufacturer’s instructions (Intron Biotech, Seoul, Korea). Polymerase chain reaction (PCR) was performed using primers displayed in Table 3, with an initial denaturation at 95ºC for 3 min, followed by 35 cycles of amplification (denaturation at 95ºC for 20 s, annealing at 55ºC for 30 s and extension at 72ºC for 1 min and 1.5 min for 16S and 28S rDNA, respectively) with a final extension at 72ºC for 10 min. Amplified PCR products were purified using a gel extraction kit (Qiagen, Valencia, CA, USA) and cloned into a T-easy vector (Promega, Madison, WI, USA) to determine DNA sequences. The phylogenetic analyses were performed using sequences for Placida spp. (Fig. 2) obtained from GenBank (NCBI 2010). Alderia spp. (A. willowi and A. modesta) and Ercolania fuscata were used as outgroups. The DNA sequences were aligned using ClustalW version 1.6 matrix (European Bioinformatics Institute, Cambridge, UK) at 50% of delay divergent cut off and the obtained alignments were manually refined. Maximum parsimony (MP) analyses were performed using Windows based program, MEGA

[Table 2.] List of algae used in the slug feeding experiment their collection site and conditions for laboratory culture and results of sacoglossan feeding experiment

List of algae used in the slug feeding experiment their collection site and conditions for laboratory culture and results of sacoglossan feeding experiment

version 4.1 (MEGA 2007, Tamura et al. 2007, Kumar et al. 2008). The MP analyses were tested for robustness by bootstrapping with 1,000 replications.

Elysia atroviridis (Fig. 1A)

This species was reported from Korea by Koh (2006) under the name E. flavomacula Jensen. The latter name,

[Table 3.] Oligonucleotide primers used for amplification of 16S and 28S rDNA

Oligonucleotide primers used for amplification of 16S and 28S rDNA

[Fig. 2.] Phylogenetic analyses in Placida dendritica.

however, was regarded as a synonym of E. atroviridis (Sea slug forum 2010).

Diet. The animals ate little but continuously when food was available. In the field, they were always found on Bryopsis spp. In the laboratory, they fed specifically on Bryopsis spp. They also fed on the sporophytes of Derbesia tenuissima, but went back to Bryopsis spp. when the algae were available (Table 2). Prolonged periods of starvation of up to 2 months could be tolerated.

Seasonality. This species appeared in Wando in December and had peak of population from January to mid

March. During this period, it was possible to collect over 40 animals during one survey. In other months of the year, this species was not detected at this site.

In vivo chlorophyll fluorescence assay. In animals starved for 1 week, F_v/F_m was 0.31-0.76, which represented 38.7-95% of the chlorophyll fluorescence values in intact plants of B. plumosa.

This species was reported from Japan and Hong Kong (Sea slug forum 2010). This is the first report of this species in Korea. Our specimens ranged from several mm to 1.7 cm in length, and ranged in color from light to very dark green when fed and pale grayish-yellow when starved. Parapodia were leaf-like when flattened, with a wide black band along the reverse side edges and thin white band along the upper side edges. The head was black, while the rhinophores and areas around eyespots were white.

Diet. In the field, the animals fed on Cladophora sakaii. In laboratory culture, they switched to feed on Chaetomorpha moniligera and Phyllodictyon orientale. Some animals (< 30%) switched to feed on Bryopsis spp. and sporophytes of Derbesia (Table 2) after 4 month-long starvation. Species C. moniligera and P. orientale did not grow in the sampling site of E. nigrocapitata.

Seasonality. In October, 58-214 animals were easily collected from a sampling site in Wando. However, from late November to mid March, only 2-6 animals were collected from the same site during each survey. In other months of the year, this species was absent from this site.

In vivo chlorophyll fluorescence assay. In animals starved for 1.5 months, F_v/F_m was 0.53-0.67, which represented 66.2-83.7% of the chlorophyll fluorescence values in intact coenocytic green algal thalli.

This species was reported from Jeju Island by Koh (2006). We herein extend knowledge on its distribution on the southern coast of Korea.

Diet. The animal ate continuously when food was available. In the field it was found on B. plumosa. In laboratory, it fed specifically on Bryopsis spp. and vesicular gametophytes of D. tenuissima (Table 2).

Seasonality. During the course of this study, we found 1 animal from Wando site on October 4, 2009 and were unable to find more specimens from the same sampling site at other times during that year.

In vivo chlorophyll fluorescence assay. In animal starved for 1 week, F_v/F_m was 0.54, which represented 67.5% of the chlorophyll fluorescence values in intact plants of B. plumosa.

Ten animals were found on C. moniligera from Gangneung and 4 animals were found in-between rhizoids of B. plumosa attached epiphytically to the red alga Plocamium sp. from Wando. It is noteworthy that Chaetomorpha spp. do not grow in that locality in Wando. E. boodleae is a common sacoglossan species in the Peter the Great Bay (Russia, Sea of Japan) (Minichev 1976). Martynov (1997) listed this species for the intertidal zone of South Sakhalin (Sea of Okhotsk). This species was reported from Jeju Island and eastern coast of Korea (Koh 2006). We herein extend knowledge on its distribution on the southern coast of Korea.

Diet.The animals were fed on C. moniligera in the laboratory and ate continuously when food was available. Two 4-5 mm long animals consumed a 20 cm-long Chaetomorpha filament within 6 h. The animals collected from Wando consumed B. plumosa occasionally, however those collected from Gangneung fed only on C. moniligera (Table 2). Short periods of starvation (4-7 days) could be tolerated.

Seasonality. In Gangneung, this species was more abundant in spring and the early summer months, correlating with the abundance of C. moniligera. In Wando, we found 4 animals on October 4, 2009 and were unable to find any more specimens from this site in other months of the year.

In vivo chlorophyll fluorescence assay. In animals fed until the analysis, F_v/F_m was 0, implying that algal protoplasm and chloroplasts were digested/damaged after consumption.

Since a number of very similar-looking taxa around the world have been identified as P. dendritica (Sea slug forum 2010), we subjected our specimens to 16S and 28S rDNA analyses. The animals from Wando showed affinity with an isolate from Spain identified as P. dendritica. However, it differed from it in both 16S and 28S rDNA (Fig. 2). The taxonomic re-appraisal of Korean species will be necessary when more sequences are available in the GenBank, however at this time we called it P. dendritica based on morphology. Our specimens were aeolid-like in shape; the anterior surface was covered with cerata down each side of the body. The body was dirty white, fairly transparent, and the color of food in the digestive system showed through the body.

This species is a new record for Korean sacoglossan fauna, although very common and distributed all around the Korean peninsula. In north Atlantic waters, P. dendritica is commonly found associated with Codium fragile (Evertsen and Johnsen 2009). During the course of this study, we collected 305 animals from various Bryopsis species. Thus, this species is more associated with Bryopsis spp. in Korea. It is noteworthy that C. fragile and C. tomentosum were very abundant in Wando sampling site, but P. dendritica was found mainly on Bryopsis plants. We found only 10 animals feeding on Codium spp. in the field (Table 1).

Diet.The animals ate continuously when Bryopsis spp. were available. In the laboratory, approximately 10% of animals switched to C. minus or Derbesia sporophytes when no Bryopsis spp. were available (Table 2), but went back to Bryopsis spp. as soon as they were provided. In the absence of food, the green coloration turned brownish within 2-3 days and then faded. The animals usually died after 10-14 days of starvation.

Seasonality.This species had peak of population in summer and early autumn. From July to mid October, it was possible to collect 50-100 animals from several plants of Bryopsis spp. during one survey. From December to March, the population declined, so that only 10-12 animals were collected from over 100 large plants of Bryopsis spp. But the host algae Bryopsis spp. were very abundant in winter months when sea slugs were reduced in number and were less abundant in summer and early autumn.

In vivo chlorophyll fluorescence assay. A difference in F_v/F_m values was observed depending on an algal diet. In animals continuously fed on B. plumosa until the assays, F_v/F_m was 0.12, but it abruptly reduced to 0 within 6-12 h after feeding. In animals continuously fed on C. minus and D. tenuissima,F_v/F_m was always 0.

This species is a new record for Korean sacoglossan fauna. It is known from Japan and Peter the Great Bay, Russia (Sea slug forum 2010). We herein extend knowledge on its distribution on the southern coast of Korea. Our specimens were aeolid-like in shape; the anterior surface was covered with irregularly arranged and irregularly sized cerata down each side of the body. The body was slightly yellowish to dirty white, fairly transparent, and the color of food in the digestive system showed through the body wall. The rhinophores were tapering and cylindrical, with a dark band about one-third of the way down from the tip.

Diet.This species has been reported to live and feed on the red algae, especially Neorhodomela larix. In the field, our specimens fed on Polysiphonia japonica var. forfex. In laboratory culture, they switched to feed on Antithamnion densum, A. glanduliferum and Aglaothamnion byssoides (Table 2). Short periods of starvation (4-7 days) could be tolerated.

Seasonality. During the course of this study, we found 5 animals from Wando on August 10, 2009 and were unable to find more specimens from the same sampling site in other months of the year.

In vivo chlorophyll fluorescence assay. In animals continuously fed on the red ceramiaceaen algae until the assays, F_v/F_m was 0, implying that algal protoplasm and chloroplasts were digested/damaged after consumption.

This species is a new record for Korean sacoglossan fauna. Three animals were found on C. sakaii from Wando. They were aeolid-like in shape and 6-7 mm long. The anterior surface was covered with irregularly arranged and irregularly sized cerata down each side of the body. The body was dirty white, fairly transparent, and the color of food in the digestive system showed through the body wall.

Diet. In the field, the animals fed on C. sakaii. In laboratory culture, they switched to feed on C. moniligera (Table 2), but did not eat it continuously and did not consume all the protoplasm from the damaged Chaetomorpha cells. It allowed C. moniligera cells to regenerate or make intracellular protoplasts.

Seasonality. During the course of this study, we found 3 animals from Wando on October 4, 2009. No other specimens were recovered from the same sampling site in other months of the year.

In vivo chlorophyll fluorescence assay. In animals continuously fed on C. moniligera until the assays, F_v/F_m was 0, implying that algal protoplasm and chloroplasts were digested/damaged after consumption.

During the course of this study, we collected 7 sacoglossan species from the same site in Wando where B. plumosa grows as a dominant species all year round, except in July and August. E. atroviridis, E. nigrocapitata, and P. dendritica were the most abundant sacoglossans in that area. In vivo chlorophyll fluorescence analysis indicated that all three species of Elysia possessed active chlorophylls and, thus, photosynthetically functional chloroplasts even after prolonged starvation. Considering that these species belong to sacoglossans, the results suggest that they have kleptoplasts (Yamamoto et al. 2009). Thus, our results add two more potential species, E. atroviridis and E. nigrocapitata, to the list of sacoglossans that are known to possess functional chloroplasts (e.g., Yamamoto et al. 2009). Moreover, in E. atroviridis and E. nigrocapitata the kleptoplasty seems to be long-term (for over 1 month to 4 months). To date, only four species are known to perform long-term maintenance of acquired plastids, including E. chlorotica, E. timida, E. crispata, and Plakobranchus ocellatus (Wagele et al. 2010).

Kawaguti and Yamasu (1965) described functional chloroplasts of C. fragile in the ‘hepatic diverticula’ of E. atroviridis by detailed investigation through electron microscopy, but did not present data to support the possibility of an endosymbiotic relationship. More detailed studies using transmission electron microscopy analysis and PAM measurements will be necessary to fully understand kleptoplasty by E. atroviridis and E. nigrocapitata.

In the case of P. dendritica fed on Bryopsis, a decrease of F_v/F_m was observed within 6-12 h after feeding, while in animals fed on Derbesia and Codium,F_v/F_m was always 0. Yamamoto et al. (2009) reported that Placida sp. fed on Codium showed a F_v/F_m value of 0.48, although it survived for less than a week during starvation. On the contrary, Evertsen and Johnsen (2009) reported that P. dendritica fed on C. fragile until assays did not have any photosynthetic responses. More studies will be necessary to confirm in vivo chloroplast functionality according to algal diet in Placida.

Also, Yamamoto et al. (2009) reported a high F_v/F_m value of 0.6 in Stiliger ornatus; however in our study, S. berghi and Stiliger sp. were negative species, together with E. boodleae.

In sea slugs feeding on green algae, the diet was restricted to either siphonous or siphonocladous species, especially Bryopsis spp. Elysia nigrocapitata could switch from its primary diet (siphonocladous green algae) to Bryopsis after being starved for 4 months, and E. boodleae occasionally fed on B. plumosa instead of Chaetomorpha. Considering that in the field three sacoglossans were constantly feeding on B. plumosa and two species could also feed on it occasionally, one might expect a rapid shortage of food resources from a small sampling site. On the contrary, B. plumosa was very abundant in that sampling site, despite the seasonal abundance of particular sea slug species. Bryopsis spp. are known for their ability to regenerate viable plants from protoplasm extruded in seawater upon wounding (Tatewaki and Nagata 1970, Pak et al. 1991, Kim et al. 2001). Protoplast formation in Bryopsis was interpreted as a method of propagation because one branch of a plant could generate hundreds of viable new cells spontaneously in seawater (Kim et al. 2001). Thus, grazing by sacoglossans might have been beneficial for Bryopsis abundance in that area. It is noteworthy that all the green algal species specifically consumed by these mollusks have ability of protoplast regeneration (Klotchkova et al. 2003, Kim and Klotchkova 2004). Wounding in Chaetomorpha can also trigger the development of aplanospores or swarmers (Klotchkova et al. 2003). Thus, the appearance of sea slugs feeding on Chaetomorpha could also contribute to its propagation to some extent. More ecological studies will be necessary to elucidate the symbiotic nature among the sea slugs and algae involved in this specific feeding and kleptoplasty.

This work was supported by Korea Research Foundation grant No. 20100020701 to GHK. We thank Dr. John A. West (University of Melbourne, Australia) for providing many algal strains used in slug feeding experiments and Dr. Nina G. Klochkova (Kamchatka State Technical University, Russia) for her useful consultations on the subject. Special thanks to Ms. Kyong Hwa Cha (Kongju National University) for assistance during sample collection.