Elysia timida (Risso, 1818) three decades of research & A. Garrote–Moreno

Elysia timida (Risso, 1818) three decades of research.— During the last 30 years, studies on Elysia timida (Risso, 1818) have addressed various aspects related to food sources, photosynthetic efficiency of kleptoplasts, population genetics, chemical ecology and reproductive biology, both in the Mediterranean Sea and in the Mar Menor coastal lagoon. E. timida shows a strong specific interaction with Acetabularia acetabulum , retaining functional chloroplasts for at least 45 days and obtaining extra energy in periods when food resources are scarce. It shows control of parapodia, avoiding pigment photodestruction under oversaturated light conditions. The chemical ecological relationships established between E. timida and its potential predator fish, Thalassoma pavo , have also been evaluated, and it has been found that that the extracts of the mollusc contain repellent and unpalatable polypropionate compounds. Population genetics has demonstrated the genetic divergence between populations showing high and significant values of FST and genetic distances, and at least six privative alleles that are not shared with Mediterranean populations have been detected in lagoon populations. This sacoglossan is a poecilogonic species, and its lagoon populations show a greater reproductive output than Mediterranean populations; they produce a greater number of egg masses and embyros per individual, and the capsules have a wider diameter.


Introduction
In recent decades, the sacoglossan Elysia timida (Risso, 1818) (Gastropoda, Opisthobranchia, Sacoglossa) has attracted the attention of numerous researchers (fig 1). The reasons for this growing interest are the easy accessibility of its habitat, the large number of individuals in a small areas (due to its gregarious behaviour associated with a low dispersal ability), the extraordinary physiological and ecological adaptations (great versatility and adaptability to stress conditions) and particular reproductive and survival strategies.

Aposematic versus cryptic coloration
The sacoglossan show a degree of feeding specialization that is similar to that of terrestrial insects. E. timida feeds with priority on the green alga, A. acetabulum. In the field, these animals are white and are conspicuous in winter or spring, but they are very cryptic when the algal food is calcified (Marín & Ros, 1992). The white colour of E. timida could serve two main purposes: chloroplasts exploitation and predation avoidance. The animal uses the white of the parapodia to control the amount of light used for the photosynthesis of kleptoplasts. The design of the mollusc coloration may appear relatively cryptic during periods of algal calcification (summer: white animal on white background) and relatively conspicuous during periods characterized by low algal calcification (autumn: white animal on green background).
Experiments conducted on feeding preference (Giménez-Casalduero, 1997b), have shown that the white models were more cryptic in the summer background than in the autumn background ( fig. 2). This suggests that the optimal colour for a possible prey is white rather than green or red, in particular for E. timida, bearing in mind its defensive characteristics   -19801981-19851986-19901991-19951996-2000-2005. Crypsis degree measured as number of attacks of the green, red and white artificial models over the three backgrounds (autumn, spring and summer), which represents the seasonal change of the algae A. acetabulum (modified from Giménez-Casalduero, 1997b).   Giménez-Casalduero, 1997b).  Fig. 3. The ability of algae to refuge measured as number of attacks of the green, red and white artificial models over the three backgrounds (autumn, spring and summer), which represent the seasonal change of the algae A. acetabulum. In this experiment the background was three-dimensional, simulating the algae structure with white plastic stick; the density of the stick was proportional to the alga density during the three seasons simulated (modified from Giménez-Casalduero, 1997b). Fig. 3. Capacidad de refugio de las algas, medido como número de ataques a los modelos artificiales verdes, rojos y blancos, en las tres estaciones (otoño, primavera y verano), los cuales representan el cambio estacional del alga A. acetabulum. En este experimento el fondo fue tridimensional, simulando la estructura de las algas mediante palitos blancos de plástico; la densidad de los palitos fue proporcional a la densidad del alga durante las tres estaciones simuladas (modificado de Giménez-Casalduero, 1997b). Spring Autumn and population dynamics. The colour design of a prey is considered to be cryptic when the potential predator cannot distinguish it from the sea floor on which it lies.
This strategy is very common in many elysiidae, one example being Bosellia mimetica (Trinchese, 1891), which feeds on Halimeda tuna (Ellis et Solander) and is virtually indistinguishable. Aposematic coloration is used as a warning to potential predators of their toxic or unpleasantness. The close relationship between E. timida and A. acetabulum, apparently favours a design that can be either cryptic or aposematic. Gendron & Staddon (1983) showed that if the probability of detecting a prey is inversely related to both the search rate and the degree of crypsis, specialist herbivores could be subject to greater selective pressure by natural enemies than generalist animals. Laboratory experiments show that E. timida is less consumed when the calcified algal biomass increases ( fig. 3) (Giménez-Casalduero, 1997b). The nutritional quality decreases, but it provides an effective refuge to avoid predation (Hacker & Steneck, 1990;Duffy & Hay, 1991).
Some sacoglossans however have adapted their diet to algae with chemical defence, because they subsequently avoid predation and accidental ingestion by macro-herbivores, or because the sequestered secondary metabolites from the algae protect them from their own predators (Hay, 1991;Hay et al., 1994) A. acetabulum contains no deterrent metabolites and E. timida must synthesize its own defence (Gavagnin et al., 1994).

Photosynthetic efficiency of kleptoplasts
The sacoglossans are specialized herbivores that can retain chloroplasts intact within the cells of the gastrointestinal tract (Marín & Ros, 1988). The manipulation of unchanged foreign structures is a complex mechanism involving both cell recognition and acceptance, and physiological adaptation within the digestive glands. Also, its operation as a photosynthetic source is an extraordinary evolutionary feature (Marín & Ros, 1991;Rumpho et al., 2000Rumpho et al., , 2001. From the point of view of chloroplasts' autonomy it is interesting to note the time period that chloroplasts retain their normal functions within their host cells. Several authors have demonstrated the photosynthetic capacity of E. timida that consists of their using symbiont chloroplasts (kleptoplasts) retained in their intestinal diverticula (Ros & Rodriguez, 1985;Marín & Ros, 1989;Wägele & Johnsen, 2001;Evertsen et al., 2007;Händerler et al., 2009Händerler et al., , 2010. In addition, Greene (1970) and Greene & Muscatine (1972) not only described the release of organic compounds in chloroplast symbionts but noted that the metabolism of symbiotic chloroplasts can be modified by the association. They found abundant evidence showing the transfer of photosynthates from the chloroplast to the host tissue sacoglossans. In the case of isolated Acetabularia, chloroplasts have been reported that C 14 labelled only chlorophyll joins the first 70', after which it is extracted from the algal tissue (Trench & Smith, 1970). The chloroplast-animal association is unable to synthesize chlorophyll, glycolipids, ribulose-bisphosphate carboxylase or membrane proteins. Therefore chloroplasts are unable to grow or have a real division when they are inside the host animal (Trench & Ohlhorst, 1976). This inability to synthesize DNA and RNA or regenerate plastidial proteins means that elysoidea must obtain chloroplasts in each generation (Marín & Ros, 1993;Wägele et al., 2010).
Effectiveness in this symbiotic system depends on the ability of the host to incorporate active chloroplasts and retain them functionally active in its glands for as long as possible (a high turnover rate would not provide benefits). Holding capacity of active chloroplasts is revealed by analysing the changes in the proportion of chlorophyll a over time compared to other pigments within the body of sacoglossa. Chlorophyll a has been described as one of least stable pigments over time as compared to carotenes or degradation products from chlorophyll, which have an extraordinary resistance to destruction. The experiment carried out by Giménez-Casalduero (1997b) on E. timida showed a lower degradation rate for chlorophyll than for the remaining pigments during the experimental time (starved and dark conditions) ( fig. 4). However, the absence of light seems to be a determining factor for the incorporation of new chloroplasts. This result was interpreted as the ability of sacoglossa to assimilate or eliminate the remaining pigments, because E. timida tends to retain and maintain high levels of chlorophyll a, declining its net concentration over time more slowly than the other pigments. In an effective system, the retained chlorophyll must endure being photosynthetically active for long periods of time. Gross oxygen production per milligram of chlorophyll reaches values between 2.910 and 3.397 mg O 2 /mg of chlorophyll a in E. timida tissues and these levels are maintained over time, even in starvation periods (table 1) (Giménez-Casalduero, 1997b)

Metabolic benefits and consequences
As mentioned above, the chloroplasts retained in the intestinal diverticula of E. timida are able to perform photosynthesis (Ros & Rodriguez, 1985;Marín & Ros, 1989). The use of photosynthetic energy has also been demonstrated by Giménez-Casalduero & Muniain research (2006, 2008. These latter works studied the photosynthetic parameters of E. timida in the Mar Menor lagoon environment, and the researchers found that the results of P-I curves were explained by the model based on the kinetics of Michaelis-Menten (1913) and described by Pérez (1989). This model showed a rapid saturation of the photosynthetic apparatus at relatively low irradiance values, seeming to contradict the animal habitat (shallow and lit areas). However, this fast saturation might be influenced by a control of parapodia, which would avoid pigment photodestruction under oversaturated light conditions. During the experimental period (Giménez-Casalduero & Muniain, 2006), E. timida individuals were observed closing and opening their parapodia, illustrating the ability of sacoglossans to regulate the photosynthetic production described by Rahat & Monseline (1979), Monseline & Rahat (1980) and Jesus et al. (2010): parapodia were completely opened when individuals were exposed to low irradiance and gradually closed with increasing the irradiance values. Furthermore, in the field, the sacoglossan is commonly found with completely closed parapodia on a well-lit day (Giménez-Casalduero & Muniain, 2006).
The production versus respiration relationship (P/R) was initially used to estimate compensation irradiance for symbiotic associations between corals and zooxanthellae: (i) P/R = 1 indicates that most of the fixed carbon during photosynthesis is consumed on respiration by the host and the 'guest'; (ii) P/R > 1 means that the amount of fixed carbon by photosynthesis exceeds the basal metabolic requirements of the host; (iii) P/R < 1 indicates that the host has to feed on other sources to fulfil the amount of carbon required for its basal metabolic needs (McCloskey et al., 1978). Theoretically, the estimated mean P/R values of E. timida populations from the Mar Menor coastal lagoon are above 0.91, these values are considered high in the coral-zooxanthellae symbioses. If we consider that sacoglossans are more active than corals and possess higher respiration values, then production values of chloroplasts should also be higher to obtain similar indices. Therefore, photosynthetic efficiency of symbiont chloroplasts on E. timida is fairly high. The parameters of the P-I curve obtained by Giménez-Casalduero & Muniain (2006) provide a useful baseline information to perform further studies on photosynthetic energy generated from the relationship between the mollusc and the acquired chloroplasts, as well as on the importance of such energy for primary metabolism of lagoon sacoglossans (Giménez-Casalduero & Muniain, 2006;Jesus et al., 2010).
Poecilogony is an intraspecific variation in the mode of larval development and can be found in single or different populations (Bouchet, 1989;Krug, 1998;Krug et al., 2007). E. timida is considered a poecilogonic species showing direct and lecitotrophic development in the Mediterranean area (Rahat, 1976;Marín & Ros, 1993;Garrote-Moreno, 2007). Direct development increases the likelihood of offspring survival in the absence of food limitation. However when the food is scarce, lecitotrophic development allows exploration of new areas with A. acetabulum. So, is there coevolution between E. timida and A. acetabulum? Or, conversely, is it an adaptive response of E. timida to environmental factors, such as the greater or lesser availability of food?
Energy invested in reproduction is also affected by the presence of 'extra energy' from symbiotic chloroplasts. In general, the number of spawn is greater if animals have extra energy input; a disturbance of normal conditions initiates a massive reproductive activity and the number of eggs per spawn is greater in starved conditions (Giménez-Casalduero & Muniain, 2008).   The sacoglossan reaction in a starved situation is to invest energy in reproduction, using a mating behaviour and a particularly highly efficient co-occurrent sperm transfer (Schmitt et al., 2007). Reproductive effort is very high the first few days, using both photosynthetic and metabolic energy (which explains a decrease of up to 20% in size the first 18 days) (Giménez-Casalduero, 2008).

Adaptations to environmental stress conditions
Estuaries and coastal lagoons constitute transitional environments whose main characteristic is the instability of their physical-chemical parameters, mainly the concentration of salt (Cognetti & Maltagliati, 2000;Gamito et al., 2005). Physiological adaptations are very important for survival in these extreme environments. These adaptations imply a high energetic cost that is reflected in growth and reproduction rates; thus genetic selection aimed at minimizing such costs can be assumed (Wright, 1977;Remmert, 1988). When environmental conditions change unpredictably in space and time, variation in life-history traits can be an adaptive response to selection (Meyers & Bull, 2002). For instance, stable dispersal dimorphisms can evolve when the quality of habitat patches varies over time and there is spatial heterogeneity in environmental fluctuations (Mathias et al., 2001). Two morphs of E. timida have been described in the Mar Menor coastal lagoon and shallow environments of the Southwest Mediterranean coast, showing differences in body size, colour and chlorophyll concentration (Giménez-Casalduero, 1997a). It is important to stress, hewever, that differences in these parameters have also been found between lagoon (regardless of morphotype) and Mediterranean populations (Giménez-Casalduero, 1999). For example, simultaneously comparing the mean dry weight and mean chlorophyll a concentration from two lagoon morphotypes and the Mediterranean population, we recognized differences between all of them ( fig. 5). These differences have recently been confirmed by Jesus et al. (2010).
Moreover, changes have been observed in the spawn of populations from Mar Menor lagoon and the Mediterranean Sea, with the reproductive effort in lagoon sacoglossa being higher than in Mediterranean populations (Giménez-Casalduero, 1997b, 1999. The variability in the number of eggs per spawn is smaller in the Mediterranean population than the lagoon one which varies from 88 to 725 eggs per spawn ( fig. 6). In some species, such as Elysia viridis (Montagu, 1804) the increase in reproductive success may result from a change in salinity of environment (Hagerman, 1970). High salinity of the Mar Menor lagoon could influence the increase in E. timida reproductive output, but it does not explain the differences among environments or among morphs.
According to Giménez-Casalduero (1997a, 1997b, the differences between brown and green lagoon animals, could have diverse explanations: i) We could be facing a new variety of E. timida, considering the Mar Menor lagoon as a shelter for this species (high salinity, extreme temperature changes, isolation from the Mediterranean Sea, few predators) and the capacity of this elysoidea to have a direct larval development and being a strong candidate to develop genetic adaptations in lagoon conditions; ii) Another possible scenario would be the higher versatility converning food in this species, with a great capacity to adapt to new food resources. González-Wangüemert et al. (2006) estimated the degree of genetic divergence between populations living inside and outside the Mar Menor coastal lagoon. They showed genetic identity values ranging from 0.9 to 0.87 in the Mediterranean and coastal lagoon populations. These values indicated a subspecies-rank separation according to several authors (Avise, 1974;Thorpe, 1983;Mariani et al., 2002). However, it is important to stress that there are different methods of delimiting species and subspecies (Sites & Marshal, 2003). However, significant genetic differences (F ST values and Nei's genetic distances) were found between lagoon and marine populations, confirmed by the principal component analysis. This genetic differentiation (Mar Menor/ Mediterranean Sea populations) has been confirmed using other molecular markers and species (González-Wangüemert & Pérez-Ruzafa, in review;Vergara-Chen et al., 2010a, 2010b. Nevertheless, no genetic differences were found between the two E. timida morphs previously described ( fig. 7). In fact, according to data from Giménez-Casalduero (1999), both morphs can occur either in the Mediterranean or Mar Menor populations, which implies The box itself represents 50% of all cases, and extends from the 25th to the 75th quartiles. The line inside the box shows the media. Points beyond the whiskers (outliers) are drawn individually. (Source of data from: Marín & Ros, 1993;Giménez-Casalduero, 1997a, 1997b, 1999unpublished data, 2007).   & Ros, 1993;Giménez-Casalduero, 1997a, 1997b, 1999datos no publicados, 2007).  Giménez-Casalduero, 1997b, 1999. Giménez-Casalduero, 1997b, 1999 Giménez-Casalduero, 1997b).

Fig. 5. Concentración media de clorofila a (A) y peso seco medio en los tejidos de las poblaciones mediterránea y de laguna de Elysia timida (modificado de
Mg chlorophyll a / dry weight (g) 4.5 Mediterranean Sea that the existence of morphotypes is not related to the lagoon environment. A sign of the versatility of the animal food was observed in the course of an experiment (Giménez-Casalduero, 1997b) in which animals kept in laboratory conditions for 28 days and feeding on A. acetabulum showed an increased concentration of chlorophyll from the start day supply of algae, while in dark conditions the values remained more or less constant over time ( fig. 8). The interpretation of these data could be the possible ability of this sacoglossa to feed on different algal species at least during stress situations (Ros & Rodriguez, 1985;Marín & Ros, 1989, 1993, and this would explain the different proportion of pigment found in dun animals as compared to the remaining populations analysed. The hypothesis of a certain food versatility is somewhat striking, given the close relationship between E. timida and A. acetabulum. It is likely that the presence of the dun morph was due to degradation of pigments in aging chloroplasts. However, the animal can feed on other algae during periods of food shortages, according to some observations in the laboratory (unpublished data).