Nudibranchia are a morphologically diverse and colorful group of non-shelled sea slugs, belonging to the Heterobranchia (Burmeister, 1837) and consisting of the suborder Doridina and the suborder Cladobranchia. Over 4,000 nudibranch species have been described and, due to their colorful appearance, they fascinate scientists and non-scientists alike and are often called “butterflies of the sea” (Anderson, 1995). Current research on Nudibranchia focuses on assessing their biodiversity (e.g., Eisenbarth et al., 2018; Fritts-Penniman et al., 2020; Korshunova et al., 2021), their developmental biology and life cycle (Page, 1993; Kristof and Klussmann-Kolb, 2010; Ahmadian et al., 2016; Togawa et al., 2019), and phylogenetic relationships within the different groups (e.g., Carmona et al., 2013; Goodheart et al., 2015a,b; Karmeinski et al., 2021Korshunova et al., 2021). Furthermore, because most Nudibranchia lost their protective shell, alternative defense strategies, such as mimicry of food sources (Gosliner and Behrens, 1990), calcareous needles (Cattaneo-Vietti et al., 1995), the synthesis of toxic metabolites (Bogdanov et al., 2017), and storing and using cnidocysts “stolen” from their cnidarian food source (Obermann et al., 2012; Goodheart et al., 2018) are investigated. Especially the potential pharmaceutical relevance of their secondary metabolites (reviewed by Cimino and Gavagnin, 2006; Putz et al., 2010; Fisch et al., 2017) makes them an interesting group for researchers. Photosymbiosis is only found in the Cladobranchia that comprise approximately 1,000 species and that can be identified by the lack of gills and their large dorsal appendices, the cerata, that also function as respiratory organ. However, photosymbiosis in Cladobranchia is not well understood.

Like most photosymbiotic animals, Cladobranchia acquire their photobionts anew in each generation (i.e., horizontally, instead of vertically from their parents), which is the most common mechanism of photobiont acquisition in animals (reviewed in Melo Clavijo et al., 2018). However, only Cladobranchia evolved two different modes of horizontal photobiont acquisition – out of the water column or from photosymbiotic cnidarians (Figure 1). Within the Cladobranchia, members of the Dendronotoidea (Allmann, 1845), such as Melibe engeli Risbec, 1937 (Figure 1), experienced morphological modifications that resulted in a fan-like mouth opening (Gosliner, 1987), allowing them to effectively ingest the photobionts out of the water column (Bleidissel, 2010; Burghardt and Wägele, 2014). To the best of our knowledge, adults of the genus Melibe are the only sea slugs to obtain their photobionts this way. The vast majority of Cladobranchia, however, acquire the photobionts by feeding on cnidarian prey and “stealing” the cnidarians’ photobionts (Rudman, 1981; Kempf, 1984; Wägele, 2004). This is a unique photobiont acquisition strategy in animals that evolved in Cladobranchia probably multiple times in the superfamilies Arminoidea, Iredale and O’Donoghue (1923), Fionoidea Gray (1827), and Aeolidioidea Gray (1827) Figure 1).

Subsequent to the uptake, Cladobranchia selectively phagocytize the photobionts into epithelial cells of their digestive gland system (DGS). The DGS branches throughout the entire body and, particularly, into the cerata (Figures 2A,B). Once the algae are phagocytized in the epithelial cells, they remain in their coccoid state and are surrounded by the phagosomal membrane (Figures 2C,D; Wägele and Johnsen, 2001; Wakefield and Kempf, 2001). In most cladobranchs, like Flabellina affinis (Gmelin, 1791) or Cratena peregrina (Gmelin, 1791) (Figure 1), the algae are then rapidly digested, while some species, like Berghia stephanieae (Valdés, 2005) (Figure 2A) are capable to maintain the algae photosynthetically active for a few days (Monteiro et al., 2019). Regarding the organismic interaction that we describe in this review we refer to the general symbiotic terminology. Within that terminology, the existing subcategories of photosymbiosis are defined based on the beneficial aspects and the time-span of interaction, following the definitions by Kempf (1991). Based on this, maintaining the photobiont for a short term is here referred to an unstable photosymbiosis, because the algae reside intracellularly, but the slugs, like B. stephanieae, tend to digest the algae within a couple of days, or expel them from the cells and secrete them in a viable state in the feces. Species that have evolved an unstable photosymbiosis do not appear to benefit from the photosynthesis performed by the acquired photobionts. They are neither able to maintain their symbionts nor their biomass when solely relying on the photobiont as source of nutrition (Kempf, 1991; McFarland and Muller-Parker, 1993; Bleidissel, 2010; Monteiro et al., 2019). Yet, some taxa have evolved the ability to establish a stable photosymbiosis, maintaining the photobionts for months. For a couple of these photosymbiotic species, like Melibe engeli and Phyllodesmium briareum (Bergh, 1896), it has been shown that the photobiont can fully support the host, enhancing growth and the ability of long-term reproduction without a reduction in the quantity and quality of egg-masses under regular light conditions (Kempf, 1984; Burghardt and Wägele, 2004, 2006, 2014; Burghardt et al., 2005, 2008a,b; Burghardt and Gosliner, 2006). These observations support the idea that at least these Cladobranchia species and Symbiodiniaceae are involved in a mutualistic symbiosis, which is unique in gastropods. More closely related Sacoglossa sea slugs, that are in a sort of photosymbiosis with chloroplasts of their algal prey, are not able to grow, or even maintain their biomass, if they are exclusively dependent on their ingested chloroplasts for more than a couple of weeks (Pelletreau et al., 2012; Christa et al., 2014). Further, they are not able to maintain the quantity and quality of egg-masses during periods of food depletion (Cartaxana et al., 2019). Hence, Cladobranchia provide a unique opportunity to understand which genomic adaptations are needed to evolve a mutualistic photosymbiosis in sea slugs. Comparative analyses of photosymbiotic and non-photosymbiotic Cladobranchia with Sacoglossa and Cnidaria might uncover if these genomic adaptations are based on convergent evolution or if, for instance, epigenetic modifications are involved in activating specific genes in photosymbiotic slugs.

The selective incorporation of photobionts in Cladobranchia is a complex process. It remains unknown how the slugs are able to distinguish between photobionts and plankton and further digest all plankton (e.g., in Melibe) or tissues of the cnidarian prey while maintaining the photobionts intact. The mechanisms of photobiont recognition by the epithelial cells of the DGS have not yet been addressed, while in cnidarians the photobiont recognition is based on a set of animal host pattern recognition receptors (PRRs). Specific microbe associated molecular patterns (MAMPs) of the photobiont (Neubauer et al., 2016, 2017; Mansfield and Gilmore, 2019) bind to the PRRs, which triggers downstream signaling cascades to maintain the photobiont (Davy et al., 2012). The PRRs of Cladobranchia might be similar to those of their cnidarian host, as they incorporate the same photobiont and hence need to recognize the same MAMPs. However, Cladobranchia could also just use different PRRs to recognize other MAMPs to identify the equivalent photobiont. The incorporation process might even be more similar to the selective uptake of chloroplasts by more closely related members of sea slugs belonging to the Sacoglossa (Chan et al., 2018; Melo Clavijo et al., 2020). Using phylogenetic and domain specific analyses of the respective receptors will help to understand if the relationship of the respective PRRs is matching the taxonomy of the host. In combination with subsequent functional analyses, for instance by gene expression and manipulation, a list of candidate genes can be generated that could be involved in photobiont recognition. Therefore, Cladobranchia are the only animals that could permit a direct comparison of photobiont recognition mechanisms, in particular PRRs, between a primary (Cnidaria) and secondary (Cladobranchia) host of the same photobiont. At the same time, Cladobranchia allow comparisons with mechanisms of chloroplast recognition in the more closely related Sacoglossa. These analyses may provide insights into the relevance of PRRs in photosymbiosis across taxonomically divergent animal taxa. Furthermore, they may elucidate how conserved the signaling pathways for initiating photosymbiosis are – independent from the taxonomic lineage of the photobiont.

The evolution of a stable photosymbiosis in Cladobranchia is still not well understood. It appears that photosymbiosis evolved several times independently in different Cladobranchia superfamilies (Figure 1). For instance, it might have evolved independently in the Dendronotoidea and a monophyletic group including Arminoidea, Fionoidea, and Aeolidioidea. This would explain the evolution of the two different acquisition modes. Alternatively, photosymbiosis could also have evolved in each of the superfamilies independently. It is furthermore unknown whether unstable and stable photosymbiosis evolved separately, or if stable photosymbiosis evolved from unstable photosymbiosis.

Commonly, photosymbiotic Cladobranchia have a highly branched digestive gland system and specialized digestive structures, such as large circular chambers, fine tubules and cisternae that are located at the tips and harbor the photobionts (Burghardt and Wägele, 2014). These structures are considered to enlarge colonizable space and optimize light attenuation, increasing photobiont density and photosynthesis, respectively (Rudman, 1991; Burghardt et al., 2008a,b; Moore and Gosliner, 2011). Yet, despite of having a highly branched digestive gland system, species of Phestilla or Limenandra (Figure 1), and species like Melibe leonina (Gould, 1852) or Phyllodesmium kabiranum Baba, 1991, are non-photosymbiotic (Figure 1). Aside of this exception the genus Phyllodesmium shows a correlation between photosymbiosis and a highly branched digestive gland system (Rudman, 1991; Burghardt et al., 2008a,b; Moore and Gosliner, 2011). A highly branched DGS might not be a prerequisite to evolve a stable photosymbiosis, but it appears to be advantageous to harbor larger numbers of photobionts and to optimize photosynthesis.

The cnidarian prey is considered important for the stability of the photosymbiosis in Cladobranchia (Wägele et al., 2010). For instance, members of the genus Phyllodesmium, in which most species have a stable photosymbiosis, feed exclusively on xeneid cnidarians belonging to the Alcyonacea. However, Pteraeolidia semperi (Bergh, 1970), also in a stable photosymbiosis, obtains the photobionts from Hydrozoa, so that the food source is not strictly connected to photosymbiosis (Figure 1). Instead, it seems to be based on obtaining specific secondary metabolites for defense purposes (Bogdanov et al., 2017) and the feeding preference is rather taxon-specific. Independent of the source of the photobiont, the efficiency of the photosymbiosis may be further influenced by the specific algal taxon. For instance, in other stable photosymbiotic systems, such as Cnidaria-Symbiodiniaceae and Bivalvia-Symbiodiniaceae associations, the animal hosts could benefit from more physiologically resilient photobionts during increased ambient temperature (Hume et al., 2016; Cziesielski et al., 2018; Mies, 2019; Cunning and Baker, 2020). When corals are in symbiosis with multiple symbiodiniacean genera, the active removal of less resilient strains results in an adaptive bleaching, which might increase the animal’s fitness and improve the stability of the symbiosis considerably (Ziegler et al., 2014; Bayliss et al., 2019; Chen et al., 2019). So far, only a few studies have investigated the diversity and composition of Symbiodiniaceae in Cladobranchia (Loh et al., 2006; FitzPatrick et al., 2012; Ziegler et al., 2014; Wecker et al., 2015; Yorifuji et al., 2015). Nevertheless, these studies have not uncovered any correlation between specific Symbiodiniaceae taxa and the ability to establish a photosymbiosis with Nudibranchia. It rather seems that symbionts are taken up from the cnidarian prey indiscriminately. Future analyses of photobiont abundance in cladobranchs in comparison to their cnidarian prey will help to understand if the slugs are able to distinguish between Symbiodiniaceae taxa and selectively expel less beneficial photobionts. Comparative metabarcoding analyses, as available for some cnidarians and their Symbiodiniaceae composition (Fujise et al., 2021), is still lacking for sea slugs, but are needed to reveal the relevance for a stable photosymbiosis in Cladobranchia. It might be further worth to investigate, if the slugs play a role in genotype dispersion of symbiodiniaceans in marine habitats. The fact that symbionts are transferred from a sessile to a motile host could potentially enhance the dispersion of symbiodiniaceans and could change the composition of clades and strains in environmental populations (Parker, 1984). This could be beneficial for other sessile photosymbiotic animals and their symbiont uptake, facing environmental changes with different adapted symbionts (Umeki et al., 2020).

Regardless of the stability of the photosymbiosis, little is known on how nutrients are exchanged between the slugs and the algae. The phagosomal membrane surrounding photobionts in cnidarians is known as symbiosome (Hill and Hill, 2012). The symbiosome plays a crucial role in the successful establishment of the symbiosis. Transporters relevant for nutrient exchange, i.e., sugars from the algae to the animal and dissolved inorganic compounds from the animal to the algae, are situated in the symbiosomal membrane (Sproles et al., 2018). In cladobranchs it is unknown if and which transporters are present on the phagosomal membrane and how comparable it is to the symbiosome in terms of functionality. Future immuno-histochemical studies could give valuable insights into the localization of photosymbiosis-relevant receptors and their role in the nutritional exchange between the two partners, while metabolomics could provide important information on the interdependent nutrient exchange between both partners. Further, it remains to be shown whether in species with a stable photosymbiosis the symbiosis is beneficial for the host and whether the symbiosis is mutualistic or even some sort of parasitism as proposed for other photosymbiotic animals by some authors (Lesser et al., 2013; Blackstone and Golladay, 2018; Androuin et al., 2020).