Photosynthesis from stolen chloroplasts can support sea slug reproductive fitness.

Some sea slugs are able to steal functional chloroplasts (kleptoplasts) from their algal food sources, but the role and relevance of photosynthesis to the animal host remain controversial. While some researchers claim that kleptoplasts are slowly digestible 'snacks', others advocate that they enhance the overall fitness of sea slugs much more profoundly. Our analysis shows light-dependent incorporation of 13C and 15N in the albumen gland and gonadal follicles of the sea slug Elysia timida, representing translocation of photosynthates to kleptoplast-free reproductive organs. Long-chain polyunsaturated fatty acids with reported roles in reproduction were produced in the sea slug cells using labelled precursors translocated from the kleptoplasts. Finally, we report reduced fecundity of E. timida by limiting kleptoplast photosynthesis. The present study indicates that photosynthesis enhances the reproductive fitness of kleptoplast-bearing sea slugs, confirming the biological relevance of this remarkable association between a metazoan and an algal-derived organelle.

Introduction

Sacoglossa is a group of sap-sucking sea slugs that feed on macroalgae. The most striking feature of some of these sea slugs is their ability to digest the algal cellular content while retaining intact functional chloroplasts (kleptoplasts) within the cells of their digestive gland [1,2]. This process of stealing plastids from algal cells (kleptoplasty) is more common in single-celled eukaryotes, such as foraminiferans, dinoflagellates and ciliates [3]. Recently, Van Steenkiste et al. [4] identified short-term functional kleptoplasts in two species of marine flatworms. However, among metazoans, the capacity for long-term maintenance (up to several months) of functional chloroplasts remains a unique feature of a few species of sacoglossans [5,6]. Functional kleptoplasty occurs despite the absence of genetic material with an important role in chloroplast regulation, as these genes have been transferred to the algal nucleus over the evolution of endosymbiosis [7]. Although horizontal gene transfer was suggested as the primary reason underlying the long-term maintenance of photosynthesis in sacoglossan sea slugs, more recent studies found no evidence of genes supporting photosynthesis in the animal nuclear DNA [8].

The importance of kleptoplasty for the nutrition and metabolism of sacoglossan sea slugs remains controversial. Most studies have shown that photosynthesis plays an important role in individual survival and fitness over periods of food scarcity [9-12], while others argue that it is not essential for slugs to endure starvation [13]. Transcriptomic data on the sea slug Elysia chlorotica show that chloroplast sequestration leads to significant changes in host gene expression patterns throughout uptake and maturation, similar to that occurring during the establishment of symbiosis in corals, and suggest parallels between these animal–algal interactions [14].

Earlier radiolabelled carbon-based studies indicate the translocation of photosynthesis-derived metabolites from functional kleptoplasts into sacoglossan sea slug tissues [15-17]. Trench et al. [15] reported 14C-labelling within 2 h of incubation in the renopericardium, the cephalic neural tissue and the mucus-secreting pedal gland of Elysia crispata and Elysia diomedea. Recently, Cruz et al. [18] have shown initial light-dependent incorporation of 13C and 15N in digestive tubules followed by a rapid translocation and accumulation in kleptoplast-free organs of Elysia viridis (i.e. in tissues involved in reproductive functions such as the albumen gland and gonadal follicles). However, no direct relation between photosynthesis and reproductive investment of kleptoplast-bearing sea slugs has been established.

In the present study, we investigated the putative role of kleptoplast photosynthesis in the reproduction of the sacoglossan sea slug Elysia timida by (i) tracking short-term light-dependent incorporation of inorganic carbon and nitrogen into animal tissues using compound-specific isotope analysis (CSIA) of fatty acid methyl esters (FAMEs) and high-resolution secondary ion mass spectrometry (NanoSIMS), and (ii) investigating the effects of limiting photosynthesis (rearing animals under reduced light levels) in the number and fatty acid (FA) composition of spawned eggs. We report strong experimental evidence for a role of photosynthesis in the reproductive investment and fitness of a kleptoplast-bearing sea slug.

Results

Light-dependent incorporation of C and N

NanoSIMS isotopic imaging

Semi-thin section imaging combined with NanoSIMS imaging showed that 13C- and 15N-labelling was not homogeneously distributed in different sea slug tissues, as 13C- and 15N-hotspots could be observed (figures 1–3). NanoSIMS images from individuals incubated in light for 6–36 h with 13C-bicarbonate and 15N-ammonium showed marked 13C- and 15N-labelling in kleptoplast-bearing digestive tubules (figure 1; electronic supplementary material, figure S1). Individuals incubated in the dark for 36 h displayed no 13C-enrichment (electronic supplementary material, figure S1). By contrast, 15N-labelling was observed in the digestive tubules of sea slugs incubated in the dark for 36 h, although at a much lower level than in conspecifics incubated under light (electronic supplementary material, figure S1).

Figure 1

13C and 15N incorporation in the digestive tubules of Elysia timida. Light microscopy picture and corresponding δ13C and δ15N NanoSIMS images of E. timida incubated in artificial seawater enriched with 2 mM NaH13CO3 and 20 µM15NH4Cl, for 6 h in the presence of light. Digestive tubules (dt) appear green on the light microscopy micrographs due to the presence of numerous chloroplasts. After 6 h of incubation strong 13C and 15N enrichment is observed in these structures. (Online version in colour.)

Figure 3

13C and 15N incorporation in the gonadal follicles of Elysia timida. Light microscopy pictures and corresponding δ13C and δ15N NanoSIMS images of E. timida in an isotopic dual labelling pulse-chase experiment incubated in artificial seawater enriched with 2 mM NaH13CO3 and 20 µM 15NH4Cl, in the presence of light for pulse (12 and 36 h) and chase (3 h), and in the dark for 36 h. fo, gonadal follicles; dt, digestive tubules; ag, albumen gland. (Online version in colour.)

Marked 13C- and 15N-labelling was also observed in the albumen gland and the gonadal follicles (both kleptoplast free) of E. timida incubated in the light for 6–36 h with 13C-bicarbonate and 15N-ammonium (figures 2 and 3; electronic supplementary material, figure S1). 13C and 15N-labelling were still observed in the chasing phase, after individuals were transferred to fresh non-labelled artificial seawater (ASW) for up to another 12 h (figures 2 and 3; electronic supplementary material, figure S1). Again, no 13C-enrichment and lower 15N-labelling was observed in the albumen gland and the gonadal follicles of sea slugs incubated in the dark for 36 h, when compared to animals incubated in the presence of light (figures 2 and 3; electronic supplementary material, figure S1).

Figure 2

13C and 15N incorporation in the albumen gland of Elysia timida. Light microscopy pictures and corresponding δ13C and δ15N NanoSIMS images of E. timida in an isotopic dual labelling pulse-chase experiment incubated in artificial seawater enriched with 2 mM NaH13CO3 and 20 µM 15NH4Cl, in the presence of light for pulse (12 and 36 h) and chase (3 h), and in the dark for 36 h. ag, albumen gland; dt, digestive tubules. (Online version in colour.)

Fatty acid analysis

The most abundant FA (greater than 5% relative abundance) observed in E. timida were the saturated FA (SFA) 16 : 0 and 18 : 0, the monounsaturated FA (MUFA) 18 : 1n−9 and the polyunsaturated FA (PUFA) 18 : 2n−6, 18 : 4n−3, 20 : 4n−6, 20 : 5n−3, 22 : 4n−6 and 22 : 5n−3 (electronic supplementary material, table S1). In the presence of light, individuals incubated in 13C-bicarbonate enriched ASW for up to 36 h showed an increasing incorporation of 13C into FA over time (figure 4; electronic supplementary material, table S2). Incorporation of 13C occurred in all of the most abundant E. timida FA, except 18 : 4n−3 (figure 4). Levels of 13C-labelling decreased from SFA and MUFA precursors to longer-chain PUFAS. However, levels of incorporation of 13C in longer-chain FA 22 : 4n−6 and 22 : 5n−3 were higher than in 20 : 4n−6 and 20 : 5n−3, respectively (figure 4). In the chasing phase, when individuals were transferred to fresh non-labelled-ASW, levels of 13C in FA stopped increasing (figure 4). When animals were kept under dark conditions during 36 h of incubation with 13C-bicarbonate enriched ASW, sea slugs’ FA showed no 13C-enrichment, with an incorporation equivalent to that of conspecifics incubated in the presence of light but in non-labelled-ASW (figure 4; electronic supplementary material, table S2).

Figure 4

13C incorporation in the main fatty acids of Elysia timida.13C (‰) in most abundant fatty acids of E. timida as a function of time (h) in an isotopic labelling pulse-chase experiment in artificial seawater enriched with 2 mM NaH13CO3 in the presence of light (open circles) or in dark-incubated specimens for 36 h (closed circles). Mean ± s.e., n = 3. (Online version in colour.)

Acetabularia acetabulum FA composition showed a lower diversity to that of E. timida (electronic supplementary material, table S1). Lower relative abundances of long-chain PUFA were found in the macroalgal tissue when compared to E. timida. PUFA 20 : 4n−6 was present in a relative abundance of 0.26% in the algae compared to 6.15% in the sea slug, while 22 : 4n−6, a major FA in E. timida (10.13%), was not present in A. acetabulum.

Effects of light treatment on egg masses

Pairs of E. timida initiated mating by meeting head-to-head and starting penis protrusion (electronic supplementary material, figure S2A). Animals mutually inserted their penis into the partner's female aperture, located at the base of the right parapodium. Spiral-shaped egg masses (electronic supplementary material, figure S2B) were spawned by E. timida in both light and quasi-dark conditions, although light treatment affected the number of spawning events. Sea slugs reared in regular light (40–160 µmol photons m−2 s−1) produced 7.5 ± 0.2 egg masses per pair (mean ± s.e.), corresponding to seven or eight egg masses over the 28-days experimental period. Spawning activity was more variable in pairs reared under reduced light levels (5 µmol photons m−2 s−1) and ranged from 1 to 5 egg masses per pair (3.0 ± 0.7, mean ± s.e.). The number of eggs produced by animals reared in regular light was significantly higher (t9 = 3.521, p = 0.007) than for sea slugs reared under quasi-dark conditions (238 ± 13 versus 129 ± 30 eggs slug−1 week−1, respectively; figure 5). FA concentrations per egg were not significantly affected by light treatments (electronic supplementary material, figure S3 and table S3). FA composition was similar in E. timida individuals and egg masses (electronic supplementary material, tables S1 and S3).

Figure 5

Fecundity of Elysia timida. Number of eggs spawned by E. timida exposed to a 14 : 10 h light/dark photoperiod and a scalar irradiance of 40–160 µmol photons m−2 s−1 (regular light) or 5 µmol photons m−2 s−1 (reduced light) for 28 days. The line is the median, the x represents the mean, top and bottom of the box are the 75% and 25% percentile, and the whiskers represent the maximum and minimum values. Animals were fed continuously with Acetabularia acetabulum. Differences between treatments were significant at p = 0.007.

Discussion

NanoSIMS isotopic imaging of the sea slug E. timida showed inorganic 13C incorporation in kleptoplast-bearing digestive tubules of light-exposed animals. However, light-dependent carbon incorporation was not restricted to these kleptoplast-bearing cells, and rapid accumulation (within 6 h) was observed in kleptoplast-free organs such as the albumen gland and gonadal follicles.13C incorporation was not detected in the tissues of E. timida exposed to full darkness. Thus, our data clearly demonstrate that inorganic 13C was photosynthetically accumulated into functional kleptoplasts and subsequently translocated to other sea slug tissues, probably through soluble C-compounds (e.g. sugars) or FA.

Translocation of photosynthetically acquired carbon to animal tissues was previously identified in other species of sacoglossan sea slugs: E. crispata, E. diomedea, E. viridis and Plakobranchus ocellatus [15-18]. Using 14C, Trench et al. [15] observed labelling of kleptoplast-bearing digestive tubules after 15 min of light incubation for E. crispata and E. diomedea. Carbon incorporation was also detected in kleptoplast-free organs, such as the renopericardium (after 1 h), the cephalic neural tissue and the mucus-secreting pedal gland (after 2 h) and the intestine (after 5 h) [15]. Trench et al. [16] observed that 14C was incorporated into glucose and galactose in E. viridis, while Ireland & Scheuer [17] reported carbon incorporation in sugars and polypropionates for P. ocellatus. Using electron microscopy combined with NanoSIMS imaging, Cruz et al. [18] observed 13C-labelling after 1.5 h in starch grains of kleptoplasts present in the kleptoplast-bearing digestive tubules of E. viridis, but 13C-labelling was also found in the cytoplasm surrounding the photosynthetic organelles. After longer incubation times (1.5–12 h), 13C-labelling was detected in E. viridis organs involved in reproduction, namely the albumen gland and gonadal follicles [18]. Evidence of fast translocation of photosynthates to kleptoplast-free animal tissues is not compatible with a previously proposed hypothesis that kleptoplasts are slowly digestible food reserves and that photosynthates produced are not continuously made available to the slug [13,19].

Sea slugs showed a much higher level of 15N enrichment in their tissues when incubated in the presence of light. Light-dependent incorporation of 15N was previously reported for E. viridis [18,20]. Teugels et al. [20] identified glutamine synthetase (GS)-glutamate synthetase (GOGAT) as the main pathway involved in N incorporation in the kleptoplasts. Hence, kleptoplasts may not only provide energy and carbon skeletons, but could also play a role in protein synthesis. De novo protein synthesis has been shown to occur for plastid-encoded membrane proteins in E. chlorotica, even after several months of starvation [21]. Contrary to 13C, our NanoSIMS imaging of E. timida recorded 15N incorporation in the dark. Nitrogen incorporated in specimens incubated in the dark (albeit significantly reduced) could result from the glutamine dehydrogenase (GDH) pathway in mitochondria [18,20].

The labelling with 13C was done in the absence of A. acetabulum, the macroalgal food source of E. timida, safeguarding that the labelled FA detected were not obtained heterotrophically (i.e. through grazing on A. acetabulum). Instead, 13C-labelled FA must have been synthesized in the kleptoplasts of the digestive tubules and eventually translocated to other animal cells. Additionally, labelled FA could have been produced in the animal cells through elongation/desaturation reactions using labelled precursors translocated from the kleptoplasts. In fact, the presence of labelled 22 : 4n−6 in E. timida is a direct evidence that the latter process occurred, as this FA was not present in A. acetabulum.

It was generally assumed that animals were unable to biosynthesize PUFA de novo since, presumably, they lacked specific desaturases required to produce 18 : 2n−6 (LA; linoleic acid) [22]. However, several findings have challenged this long-held assumption and it was recently shown that Δ desaturases genes enabling de novo PUFA biosynthesis are widespread among invertebrates [23]. De novo biosynthesis of PUFA can occur via different pathways [24]. Tracking of 13C-labelled FA allowed us to infer the main pathway of PUFA biosynthesis in E. timida (electronic supplementary material, figure S4). A general dilution of the 13C signal was observed along the FA biosynthetic pathway from saturated and monounsaturated precursors 18 : 0 and 18 : 1n–9 to longer-chain PUFA. However, an increase in the 13C signal was observed in the last elongation steps of 22 : 5n−3 and 22 : 4n−6 production from 20 : 5n−3 (eicosapentaenoic acid, EPA) and 20 : 4n−6 (arachidonic acid; ARA), respectively. This finding indicates that the carbon donor (malonyl-CoA) during this elongation process was 13C-enriched and thereby preferentially provided by kleptoplasts. Torres et al. [25] reported that methylmalonyl-CoA incorporating kleptoplast fixed-carbon is used by sacoglossan sea slugs in the synthesis of UV- and oxidation-blocking polypropionate pyrones by the action of FA synthase-like proteins. Pyrones could be critical for maintaining long-term photosynthetic activity in sacoglossan sea slugs by serving antioxidant and photoprotective roles [17,25].

Dietary PUFA have been shown to modulate marine invertebrate gametogenesis, embryogenesis and larval development [26-28]. The levels of PUFA recorded in female tissues and embryos of the sea snail Crepidula fornicata were related to its reproductive output [29]. Bautista-Teruel et al. [30] reported that the reproductive performance in the gastropod Haliotis asinina was linked to diets with increased levels of PUFA, such as EPA and ARA. The latter FAs are precursors of prostaglandins, a group of biologically active compounds participating in marine invertebrate reproduction [31,32]. Hence, the assembly of photosynthesis-driven long-chain PUFA as shown in our study for E. timida (in the kleptoplasts and elsewhere in animal cells using photosynthesis-derived precursors) are likely to play a crucial role in the reproductive output of this species and increase evolutionary fitness.

Fecundity, assessed as the number of eggs spawned by E. timida along a four week-period, was significantly higher in sea slugs exposed to regular light (40–160 µmol photons m−2 s−1) than in animals reared under reduced light (5 µmol photons m−2 s−1). Under a period of resource shortage (i.e. limited photosynthesis under reduced light levels), E. timida clearly changed its reproductive energy investment by decreasing the number of spawned egg masses. Shiroyama et al. [33] observed higher number of eggs spawned by Elysia atroviridis when fed under regular light (30 µmol photons m−2 s−1) than when kept under reduced light conditions (1 µmol photons m−2 s−1).

In our study, complete darkness was not used to inhibit photosynthesis in order not to disrupt diel biorhythms and animal behaviour. The sea slug E. viridis was reported to become inactive under full darkness, rarely seeming to feed [9]. Hence, an extremely low light intensity, enough to substantially limit kleptoplast photosynthesis, was used instead of full darkness [34]. Animals in both light treatments were observed attached to the macroalgae, and the cellular content of A. acetabulum was emptied similarly in both light treatments. This indicates that the heterotrophic feeding ability of sea slugs was not affected by the experimental conditions. Although we cannot rule out a direct effect of experimental light conditions in sea slug spawning, we hypothesize that observed differences in fecundity were due to resources provided by kleptoplast photosynthesis. While reproductive investment was reduced when photosynthesis was limited, E. timida allocated similar amounts of FA to individual eggs regardless of light treatment. This is particularly relevant for egg viability in species such as E. timida, in which offspring success depends exclusively on the parental provisioning of endogenous reserves to fuel embryonic development and early larval life, until lecithotrophic larvae are able to metamorphose to the imago of the adult and feed on exogenous food sources [35].

In conclusion, we report the allocation of photosynthates to kleptoplast-free organs involved in the reproduction of E. timida, along with photosynthesis-driven assembly of long-chain PUFA, and higher sea slug fecundity under regular light than in quasi-dark conditions. These results indicate that kleptoplast photosynthesis increases the reproductive investment of E. timida. It has been shown that kleptoplasty in Sacoglossa contributes to survival and fitness in periods of food scarcity, in some cases allowing individuals to endure several months of starvation [9,10]. We hypothesize that functional kleptoplasty in sacoglossan sea slugs may further potentiate species’ success by maximizing its reproductive output.

Material and methods

Animal collection and maintenance

Specimens of E. timida (Risso, 1818) (electronic supplementary material, figure S2A) were collected in Puerto de Mazarrón in the Mediterranean Sea, Spain. Sampling of E. timida and its algal food, A. acetabulum (Linnaeus) P. C. Silva, 1952, was done by scuba diving at a depth of approximately 2 m. Animals and macroalgae were kept in aerated seawater collected at the sampling site and transported to the laboratory within 48 h. Sea slugs and macroalgae were maintained for two weeks in a 150 l recirculated life support system (LSS) operated with ASW at 18°C and a salinity of 35. The photoperiod was kept at 14 h light : 10 h dark, with a photon scalar irradiance of 60 µmol photons m−2 s−1 being provided by T5 fluorescent lamps. Photon scalar irradiance was measured with a Spherical Micro Quantum Sensor and a ULM-500 Universal Light Meter (Heinz Walz GmbH, Germany). The laboratory adaptation period was chosen to ensure replicability in feeding and the light history of the animals at the beginning of the experiment.

Dual isotopic labelling incubations

Isotopic dual labelling pulse-chase experiments were conducted in closed-systems (1 l glass bottles, three independent containers per treatment). Labelled-ASW was prepared in accordance with Harrison et al. [36], but using NaH13CO3 (13C isotopic abundance of 99%, Sigma-Aldrich) and 15NH4Cl (15N isotopic abundance of 98%, Sigma-Aldrich) to a final concentration of 2 mM and 20 µM, respectively (labelled-ASW). Non-labelled-ASW (control-ASW) contained NaHCO3 and NH4Cl (Sigma-Aldrich) in the same concentrations as the isotopically enriched ASW. Sea slugs were incubated in labelled- and control-ASW at 18°C, in the absence of their food source, and under a photon scalar irradiance of 100 µmol photons m–2 s–1 (measured inside the glass bottles). Additionally, sea slugs in labelled-ASW were incubated in full darkness (electronic supplementary material, table S4). Dark conditions served as a control for light-independent carbon and nitrogen incorporation. The pulse of isotopic dual labelling started 1 h after the onset of the light period. A subset of three individuals kept in labelled-ASW and exposed to light were sampled after 6, 12, 24 and 36 h of incubation (pulse phase), quickly rinsed with distilled water, flash frozen in liquid nitrogen and stored at –80°C until further FA analysis. An additional individual kept in labelled-ASW and exposed to light was collected at each of the referred time points, rinsed and fixed in 0.2 M cacodylate buffer containing 4% glutaraldehyde and 0.5 M sucrose and stored at 4°C for 24 h before tissue preparation for secondary ion mass spectrometry imaging (NanoSIMS 50 l). Individuals from labelled-ASW incubated in dark conditions and from control-ASW exposed to light were sampled after 36 h as described above for FA and SIMS analysis. The remaining individuals in labelled-ASW and exposed to light were transferred to fresh control-ASW. During this chase period, a subset of individuals was collected after 3 and 12 h as described above, for FA and SIMS analysis (electronic supplementary material, table S4). In light treatments, the light was set constant throughout the 48 h experiment.

High-resolution secondary ion mass spectrometry (NanoSIMS) isotopic imaging

Tissue preparation for NanoSIMS imaging was done according to Cruz et al. [18] (see electronic supplementary material, method S1). Large areas of interest were imaged with a NanoSIMS 50 L secondary ion mass spectrometer. This allowed imaging of the subcellular distribution of 13C and 15N enrichment in the exact same areas of the imaged histological overviews described above, enabling a direct correlation of structural and isotopic images. All measurements were performed using the following analytical conditions: 16 keV primary ion beam of Cs+ focused to a beam spot of ca 100–150 nm and counting 12C2−, 13C12C−, 14N12C− and 15N12C− ions in electron multipliers at a mass resolution of greater than 8000 (Cameca definition), enough to resolve potential interferences in the mass spectra. Images captured with NanoSIMS 50 l were processed using the L'IMAGE software (Larry R. Nittler, Carnegie Institution of Washington, Washington DC, USA). Regions of interest selecting individual anatomic structures were defined, and distribution maps of 13C/12C and 15N/14N ratios were obtained by taking the ratio between the drift-corrected 13C12C− and 12C12C− images, and 15N12C− and 14N12C− images, respectively. Five stacked planes were used for each image. 13C and 15N enrichment values in the figures were expressed as delta notations, δ13C = (Cmes/Cnat − 1) × 1000 and δ15N = (Nmes/Nnat − 1) × 1000, where Cmes and Nmes are the measured 12C13C−/12C2– and 15N12C−/14N12C– ratios of the sample and Cnat and Nnat is the average 12C13C−/12C2− and 15N12C−/14N12C– ratios measured in control, non-labelled samples. A number of measurements on these controls (n = 12) yielded distributions of δ13C = 0 ± 9.6‰, and δ15N = 0 ± 21.6‰ (±2σ).

Compound-specific isotope analysis of fatty acid methyl esters

Fatty acid extraction was performed following the method of Bligh & Dyer [37] as modified by Meziane & Tsuchiya [38] and Passarelli et al. [39] (see electronic supplementary material, method S2). The CSIA of the FAME was performed by gas chromatograph/combustion/isotope ratio mass spectrometry (GC/C/IRMS) with an Agilent 6890 GC instrument coupled to a Thermo Fisher Scientific (Bremen, Germany) Delta V Plus IRMS instrument via a combustion interface III under a continuous helium flow. The GC separation was performed with the HP-FFAP column (50 m × 0.20 mm; length × inner diameter) coated with 0.33 µm nitroterephthalic acid modified polyethylene glycol stationary phase. The FAME samples were injected splitless at 230°C. After an initial period of 2 min at 100°C, the column was heated to 240°C (held 26 min) at 5°C min−1, then to 245°C (held 4 min). This GC conditions were optimized for good separation of unsaturated FAs by injection of a standard mixture of 37 FAMEs (Supelco 37 Component FAME Mix, Sigma-Aldrich, Buchs, Switzerland) containing C4–C24 homologues. For calibration and normalization of the measured FAME δ13C values were used the previously determined δ13C values (by elemental analysis/IRMS) of a mixture of deuterated carboxylic acids used as external standards. For quality control, the repeatability and intermediate precision of the GC/C/IRMS analysis and the performance of the GC and combustion interface were evaluated every five runs by injection of a carefully prepared mixture of FAMEs reference materials [40]. The standard deviation for repeatability of the δ13C values ranged between ±0.05 and ±0.5‰ for m/z 45 peak size between 15000 mV and less than 500 mV. The FA δ13C were determined from the FAME δ13C by correction for the isotopic shift due to the carbon introduced by methylation using a mass balance equation [41].

A floating tray with wells (56 mm diameter × 60 mm depth) was placed floating in the described LSS. The bottom of the wells was made of a 0.5 mm mesh to allow water exchange [42]. A re-circulating water pump was placed below the experimental tray to increase water renewal inside the wells. Twenty-four adult E. timida specimens were randomly divided in pairs and placed in individual wells. The photoperiod was kept at 14 h light : 10 h dark. Two treatments (six replicates per treatment, each replicate being a pair of sea slugs) were performed: (i) ‘regular light’ treatment, in which the sea slug specimens were subjected to a photon scalar irradiance of 40–160 µmol photons m−2 s−1, depending on the position inside the well; (ii) ‘reduced light’, in which the sea slug specimens were subjected to a photon scalar irradiance of 5 µmol photons m−2 s−1. Light treatments were achieved by placing either transparent or opaque lids over the wells. In the case of the reduced irradiance treatment, light reached the animals through the bottom mesh. Animals were fed every day with A. acetabulum grown at a photon scalar irradiance of 60 µmol photons m−2 s−1 under a 14 h light : 10 h dark photoperiod. During the experimental period, one animal died in the reduced light conditions, reducing the number of replicates in this treatment to n = 5.

Egg masses spawned by the sea slugs on the walls of the wells and, occasionally, on the net at the bottom of the wells were counted daily for 28 days and collected using a scalpel (electronic supplementary material, figure S2B). The number of eggs in each individual egg mass was counted using a Leica DMS300 digital microscope. Egg masses were gently washed in ultrapure water, frozen at −80°C and freeze-dried. The last egg mass produced in each experimental unit (well) was analysed for FA composition. Five replicates of E. timida sea slugs reared in actinic light conditions and A. acetabulum were similarly washed in ultrapure water, frozen at −80°C, freeze-dried and macerated prior to lipid extraction. After lipid extraction (see electronic supplementary material, method S3), FAs in the three biological matrices surveyed (egg masses, sea slugs and macroalgae) were transmethylated according to Aued-Pimentel et al. [43] to obtain FAME and analysed by gas chromatography—mass spectrometry (GC-MS). FAME identification was performed by comparing retention times and mass spectra with those of commercial FAME standards (Supelco 37 Component FAME Mix, ref. 47885-U, Sigma-Aldrich) and confirmed by comparison with the Wiley library and the spectral library from ‘The Lipid Web’ [44]. FA quantification was performed using calibration curves obtained from FAME standards under the same instrumental conditions. FA in E. timida and A. acetabulum were expressed as relative abundances (%). FA concentrations in the eggs were expressed as pg egg−1 dividing the FA content of the whole egg mass by the number of eggs.

Statistical analyses

The number of eggs spawned in each experimental unit (pairs of sea slugs placed on each well) was averaged to avoid pseudo-replication, and averages were treated as independent replicates [45]. Statistically significant differences in the number and FA concentrations of eggs spawned by regular versus reduced light reared animals were tested using independent samples t-tests. Normality was checked using a Shapiro–Wilk's test, while homogeneity of variances was tested using Levene's test. Statistical analyses were carried out using IBM SPSS Statistics 24.

Click here for additional data file.