Experimental designs
Acantharian photosynthesis. Experiments were designed to determine rates of photosynthetic uptake of inorganic carbon and nitrogen in photosymbiotic Acantharia (see a conceptual drawing of the experiment in Extended Data Figure 2 in the Extended data section [Mansour et al., 2022]). The used specimens all harbour intact photosynthetic symbionts that provide the hosts with photosynthates (Anderson, 1983; Decelle et al., 2012b; Probert et al., 2014; Yellowlees et al., 2008). Seawater was collected from the bay of Villefranche-sur-Mer and filtered (0.22 μm) for experimental incubations. The concentration and isotopic ratio of dissolved inorganic carbon in the seawater was measured using Delta V Plus mass spectrometer and Gas Bench II at the IRMS platform in Brest, France.
Acantharia were transferred to 6-well plates containing 10 mL FSW enriched with NaH13CO3 (99% 13C, Eurisotop) and either 15NH4Cl (99% 15N, Eurisotop) or Na15NO3 (98% 15N, Eurisotop). Each treatment was accompanied by a control incubation containing non-enriched nutrients in the same concentrations (natural medium incubations). The experimental treatments were as follows:
Treatment 1: 10 µM NaH13CO3;
Treatment 2: 10 µM NaH13CO3 and 0.3 µM 15NH4Cl;
Treatment 3: 10 µM NaH13CO3 and 0.5 µM Na15NO3;
Treatment 4: 10 µM NaH13CO3 and 0.3 µM 15NH4Cl in complete darkness. This treatment was not accompanied by a similar incubation using non-enriched nutrients.
Specimens were incubated for 4 h in an incubator set at seawater temperature (16.5°C, April 2021), with a 16:8 h light:dark cycle at 150–160 μmol photons m−2 s−1. After the incubation 60 to 100 cells were manually transferred through three series of clean filtered seawater and finally deposited one by one on a pre-combusted (450°C for 4 h) GF/F filter (25 mm, Whatman). The volumes of medium used to transfer all Acantharia were kept track of, and analytical blanks were made by wetting a separate GF/F filter with an equal amount of seawater leftover from the incubation. All filters were dried at 60°C for a minimum of 24 h and subsequently kept in sealed containers in the dark before isotopic analysis. Experiments were always initiated at the same time of day (i.e., between 15:00 and 16:00) to avoid any confounding effects owing to circadian rhythms.
Additionally, several Acantharia were reserved for measurements of photochemical efficiency, using a MICROSCOPY-PAM (Walz), before and after incubation. This provided us an indication of the symbionts’, and thereby presumably the holobiont’s, general health, and allowed insights into the treatments where photosynthetic efficiency was determined.
Symbiont (Phaeocystis) photosynthesis. Photosynthetic rate of the free-living state (i.e., in algal culture, not in symbiosis) of Phaeocystis cordata (RCC1383) was measured from cultures. Prior to the experiments, cultures were grown in 50 mL flasks, partially submerged in an aquarium with a water temperature of 20°C, a light:dark cycle 14:10 h at a photon irradiance of 150–170 µmol photons m−2 s−1 as measured in the water. The aquarium with culture flasks was placed on a glass table with light provided from below. Light was provided by cool white fluorescent tubes (OSRAM 58W, 840) and photon irradiance was measured using a light meter equipped with a spherical quantum sensor (ULM & US-SQS/L, Walz GmbH, Germany).
Cultures in exponential growth phase (as estimated by daily flowcytometric cell counts) were used for photosynthesis measurements employing the 14C methodology. Different incubations were performed for a duration of 0, 3, 6, 8 and 24 h. For each incubation-time, one set of triplicates was incubated in the light, while the other was kept in complete darkness to compensate for passive incorporation of the radioisotope.
Simultaneous with the 14C experimental procedure aliquots of the culture were taken for cell enumeration and algal chlorophyll a (Chla) content measurements. Total dissolved inorganic carbon concentrations were measured on 25 mL of the medium used for the experiments using a Shimadzu TOC L analyser.
Acantharian grazing. Grazing experiments were done with different potential prey algal cultures to measure the contribution of prey ingestion to the acantharian metabolism (see a conceptual drawing of the experiment in Extended Data Figure 3 in the Extended data section [Mansour et al., 2022]). In triplicate, Acantharia (15 to 25 cells) were transferred to 6-well plates prefilled with 1 mL FSW after which 10 mL of algal culture (from the same culture flask) was added. Control monocultures of each algal culture were also set up in triplicates allowing for the calculation of prey ingestion rate as the reduction in prey concentration in grazing treatments compared to control treatments with the prey algae alone. Incubations were done under the same conditions as for Acantharian photosynthesis measurements. At 4 h and 24 h triplicate 1 mL samples were taken fixed in 0.25% (final concentration) glutaraldehyde, and frozen at -20°C for flow-cytometry cell counts. Following the results, we note that possibly the ratios between prey and Acantharia were too high at the chosen prey densities to effectively measure the number of prey eaten from a much larger prey pool.
Additionally, to qualitatively observe ingested prey, a similar incubation was performed, though with beforehand staining of I. galbana (RCC178) cells using the fluorochrome CellTracker Blue CMAC (7 amino-4-chloromethylcoumarin, ThermoFisher), a vital cytoplasmic stain, as described by Martínez et al. (2014). Isochrysis galbana is used here as a species that has been shown to be ingested by Spumellaria (Radiolaria) (Anderson et al., 1984). Cells were stained for 4–6 h with the CellTracker at a final concentration of 10 μM. To reduce the carryover of stain that could enter Acantharia, excess stain was removed from the medium after the staining period by washing three times. This was performed by centrifugation of the live fluorescently labelled algae (LFLA) at 2,000 x g for 5 min; the supernatant was discarded and the cells were re-suspended in FSW. Acantharia were incubated with the labelled prey as before, triplicate Acantharia were individually isolated after 0, 30, 60, and 240 min; transferred to 0.5 mL tubes; fixed in a final concentration of 0.25% glutaraldehyde (Sigma, ref. G5882-100ML) and 1% paraformaldehyde (Electron Microscopy Sciences, ref. 15714, diluted in phosphate-buffered saline); and stored at 4°C for fluorescence microscopy.
Analyses
Stable isotope analysis. The C and N isotopic composition and content of the particulate organic matter was determined as CO2 and N2 released by flash combustion using an elemental analyser (EA Isolink CN/OH, Thermo Fisher) coupled to an isotope ratio mass spectrometer (IRMS, Delta V Advantage, Thermo Fisher) via an open split and dilution-controlled interface (ConFlow IV, Thermo Fisher) with a continuous flow of helium at 180 mL min-1. In addition to the samples, each run contained international isotopic standards from International Atomic Energy Agency (IAEA) (these were IAEA-600 (caffeine), and CH6 (sucrose) for C; N2 (Ammonium Sulfate) and USGS34 (Potassium Nitrate) for N). IAEA standards were used for manual normalization of the δ-unit to the Vienna Pee Dee Belemnite-limestone (VPDP) or atmospheric N2 scale, for carbon and nitrogen respectively, using a linear calibration curve (Coplen et al., 2006; Paul et al., 2007). An isotopic laboratory working standard (casein, certified Elemental Microanalysis) was run every 15 samples for verification of system stability. Though it was not needed in our case, a correction for linearity could be performed using the employed casein standard. A laboratory weight standard (acetanilide, certified Elemental Microanalysis) of different masses were used to calibrate the analyser and determine C and N content of the samples. Standard deviations were 0.11 and 0.06 µg for C and N content respectively, and 0.1 ‰, and 0.05 ‰ for δ13C and δ15N, respectively. Additionally, a quality check (urea, certified Elemental Microanalysis) for isotopic ratios and weight (urea, certified Elemental Microanalysis) was run randomly to validate the quality of analysis. Isotopic Ratio Mass Spectrometry data were processed in the manufacturer’s ISODAT 3.0 software and Microsoft Excel (RRID:SCR_016137) for the output of values expressed in delta notation (δ) in units of per mil (‰).
Carbon and nitrogen signal of the samples was corrected by accounting for the related analytical blank. The blank correction was performed for the normalized δ-values using the following equation:
For qualitative measurements of isotope uptake, or the calculation of excess, 13C and 15N of each sample relative to natural abundance δ13C and δ15N were converted to fractional abundances (F) using the following equations (Fry, 2006; Moodley et al., 2005):
Where R is the ratio of either 13C/12C or 15N/14N; δX is the normalised-blank-corrected-δ-value of the sample, either δ13C or δ15N; Rvpdp = 0.0112372, for nitrogen this is substituted with Rair = 0.0036765, and F is the fractional abundance. Finally, total uptake (I) of the heavier isotope in μg C cell-1 h-1 or μg N cell-1 h-1 was calculated as:
Here Fsample is the fractional abundance of the sample after incubation and Fcontrol the non-enriched samples. msample is the total C or N mass of the sample in μg and t is the incubation time in hours. The 13C uptake was further normalised per acantharian cell by division with the total number of Acantharia in the sample (n).
The total uptake of carbon (12C + 13C) was calculated by division of the total 13C uptake (I) by the fractional abundance of total 13C of dissolved inorganic carbon in the incubation. The dissolved inorganic carbon concentration of the seawater was measured to average 5,900 µM, with a natural abundance of 13C of 1.08%. The total fraction of 13C in the experimental incubations was thus 1.9%. For total uptake of nitrogen, the concentrations of nitrite, nitrate, and ammonium at the sampling location were considered The natural 15N fraction was assumed as 0.368% (data from https://www.somlit.fr/visualisation-des-donnees, Villefranche point B; surface), we estimated the total fraction of 15N as 43.95 and 32.53% for the nitrate and ammonium treatments, respectively.
Microscopy chlorophyll fluorescence measurements. Chlorophyll fluorescence parameters were measured using a Pulse Amplitude Modulated (PAM) fluorometer coupled to a microscope (MICROSCOPY-PAM; Walz, Effeltrich, Germany), equipped with a 10X objective lens. After a minimum of 5 min dark incubation, the basal level of fluorescence (F0) was measured under modulated light (9 µmol photons m−2 s−1, frequency: 8 Hz at 625 nm), after which the maximum fluorescence level (Fm) was determined with a saturating light pulse (1719 µmol photons m−2 s−1, during 8 × 60 ms at 625 nm). Dark-adapted maximal quantum yield (Fv/Fm) of photosystem II (PSII) was calculated as follows: Fv/Fm = (Fm − F0)/Fm.
ImagingWin (v2.46i) software (WALZ) was used to determine areas of interest (AOI) that encompass the entire Acantharia cell. Due to the patchiness of the photosynthetic activity, only pixels for which Fv/Fm > 0 were used to calculate fluorescence parameters, hence non-photosynthetically active regions (such as those of the host cell) are not taken into consideration.
Growth and ingestion rate calculations. Algal culture growth rates (µ, d−1) were measured as change in cell concentrations over time and calculated assuming exponential growth.
Where Nx and N0 are the cell concentrations in cells mL−1 at time point tx and time t0 respectively.
The ingestion rates (U = prey predator-1 h-1) were calculated by Frost (1972) equations as modified by Heinbokel (1978):
Where C is the average prey concentration in the experimental incubations with prey and grazer, calculated as:
With Xtx being the prey concentration at the end of the incubation and Xt0 at the start.
F is the clearance rate (mL grazer-1 h-1) calculated as:
k is the intrinsic growth rate of prey in monoculture incubations, calculated as:
g is the grazing coefficient, calculated as the growth rate of prey in the experimental incubations with prey and grazer:
Xt2 and Xt1 being the prey concentrations at t2 and t1, respectively.
P is the average grazer concentration in the experimental incubations with prey and grazer. Considering that Acantharia do not proliferate during the experiments (only death might be observed), P is simply:
Where A is the Acantharia count at the end of the incubation and V the volume of the incubation.
Fluorescence microscopy. Fluorescence microscopy observations of ingested prey were conducted using an inverted Leica SP8 confocal laser scanning microscope (Leica Microsystems, Germany) equipped with a compact supply unit that integrates a LIAchroic scan head and several laser lines (405 nm, 488 nm, 552 nm, 638 nm), and either an HC PL APO CS2 63 × 1.40 OIL objective or an HC PL APO CS2 40 × 1.1 WATER objective, according to specimen size. Single cells were isolated from the fixative solution and placed on LabTek II chambered coverglass (Nunc 155382; Thermo Fisher Scientific, MA, USA). To visualise the membrane and skeletal structures, the cells were additionally stained with 50 µM Poly-L-lysine conjugated with Alexa Fluor 546 for 15 min (Colin et al., 2017), and then washed by means of cell transfer through FSW. This stain binds among others strontium sulphate, protein, and polysaccharide materials (Colin et al., 2017).
Imaging was performed with excitation using the 405 nm, and 552 nm lasers, and signal collection at emission wavelengths 460–472 nm (histogram adjustment: minimum 5; maximum 105), 603–635 nm (histogram adjustments: minimum 0; maximum 210), and 693–782 nm (histogram adjustment: minimum 0; maximum 160) to capture fluorescence of the Celltracker blue stain, carotenoids/Alexa Fluor 546 stain, and chlorophyll, respectively. Image processing and 3D rendering from z-stacks was performed using the software Imaris (RRID:SCR_007370) (Bitplane AG, Zurich, Switzerland) (free alternative, FIJI ImageJ), following the guidelines of Schmied and Jambor (2020).
Algal cell enumeration. Fixed algal culture cells (glutaraldehyde 0.25% final concentration) of the grazing experiments were enumerated on an Agilent NovoCyto Advanteon flowcytometer. Except for cells < 1.5 µm (Synechococcus sp.), cells were enumerated on a Guava Easycyte Ht flowcytometer. Free-living stages of the algal symbionts (Phaeocystis) for the symbiont photosynthesis experiments, were enumerated on a Cytoflex flowcytometer (Beckman Coulter, USA). Thresholds were on red fluorescence, based on fluorescence patterns and cell size from side scatter (Olson et al., 1991).
Chlorophyll a content of Phaeocystis cultures. Samples for Chla measurements from algal suspensions were filtered or deposited on a GF/F filter and Chla was extracted by adding an equal amount of 96% ethanol (i.e., 2 mL sample with 2 mL 96% ethanol). Samples were stored at 4°C for 24 h in darkness. Fluorescence and Chla concentrations were measured using a bench-top fluorometer (Trilogy, Turner designs, CA) equipped with a manufacturer’s Chla non-acidification insert for Chla determination. The fluorometer was calibrated against a pure Chla standard (2.13 mg Chla L−1) of cyanobacterial origin (DHI, Hørsholm, Denmark).
Photosynthetic rates measurements of Phaeocystis cultures. Photosynthetic rates were measured using the 14C technique by Rivkin and Seliger (1981) and previously described by Skovgaard et al. (2000) and Hansen et al. (2016). Sample aliquots of 2 mL from algae suspensions were taken in two sets of triplicates. To each sample 20 µL of NaH14CO3 stock solution (specific activity = 100 µCi mL−1; Carbon-14 Centralen, Denmark) was added for a final concentration 1 µCi mL−1. For each experimental incubation, one set of triplicates was incubated in the light, while the other was kept in complete darkness to compensate for passive incorporation of the isotope. After incubation, a 100 µL sub-sample was transferred to a new vial containing 200 µL phenylethylamine for determination of specific activity. The remaining 1.9 mL were acidified with 2 mL 10% glacial acetic acid in methanol and evaporated overnight at 60°C to remove inorganic carbon. The residue was then re-dissolved in 1.5 mL Milli-Q water. All vials were treated with 10 mL Ultima Gold XR scintillation cocktail, and disintegrations per minute were measured using a Perkin Elmer, Tri-Carb 2910TR liquid scintillation analyzer with 10-minute acquisition time. As per Skovgaard et al. (2000), new caps (Packard poly screw caps) were mounted after adding the scintillation cocktail.
Photosynthetic activity (PA in pg C cells−1 h−1) was calculated from the disintegration per minute using the following equation.
Where DPM is disintegrations min−1 mL−1, either in the ‘light’, ‘dark’ or ‘specific activity’ vial; h is the incubation time; N is cells mL−1 and [DIC] is the concentration of dissolved inorganic carbon (pg C mL−1) in the medium. Total dissolved inorganic carbon concentrations [DIC] were measured on 25 mL of the medium used for the experiments using a TOC-L analyser, Shimadzu.
Statistical analyses. Mean δ13C (or δ15N) differences between treatments were tested by comparing means with an ANOVA and Tukey-Kramer HSD post-hoc test. Effects of Acantharia grazing were tested by comparing the grazing coefficient with the intrinsic growth rate using a Student’s t-test. Tests were done at significance level 0.01. The software SAS JMP Pro (RRID:SCR_022199) 15.0 (SAS Institute, Inc) was used for statistical analyses pertaining isotopic ratios (free alternative, the R statistical environment (R Core Team 2013, RRID:SCR_001905)). Microsoft Excel (RRID:SCR_016137) was used for statistical analyses pertaining grazing. Results are presented as mean ± standard deviation.
