Different levels of energetic coupling between photosynthesis and respiration do not determine the occurrence of adaptive responses of Symbiodiniaceae to global warming.

Disentangling the metabolic functioning of corals' endosymbionts (Symbiodiniaceae) is relevant to understanding the response of coral reefs to warming oceans. In this work, we first question whether there is an energetic coupling between photosynthesis and respiration in Symbiodiniaceae (Symbiodinium, Durusdinium and Effrenium), and second, how different levels of energetic coupling will affect their adaptive responses to global warming. Coupling between photosynthesis and respiration was established by determining the variation of metabolic rates during thermal response curves, and how inhibition of respiration affects photosynthesis. Adaptive (irreversible) responses were studied by exposing two Symbiodinium species with different levels of photosynthesis-respiration interaction to high temperature conditions (32°C) for 1 yr. We found that some Symbiodiniaceae have a high level of energetic coupling; that is, photosynthesis and respiration have the same temperature dependency, and photosynthesis is negatively affected when respiration is inhibited. Conversely, photosynthesis and respiration are not coupled in other species. In any case, prolonged exposure to high temperature caused adjustments in both photosynthesis and respiration, but these changes were fully reversible. We conclude that energetic coupling between photosynthesis and respiration exhibits wide variation amongst Symbiodiniaceae and does not determine the occurrence of adaptive responses in Symbiodiniaceae to temperature increase. ©2020 The Authors. New Phytologist ©2020 New Phytologist Trust.

Introduction

The Symbiodiniaceae family (dinoflagellates, LaJeunesse et al., 2018) comprises microalgae species capable of establishing endosymbiotic relationships with animal hosts such as Cnidaria, Porifera and Mollusca (Leggat et al., 1999; Weber & Medina, 2012; Achlatis et al., 2018). This association between Symbiodiniaceae and their hosts is key to the integrity and functioning of coral reef ecosystems world‐wide. Under current climate change conditions, alterations of global temperatures have been identified as major threats to coral reefs (Spalding & Brown, 2015; Hughes et al., 2017a,b). Increasing mean water temperatures, together with extreme climatic events such as heat waves, are disrupting the endosymbiotic relationship between Symbiodiniaceae and their cnidarian hosts, and causing reef death if these events are persistent and severe (Suggett et al., 2017; Hughes et al., 2018; Fordyce et al., 2019; Goyen et al., 2019).

In the future, the response of coral reefs to changes in temperature will depend on the ability of both Symbiodiniaceae and their hosts to adapt to the new climatic conditions (Chakravarti et al., 2017; Torda et al., 2017; Comeau et al., 2019). Because of their large population size, short generation times (at least under a free‐living lifestyle) and high genetic diversity, Symbiodiniaceae species are expected to have a faster adaptive capacity to climatic changes than their hosts (Collins, 2011; Torda et al., 2017; Comeau et al., 2019; González‐Pech et al., 2019). Thus, studies of isolated Symbiodiniaceae provide valuable contributions to the understanding of the mechanisms underpinning corals' adaptation to ocean warming. Recent studies focusing on the responses of these organisms to long‐term temperature increases have shown that some isolated Symbiodiniaceae are capable of relatively fast adaptation (c. 2.5 yr and 80 cell generations) to elevated temperature conditions (Huertas et al., 2011; Chakravarti et al., 2017; Chakravarti & van Oppen, 2018).

Adaptation (long‐term irreversible responses, usually requiring de novo mutations in the genome or species sorting) to warming conditions could imply modifications of the two most important metabolic processes of the algal cell, photosynthesis and respiration. In photosynthetic organisms (both plants and algae), photosynthesis and respiration occur in different organelles (chloroplasts and mitochondria, respectively), and their activities are very often coupled (Raghavendra & Padmasree, 2003; Cardol et al., 2009). In the green alga Chlamydomonas reinhardtii, mitochondrial respiration can supply ATP to the chloroplast, influencing net CO2 fixation (Cardol et al., 2009) and affecting the photosynthetic response under high light intensity conditions (Larosa et al., 2018). Similarly, energetic interaction (exchanging ATP and NADPH) between the mitochondrion and the chloroplast occurs in diatoms (Bailleul et al., 2015). In regard to Symbiodiniaceae and other dinoflagellates, it is still unknown whether respiration activity is energetically connected to photosynthesis. The possible presence of phenotypes with multiple degrees of photosynthesis to respiration coupling is also unexplored. Disentangling the extent of photosynthesis dependence on respiration is fundamental to clarifying the adaptive capability of photosynthesis and respiration of Symbiodiniaceae under future climatic conditions.

In relation to other species, thermal response curve (TRC) analyses (i.e. by measuring photosynthetic and respiratory responses to acute increases in temperature), showed that green algae adaptation (over several hundred cell generations) to high (above‐optimal) temperature conditions involves either increasing the rates of photosynthesis or lowering the rates of respiration (Padfield et al., 2016; Schaum et al., 2017). By contrast, in the marine diatom Thalassiosira pseudonana, adaptation to different warming scenarios caused a similar decline in both photosynthetic and respiration rates (Schaum et al., 2018). However, these studies did not establish whether the effects were due to adaptation or acclimation (i.e. short‐term reversible responses) (Raven et al., 2017). Particularly for respiration, exposure to elevated temperatures may also cause two types of response in the respiration TRCs: type I (involving changes of the slope of the TRCs) and type II (reduction in the elevation of the TRC) (Atkin & Tjoelker, 2003; Smith & Dukes, 2013). In this work, using Symbiodiniaceae as model species, we question the extent to which the degree of coupling between photosynthesis and respiration influences adaptive responses (measured through TRCs) to warming conditions (Atkin et al., 2006).

Photosynthetic adaptation to increasing temperature will also possibly require adjustments in mechanisms of inorganic carbon (Ci) acquisition. To acquire Ci, Symbiodiniaceae possess a CO2‐concentrating mechanism (CCM), that is, structural and functional components which enhance intracellular levels of Ci and favour the carboxylation activity of their form II Rubisco (Leggat et al., 1999; Brading et al., 2011, 2013; Ros et al., 2020). Generally, CCM activity is modulated by the changing of thermal conditions (Beardall & Giordano, 2002; Giordano et al., 2005), and CCM functioning could be linked to species adaptation to cold (or warm) habitats (Kranz et al., 2015; Raven et al., 2017). In the case of Symbiodiniaceae, Oakley et al. (2014a) showed that acclimation to high thermal regimes over several wk resulted in a higher photosynthetic affinity for Ci (reflecting enhanced CCM activity), which could be induced by the lower CO2 solubility in water or a lower Rubisco CO2 : O2 selectivity factor (Tcherkez et al., 2006; Raven et al., 2017). However, no information is currently available on the adaptive capability of CCMs in Symbiodiniaceae under long‐term changes in temperature conditions.

In this work, we first aimed to establish whether there is an energetic coupling between photosynthesis and respiration, and the extent to which this interplay occurs in Symbiodiniaceae (from three genera, namely Symbiodinium, Durusdinium and Effrenium). Second, we explored the consequences of two different levels of coupling between photosynthesis and respiration for the adaptive capabilities of Symbiodiniaceae to a 1‐yr increase in temperature. Finally, we assessed the extent to which adaptation of photosynthesis to warmer conditions is associated with phenotypical adjustments of CCM activity.

Materials and Methods

Organisms and culture conditions

Four species belonging to three different genera of the Symbiodiniaceae family (Symbiodinium Freudenthal, Durusdinium LaJeunesse and Effrenium LaJeunesse et Jeong) were studied. These included Symbiodinium sp. (Avir), Symbiodinium sp. (Stylodid), Durusdinium trenchii (provided by Annika Guse, University of Heidelberg), and Effrenium voratum (Table 1). To confirm the classification of these species at the genus and/or species level, the sequencing of 5.8S rDNA internal transcribed space 2 (ITS2) regions was carried out as described in Hume et al. (2018). For all the experiments, cells were cultured in F/2 medium, prepared with sea salt (Coral Pro Salt, Red Sea; Red Sea Fish Pharm, Verneuil d'Avre et d'Iton, France) and Guillard's (F/2) Marine Water Enrichment Solution (Sigma‐Aldrich), with a total dissolved inorganic carbon (Ci) content of 4 mM (CO2 = 20 μM; HCO3 − = 3495 μM; calculated using the CO2Sys.xls application, Lewis & Wallace, 1998), Alk = 4.5 mEq l−1, a salinity of 34 PSU, and buffered at pH 8.1 using 20 mM Tris‐HCl. Cultures of Symbiodinium sp. (Avir and Stylodid) and E. voratum were maintained in a plant growth chamber (Grobanks; CLF Plant Climatics GmbH, Wertingen, Germany) and exposed to a continuous photosynthetic photon flux density (PPFD) of 100 μmol photons m−2 s−1. Because this light regime was not suitable for the growth of D. trenchii, this species was maintained with a PPFD of c. 20 μmol photons m−2 s−1 under a 12 h : 12 h, light : dark photoperiod. Experiments were conducted for D. trenchii at least 3 h after the onset of the light period. For the control temperature condition, cultures of Symbiodinium sp. (Avir and Stylodid), E. voratum and D. trenchii were maintained at 25°C. During the experiments, cell concentration in the culture was maintained below 150 000–200 000 cells ml–1 by regular sub‐culturing with fresh F/2 medium. Although cultures were not axenic, bacterial concentration was always low (as verified by regular microscope inspection).

Table 1

Symbiodiniaceae species used in this study.

Culture name	Species (Former clade)	Host species (Type of host)	Geographic origin (Temperature min–max)	Collection
Avir	Symbiodinium sp. (A)	Anemonia viridis (Sea anemone)	Mediterranean Sea, France (13–25°C)	Paola Furla, University of Nice
Stylodid	Symbiodinium sp. (A)	Stylophora pistillata (Coral)	Red Sea (21–28°C)	Centre Scientifique De Monaco collection
CCMP2556	Durusdinium trenchii (D)	Montastraea faveolata (Coral)	Tennessee Reef, Florida Keys, Florida, USA (22–28°C)	Mary Alice Coffroth
CCMP421	Effrenium voratum (E)	Free‐living	Wellington, New Zealand (22–26°C)	Mary Alice Coffroth

Cell concentration was determined using a Coulter counter (Z2; Beckman, Indianapolis, IN, USA). The cell‐specific division rates (µc) of Symbiodinium sp. (Avir), Symbiodinium sp. (Stylodid), and E. voratum were calculated from the slope of the natural logarithm plot, obtained from the exponential phase of the population growth in the culture. As D. trenchii was maintained under a lower light regime and a light : dark cycle, conditions which would affect the comparison with cells exposed to continuous light (Quigg et al., 2012), its µc value was not estimated.

Long‐term exposure to elevated temperature

For the long‐term exposure to elevated temperature, three independent replicate cultures of Symbiodinium sp. (Avir and Stylodid) were transferred at 32°C and kept under this regime for 1 yr. These two species have been previously assessed for short‐term acclimation responses at 33°C (Dang et al., 2019), and as they belong to the same genus and both have a facultative symbiotic lifestyle, they provide a solid basis for the present comparison. We chose 32°C because it reflects the warming conditions that tropical algal species might experience in future years due to climate change (IPCC, 2014; Jin & Agustí, 2018; Aranguren‐Gassis et al., 2019). During the 1 yr of incubation at 32°C, cultures were diluted with new F/2 medium (pH 8.1; CO2 = 17 μM; HCO3 − = 3368 μM) every 7 d in the case of Symbiodinium sp. (Avir) and every 14 d for Symbiodinium sp. (Stylodid), according to their different µc under the 32°C treatment (Fig. 5). The µc, photosynthesis and respiration activities were analysed after 5, 7 and 12 months from the start of the 32°C incubation period. According to their µc values, the 1‐yr‐long experiment equated to c. 310 and 190 cell generations for Symbiodinium sp. (Avir) and Symbiodinium sp. (Stylodid) respectively. To test whether the physiological traits displayed during such long‐term exposure were stable and adaptive, at the end of the 1‐yr period of warming, cells were returned for at least 1 wk to the control temperature of 25°C (Chakravarti et al., 2017). Only measurements showing adaptive photosynthetic and/or respiratory adjustments were repeated.

Fig. 5

Cell‐specific division rate (μc) and gross maximal photosynthesis (Pmax) of Symbiodinium sp. (Avir) and Symbiodinium sp. (Stylodid) growing at 25°C (0 months) and during the 1‐yr‐long experiment at 32°C. Measurements were performed for at least three independent biological replicates. The connecting line fits through the raw data. Vertical bars indicate the standard deviation.

Measurements of photosynthetic and respiratory activities

Rates of photosynthetic O2 evolution as a function of irradiance (P vs I curve) and dark respiration (R d) were measured with a Clark‐type oxygen electrode (Hansatech, Norfolk, UK). P vs I curves were measured for all species grown at 25°C and Symbiodinium sp. (Avir and Stylodid) at 32°C. For each experiment, cells were concentrated through centrifugation at 2200  for 2 min. As Symbiodiniaceae cells are large (c. 10 μm, Roberty et al., 2014), harvesting by low centrifugation force allows the sinking and formation of a pellet at the bottom of the centrifuge tube, and the separation of the smaller prokaryotic cells which remain in the supernatant (Malerba et al., 2018). After centrifugation, the cells were resuspended in 1 ml of fresh culture F/2 medium (+2 mM NaHCO3) and placed into the electrode chamber. Before each P vs I measurement, the cells were dark acclimated for 15 min, with the final 5 min of this incubation period used to measure the R d. After the dark pre‐treatment, O2 evolution inside the chamber was measured over a range (4–1350 μmol photons m−2 s−1) of PPFD, provided by an LED light source peaking at 630–640 nm. Gross photosynthesis (P) was calculated as the sum of R d and net photosynthesis. During the experiments, to avoid the possibility of photorespiration (Oakley et al., 2014a; Ros et al., 2020) and a potential underestimation of O2 evolution rates due to O2 saturation, the O2 concentration in the chamber was maintained below c. 80% of air equilibrium.

Thermal response curves

We carried out TRCs as in Karsten & Holzinger (2012) and Pierangelini et al. (2019). For these analyses, cells were harvested from the culture by centrifugation (2200 , 2 min), resuspended in 1 ml of fresh culture F/2 medium (+2 mM NaHCO3), and placed into a Clark‐type oxygen electrode chamber (Hansatech). TRCs were measured for all species grown at 25°C and Symbiodinium sp. (Avir and Stylodid) at 32°C. For TRCs, cells were exposed to rising temperatures from 15 to 40°C. In the case of Symbiodinium sp. (Avir) growing under control condition of 25°C, the starting temperature for TRCs was set at 10°C. The use of this lower initial temperature did not cause changes in the shape of the TRCs (Supporting Information Fig. S1a,b). However, when growing at 32°C, photosynthesis of Symbiodinium sp. (Avir) became more susceptible to low temperature (Fig. S1c), and TRCs were started at 15°C. At each temperature, cells were initially incubated in the dark for 25 min, with the last 10 min of this incubation period used to measure respiration. After the dark period, cells were exposed to a saturating PPFD of 700 μmol photons m−2 s−1 (LED light source peaking at 630–640 nm) for 10 min, with the final 5 min used to calculate the photosynthetic O2 evolution. For each assay temperature, the oxygen electrode was calibrated (with air and N2) and the oxygen electrode drift (O2 depletion related to non‐biological processes) was estimated in order to correct the raw O2 evolution and consumption data. Gross P was calculated as the sum of R d and net photosynthesis. The activation energies (how quickly the trait varies as a function of temperature; Padfield et al., 2016) of gross P (Ea P) and respiration (Ea R) were calculated from the slopes of the TRCs, using the Arrhenius equation loge k = −Ea/R (1/T) + loge A.

Impact of a respiratory inhibitor on photosynthesis

The effects of a respiratory inhibitor on photosynthesis were measured for all species grown at 25°C and Symbiodinium sp. (Avir and Stylodid) at 32°C. For these experiments, O2 evolution and relative electron transport rate through the photosystem II (rETR) were recorded simultaneously with an optical O2 meter – FireStingO2 (Pyro Science, GmbH, Germany) and a JTS‐10 spectrophotometer (Bio‐logic, Claix, France) respectively, following the procedures of Roberty et al. (2014) and Vega de Luna et al. (2019). Prior to measurements, cells were incubated in the dark for 15 min. Salicylhydroxamic acid (SHAM) was diluted in DMSO (stock solution 1 M), and added at the beginning of the dark incubation period, with a final concentration in the cell suspension of 8 mM. This SHAM concentration is expected to cause a c. 20% reduction in respiration rates (Oakley et al., 2014b). O2 evolution and rETR were measured by exposing cells to a progressive increase (every c. 6 min) of PPFD (14—890 µmol photons m−2 s−1) provided by an LED array emitting at 660 nm. During each light step intensity, F s (stationary fluorescence level in light) and (maximum fluorescence emission induced by a saturating pulse > 5000 µmol photons m−2 s−1) were measured, and the effective quantum yield of photosystem II at steady state (φPSII) was calculated as follows: ( − F s)/ (Genty et al., 1989). The rETR was calculated as the product of φPSII and the actinic PPFD (Ralph & Gademann, 2005). Gross P was calculated as the sum of R d and net photosynthesis.

Photosynthesis as a function of dissolved inorganic carbon

The photosynthesis of Symbiodinium sp. (Avir) under increasing inorganic carbon availabilities (P vs Ci) was measured using a Clark‐type oxygen electrode (Leggat et al., 1999; Treves et al., 2016; Ruan et al., 2017). For these measurements, we prepared F/2 culture medium depleted of Ci through acidification (pH ≤ 2) and bubbling with N2 gas for at least 1 h. Shortly before the analysis, the F/2 Ci‐free medium was re‐adjusting to pH 8.1 with an NaOH pellet. For the experiment, cells were harvested from the culture, concentrated by centrifugation (2200 , 2 min), washed in 20 ml of F/2 Ci‐free, resuspended in 990 µl of F/2 Ci‐free, and placed in the electrode chamber, where they were exposed to a light saturating photosynthesis of 700 µmol photons m−2 s−1 (LED light source peaking at 630–640 nm). Photosynthesis was allowed to take place until the residual Ci in the F/2 medium and/or the intracellular Ci pool (Leggat et al., 1999) were consumed, and a stable O2 evolution rate was reached (after c. 2 h, minimizing potential photodamages related to Ci depletion). Photosynthetic rates were then measured over ascending NaHCO3 concentrations (0–2000 µM), obtained by sequential additions (1–2 µl) of freshly prepared NaHCO3 solutions (20, 100, 200 mM) into the O2 electrode chamber. To estimate the photosynthetic affinity for CO2, the slope of the CO2‐limited part of the P vs CO2 curve, and the half saturation constant for CO2‐dependent photosynthesis (K0.5 CO2), were calculated (Pierangelini et al., 2014; Ruan et al., 2017). Slope of the Ci‐limited part of the P vs Ci curves (not shown), and the half saturation constant for Ci‐dependent photosynthesis (K0.5 Ci) were also calculated.

Chl content

Chl cell contents were quantified as in Dang et al. (2019). Briefly, cells were harvested by centrifugation (2200 , 2 min), resuspended in 100% methanol and lysed in a TissueLyser II (30 Hz, 5 min, 4°C; Qiagen) in the presence of glass beads (710—1180 μm; Sigma‐Aldrich). After debris removal at 4°C (by centrifugation at 10 000 , 10 min), Chla and c concentrations were determined by spectrophotometry according to the equations of Ritchie (2006) for dinoflagellates.

Transmission electron microscopy

To evaluate whether the presence of different degrees of metabolic coupling between photosynthesis and respiration are associated with a different subcellular organization, analysis by transmission electron microscopy (TEM) was performed for Symbiodinium sp. (Avir) and E. voratum, the two most distant strains when considering interaction degree between respiration and photosynthesis. Cells samples for TEM were fixed in a vacuum bell for 2 h at 4°C with 2.5% glutaraldehyde in 0.1 M Sörensen’s phosphate buffer (pH 7.4) and postfixed for 1.5 h with 2% osmium tetroxide. After washing in Sörensen's buffer, cells were dehydrated through graded ethanol solutions and then processed for embedding in Epon. Ultrathin sections of samples were contrasted with uranyl acetate and lead citrate before examination in a JEM 1400 transmission electron microscope at 80 kV (Jeol (Europe) BV, Zaventem, Belgium). Analysis of the TEM images and measurements of the sizes of the organelles were performed using the item software (Olympus Soft Imaging Solutions GmbH, Münster, Germany). The number of mitochondria‐chloroplast surface contacts were estimated in relation to the total number of chloroplastic sections contained in the cell, using at least 10 different randomly photographed cells. To estimate chloroplastic and mitochondrial sections sizes, at least 20 measurements were performed for each organelle.

Statistical analysis

Experiments were performed with at least three (unless otherwise stated) biological replicates per species. Comparisons between two treatments/species were made using two‐tailed t‐tests. When comparison involved more than two treatments/species, one‐way ANOVA was used. Changes of parameters as a function of increasing temperature (TRCs) and changes during the yr‐long experiment were tested using repeated measures (RM) ANOVA. All analyses were followed by Tukey's multiple comparisons test. The analyses were performed using the software graphpad prism 5 (GraphPad Software, San Diego, CA, USA), setting the threshold of significance at 95%.

Results

TRCs show diverse degrees of metabolic thermal sensitivity amongst Symbiodiniaceae

To first establish whether gross O2 evolution and dark respiratory O2 consumption rates have either similar or different responses to an acute increase in temperature, we carried out TRCs. For the four Symbiodiniaceae (two Symbiodinium sp., D. trenchii and E. voratum), gross P increased from 10–15°C to 25–30°C, and declined beyond this range of temperatures (Fig. 1). The photosynthetic rates of Symbiodinium sp. (Avir and Stylodid) and E. voratum reached at 25°C were also comparable with the measures obtained with the P vs I curves (Table 2) (t‐tests: F 3,2 = 17.38, P = 0.3554; F 3,2 = 1.296, P = 0.0583; F 2,2 = 1.694, P = 0.1420, respectively), indicating that the rapid changes in temperature imposed during the TRCs did not impact the cell metabolisms. By contrast, the photosynthetic rates of D. trenchii were lower than those of the P vs I curves (Table 2) (t‐tests: F 2,3 = 2.437, P = 0.0261), reflecting a susceptibility of the species to the relatively low (15°C) initial temperature of the TRCs. In contrast to photosynthesis, respiration increased steadily during the heat exposure and did not decline at temperatures as high as 40°C. The comparison of the activation energy (Ea) parameter (in electron volts, eV), indicated that whereas the photosynthesis of the four species had different sensitivities to temperature (different Ea P) (ANOVA: F 2,7 = 91.01, P < 0.0001), sensitivities of their respiratory activities to temperature were equivalent (similar Ea R; ANOVA: F 2,7 = 2.821, P = 0.1264) (Fig. 2a). Comparing results in each species, differences were found in the temperature dependence of photosynthesis and respiration. The results showed that Ea P was higher than Ea R in both Symbiodinium sp. (Avir) (t‐tests: F 3,3 = 10.14, P = 0.0004) and D. trenchii (t‐tests: F 2,2 = 4.120, P = 0.0021). Conversely, in the case of Symbiodinium sp. (Stylodid) and E. voratum, Ea P was identical to Ea R (t‐tests: F 2,2 = 57.36, P = 0.7885; F 2,2 = 4.737, P = 0.1803), indicating that both photosynthesis and respiration were equally sensitive to the temperature increase. These differences and similarities in temperature dependence were also indicated by the R d : gross P ratios (Fig. 2b,c). Whereas for Symbiodinium sp. (Avir) and D. trenchii, the R d : gross P ratio declined between the lowest and optimal temperatures (RM ANOVA: F 4,3,12 = 9.507, P = 0.0011 and, F 2,2,4 = 14.72, P = 0.0143), the R d : gross P did not change (RM ANOVA: F 3,2,6 = 0.7479, P = 0.5620 and, F 3,2,6 = 3.342, P = 0.0972) for Symbiodinium sp. (Stylodid) and E. voratum.

Fig. 1

Thermal response curves of gross maximal photosynthesis (P max, left y‐axis) and respiration (R d, right y‐axis) for Symbiodinium sp. (Avir), Symbiodinium sp. (Stylodid), Durusdinium trenchii and Effrenium voratum grown at 25°C. Measurements were performed for at least three independent culture replicates. The connecting line fits through the raw data. Vertical bars indicate the standard deviation.

Table 2

Cell‐specific division rate (µc), photosynthetic characteristics (P vs I curves), dark respiration (R d), and Chl cell contents of Symbiodinium sp. (Avir and Stylodid), Durusdinium trenchii (CCMP2556) and Effrenium voratum (CCMP421) cultivated at 25°C.

Species	µc (d−1)	Net P max (fmol O2 cell−1 min−1)	Gross P max (fmol O2 cell−1 min−1)	R d (fmol O2 cell−1 min−1)	R d : Gross P max	Chla (pg cell−1)	Chlc (pg cell−1)
Symbiodinium sp. (Avir)	0.41 (0.01)	5.4 (0.5)	7.5 (0.5)	–2.1 (0.3)	0.28 (0.03)	1.88 (0.21)	0.36 (0.08)
Symbiodinium sp. (Stylodid)	0.45 (0.04)	4.3 (0.3)	6.3 (0.6)	–2.0 (0.4)	0.32 (0.05)	1.54 (0.23)	0.60 (0.02)
D. trenchii	‐	9.4 (1.0)	11.1 (1.3)	–1.6 (0.3)	0.15 (0.02)	0.99 (0.11)	0.34 (0.02)
E. voratum	0.58 (0.02)	9.1 (2.3)	17.5 (2.3)	–8.4 (2.2)	0.48 (0.11)	1.51 (0.44)	0.49 (0.14)

Values in brackets represent standard deviation (n ≥ 3).

Fig. 2

(a) Comparison between activation energies (in electron volts, eV) of gross photosynthesis (Ea P) and respiration (Ea R) of Symbiodinium sp. (Avir), Symbiodinium sp. (Stylodid), Durusdinium trenchii and Effrenium voratum cultivated at 25°C. The Ea were extrapolated from the slopes of thermal response curves (Fig. 1) using the Arrhenius equation. Asterisks indicate statistically significant differences of Ea R in comparison to Ea P (*, P < 0.05). (b) Respiration (R d) to gross maximal photosynthesis (P max) ratios as a function of increasing temperature in Symbiodinium sp. (Avir) and D. trenchii. (c) R d to gross P max ratios as a function of increasing temperature in Symbiodinium sp. (Stylodid) and E. voratum. All measurements were performed for at least three independent culture replicates. The connecting line fits through the raw data. Vertical bars indicate the standard deviation.

Impacts of respiratory inhibitor reveal coupling between respiration and photosynthesis

The impact of inhibition (in the presence of SHAM) of the mitochondrial alternative oxidase (AOX) respiratory pathway on maximal photosynthetic capacity is reported in Fig. 3. Although chemical inhibitors might affect cell physiology at multiple levels (e.g. Roberty et al., 2014), several studies have described SHAM as being capable of inhibiting the activity of the mitochondrial AOX pathway (but not other respiratory pathways) in several model algal species (Bailleul et al., 2015; Larosa et al., 2018; Murik et al., 2019), including Symbiodiniaceae (Oakley et al., 2014b). In our study, addition of 8 mM SHAM caused a mild 20–25% decrease in R d in Symbiodinium sp. (Avir) (t‐test: F 3,3 = 1.223, P = 0.0391), Symbiodinium sp. (Stylodid) (t‐test: F 2,2 = 1.570, P = 0.0001), D. trenchii (t‐test: F 2,2 = 1.746, P = 0.0177), and E. voratum (t‐test: F 5,5 = 1.103, P = 0.0328). In relation to photosynthesis, addition of SHAM to Symbiodinium sp. (Avir), caused a c. 30% increase in the maximal rates of gross P (t‐test: F 5,5 = 2.179, P < 0.0001) but no changes in maximal rETR (t‐test: F 5,5 = 2.465, P = 0.6271) (Fig. 3). In the case of Symbiodinium sp. (Stylodid), the addition of SHAM caused a reduction in both maximal gross P (20%, t‐test: F 5,5 = 1.702, P = 0.0339) and maximal rETR (40%, t‐test: F 5,5 = 1.306, P = 0.0047). For D. trenchii and E. voratum, the addition of SHAM caused no alteration to their maximal gross P (t‐test: F 5,5 = 1.582, P = 0.9962; F 5,5 = 4.829, P = 0.1053, respectively) but a c. 30% and 22%, respectively, decrease in the maximal rETR (t‐test: F 5,5 = 1.651, P < 0.0001; F 5,5 = 1.367, P = 0.0004, respectively).

Fig. 3

Comparison of maximal gross oxygen evolution (gross P max) and maximal relative electron transport rate of photosystem II (rETRmax) in Symbiodinium sp. (Avir), Symbiodinium sp. (Stylodid), Durusdinium trenchii and Effrenium voratum. Measurements were performed both in the absence and presence of salicylhydroxamic acid (SHAM), an inhibitor of mitochondrial alternative oxidase activity. All the measurements were performed for at least three independent culture replicates. Box plots show the interquartile range (box), median (horizontal line), and range (vertical bars). *, P < 0.05.

Subcellular organization does not reflect photosynthesis to respiration coupling

Transmission electron micrographs (Figs 4, S2) showed that both Symbiodinium sp. (Avir) and E. voratum contained an average of 10 chloroplastic sections (Lajeunesse et al., 2010) per cell (t‐test: F 13,13 = 3.736, P = 0.6960). The number of mitochondrial sections in contact per chloroplastic section was c. 0.5 (± 0.1) for Symbiodinium sp. (Avir) and 0.6 (± 0.3) for E. voratum, with no differences between the two species (t‐test: F 12,10 = 7.595, P = 0.6510). The average values of chloroplastic section area per cell were 857 (± 199) and 824 (± 202) µm2 for Symbiodinium sp. (Avir) and E. voratum, respectively, with no differences between the two species (t‐test: F 5,5 = 1.033, P = 0.7799). The average values for mitochondrial section area per cell were 258 (± 77) and 276 (± 51) µm2 for Symbiodinium sp. (Avir) and E. voratum, respectively, with no differences between the two species (t‐test: F 5,5 = 2.293, P = 0.6546).

Fig. 4

Transmission electron micrographs illustrating subcellular organization of Symbiodinium sp. (Avir) (a, c) and Effrenium voratum (b, d). Arrows show the mitochondrial–chloroplastic contacts. Bars: 2 μm (a, b); 0.5 μm (c, d). C, chloroplast; M, mitochondria with tubular cristae; P, pyrenoid.

Growth, photosynthetic and respiratory adjustments during long‐term exposure to an elevated temperature

We then exposed the two Symbiodinium sp. (Avir and Stylodid) to 32°C temperature conditions for 1 yr. Both strains were shown to be able to tolerate such temperatures over the course of a couple of wk (Dang et al., 2019) but exhibited different degrees of photosynthesis to respiration coupling under an optimal range of temperatures (Figs 2, 3); that is, photosynthetic activity is not coupled to respiration in Avir and is coupled in Stylodid. The two Symbiodinium sp. displayed differential responses to long‐term exposure to elevated temperature. We measured a stable c. 37% increase in µc in Symbiodinium sp. (Avir) throughout 1‐yr of warming (RM ANOVA: F 3,2,6 = 11.10, P = 0.0073) (Fig. 5). For Symbiodinium sp. (Stylodid), µc was lower when cells were cultivated at 32°C, although it showed a mild recovery after 1 yr at this temperature (RM ANOVA: F 3,2,6 = 9.910, P = 0.0097). With respect to photosynthesis, we measured a higher and stable maximal gross P after 7 months of cultivation at 32°C (RM ANOVA: F 2,5,10 = 5.819, P = 0.0211) for Symbiodinium sp. (Avir) but no change of maximal gross P in Symbiodinium sp. (Stylodid) (RM ANOVA: F 2,5,10 = 2.857, P = 0.1043) (Fig. 5). TRCs (Fig. 6) showed that both respiration rates and the slopes of respiratory TRCs declined (flattening of the slope and decline of Ea R) for Symbiodinium sp. (Avir) during the 1 yr of incubation at 32°C (Fig. 6b). This reflects a progressive decline in the temperature dependence of this metabolic process. By contrast, for Symbiodinium sp. (Stylodid), no change in respiration rates was observed (Fig. 6e). However, the comparison of Ea P with Ea R in Symbiodinium sp. (Stylodid) (Fig. 6f), indicates that exposure to 32°C conditions caused a divergence in the temperature dependence of photosynthesis and respiration (i.e. 2.2‐fold higher Ea P than Ea R; t‐test: F 5,5 = 9.756, P < 0.0001), with respect to the cells growing at 25°C. This metabolic uncoupling is also reflected by the slight increase in net O2 evolution (although not statistically significant after 1 yr; t‐test: F 5,5 = 1.282, P = 0.0707) in the presence of SHAM (Fig. S3), which is similar to the uncoupled phenotype of Symbiodinium sp. (Avir) (Fig. 3). However, for both Symbiodinium sp. (Avir) and (Stylodid), these photosynthetic and respiratory changes observed during the 1 yr of warming were rapidly reversed (within 1 wk) when cells were returned to the control conditions of 25°C (Figs S4, S5). For Symbiodinium sp. (Avir), the Chl‐a cell content declined from 1.9 to 1.4 pg cell−1 during the 1‐yr period at 32°C but recovered to 1.6 pg cell−1 when cells were returned to 25°C (RM ANOVA: F 3,2,6 = 6.949, P = 0.0223). For Symbiodinium sp. (Stylodid), no changes in the Chl‐a cell content were measured during the 1‐yr period at 32°C (RM ANOVA: F 2,2,4 = 1.474, P = 0.3315).

Fig. 6

Changes in thermal response curve characteristics during the 1‐yr‐long experiment at 32°C, in comparison to cells kept at 25°C. (a, d) Gross maximal photosynthesis (P max) of Symbiodinium sp. (Avir) and Symbiodinium sp. (Stylodid). (b, e) Respiration (R d) of Symbiodinium sp. (Avir) and Symbiodinium sp. (Stylodid). (c, f) Activation energies of gross photosynthesis (Ea P) and respiration (Ea R) of Symbiodinium sp. (Avir) and Symbiodinium sp. (Stylodid). All measurements were performed for at least three independent culture replicates. The connecting line fits through the raw data. Vertical bars indicate the standard deviation.

Photosynthetic affinity for CO2 is enhanced during long‐term exposure to an elevated temperature

The photosynthesis of Symbiodinium sp. (Avir) under different inorganic carbon availabilities was finally analysed (Fig. 7, Table 3). The slope of the CO2‐limited part of the curve was higher for cells grown at 32°C after 7 months and 1 yr (n = 2), than for cells cultivated at 25°C (one‐way ANOVA: F 2,6 = 8.189, P = 0.0193). The corresponding K0.5 was not different between cells at 25 and 32°C (one‐way ANOVA: F 2,6 = 1.462, P = 0.3040). Neither the slope nor the the K0.5 of the Ci‐limited part of the curve measured for cells exposed to 32°C conditions over the course of 1 yr were statistically different from the 25°C control condition (one‐way ANOVA: F 2,6 = 3.727, P = 0.0887; F 2,6 = 2.276, P = 0.1839).

Fig. 7

Photosynthetic response to inorganic carbon (P vs CO2 curves) of Symbiodinium sp. (Avir) when exposed to 25°C temperature conditions and during the 1‐yr‐long experiment at 32°C. Vertical bars indicate standard deviation of at least two independent replicates.

Table 3

Parameters extrapolated from the photosynthetic response to CO2 and inorganic carbon (Ci) of Symbiodinium sp. (Avir) exposed to temperature conditions of 25°C and 1 yr at 32°C.

	Slope CO2 (fmol O2·cell−1 min−1 (μM CO2)–1)	K0.5 CO2 (μM)	Slope Ci (fmol O2·cell−1 min−1 (μM−1 Ci)–1)	K0.5 Ci (μM)
25°C	0.78 (0.14)	11 (8)	0.0039 (0.0007)	307 (213)
32°C, 7 months	1.35 (0.25)*	5 (1)	0.0056 (0.0010)	134 (30)
32°C, 1 yr	1.35 (0.29)*	3 (1)	0.0056 (0.0012)	47 (25)

‘Slope CO2’ is the slope of the CO2‐limited part of the P vs CO2 curve (Fig. 7) (Ruan et al., 2017); K0.5 CO2 is the CO2 concentration needed to give half‐maximum photosynthetic activity; ‘Slope Ci’ is the slope of the Ci‐limited part of the P vs Ci curve (not shown); K0.5 Ci is the Ci concentration needed to give half‐maximum photosynthetic activity. Values in brackets represent standard deviation (n ≥ 2). Asterisks indicate statistically significant differences from the 25°C condition (*, P < 0.05).

Discussion

In this study we have shown the following: first, Symbiodiniaceae species have different degrees of photosynthesis to respiration coupling; second, coupling between photosynthesis and respiration is strongly controlled by temperature, influencing how species respond to temperature changes; and third, photosynthesis to respiration coupling does not define the capability of species to adapt to long‐term temperature increase.

Symbiodiniaceae have different metabolic thermal sensitivities

Thermal response curves for photosynthesis and respiration have been used to describe species' ability to cope with fluctuating temperature conditions (Karsten et al., 2016; Prelle et al., 2019) and adaptation to habitats with contrasting thermal regimes (Berry & Bjorkman, 1980; Collier et al., 2017). Our TRCs showed that Symbiodinium sp. (Avir) can tolerate a wide range (10–35°C) of temperatures and rapidly fluctuating temperatures, in comparison to other Symbiodiniaceae. This could relate to the species' association to a host – a sea anemone, Anemonia viridis, which inhabits intertidal and sublittoral zones of temperate latitudes (Suggett et al., 2012; Roberty et al., 2016) (i.e. areas characterized by relatively wide daily and seasonal changes of temperature conditions, Table 1). By contrast, species as D. trenchii, which are found in tropical regions exposed to high thermal regimes (Pettay et al., 2015; Silverstein et al., 2015), exhibited the lowest tolerance to low (15°C) and/or oscillating temperatures.

Diverse degrees of photosynthesis and respiration thermal sensitivity reveal the coupling or uncoupling between the two energetic metabolisms

With regards to the TRCs, the comparisons between the temperature‐dependent increases, in the 10–30°C range, of photosynthesis and respiration are of particular interest. In Symbiodinium sp. (Avir) and D. trenchii, respiration is less sensitive (lower Ea R, and declining R d : gross P) than photosynthesis, and so metabolic changes appear uncoupled (Figs 1, 2). This is also consistent with the early work of Iglesias‐Prieto et al. (1992) in which a lower temperature dependence of respiration in comparison to photosynthesis in S. microadriaticum was reported. By contrast, for Symbiodinium sp. (Stylodid) and E. voratum, we interpret the same temperature dependency of both respiration and photosynthesis (same Ea for R and P, and constant R d : gross P) as evidence that the two energetic processes are strongly coupled (Figs 1, 2). Analogue variations of respiration and photosynthesis in TRCs have been previously observed for green algae (Pierangelini et al., 2019) and in a marine diatom (Schaum et al., 2018), and they can be directly linked to the tight coupling of respiratory and photosynthetic activity in these organisms (e.g. the exchange of metabolic substrates and/or re‐routing of the reducing power generated in the chloroplast towards mitochondria and the import of mitochondrial ATP into the chloroplast) (Cardol et al., 2009; Bailleul et al., 2015). Above 30°C, there is a rapid decline in photosynthesis but no negative effects on respiration, reflecting the lower sensitivity of this metabolic pathway to high temperatures (Iglesias‐Prieto et al., 1992; Grégoire et al., 2017; Barton et al., 2020), and showing that such temperatures disrupt the respiration–photosynthesis coupling. Taken together, the presence of an interspecific diversity in thermal metabolic sensitivity of Symbiodiniaceae reflects a differential degree of coupling between their photosynthetic and respiratory metabolisms.

Inhibition of respiration reveals the connection with photosynthesis

To further determine whether respiratory metabolism is coupled to photosynthesis, we compared the changes in photosynthetic O2 evolution and rETR capacity caused by exposure to SHAM (Fig. 3). We reasoned that if photosynthesis depends on respiration, then an inhibition of respiration should affect photosynthesis, as well. In Symbiodinium sp. (Stylodid) and E. voratum, inhibition of the AOX respiratory pathway was concomitant with a significant decrease in photosynthetic rETRmax. Since the inhibition of respiration has a greater impact on rETR than on maximal gross P, this suggests that an electron flow from water (i.e. at the PSII donor side) to oxygen (i.e. reduced by mitochondrial AOX) takes place in the absence of an inhibitor (Cardol et al., 2008; Roberty et al., 2014) in these species. This result is also very similar to that obtained for the marine diatom Phaeodactylum tricornutum, for which an interaction between the two metabolic processes is necessary to balance the levels of ATP and NADPH in the plastid (Bailleul et al., 2015; Murik et al., 2019). In the case of Symbiodinium sp. (Avir), we found no effect related to the inhibition of AOX on rETRmax. As in Symbiodiniaceae, AOX activity accounts for c. 25–30% of the respiration (Oakley et al., 2014b); the 30% increase of Symbiodinium sp. (Avir) maximal gross P could be related to a reduced O2 consumption, rather than any effect on photosynthesis. These results provide further evidence that the photosynthetic activity of Symbiodinium sp. (Avir) is uncoupled from respiration. By contrast, photosynthesis of D. trenchii was downregulated when AOX activity was inhibited, reflecting the occurrence of metabolic coupling. However, for this species, this is in contrast to the different temperature dependences of photosynthesis and respiration observed in the TRCs. Thus, we describe D. trenchii as a phenotype with a photosynthesis coupled to respiratory activity, but its photosynthesis was easily disengaged from respiration, following exposure to (stressful) conditions such as low temperatures (onset of the TRCs).

Subcellular organization does not reflect physiological functioning

It has been suggested that a physiological coupling between photosynthesis and respiration could be favoured by an intracellular placing of mitochondria in close proximity to chloroplasts (Prihoda et al., 2012). Nonetheless, here we found that this positioning of organelles occurred in both Symbiodinium sp. (Avir) and E. voratum, regardless of their differential levels of photosynthesis–respiration interaction. This suggests that the cross‐talk between mitochondria and chloroplasts is not related to their spatial organization, but rather to different capacities to exchange metabolites.

Our ultrastructural analysis seems also to indicate that in both Symbiodinium sp. (Avir) and E. voratum, as well as in other Symbiodiniaceae (Leggat et al., 2002; Shoguchi et al., 2013; Lee et al., 2015), some pyrenoids are not immersed in the chloroplast, as in green algae (Treves et al., 2016; Mackinder et al., 2017; Pierangelini et al., 2017) and haptophytes (Stojkovic et al., 2013; Heureux et al., 2017). This may suggest that this micro‐compartment may not necessarily work as proposed for the latter, that is, being the aggregation site of Rubisco, and favouring CO2 accumulation in proximity to this enzyme (Meyer & Griffiths, 2013; Freeman Rosenzweig et al., 2017). However, we do not exclude the possibility that the pyrenoid position and functioning may change under environmental conditions regulating CCM activity or when in hospite (Leggat et al., 2002).

Photosynthesis–respiration coupling will not determine the adaptive response of Symbiodiniaceae to global warming

For species as Symbiodinium sp. (Avir), which are characterized by uncoupled photosynthesis–respiration, exposure to prolonged elevated temperature conditions (32°C) is expected to make photosynthetic and respiratory metabolisms work even more independently, without compromising cell growth. In particular, throughout the 1‐yr period at 32°C, the photosynthetic metabolism of this species responded following the ‘hotter is better’ model (Jin & Agustí, 2018), with increased performance under hotter conditions but exhibiting susceptibility to lower temperatures (10°C, Fig. S1c). The higher photosynthetic performance could also be linked to a slight enhancement of the CCM activity, reflecting a higher cell need for CO2. We must also point out that changes in both µc and photosynthetic performance showed no progressive increasing/decreasing trend during the 1‐yr period (Fig. 5). The decline of K0.5 Ci (although not statistically significant) observed here is also comparable to what was reported by Oakley et al. (2014a) after exposing Symbiodiniaceae to elevated temperature for a few wk. This indicates that the overall cell response arose shortly (within days) after exposure at 32°C temperatures, and thus, the changes are related to acclimation rather than adaptation. This is also supported by the quick recovery of photosynthesis when cells returned to 25°C conditions (Fig. S4).

In contrast to the increase in photosynthesis, respiration rates of Symbiodinium sp. (Avir) became progressively lower and less temperature dependent during the long‐term warming period. Following the TRC respiration responses (type I and II) described for plants exposed to warm conditions (Atkin & Tjoelker, 2003; Smith & Dukes, 2013), our results show that the type I response (indicated by the lower slope of TRCs) and type II response (reduction in the elevation of TRCs) co‐occur in Symbiodinium sp. (Avir) when cells are grown at an elevated temperature. Our results also show that the two responses take place over a different time scale, with the type I response occurring later than the type II response. However, neither type I nor type II responses were adaptive, and respiratory metabolism was rapidly reversed when cells were returned from 32 to 25°C conditions.

In contrast to Symbiodinium sp. (Avir), for which photosynthesis is not coupled to respiration, other Symbiodiniaceae such as Symbiodinium sp. (Stylodid), which do exhibit photosynthesis–respiration coupling, will be more negatively affected by prolonged exposure to elevated temperature. Our results suggest that when Symbiodinium sp. (Stylodid) was cultivated at 32°C, interaction between photosynthetic and respiratory metabolism is prevented (showing higher Ea P than Ea R, Fig. S3), and despite the 1‐yr acclimation time, adaptive responses to re‐establish the connection between metabolisms did not occur. This probably contributed to the observed decrease in µc. Negative effects on cell growth in Symbiodinium sp. (Stylodid) were also reported by Dang et al. (2019), after 2 wk of exposure to high temperature conditions (33°C), supporting the occurrence of the µc decline shortly after heat exposure, and lack of adaptive response in this species.

Photosynthesis–respiration coupling might determine the metabolic compatibility between Symbiodiniaceae and their hosts

The presence of either a coupled or uncoupled photosynthetic activity to respiration in Symbiodiniaceae, may also have implications for our understanding of their life in symbiosis. It is known that when in hospite, the Symbiodiniaceae cell division is strictly controlled and reduced by the animal hosts (Falkowski et al., 1993; Xiang et al., 2020). As cell division is directly linked to respiratory metabolism (which converts carbohydrates produced during photosynthesis into cellular metabolites used in the construction of cells; Falkowski et al., 1985; Geider & Osborne, 1989), respiration rates must also be kept low when in hospite (Rädecker et al., 2018; M. Pierangelini, unpublished). However, at the same time, photosynthesis must be maintained in order to guarantee the translocation of the fixed carbon to the host (Matthews et al., 2017, 2018; Krueger et al., 2020). Thus, we put forward the idea that having photosynthetic activity which can function independently from respiration could represent an advantage for Symbiodiniaceae in establishing successful endosymbiosis. Nonetheless, the coupling of photosynthesis to respiration could still be advantageous to hosts by facilitating the coordination of the overall metabolic activity of the photosymbionts; that is, by directly controlling respiration, hosts can rule on photosynthesis, as well. Overall, our observations suggest that photosynthesis–respiration interaction could play a role in determining the metabolic compatibility between Symbiodiniaceae and their hosts (Morris et al., 2019).

Conclusions

This work shows that the coupling between chloroplastic and mitochondrial energetic metabolisms occurs to different extents amongst Symbiodiniaceae. Species of this family are characterized by having photosynthetic activity that is either coupled to or uncoupled from respiratory metabolism (including intermediate phenotypes). The coupling between metabolisms is also disrupted upon changes in thermal conditions, both below and above the optimum, influencing how species perform during stable or fluctuating temperatures. During long‐term exposure to high and/or above‐optimal thermal regimes, the two Symbiodiniaceae (with a photosynthetic activity coupled or uncoupled to respiration) displayed no evidence of adaptation, despite the high numbers of cell generations (310 and 190 for Symbiodinium sp. (Avir) and (Stylodid), respectively). In both species, adjustments of energetic metabolism occurred shortly after heat exposure and were fully reversible even after the 1‐yr period of warming. This suggests that neither a coupled nor an uncoupled phenotype provides an advantage in terms of defining the capacity of a given species to adapt to global warming. However, we cannot exclude the possibility that other responses have been developed in long‐term heat‐stressed cells (Chakravarti et al., 2020). Ecologically, this lack of adaptive capabilities in Symbiodiniaceae confirms the susceptibility of coral reefs to ocean warming. At this stage, more studies are also necessary to predict the impact that one or another of these energetic metabolic traits will have on the susceptibility of the Symbiodiniaceae‐hosts’ endosymbiotic relationships to ocean warming.

Author contributions

MP and PC conceived the research, analysed the data, and wrote the manuscript; MP performed the physiological analyses; MT performed the TEM analysis; all authors reviewed the manuscript.

Fig. S1 Comparison of thermal response curves started at 10 and 15°C, for gross maximal photosynthesis and respiration of Symbiodinium sp. (Avir) cultivated at 25 and 32°C.

Fig. S2 Transmission electron micrographs illustrating subcellular organization of Symbiodinium sp. (Avir) and Effrenium voratum.

Fig. S3 Comparison of maximal net oxygen evolution and maximal electron transport rate of photosystem II in Symbiodinium sp. (Stylodid) cultivated at 32°C for 7 months and 1 yr.

Fig. S4 P vs I curves of Symbiodinium sp. (Avir) and Symbiodinium sp. (Stylodid) showing the recovery of their gross photosynthetic activities when cells cultivated at 32°C were returned to 25°C conditions.

Fig. S5 Thermal response curves for Symbiodinium sp. (Avir) showing the recovery of the respiration activity and activation energy of respiration when cells cultivated 32°C were returned to 25°C conditions.

Please note: Wiley‐Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Click here for additional data file.