Quantification of cytosolic ‘free’ calcium in isolated coral cells with confocal microscopy Free

Corals belonging to the order Scleractinia are responsible for the existence of the most biodiverse ecosystems in the ocean. Coral skeletons build vast coral reef superstructures, and coral symbiosis with intracellular photosynthetic dinoflagellates (Symbiodiniaceae) drives high coral reef primary productivity (Douglas, 2009; Sheppard et al., 2017). Physiological stress imposed on corals by climate change and other anthropogenic factors is a primary driver of coral reef degradation worldwide (Hoegh-Guldberg et al., 2017). In recent decades, the crisis facing coral reefs has intensified research into many aspects of coral biology, and there have been several calls to improve knowledge of the basic cell biology underpinning coral physiology that may ultimately provide insight into coral environmental resilience and vulnerability (Davy et al., 2012; Putnam et al., 2017; Weis, 2019). Rapid advances are being made in this field, thanks to genomic and proteomic research, but distinct gaps still lie in many aspects of fundamental coral cellular biochemistry and physiology (Melzner et al., 2022). The concentration and regulation of major intracellular ions are among these key knowledge gaps.

Intracellular Ca²⁺ ions play a pivotal role in cell biology (Berridge et al., 2000). Highly reactive with a wide range of biological macromolecules (Jaiswal, 2001), most intracellular calcium occurs in a bound form, complexed with organic compounds in specific organelles and with inorganic compounds such as phosphate or carbonate (Carafoli and Krebs, 2016; Meldolesi and Pozzan, 1998). However, a small fraction of intracellular calcium occurs in a ‘free’ ionized state, and in this form, it performs the role of intracellular messenger in all forms of life, from bacteria to mammals (Berridge et al., 2000; Case et al., 2007). Indeed, fluctuations in intracellular Ca²⁺ ion concentration ([Ca²⁺]_i) are associated with almost all cellular processes, including exocytosis, gene transcription, fertilization and cell proliferation (Berridge et al., 2003). Furthermore, increases in [Ca²⁺]_i can be cytotoxic and are associated with cellular death pathways (Orrenius et al., 2003), and as such it must be tightly regulated (Bagur and Hajnóczky, 2017). Despite its wide-reaching significance in cell biology, [Ca²⁺]_i has never been quantified in corals, or in cnidarians in general, and we know relatively little about its mechanism of regulation in this group of animals.

In model organisms and cell lines, quantification of [Ca²⁺]_i and investigations into its regulation have benefitted greatly over the years because of the development of protocols using ratiometric calcium-sensitive fluorescent dyes (Grynkiewicz et al., 1985; Zhou et al., 2021). Intracellular ratiometric calcium indicators are sensitive to Ca²⁺ in the nanomolar range and produce a signal derived from the ratio of fluorescence at two excitation or emission wavelengths, making it possible to quantify intracellular calcium despite variation in dye concentration and intensity owing to factors such as uneven dye loading, dye leakage and changes in cell volume (Takahashi et al., 1999). Ratiometric calcium indicator dyes such as Indo-1 and Fura 2 are used widely with a variety of fluorescence detection methods including laser scanning confocal microscopy, which enables calcium measurements with high spatial resolution (Habara et al., 1996; Takahashi et al., 1999). These techniques to measure [Ca²⁺]_i hold the potential to greatly facilitate investigations into calcium biology in corals, but currently remain undeveloped. Two earlier studies made measurements of Fura 2 using fluorospectrometry to approximate relative changes in intracellular ‘free’ calcium in coral cell suspensions (Fang et al., 1997; Huang et al., 1998) and another study also mapped spatial differences in the fluorescence of the ‘Calcium Orange’ indicator dye in coral larvae (Clode and Marshall, 2004). However, none of these studies made quantitative measurements of intracellular free calcium or used live cell imaging. Other previous work on intracellular calcium also includes X-ray microanalysis of total intracellular calcium, which is dominated by bound, complexed calcium rather than the free ionic form (Clode and Marshall, 2004; Marshall et al., 2007).

In the present study, we adapted procedures widely used in classic model systems (such as mammalian cell lines) to make the first quantitative measurements of intracellular free calcium [Ca²⁺]_i in corals. Working with the coral species Stylophora pistillata, we loaded the ratiometric calcium dye Indo-1 into isolated coral cell preparations and analysed them with confocal microscopy. We quantified [Ca²⁺]_i in coral cells containing dinoflagellate symbionts and those without symbionts, and verified our procedure by carrying out the same methods on Chinese Hamster Ovary (CHO) cells with a previously published [Ca²⁺]_i. We then tested whether our imaging method allowed us to detect predictable changes in [Ca²⁺]_i induced by a classical inhibitor of intracellular Ca²⁺ regulation, thapsigarin. Finally, we compared [Ca²⁺]_i in coral cells undergoing apoptosis with non-apoptotic cells, which may be useful for future investigations into the cellular mechanisms of stress responses in corals.

Experiments were conducted on Stylophora pistillata Esper 1797 maintained in the long-term coral culture facilities at the Centre Scientifique de Monaco supplied with flowing seawater from the Mediterranean Sea. The corals used in the current investigation were kept in aquaria at 25°C, pH 8, salinity 38. Continuous monitoring of temperature, pH and salinity with sensors (Ponsel, France) and a monitoring system (Enoleo, Monaco) ensured that these parameters did not vary more than 0.5°C, 0.1 pH units and salinity 0.2 outside of these values. Irradiance was supplied at 200 µmol photons m⁻² s⁻¹ on a 12 h:12 h light:dark cycle. Food was provided daily with rotifers and twice a week with Artemia salina nauplii.

Preparation and analysis of coral cells was conducted in 0.45 µm filtered seawater (FSW) except where we state artificial seawater (ASW) was used. Two types of ASW solutions were used in the study, made according to Bénazet-Tambutte et al. (1996). The first contained 10 mmol l⁻¹ calcium, consisting of 500 mmol l⁻¹ NaCl, 10 mmol l⁻¹ KCl, 10 mmol l⁻¹ CaCl₂ 2H₂O, 29 mmol l⁻¹ MgSO₄ 7H₂O, 27 mmol l⁻¹ MgCl₂ 6H₂O and 2 mmol l⁻¹ NaHCO₃. The second contained no calcium (zero calcium), consisting of 510 mmol l⁻¹ NaCl, 10 mmol l⁻¹ KCl, 29 mmol l⁻¹ MgSO₄ 7H₂O, 27 mmol l⁻¹ MgCl₂ 6H₂O, 2 mmol l⁻¹ NaHCO₃ and 9.5 mmol l⁻¹ EGTA. In both cases, pH was adjusted to 8.1 using 0.01 mol l⁻¹ NaOH and HCl.

Coral cells were prepared from branches removed from a coral colony using surgical bone cutters. The tissue removed from branches by gentle brushing with a soft-bristle toothbrush in 50 ml of 0.45 µm FSW, and the resulting suspensions of coral cells were then filtered first through a 100 µm mesh, then a 40 µm mesh, and were then centrifuged at 350 g for 5 min at 25°C. The resulting pellets were resuspended in 10 ml FSW. Aliquots of this cell suspension were used for dye loading as described in the following sections. After dye loading, cells were mounted on silane-coated slides and glass coverslips that were sealed to the slide on two sides by silicone grease (Venn et al., 2009).

All microscopy was conducted using a Leica SP8 confocal microscope (Leica, Germany) using a ×63 water immersion lens. All observations were performed in darkness and temperature was maintained at 25°C with a temperature-controlled microscope stage (PeCon, Germany).

We focused our analyses on two categories of coral cells in the isolated cell preparations: symbiont-containing cells (accommodating one to three intracellular symbiotic dinoflagellates) and symbiont-free cells (Fig. 1). Symbiont-containing cells were of oral or aboral endodermal origin, whereas symbiont-free cells may have come from the oral or aboral ectoderms, or the oral or aboral endoderm layer. Symbiont-containing and symbiont-free cells were inspected using 552 nm laser line (0.1% laser power) with images captured in the brightfield transmitted light channel to allow identification of the major intracellular compartments (Fig. 1). Brightfield microscopy allowed us to distinguish the coral host cytoplasm, the intracellular symbionts and host cell lipid droplets in accordance with previous work (Chatin et al., 2023; Luo et al., 2009; Peng et al., 2011; Venn et al., 2009) (Fig. 1). Intracellular symbionts could also be visualized by chlorophyll autofluorescence captured by confocal microscopy using the 488 nm laser (0.1% power) with emission collected in a channel between 650 and 700 nm (Fig. 1).

In order to determine the position of mitochondria in cells, preparations were stained with MitoTracker Deep Red (Thermo Fisher Scientific). Stock solutions of MitoTracker Deep Red were prepared in DMSO at 1 mmol l⁻¹ and cells were stained by incubation in a 200 nmol l⁻¹ working solution in seawater for 30 min in darkness. After dye loading, cells were rinsed via centrifugation and resuspended in 2 ml FSW. Mitochondria were imaged using a 638 nm laser (0.01%) with emission captured at 665±10 nm.

The procedure for loading cells with the calcium indicator Indo-1 was adapted from protocols for mammalian cells (e.g. Nelemans, 2006). Isolated coral cells were incubated in a solution of FSW and the cell permeant acetoxymethyl ester acetate of Indo-1 (Indo-1 AM) (Thermo Fisher Scientific) for 30 min in the dark at 21–24°C. Solutions were prepared from 5 mmol l⁻¹ stock in DMSO with working concentrations at 6.25 µmol l⁻¹ Indo-1 AM and 0.1% DMSO in FSW. Following the loading period, cells were rinsed via centrifugation (350 g, 5 min, 25°C) and resuspended in FSW. Cells were then incubated in FSW for a further 1 h in darkness to ensure complete hydrolysis of acetoxymethyl esters.

Indo-1 analysis was carried out with a Leica SP8 confocal microscope and ×63 water immersion lens, with all observations in darkness and temperature maintained at 25°C. Cell preparations were visualised in brightfield using the 552 nm laser at the low power (0.1%) to select individual cells and allow the adjustment of zoom settings without exposing the cells to UV. Once an individual cell was selected, the 552 nm laser was deactivated, and the UV 355 nm laser line was activated (1.3% power). Indo-1 fluorescence was collected with hybrid detectors in two channels at 430±10 and 460±10 nm. These wavelengths were selected on the basis of analysis of the intracellular Indo-1 spectrum at minimum (zero [Ca²⁺]) or maximum (saturating [Ca²⁺]) obtained in vivo by incubating cells in ASW containing zero or 10 mmol l⁻¹ Ca²⁺ in the presence of the ionophore ionomycin (20 µmol l⁻¹) for 20 min. Images were captured with 512×512-pixel resolution at 400 Hz. Each cell was imaged by capturing 8–10 frames in the x/y plane, throughout the depth (z plane) of the cell (z-stack profile) (Fig. S1). Acquisitions made with identical settings on cells prepared with no Indo-1 staining showed no autofluorescence.

Post image acquisition, z-stack profiles were inspected in LAS-X Lecia software to look for patterns of exclusion, compartmentalisation and variation in the fluorescence ratio of dye. Indo-1 was consistently excluded from the intracellular symbionts, and sometimes excluded or incorporated in lipid droplets (Fig. 1). Generally, Indo-1 fluorescence was homogeneously distributed throughout the cytoplasm of cells (Fig. 1). In approximately 25% of cells, higher levels of Indo-1 fluorescence were observed in in mitochondria relative to the surrounding cytoplasm, indicating low levels of compartmentalization in these organelles (Fig. 1). The degree of compartmentalization did not vary with dye loading time or dye concentration, and did not interfere with measurement of [Ca²⁺]_i. Areas where Indo-1 fluorescence was homogeneously distributed were targeted for analysis of [Ca²⁺]_i by measuring in the two channels (430±10 and 460±10 nm) within a digital region of interest drawn in this zone using the LAS-X Lecia software (Fig. S1). Between 5 and 15 individual coral cells of each category were analysed per coral branch.

Chinese Hamster Ovary (CHO-K1) cells were sourced from the American Type Culture Collection (ATCC1) cultured in Dulbecco's Modified Eagle's medium (DMEM)/F12 Glutamax with Peni Streptomycin (Gibco)+5% HyClone Fetal Bovine Calf Serum (Cytiva) at 37°C and were authenticated and tested for contamination. Cells were harvested from culture medium by treating confluent monolayers with 0.05% Trypsin EDTA (Gibco), rinsed with DMEM+5% HyClone Fetal Bovine Calf Serum (Cytiva) by centrifugation (5 min, 350 g, 37°C) and resuspended to a density of 1×10⁶ cells ml⁻¹.

CHO cells with were loaded with Indo-1 AM with same concentration and duration used for coral cells. After dye loading, cells were transferred to a POC R2 chamber (PeCon) for confocal microscopy. Confocal analysis and calibration of Indo-1 were conducted according to the same procedure as with corals cells. [Ca²⁺]_i was determined in 10–15 cells from each of six separate isolations of CHO cells from culture media.

For calibration, maximum and minimum values of R were obtained in zero and 39 µmol l⁻¹ calcium buffers provided in Calcium Calibration Buffer Kits (Thermo Fisher Scientific). The zero calcium buffer contained 10 mmol l⁻¹ EGTA in 100 mmol l⁻¹ KCl, 30 mmol l⁻¹ MOPS, pH 7.2. The 39 µmol l⁻¹ calcium buffer contained 10 mmol l⁻¹ CaEGTA in 100 mmol l⁻¹ KCl, 30 mmol l⁻¹ MOPS, pH 7.2. Temperature was maintained at 37°C for dye loading, calibration and imaging of CHO cells.

Isolated coral cells were loaded with Indo-1 as described above, centrifuged (350 g×3 min) and resuspended in ASW containing 4 µmol l⁻¹ thapsigargin and 0.5% DMSO or zero Ca²⁺ ASW containing 4 µmol l⁻¹ thapsigargin and 0.5% DMSO. Stock solutions of thapsigargin were made at 770 µmol l⁻¹ in DMSO. ASW or zero Ca²⁺ ASW controls contained 0.5% DMSO only. [Ca²⁺]_i analysis was performed on symbiont-containing cells by confocal microscopy as described above. Symbiont-free cells were not analysed in thapsigargin experiments.

In order to quantify [Ca²⁺]_i in apoptotic coral cells, we first identified apoptotic cells in our cell suspensions using the commonly used Annexin V labelling technique that has been used previously on isolated corals cells (Domart-Coulon et al., 2004). After cell isolation from branches, coral cell suspensions were incubated for 30 min in FSW containing 25 µl Annexin V–Alexa Fluor 568 (Thermo Fisher Scientific) per 100 µl and first rinsed in 2 ml FSW, and then rinsed in binding buffer according to the manufacturer's instructions. Confocal microscopy of apoptotic cells identified with Annexin V–Alexa Fluor labelling was performed with excitation at 552 nm and acquisition at 578–603 nm.

To determine the proportion of cells in our preparations that were apoptotic, the number of Annexin V-positive cells were scored against the total number of cells counted. Apoptotic cells were scored as those in which the plasmalemma membrane was clearly stained with Annexin V–Alexa Fluor 568, but in which the dye was absent from the interior of the cell (Fig. 1). Positive controls for the Annexin V staining of apoptotic cells were carried out by incubating four coral branches in the apoptosis-inducing drug 5 µmol l⁻¹ camptothecin (Sigma-Aldrich) (Traganos et al., 1996) in seawater for 24 h at 25°C before isolating the cells and counting the number of Annexin V-positive cells and comparing the proportion with cells isolated from four coral branches incubated in seawater controls. To detect necrosis, cells were incubated in FSW with 5 µmol l⁻¹ Sytox Green and rinsed in FSW. Sytox Green in necrotic cells was imaged by excitation at 488 nm and capture at 510–530 nm. Necrotic cells were those in which the Sytox Green stained the interior of the cell.

To quantify [Ca²⁺]_i in apoptotic cells, cell suspensions were co-stained with Annexin V–Alexa Fluor 568 and Indo-1 AM (Fig. 1). [Ca²⁺]_i was measured in both cells that stained positively and negatively for Annexin V–Alexa Fluor 568.

Confocal images were analyzed using LAS-X software v.3.5.2.18 (Leica). Mean [Ca²⁺]_i was determined from 5–15 cells per coral branch. Comparison between experimental treatments was made by comparing the mean [Ca²⁺]_i obtained from three separate branches in thapsigargin experiments (Fig. 2) and four separate branches in apoptosis experiments (Fig. 3) (i.e. branches were used as statistical replicates). Statistical analysis was carried out using SPSS software v. 26 (IBM). Data were tested for normality and homogeneity of variance with Shapiro–Wilk tests and Levene's tests, respectively, and compared by one-way ANOVA.

Ratiometric analysis of Indo-1 was performed in symbiont-containing and symbiont-free cells from 10 coral branches. The high spatial resolution of confocal microscopy was important for measurements in symbiont-containing cells because host cytoplasm occupied a restricted space relative to the symbionts which dominated the cell volume (Fig. 1). Previous studies of intracellular pH using the dye SNARF-1 in isolated coral cells have also benefited from confocal microscopy for the same reason (Venn et al., 2009). Our Indo-1 measurements showed that [Ca²⁺]_i was 186±61 nmol l⁻¹ (mean±s.d.) in symbiont-containing coral cells and 188±80 nmol l⁻¹ in symbiont-free coral cells (Table 1). Both values lie within the typical range for most cells, including other marine invertebrates. In neurons of the spiny lobster (Panulirus interruptus) analysed with Indo-1 and two-photon microscopy, mean [Ca²⁺]_i was reported to be 140±8 nmol l⁻¹ (Kadiri et al., 2011). In the marine sponge Suberites domuncula, [Ca²⁺]_i analysed with Fura 2 was reported to be between 100 and 200 nmol l⁻¹ (Krasko et al., 1999).

We analysed [Ca²⁺]_i in CHO cells using the same method for coral cells in order to validate our quantification procedure on a model cell type for which [Ca²⁺]_i had already been published. We obtained a mean [Ca²⁺]_i value of 58±26 nmol l⁻¹ from six separate isolations of cells from culture medium (Table 1). Previous investigations of [Ca²⁺]_i in CHO cells report resting values in the normal range of most cells (Berridge et al., 2000) of between 40 nmol l⁻¹ (Ravdin et al., 1988) and ∼120 nmol l⁻¹ (Stevenson et al., 1986). The mean value for CHO we obtained here falls within this range, which supports the validity of our [Ca²⁺]_i quantification procedure.

To investigate whether our quantification procedure was sensitive to changes in [Ca²⁺]_i induced by classical inhibition of [Ca²⁺]_i regulation, we carried out experiments with thapsigargin in ASW and zero calcium ASW (Fig. 2; Table S1). Derived from the Mediterranean plant Thapsia garganica, thapsigargin inhibits the sarcoplasmic/endoplasmic reticulum calcium ATPase (SERCA) pump, responsible for transporting Ca²⁺ from the cytoplasm into the lumen of the endoplasmic reticulum (ER) or sarcoplasmic reticulum (SR) (Jaskulska et al., 2020; Wictome et al., 1992). This pump is part of the regulatory machinery that maintains low [Ca²⁺]_i in the cytoplasm (Berridge et al., 2003). Typically, inhibition of the SERCA pump by thapsigargin leads to an increase in [Ca²⁺]_i owing to leakage from the ER into the cytoplasm (Gericke et al., 1993; Thastrup et al., 1990).

Our measurements of [Ca²⁺]_i in corals cells followed the expected pattern. In three replicate experiments (each with a separate coral branch), symbiont-containing cells loaded with Indo-1 AM exposed to 4 µmol l⁻¹ thapsigargin and 0.5% DMSO in ASW had elevated [Ca²⁺]_i relative to controls incubated in ASW+0.5% DMSO (Fig. 2). We repeated the experiments in zero calcium seawater and recorded the same pattern with thapsigargin-treated cells having a significantly higher [Ca²⁺]_i compared with DMSO-only controls in Ca²⁺-free ASW. This increase in [Ca²⁺]_i, which occurs despite the lack of external calcium, supports the classical mechanism described above, suggesting that increases in [Ca²⁺]_i come from an internal source (i.e. the ER) (Malintan et al., 2019; Thastrup et al., 1990).

Interestingly, in zero Ca²⁺ ASW experiments, both treatment and control cells had significantly lower [Ca²⁺]_i than their equivalent treatments in normal ASW (Fig. 2). This suggests that removal of calcium in the surrounding seawater reduces its influx into the cell and enhances rates of its efflux. Mechanisms of calcium influx and efflux from coral cells have not been functionally characterised, but molecular studies have identified transporters in S. pistillata and other coral species including Ca²⁺ ATPases (Barott et al., 2015; Zoccola et al., 2004) that play a well-known role in calcium efflux in model mammalian systems (Brini and Carafoli, 2011).

Apoptosis is a form of programmed cell death that functions as cell deletion mechanism destroying dysfunctional, damaged and diseased cells, and is thus fundamental in development, growth and tissue homeostasis (Cavalcante et al., 2019; Elmore, 2007). In corals and other cnidarians, there has been substantial interest in apoptosis in terms of its evolution in metazoans (Dunn et al., 2006; Quistad et al., 2014), and its role in the turnover of reproductive tissues (Shikina et al., 2020), regulation of symbiont populations (Dunn and Weis, 2009) and in stress responses (Kvitt et al., 2016; Tchernov et al., 2011). Apoptosis can be triggered by various cellular signals, especially intracellular free Ca²⁺ (Rizzuto et al., 2003; Sukumaran et al., 2021), but in corals, [Ca²⁺]_i levels have never been directly investigated in apoptotic cells. Isolated coral cell suspensions typically contain a proportion of apoptotic cells (Domart-Coulon et al., 2004; Laurent, 2013). Here, we took advantage of the presence of this small population of apoptotic cells in our isolated cell suspensions to test whether our quantification procedure could determine differences in [Ca²⁺]_i in apoptotic versus non-apoptotic coral cells.

In a first step, we quantified the proportion of apoptotic cells in our suspensions by staining with Annexin V–Alexa Fluor 568 (with no Indo-1 staining). In cell counts of 884 cells from four coral branches, 6% of cells were scored as apoptotic. In terms of cell categories, 3.5% of the total cell population were symbiont-free apoptotic cells and 2.5% were symbiont-containing cells. Sytox Green staining indicated that 5% of cells in our preparations were necrotic. This incidence of apoptosis in our preparations was similar to that found in previous studies (e.g. 5% in Domart-Coulon et al., 2004; 15% in Quistad et al., 2014; and 10% in Laurent, 2013). The proportion of Annexin V-positive cells increased three times in cells isolated from coral branches exposed to the apoptosis-inducing drug camptothecin (positive control) (Table S2).

In a second step, we co-stained coral cells with Annexin V–Alexa Fluor 568 and Indo-1 AM to allow us to target apoptotic cells for quantification of [Ca²⁺]_i. We found that mean [Ca²⁺]_i was significantly elevated three times or more in both symbiont-free and symbiont-containing cells that were undergoing apoptosis, relative to non-apoptotic cells in their respective categories (Fig. 3; Table S1). This observation of elevated [Ca²⁺]_i is consistent with the widely accepted model that increases in [Ca²⁺]_i are involved in initiating the apoptosis mechanism in many mammalian cell types (Martikainen et al., 1991; Orrenius et al., 2003; Tombal et al., 1999).

Our observations are the first to report increases in [Ca²⁺]_i associated with apoptosis in coral cells and demonstrate the potential for our technique for future studies. Specifically, our technique may be useful in deciphering the early cellular events of apoptosis involved in symbiosis dysfunction (‘coral bleaching’) caused by environmental stress such as elevated temperatures (Dunn et al., 2007; Weis, 2008). Previous non-quantitative analysis of coral [Ca²⁺]_i with Fura 2 and spectrophotometry suggested that coral cells subjected to heat stress undergo [Ca²⁺]_i increases, which may be responsible for cell apoptosis (Fang et al., 1997; Huang et al., 1998). Furthermore, previous genomic and proteomic studies on bleaching corals have reported changes in the expression of calcium-regulating proteins such as calmodulin which could be linked to apoptosis (Desalvo et al., 2008; Weston et al., 2015). Our technique has the scope to complement this earlier work by providing quantitative functional data on [Ca²⁺]_i in bleaching corals.

In the last three decades, ratiometric dyes and cell imaging have played a significant role in unveiling the role of ionic calcium in the cell biology of many animal, plant and protist systems (Zhou et al., 2021). Here, we adapted the use of the calcium dye Indo-1 and confocal microscopy to quantification of [Ca²⁺]_i in corals for the first time. Using this approach, we showed that corals maintain resting [Ca²⁺]_i in the nanomolar range typical of most cells and other marine organisms, approximately 100,000 times lower than the surrounding seawater. Next, we demonstrated that our imaging technique measures changes in [Ca²⁺]_i associated with classic inhibition of Ca²⁺ storage (by thapsigargin) and changes in the Ca²⁺ gradient with seawater. Finally, we showed the potential utility of our technique for investigating disturbances in [Ca²⁺]_i in apoptotic cells, which will interest researchers working on coral stress at the cellular level.

Looking to the future, we can build on our methods in various ways, including the application of other calcium-indicator dyes. Indo-1 was used in the present study because it is ratiometric, thus allowing accurate quantification, and we used it successfully for end-point measurements of [Ca²⁺]_i. However, it has limitations for continuous monitoring of cells because it is excitable in the UV range, which is stressful for cells, and photo-degrades relatively rapidly (Scheenen et al., 1996; Takahashi et al., 1999). Now that accurate quantification has been performed in this study, future studies on coral cells could apply dyes that are excitable at longer wavelengths and that have greater photostability. Most are not ratiometric, but they are suited to monitoring of [Ca²⁺]_i dynamics associated with signalling events. This could be advantageous in establishing how [Ca²⁺]_i dynamics are involved in host–symbiont interactions. For example, our end-point measurements with Indo-1 suggest that values of resting [Ca²⁺]_i were almost identical between symbiont-containing and symbiont-free cells, but experiments with longer wavelength dyes and high image acquisition rates (e.g. with spinning disk confocal microscopy) (Nelson et al., 2012) may reveal short-term dynamic changes in [Ca²⁺]_i that we did not capture here.

In conclusion, the present study paves the way for future functional studies on [Ca²⁺]_i in corals. It is likely that the future application of calcium imaging techniques to corals will come to the fore when used in conjunction with transcriptomic methods that may help identify the key genes and proteins associated with [Ca²⁺]_i fluctuations. Such studies could open up understanding of the role of [Ca²⁺]_i regulation in the coral cell biology of this ecologically important and environmentally sensitive group of animals.

We thank Dominique Desgré and Eric Elia for coral culture and technical assistance.