Symbiodiniaceae algal symbionts of Pocillopora damicornis larvae provide more carbon to their coral host under elevated levels of acidification and temperature

The symbiotic relationship between stony corals and their algal endosymbionts (Symbiodiniaceae) is the cornerstone underlying the success of coral reefs in the oligotrophic tropical oceans^1,2. In this partnership, Symbiodiniaceae photosynthetically fix inorganic carbon, assimilate environmental nitrogen, and provide organic nutrients (e.g., glucose, lipids, and amino acids) to the coral host, thereby supporting coral metabolic requirements for growth, skeletal formation, and reproduction^3,4. Inorganic nutrients from the host catabolism, in turn, support the growth of Symbiodiniaceae³. The delicate symbiotic nutritional balance between the coral host and Symbiodiniaceae can be impaired by climate change. Ocean acidification and warming may disrupt this symbiosis in a phenomenon referred to as coral bleaching^5,6. Understanding the combined impacts of these anthropogenic stressors on nutritional interactions could elucidate fundamental processes in the maintenance and breakdown of the symbiosis and identify the drivers shaping susceptibility to climate change^{7,8,9,10,11,12,13}.

Recent studies suggest that the cnidarian-algal symbiosis is regulated by passive feedback between carbon and nitrogen cycling between host and symbiont^14,15. Climate change has direct impacts on this carbon and nitrogen cycling, with possible implications for coral bleaching^16,17,18. Indeed, it has been increasingly suggested that elevated temperature could alter nutrient cycling in the coral host-algal symbiosis, leading to Symbiodiniaceae parasitism and nutrient starvation in the host metabolism^9,19. In particular, energy limitation and the associated increase in host catabolic activity have been recognized as early symptoms of the breakdown of symbiotic interaction during heat stress¹⁹. Understanding the effects of ocean acidification and warming on carbon and nitrogen exchange is thus central to understanding the susceptibility of the coral-algal symbiosis to future ocean conditions. However, while the importance of symbiotic carbon and nitrogen cycling in adult corals is well documented^11,12,13,20, our understanding of nutrient cycling in symbiotic coral larvae is limited.

Evaluating the mechanisms of establishing and maintaining a functional nutritional symbiosis is particularly crucial during the larval stage of corals. Successful survivorship, settlement and recruitment of coral larvae are critical for coral reef recovery following disturbance and are vital for active reef conservation²¹. A recent study observed that the nutritional input from Symbiodiniaceae to Pocillopora damicornis larvae may be minimal compared with adult colonies^22,23. As such, it is vital to understand the effect of climate change on larval nutrient assimilation to evaluate the susceptibility of the symbiosis in this crucial period.

P. damicornis is a dominant vertically transmitting broadcast spawning species in the Indo-Pacific²⁴, which releases larvae in synchrony with the new moon²⁵. P. damicornis larvae commonly settle within 7 days of spawning²⁶, representing a critical window in which environmental stress could impact survivorship. In this study, we used controlled aquaria experiments to investigate the effects of acidification and elevated temperature treatments on nutrient cycling in the larval host-Symbiodiniaceae symbiosis to elucidate the processes shaping symbiotic interactions. Using P. damicornis larvae harboring Durusdinium spp., we combined physiological measurements of larval performance with stable isotope labeling and analysis to explore the role of carbon and nitrogen assimilation in the symbiosis. First, we tested how experimental acidification (1000 µatm) and elevated temperature (32 °C) affect Symbiodiniaceae density and primary production. Second, we examined the effects of these conditions on larval carbon and nitrogen assimilation. We show that the primary production of Durusdinium spp. increased under acidification and elevated temperature. This increase in photosynthate production enhances host carbon assimilation, thereby increasing the nutritional status of coral larvae. Our findings provide insights into the carbon and nitrogen assimilation of larvae under acidification and elevated temperature, shedding light on the nutritional dynamics of host-algal symbiosis under future ocean conditions.

Larval survival, settlement, Symbiodiniaceae density, F _v/F _m, and chlorophyll a content

Neither acidification nor elevated temperature had a negative effect on coral larval survival or settlement (Tab. S1). Under all treatments, mean larval survival rate was above 95.7% and more than 72.92% of larvae settled within 24 h when exposed to the crustose coralline algae (CCA) Hydrolithon cf. reinboldii. Further, the larvae showed no signs of bleaching (Fig. 1). Under all treatments, Symbiodiniaceae communities were dominated by Clade D1 (Supplementary Fig. 3). The Symbiodiniaceae density (normalized by larva) increased slightly under the High pCO₂ and Temperature treatments (F_1,30 = 8.95, P = 0.439; F_4,30 = 0.213, P = 0.929) but remained similar to that of the ambient control conditions in the Combined treatment (F_4,30 = 0.213, P = 0.924). The changes in chlorophyll a content (normalized by larva) across the different treatments mirrored changes in the density of Symbiodiniaceae. The F_v/F_m ratio of Symbiodiniaceae was slightly lower under these conditions, but this decline was not statistically significant. Overall, elevated temperature and pCO₂ showed no significant effect on Symbiodiniaceae density, survivorship, F_v/F_m, and chlorophyll a content (Figs. 1, 2f).

Survivorship (a; n = 20), settlement (b; n = 20), Symbiodiniaceae density (c; n = 10), and Chlorophyll a content (d; n = 10) of the coral Pocillopora damicornis larvae under treatment conditions for five days. Treatment conditions: Control (29 °C and pCO₂ 460 µatm), High pCO₂ (29 °C and 1000 µatm), High Temperature (32 °C), and Combined (32 °C and pCO₂ 1000 µatm). Error bars indicate standard errors of the mean. Colors depict the treatment conditions and all black dots depict individual data points.

Full size image

Metabolic parameters

The productivity of Symbiodiniaceae increased under elevated temperatures and acidification. The P_net in the High pCO₂, High Temperature, and Combined treatments increased by 62%, 72%, and 156%, respectively, compared to the Control (Fig. 2a). Elevated pCO₂ significantly enhanced larval P_net (F_1,12 = 8.95, P = 0.011, Supplementary Table 1) and the same was true for elevated temperature (F_1,12 = 9.92, P = 0.008). The P_g of larvae in the High pCO₂, High Temperature, and Combined treatments was 67%, 57%, and 70% higher, respectively, than that in the Control (Fig. 2c). P_g was significantly affected by elevated pCO₂ (F_1,12 = 7.62, P = 0.017) but not by elevated temperature (F_1,12 = 4.19, P = 0.063). In the High pCO₂ and Combined treatments, R_d was 52% and 31% higher, respectively, than that in the Control, whereas R_d in the High-Temperature treatment remained unchanged (Fig. 2d). R_d was significantly affected by elevated pCO₂ (F_1,12 = 17.39, P = 0.001) but not by elevated temperature (F_1,12 = 1.13, P = 0.309). The P_net/R_d ratio was significantly higher under elevated temperature (F_1,12 = 13.11, P = 0.004) than under ambient temperature, with only the High Temperature and Combined treatments displaying P_net/R_d ratios > 1 (Fig. 2e). R_l was unaffected by the main effects of elevated temperature (F_1,12 = 0.04, P = 0.843) and pCO₂ (F_1,12 = 1.40, P = 0.259; Fig. 2b); however, a significant interaction was observed between pCO₂ and temperature on R_l (F_1,12 = 12.02, P = 0.005), where elevated pCO₂ enhanced R_l but only under ambient temperature. These parameters (chl a, P_g, P_net, R_l) were also shown normalized per Symbiodiniaceae cell (Supplementary Fig. 4).

Net photosynthesis (a; n = 30), light respiration (b; n = 30), gross photosynthesis (c; n = 30), dark respiration (d; n = 30), ratio of net photosynthesis to dark respiration (e; n = 30), and maximum photochemical efficiency F_v/F_m (F; n = 10) of the coral Pocillopora damicornis larvae under treatment conditions for five days. Treatment conditions: Control (29 °C and pCO₂ 460 µatm), High pCO₂ (29 °C and 1000 µatm), High Temperature (32 °C), and Combined (32 °C and pCO₂ 1000 µatm). Different lowercase letters indicate statistically significant differences at P < 0.05. Error bars indicate standard errors of the mean. Colors depict the treatment conditions and all black dots depict individual data points.

Full size image

Carbon to nitrogen ratio

The C:N ratio of the larval holobiont remained unaffected by pCO₂ (F_1,12 = 0.413, P = 0.533) but significantly increased with elevated temperature (F_1,12 = 13.12, P = 0.003), with values ranging from 16.77 ± 1.39 in the High pCO₂ treatment to 22.12 ± 1.70 in the Combined treatment (Fig. 3a).

Carbon:nitrogen ratio (C:N) of (a; n = 30) holobiont of the coral Pocillopora damicornis larvae under treatment conditions for five days. Atom percent excess (APE) enrichment of ¹³C (b; n = 30) and ¹⁵N (c; n = 30) assimilation by Symbiodiniaceae, coral host, and holobiont of the coral P. damicornis larvae during the pulse stage. Shaded gray bars indicate the proportion of algal symbiont cells to holobiont enrichment. Treatment conditions: Control (29 °C and pCO₂ 460 µatm), High pCO₂ (29 °C and 1000 µatm), High Temperature (32 °C), and Combined (32 °C and pCO₂ 1000 µatm). Different lowercase letters indicate statistically significant differences at P < 0.05. Error bars indicate standard errors of the mean. Colors depict the treatment conditions and all black dots depict individual data points.

Full size image

¹³C assimilation within the larval holobiont

All larval holobionts among the treatments fixed a significant amount of ¹³C isotope tracer, with APE ¹³C values increasing from 0.02 (Control) to 0.04 (High pCO₂) (Fig. 3b). The APE ¹³C of larvae in the High pCO₂, High Temperature, and Combined treatments was 100%, 30%, and 56% higher than that in the Control, respectively. Elevated pCO₂ significantly enhanced APE ¹³C (F_1,12 = 17.40, P = 0.014), and there was a significant interaction between temperature and pCO₂ (F_1,12 = 51.13, P = 0.002).

These changes in ¹³C assimilation at the holobiont level further translated into differences in the ¹³C enrichment by the coral host compared to that of its symbionts. Specifically, increases in ¹³C assimilation in the High pCO₂ and High Temperature treatments were largely reflected by increases in ¹³C enrichment in the coral host, while Symbiodiniaceae remained relatively stable. In contrast, increases in ¹³C assimilation in the Combined treatment were largely driven by an increase in the ¹³C enrichment by Symbiodiniaceae, while the coral host remained relatively stable. The host-to-Symbiodiniaceae ¹³C enrichment ratios were 0.59, 4.22, 1.84, and 0.42 in the control, High pCO₂, High Temperature, and Combined treatments, respectively.

¹⁵N assimilation within larval holobionts

All larval holobionts fixed significant amounts of the ¹⁵N isotope tracer in all experimental treatments, APE ¹⁵N increasing from 0.19 (Control) to 0.53 (Combined) (Fig. 3c). The APE ¹⁵N of larvae in the High pCO₂, High Temperature, and Combined treatments was 132%, 55%, and 182% higher than that in the Control, respectively. Larval APE ¹⁵N was significantly enhanced by elevated temperature (F_1,12 = 23.56, P = 0.008). There was a significant interaction between temperature and pCO₂ (F_1,12 = 143.49, P < 0.001).

These changes in ¹⁵N assimilation at the holobiont level also translated into differences in the ¹⁵N enrichment by the coral host in comparison to that by its symbionts. Specifically, High pCO₂ and High Temperature showed an additive effect on ¹⁵N enrichment in Symbiodiniaceae, with both treatments enhancing algal ¹⁵N enrichment, and the Combined treatment showing the highest relative algal ¹⁵N enrichment. In contrast, High Temperature showed no effect on coral host ¹⁵N enrichment, with only the High pCO₂ and Combined treatments showing similar increases in coral host ¹⁵N enrichment. The host-to-Symbiodiniaceae ¹⁵N enrichment ratios were 0.41, 0.73, 0.37, and 0.50 in the Control, High pCO₂, High Temperature, and Combined treatments, respectively.

Nutrient cycling between the coral host and Symbiodiniaceae forms the foundation of their mutualistic relationship² and is key to the remarkable productivity of coral reefs. Previous studies have highlighted the disruption of metabolic homeostasis in symbiosis as a key driver of dysbiosis and bleaching^14,27. In contrast to those works, we observed that Symbiodiniaceae density and primary productivity increased under acidification and elevated temperature, enhancing carbon assimilation in the larval host. The population growth of Symbiodiniaceae could stabilize symbiotic nutrient cycling during acidification and elevated temperature, thereby preventing carbon limitation and starvation in P. damicornis larvae. This notion of enhanced mutualistic nutrient cycling in P. damicornis larvae harboring Durusdinium spp. could help identify mechanisms of acclimation and adaptation to climate change.

In adult corals, the symbiotic relationship between coral hosts and Symbiodiniaceae is passively regulated by metabolic feedback between symbiotic carbon and nitrogen cycling^14,15. Reduced translocation of photosynthates by Symbiodiniaceae and host energy starvation are thus common symptoms of a destabilization of the symbiosis, e.g. during heat stress^9,10. In this light, our results clearly indicate that the treatment levels of acidification (1000 µatm) and elevated temperatures (32 °C) used herein did not impair mutualistic interactions in the symbiosis and had no negative effect on larval physiology and performance. This notion is further supported by a recent study showing that larvae of the coral Montipora capitata maintained high rates of algal photosynthate release and showed increased host ammonium assimilation when exposed to elevated temperature²⁸.

Our results, therefore, show that coral larvae may be able to tolerate acidification and elevated temperature. Indeed, previous studies showed that the temperature threshold of the larvae appears to be between 32 °C and 33 °C, as indicated by abrupt declines in Symbiodiniaceae density and downregulation of gene and protein expression related to photosynthesis at 33 °C^29,30. One possibility for this high tolerance is that the larvae primarily contain Durusdinium spp. (Supplementary Fig. 3), which are vertically transmitted from parental P. damicornis³¹. Multiple studies indicated that Durusdinium spp. enhance the heat tolerance and reduce the bleaching susceptibility of adult corals^32,33. Adult corals hosting fast-growing Durusdinium maintain higher rates of photosynthate translocation during heat stress and show reduced bleaching susceptibility^34,35. Our study found that larval P. damicornis hosts received more carbon from their endosymbiotic Durusdinium spp. under acidification and elevated temperature. Similar increases in coral host-Symbiodiniaceae carbon assimilation during heat stress have been reported for other species, suggesting that heat-exposed colonies with Durusdinium trenchii experienced less physiological stress than conspecifics with Cladocopium spp. while maintaining high carbon assimilation and nutrient transfer to the host³⁶. Therefore, fast Symbiodiniaceae population growth appears to support host nutrition and reduce bleaching susceptibility in corals.

Warming increase the metabolic energy demand of the coral host resulting in enhanced nitrogen availability for algal symbionts. In line with this, we observed increased light and dark respiration rates and associated increase in symbiont densities under High pCO₂ and Temperature indicating that the initial metabolic impacts of heat stress on the symbiosis may be similar in larval and adult coral holobionts. However, in coral larvae this increased metabolic energy demand in the host was efficiently compensated by enhanced algal photosynthesis and photosynthate translocation. This suggests that rapid algal proliferation was sufficient to stabilize mutualistic nutrient cycling in the symbiosis. Increases in C:N ratios and enhanced ¹³C and ¹⁵N assimilation further suggest that the combination of increased algal symbiont density and maintained mutualistic nutrient cycling provide a net improvement to the nutritional status of the host under High pCO₂ and Temperature conditions.

We propose that the key to the stability of symbiotic interactions in coral larvae may reflect the unique properties of this stage of the life cycle. In contrast to recruits (juveniles) and adult corals, larvae do not build a calcium carbonate skeleton³⁷. As calcification is a highly energy-demanding process³⁸, the lack of this trait may provide coral larvae with a nutritional advantage. Coupled with the large lipid reserves of coral larvae (in addition to heterotrophic feeding through the oral pore)^21,22,39, the net availability of organic carbon in the larval metabolism may thus exceed that of adult corals. This would enable coral larvae to support higher metabolic energy demands and maintain net ammonium assimilation to limit algal growth during high-temperature conditions. As a result, the symbiotic relationship between coral larvae and Symbiodiniaceae might be more resistant to metabolic imbalances than that of adult corals. Future experiments verifying the here reported patterns in other coral species locations, environmental conditions, and other stages of the coral life cycle could thus help develop a holistic understanding of the mechanisms underlying coral susceptibility to future ocean conditions.

In summary, the results of our study demonstrate that enhanced mutualistic nutrient cycling may contribute to the survival of symbiotic larvae under future climate change scenarios. As sea surface temperatures continue to increase, it is vital to understand how symbiotic relationships are affected in early life stages and to determine metabolic tipping points impacting successful coral development. Our model highlights the importance of metabolic feedback and carbon availability in maintaining the resilience of coral larvae to climate change. Understanding these dynamics will be crucial for developing strategies to protect and conserve coral reefs in the face of ongoing environmental changes.

Larval collection

In September 2018, five P. damicornis colonies were collected from a depth of 3 m at the Luhuitou fringing reef (18°12′N, 109°28′E) in Sanya, Hainan Island, China. These colonies were immediately transported to the Tropical Marine Biological Research Station in Hainan (Supplementary Fig. 1), where they were placed in individual 20-L containers with a flow-through seawater system maintained at an ambient temperature of 28.6 ± 0.2 °C within an 800-L tank. The tanks were situated in a semi-enclosed outdoor area, exposed to natural light and partially shaded to mimic the conditions of their original habitat. The seawater used in the flow-through system was sand-filtered and sourced offshore from the fringing reef. All coral experiments were conducted in accordance with the guidelines and approved by the Tropical Marine Biological Research Station in Hainan (2018-TBS-002), with collection also authorized by the station’s administration under the same permit. We have complied with all relevant ethical regulations for animal use.

Detailed procedures for larval collection have been previously described^29,30. Briefly, each container was equipped with a 180 µm mesh on the outflow pipe to facilitate the collection of coral larvae. Larvae were collected three days later, on September 15, 2018 (peak release day), and pooled and mixed for subsequent experiments.

Experimental setup

The experimental treatments (Supplementary Fig. 2) consisted of four conditions for the larvae, combing two different pCO₂ levels and two different temperature conditions as follows: Control (~29 °C and pCO₂ ~ 480 µatm), High pCO₂ (~29 °C and ~1000 µatm), High Temperature (~32 °C and ~480 µatm), and Combined (~32 °C and ~1000 µatm). The ambient temperature and pCO₂ conditions represented the average summer temperature and pCO₂ level at the coral collection site. The elevated temperature and pCO₂ conditions were selected to mimic the predicted oceanic temperature and pCO₂ level for the year 2100 under the representative concentration pathway 8.5 global warming scenario⁴⁰.

Larvae were cultured in 4-L plastic beakers under four different experimental conditions, with four replicate tanks per treatment. The beakers were placed in 20 L freshwater bath tanks maintained at the respective experimental temperatures (29 °C or 32 °C). The seawater was filtered through a 0.5 µm mesh and UV-sterilized prior to use. Illumination was provided by T5 fluorescent lamps (two aqua blue coral tubes and two actinic-blue tubes; Giesemann, Aquaristic, Nettetal, Germany) to create a mean light intensity of 150 µmol photons m⁻² s⁻¹, approximating the light intensity at the collection depth (3 m) of the parent colonies in September (147 ± 101 µmol· photons m⁻² s⁻¹).

Coral larvae were exposed to the treatment conditions for five days, reflecting the tendency of larvae to settle shortly after release, with the larval stage typically completed within a week^25,41. Ambient and elevated pCO₂ conditions were maintained using a CO₂ enricher system (CE100B; Ruihua, Wuhan, China), which controlled the bubbling of air and CO₂ into each tank based on the target pH. Ambient and elevated temperatures were maintained using a digital temperature regulator (TC-05B, SiEval, Guangzhou, China) and a 50 W heater (MX-1019, Weipro, Zhongshan, China) in each tank. Fresh water was mixed using a submerged pump (PH-1100, Weipro, Zhongshan, China). Temperatures were monitored at 15-min intervals using a HOBO Pendant temperature logger (Onset, Bourne, MA, USA) throughout the experiment.

A 30% seawater change was performed daily with temperature˗ and pH˗equilibrated seawater for each treatment. Salinity was maintained at approximately 32 ppt through daily top-offs with deionized water. Seawater samples (50 mL) were collected daily from each beaker and preserved with saturated HgCl₂ (50 μL) for the determination of total alkalinity (TA). Briefly, a 25-g seawater sample was titrated against a 0.1 mol/L HCl solution using bromocresol green indicator at pH 4.0–4.2. TA was determined by measuring the absorbance at 444 nm and 630 nm using a spectrophotometer (UV-2700; Shimadzu, Kyoto, Japan)⁴². Temperature, salinity, and pH were continuously monitored and adjusted twice daily using a pH meter (SevenGo meter; Mettler Toledo, Zürich, Switzerland). The pH was measured using the total hydrogen ion scale (pH_T). TA and pH were used to calculate carbonate parameters [i.e., dissolved inorganic carbon (DIC), aragonite saturation state (Ω_Arag) and pCO₂] using CO2SYS⁴³.

The mean temperatures for each treatment were 29.25 ± 0.01 [Control, mean ± standard error (M ± SE)], 28.90 ± 0.01 (High pCO₂), 32.24 ± 0.01 (High T), and 31.86 ± 0.01 °C (Combined) (Table 1). The temperatures did not differ between tanks within each treatment but differed between treatments (df = 8645, P < 0.001, paired samples t-test). The mean pCO₂ for each treatment was 465 ± 18 (Control), 1012 ± 42 (High pCO₂), 437 ± 16 (High T), and 966 ± 35 µatm (Combined). The pCO₂ levels did not differ between duplicate tanks within each treatment group but differed between each CO₂ treatment at each temperature (df = 23, P < 0.001, paired samples t-test).

Measurement of larval metabolism

Larval net photosynthesis in the light and respiration in the dark were derived from rates of oxygen production or consumption measured in the light and darkness, respectively. The measuring system consisted of a 1.75-mL custom-made glass vial, a water bath, a magnetic stirrer, an oxygen sensor spot (SP-PSt3-NAU, PreSens, Regensburg, Germany), and a fluorescent quenching sensor (OXY-4 mini; PreSens, Regensburg, Germany) allowing for continuous monitoring of oxygen concentration with temperature correction. Oxygen readings were calibrated by two-point calibration according to the instruction manual. Net photosynthesis (P_net), light-enhanced respiration (R_l), and dark respiration (R_d) rates were measured once by recording oxygen concentrations every 5 s over a period of 10 min on day 5 of the experiment. The temperature and light conditions for all measurements matched those of the corresponding experimental treatments.

Thirty larvae were randomly collected and transferred from their respective experimental beaker to a vial containing a miniature stir bar (3 × 5 mm) and seawater from the corresponding experimental beaker. P_net was measured for 10 consecutive min under the same light intensity as the experimental system (150 µmol photons m⁻² s⁻¹), followed by a 10-min measurement of R_l under dark conditions (0 µmol photons m⁻² s⁻¹). R_d was measured using the same method after acclimating a different batch of larvae in the dark for 2 h. Oxygen levels were sampled every 5 s during the 10-min measurement periods to detect their increases or decreases accurately.

P_net, R_l, and R_d were calculated using least squares linear regression of the O₂ concentration plotted against time and expressed in nmol O₂ larva⁻¹ min⁻¹. Gross photosynthesis (P_g) was calculated by adding P_net and R_l. The P_net/R_d ratio was calculated to assess the potential for autotrophy⁴. A P_net/R_d ratio greater than 1 was considered sufficient to support daily metabolic energy demands through organic carbon from photosynthesis. In contrast, a ratio less than 1 indicated insufficient support for daily metabolic energy demands through organic carbon from photosynthesis, suggesting a loss of endogenous reserves.

Measurement of photosynthetic efficiency, survivorship, and Symbiodiniaceae density

Following the R_d measurements, the maximum quantum yield (F_v/F_m ratio) of dark-adapted Symbiodiniaceae (ten larvae per replicate) was measured using a Diving Pulse Amplitude Modulated fluorometer (Walz, Effeltrich, Germany). Initial fluorescence (F₀) was determined by applying a weak modulated red light pulse. A 0.8-s saturating pulse of actinic light was then applied to measure the maximum fluorescence level (F_m). Both the measuring light and gain of PAM settings were adjusted to “7” to optimize the fluorescence signal. The F_v/F_m ratio was calculated as (F_m-F₀)/F_m, as previously described⁴⁴. Following PAM measurements, the larvae were flash-frozen in liquid nitrogen for subsequent Symbiodiniaceae density and chlorophyll measurement.

For survivorship assessment, larvae were placed in transparent tubes with both ends covered using a 180-µm mesh to ensure sufficient mixing with the surrounding seawater. One 100 mL tube containing 20 larvae was examined per beaker to assess survivorship, with daily observations. Settlement assays were conducted in 5.5 cm diameter petri dishes and followed by Jiang et al.⁴⁵. Briefly, Hydrolithon cf. reinboldii, one of the most abundant CCA species, was collected at depths of 2−3 m and cut into uniformly sized chips, five days before the settlement experiment. Each dish contained 15 mL of seawater and a CCA chip. Twenty actively swimming larvae were introduced into each dish, which was then floated and partially (80%) submerged in seawater to ensure temperature control. Six replicate dishes were used for each treatment. Larvae were allowed to settle for 24 h, after which settlement success was assessed under a dissecting microscope following the criteria of Heyward and Negri⁴⁶.

To determine the density of Symbiodiniaceae, larvae were ground using a handheld homogenizer with autoclavable plastic pestles. The homogenate was then centrifuged twice at 615 × g for 10 min at 4 °C to precipitate most Symbiodiniaceae. The resultant supernatant was transferred to a new tube and centrifuged twice more at 3850 × g to precipitate residual algae. The pellet was resuspended in 0.45-µm filtered and UV-sterilized seawater (FSW). The number of Symbiodiniaceae cells was determined using a hemocytometer with six replicate counts per sample, and Symbiodiniaceae density was expressed as the number of cells per larva.

For photopigment quantification, pelleted cells were lysed in 90% methanol and incubated in the dark at −20 °C for 12 h. Absorbance at wavelengths of 750, 664, and 630 nm was measured using a spectrophotometer (Varioskan LUX; Thermo Fisher Scientific, Waltham, MA, USA). Chlorophyll a content was calculated using the following Ritchie et al.⁴⁷:

Chl a (µg ml⁻¹) = 13.6849 × (OD₆₆₄ – OD₇₅₀) – 3.4551 × (OD₆₃₀ – OD₇₅₀).

Identification of Symbiodiniaceae species

To identify the Symbiodiniaceae species in larvae, we performed high-throughput sequencing of the internal transcribed spacer 2 (ITS 2) region of ribosomal RNA gene amplicons, following the protocols described by Sun et al.⁴⁸. Total DNA was extracted as described previously. The purity of the DNA was measured using a NanoDrop spectrophotometer (Thermo fisher scientific, United States). The DNA samples were amplified using polymerase chain reaction (PCR) primers employed by LaJeunesse and Trench⁴⁹. The ITS2 amplicon was approximately 320 base pairs long.

The primer sequences were:

Forward primer (ITSintfor2): 5′ - AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTN₆GAA TTGCAGAACTCCGTG-3′;

reverse primer (ITS2-reverse) 5′ -AGATCGGAAGAGCACACGTCTGAACTCCAGTCACN6 GGGATCCATATGCTTAAGTTCAGCGGGT-3′.

These primers included illumine library adapters (underlined) and a barcode (N). PCR was performed with 12.5 µL of PCR reagent (Bio-Rad, United States), 0.1 µM primer, 50 ng of DNA, and DNase-free water to make a total volume of 25 µL. The PCR conditions were initial denaturation at 94 °C for 3 min, followed by 34 cycles at 98 °C for 10 s, 51 °C for 30 s, 68 °C for 30 s, and a final extension at 68 °C for 5 min.

The PCR products were validated using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, United States) and quantified with a NanoDrop spectrophotometer. All qualified amplification products were pooled in equal amounts and sequenced on an Illumina NextSeq 2000 instrument (Illumina, San Diego, CA, United States) using a 300 × 2 paired-end configuration. Processing and analysis of the ITS2 sequences were conducted following our previously described protocol⁵⁰. The raw sequencing data have been submitted to the National center for Biotechnology information (NCBI) sequence read archive under the accession number PRJNA 1123470.

Stable isotope tracer experiment and analysis

To assess nutrient assimilation, transfer and partitioning in the symbiosis, five days after exposure to the experimental conditions, 40 larvae were randomly sampled from two beakers from each treatment and incubated in a 250-mL PC bottle (Nalgene, Rochester, NY, USA) containing experimental seawater with 100 μM NaH¹³CO₃ (99% ¹³C; Sigma-Aldrich, St. Louis, MO, USA) and 1 μM ¹⁵NH₄Cl (98% ¹⁵N, Sigma-Aldrich, St. Louis, MO, USA). Ammonium was chosen for the isotope labeling as this can be assimilated by both symbiotic partners and is the preferred source of nitrogen for Symbiodiniaceae^51,52,53.

Due to the limited number of larvae, two replicate ¹³C and ¹⁵N pulse incubations were performed per treatment. The bottles were incubated in their respective treatment tanks for 3 h under light (pulse period, 13:00–16:00). At the end of the pulse period, 40 larvae were rinsed three times with non-labeled FSW, preserved in liquid nitrogen, and stored at −80 °C for subsequent assessment of stable isotope values.

Stable isotope analysis

Stable isotope analysis was performed on entire larval holobionts, host tissue fractions, and Symbiodiniaceae fractions. For measurement of isotopic values in the larval holobiont, 10 larvae were thawed and filtered through a pre-combusted (450 °C for 4 h) Whatman GF/F filter (0.7 µm). To assess the relative contribution of the coral host and Symbiodiniaceae to holobiont stable isotope composition, 30 larvae were thawed and homogenized in FSW in a 1.5-mL centrifuge tube using a handheld homogenizer and an autoclaved pestle for 60 s. The pestle was washed thrice with FSW to ensure no host tissues or Symbiodiniaceae remained on it. The tissue slurry was homogenized and centrifuged at 615 × g for 10 min at 4 °C to precipitate most of the Symbiodiniaceae. The resultant supernatant was then transferred to a new tube and centrifuged twice more at 3850 × g to precipitate residual Symbiodiniaceae, effectively separating Symbiodiniaceae from coral host tissues⁵⁴. Subsequently, the Symbiodiniaceae fraction was decanted and filtered through a pre-combusted Whatman GF/F filter. The filter membranes for the coral host and Symbiodiniaceae fractions were collected in tin cups, freeze-dried for 24 h, rinsed with 1 N HCl to remove inorganic carbonates, and dried in an oven to achieve constant weight (60 °C for 24 h). The concentrations and isotopic values of the biological samples were analyzed via isotope ratio mass spectrometry using a Flash HT 2000 (Thermo Fisher Scientific) coupled with a Delta V Plus IRMS (Thermo Fisher Scientific) at Xiamen University. International reference materials (USGS-40 and -41) were measured every eight samples to monitor drift and ensure the accuracy of the measurements. The analysis precision was ±0.05% for carbon (C) and nitrogen (N). The reproducibility of the δ¹³C and δ¹⁵N measurements was <0.2‰.

¹³C and ¹⁵N enrichment in the larval holobiont, coral host, and Symbiodiniaceae fractions were evaluated to assess coral holobiont assimilation of carbon and nitrogen, the respective uptake of either symbiotic partner, local translocation of compounds from Symbiodiniaceae, and retention of compounds in Symbiodiniaceae after translocation to the coral host. Enriched values were reported as atom percent excess (APE) ¹³C and ¹⁵N.

For this, δ¹³C and δ¹⁵N values were determined using the following equations:

where R_sample = (¹³C/¹²C) and R_standard = 0.0112372 for carbon isotopes, and R_sample = (¹⁵N/¹⁴N) and R_air = 0.003676 for nitrogen isotopes.

Atom% ¹³C and ¹⁵N in each fraction during the pulse-chase incubation were calculated using the following equations:

APE ¹³C and ¹⁵N for each sample were calculated using the following equations:

Where atom% ¹³C_T and ¹³C_T0 are the ¹³C enrichment at the start and end of the incubation, respectively. Atom% ¹⁵N_T and ¹⁵N_T0 are the ¹⁵N enrichment at the start and end of the incubation, respectively.

Seawater column chemistry

Dissolved nutrients at the Luhuitou fringing reef were measured using a SEAL AA3 autoanalyzer (Seal Analytical, Inc., Mequon, WI, USA) following standard methods⁵⁵. The detection limits were 0.05 µM for NO₃⁻, 0.03 µM for NO₂⁻, 0.12 µM for PO₄³⁻, and 0.10 µM for NH₄⁺. Dissolved inorganic nitrogen (DIN) was calculated as [NOx + NH₄⁺], and dissolved inorganic phosphorus (DIP) was calculated as [PO₄³⁻]. DIN ranged from 1.57 µM to 3.07 µM and DIP ranged from 0.55 µM to 1.97 µM.

Statistics and reproducibility

Differences in coral larval physiology, including P_net, R_l, P_g, R_d, P_net/R_d, F_v/F_m, Symbiodiniaceae density, survivorship, chlorophyll a, C:N ratio, APE ¹³C, and ¹⁵N were determined using a two-way analysis of variance (ANOVA) followed by Tukey’s honest significant difference (HSD) post hoc test. Differences in seawater parameters between treatments were assessed using Student’s t test. These analyses were conducted using SPSS version 22.0 (IBM, Armonk, NY, USA) and Microsoft Excel (Microsoft Corporation, Redmont, WA, USA).