Acclimatization of a coral-dinoflagellate mutualism at a CO2 vent

Abstract

Ocean acidification caused by shifts in ocean carbonate chemistry resulting from increased atmospheric CO₂ concentrations is threatening many calcifying organisms, including corals. Here we assessed autotrophy vs heterotrophy shifts in the Mediterranean zooxanthellate scleractinian coral Balanophyllia europaea acclimatized to low pH/high pCO₂ conditions at a CO₂ vent off Panarea Island (Italy). Dinoflagellate endosymbiont densities were higher at lowest pH Sites where changes in the distribution of distinct haplotypes of a host-specific symbiont species, Philozoon balanophyllum, were observed. An increase in symbiont C/N ratios was observed at low pH, likely as a result of increased C fixation by higher symbiont cell densities. δ¹³C values of the symbionts and host tissue reached similar values at the lowest pH Site, suggesting an increased influence of autotrophy with increasing acidification. Host tissue δ¹⁵N values of 0‰ strongly suggest that diazotroph N₂ fixation is occurring within the coral tissue/mucus at the low pH Sites, likely explaining the decrease in host tissue C/N ratios with acidification. Overall, our findings show an acclimatization of this coral-dinoflagellate mutualism through trophic adjustment and symbiont haplotype differences with increasing acidification, highlighting that some corals are capable of acclimatizing to ocean acidification predicted under end-of-century scenarios.

Fig. 1. Ranges of measured pH_TS and Ω_arag values showing consistent decreases from Site 1 to Site 3.

The boxes indicate the 25^th and 75^th percentiles and the line within the boxes mark the medians. Whisker length is equal to 1.5 × interquartile range (IQR). Circles represent outliers. Different letters indicate statistical differences (p < 0.05; number of observations: pH_TS = 185-198 per Site; Ω_arag = 184–195 per Site).

Fig. 2. Boxplots of fluorescence parameters obtained by PAM measurements at the three Sites and at three time intervals.

a Effective quantum yield (ΔF/F_m´), b minimum fluorescence (F), and c maximum fluorescence (F_m´). For each parameter, significant differences (p < 0.01) between time intervals are indicated by different colors (significantly higher mean value is ranked as white->light gray->dark grey). The boxes indicate the 25^th and 75^th percentiles and the line within the boxes mark the medians. Whisker length is equal to 1.5 × interquartile range (IQR). Circles represent outliers. Different letters indicate statistical differences (p < 0.05; number of specimens measured is reported in Supplementary Table 3).

Fig. 3. Symbiont cells in B. europaea polyps sampled along the Panarea pH gradient.

a–c Histological transverse sections of representative polyps in Sites 1-3 (x 4 magnification). d–f Symbiont cells within the endoderm of the mesentery highlighted by x40 magnification. z: zooxanthellae; ec: ectoderm; m: mesoglea; en: endoderm. g Symbiont cell density and h chlorophyll-a (chl-a) concentration normalized over cell count or i polyp area. The boxes indicate the 25^th and 75^th percentiles and the line within the boxes mark the medians. Whisker length is equal to 1.5 × interquartile range (IQR). Circles represent outliers. Different letters indicate statistical differences (p < 0.05; number of corals = 4 per Site).

Fig. 4. δ¹³C, δ¹⁵N, C/N ratios, and proportion of carbon in the host coming from the symbionts in B. europaea tissue from Sites 1 (control), 2 (intermediate pH) and 3 (low pH).

The proportion of carbon in the host supplied by the symbionts was calculated using the formula: ${\delta }^{13}{\mathrm{C}}_{\mathrm{tissue}}=\left({\delta }^{13}{\mathrm{C}}_{\mathrm{symbiont}}\right)\mathrm{x}+\left(1-\mathrm{x}\right)\left({\delta }^{13}{\mathrm{C}}_{\mathrm{zooplankton}/\mathrm{POC}}\right)$. We assume δ¹³C_{zooplankton/POC} = −22‰. The boxes indicate the 25^th and 75^th percentiles and the line within the boxes mark the medians. Whisker length is equal to 1.5 × interquartile range (IQR). Different letters indicate statistical differences (p < 0.05) between Sites (number of corals = 4 per Site).

Fig. 5. Philozoon balanophyllum phylogeny.

Philozoon balanophyllum halpotypes in B. europaea specimens from Sites 1-3 along the Panarea pH gradient and from around the Tyrrhenian and Adriatic Seas. Haplotypes of Philozoon actinarum, the symbiont of the sea anemone Anemonia viridis, common to the Mediterranean, are provided as the outgroup. (Photograph by Francesco Sesso).

Fig. 6. Conceptual scheme summarizing the effects of life-long physiological acclimatization to low pH/high pCO₂ conditions in B. europaea at the Panarea CO₂ vent.

Under low pH conditions, coral population density decreases, net calcification is depressed, while linear extension rate is maintained constant, allowing the coral to reach critical size at sexual maturity and reproduce. Moreover, an overall rearrangement of the coral holobiont is documented by: (i) an increase in symbiont cell density, triggered by thickening of the coral tissue and the establishment of novel dinoflagellate haplotypes possibly better adapted to lower pH conditions, and (ii) an increase, within the coral tissue/mucus in microbial communities capable of dinitrogen fixation as well as N storage and mobilization. Variations displayed by corals living at average pH 7.6 compared to corals at average pH 8.0 are shown with grey symbols and are listed in the following order: skeleton (from micro to macro), symbionts (from micro to macro), trophic strategy, reproduction, tissue/mucus microbial community. Darker and lighter shades of green in the coral sketches represent higher and lower symbiont cell density. The color scale bar highlighting the seawater pH change across the gradient does not match the color scale of pH test strips. Image was assembled using Adobe Photoshop CC (19.1.6).