Coral micro- and macro-morphological skeletal properties in response to life-long acclimatization at CO2 vents in Papua New Guinea

This study investigates the effects of long-term exposure to OA on skeletal parameters of four tropical zooxanthellate corals naturally living at CO2 seeps and adjacent control sites from two locations (Dobu and Upa Upasina) in the Papua New Guinea underwater volcanic vent system. The seeps are characterized by seawater pH values ranging from 8.0 to about 7.7. The skeletal porosity of Galaxea fascicularis, Acropora millepora, massive Porites, and Pocillopora damicornis was higher (up to ~ 40%, depending on the species) at the seep sites compared to the control sites. Pocillopora damicornis also showed a decrease of micro-density (up to ~ 7%). Thus, further investigations conducted on this species showed an increase of the volume fraction of the larger pores (up to ~ 7%), a decrease of the intraskeletal organic matrix content (up to ~ 15%), and an increase of the intraskeletal water content (up to ~ 59%) at the seep sites. The organic matrix related strain and crystallite size did not vary between seep and control sites. This multi-species study showed a common phenotypic response among different zooxanthellate corals subjected to the same environmental pressures, leading to the development of a more porous skeletal phenotype under OA.

www.nature.com/scientificreports/ to have profound implications for marine calcifiers, as carbonate ions are essential for biotic calcification 3 . Coral responses to OA may be affected by several factors including colony morphology, size, skeletal mineralogy and structure, tissue thickness, symbiont types, and/or the mechanisms of nutrient acquisition 4 . Moreover, the discrepancy among responses could derive from different experimental designs and analytical methods (e.g., addition of acid vs CO 2 bubbling to mimic OA), co-limiting environmental conditions (e.g., temperature, light intensity, flow, feeding, etc.), and exposure times (days to months or even life times) 5 .
To date, most studies both under controlled conditions and under natural conditions in the field (e.g., CO 2 vents), support predictions of decreased rates of calcification and increased rates of dissolution and bioerosion as seawater pH decreases 6 . However, studies conducted using skeletal cores have shown that coral calcification rates have not declined at a constant rate as ocean pH decreased and temperatures warmed throughout the twentieth century. On the contrary, at some locations, calcification rates have remained stable and in others they have even increased over this time period [7][8][9] . Even where declines in calcification have occurred, many other factors such as ocean warming, sea level rise, changes in surface ocean productivity, as well as many localized anthropogenic disturbances co-occur with OA. These additional factors also influence coral growth and could obscure our ability to attribute changes in coral calcification solely to OA 10 .
Most of the available knowledge about OA effects on marine organisms derives from short-term laboratory or mesocosm experiments on isolated organisms 11 , which can substantially underestimate full organism acclimatization 12 . In fact, taxa that appear unaffected by high CO 2 under controlled conditions may be: (1) vulnerable in the long-term 13 , (2) affected during life stages that were not considered during the experiment 14 , or (3) be indirectly affected by OA-driven ecological changes (e.g., food webs, competition, diseases and/or community structures, habitat properties such as microbial surface biofilms) 15 . Likewise, other taxa that respond negatively to OA under controlled conditions may be capable of acclimatizing in the longer term. Thus, field experiments, where organisms are naturally exposed to OA for their entire life, as found around submarine CO 2 vents, could provide important new insights. However, vent systems are not perfect predictors of future ocean ecology owing to temporal variability in pH, spatial proximity of populations unaffected by acidification, and the unknown effects of other changing parameters (e.g., temperature, currents) 16 . Nonetheless, vents acidify sea water on sufficiently large spatial and temporal scales to integrate ecosystem processes such as reproduction, competition and predation 17 . Field-based studies conducted at volcanic CO 2 seeps in Italy [17][18][19] , Japan 20 , Mexico 21 , and Papua New Guinea (PNG) 15 provide a unique opportunity to investigate long-term effects of OA on marine ecosystems that have been naturally exposed to chronic low pH and concomitant altered carbonate chemistry parameters for years/decades. These studies have already demonstrated substantial changes in community structure and functional biodiversity 22 of benthic species, as well as an array of responses to OA spanning from sharp decrease to no effect on calcification rate 23 .
Studies conducted on corals at volcanic CO 2 vents in Papua New Guinea (PNG) have supported the mixed effects observed in laboratory experiments 15,24 . Hard coral cover is similar at acidified and control sites (33% versus 31%). However, the cover of massive Porites is doubled under OA, whereas the cover of more structurally complex corals is reduced by one third 24 . Some species are significantly less common or even absent under OA. For instance, while the coverage of Pocillopora damicornis decreases by 43% in acidified sites, in situ growth measurements have found small differences in linear extension rate 15 , but large differences in recruitment success 25 . Population reductions in situ, combined with observations of negative physiological impacts, including declines in calcification under OA, strongly suggest that low pH imposes selection pressure on less resilient taxa within the PNG system 23 .
The aim of this study was to assess the effects of long-term exposure to OA on the skeletal parameters (microdensity, porosity, bulk density) of four tropical zooxanthellate coral species Galaxea fascicularis (Linnaeus, 1767), Acropora millepora (Ehrenberg, 1834), massive Porites Link, 1807, and P. damicornis (Linnaeus, 1758), living at CO 2 vents and adjacent control sites in Milne Bay Province, PNG 15 . Additional macroscale and microscale skeletal analyses, namely Time-Domain Nuclear Magnetic Resonance (TD-NMR), Thermogravimetric Analysis (TGA), and synchrotron high-resolution powder X-ray diffraction (HRPXRD) analyses were performed on P. damicornis, the only species displaying differences in micro-density at the seep sites compared to control.

Materials and methods
Study sites and coral sampling. The study was conducted at two shallow-water (1-5 m) volcanic CO 2 seeps at ambient temperature and adjacent control sites at Milne Bay Province, PNG, namely Dobu and Upa Upasina (Fig. 1). Almost pure CO 2 (~ 99%) has been streaming from the seabed for an unknown period of time (confirmed for approximately 70 years, but likely much longer) 15 , resulting in localized acidified conditions. Environmental parameters (measured: pH, dissolved inorganic carbon, total alkalinity, salinity, and temperature; calculated with CO2SYS software: pCO 2 and aragonite saturation state) were obtained across a 4-year period (2010-2013) at 1-5 m depth in both control and seep sites 15,24 . Two-cm coral fragments, which corresponds to approximately a 1.5-year growth increment in all the investigated species 15,26,27 , were collected at 1-5 m depth from adult colonies of P. damicornis, G. fascicularis, A. millepora, and massive Porites at control and seep sites in August 2010 (N = 6-15 fragments per site, each fragment from a different colony) 28 . Tissue from the coral fragments was totally removed using established protocols applied in previous studies on corals which include immersing the samples in a solution of 10% commercial bleach for 3 days and drying them for 24 h at a maximum temperature of 40°C [29][30][31][32] . Dried fragments were kept in codified Eppendorf tubes prior to skeletal measurements.
Skeletal porosity, bulk density, and micro-density determination. The  www.nature.com/scientificreports/ obtained as follows. After determining the dry mass, the fragments were placed inside a drying chamber connected to a vacuum pump to evacuate air and water from the pores, a necessary step in order to allow effective saturation of the samples in the followig phase. After 3 h, distilled water was gently introduced to fully saturate the samples which were then weighed in air to determine the saturated weight. Buoyant weight was then measured with a hydrostatic balance (Ohaus Explorer Pro balance ± 0.0001 g) equipped with a density determination kit and used to calculate porosity, bulk density, and micro-density by means of standard calculations (details in Supplementary Methods) 29 .
Time-domain nuclear magnetic resonance for pore size distribution determination. To investigate the distribution of pore-size classes, through the analysis of the NMR transverse relaxation time T 2 distributions from control and seep site (details in Supplementary Methods), coral fragments still fully saturated with water, were placed on a wet paper to dry the excess of water on their surface. Then, every fragment was put inside a glass tube, sealed and immediately inserted into the magnetic field to be subjected to TD-NMR measurement. A home-built relaxometer based on an electromagnet JEOL C-60 (magnetic field B 0 = 0.5 Tesla) with a radiofrequency coil ≈ 8 mm in diameter, and equipped with a Spinmaster portable console (Stelar, Mede, Pavia, Italy) was used. The Carr-Purcell-Meiboom-Gill (CPMG) sequence with 200 µs echo time was used to acquire for each specimen the transverse relaxation curve. The measured multi-exponential relaxation curves, affected by unavoidable measurement noise, were transformed into distributions of the transverse relaxation time T 2 by the algorithm UPEN (Uniform-Penalty inversion algorithm) 33 , implemented in the UpenWin software 34 . The ratio between the signal under a particular portion of the T 2 distribution and the total acquired signal will correspond to the ratio of the volume of the pores with a particular pore size to the total pore volume. TD-NMR measurements were performed on 2 fragments per site per species for A. millepora, massive Porites, and G. fascicularis and on all available fragments of P. damicornis (a total of 60 fragments). The T 2 distributions showed a cut-off at 3 ms allowing to divide the pores containing water into two classes, distinguishing the smaller pores (smaller volumes, estimated pore sizes < 1 µm) from the remaining larger ones (larger volumes, estimated pore sizes > 1 µm). For the sake of simplicity, the two classes were named micro-scale and macro-scale pores 13,35 .
Thermogravimetric analysis for organic matrix content determination. To determine the intraskeletal organic matrix and water content, thermal gravimetric measurements were performed using a TA Instruments thermobalance model SDT-Q600 with 0.1 µg of balance sensitivity. Powdered subsamples (5 to 10 mg), held in alumina pans, were heated under a linear gradient from ambient (ca. 20 °C) up to 600 °C with an isotherm at 120 °C for 5 min to remove the adsorbed water; heating rate: 10 °C/min under an N 2 atmosphere,  www.nature.com/scientificreports/ with flux fixed to 100 ml/min. Two main weight loss regimes were identified: a first one in a range around 125-250° C (related to the loss of structured water molecules) followed by another thermal region between 250 and 470 °C (generally associated with organic matrix pyrolysis) 36 . A total of 32 fragments from the control and seep sites at Dobu (10 for each site) and at Upa Upasina (6 for each site) were analyzed.
Synchrotron high-resolution X-ray powder diffraction. To determine crystallite parameters, coral fragments were measured at the ID22 beamline of the European Synchrotron Radiation Facility (Grenoble, France) using a wavelength of 0.4 Å (details in Supplementary Methods). Subsamples of the fine powder were loaded into borosilicate glass capillaries of 0.7-1 mm in diameter and measured at room temperature, and again after ex-situ heating at 300 °C for 2 h, to remove the organic matrix effects on the strain 37 . Rietveld refinement was used to calculate the unit-cell parameters from the diffraction pattern profiles. The line profile analysis was applied to a specific diffraction peak to obtain the coherence length (nm) along various crystallographic directions, which was achieved by fitting the profile to a Voigt function and deconvoluting the Lorentzian and Gaussian widths. Analyses were conducted on fragments of P. damicornis from Upa Upasina in the control (N = 3) and seep (N = 3) site.

Results
Environmental parameters. The values of the environmental parameters collected at control and seep sites in Dobu and Upa Upasina over a 3-year period are summarized in Fig. S1. Briefly, pH and pCO 2 across both seep sites averaged 7.72 ± 0.23 (SD) and 1133 ± 1161 µatm, while at the control sites it averaged 7.93 ± 0.10 and 518 ± 250 µatm, respectively. The complete dataset of environmental parameters (Fig. S1) was analyzed to test for differences between sampling locations, and between control vs seep sites within each location. Effects of seasonality were also considered. PERMANOVA analyses showed that environmental conditions were different between locations and sites and that seasons did not differ significantly (Supplementary Table S1 and Fig. S1). PERMANOVA pair-wise comparisons showed that within each of the two locations control and seep sites were significantly different (Dobu: t = 3.127, P = 0.001; Upa Upasina: t = 2.547, P = 0.001). The two control sites also differed between the two locations (t = 2.112, P = 0.002), while seep sites were similar (t = 1.244, P = 0.154). The BEST routine revealed that pCO 2 and Ω AR were strongly autocorrelated with the other environmental parameters (Rho = 0.995, P = 0.001) and were therefore excluded from the following PERMANOVA analysis. PERMANOVA analyses on single environmental parameters showed that pH and total alkalinity were significantly different between control and seep sites, while temperature was significantly different between locations (Table 1 and  Supplementary Table S1). DIC was significantly different between controls of Upa Upasina and Dobu (t = 6.137, P = 0.001) while at the seep sites DIC was homogeneous (t = 0.061, P = 0.962). Salinity was unchanged between either locations or sites (Table 1 and Supplementary Table S1).

Skeletal parameters in corals sampled at the control and seep sites of Dobu and Upa
Upasina. Results for bulk density, micro-density, and porosity are reported in Fig. 2 Supplementary   Table S2. PERMANOVA analyses indicated significant differences among species in micro-density, porosity, and bulk density ( Table 2). Porosity and bulk density were also significantly different between sites (Table 2). For all species, porosity and bulk density were significantly different between control and seep sites at Upa Upasina (t = 4.752, P = 0.001 and t = 5.864, P = 0.001, respectively), with higher porosity and lower bulk density at the seep site compared to the control (Fig. 2). Bulk density was significantly lower at the seep site compared to the control also at Dobu (t = 2.675, P = 0.010; Fig. 2). Micro-density showed a significant interaction between the factor Site and Species (Table 2); indeed micro-density values assessed in P. damicornis were significantly lower at the seep site compared to the control at both locations ( Fig. 2; Table 3). Micro-density changes in P. damicornis were further explored both statistically and through additional macroscale and microscale skeletal analyses. Specifically, Time-Domain Nuclear Magnetic Resonance (TD-NMR), Thermogravimetric Analysis (TGA), and synchrotron high-resolution powder X-ray diffraction (HRPXRD) analyses were performed. TD-NMR measurements were performed on two fragments for all species to have a general overview of the T 2 distributions (Supplementary Fig. S2). Further analyses were conducted on all available fragments of P. damicornis to quantify macro-scale pore volume fraction. PERMANOVA analyses showed that macro-scale pore volume fraction was significantly higher at the seep site compared to the control in both locations (Upa Upasina: t = 2.126, P = 0.041; Dobu: t = 2.549, P = 0.028; Fig. 3 and Supplementary Table S3). The intraskeletal organic matrix (OM; t = 4.856, P = 0.004) and water content (t = 4.891, P = 0.001) were significantly different between Sites only at Upa Upasina ( www.nature.com/scientificreports/ showed lower values at the seep site compared to control, while the latter showed the opposite trend (Fig. 3).The intraskeletal OM content was significantly different also among locations (Table 4). Three P. damicornis skeletal fragments from the control and from the seep sites in Upa Upasina were analysed by HRPXRD. All HRPXRD patterns were well indexed as aragonite and no additional diffraction peaks were detected. Then, the peaks were refined using the Rietveld method 40 and lattice parameters and strain (Supplementary Table S5), and microstructural data 41 , crystallite size, and microstrain (Supplementary Table S6), were calculated. No significant differences were found between the control and seep site. To test the influence of the OM on the mineral strain, ex-situ heat treatments prior the HRPXRD measurements, which remove the effect of the OM on the strain 37 , were performed. The data showed that OM induced a positive strain on a-and c-axis and a negative one on the b-axis, but no significant differences were found between the control and seep site (Supplementary Table S5). We also measured crystallite size after the thermal annealing together with the transition to calcite (Supplementary Tables S5 and S6). These latter parameters did not show any significant difference between the control and seep sites.

Discussion
In the past decades, significant efforts have been made to quantify the ecological effects of ongoing ocean acidification (OA) in tropical regions. However, assessing the effects of OA on reef-building corals poses major challenges because multiple environmental changes, including ocean warming, are co-occurring with OA, impacting coral growth 42,43 . This study investigated the effects of long-term exposure to elevated CO 2 on skeletal properties in tropical zooxanthellate corals naturally living at CO 2 vents.
Similar to Mediterranean 13 and other tropical coral species 44 , increased porosity and decreased bulk density was observed at the seep sites compared to the control sites, with some species showing more marked trends than others, in agreement with a general decreasing trend of net calcification rates at relatively low pH conditions resembling IPCC projections 15,45 . A 2-year field transplant experiment conducted on Porites astreoides, Siderastrea siderea and Porites porites at low pH submarine springs in the Yucatán peninsula (Mexico) showed species-specific OA-related vulnerability in calcification rates which may be linked to differential growth rates, with fast-growing corals being likely more sensitive to low carbonate ion availability 46 . Species-specific sensitivities to OA also depend on its impacts on chemistry within the calcifying fluid 47,48 and/or in the diverse use of metabolic reserves 49 . Moreover, different populations of the same species might display variable responses to OA, as highlighted for instance by the intra-specific variability displayed by calcification rates of Acropora digitifera from two distinct locations after exposure to acidified conditions in aquaria 50 . All these aspects could contribute to explain the variability observed by the different species in the two locations considered in the current study.
Micro-density showed lower values at the seep sites compared to the control sites of both locations only in P. damicornis. Micro-density, which represents the mass per unit volume of the biogenic calcium carbonate composing the skeleton 51 , depends on the mineral composition of the skeleton and on intraskeletal organic matrix (OM) and water content 29 . The evaluation of additional macro-and micro-scale parameters performed in this species also revealed an increase in macro-scale pore volume fraction and intraskeletal water content and a decrease in OM, and eventually strong linked water 36 . In particular, the observed increase of intraskeletal water content at the seep site can partially justify the observed decrease in skeletal micro-density. According to literature, the observed decrease of skeletal micro-density could also stem from a variation in chemistry and micro-architecture Table 1. Means and standard deviation (in parenthesis) of the investigated environmental parameters in seep and control sites in Dobu and Upa Upasina. Lettering indicates significantly different groups (PERMANOVA on single parameters). pH TS pH in total scale, pCO 2 carbon dioxide partial pressure, Ω AR aragonite saturation, DIC dissolved inorganic carbon, TA total alkalinity, T seawater temperature, N number of measurements.   Tables 2 and 3. Plots were created with the R package ggplot2 62 . Plots were created with the R package ggplot2 62 . (photographs by co-author FR).  32 . Moreover, P. damicornis exposed to pH 7.8, 7.4 and 7.2 in aquaria for 3 weeks showed a 4 to 70-fold up-regulation of genes encoding skeleton organic matrix proteins at all pH treatments 54 . In these studies, the observed up-regulation of genes linked to OM proteins and the increase in OM content was hypothesized to promote calcification under less favorable acidified conditions. Thus, the observed decline in OM in P. damicornis in the current study suggests a possible decline in net calcification rates at the seep site, which is in agreement with the observed increase in skeletal porosity. However, considering the natural setting in which the study was performed, we cannot exclude Analysis of microscale parameters, namely intraskeletal organic matrix (OM), intraskeletal water content, and total (the sum of OM and water). The box indicates the 25th and 75th percentiles and the line within the box marks the median. Whisker length is equal to 1.5 × interquartile range (IQR). Circles represent outliers. Significant differences between control and seep sites are represented by asterisks, **P < 0.01, *P < 0.05. Plots were created with the R package ggplot2 62 . www.nature.com/scientificreports/ the influence of other covarying environmental factors in determining the observed responses (e.g., turbidity, light availability, organic/inorganic nutrient availability, feeding) [55][56][57][58] . The decrease of intra-skeletal OM content in samples from seep sites was not associated with a significant change in strain, micro-strain, or crystallite size. These observations may indicate that the amount of intracrystallite OM does not change since the crystallite sizes after the thermal annealing are the same for samples from the control and the seep sites. Thus, the observed decrease in OM is likely associated with a decrease in the inter-crystallite OM. In addition, the stability of aragonite through the transition to calcite and the lattice parameters of the calcite formed after thermal annealing did not show a significant difference between control and seep samples. The crystallographic features of aragonite from coral skeletons have been previously investigated 59 . The reef building coral Stylophora pistillata grown in aquaria under different experimental seawater acidification (pH 8.2, 7.6, and 7.3) showed anisotropic distortions of aragonite lattice parameters and a reduction of the crystallite sizes under acidified conditions 36 . In the presented study, these parameters were unaffected by decreasing pH, suggesting that biological control over calcification does not change at the nanoscale, as reported for B. europaea 13 . The fact that different species were used, but most of all that S. pistillata was exposed for 1 year (short-term acclimation) while in the current study species were exposed to acidified conditions for generations (life-long acclimatization) likely accounts for these discrepancies. The calcite phase obtained by annealing of coral samples has similar lattice parameters in samples from the control and seep sites. These parameters, when compared with those of synthetic calcite 60 , did not show differences. A different behavior was observed for calcite obtained from Desmophyllum and Favia aragonitic skeletons, which showed different strain compared with geological or synthetic calcite 59 .

Conclusions
This multi-species study showed a common phenotypic response among four zooxanthellate corals which displayed a more porous skeletal phenotype under OA. Additionally, these skeletal macromorphological adjustments led to decreased micro-density in P. damicornis but did not affect the measured crystallite features, suggesting that the fundamental structural components produced by the biomineralization process might be substantially unaffected by increased acidification 13,61 . Nonetheless, the porous phenotype here described may render structurally complex and massive corals more vulnerable to damage and bio-erosion under climate change, which in the future may lead to a weakening of the reef framework and subsequent degradation of the complex coral reef ecosystem.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. www.nature.com/scientificreports/