How calorie-rich food could help marine calcifiers in a CO₂-rich future

Abstract

Increasing carbon emissions not only enrich oceans with CO₂ but also make them more acidic. This acidifying process has caused considerable concern because laboratory studies show that ocean acidification impairs calcification (or shell building) and survival of calcifiers by the end of this century. Whether this impairment in shell building also occurs in natural communities remains largely unexplored, but requires re-examination because of the recent counterintuitive finding that populations of calcifiers can be boosted by CO₂ enrichment. Using natural CO₂ vents, we found that ocean acidification resulted in the production of thicker, more crystalline and more mechanically resilient shells of a herbivorous gastropod, which was associated with the consumption of energy-enriched food (i.e. algae). This discovery suggests that boosted energy transfer may not only compensate for the energetic burden of ocean acidification but also enable calcifiers to build energetically costly shells that are robust to acidified conditions. We unlock a possible mechanism underlying the persistence of calcifiers in acidifying oceans.

1. Introduction

Calcifying organisms (e.g. corals, shellfish and sea urchins) are diverse and widespread in the ocean, but their ability to build calcareous shells or skeletons is predicted to be impaired by ocean acidification [1]. This classic paradigm of calcification is, however, increasingly challenged by recent evidence showing that some calcifiers are able to maintain or even enhance shell growth and shell strength under ocean acidification [2–4]. In addition, boosted populations of a calcifier have been observed in naturally CO₂-enriched environments [5,6], suggesting positive rather than negative effects of ocean acidification. These counterintuitive results seem to occur when research considers either interactions between calcifiers and other components in marine ecosystems (i.e. ecological complexity) [7] or potential acclimation capacity of calcifiers (i.e. compensatory mechanisms) [4,8]. To reconcile such results, research is required to understand how ocean acidification affects calcification and the potential compensatory dynamics in naturally complex environments.

Since calcification is an energy-demanding process [9], energy intake through food consumption may play a critical role in the dynamics of shell building in acidifying oceans. Importantly, herbivorous calcifiers may indirectly benefit from CO₂ enrichment which not only boosts the biomass of their food (e.g. algae) [5,10], but also its nutritional quality [11,12]. In other words, CO₂ may act as a resource via trophic transfer from primary producers to herbivores. If this model of trophic transfer is realistic, then links should be found among CO₂ enrichment, nutritional quality of primary producers and shell quality of calcifiers.

Using natural CO₂ vents in the western Pacific, we assessed whether CO₂ enrichment acts as an additional resource which enhances the food quality (nutrients and energy content) of herbivorous calcifiers and hence their shell quality via trophic transfer. We hypothesized that increased food quality by CO₂ enrichment allows herbivorous calcifiers to adjust shell building to acidified conditions (i.e. production of more durable shells). Our assessment focused on shell properties that affect the durability of shells, including thickness, crystallinity, mechanical resilience and organic matter content. A herbivorous calcifier (gastropod Eatoniella mortoni) was chosen as the study species because its populations are able to persist in CO₂-enriched environments [6] and, counterintuitively, benefit from CO₂ vents [5,13].

2. Material and methods

(a). Study site and species

We studied the natural CO₂ vents located at Te Puia o Whakaari (White Island), Bay of Plenty, New Zealand (37°31.274′ S, 177°10.972′ E). This area represents a rocky reef habitat characterized by a mosaic of kelp (Ecklonia radiata), turf algae and urchin barrens that support a gastropod (Eatoniella mortoni). This herbivorous calcifier was chosen for this study because it can live in CO₂-enriched environments throughout its life, with limited ability to disperse, and localized populations [6], which appear to benefit from CO₂ enrichment [5] via elevated consumption rates of turf algae [13]. The CO₂ plumes at the vents extended approximately 24 × 20 m from the source at 6–8 m depth and acidified the surrounding seawater to pH levels representing approximate end-of-the-century projections (approx. pH 7.8; year 2100 under RCP 8.5 scenario) without confounding differences in temperature. The nearby control sites located approximately 25 m away from the vents represent contemporary pH levels (approx. pH 8.1). Analysis of seawater chemistry detected lower pH at the vents without a change in nutrient and mineral concentrations, albeit two exceptions detected inconsistent variation among sites that were unrelated to the vents (electronic supplementary material, tables S1 and S2). Hence, the variation in seawater chemistry between vents and controls reflected a consistent effect of ocean acidification, rather than nutrients and minerals. The spatial and temporal patterns of seawater carbonate chemistry and the turf algae at the study site were previously presented [13], where the white cyanobacterial mats that typically surround CO₂ vents were not in contact with our samples (i.e. at least 10 m away from sampling). The spatial patterns of seawater carbonate chemistry have shown consistent patterns, independent of the month sampled and time of day among the 5 years of study [13].

Gastropods were sampled with their food (same species of turf algae) from two independent vent sites (north and south) and two independent control sites (north and south) along the northeastern coast of the island (n = 5 replicate samples per site), so that the shell quality of gastropods (shell thickness, crystallinity, mechanical resilience and organic matter content) could be correlated with the nutritional quality of their food (protein, carbohydrate, lipid and energy content).

(b). Nutritional quality of turf algae

The protein, carbohydrate, lipid, energy content and C:N ratio of turf algae were analysed. Algal samples were freeze-dried and powdered prior to chemical analyses of macronutrients, including protein, carbohydrate and lipid (n = 5 replicate samples per site). Protein was extracted by 0.6 M sodium hydroxide at 40°C for 12 h [14], followed by a bicinchoninic acid assay using bovine serum albumin as the standard to determine protein content [15]. Carbohydrate content was determined by a phenol-sulfuric acid method using glucose as the standard [16], after extracting the soluble carbohydrate by boiling water for 2 h. Lipid was extracted by chloroform and methanol mixture (2 : 1, v/v) for 24 h and the lipid content was measured gravimetrically [17]. The proportion of non-energy components was calculated by subtracting protein, carbohydrate and lipid content from 100%. Energy content of algae was calculated using the energy conversion factors specified in EU Directive 90/496 [18]. C:N ratio of algae was measured using a CHNS elemental analyser (2400 Series II, Perkin Elmer, USA).

(c). Shell properties of gastropods

Individual shells of the largest size at each site were rinsed with deionized water and oven-dried at 50°C before analyses. To measure shell thickness as an indicator of capacity for precipitating calcium carbonate, the shells (shell length: approximately 1.54 mm) were mounted firmly on a plastic plate with the outer lip (i.e. growth edge) pointing vertically upward and then photographed under a stereomicroscope (SMZ25, Nikon). Three random locations on the outer lip of each individual were measured using Nikon NIS-Elements imaging software and then averaged to indicate the shell thickness of an individual. The average shell thickness of seven individuals from each sample was obtained to represent a replicate (n = 5 replicate samples per site).

Relative amorphous calcium carbonate (ACC) content was analysed to determine crystallinity using a Fourier-transform infrared spectrometer (Spectrum 100, PerkinElmer, USA). Seven individuals from the same sample were powdered to make a composite sample (n = 5 replicate samples per site), which was then transferred onto the sample holder of the spectrometer to obtain the infrared absorption spectrum, ranging from 650 to 1800 cm^–1 with background correction. The relative ACC content was estimated as the intensity ratio of the peak at 856 cm^–1 to that at 713 cm^–1 [19].

To measure mechanical resilience, shells were embedded in resin by cold mounting and then polished to expose their cross section (Struers TegraPol-11; Struers A/S, Valhojs Alle, Ballerup, Denmark). To obtain a smooth surface, a series of three polishing steps were performed first using MD Largo, MD Dac and finally MD Chem polishing discs with DiaPro Allegro/Largo 9 µm diamond suspension, DiaPro Dac 3 µm diamond suspension and OP-S 0.04 µm colloidal silica suspension, respectively. The hardness (H) and elastic modulus (E) of shells were determined by nanoindentation (IBIS, M/S Fisher-Cripps Laboratory, Australia) using a diamond Berkovich tip. The area function of the indenter tip was calibrated using standard fused silica. Load-controlled indentation with a maximum load of 50 mN was performed on the polished shell surface of the outer lip. The loading rate was set to 2.5 mN s^–1, which represents the static response of the material. Six locations on the shell were indented with a distance of 50 µm between two separate indents to avoid the interference of residual deformation from neighbouring impressions (n = 5 replicate shells per site). The method described by Oliver & Pharr [20] was used to quantify H and E, whereas the ratio of H to E, which represents the resistance of a material to elastic deformation [21], was calculated to indicate the mechanical resilience of shells.

To measure organic matter content, the shells were individually powdered in a pre-weighed crucible and the flesh was removed under a microscope (n = 5 replicates per site and n = 2 shells per replicate). The organic matter content of shell powder was determined by weight loss upon ignition at 550°C in a muffle furnace for 6 h [22].

(d). Statistical analysis

Differences between vents and controls were tested by analysis of variance (ANOVA). Two-way ANOVA treated ‘Vent’ (vent versus control) as fixed factor and ‘Site’ (north versus south) as random factor which is orthogonal to ‘Vent’. Post hoc pooling provided a more powerful test of ‘Vent’ when the interaction ‘Vent × Site’ was insignificant (p > 0.25) [23]. Prior to analysis, data were tested for homogeneity of variances using Cochran's C-Test [24]. Heterogeneous data were transformed and if heteroscedasticity was not removed by transformation, data with the more homogeneous variances were used for the analysis with the more conservative probability of 0.01. All data (i.e. seawater chemistry parameters, nutritional quality of algae and shell quality of gastropods) were sampled with the same level of replication and analysed with the same protocol.

3. Results and discussion

Energy governs the internal dynamics of organisms (i.e. homeostasis), propagating through to the dynamics of their populations and ecosystems [25]. We found that the energy content of algae was boosted by CO₂ enrichment (figure 1a; electronic supplementary material, figure S1 and table S2), but not nutrient or mineral concentrations in seawater which did not differ between controls and vents (electronic supplementary material, figure S2 and table S2). This increase in energy content was caused by greater protein and carbohydrate content, which is probably due to greater nitrate uptake and nitrate reductase activity for protein production [26–28] and photosynthesis for carbohydrate production [10,29].

Figure 1.

A conceptual diagram showing how CO₂ enrichment can benefit herbivorous calcifiers through trophic transfer. (a) At vents, the algae have a higher content of protein, carbohydrate, but not lipid, for which energy content and relative nitrogen content increase (i.e. lower C:N ratio) (mean ± S.E., n = 10 each from two vents and two controls). (b) The increased energy content is associated with the production of thicker, more crystalline (i.e. reduced relative ACC content) and more mechanically resilient shells with higher total organic matter. Each data point is derived from a single sample that contains both energy content of turf algae and shell property of gastropods. (Online version in colour.)

Both CO₂ and nitrogen are often limiting resources in marine systems, where producers and consumers are limited by CO₂ and nitrogen, respectively. For producers, the acquisition of CO₂ often requires energy expenditure for carbon concentrating mechanisms (i.e. conversion of bicarbonate ions into CO₂ for photosynthesis) [30,31]. For herbivores, nitrogen is regarded as a key element for their growth and reproduction but has widespread limitation in their diets [32]. We found that CO₂ enrichment was associated with the higher nutritional quality of algae, indicated by reduced C:N ratio (i.e. nitrogen increases relative to carbon; figure 1a; electronic supplementary material, figure S1 and table S2), which may account for observations of intensified feeding of herbivores by elevated CO₂ under field [13] and laboratory conditions [11,12].

While the energetic cost of shell building increases in acidifying oceans [33], gastropods showed their capacity to adjust shell building to acidified conditions (i.e. production of thicker, more crystalline and more mechanically resilient shells; electronic supplementary material, figure S3 and table S2), which also enrich the energy content of their food (figure 1b). The conversion of more soluble building materials (i.e. ACC) into less soluble materials (i.e. crystalline calcium carbonate) at vents indicates the adaptive response of calcifiers to ocean acidification (electronic supplementary material, figure S4). We also found that the shells were more mechanically resilient at vents. Since mechanical strength is chiefly determined by the organic matrix occluded in the shell, rather than the calcium carbonate mineral per se [34], the increased mechanical resilience is likely to be associated with a higher organic matter content of the shell, which can facilitate energy dissipation upon compression [35]. Such mechanically robust shells have higher resilience to breakage and thus help reduce shell damage and even mortality (e.g. due to predators and waves). The positive relationships between food quality and shell quality (figure 1b) not only suggest that boosted trophic transfer via CO₂ enrichment might offset the energetic burden of ocean acidification but also account for an apparent enigma where laboratory observations of impaired shell quality are seldom evident when food is available [2]. Aragonite is the most common building material of calcareous shells, but its formation is anticipated to be weakened under ocean acidification due to potential shell dissolution at low pH [1]. However, we found that aragonite can still be produced at vents (electronic supplementary material, figure S5) and the shell ultrastructure appeared unaffected (electronic supplementary material, figure S6). Together, these findings suggest that calcifiers can still build aragonitic shells without compromising shell integrity under near-future ocean acidification.

Ecological complexity involves processes that buffer organismal and community change from environmental change [7], but most research was based on simplified laboratory experiments [36]. Here, we explore the indirect effects of ocean acidification (plant–herbivore interaction) on a critical organismal function (calcification by herbivores) within the full complexity of a natural environment. We deepen this exploration by testing whether CO₂ enrichment can also boost the energy content of food (e.g. algae) for herbivores, a fundamental element required for enhanced shell building. We suggest that energy-enriched food, along with faster feeding rates [13], allows herbivorous calcifiers to build shells that are less soluble and more durable (i.e. thicker and more crystalline by precipitating more carbonate minerals and crystallizing more ACC; mechanically stronger by synthesizing more organic matrix). Physiologically, many calcifiers exhibit compensatory mechanisms (e.g. increased activity of proton and calcium pumps) to offset the reduced pH in their body fluid [8,37] so that calcification can be sustained or even enhanced [2,4]. Consuming food of higher energy content provides more energy for calcifiers to fuel these energy-requiring mechanisms. While increased food availability has been suggested to help calcifiers resist the impacts of ocean acidification [2], food availability is unlikely to be a limiting factor for many herbivorous calcifiers, particularly those feeding on expanding turf mats that are fuelled by CO₂ enrichment [13]. Therefore, the elevated energy content of food, which increases energy gain per unit of feeding effort, may be the key to enhanced calcification under ocean acidification.

Persistence of calcifiers in naturally CO₂-enriched environments suggests the existence of mechanisms that allow calcifiers to adapt to ocean acidification. By considering plant–herbivore interactions, we found that CO₂ enrichment in seawater increases the energy content of food, which is in turn linked with enhanced shell quality of a herbivorous calcifier (figure 1). Such trophic transfer may represent a key mechanism that compensates for the energetic burden of ocean acidification, allowing calcifiers not only to construct more durable shells but also to persist in acidifying oceans.

Data accessibility

The supporting data are uploaded as electronic supplementary material.

Authors' contributions

S.D.C. and J.Y.S.L. conceived the research, designed the experiments, collected and analysed the data. Y.C. performed the nanoindentation tests. J.Y.S.L. and S.D.C. wrote the manuscript. Z.A.D. helped shape the research and conceptualized the data visualization. All authors contributed to the development of paper content and commented on the final manuscript and revisions.

Competing interests

We declare we have no competing interests.

Funding

This research was supported by ARC Future Fellowships awarded to S.D.C. (FT0991953) and I.N. (FT120100183).