Mussel larvae modify calcifying fluid carbonate chemistry to promote calcification

Abstract

Understanding mollusk calcification sensitivity to ocean acidification (OA) requires a better knowledge of calcification mechanisms. Especially in rapidly calcifying larval stages, mechanisms of shell formation are largely unexplored-yet these are the most vulnerable life stages. Here we find rapid generation of crystalline shell material in mussel larvae. We find no evidence for intracellular CaCO₃ formation, indicating that mineral formation could be constrained to the calcifying space beneath the shell. Using microelectrodes we show that larvae can increase pH and [CO₃^2-] beneath the growing shell, leading to a ~1.5-fold elevation in calcium carbonate saturation state (Ω_arag). Larvae exposed to OA exhibit a drop in pH, [CO₃^2-] and Ω_arag at the site of calcification, which correlates with decreased shell growth, and, eventually, shell dissolution. Our findings help explain why bivalve larvae can form shells under moderate acidification scenarios and provide a direct link between ocean carbonate chemistry and larval calcification rate.

Fig. 1

Calcium accumulation and in vivo larval shell formation from Exp 1 and 2. a larval calcium accumulation during the first 48 hpf, N = 4 fertilizations, means and s.d. Shells are birefringent from 21 hpf onwards. b Confocal projection of early stage of shell formation in 21 hpf trochophore larva, calcein positive particles are deposited onto tissue facing side of organic shell cover and numerous calcein positive intracellular vesicles are visible underneath. Shell organic cover boundaries indicated by dashed line. c Same larva, merged transmission and calcein fluorescence images. d confocal projection of 28 hpf larva viewed from the hinge, which is not calcified yet. Shell fluorescence overexposed to visualize intracellular calcein positive vesicles. e Same larva imaged at lower intensity to illustrate shell details, calcein positive particles visible in the centre of each valve, dotted line indicates boundaries of the organic shell cover. f Close-up of calcein positive deposits attached to the organic shell cover in 21 hpf larva, organic cover boundary indicated by dotted line. g Same larva, transmission image. h Confocal projection of shell material secreted by 24 hpf larva, note deposits at the centre of the valve (c) and at the growing edge (ge). i Larvae in vivo under crossed polarized light, see also Supplementary Movie 1 for swimming and rotating larvae. j Same larvae, transmitted light. Scale bars: 20 µm b–e, 5 µm f–h, 75 µm i, j

Fig. 2

Calcein pulse-chase experiments and in vivo confocal microscopy from Exp 2. a, b Calcein fluorescence image and merged fluorescence and transmission images of larva cultured in calcein between 0–20 hpf, followed by development in filtered seawater (FSW) until 48 hpf, 1 μm thick section through body and shell to illustrate lack of calcein fluorescence in the shell. c, d Same animal, confocal projection through one entire shell valve and body to illustrate numerous calcein positive intracellular vesicles yet no fluorescence in the shell. e, f Confocal projection of calcein fluorescence of larva cultured in calcein FSW between 21–23 hpf. Shell material formed between 21–23 hpf is calcein labeled, shell material formed between 23–28 hpf is not. Calcein label of vesicles present at 28 hpf has apparently not been transferred into the shell. g, h 21 hpf animals stained with calcein for 15 min, then washed and cultured in FSW, calcein positive particles on periostracum g, shell growth bands in a slightly more advanced larva from the same fertilization h. i Confocal projection of shell calcein fluorescence of larva cultured in seawater with calcein pulse between 22–24 hpf. j Confocal projection of shell calcein fluorescence of larva cultured in seawater with calcein pulse between 27–30 hpf. c = centre of shell valve, ge = growing edge of shell valve. Scale bars: 20 µm a–f, 5 µm h, i, 10 µm j

Fig. 3

Carbonate chemistry parameters in the calcification space of PD I larvae from Exp 3. a pH_NBS. b [CO₃ ²⁻]. c Image of larva attached to a holding pipette (HP) with the tip of the measuring electrode (E) visible at the site of calcification (CS), below the shell (SH), V = velum. d [Ca²⁺]. e Ω_arag. f Setup for microelectrode measurements showing inverted microscope, the electrode, holding pipette, water bath (B), reference electrode (RE) and cooling plate (CP). Plotted data are the mean ± s.d. from N = 4 fertilizations and 5 larvae for each treatment and fertilization

Fig. 4

Impact of OA, in black on carbonate chemistry parameters at the site of calcification in red from Exp 3. a pH_NBS b ΔpH_NBS (CS-SW) c Δ [H⁺] d [CO₃ ²⁻] e Δ [CO₃ ²⁻] (CS-SW) f Ω_arag g [Ca²⁺] h Δ[Ca²⁺] i Δ log Ω_arag. Plotted data are the mean + s.d. from N = 4 fertilizations and 5 larvae for each treatment and fertilization.

Fig. 5

Impact of CO₂-driven seawater acidification on calcification in PD I larvae from Exp 3 and 4. a Shell length at PD I, veliger larval stage. Data are means ± s.d. from N = 5 fertilizations and 10–25 animals per treatment and fertilization. Linear regression: shell length (µm) = 113.3–0.0115 pCO₂ (µatm), R ² = 0.74, F_(1,22) = 63, p < 0.001. b Calcein fluorescence intensity of larval shell portions formed during 21–24 hpf imaged following 3 day exposure to OA. Data are the mean ± s.d. from one experiment. Sigmoidal regression: fluorescence intensity (a.u.) = 155.8 + 40.6/(1 + exp (−(pCO₂ − 1639)/−156.1)), R ² = 0.36, F_(3,338) = 64, p < 0.001. c Calcein fluorescence confocal projections of larvae used to measure shell dissolution. d Merged fluorescence and transmission images. Scale bars: 75 μm