Dataset: Physiological data from Zn and CO2 culture experiments on the diatom Phaeodactylum tricornutum

Growth rates: "Growth rates were calculated via a linear regression of the natural log of cell counts during their exponential growth. To monitor growth, the cell densities of all cultures were measured using a flow cytometer (BD Accuri C6 Plus) and gated according to their Forward Scatter and PerCP-A (Peridinin-Chlorophyll-Protein A) concentration. At cell densities exceeding 104 cells mL-1, the aliquot taken for flow cytometry was diluted down for an accurate cell concentration measurement." 

Measurements of seawater carbonate chemistry profile: "Seawater samples were taken for total alkalinity (TA, µmol kg-1) and DIC (µmol kg-1) were taken at the termination of the experiment to examine the carbonate chemistry profile of the seawater at the same time point at which samples were taken for CA activity proteomics. Samples were passed through a 0.2 μm filter (Whatman) to remove biomass and stored in 13 mL borosilicate vials where they were poisoned with mercuric chloride (0.02 % final concentration), sealed without headspace, and stored in the dark at room temperature prior to measurement. TA was measured using an open-system Gran titration on 1.5 mL samples brought to a total volume of 5 mL with Milli-Q water in technical duplicate using a Metrohm 805 Dosimat and a robotic Titrosampler (Borer et al., 2026). The precision of TA measurements assessed by the reproducibility of the standard was ± 2.5 μmol kg−1. DIC was measured on 1.5 mL samples in technical duplicate using an Apollo SciTech AS-C6L DIC analyzer equipped with a LI-COR 7815 CO2 detector (Wang et al., 2017). The precision of DIC measurements as calculated by the mean difference between technical duplicates was ± 4.2 μmol kg−1. Measurements of TA and DIC were calibrated against in-house standards using seawater collected from Vineyard Sound, which were previously calibrated against seawater-certified reference materials (Batch #205) (Dickson, 2026). Measured values of DIC and TA were used as input parameters along with T, Sal, Total P, and Total Si in CO2SYS (Lewis et al., 1998) (ver. 2.3; using constants from Mehrbach et al. (1973) refit by Dickson and Millero (1987)) to calculate pH, [HCO3-], [CO32-], [CO2]aq, and ppm CO2."

Carbonic anhydrase activity: "Carbonic anhydrase activity was assayed according to an 18O depletion method utilizing a membrane inlet mass spectrometer (MIMS), which measures concentrations of dissolved gases (Burlacot et al., 2020). A Pfeiffer QMG 220 mass spectrometer (closed-type ion source, 1–100 AMU mass range) was fitted with custom-made inlet consisting of a 6 inch long, 1/1600 O.D. stainless steel needle with 1 inch of perforations at the tip. Silastic tubing (0.23 mm thick) was gently slid over the needle to cover the perforations and was fixed by a knot at the tip of the needle. The needle functioned in tandem with the mass spec’s high vacuum (operating at 4×10−6 mbar) to pull dissolved gases from the sample or standard solution, without introduction of water, into the quadrupole mass spectrometer for detection based on their mass to charge (m/z) ratio.

The 18O depletion method tracks the ion currents of six different CO2 isotopes over time as a double-labeled spike solution (NaH13C18O3, m/z 49) interacts with a background buffer solution (described above). To prepare the spike solution, a 13C-labeled sodium bicarbonate salt (85 mg; MilliporeSigma 870081-58-1) was equilibrated in 1 mL of 18O-labeled water (≥ 97% labeled 18O; MilliporeSigma 14314-42-2) for 24 hr. For every sample, 15 µL of spike was pipetted into the 5 mL aliquot of a sample slurry solution (phosphate buffer medium (0.14 M sodium phosphate, pH 8) and resuspended cell pellet). The reaction was run in a conical reaction vial (MilliporeSigma Z115150) with a stir bar (MilliporeSigma 23227) gently agitating the solution and was held in a thermostat water jacket kept at 20°C. Utilizing the MIMS rapid response time, the interaction of the spike and the buffer are closely followed. 

The addition of the spike results in an initial increase in m/z 49, which is immediately diluted by the buffer’s intense background 16O signal. The fractional abundance of 18O in 13CO2 is used to calculate a rate of equilibration. Taking the slope of the fractional abundance in log space versus time yields the rate of CO2 hydration/dehydration (Subhas et al., 2019; Uchikawa & Zeebe, 2012).

Bovine carbonic anhydrase (BCA) was used to create a calibration between known concentrations of BCA and the measured rates of CA activity. Standard curves were used to assess instrument performance but were not used to calculate an effective CA concentration for samples, as the activity of a known concentration of bovine carbonic anhydrase is incomparable to phytoplankton CAs (Krishnamurthy et al., 2008). Additionally, “blank” measurements of buffer containing no standard addition of BCA were run every analytical session. The mean plus 1σ standard deviation of all the blanks provided a conservative estimate of the variability in these measurements. This value was subtracted from the sample’s measured CA activity to achieve the blank subtracted CA activity. The data in this manuscript are reported as blank subtracted CA activity normalized to the number of cells analyzed for the respective sample, with units of reactions sec-1 cell-1."