The Southern Ocean diatom Pseudo-nitzschia subcurvata flourished better under simulated glacial than interglacial ocean conditions: Combined effects of CO2 and iron.

The 'Iron Hypothesis' suggests a fertilization of the Southern Ocean by increased dust deposition in glacial times. This promoted high primary productivity and contributed to lower atmospheric pCO2. In this study, the diatom Pseudo-nitzschia subcurvata, known to form prominent blooms in the Southern Ocean, was grown under simulated glacial and interglacial climatic conditions to understand how iron (Fe) availability (no Fe or Fe addition) in conjunction with different pCO2 levels (190 and 290 μatm) influences growth, particulate organic carbon (POC) production and photophysiology. Under both glacial and interglacial conditions, the diatom grew with similar rates. In comparison, glacial conditions (190 μatm pCO2 and Fe input) favored POC production by P. subcurvata while under interglacial conditions (290 μatm pCO2 and Fe deficiency) POC production was reduced, indicating a negative effect caused by higher pCO2 and low Fe availability. Under interglacial conditions, the diatom had, however, thicker silica shells. Overall, our results show that the combination of higher Fe availability with low pCO2, present during the glacial ocean, was beneficial for the diatom P. subcurvata, thus contributing more to primary production during glacial compared to interglacial times. Under the interglacial ocean conditions, on the other hand, the diatom could have contributed to higher carbon export due to its higher degree of silicification.

Introduction

The Southern Ocean (SO) is the world’s largest high-nutrient low-chlorophyll region (HNLC) and an area where physical forcing, atmospheric pCO2, biological production and marine biogeochemical cycles are tightly linked. In this region, primary production is restricted by the bioavailability of the trace metal (TM) iron (Fe) [1-3]. Fe is an essential trace element, which is needed by phytoplankton to transfer electrons in key cellular and metabolic processes including photosynthesis, respiration, chlorophyll production, carbon (C) and nitrogen (N) fixation [4]. The availability of Fe strongly influences phytoplankton species composition and growth [5-9], and impacts the biological carbon pump and thus the global carbon cycle. Consequently, changes in Fe availability have caused feedback effects on climate over geological timescales. Furthermore, the SO is a region of high CO2 exchange between ocean and atmosphere [10, 11]. The SO has been reported to be a major sink of atmospheric CO2 during glacial periods, while it was on the other hand a source of CO2 during glacial-interglacial transitions. At present day, the SO is the major sink of anthropogenic CO2 [12, 13].

During the Last Glacial Maximum (LGM), the SO experienced changes in oceanic circulation and carbon storage. For instance, increased sea ice extent strengthened surface water stratification, thus limiting ocean ventilation and trapping more carbon in the deep ocean [14-16]. Additionally, the northward displacement of the westerly winds prevented the upwelling of CO2-rich deep water [17]. Besides physical mechanisms, the strength of the biological pump might explain 25–50% of the roughly 100 μatm pCO2 discrepancy between glacial (180 μatm pCO2) and interglacial (280 μatm pCO2) times as argued in several studies [18-21]. This supports John Martin’s ‘Iron Hypothesis’, which suggests that an increase in dust deposition during glacial times would fertilize the ocean, stimulate marine productivity, and enhance C export [1, 2, 22–24]. Indeed, analysis of sediment cores revealed a positive correlation between aeolian Fe supply and primary production during ice ages [20, 25]. Proxy data as well as model simulations showed a doubling of the global dust deposition during the last glacial climate condition, when 826 Tg/yr dust were deposited in the global ocean, compared to the 440 Tg/yr dust in pre-industrial times [26, 27]. The difference was mainly due to a dryer atmosphere and reduced vegetation cover [22]. Used a biogeochemical model to estimate the impact of Fe deposition on the global ocean. Under current conditions, 33% of the world’s oceans water masses have Fe concentrations, which limit the growth of phytoplankton. The model simulations revealed that the percentage of Fe-poor water masses decline to 25% and 13% with pre-industrial and LGM dust input, respectively. Along with the North Pacific Ocean, the SO showed the most significant difference in soluble Fe deposition during glacial and interglacial times, accordingly having the largest impact on marine biogeochemistry [22]. The dust deposition in the SO during glacial times was roughly ten times higher (0.04–0.17 Tg/yr) than in pre-industrial times (0.005–0.018 Tg/yr) [27]. However, the SO is geographically isolated from arid, dust-producing regions and is thus overall characterized by low aeolian Fe deposition [28]. Other sources of Fe include upwelling of deep nutrient-rich water, entrainment of sedimentary Fe from continental shelfs and resuspension, island-wake effects, seasonal sea ice extent and melt, as well as iceberg drift and melt [3 and references therein].

The phytoplankton community in the current SO is dominated by different diatom species and the prymnesiophyte Phaeocystis antarctica [29]. Diatoms account for 40% of the ocean’s total primary production [30-33] and dominate the export of particulate organic matter to the seafloor [34, 35]. In other words, diatoms are crucial for the ocean’s ability to sequester C to the ocean’s interior. Diatoms also have an extensive impact on the oceanic silica inventory, as they produce frustules containing silica. Some frustules are resistant to remineralization and dissolution, are well preserved in the sediment, and thus provide precious information about past oceanic biogeochemistry. Pseudo-nitzschia species have been frequently observed in today’s phytoplankton assemblages in Antarctic waters [36]. Mesoscale Fe fertilization experiments in the SO triggered massive phytoplankton blooms dominated by large diatoms like the pennate Pseudo-nitzschia sp. [7, 37]. Large diatoms in the SO appear to have a higher Fe requirement compared to smaller phytoplankton because of physical constrains in the Fe uptake process [38]. To compensate for this, they have evolved various strategies to acquire bioavailable Fe. They generally reduce their biogeochemical Fe requirement through metal or protein substitution [39] and reduce Fe-rich components of the photosystem apparatus [9]. [8] suggested that Pseudo-nitzschia is able to accumulate intracellular Fe when ambient concentrations of this TM are high, while maintaining a low Fe demand. This luxury uptake and subsequent storage of Fe supports growth in subsequent low Fe environments and enables Pseudo-nitzschia to dominate phytoplankton assemblages across a wide range of oceanic Fe concentrations.

In SO diatoms, Fe limitation often results in slower growth and reduced C fixation. The photochemical quantum efficiency, which indicates how efficiently excitation energy is transferred to the reaction centers, is usually lowered [8, 40–42]. In an Fe-poor environment, cells usually increase the functional absorption cross sectional area of their reaction centers, thereby enhancing the target area, which absorbs incoming photons [38, 41]. The absorbed photons can either drive photosynthesis, N reduction, C fixation, photorespiration or can be converted to heat (non-photochemical quenching). Fe deficiency induces changes in the photosystem II (PSII) reaction centers such as the reduction of the pigment content [43], causes less efficient electron transport [40] and increases non-photochemical quenching to dissipate the excess light energy [44].

Besides Fe limitation, phytoplankton cells have experienced variations in CO2 concentration in the past. Previous studies on the effect of high CO2 concentrations on phytoplankton reported changes in their elemental composition (e.g. [45, 46]), in cell size (e.g. [47]) and in the degree of silicification in diatoms (e.g. [48]). Furthermore, it was shown that low pCO2 levels can influence the composition of Antarctic phytoplankton communities. For example, experiments with natural phytoplankton assemblages from different regions across the SO [24, 49, 50] concluded that Pseudo-nitzschia flourishes at low pCO2 levels, while it does not do well in response to ocean acidification. Indeed, between ambient and future elevated pCO2 levels, the growth of P. subcurvata in a laboratory experiment was not stimulated under enhanced Fe supply [51]. Under similar Fe conditions, a phytoplankton community from the Ross Sea, Antarctica, responded to CO2 increase from 100 to 800 ppm with a dramatic reduction in cell abundance of P. subcurvata, being replaced by Chaetoceros species [49]. Similarly, a community from the Weddell Sea, Antarctica, shifted from Pseudo-nitzschia to Fragilariopsis after Fe addition between 390 to 800 μatm pCO2 [50], while no difference in species composition was found between the glacial (190 μatm) and the present-day (390 μatm) pCO2 levels. This implies that reduced CO2 concentrations during glacial periods potentially favored pennate diatoms such as Pseudo-nitzschia while diatom species such as Chaetoceros and Fragilariopsis became most abundant under present-day and future pCO2 levels [49]. A few studies investigated the SO phytoplankton assemblages and growth under low Fe supply in response to increasing pCO2 [24, 50, 52]. [50] observed also a CO2-dependent taxonomic shift in Fe-deplete conditions with increasing pCO2 with Pseudo-nitzschia being replaced by the pennate diatom Synedropsis between 390 and 800 μatm pCO2 levels. Similarly, when pCO2 increased from 390 to 900 μatm another SO plankton community changed from being dominated by P. prolongatoides to one, which was dominated by P. antarctica [24]. Hence, irrespective of Fe availability the genus Pseudo-nitzschia was found to be susceptible to ocean acidification pCO2 levels.

Studies that asses the effects of low pCO2 on phytoplankton often compare their results with high pCO2 levels to understand ocean acidification. However, little is known about the smaller variation from 180 (glacial) to 280 μatm (interglacial/pre-industrial) pCO2 under different Fe availability. Indeed, the potential interactive effect of low-pCO2 (180 and 280 μatm) together with different Fe availability (deplete and replete) on net primary production and export production is currently often not considered, when developing models or designing laboratory experiments simulating glacial and interglacial ocean conditions. Studies looking at N-isotopes and Th-corrected sediment accumulation rates describe large fluxes of biogenic detritus out of surface waters in the glacial ocean due to a larger amount of lithogenic Fe transported by winds [26]. The latter study indicates that increased export production and thus enhanced C storage potentially contributed to the observed lower atmospheric CO2 concentrations during glacial times [53].

The above-mentioned studies offer first insights on how some phytoplankton species cope with glacial and interglacial climatic conditions. However, studies on the ecophysiology of Antarctic diatoms subject to glacial vs. interglacial ocean conditions under reduced Fe conditions, are yet lacking. In this study, the SO bloom-forming diatom P. subcurvata was grown under Fe and CO2 conditions representative of glacial (lower CO2 and higher Fe) and interglacial (higher CO2 and lower Fe) times to untangle the influence of these two environmental factors on growth, elemental stoichiometry, photosynthetic carbon production and photophysiology. This allowed to assess its role in the paleo carbon cycle.

Material and methods

Experimental setup

Prior to the execution of the experiment, the oceanic diatom P. subcurvata (isolated by Philipp Assmy at 49°S, 2°E, R/V Polarstern cruise ANT-XXI/4, April 2004) was grown for more than one year in Antarctic seawater with a low total dissolved Fe (dFe) concentration of 0.5 nmol L-1 Preacclimation and the main experiment were carried out in Fe-poor (0.4 nmol L-1) Antarctic seawater collected at 60°32S, 26°29W (salinity of 33.8 ± 0.2), filtered through a sterilized, acid-cleaned 0.2 μm filter (Sartobran, Sartorius). This water was spiked with chelexed (Chelex® 100, Sigma Aldrich, Merck) macronutrients (100 μmol L-1 Si, 100 μmol L-1 NO3- and 6.25 μmol L-1 PO43-) and vitamins (30 nmol L-1 B1, 23 nmol L-1 B7 and 0.228 nmol L-1 B12) according to the F/2R medium [54]. In addition, a TM mix containing Zn (0.16 nmol L-1), Cu (0.08 nmol L-1), Co (0.09 nmol L-1 Co), Mn (1.9 nmol L-1), Mo (0.05 nmol L-1) in the ratio of the original F/2 recipe adjusted to 4 nmol L-1 Fe was added. As suggested by [55], in order to minimize the alteration of the natural seawater TM chemistry and ligands, no ethylenediaminetetraacetic acid (EDTA) was added. The Fe-deplete treatments (henceforth referred to as ) contained 0.4 nmol L-1 dFe while for the Fe-enriched treatments (henceforth referred to as ), 4 nmol L-1 FeCl3 were added.

To avoid Fe contamination, TM clean techniques were used according to the GEOTRACES cookbook [56]. The sampling and handling of the incubations was conducted under a laminar flow hood (Class 100, Opta, Bensheim, Germany). All equipment was soaked for one week in 1% Citranox, followed by two weeks in 1 N HCl for polycarbonate and 5 N HCl for polyethylene materials. In between and after the cleaning process, the equipment was rinsed seven times with Milli-Q (MQ, Millipore). Finally, everything was air dried under a clean bench (U.S. class 100, Opta, Bensheim, Germany) and packed in three polyethylene bags.

All Control and +Fe incubations were bubbled with humidified air containing pCO2 levels of 190 and 290 μatm, henceforth referred to as and , respectively. Using a gas flow controller (CGM 2000, MCZ Umwelttechnik, Bad Nauheim, Germany), both CO2 gas mixtures were generated by combining CO2 free air (< 1 ppmv CO2, Dominick Hunter, Kaarst, Germany) with pure CO2 (Air Liquide Deutschland Ltd., Düsseldorf, Germany) in the respective ratios. They were regularly monitored with a Li-Cor (LI6252 Biosciences, Lincoln, NE) calibrated with CO2 free air and purchased gas mixtures of 150 ± 10 and 1000 ± 20 ppmv CO2 (Air Liquide Deutschland Ltd., Düsseldorf, Deutschland). Low pCO2 and Fe input characterized the glacial ocean, which was here simulated in the treatment. Vice versa, the interglacial ocean was characterized by higher pCO2 and no Fe input and mimicked by the treatment. In addition to the incubation bottles, Fe and carbonate chemistry were determined in the culture medium which was incubated in the same way as the respective incubation bottles (pCO2 and Fe availability), to check if the different pCO2 and Fe manipulations were successful.

All incubations were placed in front of LED (light-emitting diode) lamps at 100 μmol photons m-2 s-1 under a light:dark cycle of 16:8 h. The light intensity was adjusted with a LI-1400 datalogger (Li-Cor Biosciences, Lincoln, NE, USA) with a 4π-sensor (Walz, Effeltrich, Germany). For this experiment, the long-term low Fe acclimated P. subcurvata stock culture was inoculated to the different CO2-Fe conditions and was acclimated to each experimental condition at 2°C for at least two weeks. The main experiment was carried out in triplicate 4 L acid-cleaned polycarbonate bottles for each experimental treatment. The main experiment started with initial cell densities of ~1000 cells mL-1, lasted between 8 and 9 days and reached final cell densities between 67 000 and 107 000 cells mL-1.

Trace metal chemistry

At the end of the experiment, total dissolved Fe (dFe) samples were taken from the culture medium by filtering 100 mL from each bottle through 0.2 μm HCl-cleaned polycarbonate filters (47 mm, Nuclepore, Whatman, GE Healthcare, Chicago, IL, USA) using a trace metal clean filtration system under a clean laminar flow hood (Class 100, Opta, Bensheim, Germany). The filtrate was then filled into a 125 ml HCl-cleaned PE bottle and stored triple-bagged at 2°C until analysis. Between each filtration, the filtration manifold was cleaned in an acid bath consisting of 1 M HCl and rinsed seven times with Milli-Q. Prior to the dFe analysis, 0.2 μm pre-filtered seawater samples were acidified to pH 1.75 with double distilled HNO3, minimizing the formation of Fe and Mn hydroxides. Next, samples were UV (ultraviolet) oxidized for 1.5 h using a 450 W photochemical UV power supply (photochemical lamp 7825; Power Supply 7830, ACE GLASS Inc., Vineland N.J., USA). Total dFe concentration of the seawater samples and the processed blanks were measured with a seaFAST system (Elemental Scientific, Omaha, NE, USA) [57] coupled to a sector field inductively coupled plasma mass spectrometer (ICP-MS; Element 2, Thermo Fisher Scientific; resolution of R = 4000; oxide forming rates below 0.3%). To minimize matrix effects, the seawater dFe concentrations were analyzed by standard addition. The accuracy of the dFe data was assessed by measuring NASS-6 (National Research Council of Canada) reference standards, with a recovery rate for Fe of 110%.

Carbonate chemistry

From the culture medium as well as from the incubation bottles at the end of the experiment, dissolved inorganic carbon (DIC) was filtered through 0.2 μm filters (Nalgene, Thermo Scientific) and was stored at 4°C in 5 mL borosilicate glass bottles without headspace. The colorimetric analysis was performed with a QuAAtro autoanalyzer (Seal Analytical, [58]). Again, from the culture medium as well as from the incubation bottles at the end of the experiment, samples for the total alkalinity (TA) were filtered through 0.6 μm GF/F filters (Whatman) and stored at 4°C in 150 mL borosilicate glass bottles. TA was measured via potentiometric titration [59] and the concentrations were calculated using a linear Gran Plot [60]. The pCO2 was calculated using the CO2Sys program [61] with the equilibrium constants of [62] as refitted by [63] using TA and DIC measurements, concentrations of phosphate and silicate, temperature and salinity.

Growth

Cell count samples of P. subcurvata were fixed with 10% acid lugol solution and stored at 2°C in the dark until counting. Cell numbers of P. subcurvata were enumerated according to the method by [64] using 3 ml sedimentation chambers (Hydrobios, Kiel, Germany) on an inverted microscope (Zeiss Axiovert 200) counting at least 400 cells.

The growth rates μ (d-1) were determined with where N0 and N denote the initial and the final cell concentrations of the experiments, respectively and Δt is the incubation time in days. Final harvest took place when the cells were in exponential growth and reached densities between 67 000 and 107 000 cells mL-1.

The cell volume was computed using the volume formula of a prism on parallelogram base provided by [65]. The apical and transapical axes were measured via microscopy, while the pervalvar axis was estimated to be half of the transapical axis with an average value of 1.2 μm.

Elemental composition

At the end of the experiment, particulate organic carbon (POC) and particulate organic nitrogen (PON) were measured after filtering onto pre-combusted (15 h, 500°C) GF/F filters (pore size ~ 0.6 μm, Whatman). The amount of seawater filtered ranged between 200–300 mL and was dependent on the biomass in the treatments. Filters were stored at -20°C and dried for > 12 h at 60°C. Analysis was performed using a Euro Elemental Analyzer 3000 CHNS-O (HEKAtech GmbH, Wegberg, Germany). At the end of the experiment, samples to determine biogenic silica (BSi) were filtered through a cellulose acetate filter (Sartorius, 0.6 μm) and stored at -20°C. The dried filters were submerged in 0.2 M NaOH at 95°C for 45 minutes, cooled in an ice bath for 15 minutes, neutralized with 1 M HCl according to [66] and analyzed colorimetrically for silicate using standard spectrophotometric techniques [67]. Contents of POC, PON and BSi were corrected for blank measurements and normalized to filtered volume and cell densities to obtain cellular quotas. Production rates of POC, PON and BSi were calculated by multiplying the cellular quotas with the respective growth rate.

Pigments

The amount of seawater filtered to collect pigment ranged between 200–300 mL on the GF/F filter and was dependent on the biomass in the treatments. Each pigment sample was flash frozen in liquid nitrogen and stored at -80°C until analysis. First, the pigments were homogenized and extracted for 24 h in 90% acetone at 4°C in the dark. Second, they were centrifuged for five minutes (4°C, 13000 rpm) and filtered through a 0.45 μm pore size nylon syringe filter (Nalgene, Nalge Nunc International, Rochester, NY, USA). The pigments were analyzed by reversed phase High Performance Liquid Chromatography (HPLC) on a LaChromElite system equipped with a chilled autosampler L-2200 and a DAD detector L- 2450 (VWR-Hitachi International GmbH, Darmstadt, Germany). A SpherisorbODS-2 column (25 cm × 4.6 mm, 5 μm particle size; Waters, Milford, MA, USA) with a LiChropher100-RP-18 guard cartridge was used for the separation of pigments, applying a gradient according to [68]. Peaks of light harvesting (LH) pigments chlorophyll a (Chl a) and c (Chl c), fucoxanthin (Fuco), as well as of the light protective (LP) pigments diatoxanthin (Dt) and diadinoxanthin (Dd) were detected, identified and quantified by co-chromatography with the corresponding pigment standards (DHI Lab Products, Horsholm, Denmark) using the software EZChrom Elite ver. 3.1.3. (Agilent Technologies, Santa Clara, CA, USA). Pigment contents were normalized to filtered volume and cell densities to obtain cellular quotas.

Photophysiological parameters

The efficiency of photochemistry in the PSII of P. subcurvata was assessed regularly during and at the end of the experiment by means of a Fast Repetition Rate fluorometer (FRRf, FastOcean PTX) and a FastAct Laboratory system (both from Chelsea Technologies Group ltd., West Molesey, United Kingdom). Values were obtained using the FastPro8 software (Version 1.0.50), [69]. Measurements were performed at least 2 hours after begin of the light period at 2°C after 10 minutes of dark-adaptation to ensure that all PSII reaction centers were fully oxidized and non-photochemical quenching (NPQ) was relaxed [70]. For each treatment, a 0.2 μm filtered blank was collected, measured and subtracted.

The fluorometer’s LED (wavelength 450 nm) was automatically adjusted to a light intensity of 1.2·1022 photons m-2 s-1. A single turnover flashlet was applied to cumulatively saturate PSII, thus to close all PSII reaction centers, and consisted of 100 flashlets on a 2 μs pitch, followed by a relaxation phase made of 40 flashlets on a 50 μs pitch to reopen the PSII reaction centers. The saturation phase of the single turnover acquisition, comprised 24 sequences and was fitted according to [71]. The minimum (F0) and maximum (F) Chl a fluorescence were determined and the apparent maximum PSII quantum yield (F/F) was calculated according to the equation:

Further outputs of the FastPro8 software from the single turnover measurements of dark-adapted cells were the connectivity between PSII (P, dimensionless), thus the energy transfer between PSII units, the time constant for electron transport at the acceptor side of PSII (τ, μs), the functional absorption cross section of PSII photochemistry (σPSII, nm-2) and the cellular concentration of functional PSII reaction centers (RCII, zmol cell-1).

During the photosynthesis-irradiance-curve (PE-curve), cells were exposed to eight light levels ranging from 0 to 1868 μmol photons m-2 s-1 for five minutes each. At each light level, six measurements of the light-adapted minimum (F′) and maximum (F′) Chl a fluorescence were taken and the effective PSII quantum yield (Fq′/ F′ = (F′—F′)/ F′) was calculated [72].

Cellular electron transport rates (cETR) were calculated following [73, 74] and normalized by RCII [75] using: where E (photons m-2 s-1) is the applied instantaneous irradiance, which was measured separately for each light level in seawater.

The cETR versus E curve was fitted according to [76] allowing to derive the maximum cETR (cETR), the minimum saturating irradiance (I) determined by the interception of the light-limited region with the maximum photosynthetic rate, and the maximum light utilization efficiency (α).

NPQ of Chl a fluorescence was calculated using the Stern-Volmer equation [77] at each light level:

Statistical assessment

To assess the effect of Fe concentration (Control and +Fe) and pCO2 (190 and 290) on all experimental parameters among the different treatments of P. subcurvata, we used a two-way analysis of variance (2-way ANOVA) followed by a pairwise multiple comparison test (post hoc) using the Holm-Sidak method. All statistical analyses and the curve fittings were performed using the program SigmaPlot (Version 13.0 from Systat Software, Inc., San Jose California USA, www.systatsoftware.com). Statistical significance was defined when p < 0.05.

Results

Trace metal and carbonate chemistry

The total dFe concentrations of the different culture medium showed a significant difference between the +Fe and the Control treatments (2-way ANOVA: p < 0.001, Table 1), with the +Fe treatments having higher dFe concentrations than the Control treatments. The parameters of the carbonate system are given in Table 1. TA remained constant in all culture media and incubation bottles. As expected, increasing pCO2 significantly enhanced the DIC concentration in all culture media and incubation bottles (2-way ANOVA: p < 0.001; post hoc +Fe: p < 0.001; post hoc Control: p = 0.005). While Fe availability did not alter DIC of the different culture media bottles, a significant Fe effect was found for the P. subcurvata incubations, but only for the 190 treatments (post hoc: p < 0.04). The interaction of CO2 and Fe also led to significant effects in DIC of the P. subcurvata incubations (2-way ANOVA: p < 0.02). As expected, the pCO2 and DIC in all of the 290 treatments were significantly higher than in the 190 treatments (2-way ANOVA; p<0.001; Table 1). Biologically driven changes to the carbonate chemistry were ruled out since TA, DIC, and pCO2 values did not differ between the abiotic culture medium and the corresponding P. subcurvata incubations for each treatment at the end of the experiment (Table 1).

Table 1

Total dissolved iron (dFe) concentrations and carbonate chemistry determined at the end of the experiment in the culture medium (filtered seawater without cells) and the P. subcurvata incubations of the four treatments (+Fe 190, Control 190, +Fe 290 and Control 290).

The pCO2 was calculated from measured dissolved inorganic carbon (DIC) and total alkalinity (TA). For the culture medium, dFe, TA, DIC and pCO2 values represent the range of duplicate abiotic controls. TA, DIC and pCO2 values of the P. subcurvata incubations represent the means ± SD (n = 3). Differences between the individual treatments of the P. subcurvata incubations were determined with post hoc tests, where significant statistical (p < 0.05) differences are denoted by different letters.

Parameter	Culture medium
	190		290
	+Fe	Control	+Fe	Control
dFe (nmol L-1)	2.92–3.10	0.94–1.07	1.36–1.41	0.37–0.50
TA (μmol kg-1)	2308–2318	2304–2319	2304–2323	2302–2311
DIC (μmol kg-1)	2077–2101	2058–2077	2125–2131	2131–2132
pCO2 (μatm)	208–249	201–208	269–308	296–309
	P. subcurvata incubations
Parameter	190		290
	+Fe	Control	+Fe	Control
dFe (nmol L-1)	-	-	-	-
TA (μmol kg-1)	2317 ± 11 a	2326 ± 9 a	2327 ± 13 a	2320 ± 9 a
DIC (μmol kg-1)	2046 ± 17 a	2071 ± 11 b	2138 ± 14 c	2118 ± 2 c
pCO2 (μatm)	181 ± 15 a	202 ± 24 a	287 ± 31 b	283 ± 29 b

Growth and elemental composition

The growth rates of P. subcurvata were unaffected by Fe deficiency and changes in pCO2 (Fig 1A). Similarly, cell volumes remained constant across all treatments (Table 2).

Fig 1

Effects of Fe reduction (+Fe vs Control) and pCO2 increase (190 vs 290) on (A) growth rate (μ), (B) POC production, (C) PON production and (D) BSi production in the four treatments of P. subcurvata (+Fe 190, Control 190, +Fe 290 and Control 290) at the end of the experiment. The values represent the means ± SD (n = 3). Different letters indicate significant differences between treatments (p < 0.05).

Table 2

Volume and elemental composition determined at the end of the experiment in the four treatments of P. subcurvata (+Fe 190, Control 190, +Fe 290 and Control 290).

The values represent the means ± SD (n = 3). Different letters indicate significant differences between treatments (p < 0.05).

Parameter	P. subcurvata incubations
	190		290
	+Fe	Control	+Fe	Control
Volume (μm3)	31 ± 11 a	34 ± 13 a	34 ± 16 a	32 ± 18 a
POC (pg C cell−1)	12.8 ± 0.9 a	11.7 ± 0.8 a	9.4 ± 0.9 b	8.2 ± 0.6 b
PON (pg N cell−1)	2.1 ± 0.1 b	1.7 ± 0.1 a	1.5 ± 0.2 a	1.5 ± 0.2 a
C:N (mol mol−1)	7.2 ± 0.6 a	8.1 ± 0.2 b	7.4 ± 0.4 a	6.9 ± 0.1 a
BSi (pg Si cell−1)	2.6 ± 0.2 a	2.8 ± 0.4 a	2.6 ± 0.2 a	3.1 ± 0.5 a

Cellular POC quotas (Table 2) and POC production rates (Fig 1B) in both pCO2 treatments were not affected by Fe deficiency. On the other hand, the increase of CO2 concentration resulted in a 20–30% decrease of cellular POC quotas (2-way ANOVA: p < 0.001; Table 2) and POC production (2-way ANOVA: p < 0.001; Fig 1B) in both Control and +Fe treatments.

At 190, lowered Fe concentration led to a decrease of cellular PON concentrations by 19% (post hoc: p < 0.03), while no Fe effect was observed at 290. In response to increasing pCO2, the cellular PON concentration was strongly reduced (2-way ANOVA: p = 0.005; Table 2) in the +Fe (post hoc: p < 0.004), but not in the Control treatments (Table 2). The PON production (Fig 1C) followed the same pattern as cellular PON quotas, showing a significant decrease of 15% with reduced Fe availability in the 190 treatments (post hoc: p < 0.03), while remaining constant in the 290 treatments. With increasing pCO2, a loss of 26% in PON production in the +Fe (post hoc: p < 0.02), but not in the Control treatments was observed, resulting from an interactive effect of Fe and CO2 availability (2-way ANOVA: p < 0.02; Fig 1C).

Molar C:N ratios ranged between 6.9 ± 0.1 and 8.1 ± 0.2 mol mol−1. Fe deficiency led to a 13% increase in the C:N ratio in the 190 treatments (post hoc: p < 0.04), while no such Fe effect was observed in the 290 treatments. Furthermore, the increase of CO2 concentration resulted in a decline of C:N by 15% in the Control (post hoc: p < 0.02), but not in the +Fe treatments. The interaction of Fe and CO2 altered C:N ratios significantly (2-way ANOVA: p < 0.03; Table 2).

Neither low Fe concentrations nor increased pCO2 changed the cellular BSi quota (Table 2). However, as a result of Fe deficiency the BSi production in 290 significantly increased by 35% (2-way ANOVA: p = 0.007; post hoc: p = 0.006; Fig 1D), but not in 190. A response to higher pCO2 resulted in higher BSi production only in the Control treatments (post hoc: p < 0.04).

Pigment composition

All quantified pigments, except for Chl c, were significantly affected by Fe deficiency in either the 190 or the 290 treatments (2-way ANOVA: Chl a p < 0.001; Fuco p < 0.02; Dd p < 0.02; Dt p < 0.02; Fig 2A and Table 3). At 190, reduced Fe availability resulted in a decrease of Chl a by 37% (post hoc: p = 0.002), of Fuco by 34% (post hoc: p < 0.02) and of Dd by 29% (post hoc: p = 0.03), while Dt was not affected. At 290, the reduction of Fe significantly reduced the Chl a concentration by 23% (post hoc: p = 0.03) and Dt by 60% (post hoc: p = 0.007), whereas Fuco and Dd remained constant. In response to elevated pCO2, cellular Chl a quotas of P. subcurvata were significantly reduced in the +Fe (251 ± 17 to 192 ± 19 fg cell-1 for 190 and 290, respectively; 2-way ANOVA: p < 0.03; post hoc: p = 0.02; Fig 2A), while this trend was absent in the Control. No other pigments (Fuco, Chl c, Dd or Dt) responded to changes in the pCO2.

Fig 2

Effects of Fe deficiency and pCO2 increase on (A) chlorophyll a (Chl a), (B) photosynthetic yields (Fv/Fm), (C) functional absorption cross sections (σPSII) and (D) time constants (τ) in the four treatments of P. subcurvata (+Fe 190, Control 190, +Fe 290 and Control 290) at the end of the experiment. The values represent the means ± SD (n = 3). Different letters indicate significant differences between treatments (p < 0.05).

Table 3

Pigment concentrations determined at the end of the experiment in the four treatments of P. subcurvata (+Fe 190, Control 190, +Fe 290 and Control 290).

The values represent the means ± SD (n = 3). Different letters indicate significant differences between treatments.

Parameter	P. subcurvata incubations
	190		290
	+Fe	Control	+Fe	Control
Chlorophyll c2 (fg cell-1)	28.9 ± 6.9 a	19.4 ± 5.7 a	22.3 ± 5.1 a	19.1 ± 5.0 a
Fucoxanthin (fg cell-1)	140 ± 10 a	93 ± 24 b	110 ± 16 a	86 ± 24 a,b
Diadinoxanthin (fg cell-1)	28.3 ± 3.8 a	19.4 ± 4.5 b	24.1 ± 4.3 a	17.7 ± 3.9 a,b
Diatoxanthin (fg cell-1)	1.27 ± 0.24 a	1.08 ± 0.12 a,b	1.47 ± 0.44 a	0.64 ± 0.21 b
Chl a:C (mol mol−1)	0.21 ± 0.03 a	0.14 ± 0.02 b	0.22 ± 0.04 a	0.16 ± 0.03 b

The Chl a:C ratio in P. subcurvata was significantly affected by Fe deficiency (2-way ANOVA: p = 0.005; Table 3) leading to a decrease of 33% (post hoc: p < 0.03) and 27% (post hoc: p < 0.04) in the 190 and 290 treatments, respectively. Conversely, increased pCO2 had no effect on the Chl a:C ratio.

Maximum quantum yield and changes to PSII

The photosynthetic yield of P. subcurvata (F/F) showed a significant Fe effect (2-way ANOVA: p < 0.001; Fig 2B). At 190, F/F decreased significantly by 21% in response to Fe deficiency (from 0.52 ± 0.01 to 0.41 ± 0.02 in the +Fe and Control, respectively, post hoc: p < 0.001), while no Fe effect was observed in 290. Interestingly, CO2 enhancement differently affected the photosynthetic yield of the two Fe treatments. While increasing pCO2 enhanced the F/F in the Control treatment by 15% (from 0.41 ± 0.02 to 0.47 ± 0.01, post hoc: p = 0.005), it reduced F/F in the +Fe treatments by 8% (from 0.52 ± 0.01 to 0.48 ± 0.01, post hoc: p < 0.04). Hence, there was a significant interactive effect of CO2 and Fe availability on F/F (2-way ANOVA: p = 0.002; Fig 2B).

The connectivity (P) was significantly affected by Fe deficiency (2-way ANOVA: p = 0.002; Table 4), with the Control treatment having an 11% smaller energy transfer between PSII units than the +Fe at 190 (post hoc: p = 0.002). In the 290 treatments, a similar, however, not significant, decreasing trend was seen. In contrast, no response of P to increased CO2 was observed.

Table 4

Connectivity (P), cellular concentration of functional PSII reaction centers (RCII), light utilization efficiency at low irradiance (α), maximum cellular electron transport rate (cETRmax) and minimum saturating irradiance (Ik,) of P. subcurvata in the four treatments (+Fe 190, Control 190, +Fe 290 and Control 290) at the end of the experiment.

The values represent the means ± SD (n = 3). Different letters indicate significant differences between treatments (p < 0.05).

Parameter	P. subcurvata incubations
	190		290
	+Fe	Control	+Fe	Control
P (rel. unit)	0.44 ± 0.01 a	0.39 ± 0.02 b	0.43 ± 0.01 a	0.40 ± 0.01 a,b
RCII (zmol cell−1)	515 ± 58 a	525 ± 42 a	370 ± 38 b	519 ± 47 a
α (amol e− cell−1 s−1/ μmol photons m−2 s−1)	0.75 ± 0.13 a	0.97 ± 0.14 b	0.58 ± 0.08 a	0.82 ± 0.07 b
cETRmax (amol e− cell−1 s−1)	119 ± 21 a	165 ± 26 b	85 ± 5 a	139 ± 19 b
Ik (μmol photons m−2 s−1)	155 ± 9 a	171 ± 11 a,b	143 ± 15 a	169 ± 9 b

The functional absorption cross section of PSII (σPSII) showed a significant effect to Fe deficiency (2-way ANOVA: p < 0.001; Fig 2C). While σPSII increased by 26% with reduced Fe availability in 190 (from 2.47 ± 0.03 to 3.11 ± 0.18 nm-2, respectively, post hoc: p < 0.001), this Fe effect was not seen in the 290 treatments. Furthermore, only in the Control treatments σPSII was reduced by 10% from 3.11 ± 0.21 to 2.79 ± 0.09 nm-2 between 190 and 290, respectively (post hoc: p = 0.01). Moreover, there was a synergistic effect between Fe and CO2 on σPSII (2-way ANOVA: p = 0.009; Fig 2C).

The cellular concentration of functional PSII reaction centers (RCII) was significantly altered by Fe deficiency (2-way ANOVA: p < 0.04; Table 4). This effect was only seen in 290, where RCII increased by 29% (post hoc: p < 0.02). Increasing CO2 significantly reduced the RCII concentration (2-way ANOVA: p < 0.05), but only in the +Fe treatments (post hoc: p < 0.02).

Fe deficiency differently influenced the time constant for electron transport at the acceptor of PSII (τ) in the two CO2 treatments. While lower Fe concentration reduced τ when grown at 190 μatm pCO2 (post hoc: p < 0.001), it was enhanced at 290 μatm pCO2 (post hoc: p = 0.006; Fig 2D). The effect of increased CO2 on τ was significant (2-way ANOVA: p < 0.004). In the Control treatments, τ increased from 548 ± 21 to 659 ± 23 μs from 190 to 290 μatm pCO2 (post hoc: p < 0.001) while it remained constant in the +Fe treatments. Hence, there was a strong interactive effect of Fe and CO2 on τ apparent (2-way ANOVA: p < 0.001).

PE-curve

The cellular electron transport rates (cETR) of all treatments followed the shape of a typical PE-curve (Fig 3A). The light utilization efficiency of P. subcurvata at low irradiance (α) was significantly affected by Fe deficiency (2-way ANOVA: p = 0.005; Table 4), with α increasing by 29% at 190 (post hoc: p < 0.04) and by 41% at 290 (post hoc: p < 0.02). A CO2 effect was also observed (2-way ANOVA: p = 0.02), where increased CO2 reduced α, but due to large uncertainties, the individual post hoc tests of the +Fe and Control treatments were not significant. In response to Fe deficiency, cETRmax (Table 4 and Fig 3A) was significantly enhanced (2-way ANOVA: p < 0.006) by 39% at 190 and by 64% at 290 (both post hoc: p < 0.03). The increase in CO2, however, did not lead to significant changes in cETRmax. The minimum saturating irradiance (Ik) displayed a significant Fe effect (2-way ANOVA: p < 0.02; Table 4), where Ik increased by 10% in the 290 treatments (post hoc: p < 0.04). Although not significant (p>0.05), in the 190 treatments a similar trend was observed. Ik remained unchanged by increasing CO2 irrespective of Fe availability.

Fig 3

Effects of Fe deficiency and CO2 increase on (A) cellular electron transport rates (cETR) and on (B) non-photochemical quenching (NPQ) in the four treatments with P. subcurvata (+Fe 190, Control 190, +Fe 290 and Control 290) at the end of the experiment. The values represent the means ± SD (n = 3).

The non-photochemical quenching of all treatments was similarly low at low irradiances (Fig 3B). Exposed to irradiances higher than 350 μmol photons m-2 s-1, the NPQ in P. subcurvata increased nearly linearly and then leveled off between ~1.5 and 2.5 for all treatments. No Fe or CO2 effect on NPQ was observed in any treatment.

Discussion

The ‘Iron Hypothesis’ suggests that the fertilization of the SO by increased dust deposition in glacial times promoted growth and productivity of phytoplankton. The biological pump in the SO was thus hypothesized to have reduced atmospheric pCO2. In this study, we assessed the ecophysiological response of P. subcurvata simulating glacial and interglacial climate scenarios in terms of changes in Fe and CO2 availability. It is important to note that while we manipulated two of the main environmental parameters, Fe concentrations and pCO2, other parameters (macronutrient concentrations, temperature i.e.) likely also differed between the glacial and interglacial ocean. For this study, however, the focus was on the interactive effects of Fe and pCO2.

Glacial conditions favored POC production by P. subcurvata

Between 190 and 290 μatm pCO2, no change in growth rate was observed in the +Fe treatments of P. subcurvata (Fig 1A). Previous laboratory studies with cultures of the same P. subcurvata strain also reported no changes in growth rate between 180 and 390 μatm pCO2 [51]. Similarly, growth remained unaffected in the temperate Pseudo-nitzschia pseudodelicatissima between 200 and 380 μatm pCO2 [78], in T. pseudonana, T. rotula, and T. oceanica from 230 to 350 ppm [79] and in Proboscia alata from 135 to 200 μatm pCO2 [47]. Additionally, growth rates, pigment contents, photosynthesis and photophysiology of the Antarctic diatom Chaetoceros brevis did not change between 190 and 750 ppm [80]. Differently, however, is the study by [46], which reported a stimulation of the growth rate of another P. subcurvata strain from 100 up to 450 μatm pCO2. Also, growth of the temperate Pseudo-nitzschia multiseries was enhanced between 220 and 400 ppm pCO2 [48]. It appears therefore that species- and strain-specific differences in the CO2-dependence of growth among Pseudo-nitzschia exist.

The similar growth rates at both pCO2 levels and Fe availabilities maintained by P. subcurvata in our experiment (Fig 1A) suggest the operation of carbon concentrating mechanisms (CCMs), which efficiently avoided CO2 limitation. This can also be inferred from [50], where Pseudo-nitzschia was the most abundant species within a natural Southern Ocean phytoplankton assemblage under both Fe-enriched and Fe-deplete conditions at 180 and 390 μatm. Previous studies showed that Antarctic phytoplankton species such as P. subcurvata operate very efficient CCMs, which are constitutively expressed irrespective of CO2 availability [49, 51, 81]. In addition to highest uptake rates of C and macronutrients, the temperate diatom P. pseudodelicatissima exhibited a high Fe uptake affinity at 170 ppm [78]. The latter findings indicate that Pseudo-nitzschia species can cope well with low CO2 conditions, enabling them to maintain high growth even under low CO2 conditions, as can be also seen here in P. subcurvata (Fig 1A).

In this experiment, Fv/Fm was highest in the +Fe 190 treatment (Fig 2B), indicating that P. subcurvata possessed highest photochemical fitness under simulated glacial conditions. With increasing pCO2, however, Fv/Fm declined in the diatom, but only when Fe was added (Fig 2B). Such a negative CO2 effect in Fe-enriched conditions was also observed in the Chl a content (Fig 2A) and the number of functional RCII (Table 4). Indeed, P. subcurvata cells grown in the +Fe 290 treatment had a lower Chl a content compared to ones in the +Fe 190 treatment (Fig 2A), although the Chl a:C ratios were similar.

Moreover, cellular BSi quotas and production remained constant with increasing pCO2 in the +Fe treatments (Fig 1D, Table 2) while a decline in POC and PON quotas as well as in POC and PON production rates (Fig 1B and 1C and Table 2) was found. Reducing both POC and PON quotas, P. subcurvata was able to maintain a constant C:N ratio (Table 2) in response to increasing pCO2 under Fe-enriched conditions. Considering, however, that cETRs remained similar between 190 and 290 (Fig 3A, Table 4), a reduction in POC and PON contents indicates that the contribution of linear electron transport was reduced while cycling of electron via alternative pathways was required to avoid excess light energy. These physiological characteristics resemble those observed in various field incubation experiments under ocean acidification conditions and indicate that P. subcurvata struggles when exposed to high pCO2 levels [24, 49, 50]. Overall, we can conclude that glacial conditions simulated by a low pCO2 of 190 μatm together with Fe enrichment was neither limiting growth nor POC production of P. subcurvata. On the contrary, these conditions were beneficial for biomass production and photochemical fitness of the diatom.

P. subcurvata adjusted its physiological machinery to cope with low Fe supply

Contrary to other studies, we did not observe a decrease in cell volume of P. subcurvata grown with decreasing Fe availability (Table 2) [42, 82]. This may have been masked by the fact that the P. subcurvata strain used in our experiment was acclimated to low Fe conditions for a long time. Indeed, it exhibited large and elongated cells compared to the much shorter cells of the stock culture grown in the Fe-rich F2 medium (12 μM Fe), thus increasing their surface area-to-volume ratio. Furthermore, this strain was isolated from open ocean waters in the Atlantic sector of the SO. It is well known that oceanic diatoms acclimate to Fe limitation by increasing their surface area-to-volume ratio in order to maximize the number of transporter sites and nutrient uptake kinetics [83, 84].

Many studies reported a decrease in growth rate with decreasing Fe availability [8, 39, 40, 42, 44, 85–89]. Nonetheless, some of them also observed that particular oceanic diatoms grew at comparable rates under high and low Fe conditions [8, 86], as they have evolved acclimation strategies to reduce their Fe requirement. In our experiment, the growth rate of the oceanic P. subcurvata also displayed no difference between +Fe and Control conditions at the two pCO2 levels tested (Fig 1A) [42]. Suggested that the response of physiological and biochemical parameters to Fe reduction precedes changes in growth rate. This may explain why we did not see a decrease in growth rate here, despite observing typical responses to Fe-limiting conditions as substantial reductions in photochemical quantum efficiency (Fig 2B), connectivity (Table 4) and Chl a content (Fig 2A) accompanied by large functional absorption cross sections (Fig 2C) [9, 38, 40, 41, 85, 87, 88, 90, 91].

Under Fe deficiency, lowered Fv/Fm values indicate that the excitation energy was less efficiently transferred in the antennae, due to damaged and altered parts of the photosynthetic apparatus [41]. A decrease in Fv/Fm was commonly observed in cells grown in Fe-poor environments [9, 40, 41, 44, 87, 88] and, as expected, we observed this trend in the Control 190 treatment of P. subcurvata (Fig 2B). In line with the tested P. subcurvata here, oceanic Pseudo-nitzschia species usually decouple Fv/Fm and growth rate, reducing the former while maintaining the latter [8]. This decoupling was suggested to be due to either a low energy requirement of the diatom, or a compensating mechanism that generates reducing power, thus supporting rapid growth [8].

The decrease in Fv/Fm (Fig 2B) and lowered connectivity (P, Table 4) at low pCO2 in the low Fe P. subcurvata cells indicate that the transfer of excitation energy to the reaction centers was compromised [9]. Because Fe deficiency affects the synthesis and thus cellular content of Chl a, as seen in our data (Fig 2A), light harvesting may become more difficult for the cell. While [85] held lowered pigment concentration during Fe starvation responsible for a decline in photosynthesis, we did not observe reduced POC production rates (Fig 1B). Rather P. subcurvata compensated for a low Chl a content by increasing the functional absorption cross section of PSII (σPSII), which is a measure of the target area of the light harvesting antenna (Fig 2C). In response to Fe deficiency this strategy can reduce the Fe demand and keep up the same capacity of the cell to absorb light [92]. Our results agree with literature showing an increase in σPSII with Fe reduction [9, 38, 40, 41, 75, 87–89].

These photophysiological adjustments, however, did not prevent changes in light absorption completely, as shown by the strongly impacted light use capacities of Fe-limited P. subcurvata (Table 4). Higher α values were found under Fe deficiency for both 190 and 290 treatments, indicating that cells were able to respond better to lower irradiances than Fe-replete cells. Surprisingly, this effect was not always observed for Ik values of P. subcurvata, which remained similar at 190 and was slightly higher at 290 (Table 4). Thus, while Fe deficiency at 290 resulted in a more efficient light utilization at lower irradiances (higher α), the cells required more light (higher Ik) in order to cover their photosynthetic requirement [40]. In other studies, Ik either decreased [40, 93] or remained unchanged [44, 85] under Fe reduction.

Even though POC-fixation remained constant under Fe deficiency (Fig 1B), cETRmax (Fig 3A, Table 4) and RCII concentration (only seen at 190 μatm pCO2) were enhanced (Table 4), indicating similar linear electron transport, but also cycling of electrons into alternative pathways such as cyclic electron flow within PSII [94] or Mehler reaction [95]. Considering, however, that the latter pathways are Fe-expensive, other pathways such as activity of a putative plastid plastoquinol terminal oxidase (PTOX) seem more plausible [96]. In support for this, [9] also observed constant C assimilation, but enhanced electron transport with Fe limitation in open ocean phytoplankton. Furthermore, a quicker turnover time at the acceptor side of PSII (τ) was found at 190 μatm pCO2 in the Fe deficient P. subcurvata cells (Fig 2D), supporting PTOX activity, as previously observed for the Fe-limited Antarctic diatom Chaetoceros debilis [44]. Interestingly, this was not reflected in higher NPQ activities (Fig 3B).

At low pCO2, BSi quotas and production rates of P. subcurvata remained unaltered in response to Fe deficiency (Fig 1D, Table 2), as previously observed in Chaetoceros debilis [44], Corethron pennatum [97] and Chaetoceros dichaeta [98]. Considering the importance of Fe in C and N assimilation pathways, many studies reported a decrease in C and N under Fe deficiency [40, 42]. In [99], the C quota per cell volume ranged between 0.02 and 0.03 pg μm-3 and was similar between Fe-replete and Fe-deficient treatments in the oceanic Pseudo-nitzschia fraudulenta, P. heimii, P. inflatula and P. turgidula, as well as in the coastal species P. multiseries and P. pseudodelicatissima. This matches with our results for the two tested pCO2 levels (POC per cell volume at 190 +Fe: 0.041±0.002 pg μm-3, Control: 0.035±0.005 pg μm-3 and at 290 +Fe: 0.029±0.005 pg μm-3, Control: 0.027±0.003 pg μm-3). The C:N ratio of diatoms was reported to increase [78], decrease [100] or remain unchanged [42, 44] with reduced Fe availability. We observed an increase in the C:N ratio in response to Fe deficiency at 190 μatm pCO2 (Table 2). In this case, POC quotas remained constant, whereas PON cell quotas decreased with Fe deficiency (Table 2). Literature showed that Fe limitation can affect the supply of ‘new nitrogen’ to the cell as Fe is needed in some N-rich enzymes [101, 102]. [75] observed less abundant transcripts for nitrite reductase under Fe limiting conditions in Phaeocystis antarctica. Considering this, our reduced PON-fixation in P. subcurvata under low Fe conditions in conjunction with low pCO2 could be coupled to a protein recycling process to avoid N-limitation [39, 75, 103].

We can conclude that Fe deficiency results in a less efficient transfer of excitation energy in P. subcurvata, allowing it to reduce its Fe demand. In order to keep up the same POC production, P. subcurvata needed to rely on alternative electron pathways such as cyclic electron flow as well as PTOX activity to prevent over-excitation.

Increased pCO2 weakened the effects of low Fe supply, but did not promote biomass build up

Previous experiments with Pseudo-nitzschia demonstrated on the one hand, that the cell volume of P. pseudodelicatissima increased significantly as pCO2 decreased, while, on the other hand, cell volume was found to decrease with decreasing Fe availability [42, 82]. In our experiments, the cell volume of P. subcurvata did not decrease with reduced Fe availability and increased pCO2 (Table 2), potentially due to a counteracting effect of both factors together. Moreover, Fv/Fm decreased in response to Fe reduction at 190 (Fig 2B), while such Fe-dependent decrease in Fv/Fm was not observed at 290. This indicates that increasing pCO2 had a positive effect on the maximum photochemical efficiency of low Fe P. subcurvata cells. A similar effect by high CO2 concentration was also found for σPSII in Fe-deplete cells, being much smaller (Fig 2C). Apparently, these positive CO2 effects weakened the strong Fe reduction effects previously observed at 190. Such positive response did, however, not translate into more efficient energy transfer from photochemistry to biomass production. In fact, re-oxidation of the primary electron acceptor Qa of low Fe cells was strongly compromised at 290 (Fig 2D). This was associated with reduced POC fixation and enhanced cETRs at 290 (Fig 3A, Table 4), and as a consequence, alternative electron acceptors were required. Due to a synergetic effect of reduced Fe availability and increased pCO2, in our experiment we observed the highest BSi production in low Fe high pCO2 conditions (Fig 1D). This increase in BSi production with reduced Fe concentrations at 290 hints towards stronger silicification and the production of thicker shells by P. subcurvata [104-106].

Conclusion: Glacial vs. interglacial

In our study, in a simulated Fe-fertilized glacial ocean (+Fe 190), P. subcurvata displayed similar growth rates as in interglacial ocean conditions (Control 290), despite lower Fe availability, hinting towards an efficient acclimation strategy to reduce the Fe requirement. Under glacial conditions, electrons were more efficiently channeled, leading to higher cellular POC and PON concentrations and production rates. In comparison, the interglacial conditions with higher pCO2 and reduced Fe availability resulted in reduced POC buildup of the diatom. Thus, we observed that both higher Fe availability and lower CO2 concentration as in the glacial ocean, promoted POC production by P. subcurvata. Assuming that P. subcurvata dominated phytoplankton blooms in the SO during glacial and interglacial times, we can conclude that P. subcurvata contributed more to primary production in the glacial than interglacial ocean. The higher POC production rates by the diatom under glacial conditions facilitated higher CO2 uptake from the atmosphere and potentially higher C export. This matches the ‘Iron Hypothesis’ of [1], which states that in the last glacial maximum higher Fe input from dust fertilized the SO, thus stimulating higher primary production and reducing thereby the atmospheric CO2 concentration. On the other hand, however, the thicker shells of P. subcurvata under the simulated interglacial conditions hint towards reduced grazing and thus its higher contribution to C export [107]. Biogeochemical cycles changed in the past and will change in response to future global climate change. Thus, understanding the dynamic interactions of the ocean’s biogeochemistry and phytoplankton is important in order to better simulate past and future climatic scenarios.

Cellular trace metal quotas.

Trace metal (TM) quotas without oxalate (Total TM content) and with oxalate wash (Intracellular TM content) determined at the end of the experiment in the four treatments of P. subcurvata (+Fe 190, Control 190, +Fe 290 and Control 290). The values represent the means ± SD (n = 3). Different letters indicate significant (p < 0.05) differences between treatments.

(DOCX)

Click here for additional data file.

30 Sep 2021

PONE-D-21-25243The Southern Ocean diatom Pseudo-nitzschia subcurvata flourishes better under simulated glacial than interglacial ocean conditionsPLOS ONE

Dear Dr. Koch,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration by two expert reviewers, we feel that it has considerable merit but does not yet fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Reviewer 1 asks for any evidence that the study diatom dominated the Southern Ocean over glacial and inter-glacial times.  Reviewer 1 also notes a subtle issue with estimating C assimilation rates from cellular C content which should be addressed carefully.  Reviewer 1 also offers some corrections of typos and points of clarification. Reviewer 2 points out issues with the presented metric of Non-PhotoChemical Quenching and suggests alternat or parallel presentation of a different metric (NPQ  NSV) which has been validated in similar studies. Reviewer 2 calls for clarification of exactly how the fluorescence metrics were captured, and offers a correction to the discussion of IKE. I am delighted your strong work can benefit from such constructive reviews, which I hope you find useful.

Please submit your revised manuscript by Nov 14 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Douglas A. Campbell, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

3. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

4. Thank you for stating the following in the Acknowledgments Section of your manuscript:

"This work was supported by German Federal Ministry of Education and Research (BMBF) as Research for Sustainability initiative (FONA); www.fona.de through Palmod project (FKZ: 01LP1505C). We thank (in alphabetical order) T. Brenneis, C. Völkner and D. Wilhelms-Dick for laboratory assistance and for analyzing the samples. Thanks also to K. Bischof and B. Meyer-Schlosser for the pigment analysis. ST, FK and FP were funded by the Helmholtz Association (HGF Young Investigators Group EcoTrace, VH-NG-901)."

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

"AP was supported by German Federal Ministry of Education and Research (BMBF) as Research for Sustainability initiative (FONA); www.fona.de through Palmod project (FKZ: 01LP1505C). ST, FK and FP were funded by the Helmholtz Association (HGF Young Investigators Group EcoTrace, VH-NG-901).

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

5. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this well-written paper, the authors reported the ecophysiological responses of diatom Pseudo-nitzschia subcurvata to the simulated glacial and interglacial climatic conditions i.e., no Fe or Fe addition under 190 and 290 μatm pCO2 levels. They found the varied Fe and pCO2 levels had a limited effect on growth rate, but individually or interactively altered the cell compositions including pigments, POC, PON and BSi, as well as photophysiology of P. subcurvata. They conclude the combined higher Fe availability and lower pCO2 (present in glacial ocean) was beneficial for the diatom P. subcurvata, thus contributing more to primary production during glacial than interglacial times. There are some minor revisions should be considered before accepted for publication in PLoS One journal.

I think the title exaggerates the results of the experiment; because, the authors just simulated two environmental factors i.e., Fe and pCO2 levels in glacial and interglacial times, that cannot represent all the environmental factors.

In Discussion part, the authors fully explained the data in the view of physiology, some ecological perspectives should be given as they did the simulated field condition experiments.

1. Line 26, The ecological significance of “The thicker silica shells present under interglacial conditions” has been referred in the end of the Abstract; so, this sentence “which might offer better protection against grazers” should be removed. Besides, the authors did not give any descriptions about grazers throughout the manuscript.

2. Is there any geological evidence to show the Pseudo-nitzschia species dominate the Southern Oceans in glacial and inter-glacial times? I think some information should mentioned in the Introduction part.

3. One of half brackets was missing in several places of the manuscript, e.g., Lines 105-106, Line 209, Line 220, Line 239;

4. In Materials and Methods part, clarify how much volume of culture was filtrated to measure the POC, PON, BSi and pigments.

5. In Line 195-196, the authors clarified the cell densities at the end of the cultivation were 67000 to 107000 cells per ml; I’m wondering what the cell density was at the initial inoculation. Please clarify it.

6. Line 280, The light intensity of fluorometer is so high. Please check it.

7. Line 252-253, Production rates of POC, PON and BSi were calculated by multiplying the cellular quotas with the respective growth rate.

It’s OK to calculate PON and BSi production rate. But I don’t think it’s right to use this way to calculate POC production rate; because, the cultures were grown under 16:8 light:dark cycle under 100 µmol photons m-2 s-1 light intensity, that means the POC production by photosynthesis just occurs in light duration (16 h per day), while the growth rate was calculated from the cell density changes of a whole day (24 h). So, they are mismatching.

8. Line 337-338, The authors clarified that “dFe values denote two measurements”; so, it is wrong to use the standard deviation here. I suggest to replace the SD to half of range.

Reviewer #2: The following manuscript provides details on the interactive effects of Fe and pCO2 on an Antarctic diatom, Pseudo-nitzschia subcurvata, under glacial and inter-glacial conditions. Whilst there is a wealth of evidence of understanding higher pCO2 concentrations, representative of future climate change, very little is known about how this diatom copes under these pre-industrial conditions. The manuscript is very well written, with clean concise results and figures. I recommend this paper for publication, with only minor corrections which are outlined below.

Minor Corrections:

Line 248: Should the NaOH concentration not be “M”, currently it is “N”.

Line 276: Were the samples collected during the day or night part of the light cycle? This will greatly impact the effects of dark adaptation (see Schuback et al. 2021 Frontiers for more details on this).

Line 284: No text is provided to determine whether blanks were collected, and the fluorescence subsequently corrected. This is an important step for any active chlorophyll fluorescence measurements.

Line 294: Please provide the number of measurements per light level. Was any QC applied to these measurements per light level? As stated in reference 76 (Ralph and Gademann) unlike traditional PE curves, the measurements do not reach steady state. That is why previous studies have opted for either taking the mean of the last 3 measurements per light level or applying a QC to remove potential outliers.

Line 302: More details are required here to indicate how the curve fitting was done. Was this done in FastPro8 or was another program used to fit the curves.

Line 306: Reference required for the Stern-Volner equation – Bilger and Bjorkman, 1991, doi 10.1007/BF00197951.

Line 329: This statement is a little confusing, based upon the significant differences above. Are you saying that the CO2 parameters are significantly different between 190 and 290? Or are you saying they are significantly different between the Fe and control? If the latter, what impact do you think it will have on the results if the CO2 parameters are also significantly different alongside the Fe concentrations.

Line 339: Unless I have missed it, I am struggling to find where the definition of the letters is stated. This makes it difficult to understand what was significant or not.

From my understanding:

a = all the same

b = significantly different to a

c = significantly different to a and b

Perhaps you can make it a little clearer for the reader.

Table 1: dFe concentrations – do you know what may have caused such differences? Additionally, a 1 nM concentration for the 190 control would not necessarily be limiting for SO phytoplankton species based upon measured in situ concentrations.

Line 378: Do you mean “but not in 190”?

Line 416: The connectivity parameter is most often reported as the Greek letter rho, ρ.

Table 4: Units of α, should this not be amol e- cell-1 s-1 (μmol photons m-2 s-1)-1?

Line 458: This does not make sense to me, in the 190 treatment the standard deviations are 9 and 11, whereas in the 290 treatments they are 15 and 9. So how is it that the large standard deviations in the 190 prevent statistical significance?

Line 505: I wouldn't necessarily agree with this statement. The Chl:C ratios are similar for both Fe treatments at 190 and 290. As well as σPSII being similar between Fe and control at 290. These are better indicators of potential light absorption for photosynthesis.

Line 540: Missing some more recent work: Strzepek et al. 2019 doi: 10.1073/pnas.1810886116

Line 553: See comment above - I do not think you can solely use Chl-a concentrations to infer reduced light absorption.

Line 559: Your statements above do not agree with this. Here you state clearly mechanisms that help to maintain light absorption - but above you state there is less light being absorbed due to less Chl-a.

Line 564: This statement is false. Ik is the inflection point of light-limiting versus light-saturating. Under Fe-deficiency Ik was higher, indicating that it took more light for the cells to become saturated.

Line 579: I would recommend also calculating the normalised stern-volner NPQ as well and determining whether any differences can be seen here. NPQ stern-volner has been show to have poor correlation with other photophysiological metrics, whereas the normalised stern-volner examines changes in both the dark and light-regulated states. This makes it useful to comparing samples under different Fe conditions. See Schuback et al. 2021 Frontiers for more details.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Thomas Ryan-Keogh

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

9 Nov 2021

see response to reviewers

Submitted filename: Response to reviewers_final.docx

Click here for additional data file.

15 Nov 2021

The Southern Ocean diatom Pseudo-nitzschia subcurvata flourished better under simulated glacial than interglacial ocean conditions: combined effects of CO2 and iron.

PONE-D-21-25243R1

Dear Dr. Koch,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Douglas A. Campbell, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Nice work!

Reviewers' comments:

17 Nov 2021

PONE-D-21-25243R1

The Southern Ocean diatom Pseudo-nitzschia subcurvata flourished better under simulated glacial than interglacial ocean conditions: combined effects of CO2 and iron

Dear Dr. Koch:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Douglas A. Campbell

Academic Editor

PLOS ONE