Dynamic Regulation of Extracellular Superoxide Production by the Coccolithophore Emiliania huxleyi (CCMP 374)

In marine waters, ubiquitous reactive oxygen species (ROS) drive biogeochemical cycling of metals and carbon. Marine phytoplankton produce the ROS superoxide (O2−) extracellularly and can be a dominant source of O2− in natural aquatic systems. However, the cellular regulation, biological functioning, and broader ecological impacts of extracellular O2− production by marine phytoplankton remain mysterious. Here, we explored the regulation and potential roles of extracellular O2− production by a noncalcifying strain of the cosmopolitan coccolithophorid Emiliania huxleyi, a key species of marine phytoplankton that has not been examined for extracellular O2− production previously. Cell-normalized extracellular O2− production was the highest under presumably low-stress conditions during active proliferation and inversely related to cell density during exponential growth phase. Removal of extracellular O2− through addition of the O2− scavenger superoxide dismutase (SOD), however, increased growth rates, growth yields, cell biovolume, and photosynthetic efficiency (Fv/Fm) indicating an overall physiological improvement. Thus, the presence of extracellular O2− does not directly stimulate E. huxleyi proliferation, as previously suggested for other phytoplankton, bacteria, fungi, and protists. Extracellular O2− production decreased in the dark, suggesting a connection with photosynthetic processes. Taken together, the tight regulation of this stress independent production of extracellular O2− by E. huxleyi suggests that it could be involved in fundamental photophysiological processes.


INTRODUCTION
Ubiquitous reactive oxygen species (ROS) within marine waters help drive global biogeochemical cycling. ROS include intermediates in the reduction of molecular oxygen (O 2 ) to water (H 2 O), which consist of superoxide ( O 2 -), hydrogen peroxide (H 2 O 2 ), and hydroxyl radical (OH•). These ROS shape the transformation of metal nutrients including iron (Rose, 2012) and manganese (Wuttig et al., 2013), as well as carbon (Heller and Croot, 2010b) due to their ability to act as both oxidants and reductants. Within aquatic environments, ROS are produced through both abiotic (e.g., via photodegradation of organic matter) and biotic means (Zinser, 2018). Biotic mechanisms include active extracellular production by marine microorganisms, which can be a dominant source of O 2 in natural waters (Rose et al., 2008;Hansard et al., 2010). While ROS are formed intracellularly within all aerobic organisms as metabolic by-products, a plethora of marine microorganisms actively produce ROS extracellularly as well. Although intracellular O 2 can be released into the marine environment upon cell lysis, these rates cannot account for the steady-state concentrations that have been measured in natural waters (Rose, 2012). Furthermore, within cells, O 2 exists in equilibrium with its conjugate acid, the hydroperoxyl radical (HOO•); however, with a pK a of 4.8, the O 2 anion is the dominant form at physiological pH (Bielski et al., 1985). Unlike H 2 O 2 , O 2 cannot readily diffuse across membranes due to its negative charge, short lifetime (~10 −5 s), and limited diffusive distance (~10 −7 m) (Lesser, 2006;Brown and Griendling, 2009;. Thus, the majority of microbially derived O 2 within the extracellular environment must be created on or near the cell surface via active extracellular O 2 production mechanisms . The ability to produce extracellular O 2 has been documented among heterotrophic bacteria (Diaz et al., 2013) and phytoplankton, including cyanobacteria (Rose et al., 2005(Rose et al., , 2008Godrant et al., 2009;Hansel et al., 2016), diatoms (Kustka et al., 2005;Hansel et al., 2016;Schneider et al., 2016), dinoflagellates (Saragosti et al., 2010;Zhang et al., 2016a), nontoxic microalgae (Marshall et al., 2005a), and harmful microalgae (Oda et al., 1997;Marshall et al., 2005a,b;Portune et al., 2010;. Despite advancements in identifying the extensive presence and environmental relevance of extracellular O 2 production by marine microflora, the mechanisms of extracellular O 2 production and its biological roles are not well understood. The freshwater chlorophyte Chlamydomonas reinhardtii (Anderson et al., 2015) and marine raphidophytes Chattonella marina and Chattonella ovata (Kim et al., 2000(Kim et al., , 2007 are either confirmed or thought to produce extracellular O 2 via cell membrane associated enzymes known as NADPH oxidases (Nox). These enzymes transfer electrons from cytosolic NADPH pools across cell membranes to reduce O 2 in the surrounding aqueous environment, thus creating extracellular O 2 -. The presence of these enzymes has been implicated in diatoms and dinoflagellates as well (Kim et al., 2000;Kustka et al., 2005;Hervé et al., 2006;Saragosti et al., 2010). Extracellular O 2 production is light dependent in several phytoplankton taxa (Kim et al., 1999;Marshall et al., 2002;Milne et al., 2009;Saragosti et al., 2010;Hansel et al., 2016;Schneider et al., 2016), which has led to speculation that photosynthesis may serve an indirect role in extracellular O 2 production by supplying NADPH to cell surfaceassociated NADPH-oxidizing enzymes such as Nox (Marshall et al., 2002;Saragosti et al., 2010;Schneider et al., 2016). Proposed biological roles of phytoplankton-derived extracellular O 2 are diverse . For instance, extracellular O 2 production has been implicated in harmful algal bloom toxicity (Tanaka et al., 1992;Yang et al., 1995;Kim et al., 1999;Marshall et al., 2003;Kim and Oda, 2010;Dorantes-Aranda et al., 2013Mardones et al., 2015), metal nutrient acquisition (Rose et al., 2005;Garg et al., 2007;Liu et al., 2007;Rose, 2012;Roe and Barbeau, 2014), allelopathy (Oda et al., 1992(Oda et al., , 1997Marshall et al., 2005b), and defense against grazing (Martel, 2009;Flores et al., 2012). Model phytoplankton species generate abundant extracellular O 2 even under ideal growth conditions in the absence of any obvious stressors (Kustka et al., 2005;Rose et al., 2005;Marshall et al., 2005a,b;Godrant et al., 2009;Portune et al., 2010;Diaz et al., 2013;Hansel et al., 2016;Schneider et al., 2016), suggesting an association with basal functioning. For example, extracellular O 2 regulates growth and morphology in the prolific ROS producer, C. marina (Oda et al., 1995), as well as growth in bacteria (Saran, 2003;Hansel et al., 2019) and differentiation in microbial eukaryotes (Aguirre et al., 2005). In these microorganisms, extracellular O 2 production rates are the highest during active growth and at low cell densities consistent with beneficial cell signaling and autocrine growth regulation, as also seen in plants (Mittler et al., 2011) and animals (Brown and Griendling, 2009;Aguirre and Lambeth, 2010). The accumulation of studies showing that extracellular O 2 production by diverse phytoplankton is similarly dependent on cell density (Marshall et al., 2005a;Hansel et al., 2016; and growth phase (Oda et al., 1995;Kim et al., 1999;Portune et al., 2010) has spurred speculation that extracellular O 2 production may be involved in phytoplankton cell signaling and/or growth regulation in species other than C. marina, although this possibility remains largely untested . Clarifying the cellular regulation and biological function of active extracellular O 2 production by phytoplankton is critical to understand the effects of ROS on ocean redox balance, biogeochemical cycling, and ecological interactions in marine waters. Among phytoplankton, coccolithophores are one of the most prevalent groups in the global ocean. Further, the original report of extracellular H 2 O 2 production by the coccolithophorid species Pleurochrysis carterae pioneered the recognition of microorganisms as significant sources of ROS in aquatic systems (Palenik et al., 1987). Despite this discovery, however, coccolithophorids are under explored in terms of extracellular ROS production. Therefore, this study was conducted to investigate the dynamics, cellular regulation, and biological function of extracellular O 2 production by a noncalcifying strain of Emiliania huxleyi (CCMP 374), the most prevalent coccolithophore species in modern oceans (Westbroek et al., 1989;Brown and Yoder, 1994).

MATERIALS AND METHODS
Cultivation of E. huxleyi, Growth Tracking, and Cell Counts Cultures of E. huxleyi were inoculated into f/2 growth media prepared without the addition of silicic acid (Guillard and Ryther, 1962)  To begin this experiment, 7.5 ml of media were inoculated with 300 μl stationary phase culture to give an initial concentration of ~2.3 × 10 5 cells ml −1 and grown under the conditions above in 13 mm borosilicate glass tubes (Thermo Fisher Scientific, 14-962-26D). Treatments included three different concentrations of SOD added at three different volumes and a deionized water (DI) control. A 10 kU ml −1 stock of SOD was prepared using DI. Then, treatments were performed on triplicate cultures, where each tube received daily additions of 20.25 μl, 37.5 μl, or 75 μl of the 10 kU ml −1 SOD stock, or 75 μl DI to give final concentrations of 27 U ml −1 SOD, 50 U ml −1 SOD, 100 U ml −1 SOD, or 0 U ml −1 SOD, respectively. A subsequent control experiment was performed with diafiltered SOD to ensure that SOD was responsible for potential changes observed in the cultures. For this control experiment, 7.5 ml of media were inoculated with 300 μl stationary phase culture to give an initial concentration of ~1.8 × 10 5 cells ml −1 and grown under the conditions above in 13 mm borosilicate glass tubes. To create the dialyzed SOD, a 10 kU ml −1 SOD stock prepared with DI was diafiltered by passing the SOD solution through an Amicon ultrafiltration device (10 kDa molecular weight cut-off, Millipore) at 3200 ×g for 20 min at 4°C, thereby removing the enzyme from the solution. Treatments were performed on triplicate cultures, where each tube received daily additions of 75 μl dialyzed SOD, 75 μl DI, and 75 μl SOD (100 U ml −1 SOD final concentration).

Cell Imaging and Analysis
Individual E. huxleyi cells were imaged using a FlowCam ® (Fluid Imaging Inc., Scarborough, ME), a continuous flowthrough microscope fitted with a color and monochromatic camera. To calibrate the FlowCam ® for imaging, dilutions of E. huxleyi cultures from 0 to 95% using 0.2 μm filtered seawater were analyzed in varying context settings. Final settings (Supplementary Table S1) were chosen based on their ability to differentiate cell shape and color and provide useable images for morphological and volume assessments. Culture samples of 100 μl were diluted with 500 μl of 0.2 μm filtered seawater prior to analysis on the FlowCam ® . Images were analyzed with VisualSpreadsheet ® (Fluid Imaging Inc., Scarborough, ME).
The spherical biovolume was determined using three separate

Photophysiology
Photophysiological health of E. huxleyi was monitored using the Satlantic fluorescence induction and relaxation (FIRe) fluorometer system (Sea-bird Scientific, Halifax, NS Canada). Prior to analysis, samples were allowed to dark adapt for approximately 30 min and were then diluted using 0.2 μm filtered seawater as necessary to avoid detector saturation. Samples were illuminated with a single turnover flash of blue excitation for a duration of 80 μs. The measured fluorescence response was used to calculate the maximum efficiency of photosystem II (PSII), or F v /F m , using the equation: where F m is the maximum fluorescence yield, and F o is the minimum fluorescence yield. For each sample, 20 acquisitions were obtained.

Extracellular O 2 2 --Production
Net extracellular O 2 production by E. huxleyi was measured using the flow-through FeLume (II) analytical system (Waterville Analytical, Waterville, ME) by detecting chemiluminescence emitted through the reaction of O 2 and the specific probe methyl Cypridina luciferin analog (MCLA), as previously described (Diaz et al., 2013;Schneider et al., 2016). This method allows for manipulation of experimental settings (e.g., light levels, increasing cell density) during analysis to detect changes in O 2 on immediate timescales (i.e., seconds). It has been used to measure O 2 production in pure cultures (Kustka et al., 2005;Diaz et al., 2013;Schneider et al., 2016;Zhang et al., 2016a;Hansel et al., 2019) and natural waters Hansel et al., 2016;Zhang et al., 2016b). Following the procedures of Diaz et al. (2013), MCLA blanks generated in the absence of SOD were used to determine biologically derived O 2 concentrations, in order to avoid overestimation of biological O 2 production rates due to MCLA auto-oxidation.
Briefly, cells were deposited onto an inline filter (0.22 μm), continuously rinsed (2 ml min −1 ) with a phosphate buffer (20 mM; pH = 7.6) that matched the salinity of the seawater media base (38 psu Diaz et al. (2013), the following exceptions were implemented here. First, cells deposited on the filter were exposed to ambient light (~5 μmol photons m −2 s −1 ) unless they were covered in a dark photography film-changing bag to obtain dark (0 μmol photons m −2 s −1 ) measurements, as indicated. Second, only net production rates were determined, and finally, calibration was performed using standard additions of potassium superoxide (KO 2 ) by the method of Schneider et al. (2016). A preliminary cell concentration was obtained by microscopy using a hemocytometer counting chamber to help ensure that the same number of cells was loaded onto the in-line filter from each biological replicate and on each day within an experiment. The biotic steady-state O 2 concentration was calculated by subtracting blank signals generated from the mixture of MCLA and phosphate buffer with a syringe filter inline and in the absence of SOD. Then, net O 2 production rates were calculated by multiplying the biotic steady-state O 2 concentration (pM) by the flow rate (2 ml min −1 ), dividing that value by the number of cells loaded onto the inline filter (either found using microscopy or flow cytometry), and converting to final units of amol cell −1 h −1 . All chemicals used to measure O 2 were obtained from Millipore Sigma, except for MCLA, which was obtained from Tokyo Chemical Industry Co., Ltd.

Statistical Analyses
All statistical analyses were performed using JMP Pro 13.0.0 (SAS Institute Inc., Cary, NC). Regression analyses of cellnormalized O 2 production rates as a function of time across the growth curve of E. huxleyi and cell density were performed using Spearman's rank-order correlation. This regression analysis indicates the presence or absence of monotonic relationships based on the correlation coefficient (ρ) and its level of significance (p). An independent two sample Student's t-test was used to determine potential differences between mean  Figure S1) and significantly declined as E. huxleyi grew over time (Spearman's ρ = −0.58; p < 0.0001) (Figure 1). For example, E. huxleyi produced maximum amounts of O 2 -(average ± SE) at the first time point measured in early exponential phase (4,478 ± 611 amol cell −1 h −1 , n = 3; day 2). These maximal rates were nearly 78 times higher than net production rates measured between day 8 and day 25, when average production declined to 58 ± 55 amol cell −1 h −1 (n = 30). Flow cytometry analyses revealed this decline was not due to an increase in senescent cells (i.e., cells with low chlorophyll). After day 8, net cell-normalized  (Figure 2). In exponential phase, average net per-cell O 2 production rates decreased by more than 200% from the highest (9.2 × 10 6 ) to lowest (4.6 × 10 5 ) number of cells analyzed.
To assess longer-term effects (min-h) of cell density on extracellular O 2 production, exponentially growing cells were preconditioned to lower cell densities by diluting cultures with 0.22 μm filtered, autoclaved seawater and incubating for 0-6.5 h prior to conducting O 2 measurements. In this experiment, the number of cells loaded on the FeLume filter at each dilution level was kept constant [average ± SE was 1.32 × 10 6 ± 8.47 × 10 4 cells (n = 26)]. A 10-and 100-fold dilution of E. huxleyi resulted in a 51 and 172% increase in cell-normalized net extracellular O 2 production rates, respectively.
Although average rates increased at both dilution levels, only the 100-fold dilution resulted with significantly more extracellular O 2 per cell than the undiluted control (t-test; p < 0.05) (Figure 3)

SOD Addition Experiments
To assess how extracellular O 2 influences growth and physiology, E. huxleyi cultures were grown with a range of SOD concentrations and monitored until the end of exponential phase. Since SOD is a large enzyme (>31 kDa) (Cass, 1985), it cannot passively cross cell membranes and therefore selectively targets  Table S2). In addition, specific growth rates were between 15 and 22% higher (t-test; p < 0.05) in the presence of SOD ( Figure 4B; Supplementary Table S2). Cell abundances from cultures grown with 100 U ml −1 SOD were significantly higher (t-test; p < 0.05) than cell abundances from cultures grown without SOD beginning on day 7 ( Figure 4C; Supplementary Table S2), when cell concentrations in the highest SOD addition were 41% higher than the unamended treatment.
To confirm whether these results were specifically due to SOD, a control experiment was performed in which SOD was removed via diafiltration and only the low molecular weight fraction (<10 kDa) of the SOD suspension was added to cultures. In vivo fluorescence was significantly different between treatments (mixed factor repeated measures ANOVA; p < 0.05) with significance increasing over time (mixed factor repeated measures ANOVA; p < 0.0001) (Supplementary Figure S2A; Supplementary Table S2). Cultures grown with SOD grew significantly faster (t-test; p < 0.05) than those grown with dialyzed SOD (Figure 2B; Supplementary Table S2). By the end of the control experiment (day 12), cell abundances from cultures grown with dialyzed SOD were significantly lower (t-test; p < 0.05) than those grown with SOD by about 32% (Supplementary Figure S2C; Supplementary Table S2). Thus, the effect of SOD addition on growth rates and growth yields could not be accounted for by the dialyzed SOD control (Supplementary Figure S2; Supplementary Table S2).
In addition to growth yields and growth rates, the cellular biovolume of E. huxleyi was monitored in SOD addition experiments. Biovolume was calculated using three different methods, but regardless of the calculation method used, cellular biovolume was larger in cultures grown with SOD (t-test; p < 0.0001) (Figure 5A; Supplementary Figure S3A; O production rates were measured from ~10 6 cells (n = 3 biological replicates for each day except day 6 and day 25 where three separate measurements were made on three biological replicates giving n = 9). Regression analysis on per-cell -2 O production as a function of time was performed using Spearman's rank-order correlation. The correlation coefficient (ρ) and its level of significance (p) are provided. In vivo fluorescence was normalized to the value on day 0. Error bars indicate one standard error of the mean of three biological replicates. Table S2). Adding SOD increased biovolume between 6.3 and 22.1%, depending on the SOD concentration and calculation method used. In the control experiment with dialyzed SOD, biovolume of cells grown with SOD was significantly larger (t-test; p < 0.0001) than those grown with dialyzed SOD by at least 19% (Supplementary Figure S3A; Supplementary Table S2) confirming the response was due to the presence of SOD.

Supplementary
In addition to stimulating growth, SOD improved photophysiological health, as evidenced by increased F v /F m values, which indicated more efficient light assimilation by PSII in the presence of SOD. By day 10, adding SOD at each concentration increased F v /F m between 11 and 18% compared to the 0 U ml −1 SOD control (t-test; p < 0.05) ( Figure 5B; Supplementary Table S2). A similar trend was seen in the control experiment with dialyzed SOD, where F v /F m values from cultures grown with SOD were significantly higher than those from cultures grown with dialyzed SOD starting on day 7 (t-test; p < 0.05) (Supplementary Figure S3B; Supplementary Table S2). Therefore, improvements in photophysiological health are attributed to the effects of SOD.  (Figure 6). Upon removal of light, there was an immediate decline in production, which stabilized at 1,085 ± 29 pM after ~160 s in dark conditions. This result could not be accounted for by abiotic factors, as the removal of light had insignificant effects on O 2 production in the absence of E. huxleyi cells. The drawdown of signal below the abiotic O 2 baseline by SOD confirmed that the biogenic signal was indeed due to O 2 production ( Figure 6). All biological replicates produced less O 2 in dark compared to ambient light conditions (t-test; p < 0.0001), indicating a connection with photophysiological processes (Figure 7). Dark conditions inhibited O 2 production by an average of 70%.  O production rates were measured from ~10 6 E. huxleyi cells during exponential growth phase post no dilution (1:1), 10-fold dilution (1:10), or 100-fold dilution (1:100) with 0.22 μm filtered, autoclaved seawater for 0-6.5 h. Significant differences (two sample Student's t-test) relative to the undiluted control (1:1) are indicated by asterisks, where p < 0.05 is represented by * symbol. Error bars depict one standard error of the mean of replicates (n = 8 for 1:100 and 9 for 1:1 and 1:10 dilutions).

DISCUSSION
Here, we investigated the cellular regulation and potential physiological roles of extracellular O 2 production in a noncalcifying strain (CCMP 374) of the cosmopolitan marine coccolithophore E. huxleyi to approach a broader understanding of the potential ecological and environmental impacts of phytoplankton-derived extracellular O 2 -. The ability to produce extracellular O 2 is widespread among phytoplankton, yet production rates can vary substantially within and between species   (Diaz et al., 2013). Second, the physiochemical nature of the O 2 anion prevents it from passively crossing intact cell membranes (Bielski et al., 1985;Lesser, 2006;Brown and Griendling, 2009). Therefore, the production rates measured in this study reflect active production of O 2 on or near the surface of E. huxleyi. The reported rates of extracellular O 2 production reflect the balance of gross production and decay at the cell surface, thus giving a net production rate. Therefore, any change in the net production rate of extracellular O 2 could result from a change in gross production, decay, or both. Several abiotic A C B FIGURE 4 | The effect of daily SOD additions on (A) average in vivo fluorescence, (B) specific growth rate during exponential growth phase, and (C) cell abundance of E. huxleyi. Significant differences in in vivo fluorescence between SOD additions were found using a mixed factor repeated measures ANOVA. Significant differences (two sample Student's t-test) in specific growth rate and cell abundances relative to the control (0 U ml −1 SOD) are indicated by asterisks, where p < 0.05 and < 0.01 are represented by * and ** symbols, respectively. Error bars depict one standard error of the mean of biological replicates (n = 3).
Frontiers in Microbiology | www.frontiersin.org 8 July 2019 | Volume 10 | Article 1546 and biotic factors have the potential to degrade extracellular O 2 at or near the cell surface, such as interactions with trace metals including soluble and mineral-bound iron (Fujii et al., 2006;Heller and Croot, 2010a) and expression of cell surface SODs (Aguirre et al., 2005;Oshikawa et al., 2010;Bauer, 2014).
ROS production is commonly associated with stress; however, E. huxleyi produced extracellular O 2 without any added stressors.
In fact, cell-normalized O 2 production by E. huxleyi was the highest in early exponential phase under presumably the least taxing culture conditions when nutrient concentrations were the highest (Figure 1) and when photosynthetic efficiency was at a maximum (Supplementary Figure S1). These results reflect the rates of extracellular O 2 production by cells that were removed from the prevailing culture conditions and analyzed ex situ. The ex situ analytical conditions such as pH (7.6) were identical across culture samples of all ages, yet differed from in situ levels (Supplementary Figure S4). These results therefore show that as cultures age, there is a shift toward lower O 2 production when cells are analyzed under the same conditions. The potential effect of pH on extracellular O 2 production by E. huxleyi is unknown, but an increase in pH stimulates extracellular ROS production by C. marina (Liu et al., 2007), which is opposite to the trend reported here (Figure 1; Supplementary Figure S4). The observed decline in extracellular O 2 production with culture age therefore suggests that extracellular O 2 production is physiologically driven and unlikely related to a stress response, as levels of stress would presumably increase with time in batch culture due to the depletion of resources. These findings do not rule out the potential for E. huxleyi to upregulate extracellular O 2 under stressful conditions, as seen with extracellular H 2 O 2 (Evans et al., 2006) and intracellular ROS (Evans et al., 2006;Vardi et al., 2012) during viral infection, but do demonstrate a physiological investment in the production of extracellular O 2 under ideal growth conditions, which suggests some role in basal metabolism. In order to clarify if and how E. huxleyi regulates extracellular O 2 in response to biogeochemical variability and stress, future work should consider the influence of factors such as viral infection and pH. In addition to having growth phase dependence (Figure 1) similar to other phytoplankton (Oda et al., 1995;Kim et al., 1999;Portune et al., 2010), cell-normalized net extracellular O 2 production rates by E. huxleyi were also inversely dependent on cell density over a range of timescales during exponential growth phase. For example, this trend occurred when cell density increased on time scales of seconds to minutes (Figure 2A) and when cells were preconditioned to lower cell densities on timescales of minutes to hours (Figure 3). This tight regulation may suggest a dynamic cell density-dependent signaling role for O 2 production in E. huxleyi. For instance, extracellular O 2 deriving from one cell may act as a signal between other cells or within the same cell to provide information on surrounding population density . Similar cell density dependent trends in extracellular O 2 production have been demonstrated in other phytoplankton (Marshall et al., 2005a;Hansel et al., 2016;, including C. marina (Marshall et al., 2005b). Although evidence has been accumulating that extracellular O 2 production is directly involved in growth promotion in a range of microbial cell types (Oda et al., 1995;Aguirre et al., 2005) of cells using SOD. Scavenging O 2 promoted growth (Figure 4), increased cell biovolume (Figure 5A), and improved photosynthetic efficiency ( Figure 5B). Conversely, removal of extracellular O 2 from cultures of Chattonella spp. attenuates growth (Tanaka et al., 1992;Oda et al., 1995). Specifically, in one prior study, C. marina growth was significantly hampered under similar SOD concentrations used in the present study, and the morphological state of cells was altered (Oda et al., 1995). Recently, Hansel et al. (2019) revealed that the growth of common marine bacteria from the Roseobacter clade was inhibited by SOD in a dose-dependent manner (Hansel et al., 2019). Similarly, the removal of extracellular ROS from fungi and the amoebozoan Dictyostelium discoideum is also detrimental to development (Aguirre et al., 2005). Taken (Oda et al., 1995), bacteria (Hansel et al., 2019), fungi, and protists (Aguirre et al., 2005). However, the addition of SOD not only removes O 2 but produces H 2 O 2 , which may also have impacts on E. huxleyi physiology. For example, high concentrations of H 2 O 2 are harmful to phytoplankton (Dupouy et al., 1985;Morris et al., 2011), but normal growth of C. marina is dependent on low levels of extracellular H 2 O 2 (Oda et al., 1995). In fact, in a variety of cell types, the dismutation of Nox-derived O 2 by cell surface SOD generates extracellular H 2 O 2 , which can diffuse into the cell, to elicit gene expression (Shapiguzov et al., 2012), morphogenesis (Rossi et al., 2017), and proliferation (Oshikawa et al., 2010;Bauer, 2014 (Figure 6).  This finding adds to a growing body of evidence linking modulation of extracellular O 2 production by phytoplankton to light availability and therefore photophysiology. For instance, extracellular O 2 production is light dependent in many phytoplankton including Thalassiosira spp. (Milne et al., 2009;Schneider et al., 2016), Trichodesmium , Symbiodinium (modulated on the same time scales shown here) (Saragosti et al., 2010), and Chattonella spp. (Kim et al., 1999;Marshall et al., 2002;Dorantes-Aranda et al., 2013). Thus, O 2 production may somehow be involved with light dependent processes (e.g., photosynthesis, photoacclimation physiology), and this functionality may be conserved across phytoplankton taxa. Interestingly, in a previous study, when C. marina was treated with DCMU [3-(3,4-dichlorophenyl)-1,1-dimethylurea], an electron transfer inhibitor between photosystem II and I, extracellular O 2 production was quenched to levels observed under dark conditions (Marshall et al., 2002), further illustrating a mechanistic link between extracellular O 2 production and photophysiology.
Overall, this study reveals that the stress-independent production of extracellular O 2 by E. huxleyi is dynamically regulated, and potentially part of a basal process involved with photophysiology. Extracellular O 2 production by E.
huxleyi is conceivably part of healthy cellular functioning for several reasons. First, the fact that cells diverted energy toward making O 2 in the absence of a stressor suggests its production can be unassociated with stress and probably related to basal functioning. Indeed, E. huxleyi produced maximum amounts of extracellular O 2 per cell under ideal growth conditions while cells were most metabolically active (Figure 1) and when efficiency of photosynthetic processes was the highest (Supplementary Figure S1). Further, the steady-state concentrations of O 2 generated by E. huxleyi cells are not consistent with concentrations that would be damaging (>10 −6 M) but are consistent with concentrations of biological signaling molecules (~10 −12 M) (Saran, 2003). Additionally, E. huxleyi cells tightly controlled the production of extracellular O 2 as a function of light, cell density, and growth phase, and on timescales as short as seconds, consistent with other phytoplankton (Oda et al., 1995;Kim et al., 1999;Marshall et al., 2002Marshall et al., , 2005aMilne et al., 2009;Portune et al., 2010;Saragosti et al., 2010;Dorantes-Aranda et al., 2013;Hansel et al., 2016;Schneider et al., 2016;, suggesting a potential dynamic role in signaling and photophysiology. Indeed, stress-independent extracellular O 2 production by E. huxleyi is in agreement with several other microorganisms such as phytoplankton and bacteria where its production is prolific under ideal growth conditions (Oda et al., 1995;Kustka et al., 2005;Rose et al., 2005;Marshall et al., 2005a,b;Godrant et al., 2009;Portune et al., 2010;Diaz et al., 2013;Hansel et al., 2016;Schneider et al., 2016). The fact that E. huxleyi still produces extracellular O 2 in the absence of light suggests there could be additional and/or alternative purposes for this production beyond photophysiology. Indeed, extracellular O 2 production can be produced through a variety of subcellular mechanisms and could be produced for a combination of ecophysiological functions .
huxleyi has yet to be determined but should be considered in future work.

DATA AVAILABILITY
Any materials and data will be made available to members of the scientific community upon request.

AUTHOR CONTRIBUTIONS
SP and JD conceived the study. SP conducted the experiments and data analysis, with contributions from AT. All authors contributed to interpretation of results and preparing the manuscript.