Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton

MARCHE-03155; No of Pages 12 Marine Chemistry xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Marine Chemistry journal homepage: www.elsevier.com/locate/marchem

Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton Christel S. Hassler a,b,⁎, Louiza Norman b, Carol A. Mancuso Nichols c, Lesley A. Clementson d, Charlotte Robinson b, Véronique Schoemann e, Roslyn J. Watson d, Martina A. Doblin b a

University of Geneva, Faculty of Science, Earth and Environmental Sciences, Institute F.-A. Forel, Marine and Lake Biogeochemistry, 10 rte de Suisse, 1290 Versoix, Switzerland University of Technology Sydney, Plant Functional Biology and Climate Change Cluster, PO Box 123, Broadway 2007, NSW, Australia CSIRO Materials Science and Engineering, c/o CSIRO Marine and Atmospheric Research, PO Box 1538, Hobart 7001, TAS, Australia d CSIRO Ocean and Atmosphere, PO Box 1538, Hobart 7001, TAS, Australia e Laboratory of Glaciology, Université Libre de Bruxelles, CP 160/03 Avenue F.D. Roosevelt, 50, B-1050 Bruxelles, Belgium b c

a r t i c l e

i n f o

Article history: Received 14 June 2014 Received in revised form 3 October 2014 Accepted 3 October 2014 Available online xxxx Keywords: Algae Iron Organic ligand Biochemistry

a b s t r a c t Growth limitation of marine algae due to lack of iron occurs in up to 40% of the global ocean. Despite important advances on the impact of organic compounds on iron biogeochemistry, their roles in controlling iron availability to prokaryotic and eukaryotic phytoplankton remain unclear. Whether algal and bacterial exopolymeric substances (EPS) include organic ligands which may help iron-limited phytoplankton growth remains an unknown. If so, then EPS could relieve phytoplankton iron limitation with implications for the biological carbon pump and hence the regulation of atmospheric CO2. Here we compared the biological impact of algal, bacterial and in situ EPS with model compounds, a siderophore and two saccharides on biological parameters including, iron bioavailability, phytoplankton growth, photo-physiology and community structure. Laboratory and field experiments demonstrated that EPS produced by marine microorganisms are efficient in sustaining biological iron uptake as well as algal growth, and can affect natural phytoplankton community structure. Our data suggest that natural phytoplankton growth enhancement in the presence of EPS was not solely due to highly bioavailable iron forms, but also because EPS contains other micronutrients. Stronger ligands were detected following iron-siderophore enrichments (log KFe′L = 12.0) and weaker ligands were measured in the presence of EPS (log KFe′L = 10.4– 11.0). The trend of the conditional stability constants of organic ligands did not seem to be affected as a result of biological activity and photo-chemistry during our four day incubations. The shift in the phytoplankton community observed during our field experiments was not uniformly observed between different sites rendering it difficult to extrapolate which functional group(s) would benefit the most from iron bound to EPS. © 2014 Elsevier B.V. All rights reserved.

1. Introduction The parameters that control iron bioavailability to phytoplankton are not well understood, but are strongly influenced by the physical and chemical forms of iron, its biogeochemical cycling, and the various iron requirements and uptake strategies of bacterio- and phytoplankton communities (e.g., Barbeau et al., 2001; Hutchins et al., 1999; Maldonado et al., 2005; Strzepek et al., 2005; Sunda and Huntsman, 1995). In addition a complex feedback exists between microorganisms and iron, where organic ligands can be biologically produced and impact iron chemistry with potential consequences for its bioavailability (e.g., Hassler et al., 2011a). As dissolved iron (dFe) is mainly bound to organic ligands (N99%) in the ocean, it is widely accepted that these ligands control iron biogeochemistry and bioavailability. Past studies have highlighted a link ⁎ Corresponding author. E-mail address: [email protected] (C.S. Hassler).

between phytoplankton iron limitation, biological iron uptake rates, iron requirement for growth and the presence of in situ iron-binding ligands (Gledhill and Buck, 2012; Hassler et al., 2012a; Morrissey and Bowler, 2012; Shaked and Lis, 2012 for recent reviews). Bacterially produced siderophores, which are excreted under iron-limited conditions, are commonly associated with strong iron-binding organic ligands translating to high conditional stability constants (Gledhill and Buck, 2012). Because siderophores are specifically recognized by bacteria, they allow bacterial regulation of iron-limitation (Granger and Price, 1999; Morrissey and Bowler, 2012). However, it has been demonstrated that some eukaryotic phytoplankton use iron bound to siderophores via reductive dissociation (Maldonado and Price, 2000), making the role of siderophores more complex than a simple bacterial switch. To date no eukaryotic phytoplankton have been found to excrete siderophores, thus raising the question about their ability to widely benefit from such bacterially-derived organic compounds for growth. Other organic ligands such as porphyrins, humic acids and saccharides could also control the growth of planktonic micro-organisms (Gledhill and Buck,

http://dx.doi.org/10.1016/j.marchem.2014.10.002 0304-4203/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

2

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

2012; Hassler et al., 2012a; Shaked and Lis, 2012). In the ocean, a mixture of organic ligands is likely to be present. Indeed the widely unresolved nature of organic ligands in the ocean limits our ability to identify key components for the control of iron limitation in eukaryotic phytoplankton (Gledhill and Buck, 2012; Hassler et al., 2012a; Shaked and Lis, 2012). In the open ocean, far from any continental margin, biological activity is likely a significant source of organic ligands. Most micro-organisms excrete material — often regrouped under the term exopolymeric substances (EPS) (Decho, 1990; Hoagland et al., 1993; Aluwihare et al., 1997). EPS are poorly defined macromolecules bearing various functional groups, which encompass part of the neutral and anionic polysaccharides, amino acids, and protein reported in the ocean (e.g. Biersmith and Benner, 1998; Mancuso Nichols et al., 2005). As a result of surface reaction and aggregation, EPS are also found in marine transparent EPS and marine snow (Santschi et al., 2003). As such, EPS can represent up to 50% of marine dissolved and colloidal organic carbon (Aluwihare et al., 1997; Verdugo et al., 2004). Saccharides and bacterial EPS have been demonstrated to bind iron and enhance its solubility as well as its biological uptake (viz. bioavailability) to Southern Ocean phytoplankton (Rue and Bruland, 2001; Hassler and Schoemann, 2009a, 2009b; Benner, 2011; Hassler et al., 2011a, 2011b). In addition, eukaryotic phytoplankton such as diatoms and the haptophyte Emiliania huxleyi have been shown to produce iron-binding organic ligands with a conditional stability constant (KFe′L) of 1011.5 (Boye and van den berg, 2000; Rue and Bruland, 2001; Rijkenberg et al., 2008). Finally, organic ligands have been shown to be produced in the iron-depleted Southern Ocean in response to in situ iron enrichment experiments (Rue and Bruland, 1997; Boye et al., 2005). This paper is a companion study to the composition and the impact of marine EPS on iron chemistry (Norman et al., in press). The study from Norman et al. demonstrated that material excreted by bacterial and phytoplankton cultures, as well as by natural blooms, contribute to the pool of marine iron-binding ligands with stability constants (KFe′L = 1011.2–1011.9) that were similar to what has been reported for the ocean since 20 years (reference herein). Moreover, EPS contributed to the pool of humic-like material as per Laglera and van den Berg (2009) and increased iron solubility. Finally, EPS did not contain only iron but also other nutrients (N, P, Co, Cd, Zn) that are reported to limit phytoplankton growth in some parts of the ocean. However, the impact that EPS from different origins exert on iron bioavailability, and longer term biological effects such as growth, remains mostly unknown. To shed light on the role of EPS on phytoplankton iron limitation, we studied EPS excreted by two Southern Ocean phytoplankton (E. huxleyi and Phaeocystis antarctica) cultures and one Antarctic sea-ice bacterium (Pseudoalteromonas sp.) culture as well as EPS isolated from the surface waters of the Sub-Antarctic Zone (SAZ) at the time of a phytoplankton bloom dominated by coccolithophorids during the 2010 austral summer. We compared the effect of these substances to model iron binding organic compounds on iron bioavailability, photo-physiology, and phytoplankton growth using the diatom Chaetoceros simplex. In addition, the impact of a previously characterized bacterial EPS (Pseudoalteromonas sp. isolated from the Southern Ocean, Mancuso Nichols et al., 2004; Hassler et al., 2011b) on iron bioavailability and natural phytoplankton community structure was evaluated from two sites with contrasted nutrient distribution in the Tasman Sea and in the SubAntarctic Zone (Table 1, Hassler et al., 2014).

Table 1 Average parameters for samples collected at the depth of the fluorescence maximum (Fmax) in the North Tasman Sea (P1) and the Sub-Antarctic Zone (P3) for incubation experiments during the PINTS voyage. Parameters are dissolved iron (dFe), labile iron (FeLabile) expressed relative to dFe, iron binding ligands (sumL as the sum of all detected ligands; and L1 as the strong ligand group) and their conditional stability constants with respect to inorganic iron (KFe′sumL). Labile iron is operationally defined by the experimental setup and the electrochemical technique used; it represents the iron concentration in the sample that is exchangeable towards the 10 μM TAC added following an overnight equilibration. Dissolved macronutrient concentrations as well as total chlorophyll a (Tot Chl a) are shown. The phytoplankton community was size fractionated into pico- (0.7– 2 μm), nano- (2–10 μm) and micro-phytoplankton (N10 μm) by sequential size filtration and relative contributions were calculated from their respective Chl a. bDL = below detection limit, 0.011 μM for NOx; ND = not detected. Measurements compared well with those measured from the depth profile (Hassler et al., 2014).

Depth of Fmax (m) dFe (nM) Sum of L (nM) Log KFe′sumL L1 (nM) Log KFe′L1 FeLabile (%) NOx (μM) Si (μM) PO4 (μM) FV/FM Tot Chl a (μg L−1) Picophytoplankton (%) Nanophytoplankton (%) Microphytoplankton (%)

P1 (165 0 °E, 30 0 °S)

P3 (159 5 °E, 46 2 °S)

95 0.03 2.24 11.62 ND ND 58 bDL 1.02 0.12 0.67 0.344 57.7 39.8 2.5

25 0.29 2.60 11.61 ND ND 39 2.81 0.67 0.35 0.55 0.686 29.2 62.6 8.2

0.2-μm filtered seawater collected in the Sub-Antarctic Zone (SAZSense voyage, RV Aurora Australis, 45 3 °S, 153 1 °E, 11th Feb. 2007) and enriched with micronutrients (Fe, Zn, Co = 5 nM, Cu and Ni = 2 nM, Se = 1 nM all buffered by natural ligands present), chelexed macronutrients (NOx = 30 μM, PO4 = 2 μM, Si = 30 μM) and vitamins (as per medium F/20). Axenicity was verified using DAPI stain for Phaeocystis and the absence of bacterial fatty acid signature for both strains. Natural EPS was isolated from filtered water sampled at the depth of the fluorescence maximum in the Sub-Antarctic Zone (SAZ, P3, 46 2 °S, 159 5 °E) during the PINTS voyage (RV Southern Surveyor, Jan–Feb. 2010, Table 1). The filtrate, containing EPS, was collected into a 25 L clean acid washed carboy to which NaN3 was added (Sigma) to avoid bacterial growth and kept at 4 °C in the dark until ultrafiltration (Labscale TFF system, 10 kDa MWCO PES membrane, Pellicon R XL 50 Cassette, Millipore). The NaN3 was lost during ultrafiltration and rinsing steps; 0.14 mg EPS L−1 was recovered from the SAZ sample. The bacterial strains (Pseudoalteromonas sp.) were isolated from Antarctic sea ice and Southern Ocean seawater and phenotypically characterized as described previously (Mancuso Nichols et al., 2005). Culture filtrates were diafiltered with 10 volumes of sterile Milli-Q™ water and concentrated at room temperature (VivaFlow 200 ultrafiltration unit, PES membrane, 100 kDa MWCO, Sartorius). The concentrated (50 mL) EPS solutions were frozen overnight and the water was removed by freeze-drying (Dynavac). The iron concentration associated with each EPS was analyzed by inductively coupled mass spectrometry (see Norman et al., for experimental detail); it was 393, 2274, 7112, 5267 nmol iron g− 1 EPS for EPS from sea ice bacteria, SAZ bloom, P. antarctica and E. huxleyi, respectively.

2. Methods 2.2. EPS impact on cultured and natural phytoplankton 2.1. EPS isolation EPS were isolated as per Norman et al. (in press). Algal EPS was isolated from P. antarctica (CS 243, Prydz Bay, Antarctica) and E. huxleyi (CS 812, Mercury Passage, Tasmania, Australia), maintained axenically in

To assess the biological impact of iron associated with EPS compared to inorganic iron, phytoplankton originating from an axenic culture and two natural sites were used. An Antarctic diatom was chosen as good model phytoplankton strain as diatom as present world-wide,

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

important for carbon export. The culture consisted of an iron-limited small unicellular diatom (C. simplex, CS 624, Australian National Algae Culture Collection, isolated from Prydz Bay, Antarctica). This culture is studied intensively including physiological changes in relation with iron limitation (Hassler unpublished data), iron bioavailability (Hassler and Schoemann, 2009a, 2009b; Hassler et al., 2011a) and interactions between iron and light (Petrou et al., 2014). It is maintained since 2007 in sterile seawater from the sub-Antarctic Zone (SAZ-Sense voyage, RV Aurora Australis, 45 3 °S, 153 1 °E, 11th Feb. 2007, Bowie et al., 2011) at 4 °C under 30 μmol photons m− 2 s−1 with a 16:8 h light:dark cycle and chelexed macronutrient addition as per the F/20 recipe. As dFe in this water was very low (0.3–0.4 nM), C. simplex was iron-limited under this growth condition as indicated by the low FV/ FM measured. This culture was rendered iron-limited by sequential transfers from iron rich media (F/2) to natural water and by decreasing light level to facilitate growth under iron-limited situation. This acclimation phase lasted approx. 6 months. Field experiments were carried out using water at depth of maximum fluorescence from the North Tasman Sea (Station P1, 30 00.3 °S, 165 00.0 °E, 95 m, 28 Jan. 2010) and the SAZ (Station P3, 46 17.8 °S, 159 53.1 °E, 28 m, 8 Feb. 2010) during the PINTS voyage (RV Southern Surveyor, Jan–Feb. 2010, Table 1). The conductivity, temperature, dissolved O2, photosynthetic active radiation and fluorescence depth profiles were recorded (SeaBird SBE11) 30–45 min prior to water sampling. Water was sampled using Teflon-coated Niskin X-1010 bottles (General Oceanics, USA) mounted on an autonomous rosette (Model 1018, General Oceanics, USA) and deployed using a Dynex hydroline (Dynex Dyneema 75, Hampidjan, Ltd, Nelson, New Zealand). Niskin-X bottles were mixed prior to the collection of 120 μm filtered (Nylon mesh) seawater into acid-washed polycarbonate bottles under a HEPA filter (ISO Class 5) in a clean container van. Sample acquisition and handling were as per GEOTRACES recommendations (Cutter et al., 2010), using acid-washed non-contaminating material. Previous work on this research voyage demonstrated that, at station P1, the lack of nitrogen was responsible for a deep chlorophyll maximum mostly dominated by pico-plankton of the type Prochloroccocus sp. (Table 1, Hassler et al., 2014). At this site, at depth of fluorescence maximum, light was not limiting primary productivity but low nitrate and possibly low dFe concentrations limited phytoplankton growth (Hassler et al., 2014). Experiments and data analysis suggested that light and possibly iron were limiting the growth and productivity of phytoplankton at station P3 (Hassler et al., 2014). 2.2.1. Iron bioavailability Phytoplankton iron uptake rates were used to estimate iron bioavailability (Hassler and Schoemann, 2009a; Hassler et al., 2011a; Shaked and Lis, 2012). For C. simplex, bioavailability of the iron associated with EPS was assessed in the laboratory by comparison of the uptake rate constant (kupt) obtained in the presence of inorganic iron addition and in the presence of iron associated with the EPS in 0.125–1 L polycarbonate bottles. Uptake rate constants were derived from the slope of the linear increase of cellular iron uptake rate as a function of total iron concentrations using 4–5 points measured, each in duplicate as in Hassler et al. (2011a). Experimental solutions consisted of 0.2 μm filtered Tasman Sea surface water (GP13 GEOTRACES voyage, 2011, 30 00 °S 167 00 °E, 0.56 nM dFe) spiked either with inorganic 55FeCl3 (Fe treatment; Perkin Elmer, 20.82 and 31.75 mCi mg−1 iron at the time of use in the field and the laboratory, respectively) or with 55Feprequilibrated with the EPS or other organic ligands for 1 week prior to experiment. Typically organic ligand concentrations were kept constant for each ligand and total iron concentrations were increased up to 3.56 nM. In the case of inorganic iron enrichment, total iron was only increased up to 2 nM to avoid major iron (oxy)hydroxide precipitation in the laboratory (Norman et al., in press), this treatment thus corresponded to the bioavailability of iron bound to in situ ligands. In our experiments, an EPS concentration associated with a 1 nM dFe

3

background was used; corresponding to 2.6, 0.4, 0.1, and 0.2 mg L−1 EPS and 16.3, 1.7, 1.1, and 0.9 μM carbon associated with hydrolyzable exopolysaccharides for EPS from sea-ice bacteria, SAZ bloom, P. antarctica and E. huxleyi, respectively. These carbon concentrations were smaller than reported in the Southern Ocean (20–30 μM; Pakulski and Benner, 1994), the SAZ (47–48 μM; Dumont et al., 2011) and a previous experiment with bacterial EPS (48 μM; Hassler et al., 2011a). To compare C. simplex inorganic iron uptake rate, kupt was also measured in synthetic seawater, giving equivalent kupt as in a previous study (Hassler et al., 2011a). In addition, to verify that C. simplex ability to access organically bound iron species was identical with previous works performed in our lab (Hassler and Schoemann, 2009a, Hassler unpublished data), uptake rates were measured using other iron binding model compounds including bacterial siderophores (desferrioxamine B, DFB and ferrichrome, Ferri, 15 nM), saccharides (glucuronic acid, GLU; dextran, DEX, 15 nM), protoporphyrin (Protoporphyrin IX, PIX, 15 nM), DTPA (15 nM) and EDTA (1500 nM) in synthetic seawater (AQUIL media with only major salts). In this case, relative iron bioavailability was calculated using the ratio of uptake rate to total iron concentration (1 nM dFe; Hassler and Schoemann, 2009a; Shaked and Lis, 2012). All chemicals were high purity grade (Sigma). All experiments were carried out at 4 °C under 50 μmol photons m−2 s−1. In the field, 2 nM 55FeCl3 and 2 nM 55Fe pre-equilibrated with organic ligands for at least 1 week were added to natural seawater. Experimental treatments were control (C, unenriched seawater), inorganic iron addition (Fe, FeCl3 ICP standard, Fluka), iron bound to siderophore (DFB, 15 nM), monosaccharide (GLU, 15 nM) and bacterial EPS (EPS, 0.8 nM). For field experiments, as phytoplankton EPS were not available at the time of the expedition, a previously characterized (Hassler et al., 2011b) bacterial EPS from a pelagic Pseudoalteromonas sp. was used. As this bacterial EPS contained iron associated with it (2.2 nM iron per nM EPS, Hassler et al., 2011b), the experimental treatment consisted of 1.7 nM iron associated with EPS plus 0.3 nM inorganic iron. In the laboratory experiments, C. simplex was collected by gentle filtration (2 μm polycarbonate filters, Millipore; b 5 mm Hg), rinsed and resuspended to give 20,000 cells mL−1 in experimental solution as described in Hassler et al. (2011a, 2011b). The phytoplankton cells were then incubated at 4 °C under 50 μmol photons m−2 s−1 for 24 h. Incubation conditions for field work were at in situ temperature and light for 24 h. At the end of the incubation period, the sample was gently filtered (b 5 mm Hg) and rinsed using an EDTA-oxalate solution (5 × 3 mL, total contact time 10 min) to remove the adsorbed iron (Tovar-Sanchez et al., 2003; Hassler and Schoemann, 2009b) and then filtered seawater (3 × 3 mL). Filters were then collected in scintillation vials prior to addition of the scintillation cocktail (10 mL, Ultima Gold, Perkin Elmer) and subsequent analysis using a β-scintillation counter (Perkin Elmer, Tricarb 2900). Whereas C. simplex was collected on a 1.2 μm glass fiber filter (MicroAnalytix), natural phytoplankton were sequentially filtered (10 μm, 2 μm polycarbonate filters, Millipore and 0.7 μm glass fiber filters, MicroAnalytix). Intracellular disintegrations per minute were transformed to iron concentration using a custom made quench curve, total initial radioactivity and dFe concentration. 2.2.2. Phytoplankton growth and FV/FM For the laboratory experiments, 0.2 μm filtered Tasman Sea surface water (GP13 GEOTRACES voyage, 2011, 30 00 °S 167 00 °E) was enriched with inorganic iron (ICP standard, Fluka) or EPS to provide 1 nM total iron, in addition to the 0.56 nM present in the natural seawater, and left to equilibrate at 4 °C in the dark for 24 h. After equilibration the samples were transferred to 50 mL polycarbonate bottles to which iron-limited C. simplex was added to attain a cell density of ~ 40,000 cells mL− 1. Each treatment was prepared in triplicate. The algal suspensions were incubated at 4 °C, at a continuous light level of 50 μmol photons m−2 s−1 for 8 days. Cell counts were made using an

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

4

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

electronic particle counter (Multisizer II Coulter Counter, Beckman). Estimates of the maximum quantum yield of photochemistry (FV/FM) were made on phytoplankton cultures and natural assemblages using a Pulse Amplitude Modulated fluorometer (Water-PAM; Walz GMBH, Germany, Schreiber, 2004). A 3 mL sample was placed into a cylindrical quartz cuvette and dark-adapted for 15 min. During dark-adaptation, minimum fluorescence (Fo) was measured in the absence of actinic light, while the sample was stirred continuously. Once the fluorescence signal was stable, a saturating pulse was applied to give the dark-adapted maximum fluorescence (FM). FV/FM was calculated as (FM − Fo) / FM. 2.2.3. Phytoplankton community structure Pigment samples were collected using gentle sequential filtration (see above) and stored in cryo-vials in liquid nitrogen prior to analysis on shore. The pigments were extracted in 100% methanol in the dark and at 4 °C prior to analysis by High Performance Liquid Chromatography (Waters — Alliance system, 2695XE separations module with column heater and refrigerated autosampler and a 2996 photo-diode array detector) as in Hassler et al. (2014). The separated pigments were detected at 436 nm and identified against standard spectra using Waters Empower software. Concentrations of chlorophyll a (Chl a), chlorophyll b (Chl b) and β,β-carotene (β,β-car) in sample chromatograms were determined from standards obtained from Sigma, while all other pigment concentrations were determined from standards obtained from DHI, Denmark. Chl a data were used to determine the growth rate in each experimental treatment. The presence or absence of pigments that relate specifically to an algal class is termed biomarker or diagnostic pigments, and was used to provide a simple guide to the composition of the phytoplankton community (Jeffrey and Wright, 2006). Biomarker pigments identified in this study were Fucoxanthin, 19′ butanoloxylfucoxanthin, 19′ hexanoyloxyfucoxanthin, zeaxanthin, chlorophyll b (Chl b), prasinoxanthin, peridinin and divinyl chlorophyll a (DV Chl a). In this study, peridinin has been used as an indicator of dinoflagellates, but-fucoxanthin — pelagophytes, fucoxanthin — diatoms, prasinoxanthin — prasinophytes, hex-fucoxanthin — haptophytes, zeaxanthin — cyanobacteria of the type Synechococcus, Chl b — green algae and DV Chl a — cyanobacteria of the type Prochlorococcus. Flow cytometry analysis was conducted at NIWA (NZ) as in Hassler et al. (2014). In all experiments, statistical differences from the inorganic iron enrichment were identified using an unpaired t-test with a 0.05 significance level. 2.3. EPS impact on water chemistry following field incubation 2.3.1. dFe analyses The dFe concentrations were determined by isotope dilution multiple collector inductively coupled plasma mass spectrometry (ID-MC–ICP-MS) using a 54Fe spike according to the method described into details in de Jong et al. (2008). A pre-concentration step on micro-columns filled with NTA Superflow resin (Qiagen) was applied to 50 mL samples acidified at pH 1.9. The Nu Plasma MC–ICP-MS (Nu Instruments, Wrexham, UK) was operated at low resolution in dry plasma mode using an Aridus II desolvating sample inlet system (Cetac Technologies, Omaha, NE, USA). The limit of detection (3 standard deviations of the procedural blanks) for the session during which the samples were extracted was 0.029 nM. SAFe reference seawater (Johnson et al., 2007) Surface-1 (0.094 ± 0.008 nM) and Deep-2 (0.923 ± 0.029 nM) were analyzed simultaneously for quality control. The obtained values were in good agreement with the consensus values. 2.3.2. Chemical speciation of iron binding organic ligands Iron speciation was measured by Competitive Ligand Exchange Adsorptive Cathodic Stripping Voltammetry (CLE-AdCSV) following the method of Croot and Johansson (2000), using instrumentation and

procedures as detailed in Norman et al. (in press). For determination of the isolated EPS, samples were prepared in 0.2 μm filtered Tasman Sea surface seawater (seawater; GP13 GEOTRACES voyage, RV Southern Surveyor, May–June 2011, 30 00 °S 167 00 °E) collected using noncontaminating procedures as recommended by the GEOTRACES program (Norman et al., in press). Natural seawater and iron-binding organic ligand-enriched experimental samples from the PINTS voyage (Hassler et al., 2014) were prepared in the same way as in Norman et al. (in press) after defrosting at 4 °C in the dark, followed by recovery to ambient temperature. The detection limit was 0.05 nM iron. The concentrations and conditional stability constants (KFe′L) of the Fe′-binding ligands present were determined using the non-linear fit method of Gerringa et al. (1995) and further checked to within 10% of the data using a linearization method (van den Berg, 1982). Analysis of the 0.2 μm filtered seawater used revealed that the organic ligands naturally present did not significantly affect the results obtained for the EPS as these had a weak conditional stability constant in respect of Fe′-binding when compared to the EPS experimental treatments (Normal et al., 2014).

2.3.3. Analysis of iron-binding humic substance-like substances (HS-like) Determination of HS-like material was made using the voltammetric method of Laglera and van den Berg (2009) using instrumentation and procedures as per Norman et al. (in press). Suwannee River Fulvic Acid (SRFA, International Humic Substances Society, standard 1) was used as the HS-like standard for standard addition in each sample. The detection limit of the instruments was 1.56 μg SRFA L−1.

2.3.4. Iron size fractionation Iron size fractionation was measured in parallel of bioaccumulation experiments. Particulate (N0.2 μm), colloidal (0.02–0.2 μm) and soluble iron (b0.02 μm) were calculated after sequential syringe filtration at the end of the 24 h incubation on 0.2 μm (Millipore) and 0.02 μm (Anatop, Whatman) filters. Iron concentrations were calculated using total iron concentration, custom quench curve and counts per minute as for iron bioavailability experiments.

Table 2 Iron bioavailability of organically bound iron estimated using the comparison of the uptake rate constants (kupt) for organically bound iron with inorganic iron as per Hassler et al. (2011a, 2011b) or the iron uptake rate to the total dissolved iron concentration (in italic) as per Shaked and Lis (2012). Each experiment was done using iron-limited Chaetoceros simplex incubated for 24 h in inorganic seawater at 4 °C under 50 μmol photons m−2 s−1. Bacterial EPS were from Pseudoalteromonas sp. Experiments were run in duplicate and error represents the half-interval. Iron form

Bioavailability (%)

FeCl3a

100 28 113 84 100 128 79 365 437 420 100 123 124 72 9 8 61 11

Pelagic bacterial EPSa GLU 5 nMa DEX 5 nMa In situ ligands Pacific water Sea-ice bacterial EPS SAZ EPS Phaeocystis antarctica EPS Emiliana huxleyii EPS FeCl3 FeCl3 1 nM GLU 15 nM DEX 15 nM PIX 15 nM DFB 15 nM Ferri 15 nM DTPA 15 nM EDTA 1500 nM a

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

7 2 10 9 6 14 9 7 8 12 2 5 4 2 2 1 5 0

From Hassler et al. (2011a).

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

3. Results 3.1. EPS impact on C. simplex 3.1.1. Iron bioavailability Quantifying bioavailability of EPS-bound iron is the first step towards understanding its effect on phytoplankton growth. The bioavailability of iron associated with EPS and other organic ligands relative to inorganic iron was high for all EPS studied (Table 2). In some cases, bioavailability was enhanced beyond the bioavailability of insitu iron-ligands (Table 2). The bioavailability of all EPS-associated Fe was greater than the bioavailability of iron associated with the two hydroxamate siderophores tested, which decreased iron bioavailability by a factor greater than 10 relative to inorganic iron additions. 3.1.2. Growth and FV/FM All EPS additions increased the initial FV/FM of the iron-limited C. simplex (p b 0.05) and sustained specific growth rates similar to those observed under inorganic iron enrichment (Fig. 1A and B). In the presence of EPS, C. simplex was able to sustain its growth for longer, resulting in a greater final cell density (p b 0.03, Fig. 1A). This demonstrated that iron associated with EPS is efficient in sustaining iron biological uptake and growth as well as photosynthetic efficiency for an Antarctic diatom under laboratory conditions. 3.2. EPS impact on natural phytoplankton 3.2.1. Experimental conditions In the field, samples are highly complex both chemically and biologically, possibly resulting in different observations compared to laboratory studies. EPS that was previously purified from a Southern Ocean bacterium (Pseudoalteromonas sp.) and for which impact on iron biogeochemistry had been characterized (Hassler et al., 2011b) was used. Iron associated with that EPS was shown to be less bioavailable (28% implying that Fe-EPS bioavailability was 72% lower than FeCl3 bioavailability, Table 2) to C. simplex than the algal and natural EPS studied here. The deep fluorescence maximum at P1 was mostly dominated by picophytoplankton with a relatively high FV/FM (Table 1). dFe was low and organic iron-binding ligands were detected in large excess of dFe (excess L = 2.2 nM), but strong ligands of the class L1 were not detected. At P3, a suboptimal FV/FM was measured in a bloom of

A

5

phytoplankton (mainly coccolithophorids) in the nano-phytoplankton size fraction present in surface waters (Table 1, Fig. 2). dFe was greater at P3 than at P1, but a larger excess of organic ligands (2.6 nM) was detected (Table 1). Among the macronutrients, only silicate was lower at P3 compared to P1, suggesting a potential limiting role of silicate on diatom growth in the Sub-Antarctic Zone. 3.2.2. Iron bioavailability Size fractionated iron biological uptake in the presence of EPS, GLU and DFB compared with uptake in the presence of equivalent total inorganic iron addition was performed to verify whether EPS addition can enhance iron uptake rate as observed under well controlled laboratory conditions. EPS addition enhanced iron uptake rates by 1.3–2.6-fold relative to inorganic iron additions in all treatments and phytoplankton size classes, except for micro- and nano-phytoplankton at site P1 (Fig. 3). The monosaccharide GLU enhanced iron uptake rates for all phytoplankton size classes at P3 by 1.4 to 1.9-fold. In contrast, DFB decreased iron uptake rates for pico-phytoplankton (by 7 to 17-fold), nano-phytoplankton (by 27 to 63-fold), and micro-phytoplankton (by 9 to 40-fold). High iron solubility was maintained in all experimental treatments (Fig. 4). 3.2.3. Growth and FV/FM Phytoplankton growth rate and FV/FM at the conclusion of the 4 day incubation in the presence of inorganic iron addition compared to the control (no addition) treatment was used to assess the potential for iron limitation at both sites (Fig. 5A and B). Assuming that iron bioavailability was controlling the growth of iron-limited phytoplankton, a greater growth would be expected in the EPS treatment — the most bioavailable organic iron form tested (Fig. 3). Shipboard experiments demonstrated that inorganic iron enrichment was not efficient in promoting FV/FM or pico- and nano-phytoplankton growth at P1 (Fig. 6A and C). Surprisingly, the addition of EPS, GLU and DFB promoted phytoplankton growth, suggesting that growth enhancement was not attributed to iron bioavailability. In the SAZ (P3), FV/FM as well as the growth of nano- and microphytoplankton were significantly enhanced by inorganic iron addition, suggesting that the phytoplankton were iron-limited (p b 0.05, Fig. 5B and C). However, despite a lower FV/FM, DFB supported similar growth to inorganic iron addition. The addition of EPS and GLU enhanced the growth of pico-phytoplankton and micro-phytoplankton compared to

B

Fig. 1. Biological effect of EPS. Effect of a fixed EPS concentration on the cell density (A) and FV/FM (B) of an iron-limited diatom, Chaetoceros simplex. EPS are from phytoplankton cultures (Phaeocystis antarctica; Emiliana huxleyi) as well as from a phytoplankton bloom occurring in the Sub-Antarctic Zone and sea ice bacteria Pseudoalteromonas sp. Growth in the presence of EPS was compared to growth with inorganic iron addition. Growth was quantified at 4 °C in Tasman Sea seawater (167 0 °E, 30 0 °S) under 50 μmol photons m−2 s−1. Errors bars represent the standard deviation (n = 3).

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

6

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

Fig. 2. Structure of natural phytoplankton community in the Tasman Sea. Water was sampled at the depth of maximum fluorescence in the North Tasman Sea (P1, 165 0 °E, 30 0 °S) and the Sub-Antarctic Zone (P3, 159 5 °E, 46 2 °S). The dominating groups of phytoplankton were inferred from the analysis of biomarker pigments and total chlorophyll a (Tot Chl a). Phytoplankton was size fractionated into pico- (0.7–2 μm), nano- (2–10 μm) and micro-phytoplankton (N10 μm). Average of two analyses.

inorganic iron enrichment. As the growth of pico-phytoplankton at site P3, was not responsive to inorganic iron or Fe-DFB enrichment, but was to Fe-EPS addition, this effect could not solely be related to iron. 3.2.4. Phytoplankton community structure Due to dynamic interactions between microorganisms, the phytoplankton community structure at the end of the incubation was analyzed to assess the impact of various chemical forms of iron. At P1, enrichment with inorganic and EPS-associated iron significantly (p = 0.007 to 0.020) increased the relative biomass of large diatoms (Fig. 6B, fucoxanthin). Enrichment of iron associated with EPS compared to inorganic iron was less efficient in enhancing the growth of small diatoms (p = 0.042) but further increased the growth of small haptophytes (p = 0.005, Fig. 6A). Iron associated with GLU and DFB was efficient in enhancing the growth of small diatoms. At P3, enrichment with inorganic and EPS-associated iron significantly increased the relative biomass of small diatoms (p = 0.012 and 0.006) but decreased the relative biomass of haptophytes (p = 0.015) and dinoflagellates (p = 0.006 and 0.008). In the presence of any of the organic forms of iron tested here the biomass of nanodinoflagellates (peridinin) increased, whereas the relative biomass of nano-diatoms decreased. DFB and GLU enhanced small prasinophytes, DFB enhanced the biomass of nano-haptophytes, and GLU increased

A

the relative biomass of micro-dinoflagellates. Iron associated with EPS compared to inorganic iron significantly further enhanced the growth of nano-haptophytes (p = 0.037), and micro-pelagophytes (p = 0.025) but was less efficient in promoting the growth of microdiatoms (p = 0.011 and 0.045). Analyses by flow cytometry (Fig. 7) demonstrated that inorganic iron enrichment only enhanced the growth of pico-eukaryotes at P1 (p = 0.042; p = 0.059 to 0.288 for the other groups of pico-phytoplankton analyzed) as compared to the control. Iron associated with EPS significantly enhanced the growth for Prochlorococcus and pico-eukaryotes at P1 (p = 0.014 to 0.028) and Synechococcus and pico-eukaryotes at P3 (p = 0.006 to 0.023). In contrast, in comparison with inorganic iron enrichment, DFB and GLU had no significant effect on any of the pico-phytoplankton except at P1. 3.3. EPS impact on iron chemistry in field incubations Analysis of iron chemical speciation at the end of the field incubation experiments showed similarities at P1 and P3 (Table 3). There was less dFe in C and EPS compared to any other experimental treatments. Iron lability was higher and log KFe′L was lower in these two treatments. Stronger iron binding organic ligands and lower iron lability were measured in the DFB treatments. In addition, using the log KFe′L to differentiate strong L1 and weak L2 organic ligands, the weaker ligands were

B

Fig. 3. Impact of a fixed concentration of iron added as inorganic iron (Fe), Fe-glucuronic acid (GLU), Fe-desferrioxamine B (DFB), and Fe-EPS (EPS) on biological uptake rates. Natural phytoplankton communities from two contrasted regions were used, (A) the North Tasman Sea (P1, 165 0 °E, 30 0 °S) and (B) the Sub-Antarctic Zone (P3, 159 5 °E, 46 2 °S). Phytoplankton were separated by sequential filtration to differentiate pico-phytoplankton (0.7–2 μm, red bars), nano-phytoplankton (2–10 μm, white bars), and micro-phytoplankton (N10 μm, black bars). The EPS used for these experiments was EPS isolated from pelagic bacteria Pseudoalteromonas sp. Statistical difference (p b 0.05) with Fe and other treatments is shown by “x”. Error bars represent the half interval (n = 2). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

A

7

B

Fig. 4. Impact of inorganic (A) and organic (B) iron additions on iron solubility in the Tasman Sea (P1, 165 0 °E, 30 0 °S). Experimental treatments are iron added as inorganic iron (Fe), Feglucuronic acid (GLU), Fe-desferrioxamine B (DFB), and Fe-EPS (EPS). The average of 2 analyses is presented.

measured in the GLU and EPS treatments, whereas the stronger ligands were measured in the DFB treatments (data not shown). In response to any iron enrichment, the concentration of iron binding organic ligands increased, with greatest increase in the EPS treatment. Unlike iron binding organic ligands, HS-like material was greater at P3 than at P1. Iron enrichment did not result in an increase of HS-like material, except for the DFB and EPS treatments at P3 (p = 0.041). 4. Discussion 4.1. EPS impact on cultured and natural phytoplankton 4.1.1. Iron bioavailability Determination of the bioavailability of organically bound iron relies on an accurate measure of iron bioavailability in the presence of a fully bioavailable iron form. Inorganic iron is a dynamic fraction in seawater and the use of inorganic iron addition, instead of iron maintained

A

in solution by a small excess of EDTA (e.g., Maldonado et al., 2005), to determine the fully bioavailable fraction can be debated. In this study the lowest iron concentration that could be used was constrained by the filtered seawater iron background concentration and the specific activity or the 55FeCl3 source available. Given that iron solubility in inorganic seawater increases from 0.03 nM at 25 °C to 0.5 nM at 5 °C (Liu and Millero, 2002), the concentrations used here to determine the kupt of inorganic iron (0.6–2 nM) were marginally greater than its solubility. Considering that only soluble iron (as opposed to colloidal and particulate iron) was bioavailable in our experiment and an average of 4.6% soluble iron was measured in our experimental set up, (Norman et al., in press), then our approach could result in a 22-fold under-estimation of bioavailable iron concentration, resulting in a 1.3 fold decreased iron uptake rate in this filtered seawater (according to kupt measured). However, under our experimental conditions (24 h incubation), mostly soluble and freshly formed iron hydroxides were present, which are more bioavailable than aged iron colloids (Chen and Wang, 2001),

B

C

Fig. 5. Effect of iron additions following a four day on-deck incubation on total chlorophyll a (A and B) and FV/FM (C). Experimental treatments are control (unamended, C), 2 nM FeCl3 addition (Fe) and 2 nM Fe as Fe-EPS (EPS), Fe-glucuronic acid (GLU), Fe-desferrioxamine B (DFB) additions. Incubations were carried out in duplicate using phytoplankton present at the depth of fluorescence maximum in (A, C) the North Tasman Sea (P1, 165 0 °E, 30 0 °S) and (B, C) the Sub-Antarctic Zone (P3, 159 5 °E, 46 2 °S). Phytoplankton were separated by sequential filtration to differentiate pico-phytoplankton (0.7–2 μm, red bars), nano-phytoplankton (2–10 μm, white bars), and micro-phytoplankton (N10 μm, black bars). The EPS used for these experiments was the EPS isolated from the pelagic bacteria Pseudoalteromonas sp. Statistical difference (p b 0.05) with Fe and other treatments is shown by “x”. Error bars represent half interval (panels A and B, n = 2) and standard deviation (C, n = 4). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

8

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

Fig. 6. Effect of iron additions following a four day on-deck incubation of a natural phytoplankton community. Experimental treatments are control (unamended, C), 2 nM FeCl3 addition (Fe) and 2 nM Fe as Fe-EPS (EPS), Fe-glucuronic acid (GLU), Fe-desferrioxamine B (DFB) additions. Incubations were carried out using phytoplankton present at the depth of fluorescence maximum in (A, B) the North Tasman Sea (P1, 165 0 °E, 30 0 °S) and (C, D) the Sub-Antarctic Zone (P3, 159 5 °E, 46 2 °S). The dominating groups of phytoplankton following a 4-day ondeck incubation experiment were inferred from the analysis of biomarker pigments and total chlorophyll a (Tot Chl a). Phytoplankton was size fractionated into nano- (2–10 μm, A and C) and micro-phytoplankton (N10 μm, B and D). The EPS used for these experiments was the EPS isolated from the pelagic bacteria Pseudoalteromonas sp. Average of two analyses.

therefore the 22-fold under-estimation calculated above represents an upper limit of such an effect. Iron binding by organic ligands would therefore decrease the underestimation of iron bioavailability as it will prevent the formation of iron hydroxide colloids. For the measurement of iron bioavailability of the various organic iron forms (EPS, GLU, DFB; and the other ligands used here), iron solubility data (Hassler et al., 2011a; this study, Norman et al., in press) or organic ligand conditional stability constants (Rue and Bruland, 1995; Croot and Johansson, 2000; Norman et al., in press, this study) indicate that iron was indeed associated with the organic ligands. In our laboratory experiments with C. simplex, the concentration of EPS was kept constant and total iron concentrations were increased up to 3.6 nM. Using organic ligands data from Norman et al. (in press), the EPS used in this experimental setup corresponded to 4.3–12.7 nM iron-binding organic ligands. In this study, the high solubility measured

A

in the presence of organic ligands also supported that the iron added in the presence of EPS, GLU or DFB was indeed organically bound (Norman et al., in press). In the field, the presence of 2.2–2.6 nM excess iron binding organic ligands would have prevented the precipitation of the 2 nM inorganic iron added. This treatment thus corresponded to iron weakly bound to the in situ ligands, likely to be highly bioavailable (Maldonado et al., 2005; Shaked and Lis, 2012), and reducing iron bioavailability by a factor three, rather than inorganic iron. Our experimental design could thus have led to an under-estimation of inorganic iron bioavailability, resulting in an over-estimation of organic iron bioavailability (Chen and Wang, 2001; Maldonado et al., 2005). Although this could be estimated in the laboratory, the same exercise is more difficult for natural seawater. Indeed, the destruction of organic compounds and ligands by UV photo-oxidation will dramatically change iron bioavailability due to change in its solubility (Liu and Millero, 2002) but also due to the increase

B

Fig. 7. Effect of iron additions following a four day on-deck incubation on the structure of pico-phytoplankton community. Experimental treatments are control (unamended, C), 2 nM FeCl3 addition (Fe) and 2 nM Fe as Fe-EPS (EPS), Fe-glucuronic acid (GLU), Fe-desferrioxamine B (DFB) additions. Incubations were carried out using phytoplankton present at the depth of fluorescence maximum in (A) the North Tasman Sea (P1, 165 0 °E, 30 0 °S) and (B) the Sub-Antarctic Zone (P3, 159 5 °E, 46 2 °S). Distribution of pico-phytoplankton following a 4-day on-deck incubation experiment was measured using flow cytometry. Synechoccocus sp. (black bars), Prochlorococcus sp. (red bars) and pico-eukaryotes (green bars) were differentiated. Statistical difference (p b 0.05) with Fe and other treatments is shown by “x”. The average of duplicate experiments is shown with error bars corresponding to half-interval and statistical difference at the level of 0.05 from the control (*). The EPS used for these experiments was the EPS isolated from the pelagic bacteria Pseudoalteromonas sp. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx Table 3 Iron biogeochemistry associated with enrichment experiments using phytoplankton communities collected from the Tasman Sea, P1 (30.0 °S, 156.0 °E) and P3 (46.2 °S, 159.5 °E). Concentration of organic ligands associated with strong Fe-binding affinities ([L1]), and the sum of all organic ligands ([sum L]), together with their respective calculated conditional stability constants with respect to inorganic Fe (Fe′; log KFe′L1, and log KFe′sumL), concentration of dissolved Fe (dFe, nM) and relative concentration (%) of labile Fe (FeLabile) with respect to 10 μM TAC exchange ligand are presented. Electrochemically detected humics (HS-like) are also presented as Suwanee River Fulvic Acid (SRFA) equivalent as SRFA was used for standard addition in each sample. Treatments were measured after 4 day incubation and comprised an unamended control (C), inorganic Fe (2 nM), and organic ligands desferrioxamine B (DFB [15 nM]) natural pelagic bacterial exopolymeric substances (EPS, [0.8 nM]), glucuronic acid (GLU [15 nM]). DFB, EPS, and GLU treatments also contained 2 nM inorganic Fe. Errors represent the half interval of duplicate samples. Where no error is present, the values are from a single sample. Treatment dFe (nM)

FeLabile %

P1 C Fe DFB EPS GLU

0.31 1.91 2.05 0.98 2.49

± 0.06 85 ± ± 0.01 32 ± ± 0.42 13 ± 82 ± 0.01 43 ±

5 11 5

P3 C Fe DFB EPS GLU

0.27 1.70 1.98 0.24 0.96

± 0.14 67 ± ± 0.28 28 ± ± 0.29 16 ± 84 ± 0.07 49 ±

12 9 6

[sum L] (nM)

log KFe′sumL

HS-like (μg L−1 SRFA eq.)

3.96 4.75 8.71 39.94 5.76

± ± ± ± ±

1.14 0.28 0.04 7.52 0.47

11.38 11.86 12.00 10.43 11.53

± ± ± ± ±

0.17 0.04 0.16 0.10 0.01

12.17 11.96 16.52 12.00 14.38

± ± ± ± ±

0.39 1.58 4.99 0.33 2.55

4.19 4.46 7.33 11.85 14 4.86

± ± ± ± ±

0.08 0.26 2.10 0.73 0.17

11.39 11.84 11.99 11.03 11.49

± ± ± ± ±

0.03 0.07 0.13 0.15 0.03

36.17 32.00 50.86 75.72 33.97

± ± ± ± ±

7.82 3.88 1.56 2.44 2.57

2

in the concentration of other inorganic elements known to affect iron uptake such as Cu for example (e.g., Maldonado et al., 2006). Iron bioavailability in the presence of EPS was enhanced beyond the bioavailability of in situ iron-binding organic ligands, likely due to a combination of improved iron solubility and the formation of highly bioavailable forms, as previously observed for saccharides (Hassler et al., 2011a). Indeed, the kupt for C. simplex measured here in filtered seawater (hence bioavailability of in situ ligands) is 4.2-fold smaller than the kupt measured in synthetic seawater (inorganic iron only, Hassler et al., 2011a), suggesting that the in situ organic ligands present in the filtered seawater used to measure bioavailability of iron associated with the EPS were only 23% bioavailable. In this case, the large enhanced bioavailability observed in the presence of the EPS from P. antarctica and E. huxleyi is the results of iron bioavailability close to that of inorganic iron. As previously observed in the laboratory using cultured isolates or in the field using phytoplankton communities from the Southern Ocean, DFB decreased iron bioavailability by a factor greater than 10, whereas iron associated with saccharides and EPS was highly bioavailable (Hassler et al., 2011a; Shaked and Lis, 2012). Iron bioavailability of other organic iron forms is similar to previously observed iron uptake rates for Chaetoceros sp. (Hassler and Schoemann, 2009a) which was recently identified as C. simplex (S. Trimborn, pers. comm.). It is to be noted that acetic acid reflux was not used to ensure optimal iron binding to the PIX (Hopkinson et al., 2008), which would therefore have resulted in a partial iron binding in our study and thus a potentially biased estimation of the relative iron bioavailability of iron associated with PIX. 4.1.2. Growth and FV/FM The fact that the growth rate of C. simplex was similar for all experimental treatments suggested that iron associated with EPS was able to sustain its growth equally to the inorganic iron additions. Results agree with a high bioavailability of Fe-EPS (this study, Hassler et al., 2011b). Considering that maximum yield, the cell density at which cells are in stationary phase, is a function of total Fe available, data suggested that iron bioavailability was greater in the presence of EPS than in the presence of FeCl3 additions following 7 day incubation. Following

9

inorganic iron additions, inorganic colloids could have been formed and aged during the incubation. Iron inorganic colloidal forms that were poorly bioavailable (Chen and Wang, 2001) could have thus contributed to the decrease in FV/FM and cell yield in this experimental treatment. Unfortunately, iron size fractionation was not measured at the end of the incubation, preventing the verification of this hypothesis. Despite the fact that addition of Fe-EPS enhanced the growth of phytoplankton further than inorganic Fe additions, the measured effect on growth and FV/FM in the various experimental treatments were not necessarily in accordance with the prediction based on their iron bioavailability. These observations contributed to an apparent non-iron related phytoplankton response to organic ligand additions in our field experiments, suggesting that iron bioavailability is not the only factor regulating phytoplankton growth in the Tasman Sea. Below we explore three hypotheses that could help explain this: (i) enrichment of other potentially limiting nutrients with EPS addition, (ii) shift in phytoplankton community during incubations, and (iii) change in nature of organic ligands as a result of biological activity and photo-chemistry.

4.1.2.1. Role of other nutrients. The EPS studied here (Norman et al., in press), as well as the bacterial EPS used in the field experiments (Hassler et al., 2011b), all include significant amounts of macronutrients and trace elements, other than iron, for which possible phytoplankton growth limitation has previously been reported (e.g., Morel et al., 1994; Saito et al., 2002). A significant release of 150 nM NOx (p = 0.007) was indeed measured subsequent to the bacterial EPS enrichment at station P1, which is significant in comparison to the NOx measured in the upper water column (b 10 nM, M. Woodward, personal communication). However, such NOx enrichment is unlikely to be the cause for the large phytoplankton growth observed at station P1. In addition, no significant increase in NOx was measured in the GLU and DFB treatments and no differences between FV/FM was observed in any experimental treatment. Comparing the nutrients associated with the purified SAZ EPS and their in situ concentrations (Table 4), 15%, 30% and 3% of the dissolved Zn, Co, and Cu respectively are associated with EPS, whereas EPS contribute to less than 1% of the dissolved NOx and PO4. The bacterial EPS used in the field incubation however, contributed to a large enrichment of Zn, Co and Cu. Phytoplankton Zn and/or Co growth limitation have been observed in the laboratory (Sunda and Huntsman, 1995; Saito et al., 2002; Saito and Goepfert, 2008) and speculated in the SAZ region (e.g., Hassler et al., 2012b; Sinoir et al., 2014). Despite the fact that Zn and Co can be interchanged in enzymatic reactions, a pattern in their biological requirement does exist. While diatoms cannot grow without Zn (e.g., Sunda and Huntsman, 1995), E. huxleyi or P. antarctica can replace it with cobalt (Timmermans et al., 2001; Saito and Goepfert, 2008), and cyanobacteria require Co but not Zn (Saito et al., 2002). Deck incubations are still scarce for Co and Zn studies, however, they have revealed their importance in phytoplankton communities by inducing species shifts, reduced chlorophyll a concentrations or Cd:PO4 ratios (Cullen and Sherrell, 2005; Wisniewski-Jakuba et al., 2012). Extremely low dissolved Zn and Zn′ concentrations were reported in the Tasman Sea Table 4 Dissolved trace elements measured at the depth of maximum fluorescence in the North Tasman Sea (P1, 165 0 °E, 30 0 °S) and the Sub-Antarctic Zone (P3, 159 5 °E, 46 2 °S). Zn data were from Sinoir et al. (2014) and other trace elements were determined as per O'Sullivan et al. (2013) and personal communication from E.C.V. Butler. Note that UV– photo-oxidation was not conducted prior to Co determinations. Enrichment made with the bacterial EPS was calculated as per Hassler et al. (2011b) considering a 0.8 nM EPS addition.

P1 P3 Bacterial EPS

Zn (pM)

Co (pM)

Cu (nM)

Ni (nM)

14 170 529

120 4.0 22

0.4 0.4 0.8

2.3 3.1 0.1

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

10

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

(Ellwood, 2008; Hassler et al., 2012b; Sinoir et al., 2014), suggesting that it can be limiting phytoplankton growth. The bacterial EPS addition, brings a significant manganese (Mn, 39 pM) concentration, based on previous measurements in the Australian SAZ (100–500 pM, Sedwick et al., 1997). Mn is reported as a potential limiting nutrient for the Southern Ocean (Moore et al., 2013). It is therefore likely that some of the high growth observed in response to EPS enrichment is due to the enrichment of other co-limiting nutrients. However, DFB and GLU have low background contamination b 0.1 pM for Mn, Co, Ni, Cu and Zn and resulted in negligible dFe enrichments of 1.4 pM and 0.7 pM, respectively (data not shown). In these regions the production and function of EPS may help to alleviate a degree of nutrient stress. In addition, measurements of bacterial abundance and remineralization of DOM and DOC during phytoplankton blooms and non-bloom periods indicate that EPS may also provide heterotrophic bacteria with a source of organic carbon (Amon and Benner, 1994; Morán et al., 2001; Obernosterer et al., 2008). This cannot be overlooked as iron remineralized through the grazing and viral lysis of bacteria has been found to be highly bioavailable to phytoplankton and may make a substantial contribution to the dFe pool in remote ocean regions (Barbeau et al., 1996; Poorvin et al., 2004; Sarthou et al., 2008; Strzepek et al., 2005). 4.1.2.2. Role of community structure. Because EPS contains several nutrients (Hassler et al., 2011b; Norman et al., in press), DFB strongly reduces iron bioavailability, and different algal taxa exhibit different biological nutrient requirements for growth (e.g., Twining and Baines, 2013), one would expect a shift in the algal community concomitant with the effect on algal growth or iron bioavailability. Indeed, flow cytometry data and analyses of biomarker pigments as a diagnostic of the composition of natural phytoplankton communities demonstrated a shift in phytoplankton in the presence of EPS, relative to inorganic iron that differed between sites and phytoplankton size classes. Whereas flow cytometry data demonstrated that EPS strongly impacted the structure of natural pico-phytoplankton communities, for which it had the greatest increase in growth and iron bioavailability. Biomarker analyses clearly highlighted a complex response of nano- and microphytoplankton to the experimental treatments, with differences in the relative contribution of functional groups between the organic forms of iron and sites. This response likely reflected different biological requirements and biological affinities for iron (Ho et al., 2003; Twining and Baines, 2013). However, biomarker analyses do not allow the identification of the dominant phytoplankton species or a change within a functional phytoplankton group. Unfortunately no samples were taken for microscopic identification. The fact that, in the EPS treatment, the growth of small diatoms (compared to Fe treatment) decreased at station P1, but increased at P3, and the GLU and DFB treatments enhanced the growth of small diatoms at P1 but had no significant effect at P3, demonstrates the complexity at play. 4.1.2.3. Role of chemical and biological transformation of organic ligands added. Natural communities in the field contained viruses, bacteria, phytoplankton, and possibly mixotrophs and small zooplankton — all interacting together creating a dynamic environment. Bacteria and viruses rapidly respond to changes in nutrients such as iron or organic carbon (e.g., Stocker et al., 2008), and symbiotic interactions for the acquisition of iron and carbon resources have been reported between bacteria and phytoplankton (e.g., Amin et al., 2009). Biological recycling as a result of other microorganisms present in the experimental bottles is known to happen quickly (Boyd et al., 2010) and to be efficient in recycling iron as well as organic ligands (McKay et al., 2005; Boyd et al., 2010; Hassler et al., 2012a for a review). It is therefore expected that our experimental conditions changed within the four days of incubation. Based on iron chemical speciation (Table 4), typical signatures are maintained. The concentration of organic ligands was greater when iron was added in conjunction with an organic ligand compared to inorganic iron addition, and iron addition stimulated the production

of organic ligands as previously reported (Rue and Bruland, 1997; Boye et al., 2005). Iron chemical speciation was therefore in agreement with previous data showing that saccharides and EPS are strong contributors to weak iron binding organic ligands (Hassler et al., 2011a, 2011b, Norman et al., in press). The log of KFe′L measured in the DFB treatments was close to the previously reported data using an identical method in synthetic seawater (12.28, Hassler et al., 2013) and in UV–photo-oxidized Southern Ocean water (11.8, Maldonado et al., 2005). As expected from the iron bioavailability and KFe′L, iron lability followed the order EPS N GLU N Fe N DFB. Concomitant with weak organic ligands, high iron lability and high chlorophyll a and iron uptake rates, important iron consumption were measured in the EPS treatments at P1 and P3 and in the GLU treatment at P3. These data highlight interesting differences between iron binding HS-like material and iron-binding organic ligands detected by the CLE AdCSV technique worth to be further explored in the future. 5. Conclusion This study demonstrates that EPS potentially exerts a significant effect on iron biogeochemistry in the ocean. EPS is not associated with a new iron input in surface waters, but as EPS excreted can bind iron, enhance its solubility and its bioavailability and possibly its recycling, EPS could contribute to relieve nutrient limitation, enhancing phytoplankton growth and modifying community structure. It is clear that the effect of EPS will be highly dependent on water chemistry, iron nutritive status and biological requirement of the phytoplankton community and trophic interactions at play. This study advances our understanding of the regulation of oceanic iron bioavailability from bacterial siderophores to EPS produced by photosynthetic eukaryotes. The role of EPS as iron-binding organic ligands potentially fuelling regenerated production was until now mostly ignored. Biogeochemical and ecosystem models have been actively searching a “cost effective” solution to implement the role of DOC and organic ligands in iron biogeochemistry (Ye et al., 2009; Tagliabue and Völker, 2011) with consequences for oceanic biological growth limitation, atmospheric carbon dioxide fixation and ultimately climate regulation. This study points out that EPS, as an organic material resulting from a by-product of biological activity, can be important for iron biogeochemistry. EPS by excretion and iron binding could be important for iron recycling. Their excretion and turnover rates as well as their sensitivity to photo-degradation in relation to iron chemistry still remain largely unknown. Because EPS benefited phytoplankton in most of our experiments but studies on the impact of iron-limitation on EPS excretion rate in nature have not yet been made, the question on the evolutionary benefit of EPS production and identification of EPS as a waste and storage product or a specific iron-regulation switch is still open. N limitation and P limitation have been documented to influence EPS excretion (Myklestad, 1995; Penna et al., 1999). Material excreted by microorganism is efficient in binding iron, as well as enhancing its bioavailability, a requisite to sustaining phytoplankton growth in iron-limited regions. However, our results demonstrated that iron bioavailability or speciation cannot solely be used to predict the impact on phytoplankton growth. Other parameters such as nutrient co-limitation, specific iron requirement for growth and the biological recycling of iron and ligands need to be considered. Acknowledgments We thank the Australian Research Council (Discovery Project DP1092892 and LIEF grant LE0989539) as well as the CSIRO OCE and UTS Chancellor Fellowship schemes for funding. VS was financed by a COFUND Marie Curie fellowship “Back to Belgium Grant”. Emilie Angles, Jeroen de Jong and Malcolm Woodward, Cliff Law, Jeanette O'Sullivan, Edward Butler and Peter Nichols are thanked for EPS isolation, dFe, macronutrients, flow cytometry, trace metals and lipid analyses,

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

respectively. Warm thanks also go to the CSIRO Hydrochemistry team, the officers and crew of the RV Southern Surveyor and the Marine National Facility support team. References Aluwihare, L.I., Repeta, D.J., Chen, R.F., 1997. A major biopolymeric component to dissolved organic carbon in surface seawater. Nature 387, 166–169. Amin, S.A., Green, D.H., Hart, M.C., Kupper, F.C., Sunda, W.G., Carrano, C.J., 2009. Photolysis of iron-siderophore chelates promotes bacterial-algal mutualism. Proc. Nat. Acad. Sci. 106, 17071–17076. Amon, R.M.W., Benner, R., 1994. Rapid cycling of high-molecular-weight dissolved organic matter in the ocean. Nature 369, 549–552. Barbeau, K., Moffett, J.W., Caron, D.A., Croot, P.L., Erdner, D.L., 1996. Role of protozoan grazing in relieving iron limitation of Phytoplankton. Nature 380, 61–64. Barbeau, K., Rue, E.L., Bruland, K.W., Butler, A., 2001. Photochemical cycling of iron in the surface ocean mediated by microbial iron(III)-binding ligands. Nature 413, 409–413. Benner, R., 2011. Loose ligands and available iron in the ocean. Proc. Natl. Acad. Sci. U. S. A. 108, 893–894. Biersmith, A., Benner, R., 1998. Carbohydrates in phytoplankton and freshly produced dissolved organic matter. Mar. Chem. 63, 131–144. Bowie, A.R., Griffiths, F.B., Dehairs, F., Trull, T.W., 2011. Oceanography of the subantarctic and Polar Frontal Zones south of Australia during summer: setting for the SAZ-Sense study. Deep-Sea Res. II 58, 2059–2070. Boyd, P.W., Ibisanmi, E., Sander, S.G., Hunter, K.A., Jackson, G.A., 2010. Remineralisation of upper ocean particles: implications for iron biogeochemistry. Limnol. Oceanogr. 55, 1271–1288. Boye, M., van den Berg, C.M.G., 2000. Iron availability and the release of iron-complexing ligands by Emiliania huxleyi. Mar. Chem. 70, 277–287. Boye, M., et al., 2005. Major deviations of iron complexation during 22 days of a mesoscale iron enrichment in the open Southern Ocean. Mar. Chem. 96, 257–271. Chen, M., Wang, W.-X., 2001. Bioavailability of natural colloid-bound iron to marine plankton: influences of colloidal size and aging. Limnol. Oceanogr. 46, 1956–1967. Croot, P.L., Johansson, M., 2000. Determination of iron speciation by cathodic stripping voltammetry in seawater using the competing ligand 2-(2-thiazolylazo)-p-cresol (TAC). Electroanalysis 12, 565–576. Cullen, J.T., Sherrell, R.M., 2005. Effects of dissolved carbon dioxide, zinc, manganese on the cadmium to phosphorus ratio in natural phytoplankton assemblages. Limnol. Oceanogr. 50, 1193–1204. Cutter, G., et al., 2010. Sampling and sample-handling protocols for GEOTRACES cruises. In: GEOTRACES Standards and Intercalibration Committee (Ed.), GEOTRACES Library (Available at http://www.geotraces.org/libraries/documents/Intercalibration/ Cookbook.pdf). Decho, A.W., 1990. Microbial exopolymer secretions in ocean environments: their role(s) in food webs and marine processes. Oceanogr. Mar. Biol. Annu. Rev. 28, 73–153. de Jong, J., Schoemann, V., Lannuzel, D., Tison, J.-L., Mattielli, N., 2008. High-Accuracy determination of iron in sea water by Isotope Dilution Multi Collector Induced Couple Plasma Mass Spectrometry (ID-MC-ICP-MS) using nitriloacetic acid chelating resin for pre-concentration and matrix separation. Anal. Chim. Acta 623, 126–139. Dumont, I., Schoemann, V., Jacquet, S.H.M., Masson, F., Becquevort, S., 2011. Bacterial abundance and production in epipelagic and mesopelagic waters in the Subantarctic and Polar Front zones south of Tasmania. Deep Sea Res. Part II 58, 2212–2221. Ellwood, M.J., 2008. Wintertime trace metal (Zn, Cu, Ni, Cd, Pb and Co) and nutrient distributions in the Subantarctic Zone between 40–52°S; 155–160°E. Mar. Chem. 112, 107–117. Gerringa, L.J.A., Herman, P.M.J., Poortvliet, T.C.W., 1995. Comparison of the linear Van den Berg/Ružić transformation and a non-linear fit of the Langmuir isotherm applied to Cu speciation data in the estuarine environment. Mar. Chem. 48, 131–142. Gledhill, M., Buck, K.N., 2012. The organic complexation of iron in the marine environment: a review. Front. Microbiol. 3, 69. http://dx.doi.org/10.3389/fmicb.2012.00069. Granger, J., Price, N.M., 1999. The importance of siderophores in iron nutrition of heterotrophic marine bacteria. Limnol. Oceanogr. 44, 541–555. Hassler, C.S., Schoemann, V., 2009a. Discriminating between intra- and extracellular metals using chemical extractions: an update on the case of iron. Limnol. Oceanogr. Methods 7, 479–489. Hassler, C.S., Schoemann, V., 2009b. Bioavailability of organically bound Fe to model phytoplankton of the Southern Ocean. Biogeosciences 6, 2281–2296. Hassler, C.S., Schoemann, V., Mancuso Nichols, C.A., Butler, E.C.V., Boyd, P.W., 2011a. Saccharides enhance iron bioavailability to southern ocean phytoplankton. Proc. Natl. Acad. Sci. U. S. A. 108, 1076–1081. Hassler, C.S., Alasonati, E., Mancuso Nichols, C.A., Slaveykova, V.I., 2011b. Exopolysaccharides produced by bacteria isolated from the pelagic Southern Ocean — role in iron binding, chemical reactivity and bioavailability. Mar. Chem. 123, 88–98. Hassler, C.S., Schoemann, V., Boye, M., Tagliabue, A., Rozmarynowycz, M., McKay, R.M.L., 2012a. Iron bioavailability in the Southern Ocean. Oceanogr. Mar. Biol. Annu. Rev. 50, 1–64. Hassler, C.S., Sinoir, M., Clementson, L.A., Butler, E.C.V., 2012b. Exploring the link between micro-nutrients and phytoplankton in the Southern Ocean during the 2007 austral summer. Front. Microbiol. 3, 1–26. Hassler, C.S., Legiret, F.-E., Butler, E.C.V., 2013. Measurement of iron chemical speciation in seawater at 4 °C: the use of competitive ligand exchange — adsorptive cathodic stripping voltammetry. Mar. Chem. 149, 63–73. Hassler, C.S., et al., 2014. Primary productivity induced by iron and nitrogen in the Tasman Sea—an overview of the PINTS expedition. Mar. Freshwater Res. 65, 517–537.

11

Ho, T.Y., Quigg, A., Finkel, Z.V., Milligan, A.J., Wyman, K., Falkowski, P.G., Morel, F.M.M., 2003. The elemental composition of some marine phytoplankton. J. Phycol. 39, 1145–1159. Hoagland, K.D., Rosowski, J.R., Gretz, M.R., Roemer, S.C., 1993. Diatom extracellular polymeric substances: function, fine structure, chemistry, and physiology. J. Phycol. 29, 537–566. Hopkinson, B.M., Roe, K.L., Barbeau, K., 2008. Heme uptake by Microscilla marina and evidence for heme uptake systems in the genomes of diverse marine bacteria. Appl. Environ. Microbiol. 74, 6263–6270. Hutchins, D.A., Witter, A., Butler, A., Luther III, G.W., 1999. Competition among marine phytoplankton for different chelated iron species. Nature 400, 858–861. Jeffrey, S.W., Wright, S.W., 2006. In: Subba Rao, D.V. (Ed.), Algal Cultures, Analogues of Blooms and Applications. Science Publisher, Enfield, pp. 33–90. Johnson, K.S., Boyle, E., Bruland, K., Coale, K., Measures, C., et al., 2007. Developing standards for iron in seawater. EOS 88, 131–132. Laglera, L.M., van den Berg, C.M.G., 2009. Evidence for the geochemical control of iron by humic substances in seawater. Limnol. Oceanogr. 54, 610–619. Liu, X., Millero, F.J., 2002. The solubility of iron in seawater. Mar. Chem. 77, 43–54. Maldonado, M.T., Price, N.M., 2000. Nitrate regulation of Fe reduction and transport by Felimited Thalassiosira oceanica. Limnol. Oceanogr. 45, 814–826. Maldonado, M.T., Strzepek, R.F., Sander, S., Boyd, P.W., 2005. Acquisition of Fe bound to strong organic complexes, with different Fe binding groups and photochemical reactivities, by plankton communities in Fe-limited subantarctic waters. Glob. Biogeochem. Cycles 19, GB4S23. http://dx.doi.org/10.1029/2005GB002481. Maldonado, M.T., Allen, A.E., Chong, J.S., Lin, K., Leus, D., Karpenko, N., Harris, S.L., 2006. Copper-dependent iron transport in coastal and oceanic diatoms. Limnol. Oceanogr. 51, 1729–1743. Mancuso Nichols, C.A., Garon, S., Bowman, J.P., Raguénès, G., Guézennec, J., 2004. Production of exopolysaccharides by Antarctic marine bacterial isolates. J. Appl. Microbiol. 96, 1057–1066. Mancuso Nichols, C., et al., 2005. Chemical characterization of exopolysaccharides from antarctic marine bacteria. Microb. Ecol. 49, 578–589. McKay, R., Wilhelm, S., Hall, J., Hutchins, D., Al-Rshaidat, M., Mioni, C., Pickmere, S., Porta, D., Boyd, P.W., 2005. Impact of phytoplankton on the biogeochemical cycling of iron in subantarctic waters southeast of New Zealand during FeCycle. Global Biogeochem. Cycles 19. http://dx.doi.org/10.1029/2005GB002482 (GB4S24.1-GB4S24.14). Moore, C.M., et al., 2013. Processes and patterns of oceanic nutrient limitation. Nat. Geosci. 6, 701–710. Morán, X.A.G., Gasol, J.M., Pedrós-Alió, C., Estrada, M., 2001. Dissolved and particulate primary production and bacterial production in offshore Antarctic waters during austral summer: coupled or uncoupled? Mar. Ecol. Prog. Ser. 222, 25–39. Morel, F.M.M., et al., 1994. Zinc and carbon co-limitation of marine phytoplankton. Nature 369, 740–742. Morrissey, J., Bowler, C., 2012. Iron utilization in marine cyanobacteria and eukaryotic algae. Front. Microbiol. 3, 43. http://dx.doi.org/10.3389/fmicb.2012.00043. Myklestad, S.M., 1995. Release of extracellular products by phytoplankton with special emphasis on polysaccharides. Sci. Total Environ. 165, 155–164. Norman, L., Worms, I.A.M., Angles, E., Bowie, A.R., Mancuso Nichols, C., Pham, A.N., Slaveykova, V.I., Townsend, A.T., Waite, D., Hassler, C.S., 2014. The role of bacterial and algal exopolymeric substances (EPS) on iron chemistry. Mar. Chem, (in press). Obernosterer, I., Chritaki, U., Lefèvre, D., Catala, P., Van Wambeke, F., Lebaron, P., 2008. Rapid bacterial mineralization of organic carbon produced during a phytoplankton bloom induced by natural iron fertilization in the Southern Ocean. Deep-Sea Res. II 55, 777–789. O'Sullivan, J.E., Watson, R.J., Butler, E.C.V., 2013. An ICP-MS procedure to determine Cd, Co, Cu, Ni, Pb and Zn in oceanic waters using in-line flow-injection with solid-phase extraction for preconcentration. Talanta 115, 999–1010. Pakulski, D.J., Benner, R., 1994. Abundance and distribution of carbohydrates in the ocean. Limnol. Oceanogr. 39, 930–940. Penna, A., Berluti, S., Penna, N., Magnani, M., 1999. Influence of nutrient ratios on the in vitro extracellular polysaccharide production by marine diatoms from the Adriatic Sea. J. Plankton Res. 21, 1681–1690. Petrou, K., Trimborn, S., Rost, B., Ralph, P., Hassler, C.S., 2014. The impact of iron limitation on the physiology of the Antarctic diatom Chaetoceros simplex. Mar. Biol. 161, 925–937. Poorvin, L., Rinta-Kanto, J.M., Hutchins, D.A., Wilhelm, S.W., 2004. Viral release of iron and its bioavailability to marine plankton. Limnol. Oceanogr. 49, 1734–1741. Rijkenberg, M.J.A., et al., 2008. Enhancement of the reactive iron pool by marine diatoms. Mar. Chem. 109, 29–44. Rue, E.L., Bruland, K.W., 1995. Complexation of iron(II1) by natural organic ligands in the Central North Pacific as determined by a new competitive ligand equilibration/adsorptive cathodic stripping voltammetric method. Mar. Chem. 50, 117–138. Rue, E.L., Bruland, K.W., 1997. The role of organic complexation on ambient iron chemistry in the Equatorial Pacific Ocean and the response of a mesoscale iron addition experiment. Limnol. Oceanogr. 42, 901–910. Rue, E., Bruland, K., 2001. Domoic acid binds iron and copper: a possible role for the toxin produced by the marine diatom Pseudo-nitzschia. Mar. Chem. 76, 127–134. Saito, M.A., Goepfert, T.J., 2008. Zinc–cobalt colimitation of Phaeocystis antarctica. Limnol. Oceanogr. 53, 266–275. Saito, M.A., Moffett, J.W., Chisholm, S.W., Waterbury, J.B., 2002. Cobalt uptake in Prochlorococcus. Limnol. Oceanogr. 47, 1629–1636. Santschi, P.H., et al., 2003. Control of acid polysaccharide production and 234Th and POC export fluxes by marine organisms. Geophys. Res. Lett. 30, 1044. http://dx.doi.org/ 10.1029/2002GL016046. Sarthou, G., Vincent, D., Christaki, U., Obernosterer, I., Timmermans, K.R., Brussaard, C.P.D., 2008. The fate of biogenic iron during a phytoplankton bloom induced by natural

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002

12

C.S. Hassler et al. / Marine Chemistry xxx (2014) xxx–xxx

fertilisation: impact of copepod grazing. Deep-Sea Res. 55, 734–751. http://dx.doi. org/10.1016/j.dsr2.2007.12.033. Schreiber, U., 2004. In: Papageorgiou, G.C., Govinjee (Eds.), Chlorophyll Fluorescence: A Signature of Photosynthesis. Springer, Dordrecht, pp. 279–313. Sedwick, P.N., Edwards, P.R., Mackey, D.J., Griffiths, F.B., Parslow, J.S., 1997. Iron and manganese in surface waters of the Australian subantarctic region. Deep-Sea Res. I 44, 1239–1253. Shaked, Y., Lis, H., 2012. Disassembling iron availability to phytoplankton. Front. Microbiol. 3, 123. http://dx.doi.org/10.3389/fmicb.2012.00123. Sinoir, M., et al., 2014. Zinc cycling in the Tasman Sea: distribution, speciation and relation to phytoplankton community. Mar. Chem. (in revision). Stocker, R., Seymour, J.R., Samadani, A., Hunt, D.E., Polz, M.F., 2008. Rapid chemotactic response enables marine bacteria to exploit ephemeral nutrient patches. Proc. Natl. Acad. Sci. U. S. A. 105, 4209–4214. Strzepek, R.F., Maldonado, M.T., Higgins, J.L., Hall, J., Safi, K., Wilhelm, S.W., Boyd, P.W., 2005. Spinning the “Ferrous Wheel”: the importance of the microbial community in an iron budget during the FeCycle experiment. Global Biogeochem. Cycles 19, GB4S26. http://dx.doi.org/10.1029/2005GB002490. Sunda, W.G., Huntsman, S.A., 1995. Iron uptake and growth limitation in oceanic and coastal phytoplankton. Mar. Chem. 50, 189–206. Tagliabue, A., Völker, C., 2011. Towards accounting for dissolved iron speciation in global ocean models. Biogeosci. Discuss. 8, 2775–2810.

Timmermans, K.R., et al., 2001. Not all eukaryotic algae can replace zinc with cobalt: Chaetoceros calcitrans (Bacillariophyceae) versus Emiliania huxleyi (Prymnesiophyceae). Limnol. Oceanogr. 46, 699–703. Tovar-Sanchez, A., Sanudo-Wilhelmy, S.A., Garcia-Vargas, M., Weaver, R.S., Popels, L.C., Hutchins, D.A., 2003. A trace metal clean reagent to remove surface-bound iron from marine phytoplankton. Mar. Chem. 82, 91–99 (Published erratum, Mar. Chem. 2004;85:191). Twining, B.S., Baines, S.B., 2013. The trace metal composition of marine phytoplankton. Annu. Rev. Mar. Sci. 5, 191–215. van den Berg, C.M.G., 1982. Determination of copper complexation with natural organic ligands in seawater by equilibration with MnO2 I. Theory. Mar. Chem. 11, 307–322. Verdugo, P., et al., 2004. The oceanic gel-phase: a bridge in the DOM–POM continuum. Mar. Chem. 92, 67–85. Wisniewski-Jakuba, R., et al., 2012. Dissolved zinc in the subarctic North Pacific and Bering Sea: its distribution, speciation, and importance to primary producers. Glob. Biogeochem. Cycles 26. http://dx.doi.org/10.1029/2010GB004004. Ye, Y., Völker, C., Wolf-Gladrow, D.A., 2009. A model of Fe speciation and biogeochemistry at the Tropical Eastern North Atlantic Time-Series Observatory site. Biogeosciences 6, 2041–2061.

Please cite this article as: Hassler, C.S., et al., Iron associated with exopolymeric substances is highly bioavailable to oceanic phytoplankton, Mar. Chem. (2014), http://dx.doi.org/10.1016/j.marchem.2014.10.002