UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The interaction between cadmium and iron in marine phytoplankton Lane, Erin Susan 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0150a.pdf [ 5.06MB ]
Metadata
JSON: 831-1.0053307.json
JSON-LD: 831-1.0053307-ld.json
RDF/XML (Pretty): 831-1.0053307-rdf.xml
RDF/JSON: 831-1.0053307-rdf.json
Turtle: 831-1.0053307-turtle.txt
N-Triples: 831-1.0053307-rdf-ntriples.txt
Original Record: 831-1.0053307-source.json
Full Text
831-1.0053307-fulltext.txt
Citation
831-1.0053307.ris

Full Text

THE INTERACTION BETWEEN CADMIUM AND IRON IN MARINE PHYTOPLANKTON by ERIN SUSAN LANE B.Sc, Memorial University of Newfoundland, 2003 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Oceanography) THE UNIVERSITY OF BRITISH COLUMBIA April 2007 © Erin Susan Lane, 2007 Abstract This study examined the intracellular Cd content (Cd:C) of seven species of cultured phytoplankton under varying degrees of Fe-limitation. A significant increase in Cd:C ratios was observed for all species grown under Fe-limitation, with an average 2-fold increase in Cd quotas. The mechanism behind this increase in Cd quotas was further investigated in a model diatom, Thalassiosira oceanica. A significant interaction was found between Cd and inorganic Fe(IT). Inorganic Fe(II) uptake rates by Fe-limited T. oceanica were reduced by 50% when Cd was present, while Cd uptake was inhibited 36% in the presence of Fe(II). Inorganic Fe(II) uptake rates were also 15 times faster in Fe-limited cultures than in Fe-replete cultures. Cadmium and Fe(II) appear to enter the cell through common non-specific divalent metal transporters that are up-regulated under Fe-limitation, thereby leading to enhanced Cd accumulation. In addition to the effect of Fe-limitation on Cd:C ratios of cultured phytoplankton, a species/phyla effect was observed. A greater than 65-fold difference in Cd:C ratios was also observed between species grown under identical Fe concentrations. Oceanic diatoms had the highest Cd quotas and naked pryrnesiophytes the lowest. The intracellular Cd requirements of cultured phytoplankton were compared to the Cd content of natural phytoplankton assemblages from Fe-replete and Fe-limited waters in the subarctic Pacific, as well as to a global dataset of particulate Cd:P043" ratios in surface waters. The phytoplankton from the Fe-limited station had the highest Cd:C ratios, which was attributed to a combination of species composition and Fe limitation. The same trend was found in our global data with HNLC stations having higher particulate Cd: P043" ratios than Fe-replete stations. The combined laboratory, field, and global dataset results suggest that deviations in u surface water CcLPCV" ratios can be explained by both changes in species composition well as Fe-limitation. We propose that the "kink" in the global dissolved CdiPCV" relationship is a result of the export of Fe-limited diatoms with high intristic Cd:P043" ratios in FfNLC regions. iii Table of Contents Abstract ii Table of Contents iv List of Tables vi List of Figures vii Acknowledgements vii Dedication ix Co-Authorship Statement x Chapter 1: Introduction 1.1 Cd Literature Review 1 1.1.1 Cd as a Paleonutrient Proxy 1 1.1.2 Distribution and chemical speciation of dissolved Cd in the Ocean 2 1.1.3 Biological Role of Cd 3 1.1.4 Background on the Cd:P043" Relationship 4 1.1.5 Explanations for the Kink 5 1.1.6 Factors Controlling Cd Uptake by Phytoplankton 6 1.1.7 Fe-limitation and the Kink 7 1.2 Fe Literature Review 9 1.2.1 Distribution of Fe in the Ocean 9 1.2.2 Fe Acquisition by Marine Phytoplankton 10 1.3 Research Objectives and Hypotheses 11 1.4 Figures 14 1.5 References 18 Chapter 2: The Effect of Phytoplankton Species Composition and Fe-limitation on the Cd:P043" Relationship. 2.1 Abstract 24 2.2 Introduction 25 2.3 Materials and Methods 29 2.3.1 Culturing 29 2.3.2 Growth Measurements 31 2.3.3 Fe and Cd cell quotas 32 2.3.4 Field Measurements 33 2.3.5 Global Data Set 35 2.4 Results 36 2.4.1 Growth rates 36 2.4.2 Intracellular Fe:C ratios and Fe-use efficiencies 37 2.4.3 Intracellular Cd:C ratios and steady state Cd uptake rates 38 2.4.4 Intracellular Cd:C ratios of a natural phytoplankton assemblages 40 2.4.5 Particulate Cd:P04 3" ratios from global dataset 41 2.5 Discussion 41 iv 2.5.1 Evidence for Fe-limitation 42 2.5.2 The effect of Cd on the growth rate and Fe:C ratios 43 2.5.3 Range of Cd:C ratios 44 2.5.4 Fe-limitation and Cd:C ratios 45 2.5.5 Steady-state Cd uptake 46 2.5.6 Intracellular Cd:C ratios of natural phytoplankton assemblages 47 2.5.7 Importance of the Southern Ocean 51 2.6 Oceanographic Implications 52 2.7 Figures 54 2.8 Tables 56 2.9 References 64 Chapter 3: The interaction between Inorganic Fe(II), Fe(III) and Cd uptake in the marine diatom Thalassiosira oceanica. 3.1 Abstract 71 3.2 Introduction 72 3.3 Methods 74 3.3.1 Culturing 74 3.3.2 Short-term inorganic Fe and Cd uptake rates 75 3.3.3 Effects of TTM additions on Fe(II) uptake 78 3.4 Results 78 3.4.1 Effect of Fe and Cd on growth rate 78 3.4.2 Inorganic Fe(II) vs. Fe(III) uptake rates 79 3.4.3 Inorganic Fe uptake rates in the presence of Cd 80 3.4.4 Cadmium uptake rates in the presence of inorganic Fe 80 3.5 Discussion 81 3.5.1 Inorganic Fe(II) vs. Fe(III) uptake rates 82 3.5.2 Interaction between Fe(III) and Cd 84 3.5.3 Effects of Cd on Fe(II) uptake rates 85 3.5.4 Effects of Fe(II) on Cd uptake rates 85 3.6 Oceanographic Implications 87 3.7 Figures 90 3.8 Tables 93 3.9 References 95 Chapter 4: Conclusion 4.1 Importance of Fe(II) 100 4.2 Modeling the global Cd:P043" Relationship 102 4.3 Cd as a paleonutrient proxy 104 4.4 References 1 A £ v List of Tables Table 2.1. Species used in this study along with strain designations, size range (diameter), source locations, and provenance 56 Table 2.2. Mean absolute growth rates, relative growth rates (pmax shown in bold), and Fe:C ratios for species used in this study 57 Table 2.3. Cd:C ratios, calculated Cd :P04 " ratios and steady state Cd uptake rates for species used in this study 58 Table 2.4. Regression slopes of Cd:P04 3" (nmol pmol"1) based on depth-dependent variations in dissolved Cd and PO43" in oceanic nutriclines 59 Table 3.1. Average relative growth rates (ju//imax) of T. oceanica under various Fe treatments in the absence or presence of 35nmol L"1 Cd (pCd 12) 93 Table 3.2. Effects of TTM on inorganic Fe(II) uptake rates by Fe-limited (pFe 21.5) T. oceanica at pH 6.6 93 Table 3.3. Effects of Cd on inorganic Fe(IIJ) and Fe(II) uptake rates by Fe-limited (pFe 21.5) T. oceanica 94 Table 3.4. Effects of Fe on Cd uptake rates by Fe-limited (pFe 21.5) T. oceanica at pH 6.6 94 vi List of Figures Figure 1.1. Vertical profiles of dissolved Cd and PO43" 14 Figure 1.2. Depth profiles of Cd:PcV" from the Atlantic, Indian, and Pacific oceans... 14 Figure 1.3. Compilation of dissolved Cd versus PO4" data 15 Figure 1.4. A compilation of global surface seawater Cd:PC>43" ratios plotted against latitude 16 Figure 1.5. Compilation of dissolved Cd versus PO43" data illustrating FfNLC stations and Fe-replete stations 16 Figure 1.6. A simple schematic diagram describing Fe-limitation and the kink proposed by Cullen(2006) 17 Figure 2.1. Particulate Cd:C (pimol mol"1) ratios of natural phytoplankton assemblages at P4andP26 54 Figure 2.2. A compilation of calculated surface particulate CdiPOv5" plotted against latitude 55 Figure 3.1. Inorganic Fe uptake rates by T. oceanica as a function of Fe-limited relative growth rate (100 x /*//Jma x) 90 Figure 3.2. Time course accumulation of particulate Fe by Fe-limited (pFe 21.5) T. oceanica at pH 6.6 91 Figure 3.3. Time course accumulation of particulate Cd by Fe-limited (pFe 21.5) T. oceanica at pH 6.6 92 vii Acknowledgements Foremost, I would like to thank Dr. Maite Maldonado for her guidance, inspiration, patience and the time she invested in editing this thesis. Maite is a passionate and dedicated woman and I look-up to her both personally and professionally. I am extremely honoured to be Maite's first graduate student. I would especially like to acknowledge Dr. Jay Cullen for inspiring me to investigate cadmium biogeochemistry. I would also like to give special thanks to Dr. Philippe Tortell for all of his help and feedback throughout this project. I am in gratitude to my fellow lab members, who played a pivotal role in aiding me in the completion of this thesis. I am greatly indebted to David Semeniuk for constantly challenging me with new ideas, and helping me to truly understand Fe physiology. Julie Granger's passion for knowledge in the field of biogeochemistry has been extremely inspirational. I am also extremely grateful for Kira Jang who provided me with both emotional and technical support during late hours in the lab. I would like to recognize some of the wonderful friends I have met over the past three years, who have made my experience at UBC extremely memorable. Special thanks to Desmond Chung, Martin Swift, and Muhannad Al-Darbi for their constant support and care during my stay at St. John's College. I would especially like to thank Melissa Rohde and Belinda Schubert and Laura Estrada for their unwavering friendship. Finally, I would like to thank my family and friends in Newfoundland, their belief in me inspires the successes I have achieved in my life. Dedication To (Dorothy Murphy and (Beatrice Lane-No6le, my two inspiring grandmothers... and all the other women who have touched my Rfe, your inner strength and wisdom allowed me to 6eCieve in myself and my ambitions. ix Co-Authorship Statement A version of Chapter 2 and 3 will be submitted for publication to Limnology and Oceanography. Chapter 2 in co-authored with D. Semeniuk, J. T. Cullen, and M. T. Maldonado. The inspiration to investigate the interaction between Cd biogeochemistry and Fe came from my initial discussions with Dr. J. T. Cullen and Dr. M. T. Maldonado. D. Semeniuk and M. T. Maldonado participated in the September 2006 cruise along line PO4 3" in the subarctic Pacific on board the CCGS John P. Tully. I designed the experiments that were conducted during this cruise; however they were carried out by D. Semeniuk and M. T. Maldonado. J. T. Cullen measured surface water dissolved Cd concentrations at station P4 and P26 and provided me with the Cd concentration used in culture growth media. Chapter 3 is co-authored with K. Jang, and M. T. Maldonado. K, Jang assisted me during in set-up of uptake experiments. M. T. Maldonado and D. Semeniuk critically reviewed both manuscripts in this thesis. x Chapter 1: Introduction 1.1 Cd Literature Review 1.1.1 Cd as a Paleonutrient Proxy The oceanic distribution of phosphate (PO4 3") is controlled by its removal from surface waters by phytoplankton and subsequent remineralization of sinking organic matter at depth, combined with its transport by thermohaline circulation (Redfield et al. 1963). Knowledge of past PO43" distributions therefore serves as a tracer of both paleoproductivity and paleocirculation. Changes in ocean productivity and circulation in the past had major impacts on the drawdown of atmospheric C O 2 via the biological and physical pump. Paleoreconstruction of past PO4 3 " concentrations is an important tool used to determine the oceans role in driving atmospheric C O 2 concentrations over glacial and interglacial times. A solid mechanistic understanding of how the ocean regulated atmospheric C O 2 concentrations in the past is crucial in models forecasting the possible climatic effects of anthropogenic C O 2 emissions. Phosphate is not directly preserved in the sedimentary record therefore other indirect paleonutrient proxies must be used to reconstruct its oceanic distribution in the past. Cadmium (Cd) is a nutrient-type metal and has a seawater distribution that is very similar to PO43". Dissolved Cd profiles show depleted surface concentrations that increase with depth to a maximum in the thermocline and then remain relatively constant with depth (Figure 1.1). This suggests that Cd, like PO43", is mostly controlled by phytoplankton uptake in surface waters and by sinking and remineralization of organic matter at depth (Boyle et al. 1976; Bruland et al. 1978; de Baar et al. 1994). Cadmium is also incorporated into the calcium carbonate tests of certain foraminifera in proportion to 1 the Cd concentrations of the seawater in which the foraminifera are growing (Boyle 1992). Cadmium preserved in benthic fossil foraminiferal tests has been applied, along with the strong dissolved CdiPCv3" correlation, to reconstruct past PO4 3" distributions in the deep ocean. More recently, this approach has been expanded to use Cd:Ca ratios in planktonic foraminifera to reconstruct surface water PO4 3" concentrations in the Southern Ocean during the Last Glacial Maximum (LGM) (Elderfield and Rickaby 2000). This Cd paleonutnent proxy suggests that during the LGM, PO4 " utilization in the subantarctic was similar to that of today, but was much less in the polar Southern Ocean. It has been suggested that lower glacial atmospheric CO2 was a result of restricted communication between the ocean and the atmosphere due to the expansion of sea ice in the Polar Southern Ocean, rather than an increase in the efficiency of the biological pump (Elderfield and Rickaby 2000). The use of Cd as a paleonutrient proxy however, hinges on a mechanistic understanding of the relationship between Cd and PO4 3" in seawater. 1.1.2 Distribution and chemical speciation of dissolved Cd in the Ocean Up to 70% of the total dissolved Cd (Cdj) in surface waters is organically complexed by a Cd-speciftc ligand (Bruland 1992). Inorganic Cd (Cd') is defined as the sum of inorganic species and Cd' concentrations, and free ionic Cd 2 + activities are related by the inorganic side reaction coefficient alpha ([Cd'] = alpha [Cd ]). Alpha is 34.7 in surface seawater at 25°C (Byrne et al. 1988), resulting in < 5% of Cd' being present as Cd . The remaining proportion of Cd' is present predominantly as Cd-chloride complex. 9+ 1 Surface free ionic Cd activities have been measured as low as 20 fmol L" (Bruland 1992). Both free Cd ions and Cd-chloride complexes are available for phytoplankton 2 uptake (Sunda and Huntsman 1998). Total dissolved Cd concentrations are very low in oligotrophic surface waters (1.0-5.0 pmol I/1) and high in upwelling waters (150 pmol L'1) (Bruland 1992). Cadmium also displays inter-ocean fractionation, due to superposition of the thermohaline deep water circulation on the vertical remineralization cycle (Broecker and Peng 1982). Deep water formed in the North Atlantic travels south, around Antarctica, into the Indian and Pacific Oceans accumulating nutrients and trace metals from sinking particles on its journey of-1000 years. The old waters of the deep Indian and Pacific Oceans therefore have 2-5 times more nutrients and trace metals than the young Atlantic Ocean (Figure 1.2) (Bruland and Franks 1983). 1.1.3 Biological Role of Cd Although dissolved Cd displays a nutrient-type profile in the open ocean, it was not initially thought to have a biological role. It has since been postulated that the physiological use of Cd in phytoplankton is as a substitute for zinc (Zn) in the metalloenzyme carbonic anhydrase (Price and Morel 1990; Morel et al. 1994; Lee and Morel 1995). Alternatively it can be used directly in a Cd-specific form of the enzyme (Cullen et al. 1999; Lane and Morel 2000 a, b). Carbonic anhydrase catalyzes the inter-conversion of bicarbonate (HC03~) and CO2 (Badger and Price 1994). Inorganic carbon in the ocean primarily exists as HCO3". The enzyme RubisCO used in photosynthesis to convert inorganic carbon to organic carbon, needs CO2 as its substrate. Cadmium has been found to relieve Zn-limited primary production and CO2 fixation (Lee and Morel 1995; Lane and Morel 2000). The role of Cd in carbonic anhydrase was further 3 supported by field studies off central California, which showed that the Cd content of natural phytoplankton populations increased as surface-water pC0 2 decreased (Cullen et al. 1999). Incubation experiments with natural phytoplankton from these waters confirmed that Cd uptake and the Cd content of cells is inversely related to seawater pC0 2 and Zn concentration (Cullen et al.1999; Cullen and Sherrell 2005). 1.1.4 Background on the dissolved Cd:PC«43" Relationship The first reliable vertical trace metal profiles were obtained from the Pacific Ocean (Boyle et al. 1976; Martin et al. 1976; Bruland et al. 1978; Bruland 1980; Knauer and Martin 1981). The strikingly linear correlations of Cd and PO4 3" (average slope of Cd:PO 4 3"~0.30±0.1 x 10~3, zero intercept) led early reasearchers (Boyle et al. 1976; Bruland et al. 1978) to suggest that the Pacific relationships may also apply to the global ocean. This was an attractive hypothesis, as it would extend the Redfield concept (Redfield et al. 1963) to trace metals (Morel and Hudson 1985). However, after data became available for the Atlantic Ocean, this prediction turned out to be incorrect, as the Cd:P04 " slope for the North Atlantic was 15% lower than the North Pacific (Bruland and Franks 1983). Although, the Cd:P043" relationship of the whole water column at individual stations throughout the ocean exhibited a linear relationship with PO43", combining all data from the global ocean (below the mixed layer) onto one plot showed a kink at a P043" concentration of ~1.3pmol L"1 P043" (Boyle 1988; Cullen 2006) (Figure 1.3). This led Boyle (1988) to suggest two relationships for the world ocean, one for PO4 3" < 1.3pM (all upper ocean waters and most North Atlantic deep water) and one for PO4 3" > 1.3pM (largely North Pacific deep waters). Accepting the regional differences in 4 -5 relative Cd:P04 " ratios in the Atlantic and Pacific ocean basins (Figure 1.2) the ratio of dissolved Cd to PO4 3" is quite constant in deep waters (>1000m) (de Baar et al. 1994). In contrast significant variability exists in surface-water ratios (Figure 1.2, Figure 1.4) (de Baar et al. 1994; Rutgers van der Loeff et al. 1997), especially in areas where phytoplankton growth is known to be Fe-limited (Martin et al. 1989; Martin et al. 1990). 1.1.5 Explanations for the Kink The mechanism behind the kink in the Cd:PC»4 " relationship and heterogeneous nature of Cd:P043" ratios in surface waters has fuelled much debate. Boyle (1988) suggested that the kink in the Cd:PC>43" relationship was due to Cd in sinking organic matter being regenerated at greater depths than PO4 ". This explanation seems unlikely due to the virtual coincidence of subsurface Cd and PO4 3" concentration maxima and the fact that vertical Cd and PO4 3" profiles are almost never decoupled. The kink has also been attributed to mixing of water masses with different preformed Cd:P043" ratios (Westerlund and Ohman 1991; Doroshevich and Boyle 1992; Frew and Hunter 1992). If surface waters sink to form deep water masses, the surface water Cd and PO4 3" concentrations and the Cd:PC>43" ratio are retained. It was suggested that the input of subantarctic waters, which are depleted in Cd relative to PO4 3" (Antarctic Intermediate Water (AAIW)) into the intermediate waters causes an inflection due to mixing in a three-end member system, further consisting of North Atlantic Deep Water (NADW) and Antarctic Bottom Water (AABW) (Doroshevich and Boyle 1992; Frew and Hunter 1992). Although an elegant hypothesis, the subantarctic data used by Doroshevich and Boyle (1992) and Frew and Hunter (1992) does not represent the AAIW 5 source area (Saager 1994). In response to this, Frew (1995) alternatively proposed that the formation of AABW from high Cd ' .PCV" surface waters close to the Antarctic continent may cause the kink. This explanation helps to account for elevated ratios in the deep ocean but it can not explain the low ratios and variability in dissolved C d i P C V " in oceanic surface waters. The role of phytoplankton in explaining the kink has also been investigated. Many observations have led to the suggestion that Cd may be preferentially removed from surface waters by phytoplankton relative to PO4 3" (Ldscher et al. 1998; Saager and de Baar 1993; Elderfield and Rickaby 2000). To account for the variability in surface water dissolved Cd:PC>43" in the global ocean, Saager and de Baar (1993) modeled global Cd:PC>43" distributions assuming regionally specific phytoplankton Cd:PC»43" composition. In contrast, Elderfield and Rickaby (2000) treated Cd:PC>43" ratios in a manner analogous to nutrient isotope systems (e.g. C, N, silicon) and a applied a Rayleigh fractionation model to surface water Cd ' .PCV" distributions. This model assumes a constant fractionation factor between seawater (SW) and particulate organic matter (POM) for all regions of the ocean (Ofcd:P043 = Cd:PpoM/Cd:Psw —2.5). Both approaches proved fairly successful at describing the distribution of Cd:P043" in the modern ocean. However, these models provided no mechanistic understanding of the processes controlling the relative rates of Cd and PO4 " uptake by phytoplankton or why regional differences in particulate and dissolved Cd:P043" might exist. 1.1.6 Factors Controlling Cd Uptake by Phytoplankton Differences in the removal of dissolved Cd relative to PO4 3" removal by 6 phytoplankton in surface waters could help to explain the vertical and horizontal variation of seawater Cd:P043" ratios. Laboratory studies have shown that the Cd content of several marine phytoplankton species are controlled by aqueous Cd ion concentrations but are also inversely related to manganese (Mn) and Zn ion concentrations, likely because of competitive inhibition at cellular uptake sites (Lee et al. 1995; Sunda and Huntsman 1996, 1998, 2000). As well, aqueous CO2 concentrations have been shown to control phytoplankton Cd quotas in culture (Lane and Morel 2000). The few field studies with natural algal assemblages tend to support the controlling factors identified in culture (Cullen et al. 1999; Cullen and Sherrell 2005). There is also evidence that cellular Cd:P043" ratios increase within Fe-limited phytoplankton. Sunda and Huntsman (2000) demonstrated that Fe-limited populations of the cultured oceanic diatom Thalassiosira oceanica had significantly elevated cellular Cd content relative to Fe-replete treatments. Cullen et al. (2003) during shipboard incubation experiments, studied the effects of Fe-limitation on the Cd:PC»4 " composition of natural assemblages of Fe-limited marine •a t phytoplankton in Southern Ocean. Particulate Cd:P04 " ratios decreased from control values with increasing initial dissolved Fe concentrations by factors of -2-10 at highest Fe additions (Cullen et al. 2003). 1.1.7 Fe-limitation and the Kink The kink in the Cd:P043" relationship has also been linked to High Nutrient Low Chlorophyll (HNLC) regions. HNLC regions are areas of the ocean such as the subarctic Pacific, the equatorial Pacific and the Southern Ocean, where phytoplankton growth is limited by Fe. The kink in the Cd:P043" relationship occurs at PO4 3" concentrations that 7 are typical for surface waters in Fe-limited HNLC areas (Cullen 2006). Several studies have noted significant deviations from linearity in the global dissolved Cd versus PO4 3" relationship in high-latitude FfNLC regions of the Southern Ocean (Martin et al. 1990; Nolting et al. 1991; Frew 1995; Ldscher 1998; Fitzwater et al. 2000; Ellwood 2004) and the subarctic Pacific (Martin et al. 1989). This non-linearity is marked by a kink in the dissolved Cd versus PO4 3" relationship at PO4 3" concentrations equal to -1.3 /miol L"1, low dissolved Cd: PO4 3" ratios in surface waters, and higher than average Cd: PO4 3" slopes and significant negative y-intercepts in nutricline waters (Figure 1.5) (de Baar et al. 1994). Cullen (2006) made a distinction between Fe-replete and HNLC stations and plotted Cd versus PO4 3" concentrations from various stations throughout the global ocean onto one plot (Figure 1.5). This plot illustrates that the kink in the Cd: PO4 3" relationship is only apparent in the Fe-limited stations. The dissolved Cd versus PO4 3" can therefore be described by a linear relationship for Fe-replete areas and by two different slopes with a kink at PO4 3" concentrations equal to ~1.3 jiimol L"1 for FfNLC regions (Figure 1.5). Sunda and Huntsman (2000) modeled oceanic diatom Cd:P04 " ratios based on variations in Cd and Zn concentrations and their results showed good agreement with slopes of Cd versus PO4 3" in different oceanic regimes. Their model suggested that the high Cd versus PO4 3" slopes found in the nutricline of Fe-limited waters of the Southern Ocean and subarctic Pacific resulted from high levels of Zn depletion in these waters. They therefore attributed the kink to the export of high Cd:P04 " phytoplankton growing in waters with depleted Zn concentrations, resulting from the high rates of Zn uptake by Fe-limited phytoplankton. Cullen (2006) suggests that the kink may be due to the preferential accumulation 8 o f C d by Fe- l imited phytoplankton in high latitude H N L C regions. Hyper-accumulat ion o f C d b y h igh latitude Fe- l imi ted phytoplankton has been attributed to a reduced growth rate at constant C d uptake rate (Sunda and Huntsman 2000; C u l l e n 2006). Enhanced intracellular C d quotas b y phytoplankton leads to the reduction o f dissolved C d : P 0 4 3 " ratio o f surface waters. T h e export and subsequent remineralization o f this high Cd:P04 3 " particulate matter elevates the dissolved Cd:P04 3 " ratio in the subsurface nurticline, accentuating the nonlinearity o f vertical profiles from Fe- l imited region (Figure 1.6) (Cul len 2006). In order to establish i f accumulation o f C d b y Fe- l imited phytoplankton is the root cause for the kink in the C d : P 0 4 3 " relationship and variability o f dissolved C d : P 0 4 3 " ratios in surface waters, a thorough understanding o f the interaction between F e and C d seawater chemistry and uptake b y phytoplankton is needed. 1.2 Fe Literature Review 1.2.1 Distribution of Fe in the Ocean In recent years improved trace metal clean and analytical techniques have provided accurate measurements o f oceanic F e concentration and speciation, thereby changing our understanding o f phytoplankton F e transport mechanisms. Iron i n open ocean surface waters is present at total concentration o f -0.07 n M (Johnson et al. 1997). T h e role o f iron in l imit ing pr imary production in the oceans is n o w we l l established (Coale et al. 1996; B o y d et al. 2000). Iron(HI) has a very low solubil ity in seawater and is rapidly hydrolysed and nucleated to particulate Fe(IIT) hydroxides ( L i u and M i l l e r o 2002). In oxic seawater around p H 8, Fe(UI) is the dominant redox species and is present 9 predominantly in the particulate iron oxyhydroxide . The overall dissolved speciation of Fe(III) in seawater has been shown to be dominated by complexation with organic ligands (Rue and Bruland 1995). These organic ligands are predominately siderophore-like molecules that are believed to be produced by bacteria or phytoplankton to increase the solubility of Fe in seawater (Kuma et al. 1996; Millero 1998). Iron(H) is more soluble than Fe(III) at alkaline pH, however in natural seawater Fe(II) is rapidly oxidized by O 2 and H 2 O 2 (Millero et al. 1987; Millero and Sotolongo 1989; Millero 1989; King et al. 1991; King et al. 1995; King 1998). Photoreduction of Fe(IU) to Fe(U), is a possible mechanism by which colloidal Fe is made more bioavailable to phytoplankton (Croot et al. 2001). Iron(II) has been viewed as a short lived intermediate in iron cycling, existing at low concentrations (pmol L"1 or less) in warm tropical waters (Johnson et al. 1994). Cold temperatures however, lead to slower kinetics for the oxidation of Fe(II) (Millero et al. 1987; Millero and Sotolongo 1989; King et al. 1995; King 1998; King and Farlow 2000) and studies in the eastern and western North Atlantic have measured Fe(II) concentrations as high as 50% of the total dissolved Fe in seawater (Zhuang et al. 1995; Boye et al. 2003). During a recent Fe fertilization experiment in the Southern Ocean (SOIREE), Fe(II) existed for several hours at elevated concentrations (~1 nmol L"1) (Croot et al. 2001). There is now growing evidence that Fe(II) can also be organically complexed, which would further retard the oxidation step (Croot et al. 2001). 1.2.2 Fe Acquisition by Marine Phytoplankton It was once thought that Fe uptake rates were dependent on the concentration of 10 inorganic Fe species (Fe') (Hudson and Morel 1990; 1993; Sunda and Huntsman 1995). It is now known, that the dissolved Fe concentrations are low in surface waters of the ocean and excess natural chelators buffer a remarkably low Fe' concentration that is unable to support phytoplankton growth (Gledhill and van Den Berg 1994; Rue and Bruland 1995; Wu and Luther 1995). In response to such low Fe' concentrations, phytoplankton have evolved high affinity uptake systems allowing them to acquire the Fe that is bound to these strong organic ligands (Shaked et al. 2005; Maldonado et al. 2006; Kustka et al. 2007). This uptake process first requires the dissociation of Fe from these ligands. This high affinity Fe uptake system is thought to be similar to those identified in the yeast Saccharomyces cerevisiae (Askwith et al. 1994) and the green alga Chlamydomonas reinhardtii (La Fontaine et al. 2002; Herbik et al. 2002a, b). This system is well characterized in Saccharomyces cerevisiae and involves the reduction of Fe(III) to Fe(II) by plasma membrane reductases (Frelp and Fre2p), followed by a coupled oxidation of Fe(II) by a multi-copper oxidase (Fet3p), and uptake of Fe(III) by an Fe permease (Ftrlp) (reviewed by Eide 1998; Van Ho et al. 2002). Genes homologous to those that encode for these proteins have been identified in the recently sequenced geome of the diatom Thalassiosira pseudonana (Ambrust et al. 2004). The transcript abundance of reductase and permease genes in T. pseudonana was highly regulated by the cellular Fe status (Kustka et al. 2007). A putative gene for a Fe(II) transport protein was also identified in T. pseudonana (7)?Nramp), which was dramatically up-regulated under Fe-limitation (Kustka et al. 2007). The Fe(II) transporter identified in T. pseudonana is a member of the NRAMP's family. Several NRAMP transporters have also been identified in yeast 11 and plants (Eide 1998, Van Ho 2002). The accumulation of Cd by Fe-deficient plants has been linked to the up-regulation of NRAMP's transporters that have a broad substrate range for divalent metals (Curie et al. 2000, Thomine et al. 2000). 1.3 Research Objectives and Hypotheses The main objective of this thesis was to understand why differences in both dissolved and particulate Cd:PC»43~ ratios exist between Fe-limited and Fe-replete oceanic waters. Accurate knowledge of the modern dissolved Cd:PC>43" relationship in surface waters and our ability to model how dissolved Cd: P0 4 " ratios change temporally and spatially in the ocean is crucial for reconstructions of past surface water PO4 3" concentrations using Cd preserved in fossil planktonic foraminifera. While shipboard incubation experiments show decreasing particulate Cd:P043" ratios upon Fe supplementation, it is unclear whether this is due to a shift in species composition or an underlying physiological mechanism (Cullen et al. 2003). The one laboratory experiment where intracellular Cd content was investigated in relation to Fe status only used a single species (Thalassiosira oceanica, Sunda and Huntsman 2000). In order to systematically decipher the effect of Fe-limitation on particulate Cd:PC>43" ratios, I measured Cd:C ratios of several species of cultured phytoplankton under varying degrees of Fe limitation and of natural phytoplankton assemblages from an Fe-limited and an Fe-sufficient station in the subarctic Pacific. These measured Cd quotas were then compared with estimated particulate Cd:P043" ratios determined from dissolved Cd and P043" concentrations in the nutriclines of a global data set. I also investigated the physiological mechanism that mediated the accumulation of Cd in Fe-12 limited phytoplankton. Given the empirical link between Fe-limitation and high intracellular Cd, I hypothesized that Cd and Fe enter the cell via a common non-specific divalent cation transporter that is up-regulated under Fe-limitation. The up-regulation of this transporter, may lead to enhanced Cd accumulation by Fe-limited phytoplankton. I hypothesize that the export of Fe-limited phytoplankton with high intracellular Cd content was the root cause of the kink in the Cd:PC»43" relationship. Specifically, I: 1. Compared Fe-replete and Fe-limited growth rates of several cultured oceanic phytoplankton species from two phyla in the presence and absence of Cd, at a free Cd activity characteristic of Fe-limited surface oceanic waters. 2. Determined the cellular Cd:C and Fe:C ratios of these species under Fe-limited and Fe-replete conditions. 3. Compared Cd:C ratios of natural phytoplankton assemblages at a Fe-replete and Fe-limited station in the subarctic Pacific Ocean, an Fe-limited region. 4. Compiled a global dataset of surface particulate Cd:P043" ratios estimated from dissolved Cd:P043" ratios in oceanic nutriclines from all the available literature where Cd and PO4 3" vertical profiles were measured simultaneously. 5. Investigated if Thalassiosira oceanica has a non-specific Fe(II) transporter that is up-regulated under Fe-limitation. In particular, I elucidated whether a competitive uptake mechanism between Fe(II) and Cd could explain the enhanced accumulation of Cd in Fe-limited Thalassiosira oceanica. 13 1.4 Figures |Cdi irunoi kg"'} 0.3 OA PO*"'] (iimol kg" :j i .000 - N*. -2.000 y 0 0 0 % a o o 0 - o o o 0 - o o o i 1 1 . L i„„ 1 1 S Figure 1.1. Vertical profiles of dissolved Cd and PO4 3" (from Puyseger trench) (Figure taken from Frew and Hunter 1992). Cd/P (nmol/pmol) 0 0.1 0 . 2 0 . 3 0.4 Figure 1.2. Depth profiles of Cd:P043" from the Atlantic (closed circles), Indian (closed triangles), and Pacific oceans (open circles) (Figure taken from Elderfield and Rickaby 2000). 14 1.4 • Yeats and Wester I und (1991) v Bruland (1980) • Bruland (1983) O Danielsson and Wcsterlund (1983) A Frew and Hunter (1992) O Nolting and De Baar (1994) • Frew and Hunter (1995) v Kremling and Pohl {1989) • Bruland etal. (1978) O Knauer and Martin (1981) A Eli wood (2004) O o Fitzwater et al. (2000) O • Abe (2001) v Abe (2002a) ^ • Abe (20026) <s5*c5^ O Chen et al. ( 20051^^ o<£<T#<> 0.0 H > , , , : .- : , _ _ , ( 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 P0 4 3 (umot L*1) Figure 1.3. Compilation of dissolved Cd versus P043~ data, lines represent the two distinct relationships for [P043"]< 1.3pmol L"1, primarily Atlantic waters, and for [P043"]> 1.3pmol L"1, primarily Indian-Southern-Pacific Ocean waters proposed by Boyle (1988) (Figure taken from Cullen 2006). 15 0.5 0.4 o f_ 0.3 o E c QL O 0.2 0.1 0.0 Sur face Wate r Dissolved C d / P 4' 8 \ * • -60 -40 -20 0 20 Latitude (deg) 40 60 Figure 1.4. A compilation of global surface seawater CcbPCV" (nmol/pmol) ratios plotted against latitude (Figure taken from Elderfield and Rickaby 2000). 1,4 12 0.6 -\ o v o o o 0 V • • • A 0.4 H 0.2 0.0 Bruland (1983) Bruland etal, (1978) Chen ct ai. (2005 ) Abe (2002a) This Study Abe (2002*) Krcmling and Pohl (1989) Yeats and Westcrlund (1991) Ollwood (2004) Fitzwater et ai. (2000) g Martin cl al. (1989) Frew and H unier (1995) $ c? Thissiudy j&&0<?' 0.0 0.5 1.0 1.5 2.0 PO> (pmol L"'y 2.5 3.0 3.5 Figure 1.5. Compilation of dissolved Cd versus PO4 3" data, solid symbols represent stations from high-latitude HNLC areas for which the growth of phytoplankton is thought to be limited by Fe availability, and open symbols are from stations in which surface dissolved PO4 3" concentrations have been drawn down to > 0.3 mmol L"1 by the growth of phytoplankton (Figure taken from Cullen 2006). 16 Iron limited HNLC area * s phytoplankton surface water 'Cd:^ C d : P ° 4 nutricline Cd:P04 sinking and remineralization , Iron sufficient area phytoplankton surface water Cd:PQ 4 nutricline Cd:P0 4 sinking and remineralization P0 4 3 -(p:molL- 1 ) P0 4 3-(p.mol L" 1) Figure 1.6. A simple schematic diagram summarizing the influence of Fe availability on the Cd:P043" composition of marine phytoplankton and the resulting effects on the dissolved Cd:P04" ratio of surface waters and the nutricline proposed by Cullen (2006) (Figure taken from Cullen 2006). 17 1.5 References Armbrust, E. V., and others. 2004. The genome of the diatom Thalassiosira pseudonana: Ecology, evolution, and metabolism. Science 306: 79-86. Askwith, C , and others. 1994. The FET3 gene of S. cerevisiae encodes a multicopper oxidase required for ferrous iron uptake. Cell 76: 403-410. Badger, M. R., and G. D. Price. 1994. The role of carbonic anhydrase in photosynthesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 45: 369-392. Boyd, P. W., and others. 2000. A mesoscale phytoplankton bloom in the polar Southern Ocean stimulated by iron fertilization. Nature 407: 695-702. Boye, M., A. P. Aldrich, C. M. G. van den Berg, J. T. M. de Jong, M. Veldhuis, and H. J. W. de Baar. 2003. Horizontal gradient of the chemical speciation of iron in surface waters of the northeast Atlantic Ocean. Mar. Chem. 80: 129-143. Boyle, E. A. 1988. Cadmium: Chemical tracer of deepwater paleoceanography. Paleoceanography 3: 471-489. Boyle, E. A. 1992. Oceanic chemical distributions during the stage 2 glacial maximum: Cadmium and 5 1 3C evidence compared. Earth Planet. Sci. Lett. 20: 245-287. Boyle, E. A., F. Sclater, and J. M. Edmond. 1976. On the marine geochemistry of cadmium. Nature 263: 42-44. Broecker, W. S., and T. H. Peng. 1982. Tracers in the Sea. LDGO, New York, 690 pp. Bruland, K. W. 1980. Oceanographic distributions of cadmium, zinc, nickel, and copper in the North Pacific. Earth Planet. Sci. Lett. 47: 176-198. Bruland, K. W. 1992. Complexation of cadmium by natural organic ligands in the central North Pacific. Oceanography 37: 1008-1017. Bruland, K. W., and R. P. Franks. 1983. Mn, Ni, Cu, Zn and Cd in the western north Atlantic. In C.S. Wong (ed.), Trace Metals in Seawater. NATO Conference Series IV; Marine Sciences: 394-414. Bruland, K. W., G. A. Knauer, and J. H. Martin. 1978. Cadmium in northeast Pacific waters. Limnol. Oceanogr 23: 618-625. Coale, K. H. and others. 1996. A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean. Nature 383: 495-501. 18 Croot, P. L., A. R. Bowie, R. D. Frew, M. Maldonado, J. A. Hall, K. A. Safi, J. La Roche, P. W. Boyd, C. S. Law. 2001. Retention of dissolved iron and Fe11 in an iron induced Southern Ocean phytoplankton bloom. Geophys. Res. Lett. 28: 3425-3428. Cullen, J. T. 2006. On the nonlinear relationship between dissolved cadmium and phosphate in the modern global ocean: Could chronic iron limitation of phytoplankton growth cause the kink? Limnol. Oceanogr. 51: 1369-1380. Cullen, J. T., Z. Chase, K. H. Coale, S. Fitzwater, and R. M. Sherrell. 2003. Effect of iron limitation on the cadmium to phosphorus ratios of natural phytoplankton assemblages from the Southern Ocean. Limnol. Oceanogr. 48: 1079-1087. Cullen, J. T., T. W. Lane, F. M. M. Morel, and R. M. Sherrell. 1999. Modulation of cadmium uptake in phytoplankton by seawater C0 2 concentration. Nature 402: 165-167. Cullen, J. T. and R. M. Sherrell. 2005. Effects of dissolved carbon dioxide, zinc, and manganese on the cadmium to phosphorus ratio in natural phytoplankton assemblages. Limnol. Oceanogr. 50: 1193-1204. Curie C , J. M. Alonso, M. Le Jean, J. R. Ecker, and J. F. Briat. 2000. Involvement of NRAMP1 from Arabidopsis thaliana in iron transport. Biochem. J. 347: 749-755. de Baar, H. J. W., P. M. Saager, R. F. Notling, and J. Van Der Meer. 1994. Cadmium versus phosphate in the world ocean. Mar. Chem. 46: 261-281. Doroshevich, P. and E. A. Boyle. 1992. Cadmium in surface waters of the Antarctic Ocean (Atlantic sector). EOS, Trans. Am. Geophys. Union, 44. Eide, D. J. 1998. The molecular biology of metal ion transport in Saccharomyces cerevisiae. Ann. Rev. Nutrition 18: 441-469. Elderfield, H., and R. E. M. Rickaby. 2000. Oceanic Cd/P043" ratio and nutrient utilization in the glacial Southern Ocean. Nature 405: 305-310. Ellwood, M. J. 2004. Zinc and cadmium speciation in subantarctic waters east of New Zealand. Mar. Chem. 87: 37-58. Fitzwater, S. E., K. S. Johnson, R. M. Gordon, K. H. Coale and W. O. Smith, Jr. 2000. Trace metal concentrations in the Ross Sea and their relationship with nutrients and growth. Deep-Sea Research II47: 3159-3179. Frew, R. D. 1995. Antarctic Bottom Water formation and the global cadmium to phosphorus relationship. Geophysical Research Letters 22: 2349-2352. Frew, R., and K. A. Hunter. 1992. Influence of Southern Ocean waters on the cadmium-phosphate properties of the global ocean. Nature 360: 144-146. 19 Gledhill, M., and C. M. G. Van den Berg. 1994. Determination of complexation of iron (III) with natural organic complexing ligands in seawater using cathodic stripping voltammetry. Mar. Chem. 47: 41-54. Herbik, A., C. Boiling, and T. J. Buckhout. 2002a. The involvement of a multicopper oxidase in iron uptake by the green algae Chlamydomonas reinhardtii. Plant Physiol 130: 2039-2048. Herbik, A., S. Haebel, and T. J. Buckhout. 2002b. Is a ferroxidase involved in the high-affinity iron uptake in Chlamydomonas reinhardtii? Plant Soil 241: 1-9. Hudson, R. J. M. and F. M. M. Morel. 1990. Iron transport in marine phytoplankton: Kinetics of cellular and medium coordination reactions. Limnol. Oceanogr. 35: 1002-1020. Hudson, R. J. M. and F. M. M. Morel. 1993. Trace metal transport by marine microorganisms: Implications of metal coordination kinetics. Deep-Sea Res. 40: 129-150. Johnson, K. S., K. H. Coale, V. A. Elrod, N. W. Tindale. 1994. Iron Photochemistry in seawater from the Equatorial Pacific. Mar. Chem. 46: 319-334. Johnson, K. S., R. M. Gordon, and K. H. Coale. 1997. What controls dissolved iron concentrations in the world ocean? Mar. Chem. 57: 137-161. King, D. W. 1998. Role of Carbonate Speciation on the Oxidation Rate of Fe(II) in Aquatic Systems. Environ. Sci. Technol. 32: 2997-3003. King, D. W. and R. Farlow. 2000. Role of carbonate speciation on the oxidation of Fe(II) byH 20 2 . Mar. Chem. 70: 201-209. King, D. W., J. Lin, D. R. Kester. 1991. Determination of Fe(II) in seawater at nanomolar concentrations. Analytica Chimica Acta, 247(1), 125. Anal. Chim. Acta 247: 125-132. King, D. W., H. A. Lounsbury, F. J. Millero. 1995. Rates and mechanism of Fe(II) oxidation at nanomolar total iron concentrations . Environ. Sci. Technol. 29: 818-824. Kuma, K., J. Nishioka, K. Matsunaga. 1996. Controls on iron(HI) hydroxide solubility in seawater: The influence of pH and natural organic chelators. Limnol. Oceanogr. 41: 396-407. Kustka, A. B., A. E. Allen, F. M. M. Morel. 2007. Sequence analysis and transcriptional regulation of Fe acquisition genes in two marine diatoms. J. Phycol. (In press). 20 La Fontaine, S., J. M. Quinn, S. S. Nakamoto, M. D. Page, V. Gohre, J. L. Mosely, J Kropat, and S. Merchant. 2002. Copper-dependent iron assimilation pathway in the model photosynthetic eukaryote Chlamydomonas reinhardtii. Eukaryotic Cell 1: 736-757. Lane, T. W., and F. M. M. Morel. 2000a. A biological function for cadmium in marine diatoms. Proceedings of the National Academy of Science USA 97: 4627-4631. Lane, T. W., and F. M. M. Morel. 2000b. Regulation of carbonic anhydrase by zinc, cobalt, and carbon dioxide in the marine diatom Thalassiosira weissflogii. Plant Physiol. 123: 345-352. Lee, J. G., and F. M. M. Morel. 1995. Replacement of Zn by cadmium in marine phytoplankton. Mar. Ecol. Progr. Ser. 127: 305-309. Lee, J. G., S. B. Roberts, and F. M. M. Morel. 1995. Cadmium: a nutrient for the marine diatom Thalassiosira weissflogii. Limnol. Oceanogr. 40: 1056-1063. Liu, X. and F. J . Millero. 2002. The solubility of iron in seawater. Mar. Chem. 77: 43-54. Ldscher, B. M., J. T. M. D. Jong, and H. J. W. de Baar. 1998. The distribution and preferential biological uptake of cadmium at 6°W in the Southern Ocean. Mar. Chem. 62: 259-286. Martin, J. H., R. M. Gordon, and S. Fitzwater. 1990. Iron in Antarctic waters. Nature 345: 156-158. Martin, J. H., R. M. Gordon, S. Fitzwater, and W. W. Broenkow. 1989. VERTEX: Phytoplankton/iron studies in the Gulf of Alaska. Deep-Sea Res. 36: 649-680. Maldonado, M. T., A. E. Allen, J. S. Chong, K. Lin, D. Leus, N. Karpenko, and S. L. Harris. 2006. Copper-dependent iron transport in coastal and oceanic diatoms. Limnol. Oceanogr. 51: 1729-1743. Millero, F. J. 1998. Solubility of Fe(III) in seawater. Earth Planet. Sci. Lett. 154: 323-329. Millero, F. J. and S. Sotolongo. 1989. The oxidation of Fe(II) with H202 in seawater. Geochim. Cosmochim Acta 53: 1867-1873. Millero, F. J., S. Sotolongo, M. Izaguirre. 1987. The oxidation-kinetics of Fe(II) in seawater. Geochim. Cosmochim. Acta. 51: 793-801. Morel, R. M. & R. J. Hudson, 1985. The geobiological cycle of trace elements in aquatic systems: Redfield revisited. In W. Stumm (ed.), Chemical Processes in Lakes. J. Wiley & Sons, New York: 251-281. 21 Nolting, R. F., H. J. W. de Baar, A. J. Van Bennekom, and A. Masson. 1991. Cadmium, copper and iron in the Scotia Sea, Weddell Sea and Weddell/Scotia Confluence (Antarctica). Mar. Chem. 35: 219-243. Price, N. M., and F. M. M. Morel. 1990. Cadmium and cobalt substitution for zinc in a marine diatom. Nature 344: 658-660. Redfield, A. C , B. H. Ketchum, and F. A. Richards. 1963. The influence of organisms on the composition of seawater. In M. N. Hill (ed.), The Sea, Vol 2, Wiley Interscience, New York: 26-77. Rue, E. L., and K. W. Bruland. 1995. Complexation of Fe(III) by natural organic ligands in the central North Pacific as determined by a new competitive ligand equilibration/absorptive cathodic stripping voltammetric method. Mar. Chem. 50: 117-138. Rutgers Van Der Loeff, M., E. Helmers, and G. Kattner. 1997. Continuous transects of cadmium, copper, and aluminium in surface waters of the Atlantic Ocean, 50°N to 50°S: Correspondence and contrast with nutrient-like behaviour. Geochimica et Cosmochimica Acta 61: 47-61. Saager, P. M. 1994. On the relationships between dissolved trace metals and nutrients in seawater. Free University of Amsterdam (Ph.D. thesis) 240pp. Saager, P. M. and H. J. W. de Baar. 1993. Limitations to the quantitative application of Cd as a paleoceanographic tracer, based on results of a multi-box model (MENU) and statistical considerations. Global and Planetary Change 8: 69-92. Shaked, Y., A. B. Kustka, and F. M. M. Morel. 2005. A general kinetic model for iron acquisition by eukaryotic phytoplankton. Limnol. Oceanogr. 50: 872-882. Sunda, W. G., and S. A. Huntsman. 1995. Iron uptake and growth limitation in oceanic and coastal phytoplankton. Mar Chem. 50: 189-206. Sunda, W. G., and S. A. Huntsman. 1996. Antagonisms between cadmium and zinc toxicity and manganese limitation in a coastal diatom. Limnol. Oceanogr. 41: 373-387. Sunda, W. G., and S. A. Huntsman. 1998. Control of Cd concentrations in a coastal diatom by interactions among free ionic Cd, Zn, and Mn in seawater. Environ. Sci. Technol. 32: 2961-2968. Sunda, W. G., and S. A. Huntsman. 2000. Effect of Zn, Mn, and Fe on Cd accumulation in phytoplankton: Implications for oceanic Cd cycling. Limnol. Oceanogr. 45: 1501-1516. 22 Thomine, S., R. C. Wang, J. M. Ward, N. M. Crawford and J. I. Schroeder. 2000. Cadmium and iron transport by members of a plant metal transporter family in Arabidopsis with homology to Nramp genes. Proc. Nat. Acad. Sci. 97: 4991-4996. Van Ho, A., D. M. Ward, and J. Kaplan. 2002. Transition metal transport in yeast. Annu. Rev. Microbiol. 56: 237-261. Westerlund, S., and P. Ohman. 1991. Cadmium, cooper, cobalt, nickel, lead, and zinc in the water column of the Wedell Sea, Antarctica. Geochimica et Cosmochimica Acta 55: 2127-2146. Wu, J., and G. W. Luther. 1995. Complexation of Fe(III) by natural organic ligands in the Northwest Atlantic Ocean by a competitive ligand equilibrium method and a kinetic approach. Mar. Chem. 50: 159-177. Zhuang, G., Z. Yi, and G. T. Wallace. 1995. Iron (II) in rainwater, snow, and surface seawater from a coastal environment. Mar. Chem. 50: 41-50. 23 Chapter 2: The Effect of Phytoplankton Species Composition and Fe-limitation on the Cd:P043" Relationship1. 2.1 Abstract This study compares intracellular Cd content (Cd:C) of seven species of cultured marine phytoplankton from two phyla, under varying degrees of Fe limitation, with those of natural phytoplankton assemblages from Fe-replete and Fe-limited waters in the northeast subarctic Pacific. We also estimated surface particulate CdiPCM3" ratios from a global dataset and compared these values with the laboratory and field measurements. A greater than 65-fold difference in Cd:C ratios was observed between species grown under identical Fe concentrations. Oceanic diatoms had the highest Cd quotas and naked prymesiophytes the lowest. All species significantly increased their Cd:C ratios under Fe-limitation (on average 2-fold). The phytoplankton at the Fe-limitited station had higher Cd:C ratios than the Fe-replete station. The field results were partly attrituded to Fe-limitation as well as a flourstic shift to phytoplankton with higher intristic Cd:C ratios. The global data set also showed that the mean Cd:P ratio of surface water particulates in HNLC regions were approximately 2-fold higher than non-HNLC regions. The two extremes in the range of particulate Cd:P values our global dataset can be explined by the export blooms of either diatoms with high Cd:P ratios or naked prymesiophytes with low Cd:P ratios. We propose that the kink in the global dissolved Cd:P04 " relationship is a result of the export of Fe-limited phytoplankton with high Cd:P ratios in HNLC regions. 1 A version of this chapter will be submitted for publication. Lane, E. S., D. Semeniuk, J. T. Cullen and M. T. Maldonado (2007) The Effect of Phytoplankton Species Composition and Fe-limitation on the Cd:P043" Relationship. Limnology and Oceanography. 24 2.2 Introduction Cadmium (Cd) has a seawater distribution that is very similar to phosphate (PO43" ), suggesting that Cd, like PO4 ", is controlled by phytoplankton uptake in surface waters and the sinking and remineralization of organic matter with depth (Boyle et al. 1976; Bruland et al. 1978; de Baar et al. 1994). Cadmium is also incorporated into the calcium carbonate tests of certain foraminifera in proportion to the dissolved Cd concentrations of the seawater in which the foraminifera are growing (Boyle 1992). As a result, Cd:Ca ratios preserved in fossil foraminifera tests has been applied along with the strong Cd:P043" correlation to reconstruct past PO4 3" distributions in the deep ocean. More recently, Elderfield and Rickaby (2000) expanded the use of Cd:Ca ratios as a paleonutrient proxy to surface waters by using Cd:Ca ratios preserved in planktonic foraminifera. Elderfield and Rickaby (2000) using this proxy estimated PO4 3" utilization in surface waters of the Southern Ocean during the last glacial maximum (LGM) as a means of evaluating the Southern Ocean's role in regulating atmospheric CO2 concentrations across glacial transitions. In order to use Cd:Ca ratios as an accurate paleonutrient proxy, the Cd:P04 " relationship must remain constant both temporally and spatially (de Baar et al. 1994; Saager and de Baar 1993). However, in general surface waters have lower dissolved Cd:P043" ratios (de Baar et al. 1994) and show significant spatial variability (Elderfield and Rickaby 2000) in comparison to ratios in underlying deep waters, resulting in a non-linearity or "kink" in the global Cd:P043" relationship. Several factors that may affect dissolved Cd:P043" variability in surface waters have been identified. Laboratory studies (Lee and Morel 1995; Sunda and Huntsman 25 1996, 1998, 2000) and shipboard incubation experiments (Cullen et al. 1999; Cullen and Sherrell 2005) have demonstrated that the uptake of Cd by phytoplankton is directly related to ambient Cd concentrations and inversely related to Zn and Mn free ion concentrations. However, our understanding of how these metals effect Cd:P varablity in the open ocean is limited by our understanding of trace-metal speciation of Cd, Zn, and Mn and by lack of direct measurements of free ion concentrations for the salient metals in the oceanic euphotic zone. Therefore, enhanced intracellular Cd levels in indigenous phytoplankton as a result of low Zn or Mn concentrations along natural gradients have never been observed in the ocean, despite their importance on phytoplankton Cd quotas in well-controlled laboratory and field manipulation experiments. Aqueous C0 2 concentrations in the growth media have also been shown to control phytoplankton Cd levels in culture (Lane and Morel 2000). Shipboard incubation experiments showed that the particulate Cd: PO4 3" ratios of phytoplankton were two to five times higher for cells grown at low pCC»2 than for cells acclimated to high pCC>2 (Cullen and Sherrell 2005). In support of this, the Cd content of natural phytoplankton increased with decreased surface water pCC»2 in the intensely productive upwelled waters off the coast California (Cullen et al. 1999). The range of CO2 concentrations (250 - >800ppm) found in the surface waters off the coast of Califorina are considerably larger than natural gradients found in the open ocean. Based on the combined laboratory and field results it is unlikely that CO2 could account for the Cd:P varabity found in surface waters of the open ocean. Incidentally, dissolved Cd:PCV" ratios are especially low in surface waters of Fe-limited regions of the ocean (Martin et al. 1989; Martin et al. 1990). The kink in the global dissolved Cd versus PO4 " relationship has been proposed to result from chronic 26 Fe-limiting conditions in high-latitude HNLC areas in the modern ocean (Cullen 2006). This suggests that Cd may be preferentially removed relative to PO4 3" by phytoplankton in Fe limited waters (Cullen 2006). In support of this, Sunda and Huntsman (2000) demonstrated that Fe-limited cultures of Thalassiosira oceanica have significantly elevated cellular Cd content in comparison to Fe-replete cultures. This increase in Cd quotas was hypothesised to result from the depletion of ambient Zn concentrations in response to Fe limitation and to a lesser extent from biodilution. The hypothesis of biodilution suggests that phytoplankton accumulate higher intracellular Cd under Fe-limitation because Cd uptake rates are kept constant while growth rate is reduced. Biodilution is characterized by the equation P = V>Q where p is the normalized cellular uptake rate (mol Cd cell"1 time"1), JX is the specific growth rate (time"1), and Q is the cellular quota (mol Cd cell"1). In the field, the idea of biodilution was also used to explain the decrease in particulate Cd-.PCV" ratios upon Fe supplementation during a shipboard incubation study with Fe-limited natural phytoplankton assemblages from the Southern Ocean (Cullen et al. 2003). Contrary to Sunda and Huntsman (2000), increased Cd accumulation by Fe-limited phytoplankton occurred despite the fact that total dissolved and free ion Zn concentrations were high. This seems to refute the idea that depletion of Zn by Fe-limited phytoplankton is the main factor controlling preferential Cd uptake by phytoplankton in Fe-limited waters. Biodilution can only occur when growth rates decline more rapidly than metal transport rates (Strzepek and Price 2000). Sunda and Huntsman (2000) measured a -50% reduction in the growth rate of Thalassiosira oceanica due to Fe limitation associated 27 with a 20-160% increase in intracellular Cd:C ratios. This goes against the biodilution hypothesis, as in many cases Cd uptake rates had to be faster than the reduction in growth rates to account for the observed increase in Cd quotas. Alternatively, the higher Cd quotas in Fe-limited phytoplankton may result from the transport of Cd through non-specific divalent metal transporters, which are up-regulated under Fe-limitation, as suggested by previous physiological data from our laboratory. Thalassiosira oceanica Fe(II) uptake rates increased up to 15-fold under severe Fe-limitation and a significant antagonistic interaction between Fe(U) and Cd uptake occurred (Lane et al., Chapter 3). Further evidence for an Fe(II) transporter in diatoms was found in a recent study, where a putative gene for a divalent metal transporter (TpNRAMP) was identified in the genome of Thalassiosira pseudonana (Kutska et al. 2007). The expression of this gene was also dramatically up-regulated under Fe-limitation (Kustka et al. 2007) and prompted the use of Thalassiosira .pseudonana in our study. Thalassiosira pseudonana was the only coastal species used in this study, as the main objective of this research was to investigate the effects of Fe-limitation on open ocean phytoplankton. Although numerous studies have examined factors that control Cd content in marine phytoplankton (Lee and Morel 1995; Sunda and Huntsman 1996, 1998, 2000; Lane and Morel 2000; Cullen et al. 1999; Cullen et al. 2003; Cullen and Sherrell 2005), this is the first study to systematically investigate the effect of Fe-limitation on the Cd quotas of more than one species/phyla of cultured phytoplankton. We cultured seven species of phytoplankton from two different taxa under varying concentrations of Fe and determined their corresponding growth rates, Fe:C and Cd:C ratios. These measured 28 Cd:C ratios were then compared to the Cd:C ratios of natural phytoplankton assemblages from an Fe-replete and an Fe-limited station in the subarctic Pacific Ocean. Both laboratory and field Cd:C ratios were normalized to PO4 3" and were compared to a global oceanic dataset of calculated surface waters particulate Cd:P043~ ratios. 2.3 Materials and Methods 2.3.1 Culturing Most of the phytoplankton isolates were obtained from the northeast Pacific Culture Collection (NEPCC; University of British Columbia, Vancouver, Canada) and the Center for Culture of Marine Phytoplankton (CCMP; Bigelow Laboratory, West Boothbay Harbor, ME, USA). Robert Strzepek (University of Otago, New Zealand) kindly donated the Southern Ocean isolates (Table 2.1). The chemically well-defined artificial seawater medium AQUIL (Price et al. 1988/89) was used as culture media. All culture containers were washed in X-Tran detergent to remove organic matter, and then soaked in 10% HC1 for at least 24 hours, before being washed with purified Milli-Q water and sterilized (Price et al. 1988/89). Synthetic ocean water (SOW) at pH 8.2, containing standard additions of phosphate (10 /miol IS PO4 ") and silicic acid (100 /miol IS Si03 ') was prepared and was chelexed to remove trace metals. This basal media was microwave-sterilized in acid-washed Teflon bottles, and enriched with filter-sterilized (0.2 pm, Acrodisc) chelexed nitrate (300 /unol L"1 NO3-,), vitamins (biotin, thiamine and B12), and trace metals. Trace metal concentrations were buffered with 100 /miol L"1 ethylene-diamine-tetra-acetate (EDTA) in all cultures media, except for those of the Southern Ocean species, which were 29 buffered with 10 /rniol L"1 EDTA. In all cases, trace metal concentrations were added such that Cu, Mn, Zn and Co were present at free-ion concentrations of 10~1319, 10"821, 10"10'88, 10"10 8 8 mol L"1, respectively (Maldonado and Price 1996). Selenium and molybdenum were added at total concentrations of 10"8 and 10"7 mol L~ l, respectively. Trace metal concentrations were calculated using the MINEQL model (version 2.0, Westall et al. 1976). Premixed Fe-EDTA (1:1.05) was added separately to achieve total Fe concentrations ([Fe]T) of 1.37 jumol L"1 (pFe 19, pFe = -log[Fe+3]), 42 nmol L"1 (pFe 20.5), 12.9 nmol L"1 (pFe 21), and 4.2 nmol L"1 (pFe 21.5). Cultures were grown in pFe 19 medium to obtain Fe-replete growth, whereas pFe 20.5, pFe 21 and pFe 21.5 media were used to obtain Fe-limited growth. Southern Ocean species were grown in Fe-replete media, containing 4 nmol L"1 Fe (pFe 20.74). In order to elicit Fe-limitation in the Southern Ocean isolates, Fe was added bound to DFB (desferriferrioxime B). Phaeocystis antarctica was Fe-limited with an addition of 4 nmol L"1 Fe bound to 400 nmol L"1 DFB, whereas Proboscia inermis was grown with 4 nmol L"1 Fe bound to 40 nmol L"1 DFB. A duplicate of each culture medium was made containing Cd (pCd=12, pCd = -log[Cd2+]). Cadmium was added as CdCl2, at a total concentration of 35nmol L"1 in order to mimic in situ oceanic pCd levels in typical HNLC waters. Cullen (2006) measured a total Cd concentration ([Cd]x) of 0.25 nmol L"1 at a typical HNLC surface station in the Bering Sea. About 70% of [Cd]t is present as organic complexes, leaving -30% as inorganic species ([Cd']) (Bruland 1992). Approximately 3 % of the inorganic Cd species in seawater is present as free Cd 2 + (Byrne et al. 1988). Therefore, a surface [Cd2+] of -2.2 pmol L"1 was present at Cullen's (2006) HNLC station, which corresponds to a pCd 30 of 11.7. The pCd = 12, used in this study is a conservative value and lies on the lower 2"f" end of the free Cd range likely to be encountered in high latitude HNLC regions. Since, Southern Ocean species were grown with 10 times lower EDTA concentration, only 3.5 nmol L"1 Cd was added. Metal-EDTA reactions in the media were allowed to equilibrate overnight before media was used. 2.3.2 Growth Measurements All cultures were grown in acid-washed 28 ml polycarbonate tubes. Southern Ocean cultures were incubated at 4°C, under 24h photon flux density of 100 pmol quanta m"2 s"1. All other cultures were grown at 19 ± 1°C under a continuous saturating light 9 1 level of 175 jumol quanta m" s" . Cultures were constantly maintained in exponential phase by serial dilution into fresh medium as needed. Culture growth was monitored daily, using fluorescence as a proxy for chlorophyll a. In vivo fluorescence was measured by a Turner 10-AU Fluorometer. Absolute growth rates (p) were determined as doublings per day (dd-1) from linear regressions of the log2 of in vivo fluorescence versus time during the exponential growth phase of acclimated cells, after confirming a proportional relationship between fluorescence and cell concentration. Cultures were considered acclimated when growth rates in successive transfers varied by less than 10% (typically 4-5 transfers). The r values of in vivo fluorescence versus cell concentration standard curves for phytoplankton species at each Fe treatment were >0.96 (data not shown). 31 2.3.3 Fe and C d cell quotas Intracellular Fe:C ratios were determined using an 5 5Fe and 1 4 C dual label, radiotracer technique (Tortell et al. 1996). For the Fe-replete medium (pFe 19), 1% of the total Fe was added as the 55FeCl3 (specific activity of stock, 70 MBq mmol"1, PerkinElmer). For the low-Fe medium, 10% of the total Fe was 5 5Fe for pFe 20.5 and 5 0% for pFe 21 and pFe 21.5. Fe additions were pre-mixed with EDTA (1:1.5) in Teflon vials. For Southern Ocean treatments, 50% of the total Fe (2nmol L"1) was added as 5 5Fe pre-complexed with either EDTA or DFB. In addition, 0.74 MBq L"1 of H 1 4C0 3" (specific activity of stock, 1665 MBq mmol"1, PerkinElmer) was added to all treatments to enable the simultaneous measurement of intracellular Fe and C content. Intracellular Cd:C ratios were determined using the radioisotope 1 0 9 Cd (specific activity of stock 133 MBq mmol"1, Amersham). The percentage of 1 0 9 Cd added to the media was calculated in order to achieve between 0.67 and 0.84 MBq L"1. Due to the fact that 1 0 9 Cd and 1 4 C have overlapping energy windows, dual labeling is not possible. Therefore, the Cd cell"1 determined from the 1 0 9 Cd cultures was normalized by the C cell" 1 value obtained from its corresponding 5 5Fe: 1 4C quota measurements. To initiate experiments, 1000-5000 cells in late exponential phase, pre-acclimated to that specific Fe concentration, were inoculated in triplicate into an acid washed, 28-mL polycarbonate tube containing identical pFe equivalent medium, with tracer amounts of radionuclides. At least eight cell divisions occurred before harvesting the culture to ensure that 99% of the intracellular metal content within a cell was radio-labeled. Growth rates were measured using in vivo fluorescence as described above. When cultures reached mid- to late-exponential phase, the cells were harvested onto 2.0 pm 32 AMD polycarbonate filters and washed for 5 minutes with titanium-EDTA-citrate reducing solution (Hudson and Morel 1989) or 1 mmol L"1 DTP A (Lee and Morel 1995) (dissolved SOW, pH adjusted to -8.1) to remove extracellularly bound Fe or Cd, respectively. Incorporated 5 5Fe and 1 4 C and 1 0 9 Cd were then measured with a liquid scintillation counter (Beckman, LS6500). To correct for the interference between 5 5Fe and 1 4 C radionuclides, a dual-labeled quench curve was performed. To determine total cell concentrations in these cultures, radioactive cell samples were preserved with Lugol's solution and counted using a Zeiss light microscope. 2.3.4 Field Measurements The subarctic Pacific is a well known Fe-limited HNLC region (Harrison 2002). Line P is a well studied transect from the coastal, Fe-replete waters of British Columbia to the open ocean Fe-limited site, P26 (Harrison 2002). All samples were collected in September 2006 from the CCGS John P. Tully. Station P4 is located off the coast of British Columbia and P26 is the site of Ocean Station PAPA, OSP (Whitney and Freeland 1999). The ship's coordinates were 48°39.008 N and 126°39.041 W for P4 and 49°59.993 N and 144°18.199 W for P26. Unfiltered seawater was collected for chlorophyll a, dissolved cadmium, phytoplankton species counts and experiments, within the mixed layer at a depth of 25 m at P4 and 10 m at P26. This seawater was collected using a trace metal clean Teflon peristaltic pump and Teflon tubing system suspended on a Kevlar line. The seawater was pumped directly into a laminar flow hood where it was distributed into either HDPE trace metal clean bottles for dissolved Cd determinations, or into 1 L polycarbonate bottles for the Cd:C quota determinations. The seawater was 33 stored at 4 °C prior to analysis (dissolved Cd) or to the start of all experiments (Cd:C ratios). All bottles were cleaned with a combination of strong and weak HC1 and HNO3 over three weeks before departing. Initial samples for nutrients were drawn directly from 30-liter Teflon-coated GO-Flo bottles. Nutrient samples (NO3" and PO43") were analyzed fresh on board using a Technicon AAII auto-analyzer following methods described in Barwell-Clarke and Whitney (1996). Size fractionated chlorophyll a concentrations were measured in acetone extractions using a Turner Designs 10-AU-005-CE fluorometer (Price et al. 1994). Dissolved Cd was measured according to Cullen (2006). Phytoplankton were counted and identified to species/genus using light microscopy. Phytoplankton biomass (fig C L"1) from cell counts was determined using cell biovolumes estimated by Haigh (1992). Experimental manipulations were carried out in a portable laminar flow hood. Size-fractionated Cd:C ratios of the phytoplankton community were determined in duplicate with 1 0 9 Cd (specific activity of stock 133 MBq mmol"1, Amersham). Before 1 0 9 Cd was added to incubation bottles, it was equilibrated in seawater for at least 30 minutes. Bottles were capped and sealed with parafilm, and incubated on deck for 24 hours under in situ light and temperature using neutral density screening and continuously flowing seawater pumped from 5 m depth. After 24 hours, the content of each bottle was filtered onto Poretics polycarbonate filters (porosities of 2, and 20 pm) separated with nylon drain disks and rinsed for 5 minutes with 1 mmol L"1 DTPA (Lee and Morel 1995) (dissolved in SOW, pH adjusted to -8.1) to remove extracellularly bound Cd. Incorporated 1 0 9 Cd was measured with a liquid scintillation counter (Beckman, LS6500). Cadmium quotas were normalized to C quotas determined in 34 parallel bottles using H ,4C03" (Semeniuk et al., unpub.). 2.3.5 Global Data Set Relying on the assumption that dissolved Cd:P043" ratios in the oceanic nutriclines are the result of the remineralization of Cd and PO43" in sinking particulate organic matter (Broecker and Peng 1982), surface water particulate Cd:P043" ratios were estimated from dissolved Cd:P043" ratios in the nutricline. Data from stations in the Atlantic, Indian, Southern and Pacific Oceans was complied from all the literature where dissolved Cd and PO43" were measured simultaneously at individual stations. Dissolved Cd and P043" concentrations were plotted as a function of depth and only stations that showed nutrient like profiles for both Cd and PO43" were used (a total of 138 stations). The depth in each profile where the dissolved concentration strongly increased to the depth where it reached a maximum and then remained constant below this depth was taken to be the nutricline. Cadmium concentrations were then plotted against PO43" concentrations for each depth within the nutricline. A linear regression was performed on each plot and the slope of the Cd versus PO43" relationship was determined to be the particulate Cd: PO4 "ratio in surface water. Stations were considered to be Fe-replete when surface PO43" was drawn down to < 0.3 pmol L"1 by the growth of phytoplankton and Fe-limited when PO43" concentrations were > 0.3 pmol L' 1 (Cullen 2006). Stations that did not have corresponding surface PO43" concentrations were classified as Fe-replete or Fe-limited based on the Fe regime in the general area where the station was located. 35 2.4 Results 2.4.1 Growth rates In order to determine if Fe-limited phytoplankton have higher Cd quotas than their Fe-replete counterparts, seven species of phytoplankton (Table 2.1) were grown under Fe replete and limiting conditions in the presence and absence of Cd (Table 2.2). The physiological status of each species in relation to Fe and Cd nutrition was assessed by measuring growth rates. In all cases, decreasing Fe availability reduced the growth rates (p < 0.01, t-test), although species differed in the magnitude of their response to Fe-limitation (Table 2.2). Thalassiosira pseudonana, the only coastal species used in this study, had the fastest relative growth rate in Fe-replete media, and reduced its growth -rates by 33% when grown in pFe 20.5 media. In order to elicit further Fe-limitation, Thalassiosira pseudonana was grown at pFe 21, however, it was unable to survive due to severe Fe limitation. In contrast, only a -20% reduction in the growth rate of .Thalassiosira oceanica was observed when grown at pFe 21. Thalassiosira oceanica was therefore subjected to further Fe-limitation (pFe 21.5), causing a 46% reduction in its growth rate. Emiliania huxleyi also showed a 46% reduction in growth rate under Fe-limitation (pFe 21). The two subarctic Pacific naked Prymniosophytes, Phaeocystis pouchetti and Chrysochromulina polylepis, proved difficult to Fe-limit and showed only a -25% growth rate reduction in pFe 21.5 media. The two Southern Ocean isolates, Phaeocystis antarctica and Proboscia inermis achieved the slowest relative growth rates, showing a -40% and a -30% reduction in growth rate, respectively, when grown in the presence of DFB (4:400 and 4:40 nmol L"1 Fe: nmol L"1 DFB, respectively). All Fe growth treatments were replicated in the presence of pCd 12. With the exception of 36 Phaeocystis antarctica, Cd had no significant effect on the growth rates of the isolates when grown in either high Fe or low Fe media. In the presence of Cd, Phaeocystis antarctica surprisingly increased its Fe-replete growth rate from 0.43 to 0.51 dd"1 and its Fe-limited growth rate from 0.26 to 0.31 dd"1 (p < 0.05, t-test). 2.4.2 Intracellular Fe:C ratios and Fe-use efficiencies A wide range of Fe quotas were found among the seven phytoplankton species. In Fe-replete conditions, the Fe:C ratios of all species ranged from 2.7±0.029 to 108.9±2.1pmol Fe mol C"1. In low-Fe conditions, cellular Fe:C ratios of all isolates decreased significantly (p < 0.01, ANOVA) compared to those of Fe-replete cultures. The Fe:C ratios of species ranged from 0.61±0.020 to 19.3±0.39 umol Fe mol C 1 among the different low-Fe treatments. The coastal strain Thalassiosira pseudonana had the highest Fe quotas for both Fe-replete (108.9±2.1 pmol Fe mol"1 C) and low-Fe (19.3±0.39pmol Fe mol"1 C) treatments. In contrast, the oceanic strain Thalassiosira oceanica reduced its Fe:C ratio from 88.8±1.7 to 7.5±0.78 to 3.1±0.094 pmol Fe mol"1 C, as it was subjected to greater Fe-limitation. The prymnesiophytes-, (Emiliania huxleyi, Phaeocystis pouchetti, and Chrysochromulina polylepis) all had a similar Fe quota of 55.0 ±1.5 pmol Fe mol"1 C at pFe 19. In low-Fe media Emiliania huxleyi had an Fe:C ratio of 12.3±0.56 pmol Fe mol"1 C whereas, Phaeocystis pouchetti and Chrysochromulina polylepis were able to achieve much lower Fe-quotas (5.8±0.11 and 6.2±0.36 pmol Fe mol"1 C, respectively). Overall the Southern Ocean isolates had the lowest Fe quotas. In Fe-replete media, Phaeocystis antarctica and Proboscia inermis had Fe:C ratios of 3.3±0.083 and 2.7±0.029 pmol Fe mol"1 C, which were reduced to 37 0.61±0.020 and 0.65±0.032 pmol Fe mol"1 C, respectively, under Fe-limitation. Cadmium had no significant effect on the Fe quotas of all the isolates investigated (ANOVA, p > 0.05, Tukey test), except for Phaeocystis antarctica. This species, in the presence of Cd, increased its Fe quota in high Fe media from 3.3±0.083 to 5.5±0.3pmol Fe mol"-1 C and in low Fe media from 0.61±0.020 to 0.84±0.03 pmol Fe mol"1 C (p < 0.05, ANOVA). Iron-use efficiency (IUE) is defined as the rate of assimilated carbon per unit of cellular Fe (mol C mol Fe"1 d"1). The IUEs of all isolates increased when grown in low Fe conditions (p < 0.01, ANOVA) (Table 2.2). The Southern Ocean species had remarkably high IUEs under both high and low Fe conditions. Except for Phaeocystis antarctica which decreased its IUE in the presence of Cd (ANOVA, p < 0.05,Tukey test), no significant difference in the IUE was measured in the presence of Cd for the phytoplankton species studied (p > 0.05, ANOVA). 2.4.3 Intracellular Cd:C ratios and steady state Cd uptake rates Cadmium quotas were variable among the phytoplankton species (Table 2.3). The seven species used in this study were separated into four groupings (coastal diatoms, oceanic diatoms, coccolithophores and naked prymesiophytes). Both grouping and Fe-status were found to have a significant effect on the Cd:C ratios (p < 0.01, ANOVA). With the exception of the coastal diatoms and naked prymnesiophytes, the Cd:C ratio of each group was statistically different from the others (ANOVA, p < 0.01, Tukey test). The oceanic diatoms had the highest Cd:C ratios followed by coccolithophores, and coastal diatoms and the naked prymnesiophytes. 38 The Cd:C quotas of Fe-replete cultures ranged from 0.12±0.007 to 7.9±0.6 pmol Cd mol C"1, with an average Cd quota of 2.46 pmol Cd mol C"1. In low Fe media the quotas ranged from 0.22±0.02 to 12.1±0.3pmol Cd mol C"1, with an average Cd:C ratio of 4.68 pmol Cd mol C"1. All species significantly increased their Cd quota under Fe-limitation. The lowest Cd:C ratios were measured for two subarctic Pacific prymnesiophytes, Phaeocystis pouchetti and Chrysochromulina polylepis, in pFe 19 media (0.12±0.007 and 0.15±0.01 pmol Cd mol C"1, respectively). The Southern Ocean prymnesiophyte, Phaeocystis antarctica also had a relatively low Cd quota in high Fe media (0.40±0.05 pmol Cd mol C"1) but showed the highest increase (-80%) in intracellular Cd when Fe limited. Emiliania huxleyi had much higher Cd quotas (3.1±0.2 and 5.8±0.1 pmol Cd mol C"1 in pFe 19 and pFe 21 media, respectively) than the other prymnesiophytes. However, the highest Cd quotas were measured in the oceanic diatoms. Thalassiosira oceanica increased its Cd quota to 10.3±0.2 pmol Cd mol C"1, as it was subjected to severe Fe-limitation. The highest Cd quota ever measured corresponded to the Southern Ocean diatom Proboscia inermis and was 12.1±0.3 pmol CdmolC"1. The calculated steady-state Cd uptake rates ranged from 0.11±0.0002 to 6.97±0.1 pmol Cd mol C 1 d"1 (Table 2.3). For all species, the steady-state Cd uptake rates were markedly higher for Fe-limited than Fe-sufficient cultures (p < 0.01, ANOVA). The only culture that did not show this trend was mildly Fe-limited Thalassiosira oceanica grown in pFe 21 media, as its steady-state uptake rate was not significantly different than that of the Fe-replete culture (ANOVA, p > 0.05, Tukey test). However, when T. oceanica was grown in pFe 21.5 media its steady-state Cd uptake rate increased significantly. 39 2.4.4 Intracellular Cd:C ratios of a natural phytoplankton assemblages The Cd quotas of two natural phytoplankton assemblages in the subarctic Pacific Ocean (P4, an Fe-replete station and P26, an Fe-limited station) were determined in September 2006, in order to compare our laboratory finding to the natural environment. Station P4 had relatively low macronutrient concentrations, with surface NO3" and PO4 3" concentrations of 2.7 and 0.65 pmol L"1, respectively. Station P26 had higher macronutrient concentrations, with a surface NO3" value of 9.8 pmol L"1 and a PO4 3" concentration of 1.1 pmol L"1. Dissolved surface Cd concentrations were 0.156 pmol L"1 at P4 and 0.2 nmol L"1 at P26. Size fractionated surface chlorophyll a concentrations decreased from P4 (0.27 and 0.20 ixg L"1 in the 2-20 and > 20 size fraction, respectively) to P26 (0.12 and 0.04 pg L' 1 in the 2-20 and > 20 size fraction, respectively). Phytoplankton abundance increased from 6.35 x 104 cells L"1 at P4 to 11.2 x 104 cells L"1 at P26. Small phytoplankton dominated the phytoplankton abundance at P26, however only accounted for a small proportion of the biomass. Biomass was 24.14 \xg C L"1 at P4 and 88.58 jug C L"1. The > 20 um size fraction at P4 consited of dinoflagellates and oceanic diatoms. Large dinoflagellates accounted for a large proportion of the total biomass in the > 20 /mi size fraction at P26. Naked flagellates were present in equal biomass in the 2-20 um size fraction at both sites with the remaining biomass consisting predominantly of dinoflagellates at P4, and approximately equal proportions of diatoms, coccolithophores, and dinoflagellates at P26. There was a significant increase in particulate Cd:C ratios in both the 2pm and 20pm size fractions (p < 0.01, t-test) from P4 to P26 (Figure 1.2). Particulate Cd:C ratios 40 of phytoplankton in the 2-20 pm size class increased from 0.51±0.23 to 2.28±0.22 pmol Cd mol C"1. Larger phytoplankton in the > 20pm increased their Cd:C ratios from 0.62±0.27 to 2.27±0.27 pmol Cd mol C"1. 2.4.5 Particulate Cd:P043~ ratios from global dataset The particulate Cd:PC>43" ratios calculated from dissolved Cd:PC»43" ratios in oceanic nutriclines in our global dataset ranged from 0.091 to 1.87 nmol Cd pmol"1 PO4 3 " (Table 2.4, Figure 2.2). The average Cd:PC>43" ratio for Fe-replete stations was 0.241 nmol Cd pmol"1 PO4 3 ". Fe-limited stations had a significantly higher (p < 0.001, t-test) Cd:PC>43" ratios with an average ratio of 0.543 nmol Cd pmol"1 PO4 3 ". When broadly classifying all stations based on the Fe-regime of the general area where the station is located (Southern Ocean HNLC, Sub-arctic Pacific HNLC, and other HNLC stations, non-HNLC regions) it is apparent that Fe-regime is the dominant player controlling surface particulate Cd:P043' ratios. All HNLC stations are statistically different from non-HNLC stations. However, the Southern Ocean HNLC stations and Sub-arctic Pacific HNLC stations are different from each other (ANOVA, p < 0.05, Tukey test), but not from the other HNLC stations (ANOVA, p > 0.05, Tukey test). 2.5 Discussion Several studies, both in the field and laboratory, have suggested that phytoplankton accumulate higher intracellular Cd quotas under Fe-limitation. This study however, is the first to combine measured Cd quotas of several species of cultured 41 phytoplankton under varying degrees of Fe limitation; of natural phytoplankton assemblages from an Fe-limited and an Fe-sufficient station in the subarctic Pacific; as well as those calculated from dissolved Cd and PO43" in the nutriclines of a global data set. All three means of determining phytoplankton Cd quotas show that particulate Cd:PC>4 ' increases with Fe-limitation. The export of phytoplankton species with high particulate Cd quotas could be a major factor controlling dissolved Cd:P ratios in surface waters and nutriclines of oceanic HNLC regions. 2.5.1 Evidence for Fe-limitation All phytoplankton decreased their growth rates by at least 20% in response to reduced Fe availability. As a result, all species were considered significantly Fe-limited in their respective low Fe media (Table 2.2). The reduction in their corresponding cellular Fe requirements (Fe:C ratios), provides further evidence for the Fe-limited condition of the cultures in low Fe media (Table 2.2). Thalassiosira pseudonana was affected the most by Fe-limitation, and was unable to survive in pFe 21 media. The highest Fe quotas were also measured for Thalassiosira pseudonana in both Fe-replete and Fe-limiting media (Table 2.2). These results are consistent with other studies (Brand 1991; Sunda et al. 1991; Maldonado et al. 2006), which show coastal species are more vulnerable to Fe-limitation than their oceanic couterparts. The average minimum Fe:C ratio measured for an oceanic phytoplankton is ~3 /xmol Fe mol"1 C (Sunda and Huntsman 1995a; Maldonado and Price 1996). With the exception of the Southern Ocean species, all phytoplankton in this study were found to have minimum Fe quotas in this range (Table 2.2). The Southern Ocean species, 42 Phaeocystis antarctica and Proboscia inermis, have the lowest Fe quotas ever measured (-0.6 pmol Fe mol"1 C). These low Fe quotas are well below the theoretical phytoplankton minimal Fe requirements and allowed Phaeocystis antarctica and Proboscia inermis to grow at 50% and 67% of ju^x, respectively. The non-Fe-limited quotas of both Southern Ocean species were also very low (~3 /rniol Fe mol"1 C) and were within the lower range of Fe quotas for severely Fe-limited oceanic phytoplanton (Sunda et al. 1991; Sunda and Huntsman 1995a; and Maldonado and Price 1996; Marchetti et al. 2006). Southern Ocean phytoplankton experience prolonged periods of Fe-limitation and appear to have evolved mechanisms for lowering their cellular Fe requirements. The means by which these Southern Ocean isolates achieve such amazingly low Fe quotas merits further investigation. 2.5.2 The effect of Cd on the growth rate and Fe:C ratios Cadmium additions had no significant effect on the growth rates or Fe quotas of most species. Previous work in our laboratory has shown that Cd has a significant antagonistic effect on inorganic Fe(II) uptake but not on inorganic Fe(III) uptake (Lane et al., chapter 3). Thus, in order to see a significant effect of Cd in the Fe quotas, a considerable concentration of Fe(II) relative to Fe(III) in our culture media is necessary. Since the highest Fe treatment in this study is pFe 19, this treatment is expected to have the highest Fe(II) concentration. Our calculated inorganic Fe (Fe') concentration in pFe 19 media, as a result of thermodynamic and photo-mediated dissociation of FeEDTA was estimated as 1 nmol L"1 (Hudson and Morel 1990; Sunda and Huntsman 1995a). Only 0.1%) of this Fe' fraction was estimated to be present as inorganic Fe(II) (Sunda and 43 Huntsman 1995a). Therefore, it is not surprising that Cd had no effect on the species growth rates or Fe quotas, as there was virtually no Fe(II) present, and Cd has been found to only interfere with Fe(IT) uptake (Lane et al., chapter 3). Phytoplankton may have non-specific divalent metal transporters in order to acquire other essential divalent metals even when no Fe(II) is present in the media. Furthermore, Kustka et al. (2007) found TpNRAMP was regulated by Fe' concentrations. The fact that Cd had no effect on the phytoplankton growth rates and Fe quotas also suggests that Cd was well below concentrations previously shown to be toxic (Payne and Price 1999). Unexpectedly, Cd significantly increased growth rates and Fe quotas of Phaeocystis antarctica in both high and low Fe treatments. The mechanism mediating this positive physiological effect on Phaeocystis antarctica is presently unknown. 2.5.3 Range of Cd:C ratios Our measured Cd quotas varied over 65-fold among species grown under identical high Fe concentrations. Among all phytoplankton species, oceanic diatoms had the highest Cd quotas followed by the coccolithophore Emiliania huxleyi, and the naked prymesiophytes. Higher Cd quotas were also measured in the oceanic diatom Thalassiosira oceanica compared to its coastal counterpart Thalassiosira pseudonana. The variability in Cd quotas and the absolute Cd:C ratios presented here are in good agreement with those previously reported (Sunda and Huntsman 1998; Sunda and Huntsman 2000; Ho et al. 2003). Ho et al. (2003) reported that the Cd content of 15 marine phytoplankton species varied by over two orders of magnitude. Their average Cd quota (1.43 pmol Cd mol C"1, normalizing their average Cd quota to average C quota) is 44 lower than our average Cd:C ratio (3.65 pmol Cd mol C"1). The difference between these mean average Cd quotas may simply reflect the differences in Cd requirements between coastal and oceanic species (Ho et al. 2003), as 11 out of the 15 species used by Ho et al. (2003) were coastal/estuarine species, while ours were predominately oceanic strains. In contrast to our study, Ho et al. (2003) measured higher Cd quotas in coccolithophores than in diatoms. The authors note that all the coccolithophore isolates were oceanic, whereas the majority of diatom isolates were from coastal/estuarine waters. This habitat difference between the coccolithophores and diatoms examined by Ho et al. (2003), may explain the lower Cd quotas observed in their diatoms. The one oceanic diatom investigated (Ditylum brightwellii), had relatively high Cd quotas compared with the coastal diatoms (Ho et al. 2003). This result is consistent with our Cd quotas for Thalassiosira oceanica and Thalassiosira pseudonana. 2.5.4 Fe-limitation and Cd:C ratios The average Cd:C ratios of all isolates were significantly higher under Fe-limitation (4.68 pmol Cd mol C"1) than their Fe-replete controls (2.46 pmol Cd mol C"1). Assuming Redfield stoichiometry (C:PC>43" = 106:1), the corresponding Cd:P043" ratios were 0.496 nmol Cd pmol"1 PO4 3" and 0.261 nmol Cd pmol"1 PO4 3", respectively. These values agree nicely with the mean calculated particulate Cd:PC>43" ratios in our global dataset (0.543 versus 0.241 nmol Cd pmol"1 PO4 3" for Fe-limited and Fe-replete stations, respectively). However, assuming Redfield stoichiometry may not be the most accurate method for converting our measured Cd:C ratios to Cd:PC»43" ratios. Iron-limitation has been shown to decrease the C:PC>4 "ratios of Thalassiosira weissflogii, due to an increase 45 in the volumetric PO4 3" content of the cells (Price 2005). Based on this study, our calculated Cd:P043~ ratios may be slightly overestimated. Nonetheless, the measured Cd:C ratios in all species increased significantly under Fe-limitatation. The increase in intracellular Cd must be greater than the increase in intracellular PO4 3" in order to account for the high Cd:P043" ratios found in Fe-limited phytoplankton (Cullen et al. 2003). Depending on the species, Cd:P043" ratios have also been shown to deviate from the Redfield ratio (Ho et al. 2003; Geider and La Roche 2002). Some of our converted Cd:P043" ratios may have therefore been under or over estimated. Phaeocystis antarctica showed the greatest increase (4.75-fold) in its Cd quota under Fe-limitation. Previous uptake experiments suggest that Cd and Fe(II) enter the cell via a non-specific divalent metal transporter that is up-regulated under Fe-limitation (Lane et al., Chapter 3). This may imply that Phaeocystis antarctica is better suited to up-regulate its non-specific divalent metal transporters under Fe-limitation. Iron(II) has been found to be relatively stable in Southern Ocean waters (Croot et al. 2001), due to high UV radiation (Rijkenberg et al. 2005), cold temperatures (Millero et al. 1987), and low hydrogen peroxide levels (Yocis 2000). Up-regulation of non-specific divalent metal transporters would therefore be advantageous for Fe-limited Southern Ocean phytoplankton as a means of acquiring Fe, during Fe(II) deposition events. 2.5.5 Steady-state Cd uptake Two previous studies have also shown that the Cd quotas of marine phytoplankton are enhanced under Fe-limitation (Sunda and Huntsman 2000; Cullen et al. 2003). This observation was previously attributed to growth rate biodilution effects 46 on intracellular Cd content. According to Sunda and Huntsman (2000), a -50% reduction in the growth rate of Thalassiosira oceanica due to Fe limitation, while keeping a constant Cd uptake rate would lead to the observed increase in cellular Cd quotas. However, this -50% reduction in the growth rate of Thalassiosira oceanica due to Fe limitation was associated with anywhere from a 20 to 160% increase in intracellular Cd:C ratios suggesting that Cd uptake rates were highly up-regulated and not constant. Contrary to this biodilution hypothesis, we observed faster steady state Cd uptake rates for all Fe-limited cultures. This is consistent with other physiological studies that indicate metal uptake by phytoplankton may be up or down-regulated within hours after a change in total metal concentration in the media (Kustka et al. 2007). Therefore, we believe that the up-regulation of non-specific divalent metal transporters observed under Fe-limitation may explain these faster rates. 2.5.6 Intracellular Cd:C ratios of natural phytoplankton assemblages Stations P4 and P26 are located along line P, a well-studied coastal-oceanic transect in the subarctic Pacific, a HNLC region (Harrison 2002). It is well established that P4 and P26 are Fe-replete and Fe-limited stations, respectively, with a 10-year average of dissolved surface Fe of 0.17 ±0.09 nmol L"1 at P4 and 0.026±0.009 nmol L"1 at P26 (Johnson, K., pers. comm.). Combined data (nitrate, phosphate, and chlorophyll) from the September 2006 cruise also suggest that P4 and P26 were Fe-replete and Fe-limited, respectively. The Fe quota determinations in this cruise along a transect from P4 to P26 also decreased appreciably (Semeniuk, D., pers. comm.), providing further evidence that the phytoplankton community was considerably more Fe-deprived at P26 47 than at P4. The range of Cd quotas measured at P4 and P26 (0.51 to 2.28 pmol Cd mol C"1) are within the range of values measured in the laboratory (0.12 to 12.lu.mol Cd mol C"1). The average Cd:C ratio for the two size classes were 0.57 and 2.28 pmol Cd mol C"1 at P4 and P26, respectively. These values are lower than the average Cd:C ratios of cultured phytoplankton (2.46 and 4.68 pmol Cd mol C"1 for Fe-replete and limited cultures, respectively) and may be explained by the large proportion of dinoflagellates in these waters. Though dinoflagellates were not investigated in this study, Ho et al. (2003) measured a Cd:C ratio of -0.98 pmol Cd mol C"1 for an oceanic dinoflagellate. This value is considerably lower than the Cd:C ratios measured for oceanic diatoms and coccolithophorids in our study. The high proportion of dinoflagellates may thereby lower the overall Cd:C ratios at these two stations. Converting the average Cd:C ratios to Cd:P043" ratios yields values of 0.06 and 0.241 nmol Cd pmol"1 P0 4 3 \ for P4 and P26, respectively. The lowest Cd:P043~ ratio determined from our global dataset of mainly oceanic stations was 0.091 nmol Cd pmol"1 P043". Coastal phytoplankton have been reported, here and in previous studies, to have lower Cd quotas than oceanic phytoplankton (Sunda and Huntsman 2000; Ho et al. 2003). Thus, the fact that P4 is a coastal station may explain its low Cd:P043" ratio relative to the global dataset. In contrast, the Cd:P043" ratio at P26 falls into the lower end of the Fe-limited values calculated in our global dataset. A significant increase in Cd quotas was found in both the 2-20pm and >20pm size fraction from P4 to P26. This result may be explained by both the change in species composition and Fe concentration between the two stations. We believe that total 48 biomass was overestimated at P26, however the species composition was consistent. Biovolume was not measured, however taken from biovolume estimates determined by Haigh (1992) for same phytoplankton species. The phytoplankton used by Haigh (1992) to determine biovolume were collected from Jervis Inlet, a fjord located on the British Columbia coast. These phytoplankton are not Fe-limited and therefore are expected to be larger than the Fe-limited phytoplankton at P26. The 2-20 pm size fraction at both P4 and P26 was composed of similar proportions of naked flagellates, whereas the remaining fraction consisted of mostly dinoflagellates at P4, and approximately equal proportions of diatoms, coccolithophores, and dinoflagellates at P26. Given the relative differences in Cd:C ratios of diatoms, coccolithophores, and dinoflagellates (listed in decreasing order) in this study and the work of Ho et al. (2003), species composition may have driven some of the increase in the Cd:C quotas observed in 2-20 pm size fraction at P26. In contrast, the change in phytoplankton species composition in the > 20 pm size class does not support the higher Cd quotas found at P26. Dinoflagellates accounted for 70% and 99% of the biomass at P4 and P26, respectively. The remaining -30% of the biomass at P4 consisted of oceanic diatoms. Oceanic diatoms have the highest Cd quotas in this study, and if anything would have increased the Cd quota at P4. Therefore, in the > 20 pm size fraction, Fe-limitation appears to be the main driving force controlling high phytoplankton Cd:C ratios at P26 relative to P4. In order to determine if export production of Fe-limited phytoplankton could provide an explanation for low dissolved Cd:P043" ratios in HNLC waters, Cullen (2006) estimated a late summer surface water dissolved Cd:P043" of -0.3 nmol Cd pmol"1 P043" for Station PAPA (P26). For this calculation, Cullen assumed a Cd:P043" ratio of 1.3 49 nmol Cd pmol"1 P043" for Fe-limited phytoplankton at Station PAPA. The Cd:P04 3' ratio that we measured at this station in September 2006 (0.241 nmol Cd pmol"1 P043") indicates that the value used by Cullen might have been an overestimate. Cullen (2006) obtained this phytoplankton Cd:P043" ratio from the Southern Ocean, where the highest Cd:P04 " ratios were observed in our global dataset. The only phytoplankton culture having such high Cd:P043" ratio was the Fe-limited Southern Ocean diatom (1.28 nmol Cd pmol'1 P043"). This suggests that phytoplankton in the Southern Ocean have the highest Cd:P043" ratios and that extrapolating Southern Ocean particulate Cd:P043" ratios to other FfNLC regions may not be adequate. However, the dissolved surface Cd concentration measured at Station PAPA in September 2006 (0.20 nmol L"1) matched closely to that estimated by Cullen (0.25 nmol L"1). The resulting dissolved Cd:P043" ratio (0.18 nmol pmol"1) in September 2006 was even lower than the dissolved Cd:P043" ratio estimated by Cullen (2006), assuming the export of phytoplankton with extremely high Cd:P04 " ratios. The low dissolved Cd:P04 " ratios measured at P26 may be the repercussion of an early bloom of phytoplankton with fundamentally high Cd:P043" ratios, such as large pennate diatoms. Large pennate diatoms are commom bloom formers in the subarctic Pacific (Marchetti et al. 2006) and they have the potential to have 3 • * high Cd:P04 " ratios. This suggests that Cullen's estimates may hold, however it appers that the export of diatoms with high Cd:P043" ratios are the main driving force in lowering dissolved Cd:P043" ratios. It is plausible that the export of diatoms with high Cd:P043" ratios in Southern Ocean and perhaps in other HNLC waters may lead to the low dissolved Cd:P04 " ratios measured in surface waters of these regions. 50 2.5.7 Importance of the Southern Ocean Analysis of our global data set suggests that the trend in high particulate Cd:P043" ratios in HNLC regions is dominated by the high values found in the Southern Ocean (Figure 2.2). The Southern Ocean HNLC and subarctic Pacific HNLC stations are also shown to be statistically unique from one another, however not from other HNLC stations. The difference between the Southern Ocean and the subarctic Pacific may be biased because the Southern Ocean with respect to Cd and PO4 3" chemistry has been more extensively studied than the subarctic Pacific. The full range in particulate Cd:P043" ratios may have been missed during sampling periods in the subarctic Pacific. Alternatively, this may be a true reflection of geographical differences that exist between the Southern Ocean and the subarctic Pacific. The average estimated surface particulate Cd:P043" ratio (0.6 nmol Cd pmol"1 P043') in our global dataset for only Southern Ocean stations is remarkably similar to the mean Cd:P043" ratio (0.59 nmol Cd pmol"1 PO43") measured in culture for the Southern Ocean isolates (a Southern Ocean diatom and Phaeocystis). This is not surprising as diatoms and prymesiophytes (mostly Phaeocystis antarctica) dominate Southern Ocean waters in approximately equal proportion. However, both groups show spatial variations and a seasonal progression within water masses (Marchant 1993; Marchant and Thomsen 1994; Arrigo et al. 1999). Diatom blooms have mainly been observed in frontal regions, such as the Antarctic Polar front (APF) (Sakshaug et al. 1991, Laubscher et al. 1993, Smetacek et al. 1997), and during summer in certain regions of the Ross Sea (Arrigo et al. 1999). Phaeocystis antarctica blooms have only been observed south of the Antarctic divergence; and in some regions of the Ross Sea during the spring (DiTullio and Smith 51 1996, Arrigo et al. 1999). However, the Southern Ocean diatom had a higher Cd:C ratio than the Southern Ocean Phaeocystis. Thus, species composition in the Southern Ocean may affect Cd:P043" ratios in the region. Diatoms are believed to be responsible for much of the carbon export from surface to deep waters (Armstrong et al. 2002). Recently, it has also been shown that Phaeocystis antarctica blooms in the Southern Ocean can also be rapidly exported from surface waters (DiTullio et al. 2000). The two extremes in the Cd:P04 " ratios in our global dataset may therefore be explained by the export of blooms of either species. The highest Cd:P043" ratio in our cultures was for the diatom Proboscia inermis (1.28 nmol Cd pmol"1 PO4 3"), which is within range of the maximum particulate Cd.P043" ratios from our global dataset (Table 2.4 and Figure 2.2). Interestingly, the highest estimated particulate Cd:P04 " ratios from our global dataset were from dissolved Cd and PO43" data in the APF region during a diatom dominated spring bloom (Ldscher et al. , 1998) (Table 2.4). In the APF region where primary production was the highest, dissolved surface Cd:P043" ratios were less than half the ratio in the southern Antarctic Circumpolar Current area, which was characterized by low biological productivity (Ldscher et al. 1998). Therefore, the export of diatoms with high Cd:P043" ratios, as seen in culture, would have resulted in low dissolved Cd:P043' ratios in surface waters and the high dissolved Cd:P043" ratios in the nutricline at these stations. 2.6 Oceanographic Implications Regardless of the phyla, Fe-limitation throughout most of the year in HNLC regions would lead to high Cd:P043" ratios in resident phytoplankton communities. 52 However, only phytoplankton that are exported out of the mixed layer would leave an imprint on dissolved Cd:P043~ ratios. As large high-Cd diatoms account for a disproportionate fraction of export production (Goldman 1993; Dugdale and Wilkerson 1998), dissolved Cd would be removed preferentially to PO4 " in HNLC surface waters. The sinking and remineralization of these diatoms would lead to enhanced dissolved Cd:P043" ratios in the nutricline. In all HNLC regions, Fe fertilization has always led to growth enhancement (de Baar et al. 1990; Martin et al. 1990; Timmermans et al. 1998), if not blooms of large chain forming diatoms (de Baar and Boyd 2000). The termination of these diatoms blooms can be partly attributed to Fe-limitation (Boyd et al. 2000; Boyd et al. 2005). The sedimentation of Fe-limited diatoms with high signature Cd:P043" ratios at the end of a bloom may further enhance the discrepency of dissolved Cd:PC>43" ratios between surface and nurticline waters in HNLC regions. The export of Fe-limited diatoms with high intracellular Cd content may therefore explain pronounced regional .differences in Cd:PC>4 " cycling indicated by the kink in the global dissolved Cd:P04 " relationship. The results of this study indicate that both Fe-limitation and species composition fundamentally change the dissolved Cd:P043" ratio in surface waters and nutriclines. The Cd preserved in planktonic foraminifera as a paleonutrient proxy for surface water PO4 3" concentrations should be used with caution until changes in Cd:P043~ ratios due to species composition and Fe-limitation are fully accounted for. 53 2.7 Figures 3.0 2.5 L_ 2.0 o B 3 L ( H 0.5 H warn 2 u M paaatsi 20 u.M 0.0 P26 Station Figure 2.1. Particulate Cd:C (/txnol mol"1) ratios of natural phytoplankton assemblages in two-size fractions (2-20 pm and > 20 pm) from an Fe-replete coastal station (P4) and an Fe-limited oceanic station (P26) in the subarctic Pacific. Samples were collected in September 2006 from the CCGS John P. Tully. Error bars represent the mean error of replicate bottles at each station. 54 o a i a U 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 74 o o o • Fe-limited O Fe-replete 8 0 0 ^ 0 % ) o o ^ |fe o 0 o o ^8 u~1 1 1 1 1 1 1 1— •80 -60 -40 -20 0 20 40 60 Latitude —1— 80 Figure 2.2. A compilation of calculated surface particulate CdiPCV" (nmol pmol"1) ratios based on dissolved Cd and PO43" profiles from oceanic nutriclines plotted against latitude (Table 2.4). 55 Table 2.1. Species used in this study along with strain designations, size range (diameter), source locations, and provenance. Species Strain Size (um) Source Location Environment Thalassiosira oceanica CCMP 1003 4.9-7.7 36°18N69°58W Sargasso Sea Oceanic Thalassiosira pseudonana (3H) NEPCC 58 3.9-4.9 40°75 N 72°82 W Moriches Bay, NY Coastal Emiliania huxleyi NEPCC 732 5.9-14.0 50° N145° W subarctic Pacific Oceanic Phaeocystis pouchetti NEPCC 225 2.6-5.6 49°52 N 142°40 W subarctic Pacific Oceanic Chrysochromulina polylepis NEPCC 242 2.6-5.6 49°36 N 140°37 W subarctic Pacific Oceanic Phaeocystis antarctica SX-9 2.2-4.3 65°09 S 174°09 E Southern Ocean Oceanic Proboscia inermis (alata) Prob -270 fl) x 45 fw)a 57°51 S 139°51 E Southern Ocean Oceanic a Mean length and width of 20 cells. Table 2.2. Mean absolute growth rates, relative growth rates (um a x shown in bold), and Fe:C ratios for species used in this study. Cultures were grown in Fe-replete (pFe 19 or 20.74) and low-Fe (pFe 20.5, 21, 21.5, 4nM Fe: 40nM DFB or 4nM Fe: 400 nM DFB) conditions in the absence or presence of 35nmol L"1 Cd(pCd 12). Errors represent ± standard error associated with the mean of triplicate measurements. Species pFe Fe:DFBa Growth Rateb dd"' U/Hmax Fe:C (pmol mol"1) Fe-use efficiency*1 (xlO5 mol C mol Fe"1 d"1) T. pseudonana 19 2.46 0.99 109±2 0.16±0.004 19 Cd 2.48 1.00 107±3 0.17±0.008 20.5 1.65 0.67 19.3±0.4 0.59±0.07 20.5_Cd 1.66 0.67 19.9±0.7 0.57±0.07 T. oceanica 19 1.68 1.00 88.8±2 0.15±0.008 19 Cd 1.68 1.00 85.5±1 0.16±0.002 21 1.33 0.79 7.47±0.8 1.1±0.03 21 Cd 1.33 0.79 9.30±0.3 1.1±0.07 21.5 0.91 0.54 3.06±0.09 2.0±0.09 21.5_Cd 0.90 0.54 3.20±0.1 2.0±0.09 E. huxleyi 19 0.90 1.00 53.4±2 0.12±0.006 19 Cd 0.90 1.00 58.5±2 0.11±0.006 21 0.49 0.54 12.3±0.6 0.28±0.02 21_Cd 0.51 0.57 13.2±1 0.26±0.01 P. pouchetti 19 1.37 1.00 58.9±4 0.16±0.01 19 Cd' 1.36 0.99 60.1±3 0.16±0.01 21.5 1.07 0.78 5.77±0.1 1.2±0.07 21.5_Cd 1.08 0.79 5.25±0.07 1.3±0.03 C. polylepis 19 1.35 0.98 52.6±1 0.18±0.008 19 Cd 1.38 1.00 56.5±2 0.17±0.01 21.5 0.99 0.71 6.16±0.4 1.0±0.09 21.5_Cd 0.95 0.69 6.18±0.2 0.98±0.07 P. antarctica 20.74 0.43 0.84 3.34±0.08 1.0±0.02 20.74 Cd 0.51 1.00 5.53±0.3 0.64±0.02 4:400a 0.26 0.50 0.610±0.02 3.6±0.2 4:400_Cda 0.31 0.61 0.845±0.03 2.4±0.02 Proboscia inermis 20.74 0.49 0.96 2.70±0.03 1.3±0.07 20.74 Cd 0.51 1.00 2.86±0.4 1.2±0.05 4:40a 0.34 0.67 0.655±0.03 3.6±0.2 4:40 Cd a 0.37 0.73 0.787±0.05 3.5±0.3 a Represents Fe:DFB ratio in nmol nmol"1 and includes Fe contamination in Aquil. Error associated with growth rate is <10% and represents the average of a least 6 measurements 0 Mean Fe-use efficiencies listed as the average of independent Fe-use efficiencies calculated using the individual Fe:C ratios and corresponding growth rates. 57 Table 2.3. Cd:C ratios, calculated Cd:P043" ratios and steady state Cd uptake rates for species used in this study. Cultures were grown in Fe-replete (pFe 19 or 20.74) and low-Fe (pFe 20.5, 21, 21.5, 4nM Fe: 40nM DFB or 4nM Fe: 400 nM DFB) conditions presence of 35nmol L"1 Cd (pCd 12). Errors represent ± standard error associated with the mean of triplicate measurements Species pFe Fe:DFBa Cd:C (pmol mol'1) Cd:P043" (nmol pmol"1) Cd uptakeb (pmol Cd mol C"1 d"1) T. pseudonana 19 0.48±0.004 0.051±0.0003 0.82±0.005 20.5 0.85±0.03 0.090±0.004 . 0.98±0.04 T. oceanica 19 5.1±0.3 0.54±0.03 6.1±0.1 21 6.0±0.09 0.64±0.009 5.8±0.08 21.5 10±0.2 1.1±0.02 7.0±0.1 E. huxleyi 19 3.1±0.2 0.32±0.003 1.9±0.02 21 5.8±0.1 0.61±0.0053 2.1±0.001 P. pouchetti 19 0.12±0.007 0.013±0.00004 0.11±0.0002 21.5 0.22±0.02 0.023±0.002 0.18±0.01 C. polylepis 19 0.15±0.01 0.016±0.001 0.13±0.001 21.5 0.28±0.04 0.030±0.004 0.20±0.005 P. antarctica 20.74 0.40±0.05 0.042±0.005 0.15±0.006 4:400a 1.9±0.3 0.20±0.01 0.39±0.03 P. inermis 20.74 7.9±0.6 0.84±0.07 2.5±0.1 4:40a 12±0.3 1.28±0.03 3.2±0.005 a Represents Fe:DFB ratio in nmol nmol" and includes Fe contamination in Aquil. b Mean Cd uptake rates listed as the average of independent Cd uptake rates calculated using the individual Cd:C ratios and corresponding growth rates. 58 Table 2.4. Regression slopes of Cd:P04 " (nmol pmol"1) based on depth-dependent variations in dissolved Cd and P043" in oceanic nutnclmes. Fe-hmited stations are shown in bold italics^ Statidris; were considered Fe-limited when surface P043" concentrations were > 0.3 /xmol L , when surface P0 4 concentrations were not available classification was based on the Fe-regime in the general area of the station. e Reference Abe (2001) Abe (2002) Abe (2002b) Boyle etal. (1976) Boyle etal. (1984) Bruland (1980) Bruland & Franks (1983) Chen et al. (2005) Cullen (2006) Lat Long Surface P0 4 pmol L"1 Depth Cd:P043" n r 2 Location (m) nmol pmol I 00°00 N 147°00 E 0.079 90-350 0.240 6 0.960 Equatorial Pacific 00°00 N 178°00E 0.539 99-399 0.274 5 0.979 Equatorial Pacific 00°00 N 161°00E 0.213 99-398 0.288 8 0.952 Equatorial Pacific 44°00 N 155°00E 0.84 20-298 0.365 12 0.988 NW Pacific 45°25 N 145°05 E 79-794 0.441 12 0.911 NW Pacific 25°40 N 125°15E 350-900 0.351 8 0.987 NW Pacific 23°00 N 126°20 E 350-700 0.384 10 0.988 NW Pacific 25°12 N 123°41 E 250-700 0.385 12 0.983 NW Pacific 30°34 N 170°36E 0.02 82-687 0.292 7 0.734 NW Pacific 53°60 N 177°17 W 1.6 5-349 0.406 3 0.998 NW Pacific 52°40 S 178°50 W 0.72 216-1290 0.775 6 0.901 SW Pacific 27°54 N 86°52 W 0.05 100-300 0.100 4 0.928 Gulf of Mexico 26°54 N 91°24 W 0.02 90-200 0.167 5 0.945 Gulf of Mexico 29°43 N 88°40 W 0.02 100-200 0.183 4 0.957 Gulf of Mexico 25°59 N 86°22 W 0.02 100-400 0.187 4 0.976 Gulf of Mexico 27°31 N 91°19 W 0.02 100-200 0.204 3 0.988 Gulf of Mexico 24°10N 84°49 W 0.07 100-750 0.257 6 0.991 Gulf of Mexico 32°41 N 144°60 W 0.06 75-595 0.314 4 0.990 NE Pacific 36°52 N 122°53 W 0.6 130-500 0.367 3 0.975 NE Pacific 37°00 N 124°12 W 0.47 75-490 0.409 4 0.994 NE Pacific 34°06 N 66°07 W 0.03 375-715 0.236 3 0.970 NW Atlantic 18°00N 115°50E 150-400 0.259 4 0.977 South China Sea 15°70N 116°70E 150-800 0.352 6 0.935 South China Sea 18°00N 117°70E 0.2 100-500 0.396 6 0.938 South China Sea 22°50 N 119°67E 0.02 100-600 0.421 4 0.991 South China Sea 21°42 N 119°47E 0.64 100-300 0.462 4 0.979 South China Sea 20°25 N 118°58E 0.44 100-500 0.508 5 0.959 South China Sea 56°33 N 171°60 W 0.42 15-50 0273 4 0.978 Bering Sea 55°00 N 178°99 W 1.31 70-300 0.460 4 0.929 Bering Sea Table 2.4. Continued Reference Lat Long Danielsson &Westerlund (1983) 82°53 N 43°97 E Danielsson et al. (1985) 59°00 N 20°00 W 53°00 N 20°00 W - 40°30 N 15°00 W 48°30 N 20°00 W 64°10N 05°40 W 62°30 N 00°30 E 60°47 N 14°00 W Fitzwater et al. (2000) 72°30 S 174°00 W Frew (1995) 62°10 S 83°20 E Frew and Hunter (1992) 48°05 S 164°30 E Frew and Hunter (1995) 46°80 S 167°00E 47°60 S 170°00E 48°20 S 164°50E Hunter and Ho (1991) 37°90 S 166°80E 38°10S 168°01 E 34°30 S 171°42 E 38°45 S 169°18E 34°47 S 170°30E 35°50 S 162°40E 34°38 S 171°26E 35°11 S 167°09E 34°29 S 171°46E 39°14S 170°59E 34°29 S 171°53E 34°29 S 172°02E 34°24 S 172°18E Jones and Murray (1984) 47°00 N 131.50 W 47°00 N 128.00 W 47°00N 127.00 W 47°0ON 125.50 W Knauer and Martin (1981) 36°50 N 123°00 W Surface P0 4 3 Depth Cd:P043 n r 2 Location pmolL'1 (m) nmol pmol'1 0.37 100-1000 0.226 4 0.931 Arctic Ocean 200-1000 0.138 3 0.810 NE Atlantic 100-500 0.171 3 0.993 NE Atlantic 0.1 100-1000 0.187 4 0.978 NE Atlantic 0.08 100-1000 0.252 4 0.997 NE Atlantic 0.53 100-500 0.254 3 0.885 NE Atlantic 0.41 100-400 0.295 3 0.900 NE Atlantic 0.68 100-500 0.672 3 0.998 NE Atlantic 1.83 100-275 0.757 3 0.990 Southern Ocean 1.49 54-155 0.421 2 0.999 Indian Ocean 0.71 100-1300 0.350 18 0.953 Southern Ocean 0.51 100-300 0.161 4 0.981 •Doubtful Sound 200-800 0.330 4 0.981 Foveaux Strait 0.71 75-1590 0.350 22 0.964 -Puyseger Trench 250-1250 0.162 7 0.907 -Tasman Sea 0.41 150-600 0.258 6 0.951 Tasman Sea 0.26 22-520 0.272 8 0.795 "Tasman Sea: 0.44 98-443 0.276 5 0.955 Tasman Sea 0.28 191-1144 0.292 7 0.889 •-Tasman Sea 0.16 100-1395 0.295 16 0.850 Tasman Sea 0.42 100-897 0.324 6 0.965 Tasman Sea 0.31 394-1280 0.441 7 0.889 -Tasman Sea 0.37 49-519 0.441 9 0.935 Tasman Sea 0.26 200-600 0.578 5 0.985 -Tasman Sea 0.71 74-222 0.718 4 0.842 Tasman Sea 0.48 94-282 0.974 5 0.616 Tasman Sea 0.53 50-100 1.50 3 0.923 Tasman Sea 0.67 50-900 0.232 4 0.983 NE Pacific 0.46 50-300 0.252 4 0.960 NE Pacific 0.37 75-300 0.570 4 0.985 1NE Pacific 0.31 50-200 0.697 3 0.960 NE Pacific 0.91 65-350 0.444 3 0.994 NE Pacific Table 2.4. Continued Reference Loscheret al. (1998) Martin et al. (1989) Martin etal. (1990) Martin et al. (1993) Lat Long 55°01 S 06°00 W 57°03 S 06°01 W 55°00 S 06°01 W 47°00 S 06°00 W 49°00 S 06°00 W 49°00 S 06°00 W 56°01 S 06°00 W 57°29 S 06°00 W 55°59 S 06°04 W 53°00 S 06°00 W 48°41 S 05°59 W 48°00 S 06°00 W 48°00 S 06°00 W 52°00 S 06°00 W 58°00 S 06°00 W 51°00S 06°00 W 56°09 S 15°26 W 49°59 S 06°00 W 50°00 S 06°00 W 46°52 S 06°00 W 54°00 S 06°00 W 48°00 S 06°00 W 57°00 S 23°19 W 39°60 N 140*77 W 45°00 N 142°87 W 50°00 N 145°00 W 55°50 N 147°50 W 58°68 N 147°95 W 60°46 S 63°26 W 47°00 N 20°00 W Surface P0 4 3 Depth Cd:P043_ n r 2 Location pmol L"1 (m) nmol nmol"1 1.79 150-30C ) 0.245 3 0.984 Southern Ocean 1.95 95-190 0.324 3 0.927 Southern Ocean 1.89 100-30C ) 0.356 4 0.997 Southern Ocean 1.57 85-255 0.376 4 0.922 Southern Ocean 1.15 100-30C 1 0.420 3 0.966 Southern Ocean 1.22 97-291 0.429 4 0.992 Southern Ocean 1.88 100-200 0.431 3 0.995 Southern Ocean 1.92 100-200 0.441 3 0.881 Southern Ocean 1.84 95-284 0.454 4 0.999 Southern Ocean 1.87 100-200 0.458 3 0.961 Southern Ocean 1.32 97-194 0.492 3 0.998 Southern Ocean 1.62 100-300 0.519 4 0.914 Southern-Ocean 1.65 74-277 0.569 3 0.999 Southern Ocean 1.9 100-300 0.615 4 0.805 "Southern Ocean 1.95 100-300 0.619 3 0.865 Southern Ocean 1.88 100-300 0.622 4 0.998 Southern Oceans 1.91 100-300 0.639 4 0.883 Southern Ocean 1.11 98-294 0.784 3 0.978 Southern Ocean. 1.71 100-300 0.845 4 0.947 S^outhern Ocean -100-200 0.880 3 0.969 S^outhern Ocean 1.9 98-197 1.01 3 0.955 Southern Ocean 1.5 100-200 1.13 3 0.970 Southern Ocean 1.87 100-300 1.87 4 0.833 ;Southern Ocean 0.2 100-690 0.343 9 0.952 iNE Pacific 0.63 20-580 0.367 10 0.934 ^NE Pacific 0.72 : 20-290 0.432 7 0.974 NE Pacific 0.39 : 20-150 0.478 4 0.996 NE Pacific 0.73 : 20-150 0.500 4 0.999 NE Pacific 1.78 ( 50-300 0.694 0.992 Drake Passage 0.05 75-600 0.238 3 0.961 N. Atlantic Table 2.4. Continued Reference Lat Long Morleyet al. (1993) 08°27 S 52°43 E 12°14S 55°02 E 06°09 S 50°53 E 18°36S 55°36 E 27°00 S 56°58 E Nolting & de Baar (1994) 59°22 S 48°44 W 61°60S 48°60 W 57°03 S 48°51 W 58°00 S 48°59 W 59°60 S 48°59 W Noltingetal. (1991) 60°52 S 45°22 W 57°00 S 48°24 W 61°59 S 48°59 W Orren & Monteiro (1985) 51°40S 32°00E Pai & Chen (1994) 21°45 N 124°00E 21°45N 124°00E 16°35 N 119°00E 2T45N 127°00 E 24°50 N 123°10E Saager etal. (1992) 14°30 N 67°00 E 21°16N 63°22 E 22°30 N 60°40 E Saager et al. (1997) 40°33 N 20°08 W 40°30 N 20°03 W 32°59 N 20°00 W 58°30 N 20°30 W 32°58 N 19°58W Sakamoto-Arnold et al. (1987) 36°49 N 73°33 W 37°38 N 73°38 W 37°24 N 73°70 W 37°33 N 73°20 W 38°52 N 71°39 W Surface P0 4 3 Depth Cd:P04 3 n r 2 Location limol L"1 (m) nmol nmol"1 0.24 180-957 0.091 4 0.862 SW Indian Ocean 0.36 230-1000 0.160 7 0.816 SW Indian Ocean 0.07 112-965 0.166 12 0.885 SW Indian Ocean 0.17 115-1000 0.324 8 0.949 SW Indian Ocean 0.27 298-1080 0.332 8 0.994 SW Indian Ocean 1.49 40-80 0.590 3 0.977 Southern Ocean 2.01 100-300 0.613 3 0.648 Southern Ocean 1.6 60-200 0.772 6 0.866 Southern Ocean 1.76 80-120 0.808 3 0.916 Southern Ocean 1.85 30-60 0.934 4 0.737 Southern Ocean 1.82 100-217 0.535 4 0.979 Antarctic Atlantic 1.77 60-200 0.705 7 0.824 Antarctic Atlantic 2.01 60-150 1.19 4 0.926 Antarctic Atlantic 2.81 184-720 0.379 3 0.965 Southern Ocean 254-1002 0.356 6 0.995 Philippine Sea 301-801 0.364 6 0.990 Philippine Sea 101-500 0.364 4 0.997 South China Sea 252-798 0.365 5 0.995- Philippine Sea -202-801 0.393 4 0.992- - East China Sea-0.42 80-200 0.398 4 0.968 Indian Ocean 0.48 100-1200 0.506 7 0.930 Indian Ocean 0.58 130-1200 0.655 8 0.967 Indian Ocean 0.2 20-150 0.094 5 0.999 NE Atlantic 0.03 200-1500 0.125 6 0.866:. NE Atlantic 0.04 150-750 0.211 7 0.945 NE Atlantic 0.52 75-900 0.246 10 0.721 NE Atlantic 0.03 150-800 0.253 5 0.970 NE Atlantic 0.1 302-587 0.136 6 0.937 NW Atlantic 0.05 245-490 0.156 3 0.721 NW Atlantic 0.1 199-369 0.170 4 0.991 NW Atlantic 0.1 83-398 0.179 4 0.969 NW Atlantic 0.25 347-543 0.189 3 0.986 NW Atlantic Table 2.4. Continued Reference Lat Long Surface P0 4 3 Depth Cd:P043- n r2 Location pmol L"1 (m) nmol pmol" I Sakamoto-Arnold et al. (1987) 39°24 N 71°90W 0.1 66-269 0.190 8 0.845 NW Atlantic 37°70 N 73°90 W 0.15 119-243 0.192 3 0.192 NW Atlantic 37°19N 72°44 W 0.1 106-242 0.203 5 0.988 NW Atlantic 37°34 N 74°80 W 103-239 0.232 4 0.943 NW Atlantic Westerlund & Ohman (1991) 72°12 S 16°41 W 2.01 200-800 0.322 5 0.687 Weddell Sea 73°55 S 39°39 W 2.11 100-400 0.541 3 0.999 Weddell Sea 72°19 S 34°25 W 100-300 0.542 3 0.959 Weddell Sea 72°01 S 16°50W 100-600 0.557 4 0.932 Weddell Sea 76°10 S 31°46 W 100-300 0.720 3 0.999 Weddell Sea Yeats (1988) 64°00 S 59°39 W 200-400 0.813 3 0.912 Weddell Sea 32°00 N 64°50 W 0.06 95-1000 0.276 7 0.905 Sargasso Sea Yeats & Campbell (1983) 35°20 N 62°30 W 0.34 200-1200 0.402 4 0.965 Sargasso Sea 45°00 N 45°00 W 0.36 101-507 0.293 4 0.994 NW Atlantic 53°00 N 41°00 W 0.86 47-191 0.380 3 0.997 NW Atlantic 50°15 N 35°30 W 0.74 100-507 0.415 3 0.711 NW Atlantic 2.9 References Abe, K. 2001. Cd in the western equatorial Pacific. Mar. Chem. 74: 197-211. Abe, K. 2002a. Preformed Cd and P04 and the relationship between the two elements in the northwestern Pacific and the Okhotsk Sea. Mar. Chem. 79: 27-36. Abe, K. 2002b. Relationship between Cd and P04 in the subtropical sea near the Ryukyu Islands. J. Oceanogr. 58: 577-588. Armstrong, R. A., C. Lee, J. I. Hedges, S. Honjo, and S. G. Wakeham. 2002. A new, mechanistic model for organic carbon fluxes in the ocean based on the quantitative association of POC with ballast minerals. Deep Sea Res. II49: 219-236. Arrigo, K. R., D. H. Robinson, D. L. Worthen, R. B. Dunbar, G. R. DiTuillo, M. VanWoert, and M. P. Lizotte. 1999. Phytoplankton community structure and the drawdown of nutrients and CO2 in the Southern Ocean. Science 283: 365-368. Barwell-Clarke, J., and F. Whitney. 1996. Institute of Ocean Sciences Nutrient Methods and Analysis. Can. Tech. Rep. Hydrogr. Ocean Sci. 182: 43 pp. Boyd, P. W., and others. 2005. The evolution and termination of an iron-induced mesoscale bloom in the northeast subarctic Pacific. Limnol. Oceanogr. 50: 1872-1886. Boyd, P. W., and others. 2000. A mesoscale phytoplankton bloom in the polar Southern Ocean stimulated by iron fertilization. Nature 407: 695-702. Boyle, E. A. 1992. Oceanic chemical distributions during the stage 2 glacial maximum: Cadmium and 5 1 3C evidence compared. Annual Review of Earth and Planetary Science Letters 20: 245-287. Boyle, E. A., D. F. Reid, S. S. Huested, and J. Hering. 1984. Trace metals and radium in the Gulf of Mexico: An evaluation of river and continental shelf sources. Earth Planet. Sci. Lett. 69: 69-87. Boyle, E. A., F. Sclater, and J. M. Edmond. 1976. On the marine geochemistry of cadmium. Nature 263: 42-44. Brand, L. E. 1991. Minimum iron requirements of marine phytoplankton and the implications for the biogeochemical control of new production. Limnol. Oceanogr. 36: 1756-1771. Broecker, W. S., and T. H. Peng. 1982. Tracers in the Sea. LDGO, New York, 690 pp. Bruland, K. W. 1980. Oceanographic distributions of cadmium, zinc, nickel, and copper in the North Pacific. Earth and Planetary Science Letters 47: 176-198. 64 Bruland, K. W. 1992. Complexation of cadmium by natural organic ligands in the central North Pacific. Oceanography 37: 1008-1017. Bruland, K. W., and R. P. Franks. 1983. Manganese, nickel, copper, zinc and cadmium in the western north Atlantic. Trace Metals in Seawater. NATO Conference Series IV; Marine Sciences: 394-414. Bruland, K. W., G. A. Knauer, and J. H. Martin. 1978. Cadmium in northeast Pacific waters. Limnol. Oceanogr 23: 618-625. Brzezinski, M.A., D. M. Nelson, V. M. Franck, and D. E. Sigmon. 2001. Silicon dynamics within an intense open-oceandiatom bloom in the Pacific sector of the Southern Ocean. Deep-Sea Res. II49: 3997-4018. Chen, H. Y., T. H. Fang, and L. S. Wen. 2005. A preliminary study of the distributions of Cd in the South China Sea. Cont. Shelf Res. 25: 297-310. Croot, P. L., A. R. Bowie, R. D. Frew, M. Maldonado, J. A. Hall, K. A. Safi, J. La Roche, P. W. Boyd, and C. S. Law. 2001. Retention of dissolved iron and Fe" in an iron induced Southern Ocean phytoplankton bloom. Geophys. Res. Lett. 28: 3425-3428. Cullen, J. T. 2006. On the nonlinear relationship between dissolved cadmium and phosphate in the modern global ocean: Could chronic iron limitation of phytoplankton growth cause the kink? Limnol. Oceanogr. 51: 1369-1380. Cullen, J. T., Z. Chase, K. H. Coale, S. Fitzwater, and R. M. Sherrell. 2003. Effect of iron limitation on the cadmium to phosphorus ratios of natural phytoplankton assemblages from the Southern Ocean. Limnol. Oceanogr. 48: 1079-1087. Cullen, J. T., T. W. Lane, F. M. M. Morel, and R. M. Sherrell. 1999. Modulation of cadmium uptake in phytoplankton by seawater C02 concentration. Nature 402: 165-167. Cullen, J. T., and R. M. Sherrell. 2005. Effects of dissolved carbon dioxide, zinc, and manganese on the cadmium to phosphorus ratio in natural phytoplankton assemblages. Limnol. Oceanogr. 50: 1193-1204. Danielsson, L., B. Magnusson, and S. Westerlund. 1985. Cadmium, copper, iron, nickel and zinc in the north-east Atlantic Ocean. Mar. Chem. 17: 23-41. Danielsson, L., and S. Westerlund. 1983. Trace metals in the Arctic Ocean. In C. S. Wong, E. D. Goldberg, K. W. Bruland, J. D. Burton, and E. A. Boyle (eds.), Trace Metals in Seawater, Plenum Press: 85-96. 65 de Baar, H. J. W., and P. W. Boyd. 2000. The role of iron in plankton ecology and carbon dioxide transfer of the global oceans. In Hanson, R. B., H. W. Ducklow, and J. G. Field (eds.), The Dynamic Ocean Carbon Cycle: A Midterm Synthesis of the Joint Global Ocean Flux Study, Cambridge University Press, Cambridge: 61-140. de Baar, H. J. W., A. G. J. Buma, R. F. Nolting, G. C. Cadee, G. Jacques, and P. J. Treguer. 1990. On iron limitation of the Southern Ocean: experimental observations in the Weddell and Scotia Seas. Mar. Ecol. Prog. Ser. 65: 105-122. de Baar, H. J. W., P. M. Saager, R. F. Notling, and J. Van Der Meer. 1994. Cadmium versus phosphate in the world ocean. Mar. Chem. 46: 261-281. DiTullio, G. R., and others. 2000. Rapid and early export of Phaeocystis antarctica blooms in the Ross Sea, Antarctica. Nature 404: 595-598. DiTullio, G. R., and W. O. Smith. 1996. Spatial patterns in phytoplankton biomass and pigment distributions in the Ross Sea. J. Geophys. Res. 101: 467-477. Dugdale, R. C , and F. P. Wilkerson. 1998. Silicate regulation of new production in the equatorial Pacific upwelling. Nature 391: 270-273 (1998). Elderfield, H , and R. E. M. Rickaby. 2000. Oceanic Cd/P043" ratio and nutrient utilization in the glacial Southern Ocean. Nature 405: 305-310. Fitzwater, S. E., K. S. Johnson, R. M. Gordon, K. H. Coale, and W. O. Smith, Jr. 2000. Trace metal concentrations in the Ross Sea and their relationship with nutrients and growth. Deep-Sea Res. II 47: 3159-3179. Frew, R. D. 1995. Antarctic Bottom Water formation and the global cadmium to phosphorus relationship. Geophysical Research Letters 22: 2349-2352. Frew, R., and K. A. Hunter. 1992. Influence of Southern Ocean waters on the cadmium-phosphate properties of the global ocean. Nature 360: 144-146. Frew, R., and K. A. Hunter. 1995. Cadmium-phosphorus cycling at the subtropical convergence south of New Zealand. Mar. Chem. 51: 223-237. Geider, R. J. and La Roche, J. 2002. Redfield revisited: variability of C:N:P in marine microalgae and its biochemical basis. Eur. J. Phycol. 37: 1-17. Goldman, J. C. 1993. Potential role of large oceanic diatoms in new primary production. Deep-Sea Res. 140: 159-168. Haigh, R., F. J. R. Taylor, and T. F. Sutherland. 1992. Phytoplankton ecology of Jervis Inlet, a fjord system on the British Columbia coast. I. General features of the nano- and microplankton. Marine Ecology Progress Series 89: 117-134. 66 Harrison, P. J. 2002. Station Papa time series: Insights into ecosystem dynamics. Journal of Oceanography 58: 259-264. Ho, T., and others. 2003. The elemental composition of some marine phytoplankton. J. Phycol. 39: 1145-1159. Hudson, R. J. M. and F. M. M. Morel. 1990. Iron transport in marine phytoplankton: Kinetics of cellular and medium coordination reactions. Limnol. Oceanogr. 35: 1002-1020. Hunter, K. A., and F. W. T. Ho. 1991. Phosphorus-cadmium cycling in the northeast Tasman Sea, 3'5-40'S. Mar. Chem. 33: 279-298. Jones, C. J., and J. W. Murray. 1984. Nickel, cadmium, and copper in the northeast Pacific off the coast of Washington. Limnol. Oceanogr. 29: 711-720. Knauer, G. A., and J. H. Martin. 1981. Phosphorus-cadmium cycling in northeast Pacific waters. J. Mar. Res. 39: 65-76. Kustka, A. B., A. E. Allen, F. M. M. Morel. 2007. Sequence analysis and transcriptional regulation of Fe acquisition genes in two marine diatoms. J. Phycol. (In press). Lane, T. W., and F. M. M. Morel. 2000. A biological function for cadmium in marine diatoms. Proc National Academy of Science USA 97: 4627-4631. LaRoche, J., P. W. Boyd, R. M. L. McKay, and R. J. Geider. 1996. Flavodoxin as an in situ marker for iron stress in phytoplankton. Nature 382: 802-805. Laubscher, R. K., Perissinotto, R. & McQuaid, C. D. 1993. Phytoplankton production and biomass at frontal zones in the Atlantic sector of the Southern Ocean. Polar Biol. 13: 471-481. Lee, J. G., and F. M. M. Morel. 1995. Replacement of Zn by cadmium in marine phytoplankton. Mar. Ecol. Progr. Ser. 127: 305-309. Ldscher, B. M., J. T. M. D. Jong, and H. J. W. de Baar. 1998. The distribution and preferential biological uptake of cadmium at 6°W in the Southern Ocean. Mar. Chem. 62: 259-286. Maldonado, M. T., and N. M. Price. 1996. Influence of N substrate on Fe requirements of marine centric diatoms. Mar. Ecol. Prog. Ser. 141: 161-172. Marchant, H. J. 1993. Antarctic marine nanoplankton. Current Topics in Botanical Research, ed. By J. Menon. 1: 189-201. 67 Marchant, H. J., and H. A. Thomsen. 1994. Haptophytes in polar waters. In J.C. Green and B.S.C. Leadbeater (eds), The Haptophyte Algae, Clarendon Press, Oxford: 209-228. Marchetti, A., M. T. Maldonado, E. S. Lane, and P. J. Harrison. 2006. Iron requirements of the Pennate diatom Pseudo-nitzschia: Comparison of oceanic (high-nitrate, low-chlorophyll waters) and coastal species. Limnol. Oceanogr. 51: 2092-2101. Martin, J. H. 1990. Glacial-interglacial C0 2 change: the iron hypothesis. Paleoceanography 5: 1-3. Martin, J. H., R. M. Gordon, and S. Fitzwater. 1990. Iron in Antarctic waters. Nature 345: 156-158. Martin, J. H., S. E. Fitzwater, R. M. Gordon, G. N. Hunter, and S. J. Tanner. 1993. Iron, primary production and carbon-nitrogen flux studies during the JGOFS North Atlantic Bloom Experiment. Deep-Sea Res. II40: 115-134. Martin, J. H., R. M. Gordon, S. Fitzwater, and W. W. Broenkow. 1989. VERTEX: phytoplankton/iron studies in the Gulf of Alaska. Deep-Sea Res. 36: 649-680. Millero, F. J., S. Sotolongo, and M. Izaguirre. 1987. The oxidation-kinetics of Fe(II) in seawater. Geochim. Cosmochim. Acta. 51: 793-801. Morley, N. H., P. J. Statham and J. D. Burton. 1993. Dissolved trace metals in the southwestern Indian Ocean. Mar. Chem. 40: 1043-1062. Nolting, R. F., and H. J. W. de Baar. 1994. Behaviour of nickel, copper, zinc and cadmium in the upper 300 m of a transect in the Southern Ocean (57u-62uS, 49uW). Mar. Chem. 45: 225-242. Nolting, R. F., H. J. W. de Baar, A. J. Van Bennekom, and A. Masson. 1991. Cadmium, copper and iron in the Scotia Sea, Weddell Sea and Weddell/Scotia Confluence (Antarctica). Mar. Chem. 35: 219-243. Orren, M. J, and P. M. S. Monteiro. 1985. Trace element geochemistry in the Southern Ocean. In Siegfried WR, Condy PR, Laws RM (eds), Antarctic nutrient cycles and food webs. Proc 4th SCAR Symp Antarct Biol, Springer, Berlin Heidelberg New York: 30-37. Pai, S. C , and H. Y. Chen. 1994. Vertical distribution of cadmium in marginal seas of the western Pacific Ocean. Mar. Chem. 47: 81-91. Payne, C. D., and N. M. Price. 1999. Effects of cadmium toxicity on growth and elemental composition of marine phytoplankton. J. Phycol. 35: 293-302. 68 Price, N. M., B. A. Ahner, and F. M. M. Morel. 1994. The equatorial Pacific Ocean: grazer-controlled phytoplankton populations in an iron-limited ecosystem. Limnol. Oceanogr. 39: 520-534. Price, N. M., G. I. Harrison, J. G. Hering, R. J. Hudson, P. Nirel, B. Palenik, and F. M. M. Morel. 1988/89. Preparation and chemistry of the artificial algal culture medium Aquil. Biol. Oceanogr. 5: 43-46. Rijkenberg, M. J. A., A. C. Fischer, J. J. Kroon, L. J. A. Gerringa, K. R. Timmermans, H. T. Wolterbeek, and H. J. W. de Baar. 2005. The influence of UV irradiation on the photoreduction of iron in the Southern Ocean. Mar. Chem. 93: 119-129. Saager, P. M. and H. J. W. de Baar. 1993. Limitations to the quantitative application of Cd as a paleoceanographic tracer, based on results of a multi-box model (MENU) and statistical considerations. Global and Planetary Change 8: 69-92. Saager, P. M., H. J. W. de Baar, J. T .M. de Jong, R. F. Nolting, and J. Schijf. 1997. Hydrography and local sources of dissolved Mn, Ni, Cu and Cd in the northeast Atlantic Ocean. Mar. Chem. 57: 195-216. Saager, P. M., H. J. W. de Baar, and R. J. Howland. 1992. Cd, Zn, Ni and Cu in the Indian-Ocean. Deep-Sea Res. A 39: 9-35. Sakamoto-Arnold C. M., A. K. Hanson, D. L. Huizenga, and D. R. Kester. 1987. Spatial and temporal variability of Cd in Gulf Stream warm-core rings and associated waters. J. Mar. Res. 45: 201-230. Sakshaug, E., D. Slagstad, and O. Holm-Hansen. 1991. Factors controlling the development of phytoplankton blooms in the Antarctic Ocean—a mathematical model. Mar. Chem. 35:259-271. Smetacek, V., and others. 1997. Ecology and biogeochemistry of the Antarctic Circumpolar Current during austral spring: A summary of Southern Ocean JGOFS cruise ANTX/6 of R.V. Polarstern. Deep-Sea Res. II44: 1-22. Strzepek, R. F., and N. M. Price. 2000. Influence of irradiance and temperature on the iron content of the marine diatom, Thalassiosira weissflogii (Bacillariophyceae). Mar. Ecol. Prog. Ser. 206: 107-117. Sunda, W. G., and S. A. Huntsman. 1995a. Iron uptake and growth limitation in oceanic and coastal phytoplankton. Mar Chem. 50: 189-206. Sunda, W. G., and S. A. Huntsman. 1996. Antagonisms between cadmium and zinc toxicity and manganese limitation in a coastal diatom. Limnol. Oceanogr. 41: 373-387. 69 Sunda, W. G., and S. A. Huntsman. 1998. Control of Cd concentrations in a coastal diatom by interactions among free ionic Cd, Zn, and Mn in seawater. Environ. Sci. Technol. 32: 2961-2968. Sunda, W. G., and S. A. Huntsman. 2000. Effect of Zn, Mn, and Fe on Cd accumulation in phytoplankton: Implications for oceanic Cd cycling. Limnol. Oceanogr. 45: 1501-1516. Sunda, W. G., D. G. Swift, and S. A. Huntsman. 1991. Low iron requirement for growth in oceanic phytoplankton. Nature 351: 55-57. Timmermans K. R., M. A. van Leeuwe, J. T. M. de Jong, R. M. L. McKay, R. F. Nolting, H. J. Witte, J. van Ooyen, M. J. W. Swagerman, H. Kloosterhuis, and H. J. W. de Baar. 1998. Iron stress in the Pacific region of the Southern Ocean: Evidence from enrichment bioassays. Mar. Ecol. Ser. 166: 27-41. Tortell, P. D., M. T. Maldonado, and N. M. Price. 1996. The role of heterotrophic bacteria in iron-limited ocean ecosystems. Nature 383: 330-332. Westall, J. C , J. L. Zachary, and F. M. M. Morel. 1976. MINEQL: A computer program for the calculation of chemical equilibrium composition of aqueous systems. R. M. Parsons Lab for Water Resources and Environmental Engineering, Department of Civil Engineering, MIT. Westerlund, S., and P. Ohman. 1991. Cadmium, cooper, cobalt, nickel, lead, and zinc in the water column of the Wedell Sea, Antarctica. Geochimica et Cosmochimica Acta 55: 2127-2146. Whitney, F., and H. J. Freeland. 1999. Variability in upper ocean water properties in the NE Pacific Ocean. Deep-Sea Res. II46: 2351-2370. Yeats, P. A. 1988. The distribution of trace metals in ocean waters. The Science of the Total Environment 72: 131-149. Yeats, P. A., and J. A. Campbell. 1983. Nickel, Copper, Cadmium and Zinc in the Northwest Atlantic Ocean. Mar. Chem. 12: 43-58. Yocis, B. H., D. J. Kieber, and K. Mopper. 2000. Photochemical production of hydrogen peroxide in Antarctic waters. Deep-Sea Res. 47: 1077-1099. 70 Chapter 3: The interaction between Inorganic Fe(II), Fe(III) and Cd uptake in the marine diatom Thalassiosira oceanica2. 3.1 Abstract This study investigates the interactions between the cellular uptake of inorganic Fe(II), Fe(III) and Cd by the oceanic diatom Thalassiosira oceanica. Inorganic Fe(II) uptake rates of Fe-limited cultures were 15 times faster than Fe-replete cultures. In contrast, the rates of inorganic Fe(III) uptake were up-regulated only 5 fold under Fe-limitation. The addition of Cd to the uptake media resulted in a 50% reduction of inorganic Fe(II) uptake rates by Fe-limited T. oceanica compared to the Cd-free medium. Similarly, Cd uptake was inhibited 36% in the presence of Fe(II). In contrast to Fe(II), no interaction between Cd and inorganic Fe(HI) transport was observed. These findings suggest that T. oceanica has separate transporters for inorganic Fe(HI) and Fe(II). Moreover, Cd and Fe(II) appear to enter the cell through a common, non-specific divalent metal transporter that is up-regulated under Fe-deficiency. This non-specific transporter may be a putative NRAMP trans-membrane protein. The uptake of both Cd and Fe(II) via a non-specific divalent metal transporter under low Fe conditions provides a mechanistic explanation for higher phytoplankton Cd quotas observed under Fe-limitation in both field and laboratory studies. 2 A version of this chapter will be submitted for publication. Lane, E. S., K. Jang, and M. T. Maldonado (2007) The interaction between Inorganic Fe(II), Fe(III) and Cd uptake in the marine diatom Thalassiosira oceanica. Limnology and Oceanography. 71 3.2 Introduction The seawater distributions of dissolved cadmium (Cd) and phosphate (PO4 3 ) throughout the water column are closely correlated (Boyle et al. 1976; Bruland et al. 1978; de Baar et al. 1994). This close correlation between Cd and PO4 3" distribution in the ocean suggests that dissolved Cd, like PO43", is controlled by its removal from surface waters by phytoplankton and subsequent remineralization of sinking organic matter at depth. Based on the assumption that the relationship between Cd and PO4 3" remains constant both spatially and temporally in the global ocean, Cd preserved in fossil benthic foraminifera tests have been used to reconstruct past oceanic PO4 3" concentrations in deep waters (Boyle 1988). More recently, this approach has been applied to surface waters and has yielded new insights into nutrient utilization and primary productivity in the glacial Southern Ocean (Elderfield and Rickaby 2000). One of the main caveats in expanding the use of the Cd:Ca paleonutrient proxy to surface waters is derived from the heterogeneous distribution of modern surface Cd:P043" relationship. Several factors have been identified which affect surface Cd:P043" variability. Laboratory studies have demonstrated that the uptake of Cd by phytoplankton is directly related to ambient Cd concentrations and inversely related to Zn and Mn free ion concentrations, likely because of competitive inhibition at cellular uptake sites (Sunda and Huntsman 1996, 1998, 2000). Under low CO2 concentrations phytoplankton cellular Cd content increased, and was attributed to the role Cd plays in carbonic anhydrase (Cullen et al. 1999). Phytoplankton Cd:P043" ratios in Fe-limited regions are higher than those of Fe sufficient areas (Martin et al. 1989; 1990). In support of this, laboratory studies have demonstrated that Fe-limited T. oceanica (Sunda and Huntsman 2000), as well as other phytoplankton 72 taxa (Lane et al., Chapter 2) have significantly elevated cellular Cd content relative to Fe-replete cultures. Cullen et al. (2003) performed shipboard incubation studies with natural assemblages of Fe-limited phytoplankton from the Southern Ocean and found that particulate Cd:P043" ratios decreased significantly with Fe supplementation. Recent work in the Subarctic Pacific Ocean has confirmed these findings, reporting that Cd quotas of natural phytoplankton assemblages are greater at the classic Fe-limited station P26, than at a Fe-replete coastal station (Lane et al., Chapter 2). While observations of enhanced Cd accumulation in Fe-limited phytoplankton appear robust, the physiological mechanism underlying this interaction is not understood. Given this empirical link between Fe-limitation and high intracellular Cd, we hypothesized that Cd and Fe enter the cell via a non-specific divalent cation transporter that is up-regulated under Fe-limitation. One such group of transporters is the NRAMP superfamily, which are widely distributed among both prokaryotic and eukaryotic organisms (Cellier et al. 1995). These transporters are known to be up-regulated under Fe-limitation and are capable of transporting multiple divalent metals (Cd, Mn, Zn, Co) (Gunshin et al. 1997; Thomine et al. 2000; Rosakis and Koster 2005; Agranoff et al. 2005). Interestingly, in higher plants, Fe deficiency leads to a hyper-accumulation of Cd via NRAMP transporters (Thomine et al. 2000). A putative gene for a divalent metal transporter belonging to the NRAMP superfamily (7/»NRAMP) has recently been identified in the marine diatom Thalassiosira pseudonana genome (Kustka et al. 2007). The expression of this gene was found to be dramatically up-regulated under Fe-limitation (Kustka et al. 2007). We thus hypothesized that a TpNRAMP-like gene encodes for a non-specific divalent transporter that can directly access Fe(II) as well as 73 Cd in T. oceanica. The main goal of this study was to investigate the relationship between Fe nutrition and Cd uptake. We therefore determined the physiological mechanism of Cd acquisition in Fe-limited marine diatoms, using T. oceanica as a model diatom. Inorganic Fe(TJ), Fe(HI) and Cd uptake rates were determined for Fe-limited and Fe-sufficient T. oceanica and the antagonistic interaction between Fe and Cd during transport was investigated. 3.3 Methods 3.3.1 Culturing Thalassiosira oceanica (clone 1003), a small 6/rni diameter centric diatom isolated from the Sargasso Sea, was obtained from the Center for Culture of Marine Phytoplankton (CCMP) (Bigelow Laboratory for Ocean Sciences, West Boothbay Harbor, ME). T. oceanica was grown in artificial seawater medium Aquil (Price et al. 1988/1989) at ~19°C under a continuous photon flux density of 175 /xmol quanta m 2 s"1. Aquil medium was prepared using chelexed synthetic ocean water (SOW) at pH 8.2, enriched with standard additions of nitrate (300 /miol L"1 NO3"), phosphate (10 /miol L"1 PO43"), silicic acid (100 /miol L"1 Si032"), and vitamins (biotin, thiamine and Bi2). The trace elements Cu, Mn, Zn, Co, Mo, and Se were added bound to 100 /nmol L"1 ethylene-diamine-tetra-acetate (EDTA), such that Cu, Mn, Zn, and Co were present at free-ion concentrations of 10"1319, 10"827,10"1088, 10"10 8 8 mol L"1, respectively. Total Mo and Se concentrations were 10"7 and 10~8 mol L"1, respectively. Premixed Fe-EDTA (1:1.05) was added separately to achieve a total Fe concentration of 1.37 /imol L"1 (pFe 19; pFe = -74 log[Fe+3]) or 4.2 nmol L"1 (pFe 21.5) for the Fe-sufficient or Fe-limiting media, respectively (speciation calculated using MINEQL) (Westall et al. 1976). Cadmium was added as CdCL., at a total concentration of 35 nmol L"1 in order to mimic in situ oceanic pCd levels in typical HNLC waters. Cullen (2006), measured a total Cd concentration ([Cd]T) of 0.25 nmol L"1 at a typical HNLC surface station in the Bering Sea. About 70% of [Cd]j is present as organic complexes leaving 30% as inorganic species ([Cd']) (Bruland 1992). Approximately 3% of the inorganic Cd species in seawater is present as free Cd 2 + (Byrne et al. 1988). Therefore, a surface [Cd2+] of ~ 2.2 pmol L"1 was present at Cullen's (2006) HNLC station, which corresponds to a pCd = 11.7. Sterile and trace metal-clean techniques were used during all manipulations, and the media was allowed to equilibrate chemically overnight before use. Culture growth was monitored by in vivo fluorescence using a Turner 10-AU Fluorometer and log2 of fluorescence versus time was used to calculate growth rates in doublings per day (dd"1). Growth rates are expressed as the specific growth rate measured divided by the maximum growth rate achieved by T. oceanica during experiments (pt/ju.max)- Cultures were considered acclimated when growth rates in successive transfers varied by less than 10%. 3.3.2 Short-term inorganic Fe and Cd uptake rates Short-term Fe uptake rates of inorganic Fe by Thalassiosira oceanica were determined using a basal media, consisting of sterile SOW (no trace metals or vitamins 3 2 1 were added and PO4 " and Si03 "were present at a concentration of 10 umo\ L" and 100 ixmol L"1, respectively) containing 20 nmol L"1 total inorganic Fe(II) or Fe(III), 50% of the total Fe was added as Fe 5 5 (specific activity of stock, 3.83 MBq pg"1, PerkinElmer) in 75 the presence or absence of 20 nmol L"1 Cd. Cadmium uptake experiments were also conducted using 20 nmol L"1 total Cd, 50% of which was C d 1 0 9 (specific activity of stock 14.53 MBq pg"1, Amersham) with or without the addition of 20 nmol L"1 cold inorganic Fe(ID or Fe(IJJ) in the basal uptake media. To avoid the rapid oxidation of Fe(II) and precipitation of Fe(IJI), all experiments were performed at either pH 6.6 or 7.5. Cells are not physiologically stressed over the course of the experiment in this pH range (Anderson and Morel 1982; Maldonado and Price 2000; Maldonado et al. 2006). To prepare pH 6.6 and 7.5 seawater, the seawater was first buffered with either 1 mmol L"1 of chelexed Piperazinel,4 bis-(2-ethanesulfonic acid) (PIPES, pKa 6.8) or N-(2-hydroxyethyl)-piperazine-N'-2-ethanesulfonic acid (HEPES, pKa 7.55), respectively. The pH was then adjusted using 10% trace metal grade HCL (Seastar Chemicals Inc.) while bubbling the seawater with 0.2 /xm filtered air; thereby accelerating the degassing of CO2 from seawater and lowering its buffering capacity. The seawater was bubbled until it reached equilibrium with the air and remained stable for months at the desired pH. Iron(II) was prepared by reducing an Fe(HT) dissolved in 0.1 mol L"1 NaCl stock solution with sulfur dioxide gas as the reducing agent (Hudson et al. 1992, Maldonado et al. 2006). Before the Fe(IT) was added to the uptake medium, the reductant was removed from the Fe(II) stock solution by bubbling it with 0.2 fim filtered nitrogen gas. The half-life of Fe(II) in SOW at pH 6.6 and 7.5 was determined spectrophotometrically using the Fe(II) specific ligand ferrozine (3-(2-pyridyl)-5,6-bis(4-phenyl-sulfonic acid)-l,2,4-triazine; Sigma Chemical) (Stookey 1970). The corresponding half-lives were 15 hours at pH 6.6 and 1 hours at pH 7.5 (data not shown). Accordingly, uptake experiments were conducted for 1.5 hours at pH 6.6 and for 0.5 hours at pH 7.5. Approximately 93% and 70% of the 76 added Fe(II) was remaining at the end of the experiments at pH 6.6 and 7.5, respectively. Batch cultures were grown in Fe-replete or Fe-limited medium. During mid-exponential phase, cells were harvested by gentle filtration onto acid-washed 3-/mi polycarbonate filters (Poretics) and immediately resuspended in the basal media. Cell concentration and average cell size in each treatment were determined using a Coulter Counter (model Z2). All experiments were conducted under ambient temperature and light conditions using acid washed polycarbonate bottles. Experiments were initiated by the addition of 5 5Fe and/or 1 0 9 Cd. Each resuspended culture was then sampled in duplicate approximately every 15 minutes by filtering 4 ml aliquots onto a 3 /mi polycarbonate filter. The pH was monitored throughout the experiments using a Corning 350 pH meter and varied < 1%. Before running dry, the cells were washed with either a 5 ml Ti(III) EDTA-citrate solution (Hudson and Morel 1989) or a 1 mmol L"1 DTP A rinse (Lee and Morel 1995) (dissolved SOW, pH adjusted to -8.1) to remove extracellularly bound Fe and Cd, respectively. Measurements df Fe and Cd uptake were replicated in independent cultures of T. oceanica. All filters with cells were immersed in liquid scintillation cocktail (ScintiSafe Plus 50% Cocktail, Fisher Scientific), and measured using Beckman-Coulter LS 6500 liquid scintillation counter. Uptake rates were calculated from linear regression of particulate metal concentrations as a function of incubation time (hours (h)) and normalized to cellular surface area (mol metal /xm"2 h"1). 77 3.3.3 Effects of TTM additions on Fe(II) uptake To test whether Fe(II) was entering the cell via the multicopper oxidase (MCO) /permease complex or a separate Fe(II) transporter, MCO activity in T. oceanica was inhibited with TTM (ammonium tetrathiomolybdate, (TSfH^MoS^ Alfa Aesar) (Maldonado et al. 2006, Chidambaram et al. 1984, Bissig et al. 2001). Ammonium tetrathiomolybdate selectively inhibits a variety of cupro-oxidases, including the multi-Cu ferroxidase (Chidambaram et al. 1984). Reportedly, TTM inhibits Cu-requiring enzymes by forming of a reversible ternary complex with enzyme-bound Cu (Bissig et al. 2001). During mid-exponential phase growth T. oceanica was exposed to 25 pmol L"1 TTM. After three hours of exposure to TTM, the cells were harvested gently, rinsed with SOW, and rapidly resuspended in uptake media without TTM to measure Fe(II) uptake. 3.4 Results 3.4.1 Effect of Fe and Cd on growth rate In all experiments T. oceanica was grown in the presence of 35 nmol L"1 Cd (pCd 12), the observed growth rate of T. oceanica in Fe-replete and Fe-limiting media was unaffected by the presence of Cd (Table 3.1). The fastest single growth rate (pmax) measured for T. oceanica was 1.90 dd"1 in high iron (pFe 19) media. Rate of growth with or without Cd averaged 1.67 dd"1 (p/pm ax- 0.88) in pFe 19 media, 1.33 dd"1 (p/pmax = 0.70) in pFe 21 media, and 0.893 dd"1 (p/pmax = 0.47) in pFe 21.5 media. 78 3.4.2 Inorganic Fe(II) vs. Fe(III) uptake rates In order to determine whether T. oceanica has an Fe(II) transporter able to directly access Fe(II) and whether it is up-regulated under Fe-limitation, we first examined the rates of inorganic Fe(II) uptake by cultures grown under varying degrees of Fe-limitation. T. oceanica had an average Fe(II) uptake rate of 2.02 ± 0.07 x 10"21 mol Fe um"2 h"1 when grown in Fe-replete media (pFe 19, p/pm a x = 0.87). As cultures were subjected to Fe-limitation, Fe(II) uptake rates increased significantly (Figure 3.1 A), up to a maximum of 31.20 x 10"21 mol Fe um"2 h"1 for the most severely Fe-limited culture (pFe 21.5, p/pmax- 0.36). T. oceanica therefore increased its Fe(II) uptake rate ~15 fold when the Fe-limited growth rate decreased by -60%. The rates of inorganic Fe(IH) uptake by T. oceanica were also up-regulated under Fe-limitation, however not to the extent of Fe(II) uptake (Figure 3.IB). The rate of Fe(III) uptake by Fe-sufficient cells (pFel9) was 4.93 x 10"21 mol Fe /mi"2 h"1, 2.4 times higher than its Fe(II) uptake rates. The rates of Fe(III) uptake increased -5 fold when cells were severely Fe-limited (pFe 21.5, p/pmax=0.36), to 26.7 x 10"21 mol Fe /tm"2 h'1. Thus, severely Fe-limited T. oceanica achieved similar rates of Fe(III) and Fe(II) uptake. To test whether Fe(n) was entering the cell via the MCO/permease complex (Maldonado et al. 2006) or via a non-specific Fe(U) transporter, we attempted to inhibit the MCO with TTM. No significant effect by TTM on Fe(II) transport was found when T. oceanica was grown under Fe-limiting conditions (pFe 21.5) (p = 0.714, paired t-test) (Table 3.2). 79 3.4.3 Inorganic Fe uptake rates in the presence of Cd To establish the competitive interaction between Fe and Cd uptake, we determined Fe(III) and Fe(Ii) uptake rates in the presence of equimolar concentrations of Cd. Cadmium had no significant effect on Fe(IJJ) uptake (p = 0.275, paired t-test) (Table 3.3 and Figure 3.2A). However, Cd had a significant negative effect on Fe(II) uptake (p = 0.036, paired t-tet) (Table 3.3 and Figure 3.2B). In the presence of Cd the average Fe(II) uptake rate by Fe-limited T. oceanica (pFe 21.5) decreased by 50%, from 12.43 ± 2.77 x 10"21 mol Fe um'2 h"1 to 6.35 ±1.05 x 10"21 mol Fe /nm"2 h"1. Similarly, in one experiment for an Fe-sufiicient T. oceanica culture (pFe 19) (results not shown), a 30% decrease in Fe(II) uptake rate from 2.02 to 1.39 x 10"2! mol Fe um'2 h"1 was observed in the presence of Cd. Iron(II) uptake experiments with Fe-limited T. oceanica, were also conducted at pH 7.5 to investigate the effect of low pH on inorganic Fe(U) uptake. Iron(II) uptake rates were -65% faster at pH 7.5 than at pH 6.6, with an average uptake rate of 36.31 ± 2 1 4.42 mol Fe um h" . In the presence of equimolar Cd concentrations the average Fe(II) uptake rate for Fe-limited T. oceanica at pH 7.5 was also reduced by -35% (p = 0.038, paired t-test) to 23.83 ± 2.93 x 10"21 mol Fe /mi"2 h"1 (Table 3.3). 3.4.4 Cadmium uptake rates in the presence of inorganic Fe Based on the observation that inorganic Fe(II) uptake rates were reduced by nearly 50% in the presence of equal concentrations of Cd, we hypothesized that Fe(II) would have an antagonistic effect on Cd uptake. Therefore, we determined Cd uptake rates in the presence of equal concentrations of Fe(II) or Fe(III). The rates of Cd uptake 80 alone (26.66 ± 3.14 x 10"21 mol Cd /mi"2 h"1) were approximately two times faster than Fe(II) uptake in the absence of Cd (12.43 ± 2.77 x 10"21 mol Cd /mi"2 h"1). Iron(in) had no significant effect on Cd uptake rates (p = 0.554, paired t-test) (Table 3.3, Figure 3.3). In contrast, Cd uptake rates were reduced by 35% in the presence of Fe(II) from 26.66 ± 3.14 to 17.25 ± 2.01 x 10"21 mol Fe /mi"2 h"1 (p = 0.005) (Table 3.4, Figure 3.3). 3.5 Discussion It is well established that Fe-limited marine phytoplankton accumulate higher intracellular Cd concentrations than Fe-replete phytoplankton (Sunda and Huntsman 2000; Cullen et al. 2003; Lane et al., Chapter 2). This study however, is the first to investigate the physiological mechanism behind enhanced cellular Cd quotas in Fe-limited oceanic waters. In this study, T. oceanica was used as a model diatom, and we present physiological evidence for a putative non-specific divalent metal transporter. Much information is available on other proposed Fe-uptake mechanisms by this diatom (Maldonado and Price 2001; Peers et al. 2005; Shaked et al. 2005; Maldonado et al, 2006). However, this work provides insight into the mechanisms of inorganic Fe(II) and Cd(IT) uptake in diatoms. Our findings suggest that both metals are taken up via a non-specific divalent metal transporter that is up-regulated under Fe-limitation. The physiological data presented here for T. oceanica complements the genomic data for T. pseudonana (Kustka et al. 2007). This study also suggests that inorganic Fe(II), if available, might be a significant source of Fe to Fe-limited marine diatoms. 81 3.5.1 Inorganic Fe(II) vs. Fe(III) uptake rates Previous genomic work with the marine diatom T. pseudonana has found genes encoding for a non-specific Fe(IT) transporter (TpNRAMP), as well as a high-affinity uptake Fe(III) system. This latter system is similar to that of yeast and is comprised of putative reductases, a MCO, and permease complex (Kustka et al. 2007). Kustka et al. (2007) also found that the expression of the genes that encode for all these transporters were up-regulated under Fe-deficiency. In our study, when T. oceanica is not Fe-limited, Fe(II) uptake was slower than Fe(III) uptake. Similar results were found by Maldonado et al. (2006), who found slower rates of inorganic Fe(II) uptake compared to Fe(HI) uptake (3.51and 29.8 x 10"21 mol Fe um'2, respectively at 70 nmol L"1 Fe) by mildly Fe-limited (pFe 20.5) T. oceanica . These rates are within the range of those measured in our experiments. However, severely Fe-limited T. oceanica had similar rates of inorganic Fe(II) and Fe(ffl) uptake. Iron-limited T. weissflogii has also been shown to achieve comparable inorganic Fe(II) and Fe(III) uptake rates at pH 6.3 (see Figure 10, Anderson and Morel, 1982). Iron-limitation elicited a -15 fold increase in the Fe(II) uptake rates and a ~5 fold increase in the Fe(HI) uptake rates. These results are consistent with T. pseudonana genomic data (Kustka et al. 2007). Kustka et al. (2007) showed that the expression of the non-specific Fe(II) transporter gene, TpNRAMP, increased dramatically under Fe-limitation and was down-regulated 25-fold upon Fe addition. Even though it is well established that inorganic Fe(II) and Fe(III) may be taken up by diatoms, the physiological evidence for the specific transporters used to internalize ferric and ferrous ions is more controversial. Both inorganic and organically bound Fe(III) are believed to be accessible to phytoplankton via enzymatic reduction of Fe(III), 82 followed by oxidation of Fe(II) and subsequent internalization (Shaked et al. 2005; Maldonado et al. 2006). The relative rates of Fe(III) uptake from inorganic and organic complexes are determined by the degree of Fe-limitation experienced by the cells (Maldonado and Price 2001). This complex Fe(III) uptake mechanism imparts high specificity to the transporter. In contrast, the mechanism of Fe(II) transport is less understood. In order to determine whether Fe(II) was transported through a non-specific Fe(II) transporter (such as NRAMP's) or through the MCO/permease complex transporter, we determined Fe(IT) uptake rates by cells pre-incubated with TTM, a well-known inhibitor of MCOs. This study found that TTM caused no inhibition of Fe(II) uptake rates by severely Fe-limited cultures, suggesting that the Fe(II) was transported ; directly into the cell through a divalent metal transporter. However, Maldonado et al. (2006) also using T. oceanica cultures found, in a single experiment, that TTM had a remarkable inhibitory effect on Fe(II) uptake rates. They argued that Fe(II) was being ^oxidized and internalized by the oxidase/permease complex. It is expected that the reductases and the MCO/permease complex are a highly coupled transport system to ensure its high affinity and specificity for Fe(III). Thus, it is surprising that Fe(II) can access the MCO complex. Indeed, more recent physiological data suggest that the only transition metals able to access this complex are biologically reducible metals such as Cu(II) (Kustka et al. 2007). We believe that extremely high Fe(II) concentrations (10 pmol L"1) used in the Maldonado et al. (2006) experiment, and the high specific growth rate of their culture (1.89 d"1), may account for the conflicting result. It is plausible that under mildly Fe-limited growth conditions few Fe(II) transporters are present. Moreover, at a concentration of lOpmol L"1 Fe(IT), Fe(II) might have been able to access and saturate 83 the MCO component of the high-affinity transport complex. Therefore, in the Maldonado et al. (2006) experiments, added Fe(U) may have been oxidized and taken-up by the MCO/permease complex. The paired reduction and oxidation/transport steps provide a degree of specificity for the high-affinity Fe(III) transporter under most conditions and Fe(II) uptake measured in our experiments was presumably through a non-specific divalent metal transporter. 3.5.2 Interaction between Fe(III) and Cd Our results suggest that there is no interaction between Cd and inorganic Fe(III) transport. It has been suggested that Fe(III) must be transported via the high-affinity Fe uptake system and thus must be reduced before uptake (Shaked et al. 2005). Given that Cd had no effect on Fe(JJI) uptake in our experiments, it suggests that Cd does not interfere with the high-affinity Fe uptake system. Recent work in our laboratory has shown that Cd does not have an effect on Fe-DFB uptake, which presumably must be transported via the high-affinity Fe uptake system (Semeniuk, D., pers. comm.). These results are consistent with those of Kustka et al. (2007), showing that micromolar concentrations of Cd and other divalent metals (Mn and Zn) could not competitively inhibit inorganic Fe(III) uptake by T. pseudonana. The only metal found to inhibit Fe(III) uptake was Cu(II). Interestingly, Fe(IIi) uptake rates were restored by the addition of the Cu(I)-binding ligand BCDS (bathocuproine disulfonic acid), suggesting that that Cu(II) was reduced and the effect of Cu on Fe(III) uptake was caused by Cu(I) binding the oxidase. Similarly, the only other known metal to compete for Fe uptake mediated by the high-affinity Fe uptake system in yeast is Cu. At high concentrations, 84 Cu can be reduced by the reductases (Eide 1998; Jones et al. 1987) and oxidized by the MCO (Shi et al. 2003; Stoj and Kosman 2003). These combined results suggest that only metals that are first reduced can interact with MCO/permease transporter, thus Cd, a non-reducible metal, must enter the cell through an alternative transporter. 3.5.3 Effects of C d on Fe(II) uptake rates In contrast to Fe(III) uptake, inorganic Fe(II) uptake rates in our experiments were reduced by -50% in the presence of Cd, suggesting that marine phytoplankton have separate transporters for inorganic Fe(III) and Fe(II) and that Cd and Fe(II) enter the cell through the same transporter. Iron(II) uptake experiments were also conducted at pH 7.5 to investigate how changes in pH may effect the Fe(II) or Cd specificity of the non-specific divalent metal transporter. The results of these higher pH experiments would also allow us to better estimate Fe(II) uptake rates at the natural seawater pH. Iron(II) uptake rates were -3 times higher at pH 7.5 than at pH 6.6. Since 30% of the Fe(IT) was converted to Fe(III) in the course of the experiments, the rates of Fe uptake may have included simultaneous Fe(II) and Fe(III) uptake. However, at this pH (7.5) Fe(II) uptake rates were reduced by -35% in the presence of Cd, further confirming the antagonistic effect of Cd on Fe(II) uptake. 3.5.4 Effects of Fe(II) on C d uptake rates Based on our results of the inhibitory effects of Cd on Fe(II) uptake we expected Cd uptake rates to be similarly inhibited by Fe(II). Indeed, the presence of Fe(II) inhibited Cd uptake by -35% (Table 3.4, Figure 3.3). However, absolute Cd uptake 85 rates, were ~2 times faster than the rates of Fe(II) uptake. These findings suggest that there may be more than one type of transporter for Cd; one that can take up both Cd and Fe(H) and one that takes up Cd with a higher specificity than Fe(IJ), or perhaps does not transport Fe(H) at all. Yeast and plants have two different families of non-specific divalent metal transport proteins: NRAMP's and ZIP's (ZRT, IRT-like Protein), with family members differing in their substrate range and specificity. In yeast, three NRAMP's have been identified (SMF 1-3), and have been shown to mediate the uptake of Mn(II), Cu(II), Co(H), Cd,(II) and Fe(II) (Supek et al. 1997; Liu et al. 1997; Chen et al., 1999). Several members of the NRAMP family have also been identified in plants. Many of these NRAMP's are up-regulated under conditions of Fe deficiency and some have been linked to enhanced Cd uptake and sensitivity in plants (Curie et al. 2000; Thomine et al. 2000). AtlRTl, a member of the ZIP family cloned from Arabidopis, was first thought to be a Fe transporter, but is now known to transport Fe(II), Zn(II), Mn(II) and Cd(II). Several pieces of evidence point to a role of IRT1 in mediating the accumulation of Cd in Fe-deficient plants: cadmium was shown to compete with Fe for uptake in yeast expressing IRT1 (Eide et al. 1996) and Fe-limitation increased their Cd sensitivity (Guerinot et al. 2000). Under Fe-limitation plants over-expressing the IRT1 gene have been found to accumulate higher concentrations of Cd and Zn than wild types (Guerinot et al. 2000). ZRT1, a gene similar to IRT1, encodes for a high-affinity Zn transport system in yeast but, it does not play a role in Fe uptake in yeast (Zhao et al. 1996). These studies suggest that multiple non-specific divalent metal transporters are common in yeasf and plants and many of these transporters are up-regulated under Fe-limitation often leading to a hyper-86 accumulation of Cd. Other phytoplankton studies have also suggested that cellular uptake of Cd is controlled by two inducible transport systems: the Mn system which is enhanced at low Mn(II) concentrations and a separate system induced at low cellular Zn (Sunda and Huntsman 1996, 1998, 2000). Based on what we know about divalent trace metal transporters in yeast and plants (see above) it is likely that the high affinity Mn(II) transporter proposed in these studies is a NRAMP like transporter and the system induced at low cellular Zn is perhaps a ZEP-like transporter. More genomic work needs to be completed in order to elucidate the types of divalent metal transporters found in phytoplankton and the degree of metal specificity of the transporters. 3.6 Oceanographic Implications This study provides a mechanistic understanding behind enhanced Cd accumulation by Fe-limited marine phytoplankton. The up-regulation of non-specific divalent metal transporters by Fe-limited phytoplankton may lower the concentration of Cd as well as other divalent metals in the surface ocean. This has major implications for the use of the CdVCa paleo-nutrient proxy, which relies on the assumption that dissolved •a Cd versus PO4 " concentrations are constant in oceanic surface waters. Further investigation into how Fe-limited marine phytoplankton control divalent metal chemistry in seawater is needed. Phytochelatin is an intracellular metal-binding polypeptide, synthesized in phytoplankton in response to toxic metals such as Cd (Ahner et al. 1998). Intracellular phytochelatin concentrations in particulate samples collected from the equatorial Pacific 87 were found to be unexpectedly high, compared to phytochelatin levels found in cultured oceanic phytoplankton (Ahuer et al. 1998). The average phytochelatin concentration in phytoplankton from the equatorial Pacific was slightly higher than in coastal areas, where Cd concentrations are typically much higher (Ahner et al. 1998). The equatorial Pacific is a well know HNLC area (Martin et al. 1994; Coale et al. .1996). Low ambient Fe concentrations in the equatorial Pacific may led to the up-regulation of non-specific divalent metal transporters by phytoplankton, resulting in enhanced Cd accumulation. High phytochelatin production is likely to be a direct response to the high intracellular Cd concentrations observed in Fe-limited oceanic phytoplankton (Lane et al., Chapter 2). The results from this study show that Fe(H), if available, could prove to be a significant source of Fe for Fe-limited phytoplankton. Zhuang et al. (1995) and Boye et al. (2003) observed that as much as half of the dissolved Fe in seawater was Fe(II) in the eastern and western North Atlantic. Atmospheric aerosols, including dust, contained a significant amount of Fe(II) (Zhuang et al. 1992a,b; Zhu et al. 1997). Rain has been found to contain nanomolar concentrations of Fe(II) and may provide a significant source of Fe(IT) to the surface ocean (Zhaung 1995; Willey et al. 2000, 2004; Kieber et al. 2001a,b, 2003). Iron(TI) in seawater also remains longer in solution in the presence of rainwater (Kieber et al. 2001a). Iron(II) added during the large scale Fe-fertilization experiment SOIREE, was found to be very stable in the Southern Ocean (Croot et al. 2001). This was attributed the Southern Ocean having high UV radiation, which causes enhanced photoreduction (Rijkenberg et al. 2005), as well as cold temperatures (Millero et al. 1987) and low hydrogen peroxide levels (Yocis 2000), which both prevent the rapid oxidation of Fe(II). Up-regulation of Fe(II) transporters would therefore be 88 advantageous under various ambient physical and chemical conditions, especially during an episodic Fe(II) supply from dust or rain. Even if background levels of inorganic Fe(II) in the modern ocean are low, non-specific divalent metal transporters may be up-regulated under low Fe conditions to enhance uptake of other essential metals such as Cu(II), which is needed for the high-affinity Fe uptake system. 89 3.7 Figures E PH 6 15 6 -*-» a, 3 Relative growth rate (100 x (a/^ max) Figure 3.1. Inorganic Fe uptake rates by T. oceanica as a function of Fe-limited relative growth rate (100 x ix/fimax). Experiments were conducted at pH 6.6 in the presence of 20 nmol L"1 inorganic Fe(H). 90 E a. u o X KL, OH 0) P L ID o 1 OH 25 20 15 10 • Fe(III) • Fe(III) + Cd A i i 1 1 0 0.2 0.4 0.6 0.8 1 1.0 1. Time (hours) 1.0 1.2 Time (hours) Figure 3.2. Time course accumulation of particulate Fe by Fe-limited (pFe 21.5) T. oceanica at pH 6.6. Experiment was conducted in the presence of 20 nmol L"1 inorganic Fe(ffl) or Fe(U) (50% 55Fe), with or without additions of 20 nmol L"1 Cd. Points represent the mean ± range of duplicate analyses of a single culture. The rates of inorganic Fe uptake were calculated by least-squares regression (A) Inorganic Fe(III) uptake rates were 16.03 (r2 = 0.97) and 15.98 x 10"21 mol Fe am2 h"1 (r2 = 0.98) for Fe (III) and Fe(III) + Cd treatments, respectively. (B) Inorganic Fe(II) uptake rates were 17.96 (r2 = 0.92) and 8.44 x 10"21 mol Fe um2 h"1 (r2 = 0.96) for Fe (II) and Fe(H) + Cd treatments, respectively. 91 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Time (hours) Figure 3.3. Time course accumulation of particulate 1 0 9 Cd by Fe-limited (pFe 21.5) T. oceanica at pH 6.6. Experiment was conducted in the presence of 20 nmol L"1 Cd (50% 1 0 9Cd), in the absence of Fe or with a 20 nmol L"1 addition of Fe(III) or Fe(II). Points represent the mean ± range of duplicate analyses of a single culture. The rates of Cd uptake were calculated by least-squares regression and were 28.53 (r2 = 0.97), 25.95 (r2 = 0.99) and 17.24 x 10"21 mol Fe am'2 h"1 (r2 = 0.97) for Cd with no Fe addition, Cd + Fe(III) and Cd + Fe(II) treatments, respectively. 92 3.8 Tables Table 3.1. Average relative growth rates (/1//W) of T. oceanica under various Fe treatments in the absence or presence of 35 nmol L"1 Cd(pCd 12). The maximum growth rate ( / w ) of T. oceanica was measured as 1.90 dd"1, in pFe 19 media. The p values represent two-tailed unpaired t-test analysis (assuming equal variance) comparing the effect of Cd on growth rates within each Fe growth treatment. T. oceanica average relative growth rates (fi/umax) Growth treatment [i/umax Std. Error n t-test p value pFe 19 0.88 0.047 21 0.952 pFe 19 + Cd 0.88 0.037 21 pFe21 0.70 0.042 20 0.973 pFe21 +Cd 0.70 0.039 19 pFe21.5 0.47 0.029 20 0.917 pFe21.5 + Cd 0.47 0.037 20 Table 3.2. Effects of TTM on inorganic Fe(H) uptake rates by Fe-limited (pFe 21.5) T. oceanica at pH 6.6. Half of the culture (+TTM) was exposed to 25pmol L"1 TTM for 3 h prior to the experiment. Experiments were conducted with 20 nmol L"1 (50% 55Fe) inorganic Fe(II). The p value between two columns represents a two-tailed paired t-test analysis. T. oceanica Fe uptake rates (x 10"21 mol Fe /mi"2 h"1) Growth media -TTM +TTM pFe21.5 20.3 23.3 18.0 16.9 Average 19.1 20.1 Std. Error 1.16 3.18 t-test p value 0.714 93 Table 3.3. Effects of Cd on inorganic Fe(III) and Fe(II) uptake rates by Fe-limited (pFe 21.5) T. oceanica. Experiments were conducted with 20 nmol L"1 (50% 55Fe) inorganic Fe(III) or Fe(II), with or without additions of 20 nmol L"1 Cd. The p value between two columns for Fe(III) uptake treatments represents a two-tailed paired t-test analysis. The p value between two columns for Fe(II) uptake treatments represents a one-tailed paired t-test analysis. T. oceanica Fe uptake rates (x 10"21 mol Fe um'2 h"1) Fe(III) Fe(II) -Cd +Cd -Cd +Cd pH 6.6 16.0 16.0 16.0 8.44 Shown in Fig 3.2 16.2 16.0 9.72 5.10 9.46 8.83 9.61 5.52 Average 13.89 13.62 12.4 6.35 Std. Error 2.22 2.40 2.77 1.05 t-test p value 0.275 0.0360 pH 7.5 31.9 20.9 40.7 26.8 Average 36.3 23.8 Std. Error 4.42 2.93 t-test p value 0.0380 Table 3.4. Effects of Fe on Cd uptake rates by Fe-limited (pFe 21.5) T. oceanica at pH 6.6. Experiments were conducted with 20 nmol L"1 Cd (50% 1 0 9Cd) in the absence of Fe or with 20 nmol L"1 inorganic Fe(III) or Fe(II). The p value under the +Fe(III) column represents a two-tailed paired t-test analysis comparing the Cd uptake rates of the -Fe and +Fe(III) experiments. The p value under the +Fe(II) column represents a one-tailed paired t-test analysis comparing the Cd uptake rates of the -Fe and +Fe(II) experiments. T. oceanica Cd uptake rates (x 10"21 mol Cd um'2 h"1) -Fe +Fe(III) +Fe(II) Shown in Fig 3.3 28.5 25.9 17.2 21.8 21.6 13.8 31.7 32.4 20.8 Average 27.4 26.7 17.3 Std. Error 2.93 3.14 2.01 t-test P value 0.554 0.005 94 3.9 References Agranoff, D., L. Collins, D. Kehres, T. Harrison, M. Maguire, and S. Krishna. 2005. The Nramp orthologue of Cryptococcus neoformans is a pH-dependent transporter of manganese, iron, cobalt and nickel. Biochem. J. 385: 225-232. Anderson, M. A. and F. M. M. Morel. 1982. The influence of aqueous iron chemistry on the uptake of iron by the coastal Thalassiosira weissflogii. Limnol. Oceanogr. 27: 789-813. Bissig, K. D., T. C. Voegelin, and M. Solioz. 2001. Tetrathiomolybdate inhibition of the Enterococcus hirae CopB copper ATPase. FEBS Lett 507: 367-370. Boye, M., A. P. Aldrich, C. M. G. van den Berg, J. T. M. de Jong, M.Veldhuis, and H. J. W. de Baar. 2003. Horizontal gradient of the chemical speciation of iron in surface waters of the northeast Atlantic Ocean. Mar. Chem. 80: 129-143. Boyle, E. A. 1988. Cadmium: Chemical tracer of deepwater paleoceanography. Paleoceanography 3: 471-489. Boyle, E. A., F. Sclater, and J. M. Edmond. 1976. On the marine geochemistry of cadmium. Nature 263: 42-44. Bruland, K. W., G. A. Knauer, and J. H. Martin. 1978. Cadmium in northeast Pacific waters. Limnol. Oceanogr 23: 618-625. Cellier, M., G. Prive, A. Belouchi, T. Kwan, V. Rodrigues, W. Chia, and P. Gros. 1995. Nramp defines a family of membrane-proteins. Proc. Nat. Acad. Sci. 92: 10089-10093. Chen X. Z., J. B. Peng, A. Cohen, H. Nelson, N. Nelson, M. A. Hediger. 1999. Yeast SMF1 mediates H(+)-coupled iron uptake with concomitant uncoupled cation currents. J. Biol. Chem. 274: 35089-35094 Chidambaram, M. V., G. Barnes, and E. Frieden. 1984. Inhibition of ceruloplasmin and other copper oxidases by thiomolybdate. J. Inorg. Biochem. 22: 231-239. Cullen, J. T., Z. Chase, K. H. Coale, S. Fitzwater, and R. M. Sherrell. 2003. Effect of iron limitation on the cadmium to phosphorus ratios of natural phytoplankton assemblages from the Southern Ocean. Limnol. Oceanogr 48: 1079-1087. Curie C , J. M. Alonso, M. Le Jean, J. R. Ecker, and J. F. Briat. 2000. Involvement of NRAMP1 from Arabidopsis thaliana in iron transport. Biochem. J. 347: 749-755. de Baar, H. J. W., P. M. Saager, R. F. Notling, and J. Van Der Meer. 1994. Cadmium versus phosphate in the world ocean. Mar. Chem. 46: 261-281. 95 Eide, D. J. 1998. The molecular biology of metal ion transport in Saccharomyces cerevisiae. Ann. Rev. Nutrition 18: 441-469. Eide, D., M. Broderius, J. Fett, and M. L. Guerinot. 1996. A novel iron-regulated metal transporter from plants identified by functional expression in yeast. Proc. Nat. Acad. Sci. 93: 5624-5628. Elderfield, FL, and R. E. M. Rickaby. 2000. Oceanic Cd/P043" ratio and nutrient utilization in the glacial Southern Ocean. Nature 405: 305-310. Gunshin, H. and others. 1997. Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature 388: 482-488. Guerinot, M. L. 2000. The ZIP family of metal transporters. Biochim. Biophys. Acta 1465: 190-198. Hudson, R. J. M. and F. M. M. Morel. 1989. Distinguishing between extra- and intracellular iron in marine phytoplankton. Limnol. Oceanogr. 34: 1113-1120. Hudson, R. J. M., D. T. Covault, and F. M. M. Morel. 1992. Investigations of iron coordination & redox reactions in seawater using 59Fe radiometry & ion-pair solvent extraction of amphiphilic iron complexes. Mar. Chem. 38: 209-235. Jones, G. J., B. Palenik, and F. M. M. Morel. 1987. Trace metal reduction by phytoplankton: The role of plasmalemma redox enzymes. J. Phycol. 23: 237-244. Kieber, R. J., B. Peake, J. D. Willey, and B. Jacobs. 2001b. Iron speciation and hydrogen peroxide concentrations in New Zealand rainwater. Atmos. Environ. 35: 6041-6048. Kieber, R. J., J. D. Willey, and G. B. Avery. 2003. Temporal variability of rainwater iron speciation at the Bermuda Atlantic Time Series Station. J. Geophys. Res. 108: 1-7. Kieber, R. J., K. H. Williams, J. D.Willey, S. A. Skrabal, and G. B. Avery. 2001a. Iron speciation in coastal rainwater: Concentration and deposition to seawater, Mar. Chem. 73: 83-95. Kustka, A. B., A. E. Allen, F. M. M. Morel. 2007. Sequence analysis and transcriptional regulation of Fe acquisition genes in two marine diatoms. J. Phycol. (In press). Lee, J. G., and F. M. M. Morel. 1995. Replacement of Zn by cadmium in marine phytoplankton. Mar. Ecol. Progr. Ser. 127: 305-309. Liu X. F., F. Supek, N. Nelson, V. C. Culotta. 1997. Negative control of heavy metal uptake by the Saccharomyces cerevisiae BSD2 gene. J. Biol. Chem. 272: 11763-11769. 96 Martin, J. H., R. M. Gordon, S. Fitzwater, and W. W. Broenkow. 1989. VERTEX: Phytoplankton/iron studies in the Gulf of Alaska. Deep-Sea Res. 36: 649-680. Martin, J. F£., R. M. Gordon, and S. Fitzwater. 1990. Iron in Antarctic waters. Nature 345: 156-158. Maldonado, M. T., and N. M. Price. 2000. Nitrate regulation of Fe reduction and transport by Fe-limited Thalassiosira oceanica. Limnol. Oceanogr. 45: 814-826. Maldonado, M. T., and N. M. Price. 2001. Reduction and transport of organically bound iron by Thalassiosira oceanica (Bacillariophyceae). J. Phycol. 37: 298-309. Maldonado, M. T., A. E. Allen, J. S. Chong, K. Lin, D. Leus, N. Karpenko, and S. L. Harris. 2006. Copper-dependent iron transport in coastal and oceanic diatoms. Limnol. Oceanogr. 51: 1729-1743. Millero, F. J., S. Sotolongo, M. Izaguirre. 1987. The oxidation-kinetics of Fe(II) in seawater. Geochim. Cosmochim. Acta. 51: 793-801. Peers, G., S. A. Quesnel, and N. M. Price. 2005. Copper requirements for iron acquisition and growth of coastal and oceanic diatoms. Limnology and Oceanography 50: 1149-1158. Shaked, Y., A. B. Kuska, and F. M. M. Morel. 2005. A general kinetic model for iron acquisition by eukaryotic phytoplankton. Limnol. Oceanogr. 50: 872-882. Shi, X. L., C. Stoj, A. Romeo, D. J. Kosman, and Z. W. Zhu. 2003. Frelp Cu2+ reduction and FET3p Cul+ oxidation modulate copper toxicity in Saccharomyces cerevisiae. J. Biol. Chem. 278: 50309-50315. Stoj, C. and D. J. Kosman. 2003. Cuprous oxidase activity of yeast Fet3p and human ceruloplasmin: implication for function. Febs. Letters 554: 422-426. Stookey, L. L. 1970. Ferrozine- a new spectrophotometric reagent for iron. Anal. Chem. 42: 779-781. Supek F., L. Supekova, H. Nelson, N. Nelson. 1996. A yeast manganese transporter related to the macrophage protein involved in conferring resistance to mycobacteria. Proc. Natl. Acad. Sci. USA 93: 5105-5110. Price, N. M., G. I. Harrison, J. G. Hering, R. J. Hudson, P. Nirel, B. Palenik, and F. M. M. Morel. 1988/89. Preparation and chemistry of the artificial algal culture medium Aquil. Biol. Oceanogr. 5: 43-46. 97 Rijkenberg, M. J. A., A. C. Fischer, J. J. Kroon, L. J. A. Gerringa, K. R. Timmermans, H. T. Wolterbeek, and H. J. W. de Baar. 2005. The influence of UV irradiation on the photoreduction of iron in the Southern Ocean, Mar. Chem., 93, 119-129. Rosakis, A. and W. Koster. 2005. Divalent metal transport in the green microalga Chlamydomonas reinhardtii is mediated by a protein similar to prokaryotic Nramp homologues. Biometals 18: 107-120. Shaked, Y., A. B. Kustka, and F. M. M. Morel. 2005. A general kinetic model for iron acquisition by eukaryotic phytoplankton. Limnol. Oceanogr. 50: 872-882. Sunda, W. G., and S. A. Huntsman. 1996. Antagonisms between cadmium and zinc toxicity and manganese limitation in a coastal diatom. Limnol. Oceanogr. 41: 373-387. Sunda, W. G., and S. A. Huntsman. 1998. Control of Cd concentrations in a coastal diatom by interactions among free ionic Cd, Zn, and Mn in seawater. Environ. Sci. Technol. 32:2961-2968. Sunda, W. G., and S. A. Huntsman. 2000. Effect of Zn, Mn, and Fe on Cd accumulation in phytoplankton: Implications for oceanic Cd cycling. Limnol. Oceanogr. 45: 1501-1516. Thomine, S., R. C. Wang, J. M. Ward, N. M. Crawford and J. I. Schroeder. 2000. Cadmium and iron transport by members of a plant metal transporter family in Arabidopsis with homology to Nramp genes. Proc. Nat. Acad. Sci. 97: 4991-4996. Westall, J. C , J. L. Zachary, and F. M. M. Morel. 1976. MINEQL: A computer program . for the calculation of chemical equilibrium composition of aqueous systems. R. M. Parsons Lab for Water Resources and Environmental Engineering, Department of Civil Engineering, MIT. Willey, J. D., R. J. Kieber, K. H. Williams, J. S. Crozier, S. A. Skrabal, and G. B. Avery. 2000. Temporal variability of iron speciation in coastal rainwater. J. Atmos. Chem. 37: 185-205. Willey, J. D., R. J. Kieber, and G. Brooks Avery Jr. 2004. Effects of Rainwater Iron and Hydrogen Peroxide on Iron Speciation and Phytoplankton Growth in Seawater near Bermuda. J. Atmos. Chem. 47: 209-222. Yocis, B. H., D. J. Kieber, and K. Mopper. 2000. Photochemical production of hydrogen peroxide in Antarctic waters. Deep-Sea Res. 47: 1077-1099. Zhao, H. and D. J. Eide. 1996. The ZRT2 gene encodes the low affinity zinc transporter in Saccharomyces cerevisiae. J. Biol. Chem. 271: 23203-23210. 98 Zhu, X. R., J. M. Prospero, and F. Millero. 1997. Diel variability of soluble Fe(II) and soluble total Fe in North African dust in the trade winds at Barbados. J. Geophys. Res. 102: 21297-21305. Zhuang, G., Z. Yi, R. A. Duce, and P. R. Brown. 1992a. Chemistry of iron in marine aerosols, Glob. Biogeochem. Cyc. 6: 161-173, Zhuang, G., Y. Zhen, R. A. Duce, and P. R. Brown. 1992b. Link between iron and sulphur cycles suggested by detection of Fe(U) in remote marine aerosols. Nature. 355: 537-539. Zhuang, G., Z. Yi, and G. T. Wallace. 1995. Iron (II) in rainwater, snow, and surface seawater from a coastal environment. Mar. Chem. 50: 41-50. 99 Chapter 4: Conclusion The findings of this thesis provide a better understanding of the mechanisms of Fe acquisition in Fe-limited marine phytoplankton, as well as the factors controlling Cd biogeochemistry in surface waters. Iron limitation, both in the lab and field, was shown to affect the intracellular Cd content of phytoplankton. Physiological data suggests that Cd is internalized through a non-specific divalent metal transporter, which is highly up-regulated under low Fe. In addition, large differences in Cd:C ratios among phytoplankton species/phyla were observed. These combined results suggest that Cd biogeochemistry in surface oceanic waters may be largely controlled by phytoplankton species composition and Fe bioavailability. 4.1 Importance of Fe(II) For more than 30 years, phytoplankton physiologists have tried to elucidate the mechanism of Fe uptake by marine phytoplankton. It is thought that inorganic Fe(III) is the dominant Fe species taken-up by phytoplankton under Fe-sufiicient conditions (Anderson and Morel 1982). However, under Fe-limiting conditions Fe(III) bound to organic complexes may also serve as an Fe source (Maldonado and Price 2000). Both inorganic and organic forms of Fe(III) are thought to be transported by a high-affinity uptake system (Shaked et al. 2005). This high-afiinity system involves an initial reduction of Fe(III) to Fe(II) at the cell surface permeases, followed by the coupled oxidation of Fe(II) and transport of Fe(III) at a MCO/permease complex (Shaked et al. 2005; Maldonado et al. 2006; and Kustka et al. 2007). This system has been identified as a high-affinity Fe transport complex because it is highly enhanced under Fe-limitation 100 and allows phytoplankton to access Fe when concentrations are low and organic complexion is high (Maldonado and Price 2001). Despite the importance of Fe(III) as the dominant Fe species used for marine phytoplankton nutrution, inorganic Fe(II) uptake has also been observed, however the mechanism for Fe(II) transport is unknown (Anderson and Morel 1982; Maldonado et al. 2006). This thesis shows the relative importance of inorganic Fe(ffl) versus Fe(II) transport depending on the Fe status of the cell. Iron(III) was shown to be the preferred form of Fe for Fe-sufficient phytoplankton. However, when phytoplankton are severely Fe-limited, Fe(II) and Fe(III) uptake rates are approximately equal. The competitive inhibition found between Cd and Fe(II) in T. oceanica in this study, points to the role that TpNRAMP may play in Fe(II) uptake in marine diatoms. The work of this thesis reconciles the findings of Shaked et al. (2005); Maldonado et al. (2006); and Kustka et al. (2007) by suggesting that all forms of Fe(III) must be reduced and transported via the high-affinity uptake system, however Fe(II) may be taken up directly by a non-specific divalent metal transporter. This dissertation also suggests that Fe(II) may play a more important role in satisfying the Fe requirements of Fe-limited phytoplankton than previously thought. Up-regulation of Fe(II) transporters would be advantageous to marine phytoplankton under ambient conditions that support Fe(II), such as in the Southern Ocean. Up-regulation of these non-specific divalent metal transporters during Fe-limitation would also lead to the enhanced accumulation of other divalent metals such as Cd and Zn. 101 4.2 Modeling the global Cd:P043" Relationship Both Sunda and Huntsman (2000) and Cullen et al. (2006) modeled how preferential accumulation of Cd by phytoplankton in Fe-limited waters could account for high slopes in the dissolved Cd:PC>4 " relationship in nutricline waters of HNLC regions. Sunda and Huntsman (2000) based their model on how free ion concentrations of Zn, Mn, and Cd changed the Cd:C ratios of T. oceanica in culture, combined with estimates of the free ion concentrations of these metals in seawater. The laboratory results of their study indicated that the Cd content of marine diatoms is related to the Cd ion concentration and inversely related to the Zn and Mn ion concentrations, through competitive inhibition at cellular uptake sites. In contrast, Cullen (2006) predominantly based his model on the results of a single field incubation experiment in the Southern Ocean. This bottle experiment demonstrated that Fe-supplementation decreased the Cd:P043" ratios of an Fe-limited natural phytoplankton assemblage. Both models present conflicting ideas on the mechanism behind these high slopes in nutricline waters of HNLC regions, which this thesis resolves. Sunda and Huntsman (2000) suggest that Fe-limitation causes a chain of events that leads to Cd accumulation in phytoplankton. First, Fe-limited growth causes Zn:P043" ratios in phytoplankton to increase. Iron-limitation has been shown to increase diatoms Si:P043" ratios (Hutchins and Bruland 1998; Takeda 1998) and dissolved Zn and Si concentrations are closely correlated in the ocean (Bruland 1980). Therefore, Sunda and Huntsman (2000) suggested that Si and Zn uptake may be linked, and Fe-limitation may not only cause Si: PO4 3" ratios to in diatoms to increase but Zn:P04 " ratios as well. The high Zn:P04 " ratios in diatoms would lead to a decrease in free Zn ion concentrations [Zn2+] in surface waters. The decrease in [Zn2+] would then 102 cause the induction of a Cd/Co transporter in phytoplankton that increases Cd uptake rates, and thus phytoplankton Cd:PC>43" ratios. Later, Cullen et al. (2003) dismissed the idea that low [Zn ] is the main factor controlling Cd uptake and measured increased Cd accumulation by Fe-limited phytoplankton despite high [Zn2+]. Cullen (2006) suggested that the increase in Cd versus PO4 3" slopes in HNLC sub-surface waters was a direct result of Fe-limitation. The biodilution hypothesis was used by Cullen (2006) to explain enhanced Cd accumulation in Fe-limited phytoplankton. However, Cullen (2006) provided no supporting experimental evidence for biodilution. This study rejected the biodilution hypothesis and found an increase in steady state Cd uptake rates under Fe-limitation despite reduced growth rates. The mechanism presented in this thesis can explain the results of both manuscripts. The up-regulation of non-specific divalent metal transporters as a direct result of Fe-limitation would result in the increase in both phytoplankton Cd:P0 4 3" and Zn:PC>43" ratios. The competitive inhibition Sunda and Huntsman (2003) found in laboratory cultures between Cd, Zn and Mn can be explained by the uptake of these metals through this non-specific divalent metal transporter. Cullen (2006) further suggested that the export of Fe-limited phytoplankton with high C d i P C V " ratios is enough to cause the kink in the C d i P C V " relationship and dissolved surface water Cd:P043" variability. The combined laboratory, field, and global dataset results of this thesis are in accordance with Cullen's (2006) hypothesis, and the export of Fe-limited diatoms seems to be driving the kink in the Cd:PC»43" relationship. Given the range of Cd quotas found among different phytoplankton phyla, it is also likely that export of phytoplankton with varying intracellular C d : P C V " content is causing much of the surface water Cd:P043~ variability seen throughout the ocean. In order to 103 determine whether Fe-limitation and species composition are indeed the culprits controlling dissolved surface water Cd:P04 3" variability in the world's oceans, we plan to use culture data obtained from this thesis to produce an empirical relationship between dissolved Fe concentrations, species composition and phytoplankton Cd:P04 3 " ratios. This empirical relationship will then be used in a model to determine whether we can reproduce the modern dissolved C&.FO43' distribution in the global ocean. This will show whether or not Fe-limitation and species composition are truly driving surface water variability in dissolved Cd:P04 3" ratios in oceanic surface waters and the bimodality in the dissolved Cd:PC>43" relationship in HNLC regions. 4.3 Cd as a paleonutrient proxy Accurate knowledge of the modern dissolved Cd:PC>43" relationship in surface waters and our ability to model how Cd:P043" ratios change temporally and spatially in surface waters of the ocean is crucial for reconstructions of past surface water PO4 3" concentrations. This study is directly relevant for validating the assumptions of predictive models that attempt to account for the variability in surface water dissolved Cd:P043" ratios in the global ocean (Elderfield and Rickaby 2000; Saager and Baar 1993). The modulation of Cd uptake rates by the Fe status of phytoplankton and the large variability in the intrinsic Cd content of different phytoplankton taxa is inconsistent with a Rayleigh model, which uses a constant fractionation factor (Elderfield and Rickaby 2000). The control that Fe has on phytoplankton Cd uptake and species composition introduces an additional level of complexity to reconstructions of past surface water dissolved PO43" concentrations. The two factors identified in this thesis that control Cd 104 biogeochemistry show marked fluctuations both regionally and globally and throughout geologic time. The reduction of Fe-limitation and an enhancement in export primary production has been proposed as a major contributor in lowering atmospheric C O 2 concentrations during the Last Glacial Maximum (LGM) (Martin 1990). It has since been confirmed that the supply of Fe-rich dust increased during glacial periods (Rea 1994; Petit et al. 1999). There is also evidence for a switch in species composition in the Southern Ocean and reduced diatom export production south of the APF during the LGM (Mortlock et al. 1991). Under these conditions, Cd removal by phytoplankton would decrease, leading to higher dissolved Cd:P043~ ratios in surface waters. However, previous reconstructions of surface water PO4 " concentrations based on planktonic foraminiferal Cd:Ca ratios (Elderfield and Rickaby 2000) would underestimate nutrient drawdown and the efficiency of the biological carbon pump during the LGM. Indeed, Elderfield and Rickaby (2000) concluded that during the LGM P O 4 3 " ultilization was low, suggesting that biological carbon pump was inefficient. Elderfield and Rickaby (2000) provided an alternative hypothesis to reconcile their findings with low atmospheric C O 2 concentrations and proposed that an increase in sea-ice cover prevented outgassing of C02-rich water around Antarctica during glacial times. Based on the effects that Fe-limitation and species composition have on surface water dissolved Cd:P043" ratios, re-modeling nutrient utilization in the Southern Ocean during the LGM may suggest that P043" utilization was higher during the LGM than predicted by Elderfield and Rickaby (2000). This would better reconcile the various paleoproxies that indicate enhanced nutrient ultilization south of the APF region in the Southern Ocean during the LGM (Elderfield and Rickaby 2000). 105 4.4 References Anderson, M. A. and F. M. M. Morel. 1982. The influence of aqueous iron chemistry on the uptake of iron by the coastal Thalassiosira weissflogii. Limnol. Oceanogr. 27: 789-813. Cullen, J. T. 2006. On the nonlinear relationship between dissolved cadmium and phosphate in the modern global ocean: Could chronic iron limitation of phytoplankton growth cause the kink? Limnol. Oceanogr. 51: 1369-1380. Elderfield, H., and R. E. M. Rickaby. 2000. Oceanic Cd/P043" ratio and nutrient utilization in the glacial Southern Ocean. Nature 405: 305-310. Kustka, A. B., A. E. Allen, F. M. M. Morel. 2007. Sequence analysis and transcriptional regulation of Fe acquisition genes in two marine diatoms. J. Phycol. (In press). Maldonado, M. T., A. E. Allen, J. S. Chong, K. Lin, D. Leus, N. Karpenko, and S. L. Harris. 2006. Copper-dependent iron transport in coastal and oceanic diatoms. Limnol. Oceanogr. 51: 1729-1743. Maldonado, M. T., and N. M. Price. 2001. Reduction and transport of organically bound iron by Thalassiosira oceanica (Bacillariophyceae). J. Phycol. 37: 298-309. Martin, J. H. 1990. Glacial-interglacial C02 changes: The iron hypothesis. Paleoceanography 5: 1-13. Mortlock, R. A., C. D. Charles, P. N. Froelich, M. A. Zibello, J. Saltzman, J. D. Hays; L. H. Burckle. Evidence for lower productivity in the Antarctic Ocean during the last glaciation. Nature 351: 220-223. Petit, J. R., and others. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice, Antarctica. Nature 399: 429-436. Rea, D. 1994. The paleoclimatic record provided by eolian deposition in the deep sea: The geologic history of wind. Rev. Geophys. 32: 159-195. Saager, P. M. and H. J. W. de Baar. 1993. Limitations to the quantitative application of Cd as a paleoceanographic tracer, based on results of a multi-box model (MENU) and statistical considerations. Global and Planetary Change 8: 69-92. Shaked, Y., A. B. Kuska, and F. M. M. Morel. 2005. A general kinetic model for iron acquisition by eukaryotic phytoplankton. Limnol. Oceanogr. 50: 872-882. Sunda, W. G., and S. A. Huntsman. 2000. Effect of Zn, Mn, and Fe on Cd accumulation in phytoplankton: Implications for oceanic Cd cycling. Limnol. Oceanogr. 45: 1501-1516. 106 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0053307/manifest

Comment

Related Items